Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.enzyme.chem.msu.ru/ryabov/pdf/JACS_00.pdf
Äàòà èçìåíåíèÿ: Mon Jul 24 12:37:06 2000
Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 21:44:41 2012
Êîäèðîâêà:

Ïîèñêîâûå ñëîâà: m 63
J. Am. Chem. Soc. 2000, 122, 5189-5200

5189

Mechanism of Biologically Relevant Deoxygenation of Dimethyl Sulfoxide Coupled with Pt(II) to Pt(IV) Oxidation of Orthoplatinated Oximes. Synthetic, Kinetic, Electrochemical, X-ray Structural, and Density Functional Study
Larissa Alexandrova,, Oleg G. D'yachenko, Grigory M. Kazankov, Vladimir A. Polyakov,§ Pavel V. Samuleev, Enrique Sansores, and Alexander D. Ryabov*,
Contribution from the Instituto de InVestigaciones en Materiales, UNAM, Circuito Exterior s/n, Ciudad UniVersitaria, Apdo. Postal 70-360, Coyoacan CP 04510, Mexico, D.F., Mexico, ´ ´ Department of Chemistry, M. V. LomonosoV Moscow State UniVersity, 119899 Moscow, Russia, and D. I. MendeleeV Moscow UniVersity of Chemical Technology, Miusskaya sq. 9, 125820 Moscow, Russia ReceiVed December 13, 1999

Abstract: Orthometalated aryl oxime complexes cis-(C,S)-[PtII(C6H3-2-CMedNOH-5-R)Cl(Me2SdO)] (1, R ) H (a), MeO, Me, F, and Cl) undergo deoxygenation of dimethyl sulfoxide (DMSO) in methanol in the presence of HCl to afford the Pt(IV) dimethyl sulfide complexes fac-[PtIV(C6H3-2-CMedNOH-5-R)Cl3(Me2S)] (2), the composition of which was confirmed by an X-ray structural study of 2a. The mechanism of the deoxygenation coupled with the oxidation of Pt(II) to Pt(IV) was investigated using cyclic voltammetry, UVvis, and 1H NMR spectrometry techniques at 40-60 °C in the presence of HCl, LiCl, and NaClO4. The conversion of 1 into 2 does not occur intramolecularly and involves two time-resolved phases which were studied independently. The first is the substitution of chloride for DMSO to afford the anionic reactive complexes cis-[Pt(C6H3-2-CMedNOH-5-R)Cl2]- (1Cl), which are involved in the acid-promoted interaction with free DMSO in the second phase. The formation of 1Cl follows the usual two-term rate law kobs1 ) ks + kCl[LiCl], the kCl-driven pathway being negligible for the electron-rich complex with R ) MeO. Thus-generated complexes 1Cl, in contrast to their precursors 1, are more susceptible to oxidation, and the irreversible peak for 1Cl, E(p1), is observed ca. 300 mV more cathodically compared to that of 1. The second phase is acid-catalyzed and at low LiCl concentrations follows the rate expression kobs2[H+]-1 ) k10 + k10[LiCl]. The complexes with the electron-withdrawing substituents R react faster, and there is a linear correlation between log k10 and E(p1). The first-order in the acid is discussed in terms of two kinetically indistinguishable mechanisms involving the rate-limiting either electron transfer from 1Cl to protonated DMSO (mechanism 1) or insertion of the SdO bond of free DMSO into the platinum-hydride bond of the reactive hydride complex of Pt(IV), cis-[Pt(C6H32-CMedNOH)(H)Cl2], to afford a {Pt-SMe2-OH} fragment. Its protonation by HCl and dissociation of water gives the final product 2 (mechanism 2). 1H NMR evidence is presented for the formation of the hydride species on protonation of a Pt(II) complex, whereas a density functional study of the two mechanisms indicates that mechanism 2 is less energy demanding. The system studied is viewed as a functioning mimetic of the Mo-dependent enzyme DMSO reductase because of several common features observed in catalysis.

Introduction Metal-ion-mediated deoxygenation of sulfoxides into corresponding sulfides has recently attracted much attention in connection with its synthetic virtue,1,2 the structural and mechanistic behavior of sulfoxide ligands in the coordination sphere of transition metal complexes,3 and the biological
* Address correspondence to this author (temporary address) at Department of Chemistry, Carnegie Mellon University, 4400 Fifth Ave., Pittsburgh, PA 15213. Fax: (412) 268-1061. E-mail: ryabov@enz.chem.msu.ru. Ciudad Universitaria. M. V. Lomonosov Moscow State University. § D. I. Mendeleev Moscow University of Chemical Technology. Visiting scientist at the Department of Chemistry, M. V. Lomonosov Moscow State University. (1) Kukushkin, V. Y. Usp. Khim. 1990, 59, 1453-1468. (2) Kukushkin, V. Y. Coord. Chem. ReV. 1995, 139, 375-407. (3) Calligaris, M.; Carugo, O. Coord. Chem. ReV. 1996, 153, 83-154. Calligaris, M. Croat. Chem. Acta 1999, 72, 147-169.

deoxygenation of dimethyl sulfoxide catalyzed by the molybdenum-dependent enzyme.4 Since the pioneering work of Kukushkin et al.,5 much effort has been devoted to the investigation of the deoxygenation promoted by acido and related complexes of platinum(II), which is accompanied by the oxidation of Pt(II) into Pt(IV) (eq 1). The process shown

[PtIIX3(SOMe2)]- + 2HCl f [PtIVCl2X3(SMe2)]- + H2O (1)
by eq 1 has been demonstrated by several groups of authors6-8 and confirmed by X-ray structural characterization of the
(4) Weiner, J. H.; Rothery, R. A.; Sambasivarao, D.; Trieber, C. A. Biochim. Biophys. Acta 1992, 1102, 1-18. (5) Kukushkin, Y. N.; Vyazmenskii, Y. E.; Zorina, L. I.; Pazukhina, Y. L. Zh. Neorg. Khim. 1968, 13, 1595-1599. (6) Pakhomova, I. V.; Konovalov, L. V.; Komyagin, N. T.; Yanovsky, A. I.; Struchkov, Y. T. Zh. Obsch. Khim. 1988, 58, 733-738.

10.1021/ja994342k CCC: $19.00 © 2000 American Chemical Society Published on Web 05/17/2000


5190 J. Am. Chem. Soc., Vol. 122, No. 21, 2000 products formed. Nevertheless, reaction 1 seems to be mechanistically rather controversial, if one considers, for example, the known standard redox potentials of the corresponding halfreactions 2 and 3.

AlexandroVa et al. complexes. Orthometalated complexes are superior for mechanistic studies12 because of advantages utilized by us previously in studies of mechanisms of reactions of platinacycles.13-17 First, the covalently -bound, substitutionally inert orthoplatinated ligand limits possible substitution and exchange processes at Pt(II). As a result, mechanistic information obtained by different techniques is more straightforward and easier to rationalize. Second, electron-donating or -withdrawing groups at the -bound aryl ring cause variations in the electronic features of the entire molecule, the steric properties being unaffected. Third, an orthoplatinated organic ligand is located in the Pt(II) plane, and there is no steric retardation for incoming ligands to approach the Pt(II) center. Apart from several special cases,18-20 the attack occurs via associative mechanisms.21,22 In addition, platinacycles 1 further demonstrate wide applicability of oxime ligands in the comprehensive coordination chemistry.23,24 Thus, in this work we report on a detailed mechanistic, structural, and density functional study of reaction 1, using as an example the square-planar sulfoxide platina(II)cycles 1 which transform into octahedral Pt(IV) complexes 2. The investigation includes synthesis of complexes 1 and 2, an X-ray structural study of 2a, spectral (UV-vis, 1H NMR) and cyclic voltammetry (CVA) examination of 1 and 2 in the presence of LiCl and HCl, and a kinetic study of DMSO substitution for chloride in 1, as well as of the acid-promoted formation of the final reaction products 2. The mechanistic conclusions are discussed together with results of the quantum-chemical simulations using a density functional theory. Results Synthesis and Properties of Platina(II)cycles 1. Sulfoxide platinum(II) complexes 1 were prepared from the corresponding ring-substituted acetophenone oximes and cis-[PtCl2(SOMe2)2]25 in refluxing methanol (30-36 h) as described in previous work for the unsubstituted oxime.11 Platinacycles 1 precipitated from the reaction solution on cooling as spectrally pure yellow crystals. Isolated yields were usually 45-67% (Table 1). The somewhat lower yield for fluoro-substituted compound 1d (23%) is probably due to its better solubility in MeOH. In the previous work, we have concluded that the steric effect is responsible
(12) Canty, A. J.; van Koten, G. Acc. Chem. Res. 1995, 28, 406-413. (13) Ryabov, A. D.; Kuzmina, L. G.; Dvortsova, N. V.; Stufkens, D. J.; van Eldik, R. Inorg. Chem. 1993, 32, 3166-3174. (14) Schmulling, M.; Ryabov, A. D.; van Eldik, R. J. Chem. Soc., Dalton Trans. 1994, 1257-1263. (15) Ryabov, A. D.; Kuz'mina, L. G.; Polyakov, V. A.; Kazankov, G. M.; Ryabova, E. S.; Pfeffer, M.; van Eldik, R. J. Chem. Soc., Dalton Trans. 1995, 999-1006. (16) Krooglyak, E. V.; Kazankov, G. M.; Kurzeev, S. A.; Polyakov, V. A.; Semenov, A. N.; Ryabov, A. D. Inorg. Chem. 1996, 35, 4804-4806. (17) Gossage, R. A.; Ryabov, A. D.; Spek, A. L.; Stufkens, D. J.; van Beek, J. A. M.; van Eldik, R.; van Koten, G. J. Am. Chem. Soc. 1999, 121, 2488-2497. (18) Frey, U.; Helm, L.; Merbach, A. E.; Romeo, R. J. Am. Chem. Soc. 1989, 111, 8161-8165. (19) Romeo, R.; Alibrandi, G. Inorg. Chem. 1997, 36, 4822-4830. (20) Romeo, R.; Plutino, M. R.; Elding, L.-I. Organometallics 1997, 16, 5909-5916. (21) Basolo, F.; Pearson, A. G. Mechanism of Inorganic Reactions. A Study of Metal Complexes in Solution, 2nd ed.; Wiley: New York, 1967. (22) Kotowski, M.; van Eldik, R. In Substitution, Isomerization and Related Reactions of Four-Coordinate Complexes; van Eldik, R., Ed.; Elsevier: Amsterdam, Oxford, New York, Tokyo, 1986; Vol. 7, pp 219271. (23) Kukushkin, V. Y.; Tudela, D.; Pombeiro, A. J. L. Coord. Chem. ReV. 1996, 156, 333-362. (24) Kukushkin, V. Y.; Pombeiro, A. J. L. Coord. Chem. ReV. 1999, 181, 147-175. (25) Ranatunge-Bandarage, P. R. R.; Robinson, B. H.; Simpson, J. Organometallics 1994, 13, 500-510.

[PtCl6]

2-

+ 2e f [PtCl4]

2-

E2°

(2)

Me2SdO + 2H+ + 2e f Me2S + H2O

E3° (3)

The values available, viz., E2° and E3° of 0.749 and 0.16 V (pH 7),10 respectively, versus NHE indicate that reaction 1 is formally thermodynamically unfavorable, the uphill process with G° ¨ ca. 110 kJ mol-1. Thus, it was naive to expect a simple reaction mechanism for the Pt(II)-induced deoxygenation of sulfoxides involving, for example, a transfer of two electrons from Pt(II) to S(IV). Rather, a nonconcerted, multistep pathway would have been anticipated, and our understanding of it could throw light on the mechanism of the related enzymatic reaction.4

It should, however, be mentioned that in contrast to the enzymatic deoxygenation, the Pt(II)-promoted reactions often proceed under harsh conditions. The formation of dimethyl sulfide occurs under reflux in the presence of high concentrations of HCl.2 Our recent work suggested an approach to make the reaction conditions milder.11 When 2-methyl-1-phenylbutan-1one oxime was reacted with cis-[PtCl2(SOMe2)2] in methanol, the major reaction product was the orthometalated Pt(IV) complex with coordinated dimethyl sulfide, formulated as 2g on the basis of an X-ray structural investigation. Together with 2g, a small amount of dimethyl sulfoxide Pt(II) complex 1g was isolated, showing that 1 could be a precursor of 2. Direct proof is gained in this work. Less bulky ring-substituted oxime complexes 1 have been shown to transform cleanly into dimethyl sulfide Pt(IV) complexes 2 in MeOH at 40-60 °C in the presence of a small amount of HCl. Reaction 4 afforded 2a in good yield and provided an object for the mechanistic investigation of the deoxygenation of sulfoxides in the presence of Pt(II)
(7) Fanizzi, F. P.; Natile, G.; Maresca, L.; Manotti-Lanfredi, A. M.; Tiripicchio, A. J. Chem. Soc., Dalton Trans. 1984, 1467-1470. (8) Kukushkin, V. Y.; Belsky, V. K.; Konovalov, V. E.; Aleksandrova, E. A.; Pankova, E. Y.; Moiseev, A. I. Phosphorus, Sulfur, Silicon 1992, 69, 103-117. (9) Bard, A. J.; Faulkner, L. R. Electrochemical Methods. Fundamentals and Applications; John Wiley & Sons: New York, 1980. (10) Wood, P. M. FEBS Lett. 1981, 124, 11-14. (11) Ryabov, A. D.; Kazankov, G. M.; Panyashkina, I. M.; Grozovsky, O. V.; Dyachenko, O. G.; Polyakov, V. A.; Kuzmina, L. G. J. Chem. Soc., Dalton Trans. 1997, 4385-4391.


Biologically ReleVant Deoxygenation of DMSO
Table 1. Yields, UV-Vis, and 1H NMR (CDCl3) Spectral Data for Platinacycles 1 complex (R, yield, %) 1a (H, 67) UV-vis (MeOH)a (max ( ,M-1 cm-1)) 303 (5100), 319 (4800) 304,b 322 1b (MeO, 45) 1c (Me, 48) 1d (F, 23) 1e (Cl, 51) 3
a b

J. Am. Chem. Soc., Vol. 122, No. 21, 2000 5191

C-CH 2.39 (J ) 2.44 (J ) 2.45 (J ) 2.37 (J ) 2.38 (J ) 2.37 (J )

3

S-CH

3

H NMR (CDCl3, J in Hz) N-OH R 10.15 (J ) 6.6) n. o. n. o. 10.04 (J ) 6.4) 10.01 (J ) 6) 10.10 (J ) 6.1) -

1

Ar

6.3) 6.4)
c

314 (5500), 326 (5800) 308 (3900), 323 (4100) 302 (4200), 317 (4100) 306 (5800), 322 (5650) 307 (5400), 329 (5500), 370 sh, 308,b 332b

5.9)c 6) 6.4) 6.1)

3.54 J ) 23.8) 3.58 (J ) 24.8) 3.61 (J ) 24.5) 3.53 (J ) 24) 3.54 (J ) 24) 3.54 (J ) 24.2)

3.84 (OCH3) 2.38 (ArCH3) -

7.1-7.15 (m, H3-H5), 8.03 (d, J ) 8, JPt-H 51) 7.0-7.35 (m, H3-H5), 8.0 (d, J ) 8, JPt-H ) 44) 6.73 (dd, J ) 8.4, 2.6, H4), 7.28 (d, J ) 8.4, H3), 7.70 (d, J ) 2.6, JPt-H ) 57, H6) 6.95 (d, J ) 8.1, H3), 7.06 (d, J ) 8.1, H4), 7.86 (s, JPt-H ) 48, H6) 6.83 (td, JHH 2.6, JHF ) 8.4, H4), 7.2 (dd, H3), 7.82 (dd JHH ) 2.6, JHF ) 10.4, H6) 7.1-7.3 (m, H3, H4), 8.05 (d, J ) 1.95, JPt-H ) 55.2, H6)

All complexes have a shoulder at ca. 350 nm. b Chloroform solvent. c Methanol-d4 solvent.

(JPtH ) 25 Hz) (Figure 1a), and a weak with the satellites. The formation of free the minor product formed in eq 6a, since the 1H NMR spectrum of the methanolic (SOMe2)2], whereas the 3.55 signal is

signal around 3.55 DMSO is ascribed to free DMSO is seen in solution of cis-[PtCl2due to the solvolysis

Figure 1. 1H NMR spectra of complex 1a obtained at different LiCl concentrations at 64 °C in methanol-d4.

reaction 6b. Evidence for this is its disappearance on addition of 0.047 M LiCl. Added LiCl increases the fraction of unbound DMSO ( 2.68) and gives rise to a new singlet at 2.19 (JPtH ) 10.2 Hz) (Figure 1b). There is an additional weak singlet at 3.42 (JPtH ) 22 Hz), the origin of which is not completely clear. It could tentatively be ascribed to the product formed in equilibrium 7. At 0.14 M LiCl (Figure 1c), the intensity of new

for the dominant formation of complex 2g, whereas 1g was isolated only in a very low yield.11 In this study, we tested the long-chain phenyl n-heptyl ketone oxime as a sterically hindered but less bulky molecule. In accord with the previous proposal, formation of both 1f and 2f was observed. However, the reaction between cis-[PtCl2(SOMe2)2] and phenyl n-heptyl ketone oxime is time-controlled, and complex 1f is the only species that can be formed. Platinacycles 1 were characterized by UV-vis and 1H NMR spectral data (Table 1). The structure of 1c was confirmed by an X-ray single-crystal investigation.26 Reaction of 1 with LiCl. Complexes 1 react with LiCl in MeOH according to eq 5, which was investigated by 1H NMR, cyclic voltammetry, and UV-vis techniques. The 1H NMR

resonances at 2.19 and 2.68 increases further with a decrease of those at 2.44 (JPtH ) 6.4 Hz) and 3.58 from the oxime methyl and bound DMSO, respectively. The tendency holds at 0.24 M LiCl (Figure 1d). It should be mentioned that (i) the changes are rather fast at 64 °C and the spectrum does not practically fluctuate until next addition of LiCl and (ii) the signal intensities from the coordinated and free DMSO, as well as from the oxime methyls ( 2.19 and 2.44), vary correspondingly. The spectral changes suggest that the substitution of chloride for DMSO (eq 5) is a dominant process under these conditions. The electrochemical properties of platinacycles is a subject of independent interest,27-32 and CVA was applied to 1a-e (MeOH, 0.1 M NaClO4, 25 and 55 °C). The voltammograms
(27) Headford, C. E. L.; Mason, R.; Ranatunge-Bandarage, P. R.; Robinson, B. H. J. Chem. Soc., Chem. Commun. 1990, 601-603. (28) Braterman, P. S.; Song, J.-I.; Wimmer, F. M.; Wimmer, S.; Kaim, W.; Klein, A.; Peacock, R. D. Inorg. Chem. 1992, 31, 5084-5088. (29) Minghetti, G.; Pilo, M. I.; Sanna, G.; Seeber, R.; Stoccoro, S.; Laschi, F. J. Organomet. Chem. 1993, 452, 257-261. (30) Kotler, A. N.; Puzyk, K. G.; Balashov, V. T. Elektrokhimia 1995, 31, 746-749. (31) Crespo, M.; Grande, C.; Garcia, M. M.; Klein, A.; Font-Bardia, M.; Solans, X. J. Organomet. Chem. 1998, 563, 179-190.

spectrum of 1a (0.02 M) recorded in D3COD at 64 °C contains extra resonances as compared to those in CDCl3 (Figure 1). There is a weak singlet at 2.68 without Pt satellites from free DMSO, a strong resonance from coordinated DMSO at 3.58
(26) Revenco, M.; Alexandrova, L.; Ryabov, A. D. To be published.


5192 J. Am. Chem. Soc., Vol. 122, No. 21, 2000

AlexandroVa et al.

Figure 2. line) and at 55 °C, methanol see text.

Cyclic 75 min 0.1 M solvent.

voltammograms of complex 1a in the absence (broken after addition of 0.023 M LiCl (solid line) obtained NaClO4, [1a] 0.0006 M, at a scan rate 0.1 V s-1 in (Inset) Variation of E(p1) with time. For more details,

Figure 3. Spectral changes of complex 1a (0.00025 M) with time in the presence of 0.005 M LiCl at 60 °C in MeOH. Spectrum a was recorded at time t ) 0 after addition of LiCl; other spectra were obtained at t ) 5, 15, 35, 50, 80, and 140 min, respectively.

Table 2. Oxidation Potentials p1 and p2 (in Volts) for Complexes 1 with and without 0.02 M LiCl in MeOH at 55 °C (0.1 M NaClO4, SCE) in the absence of LiCl complex (R) 1a 1c 1d 1b 1e
a

in the presence of LiCl p1 0.875 0.83 0.905 0.785 0.93 p2 1.20 1.17 1.26 (shoulder) 1.03; 1.14 1.25

p1 0.81 n.o. 0.88 n.o. n.o.
a a

p2 1.18 1.17 1.17 1.04; 1.40 1.26

(H) (Me) (F) (MeO) (Cl)

Poorly resolved.

for 1a with and without added LiCl are shown in Figure 2. There is one irreversible oxidation peak, p2, around 1.2 V (versus SCE), together with a poorly resolved shoulder, p1, at ca. 0.8 V at [LiCl] ) 0 (broken line). Addition of LiCl (0.02 M) develops the oxidation peak p1. Its position is time-dependent (inset to Figure 2), and a constant value is reached after 10-15 min at 55 °C. The final voltammogram (solid line) contains two irreversible peaks, p1 and p2, at 0.87 and 1.2 V, which are ascribed to the 2e oxidations of 1aCl and 1a, respectively, plausibly via transient Pt(III) complexes.33 The LiCl-induced effects are fully consistent with eq 5. The CVA data suggest that, in contrast to neutral complexes 1, anionic species 1Cl undergo oxidation at ca. 300 mV more negative potentials. Ringsubstituted complexes 1b-e were similarly tested, and the p1 and p2 values are given in Table 2. The p1 values for 1Cl depend on the nature of R, and the peak potentials are more cathodic for the electron-rich complexes, the largest difference being 145 mV for 1b and 1e. Reaction 5 is observed by UV-vis spectroscopy (Figure 3). In MeOH as solvent, complex 1a shows two absorption maxima at 303 and 319 nm, together with a broad plateau at 350 nm. Addition of LiCl develops a band at 350 nm, and the maxima at 303 and 319 nm diminish their intensity. Isosbestic points at 282 and 328 nm suggest that two absorbing species dominate in solution. The solvento species17 (eq 6b) plays no important role here. To give more evidence for this, the structural analogue of 1a, trans-(N,N)-[PtII(C6H4-2-CMedNOH)Cl(py)] (3), was prepared and investigated by UV-vis spectroscopy. The spectra
(32) Balashev, K. P.; Puzyk, M. V.; Kotlyar, V. S.; Kulikova, M. V. Coord. Chem. ReV. 1997, 159, 109-120. (33) Bandoli, G.; Caputo, P. A.; Intini, F. P.; Sivo, M. F.; Natile, G. J. Am. Chem. Soc. 1997, 119, 10370-10376.

Figure 4. Pseudo-first-order rate constants kobs1 as a function of LiCl concentration for the substitution of chloride for dimethyl sulfoxide in complex 1a at different temperatures in methanol solvent.

of 1a and 3 recorded in CHCl3 are similar (Table 1). More important is that their spectra in CHCl3 and MeOH are also similar. Additionally, the spectrum of 3 in MeOH is practically insensitive to LiCl at 0.001-5.0 M, even at 55 °C, proving the insignificance of eq 6b. The LiCl-induced absorbance changes depend on the concentration of LiCl added and can be reversed by addition of extra DMSO after the equilibration of 1a with LiCl. An attempted isolation of 1aCl after the methanol solution of 1a was refluxed in the presence of a 10-fold excess of LiCl for 30 h resulted, after cooling, in a quantitative recovery of 1a, confirming the reversible nature of eq 5. Equilibria and Kinetics of Substitution of DMSO by Clin Complexes 1. The kinetics of reaction 5 was studied as a function of the LiCl, NaClO4, and HCl concentrations in MeOH at 42-60 °C. It was monitored at 353 nm for all complexes, except for 1b (327 nm), at [LiCl]t . [1] to ensure the pseudofirst-order conditions. The values of kobs1 were evaluated at different concentrations of NaClO4 (0-0.1 M), HCl (0.0050.05 M), and LiCl (0.003-0.02 M). The rate constants kobs1 were insensitive to NaClO4 and HCl concentrations. The dependence of kobs1 on [LiCl] is demonstrated in Figure 4. As can be seen, the kobs1 is practically independent of [LiCl] at 42 and 50 °C, whereas the LiCl-dependent pathway manifests at higher temperatures, viz., 55 and 60 °C. These facts are in


Biologically ReleVant Deoxygenation of DMSO

J. Am. Chem. Soc., Vol. 122, No. 21, 2000 5193
Table 3. Kinetic Parameters ks and kCl for the Substitution of Chloride for DMSO in Complexes 1 (Eq 8) and k10 and k10 for the Conversion of 1 into 2 (Eq 10) in MeOH
complex (R) 1a (H) T, °C 104ks, s-1 2.00 ( 0.16 3.2 ( 0.28 4.87 ( 0.28 5.5 ( 0.3 3.67 ( 0.3 1.52 ( 0.28 1.40 ( 0.03 4.17 ( 0.03
a

102kCl, M-1 s-1 0 0 0.83 ( 0.33 2.2 ( 0.3 1.4 ( 0.3 2.38 ( 0.25 0.55 ( 0.03 3.28 ( 0.33

102k10, M-1 s-1

M

k10, -2 s-1

42 50 55 60 1c (Me) 55 1d (F) 55 1b (MeO) 55 1e (Cl) 55
a

2.4 ( 0.2 2.4 ( 0.1 2.97 ( 0.03 1.9 ( 0.1 3.3 ( 0.1

0.37 ( 0.07 0.29 0.48 0.10 0.70
-1

( ( ( (

0.02 0.01 0.02 0.03

Hsq ) 24.5 ( 2.8 kJ mol-1, Ssq ) -81 ( 14 J K

mol-1.

Figure 5. Dependencies of log ks and log kCl on the E(p1) potential for the substitution of chloride for dimethyl sulfoxide in complexes 1 at 55 °C in methanol solvent.

work,37 the equilibrium concentrations of Cl- were calculated using the KLiCl which were then used in the fitting of the data to eq 9 (for evaluation of eq 9, see Appendix). Here, Ao is the

(A - Ao) )

agreement with the accepted scheme for the ligand substitution in square-planar complexes,21,34 according to which the rate law is given by eq 8. Here, kCl refers to the chloride-dependent

(

0.25Kx2[Cl-]2 + [Pt]tKx[Cl-] -

Kx[Cl-] 2 (9)

)

kobs1 ) ks + kCl[LiCl]

(8)

substitution pathway. Since the reaction rate is independent of chloride concentrations in several cases (Figure 4 and Figure 1S in the Supporting Information), ks is the sum of the rate constants for the reversible reaction and the solvent-dependent substitution pathway; thus, the relative contribution of each path is difficult to estimate. Likely, the latter dominates, since reaction 5 proceeds at appreciable rates even when zero-order in LiCl holds. The rate law 8 is valid for ring-substituted complexes (Figure 1S), and the rate constants ks and kCl for 1a-e are summarized in Table 3. These do not correlate with the Hammett constants, presumably due to the fact that the R group affects the metal center via two channels, viz., the covalent -Pt-C and coordinative PtrN bonds, the contribution of each being difficult to take into account.35 Therefore, some other effective parameter related to the electronic density at the entire complex could be more useful. Such proved to be the p1 peak potentials for 1Cl (Table 2). As can be seen in Figure 5, there is a correlation between log kCl and E(p1), with the anticipated positive slope of 6 ( 2 V-1, since the electron-deficient compounds should react faster with the negatively charged incoming ligand. The slope is much lower for log ks (1 ( 2 V-1). The pathway is less electronically sensitive to the electroneutral solvent molecule. The activation enthalpy (Hsq) and entropy (Ssq) for the ks pathway were calculated from the linear ln(ks/T) versus T -1 plot for 1a (Figure 2S, Supporting Information). The numerical values are given in Table 3. The calculation of kobs1 using the nonlinear least-squares curve-fitting provided the infinite absorbances (A) at a given concentration of LiCl. These were dependent on [LiCl] added (Figure 5S, Supporting Information) and were used for estimating the equilibrium constant Kx (eq 5). The dissociation of LiCl in MeOH should be taken into account, the KLiCl for which equals 1.58 â 10-2 M,36 since the experimental data were evaluated at [LiCl]t lower than 0.02 M. As in our previous
(34) Wilkins, R. G. Kinetics and Mechanism of Reactions of Transition Metal Complexes, 2nd ed.; VCH: Weinheim, 1991. (35) Ryabov, A. D. Chem. ReV. 1990, 90, 403-424. (36) Janz, C. J.; Tomkins, R. P. T. Nonaqueous Electrolytes Handbook; Academic Press: New York, 1972.

initial absorbance, is the difference in the extinction coefficients for 1Cl and 1, [Pt]t is the total concentration of Pt(II), and [Cl-] is the calculated equilibrium concentration of chloride. The data such as those in Figure 3S (Supporting Information) were fitted to eq 9, and the best-fit values of Kx were equal to 0.15 ( 0.05, 0.17 ( 0.06, and 0.19 ( 0.12 at 50, 55, and 60 °C, respectively. The equilibrium constants increase slightly with the temperature, the reaction enthalpy H° being 10.6 ( 0.2 kJ mol-1 (calculated from the ln Kx versus T -1 plot; S° ) 25 ( 1 J K-1 mol-1). Protonation of Complex 1c. To demonstrate the ability of Pt(II) complexes 1 to generate the Pt(IV) hydride species on interaction with strong acids, complex 1c, as the most electronrich one with the lowest number of oxygen atoms, was treated with triflic acid in methanol. Compound 1c (2.5 mg) was reacted with 15 µL of the acid in 1 mL of CH3OH for 15 min. The solvent was then removed in a vacuum, and the 1H NMR spectrum of the residue was recorded immediately after addition of dry methanol-d4 showed the characteristic resonance at -14.8 with JPtH ) 1180 Hz. Its integral intensity corresponded to one proton as compared with other signals, which were found to have chemical shifts similar to those of the parent complex 1c, and decreased with time due to instability of the compound and/or exchange of the hydride with deuterium atoms from the solvent. Preparation of Complexes 2 from 1 in the Presence of HCl in MeOH. Sulfoxide Pt(II) complexes 1 transform readily into Pt(IV) sulfide complexes 2 in the MeOH-HCl system. The reaction usually takes 5-6 h, and the product precipitates from the solution on cooling. No additional crystallization is needed for purification. Complexes 2 were characterized by the analytical and 1H NMR data, whereas the structure of 2a was confirmed by a single-crystal X-ray structural study. The 1H NMR spectrum of 2a recorded in CDCl3 is simple compared to that of its analogue 2g reported previously.11 It consists of three methyl resonances at 2.20, 2.62, and 2.73 with 195Pt satellites. The Pt-H coupling constants equal 40, 35.5, and 8.6 Hz, respectively, indicating that the former two refer to the diastereotopic methyls of coordinated Me2S and the complex has no plane of symmetry. The signal at 2.73 is ascribed to the methyl group of acetophenone oxime. It is worth
(37) Ferstl, W.; Sakodinskaya, I. K.; Beydoun-Sutter, N.; Le Borgne, G.; Pfeffer, M.; Ryabov, A. D. Organometallics 1997, 16, 411-418.


5194 J. Am. Chem. Soc., Vol. 122, No. 21, 2000
Table 4. Yields and 1H NMR (CDCl3-D3COD) Spectral Data for Platinacycles 2
complex (R) 2a (H) 2b (MeO) 2c (Me) 2d (F) 2e (Cl)
a

AlexandroVa et al.

isolated yield, % 72 34 48 21 43

1H

NMR C-CH3 8.6)a

S-CH

3

R 3.81 (OCH3) 2.49 (ArCH3) -

Ar 7.34-7.45 (m, H3-H5), 7.88 (d, J ) 8, JPt-H ) 23.4, H6) 7.2-7.6 (m, H3-H5), 7.87 (d, J ) 7.5, JPt-H ) 25.6, H6) 6.74 (dd, J ) 8.54, 2.44, H4), 7.23 (d, J ) 8.54, H3), 7.28 (d, J ) 2.44, JPt-H ) 30, H6) 7.16 (d, J ) 7.8, H4), 7.30(d, J ) 7.8, H3), 7.66 (s, JPt-H ) 28, H6) 7.14 (ddd, JHH ) 8.3, 0.4, JHF ) 8.1, H3), 7.58 (dd, JHH ) 8.3, 1.5, JHF ) 8.1, H4), 7.59 (ddd, JHH ) 1.5, JHF ) 8.1, JPt-H ) 28, H6) 7.38-7.41 (m, H3, H4), 7.85 (d, JHH ) 2, JPtH ) 28, H6)

2.73 (J ) 2.74 (J ) 8.7)b 2.57 (J ) 8.3) 2.72 (J ) 8.0) 2.78 (J ) 8.6) 2.75 (J ) 8.0)
b

2.20 (J ) 40.0), 2.62 (J ) 35.5) 2.24 (J ) 40.2), 2.35 (J ) 38.8) 2.15 (J ) 40.0), 2.40 (J ) 37.1) 2.25 (J ) 40.6), 2.53 (J ) 36.7) 2.32 (J ) 40.5), 2.53 (J ) 37.7) 2.29 (J ) 40.0), 2.53 (J ) 36.4)

In CDCl3 as solvent, (NOH) 11.1. In pure methanol-d4.

Figure 6. Crystal structure of complex 2a. Table 5. Bond Distances (in å) and Bond Angles (in Degrees) for Complex 2aa
Pt1-Cl1 Pt1-Cl2 Pt1-Cl3 Pt1-S1 Pt1-N1 Pt1-C1 S1-C9 S1-C10 O1-N1 Cl1-Pt1-Cl2 Cl1-Pt1-S1 Cl1-Pt1-N1 Pt1-C1-C2 C3-C2-C7 Cl2-Pt1-N1 Cl2-Pt1-C1 Cl3-Pt1-N1 C2-C1-C6 C2-C3-C4 N1-Pt1-C1 Pt1-S1-C10 C6-C1-C7 Pt1-N1-O1 O1-N1-C7 N1-C7-C8 C2-C7-C8 N1-C7-C2 2.306(2) 2.330(2) 2.444(2) 2.336(2) 2.019(6) 2.017(5) 1.807(9) 1.788(8) 1.385(7) 91.64(8) 92.44(8) 174.6(2) 111.1(4) 123.9(7) 89.5(2) 87.3(2) 91.4(2) 121.4(6) 118.9(8) 80.1(2) 108.4(3) 153.5(5) 121.7(4) 119.3(6) 123.1(6) 124.3(6) 112.7(7) N1-C7 C1-C2 C1-C6 C2-C7 C2-C3 C3-C4 C4-C5 C5-C6 C7-C8 Cl1-Pt1-Cl3 C1-C2-C7 Cl1-Pt1-C1 Cl2-Pt1-Cl3 Cl2-Pt1-S1 Pt1-C1-C6 Cl3-Pt1-S1 Cl3-Pt1-C1 S1-Pt1-N1 S1-Pt1-C1 Pt1-S1-C9 C9-S1-C10 C3-C4-C5 Pt1-N1-C7 C1-C2-C3 C4-C5-C6 C1-C6-C5 1.275(9) 1.41(1) 1.37(1) 1.46(1) 1.409(9) 1.40(1) 1.39(1) 1.40(1) 1.51(1) 93.82(7) 117.0(5) 94.7(2) 91.60(8) 175.90(8) 127.5(5) 87.73(7) 171.4(2) 86.4(2) 92.7(2) 108.0(4) 99.3(4) 121.1(7) 119.0(5) 119.1(7) 120.1(8) 119.4(8)

Figure 7. 1H NMR spectral changes of complex 1a with time observed in the presence of 0.25 M DCl at 64 °C in methanol-d4.

a Numbers in parentheses are estimated standard deviations in the least significant digits.

noting that the Pt-H coupling constants for the S-CH3 resonances differ significantly in complexes 1 and 2 (cf. with JPtH ) 24 Hz in 1a). The 1H NMR spectral data for other ringsubstituted complexes 1b-e are also tabulated (Table 4). The crystal structure of Pt(IV) complex 2a is shown in Figure 6, whereas bond lengths and bond angles are collected in Table 5. The chloro ligands are facially arranged, the Pt-Cl bond distances being 2.306(2), 2.330(2), and 2.444(2) å for chloride

trans to N, S, and C, respectively, similar to those in 2g.11 The bond distances match the ground-state trans influence which increases in the series N < S < C.21 Other structural features of 2a are similar to those of 2g and related Pt(IV) complexes which were discussed in detail previously (see ref 11 and references cited therein). The X-ray crystal structure of the relevant Pt(IV) complex trans-(S,N)-[Pt(OC6H2Cl2CHdNOH)Cl3(Me2SO)] has recently been published by Kukushkin et al.38 Its Pt-S and Pt-N bond distances of 2.301 and 2.036 å, respectively, are comparable with those in 2a, but the ligand arrangement is different. The Pt(IV) sulfoxide compound adopts the meridional coordination of chloro ligands. 1H NMR and UV-Vis Studies of Conversion of 1 into 2. When DCl is added instead of LiCl to the solution of 1a in methanol-d4, the 1H NMR spectra of 1a are similar at the early stages in both cases (cf. Figures 1 and 7b,c). After 15 min, the substitution of chloride for DMSO is noticeable due to the singlet at 2.19. Product 2a shows up after 15-30 min, as suggested by three new resonances at 2.24, 2.35, and 2.74 (Figure 7b,c) from two diastereotopic S-CH3 and C-CH3 methyls, respectively. The spectra recorded after 45 and 190 min demonstrate that the Pt(II) species transform into 2a, which dominates after 3 h. The integration allows us to follow the concentration of each species in solution. The concentration profiles in Figure 8 deserve comment. First, the amount of 1a decreases gradually as evidenced by the signals from the coordinated DMSO and the methyl group of the orthoplatinated aryl oxime. Second, the 1aCl profile exhibits a maximum at the same point on the time scale as the inflection point at the 2a profile. Thus, complex 1aCl is a conceivable precursor of 2. It
(38) Kukushkin, Y. N.; Krylov, V. K.; Kaplan, S. F.; Calligaris, M.; Zangrando, E.; Pombeiro, A. J. L.; Kukushkin, V. Y. Inorg. Chim. Acta 1999, 285, 116-121.


Biologically ReleVant Deoxygenation of DMSO

J. Am. Chem. Soc., Vol. 122, No. 21, 2000 5195

Figure 10. Pseudo-first-order rate constants kobs2 for the second phase as a function of HCl concentration for conversion of complex 1a into 2a in the absence (a) and in the presence of 0.1 M NaClO4 (b) at 55 °C in methanol solvent.

Figure 8. Concentration profiles obtained from the data analogous to those shown in Figure 7 for various species formed during the conversion of 1a into 2a. Profiles for 1a (9 and O) are obtained from resonances from coordinated DMSO and C-CH3 ( 2.44), respectively; profile for 1aCl (0) is an average from resonances from two diastereotopic S-CH3 groups and C-CH3 group; 0.25 M DCl at 64 °C in methanol-d4.

Figure 11. Pseudo-first-order rate constants kobs2 for the second phase as a function of LiCl concentration for conversion of complexes 1 into 2; 0.03 M HCl, 0.1 M NaClO4, 55 °C, methanol solvent.

Figure 9. Spectral changes of complex 1a (0.00025 M) with time in the presence of 0.03 M HCl and 0.1 M NaClO4 at 60 °C in MeOH. Spectrum a was recorded at time t ) 0 after addition of HCl; other spectra were obtained at t ) 5, 10, 15, 20, 30, and 50 min, respectively. (Inset) Biphasic nature of the deoxygenation, shown by following the absorbance change at 350 nm at 0.008 M HCl and 0.094 M NaClO4 at 60 °C in MeOH.

is worth noting that the inset in Figure 9 matches the concentration profile for 1aCl in Figure 8, and this provides extra evidence for 1Cl as the key intermediate. Third, the concentration of free DMSO is practically constant over the course of the experiment. The same run carried out at 0.1 M NaClO4 showed a 2-fold rate increase. The conversion of 1a into 2a can be clearly observed by UVvis spectroscopy at higher HCl concentrations (Figure 9). The process is characterized by the absorbance decrease at 320 and 350 nm, the isosbestic point being held at 301 nm. There are two species involved, viz., 1aCl and 2a, since the final spectrum corresponds to that of 2a under the identical conditions.39 The

absorbance variation at 350 nm (inset to Figure 9) at lower [HCl] (0.008 M) confirms the biphasic conversion of 1 into 2. Kinetics of Formation of Complexes 2. At [HCl] > 0.008 M, the first phase proceeds rather fast when followed by conventional spectrophotometry, and the buildup in absorbance at 350 nm does not occur. Thus, the kinetics of the second phase can directly be measured. The largest absorbance change for 1a is observed at 320 nm, and this wavelength was chosen for collecting kinetic data. The pseudo-first-order rate constants kobs2 were evaluated at 55 °C as functions of HCl, LiCl, and NaClO4 concentrations. Pronounced acid catalysis emphasizes the second phase. Figure 10 shows the kobs2 versus [HCl] plot with and without 0.1 M NaClO4. There is a first-order dependence in HCl, and NaClO4 accelerates the reaction. The slopes of the linear plots equal (2.56 ( 0.09) â 10-2 and (1.47 ( 0.06) â 10-2 M-1 s-1 with and without NaClO4, respectively, the ratio being 1.74, in agreement with the 1H NMR data. Plots of kobs2 as a function of [LiCl] are shown in Figure 11. At [LiCl] < 0.06 M, the plots match those obtained for kobs1 (cf. data for 1a in Figure 4). Since the second phase does not occur at [HCl] ) 0 and kobs2 is independent of the concentration of added DMSO (0.0028-0.012 M), rate law at [LiCl] < 0.06
(39) This is important, since the spectrum of 2a in MeOH was found to be dependent on concentration of HCl (see Figures 4S and 5S of the Supporting Information).


5196 J. Am. Chem. Soc., Vol. 122, No. 21, 2000

AlexandroVa et al. form;42 (ii) separates the reacting fragments, ruling out an intramolecular mechanism; (iii) adjusts them for the subsequent redox transformation; and (iv) forces the reaction to proceed intermolecularly. Rate expression 10 indicates that a proton is included in the transition state and either DMSO (mechanism 1, MI) or complex 1Cl (mechanism 2, MII) can be protonated. The simplest stoichiometric mechanism of the reaction with a proton ambiguity is given by eqs 12-16. Here, eq 13 shows
MI MII Me2SdOH+ h or 6 h 1Cl + H+ Me2SdO + H+ 1 h 1Cl 1 + Cl- h 1Cl + Me2SdO 1Cl + Me2SdOH+ f or 6 + Me2SdOH f [PtIIIsO(H)sSMe2] [PtsSMe2sOH] [PtIIIsO(H)sSMe2] f 2 or [PtsSMe2sOH] f 2 K (12) (13) (14) (15) (16)

a

ks, k-s kCl, kk15 fast

Cl

Figure 12. Dependencies of log k10 (O) and log k10 (b) (eq 10) on E(p1) for the formation of 2 at 55 °C in methanol solvent.

M is given by eq 10. The corresponding numerical values k10

kobs2[H+]-1 ) (k10 + k10[LiCl])

(10)

and k10 for complexes 1a-e are placed in Table 3. The kinetic data in Figure 11 show, however, that at [LiCl] > 0.06 M, the kobs2 values level off as the concentration of LiCl increases, indicating that eq 11 may hold. The data in Figure 11 were

schematically substitution of DMSO by chloride via the solventdependent pathway; eq 15 is the rate-limiting step, the nature of which is specified below; and eq 16 is the fast step, leading to 2. Applying the steady-state approximation with respect to 1Cl and taking into account the mass balance equation [Pt]t ) [1] + [1Cl], eqs 12-16 lead to expression 17 for kobs2 in terms of MI. Here, [Cl-]t and [Me2SO]t denote total concentrations

kobs2 )

(ks + kCl[Cl-]t)k15Ka-1[Me2SO]t[H+] ks + k
-s

+ kCl[Cl-]t + (k

-Cl

+ k15[H+]/Ka)[Me2SO]

t

k

obs2

[H+]

-1

)

a + b[LiCl] c + [LiCl]

(17) (11)
of chloride and DMSO, respectively. Since in the concentration range studied the reaction is strictly first-order in [H+], eq 17 should be rewritten as eq 18 on the assumption that k-Cl . k15[H+]/Ka, which seems to be valid since the basicity of DMSO is low. The rate expression 18 is in agreement with the observed

fitted to eq 11, and the values of a, b, and c are summarized in Table 1S (Supporting Information). Moreover, rate retardation by LiCl is observed for electron-poor complexes 1d,e when the concentration of LiCl approaches 0.1 M. The effect of substituents on k10 and k10 is shown in Figure 12, where log k10 and log k10 are plotted against E(p1). The LiCl-dependent pathway is again more sensitive to the electronic effects and the slopes equal (1.50 ( 0.25) and (5.2 ( 0.8) V-1 for the k10 and k10 pathways, respectively. When log b was plotted against E(p1), the slope was 5.1 ( 0.9 V-1; i.e., there is a clear parallel between the kinetic parameters k10 and b. Discussion General Comments and Stoichiometric Mechanism. Spectral, electrochemical, and kinetic data obtained in this work emphasize a nonconcerted mechanism of reaction 4. The particular fragments of complex 1 which undergo transformation, viz., coordinated DMSO and the Pt(II) center, are insufficiently tuned for the redox reaction, and, therefore, several preliminary steps are required to adjust their reactivity. In particular, neutral Pt(II) complexes 1 must be converted into the easier-to-oxidize anionic species 1Cl in reaction 5, in which the orthoplatinated ligand plays a key role. It is well documented that DMSO is easily substituted by chloride in Pt(II) complexes when activated by an appropriate ligand located cis to it.40,41 It is thus the -bound phenyl ring of the orthoplatinated oxime that is a perfect cis-labilizing ligand. It is equilibrium 5 that (i) generates the reactive species, viz., 1Cl and unbound DMSO, which is easier to protonate in a free
(40) Lanza, S.; Minniti, D.; Romeo, R.; Tobe, M. Inorg. Chem. 1983, 22, 2006-2010. (41) Romeo, R.; Nastasi, N.; Scolaro, L. M.; Plutino, M. R.; Albinati, A.; Macchioini, A. Inorg. Chem. 1998, 37, 5460-5466.

k

obs2

)

(ks + kCl[Cl-]t)k15K ks + k
-s

-1 a

[Me2SO]t[H+]
t

+ kCl[Cl-]t + k-Cl[Me2SO]

(18)

dependencies of kobs2 on the concentrations of added chloride and DMSO. One arrives directly at eq 18 in terms of MII on the assumption that the concentration of 6 is much lower than those of 1 and 1Cl. As emphasized above, the reaction rate is essentially independent of [Me2SO]t, and in terms of eq 18 this suggests that at low chloride concentrations (ks + k-s + kCl[Cl-]t) , k-Cl[Me2SO]t, providing further simplification:

k

obs2

) (ks + kCl[Cl-]t)k15(k

-ClKa

)-1[H+]

(19)

Thus, the dependence of kobs2 on [Cl-]t at low LiCl concentrations should be linear with a positive intercept (eq 19), and the data in Figure 11 demonstrate that this is the case. However, when the relation (ks + k-s + kCl[Cl-]t) , k-Cl[Me2SO]t does not hold, kobs2 should level off as the chloride concentration becomes higher (eq 18). The experimental evidence for this is found in Figure 11 and eq 11. Thus, the stoichiometric mechanisms MI and MII are in accord with the experimental kinetic data. The rate law 10 corresponds to eq 19 derived from eqs 12-16. The rate constants k10 and k10 in eq 10 must be associated with ksk15/k-ClKa and kClk15/k-ClKa (eq 19), respectively.
(42) The reversible protonation of DMSO in MeOH is observed by UVvis spectroscopy at 202 nm. The estimated value of pKa is around 0. The data are shown in Figure 6S.


Biologically ReleVant Deoxygenation of DMSO Intimate Mechanistic Features: Substitution of DMSO in Complexes 1. The mechanism of substitution of DMSO in Pt(II) complexes by various ligands was intensively studied during past three decades (see ref 43 and the literature cited therein). Efforts were aimed at differentiation between associative and dissociative substitution mechanisms.18-20 In this work, the previous knowledge is used for assigning mechanistic steps prior to reduction of DMSO. The orthoplatinated ligand in 1 affects the rate law and contributions of the solvent- and chloride-driven pathways into the overall rate. In fact, the substitution of DMSO by chloride in complexes [Pt(en)(DMSO)2]2+ 40 and [PtMe(LL)(DMSO)]+ (LL ) bis-2-pyridylamine)41 in MeOH follows the first-order kinetics in chloride without intercept, showing the insignificance of the solvolitic pathway. In contrast, the latter dominates for complexes 1. The kCl pathway is unambiguous for electron-poor complexes and negligible for MeO- and Me-substituted molecules 1b,c. This trend is also seen in Figure 5, and the log kCl, not log ks, versus E(p1) plot has a meaningful positive slope. As mentioned above, different sensitivity of ks and kCl to electronic characteristics of 1 is due to different charges of methanol and chloride. Negatively charged chloride is obviously more sensitive to the redox potential of the target complex. The kinetic behavior of 1 resembles that of complexes trans-[Pt(aryl)Cl(SMe2)2], the rate of substitution at which is also independent of the concentration of such nucleophiles as Br- and N3-.44 The activation enthalpy of 24.5 kJ mol-1 for the ks pathway for 1a is too low to mirror the rate-limiting dissociative Pt-Cl bond cleavage, and the associative mechanism must be operative. In accord with this is the Sq of -81 J K-1 mol-1, typical of a bimolecular reaction with the entropy loss in the transition state. The same mechanism is valid for the kCl pathway, since the electron-deficient complexes 1 react faster. Thus, the mechanistic information on reaction 5 supports previous conclusions14,15 that -bound orthoplatinated ligands, although possessing strong ground-state trans influence and speeding up substitution and bridge splitting, do not bring about the associative f dissociative mechanistic changeover. Intimate Mechanistic Features: Deoxygenation of DMSO. The intriguing question here is why the reaction does not occur intramolecularly within 1, but rather intermolecular activation, at first glance less favorable, is preferred. The necessity of redox tuning of the Pt(II) species has already been emphasized. It should also be taken into account that, in addition to electron transfer, the reduction of DMSO involves SdO bond cleavage. The binding of Pt(II) with DMSO could, in principle, elongate the SdO bond and thus favor the reaction. However, thorough analysis of the structural data3 has led to the conclusion that the coordination of DMSO with Pt(II) via sulfur shortens the SdO bond in the ground state. The average bond distances in free and S-bound DMSO are 1.492 and 1.471 å, respectively. The bond length increases when DMSO is bound to Pt(II) via oxygen (1.540 å),3 but the largest elongation (1.585 å) occurs on protonation of the sulfoxide oxygen. Thus, DMSO dissociates from 1 to generate a species with a weaker sulfur-oxygen bond. It was not, therefore, surprising to observe the acid catalysis with plausibly protonated DMSO as a reactive species in terms of MI (Scheme 1). Further elongation of the SdO bond is feasible in the Pt-O-bonded intermediate 4, with the weakest sulfur-oxygen bond due to both protonation and O-coordination. The rate-limiting first electron transfer affords the Pt(III)
(43) Romeo, R.; Scolaro, L. M.; Nastasi, N.; Mann, B. E.; Bruno, G.; Nicolo, F. Inorg. Chem. 1996, 35, 7691-7698. (44) Wendt, O. F.; Oskarsson, å.; Leipoldt, J. G.; Elding, L.-I. Inorg. Chem. 1997, 36, 4514-4519.

J. Am. Chem. Soc., Vol. 122, No. 21, 2000 5197 Scheme 1

Scheme 2

intermediate 5 with a weakened sulfur-oxygen bond. The second electron transfer, to give Pt(IV), leads to S-O bond cleavage, and the oxygen dissociates as water in the presence of HCl. Five-coordinated species are not unusual and are well documented45-48 not only as intermediates.49 A squarepyramidal geometry of 4 as shown in Scheme 1 is one of the possibilities. The orthometalated C,N-chelate, which is a -acceptor due to the -bound aryl ring, and two chloro ligands are capable of creating a trigonal-bipyramidal structure, with two chlorides occupying axial positions.47 The kinetically indistinguishable mechanism 2 is shown is Scheme 2. It involves the protonation of the Pt(II) center to afford the platinum(II)-proton or platinum(IV)-hydride moiety followed by the rate-limiting insertion of the SdO double bond into the PtsH bond of intermediate 6. Hydride complexes of Pt(IV) are now widely invoked to rationalize mechanisms of catalytic reactions.12,50 Their existence has been confirmed previously51-54 and in this work by the example of protonation of complex 1c (anionic species 1Cl cannot be generated in the absence of DMSO, see above). A five-coordinate hydride Pt(IV) intermediate similar to 6 has recently been proved to exist in the elimination reaction of methane from [Pt(H)Me2(L)]+ (L ) bis(2-pyridylmethyl)amine).55 The rate-limiting step, which
(45) Wehman-Ooyevaar, I. C. M.; Grove, D. M.; de Vaal, P.; Dedieu, A.; van Koten, G. Inorg. Chem. 1992, 31, 5484-5493. (46) Fanizzi, F. P.; Intini, F. P.; Maresca, L.; Natile, G.; Lanfranchi, M.; Tiripicchio, A. J. Chem. Soc., Dalton Trans. 1991, 1007-1015. (47) Fanizzi, F. P.; Maresca, L.; Natile, G.; Lanfranchi, M.; Tiripicchio, A.; Pacchioni, G. J. Chem. Soc., Chem. Commun. 1992, 333-335. (48) Albano, V. G.; Natile, G.; Panunzi, A. Coord. Chem. ReV. 1994, 133, 67-114. (49) Otto, S. Structure and Reactivity Relationships in Platinum(II) and Rhodium(I) Complexes. Ph.D. Thesis, University of the Free State, Bloemfontein, South Africa, 1999. (50) Rendina, L. M.; Puddephatt, R. J. Chem. ReV. 1997, 97, 17351754.


5198 J. Am. Chem. Soc., Vol. 122, No. 21, 2000 affords 7, is analogous to insertion of carbon-carbon double bonds into the M-H bonds of transition metal complexes. Subsequent protonation of the S-hydroxyl generates a good leaving group, and the water departure accomplishes MII. To our knowledge, the insertion of the SdO bond into the metalhydride one has never been proposed to account for the mechanism of reaction of type 1. However, it seems to be due to the acid catalysis. Key steps of MII match those for related processes such as platinum-catalyzed hydrogenation and hydrosilylation of alkenes.56,57 Scheme 2 fits also the biphasic nature of the process, since the first step is needed to generate 1Cl and free DMSO. Anionic complex 1Cl must be more susceptible to protonation than its neutral precursor 1. The rate retardation at high LiCl concentrations is readily accountable in terms of both MI and MII, since the formation of 4 or 6 is disfavored when the fifth coordination site is occupied by the chloro ligand. Five-coordinated species of the type [PtI2(PR3)3]I generated by attack of halides at square-planar Pt(II) complexes have recently been isolated and characterized by a single-crystal X-ray study.49 Electron-poor complexes 1 react faster (Figure 12). This observation is controversial, since they are more difficult to oxidize. The discrepancy is also rationalizable in terms of MI and MII. In fact, several phenomena contribute to the overall rate, i.e., generation of 1Cl and its subsequent transformations into 4 and 5 (MI) or 6 and 7 (MII). The formation of 1Cl and the following steps should have opposite sensitivity to the electronic effects. The electron-deficient species have higher affinity to chloride, and their corresponding rate constants kCl differ by a factor of 6 in the series (Table 3). On the condition that this process contributes most into the electronic effects in formation of 2 (k10 ) kClk15[H+]/k-ClKa), the tendency observed is easy to understand. Theoretical Modeling. A Density Functional Study. Discrimination between MI and MII is difficult on the basis of exclusively kinetic data. Therefore, an attempt was made to probe the mechanisms by the quantum-chemical analysis using density functional theory (DFT), which proved to be useful for theoretical treatments of transition metal complexes. First, the orbital analysis has been performed with a goal of seeing any orbital advantage for the interaction between the elucidated reactive species, viz., 1Cl/DMSOH+ (MI) and 6/DMSO (MII). A LUMO/HOMO diagram for the species indicated is shown in Figure 13. It should first be mentioned that the DFT results in Figure 13 support the conclusions derived from the electrochemical data in Table 2. The HOMO of 1Cl is higher than that of 1, accounting for the easier oxidation of the anionic complexes. As far as the mechanistic discrimination is concerned, the data in Figure 13 suggest that MII involving 6 and free DMSO is better. Comparison of the 1Cl + DMSOH+ and 6 + DMSO pairs of reactants indicated that the gap between 1Cl HOMO and DMSOH+ LUMO (0.1918 au) is much bigger than the gap between DMSO HOMO and 6 LUMO (0.0930
(51) Hill, G. S.; Vittal, J. J.; Puddephatt, R. J. Organometallics 1997, 16, 1209-1217. (52) Jenkins, H. A.; Yap, G. P. A.; Puddephatt, R. J. Organometallics 1997, 16, 1946-1955. (53) Canty, A. J.; Fritsche, S. D.; Jin, H.; Patel, J.; Skelton, B. W.; White, A. H. Organometallics 1997, 16, 2175-2182. (54) Prokopchuk, E. M.; Jenkins, H. A.; Puddephatt, R. J. Organometallics 1999, 18, 2861-2866. (55) Fekl, U.; Zahl, A.; van Eldik, R. Organometallics 1999, 18, 41564164. (56) Kwiatek, J. In Hydrogenation and Dehydrogenation; Schrauzer, G. N., Ed.; Marcel Dekker: New York, 1971; pp 13-57. (57) Creve, S.; Oevering, H.; Coussens, B. B. Organometallics 1999, 18, 1967-1978.

AlexandroVa et al.

Figure 13. Calculated molecular orbital diagram of species involved in mechanisms MI and MII (au ) atomic units).

Figure 14. Calculated B3LYP energy profiles for MI and MII showing that MII requires less energy. Numbers show energy in kilojoules per mole.

au). However, there is inverse order of MO's in the two schemes. But this most likely does not play a crucial role, since it is already Pt(IV) in complex 6. Results of comparative analysis of energies of conceivable intermediates in MI and MII, viz., 1Cl/DMSOH+ and 6/DMSO, are presented in Figure 14. These are in accord with conclusions derived from the MO energies. The MII pathway demands much less energy for generation of the 6/DMSO pair of intermediates compared to 1Cl/DMSOH+ considering complex 1 and HCl as the initial state. The energy difference (236 kJ mol-1) is striking, and although the energies of intermediates, in contrast to the transition state energies, do not provide straightforward information about the activation barriers, one may state with confidence that the MII mechanism is definitely energetically preferred. This conclusion is somewhat surprising since it is still more customary to treat nonmetal atoms as better bases compared to transition metal ions. A question which deserves comment is how the system copes with the unfavorable thermodynamics mentioned in the Introduction. The DFT study suggests an energy gain in 260 kJ mol-1. It is presumably due to the favorable binding of Me2S to Pt(IV), since the sulfides are better ligands for Pt(II) and Pt(IV) compared to sulfoxides. Therefore, the formation of a new Pt-S bond makes the entire process 1 thermodynamically favorable. It should also be added that the Pt(II) f Pt(IV) oxidative transitions induced by mild oxidants have been documented. The most unexpected and striking is the oxidation by water of electron-rich Pt(II) complexes exemplified by the exciting works from the Canty group.12,53,58 It has recently been shown that even cis-[PtCl2(SOMe2)2] is oxidized by dioxygen under certain conditions.59


Biologically ReleVant Deoxygenation of DMSO Biomimetic Relevance of the Reaction. There is an analogy between the Pt(II)-promoted deoxygenation reported here and the enzymatic reaction catalyzed by the Mo-dependent DMSO reductase.60,61 The stoichiometry of the enzymatic process (eq 20) parallels eq 1. The enzymatic reduction is a multistep process

J. Am. Chem. Soc., Vol. 122, No. 21, 2000 5199
and then converted into corresponding oximes 4-RC6H4C(Me)dNOH by the action of hydroxylamine according to the classical procedure.63 Complexes 1 were prepared from the oximes and cis-[PtCl2(SOMe2)2] in refluxing methanol as described previously.11 Methanol of the highest purity (Khimed), used as a reaction medium, was utilized as received. Other reagents used in this work were of the highest quality available. Synthesis of Complex 1f. A 18.2 g (0.126 mol) portion of octanoic acid (Reakhim) was refluxed with SOCl2 (22.5 g, 0.189 mol) for 2 h, and then acid chloride formed was distilled in a vacuum (82 °C, 16 mmHg). Yield 85%. Subsequent Friedel-Crafts acylation of benzene (30 mL) in the presence of AlCl3 (6.7 g, 0.05 mol) by octanoic acid chloride (8.1 g, 0.05 mol) for 4 h at ambient temperature as described elsewhere64 afforded phenyl n-heptyl ketone (80%). The latter was converted into phenyl n-heptyl ketone oxime (85%) by reaction with hydroxylamine hydrochloride.63 1H NMR (CDCl3): 0.87 (t, 3H, CH3), 1.26 (m, 8H), 1.49 (m, 2H, NCH2CH2), 2.80 (t, 2H, NCH2), 7.347.41 (m, 3H, Ar), 7.57-7.63 (m, 2H, Ar), 8.62 (bs, 1H, OH). n-Heptyl ketone oxime (0.0508 g, 0.23 mmol) and cis-[PtCl2(SOMe2)2] (0.1 g, 0.23 mmol) in 20 mL of MeOH were refluxed for 12 h, and the solvent was then evaporated to dryness. The residue was separated by TLC using Silufol plates and benzene-ethyl acetate (15:1) as eluent. The first band was separated to yield 1f (35%). Anal. Calcd for C16H26ClNO2PtS: C, 37.5; H, 5.0. Found: C, 36.9; H, 5.1. 1H NMR (CDCl3): 0.89 (t, 3H, CH3), 1.28 (m, 8H), 1.65 (m, 2H, NCH2CH2), 2.79 (t, 2H, NCH2), 3.54 (s, 6H, JPtH ) 24 Hz, SCH3), 7.15 (m, 4H, Ar), 8.04 (dd, 1H, JPtH ) 52 Hz, H6), 10.11 (s, 1H, OH). Synthesis of Complexes 2 (by the Example of 2a). Complex 1a (0.0394 g, 0.089 mmol) was dissolved in 6 mL of methanol, and then 28 µL of concentrated HCl was introduced. The mixture was refluxed for 5 h and then cooled. Yellow crystals precipitated were separated, washed with cold methanol, and air-dried. Yield 67.2% (0.033 g). Anal. Calcd for C10H14Cl3NOPtS: C, 24.13; H, 2.84; Pt, 39.19. Found: C, 24.25; H, 2.64; Pt, 39.60. Exactly the same procedure was employed for the preparation of other complexes 2; yields and their characteristics are summarized in Table 4. Satisfactory analytical data (C, H, N, and Cl) were obtained for complexes 1b-e. Synthesis of Complex trans-(N,N)-[PtII(C6H4-2-CMedNOH)Cl(py)] (3). Compound 3 was obtained by refluxing 1a (0.0344 g, 0.077 mmol) with pyridine (12 µL, 0.15 mmol) in 6 mL of MeOH for 1.5 h. Dark-yellow crystals that precipitated after cooling of the solution were filtered off, washed with cold methanol, and air-dried. Yield 87%. The 1H NMR spectrum of the material was identical with that of complex 3 which was prepared using another procedure and characterized by an X-ray structural study.15 Kinetic Measurements. Stock solutions of complexes 1 (ca. 1.3 â 10-3 M) were prepared in methanol solvent. UV-vis measurements indicated that the spectra of the solutions remained unchanged over several weeks. An aliquot of such a solution (ca. 0.4 mL) was introduced to a 1-cm quartz cuvette, followed by addition of the required amount of LiCl, HCl, or NaClO4 dissolved in MeOH. The total concentration of 1 in the cell was usually around 2.5 â 10-4 M; concentrations of the inorganic additives used are indicated in the legends to the figures throughout the paper. Different wavelengths were chosen for monitoring the first and the second phases of the deoxygenation reaction. The substitution of DMSO by chloride was usually monitored at 353 nm. Since the spectrum of the methoxy-substituted complex 1b was somewhat different compared with those of other complexes, the wavelength of 327 nm was selected. The kinetics of the second phase was monitored by following a decrease in absorbance of a longwavelength maximum of the corresponding compound (Table 1). Good first-order kinetic curves were usually obtained for the first and the second phases, and they were quantified by nonlinear least-squares fitting of the absorbance versus time plots to the equation A ) A + (Ao - A){exp(-kobst)}, where A, A, and Ao are absorbances at time t, , and t ) 0, respectively. Satisfactory first-order dependencies were normally observed in a matter of 4-5 half-lives. All calculations were performed using a SigmaPlot 4.0 package.
(63) Beckman, E. Chem. Ber. 1890, 23, 1680-1692. (64) Organikum. Organisch-Chemisches Grundpraktikum; VEB Deutscher Verlag der Wissenschaften: Berlin, 1976.

enzymesMo(IV) + Me2SdO + 2H+ f enzymesMo(VI) + Me2S + H2O (20)
involving two, one-electron steps. At pH 8.5, two oxidations for Rhodobacter sphaeroides DMSO reductase observed at +37 and +83 mV (versus NHE) are associated with the Mo(IV/V) and Mo(V/VI) redox transitions, respectively,61 which, of course, occur more cathodically compared to the Pt(II) mimetic reported here. According to the present state of knowledge, DMSO is coordinated with Mo via oxygen in the key enzymatic intermediate. A proton is also believed to play an important role in the enzymatic system.60 In general, the mechanism of the bioreaction matches better the MI mechanism which seems, however, less probable than MII in light of the DFT study. Thus, there is a similar pathway in the unnatural Pt(II) system, although the soft sulfur center of DMSO is more useful for binding with a soft Pt(II) center compared to hard oxygen. The acid catalysis is a key feature in the platinum(II)-promoted reaction. These analogies make it possible to view the platinum system reported in this work as a true functioning mimetic of the Mo-dependent dimethyl sulfoxide reductase. Conclusion. The system described represents an interesting mechanistic example in which the intuitively intramolecular conversion of 1 into 2 does not occur as such because the participating fragments of the starting complex 1 are insufficiently tuned to undergo deoxygenation of DMSO coupled with the oxidation of Pt(II) into Pt(IV). The system adopts, at first glance, a less advantageous intermolecular pathway which makes it possible to adjust independently the reactivities of the two participating species. A novel mechanism of deoxygenation is proposed involving the insertion of the sulfoxide SdO bond into the platinum-hydride moiety formed on protonation of the anionic intermediate 1Cl. Experimental Section
Methods. Spectrophotometric and kinetic measurements were carried out on a Shimadzu UV-160A spectrophotometer equipped with a CPS240A cell positioner/temperature controller or a Hitachi 150-20 instrument with a thermostated by circulating water cell compartment. 1H NMR spectra were recorded on CXP-200 Bruker and Varian UNITY 300 instruments with a residual solvent signal as internal standard. Electrochemical measurements were made in a three-electrode cell with a pyrolitic graphite working electrode at 25 and 55 °C. A potential sweep was generated with an IPC-3 instrument interfaced with a personal computer. The working electrode was polished with a diamond paste before every new measurement. Unless otherwise indicated, potentials are versus saturated calomel electrode (SCE) throughout. Reagents. cis-[PtCl2(SOMe2)2] was obtained according to the procedure of Price et al.62 Ring-substituted acetophenones 4-RC6H4COMe (R ) H, MeO, Me, F, and Cl) were purchased from Lancaster
(58) Canty, A. J.; Honeyman, R. T.; Roberts, A. S.; Traill, P. R.; Colton, R.; Skelton, B. W.; White, A. H. J. Organomet. Chem. 1994, 471, C8C10. (59) Davies, M. S.; Hambley, T. W. Inorg. Chem. 1998, 37, 5408. (60) Baugh, P. E.; Garner, C. D.; Charnock, J. M.; Collison, D.; Davies, E. S.; McAlpine, A. S.; Bailey, S.; Lane, I.; Hanson, G. R.; McEwan, A. G. J. Biol. Inorg. Chem. 1997, 2, 634-643. (61) George, G. N.; Hilton, J.; Temple, C.; Prince, R. C.; Rajagopalan, K. V. J. Am. Chem. Soc. 1999, 121, 1256-1266. (62) Price, J. H.; Williamson, A. N.; Schramm, R. F.; Wayland, B. B. Inorg. Chem. 1972, 11, 1280-1284.


5200 J. Am. Chem. Soc., Vol. 122, No. 21, 2000
Quantum-Chemical Calculations. These were performed using Gaussian 9465 at the B3LYP level of theory. The basis used for carbon, nitrogen, oxygen, and hydrogen atoms was 6-31G with the polarization functions (6-31G**). For platinum, sulfur, and chlorine the basis used was LANL2DZ66-68 with pseudopotential. A complete list of calculation results can be found in Table 4S of the Supporting Information. Crystallography. Data Collection and Structure Determination of 2a. A yellow parallelepiped-shaped crystal was glued on top of a cactus needle and transferred to an Enraf-Nonius CAD4 diffractometer for data collection (graphite-monochromatized Mo KR radiation). Unit cell dimensions were determined from a least-squares treatment of the setting angles of 24 reflections with 17 < < 19. The X-ray singlecrystal data were corrected for Lorentz, polarization, and absorption effects (empirical correction based on -scans). The atomic scattering factors were taken from the International Tables for X-ray Crystallography.69 All calculations were performed with the CAD4-SDP program package. The Pt, S, and Cl atoms were found by direct methods, and the remaining non-hydrogen atoms were located from the subsequent Fourier synthesis. Refinement on F was carried out by full-matrix least-squares techniques. All non-hydrogen atoms were refined with anisotropic thermal parameters. Hydrogen atoms were not included in the refinement. Additional details of structure determination are given in Table 6.

AlexandroVa et al.
Table 6. Crystal Data and Details of the Structure Determination for 2a Crystal Data empirical formula C10H14Cl3NOPtS formula weight (g mol-1) 497.726 crystal system P1 triclinic h a (å) 7.1808(7) b (å) 8.317(1) c (å) 12.725(2) R (deg) 93.87(1) (deg) 98.78(1) (deg) 105.75(1) V (å3) 718.1(2) Z 2 D(calcd) (g cm-3) 2.301 F(000) 497.74 µ(Mo KR) (cm-1) 105.538 crystal size (mm) 0.06 â 0.08 â 0.20 Data Collection temperature (K) 293 radiation (å) 0.71069 (Mo KR) min, max (deg) 2, 30 scan (deg) 0.90 + 0.35 tan data set 0/10, -11/ 11, -17/17 total no. of unique data 3842 no. of obsd data [I > 3(I)] 3733 Refinement Nref, R,w w max min, N R
par

Acknowledgment. The research described in this publication was made possible in part by financial support from the Russian Foundation for Basic Research (98-03-33023a) and INTAS (970166). We thank Irina Panyashkina and Valentina Ryzhova for experimental assistance. Appendix According to eq 5, the equilibrium constant Kx is given by eq 1A.

shift/error max resd dens (e/å3)

3733, 155 0.0355, 0.0389 1/2(F) 0.00 -2.16, 1.70

Kx )

[1Cl]

2

Kx )

[1Cl][DMSO] [1][Cl-]

([Pt]t - [1Cl])[Cl-]

(3A)

(1A)

The absorbance is a sum of two terms:

Since [Cl-] > [Pt]t, the solution of the quadratic equation makes it possible to calculate the equilibrium concentration of 1Cl (eq 4A).

A)

1[

1] +

Cl[1Cl

]

(2A)

[1Cl] )-

Kx[Cl-] + 2

0.25Kx2[Cl-]2 + [Pt]tKx[Cl-] (4A)

Using the mass balance equation ([Pt]t ) [1] + [1Cl]) and since [1Cl] ) [DMSO], one arrives at eq 3A:
(65) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.; Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T. A.; Petersson, G. A.; Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski, V. G.; Ortiz, J. V.; Foresman, J. B.; Cioslowski, J.; Stefanov, B. B.; Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen, W.; Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. P.; HeadGordon, M.; Gonzalez, C.; Pople, J. A. Gaussian 94 (Revision D.4); Gaussian Inc.: Pittsburgh, PA, 1995. (66) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270-283. (67) Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284-298. (68) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299-310. (69) International Tables for X-Ray Crystallography; Kynoch Press: Birmingham, 1974; Vol. IV.

Substitution of the expression for [1Cl] into eq 2A gives the final eq 9, where Ao ) 1[1] and ) Cl - 1. Supporting Information Available: Tables of crystal data, positional parameters, general displacement expressions, results of quantum-chemical calculations, calculated kinetic parameters of eq 11, UV-vis spectra of 2a recorded under various conditions, dependence of kobs1 against [LiCl] for 1b-e, the Eyring plot for ks, absorbance change vs [LiC] for eq 5, and absorbance change of DMSO in the presence of HCl (PDF). This material is available free of charge via the Internet at http://pubs.acs.org.
JA994342K