Detection, Isolation, and Characterization of Intermediates in Oxygen Atom Transfer Reactions in Molybdoenzyme Model Systems Paul D. Smith, 1a Andrew J. Millar, 1a Charles G. Young,* ,1a Amit Ghosh, 1b and Partha Basu* ,1c School of Chemistry, UniVersity of Melbourne Victoria 3010, Australia Chemical Group Technical Center, PPG Industries Inc. Pittsburgh, PennsylVania 15146 Department of Chemistry and Biochemistry Duquesne UniVersity, Pittsburgh, PennsylVania 15282 ReceiVed March 8, 2000 The pterin-containing molybdoenzymes catalyze the net ex- change of an oxygen atom between water and substrate and there is evidence to support the involvement of oxygen atom transfer (OAT or oxo transfer) in the reactions of dimethyl sulfoxide reductases (DMSOR), sulfite oxidase, and nitrate reductase. 2-7 Accordingly, many model studies have focused on OAT reactions, most notably the reduction of dimethyl sulfoxide by oxo-Mo- (IV) complexes and the oxidation of tertiary phosphines by dioxo-Mo(VI) complexes. 3 Indeed, Schultz et al. 4 have shown that the DMSOR from Rhodobacter sphaeroides couples these reactions during catalysis of OAT from Me 2 SO to the water- soluble tertiary phosphine 1,3,5-triaza-7-phosphatricyclo[3.3.1.1 3,7 ]- decane. As well, the crystal structure of Me 2 S-soaked crystals of oxidized DMSOR from R. capsulatus has revealed the presence of an Mo-bound Me 2 SO molecule formed upon incomplete OAT from Mo to Me 2 S. 5 Putative oxo(phosphine oxide) intermediates formed during the oxidation of phosphines by enzyme or model systems have never been detected or isolated. The reactions of LMo VI O 2 X (L ) hydrotris(3,5-dimethylpyra- zol-1-yl)borate) with PPh 3 are second-order and produce OPPh 3 and “LMo IV OX”, which may be trapped, e.g., as LMo IV OX- (solvent) or LMo IV OX (as monodentate X becomes bidentate). 8-10 An associative mechanism has been proposed for the overall OAT reaction. 9 However, it is not clear whether the intermediate, LMoOX(OPPh 3 ), gives the product by an associative or dissocia- tive mechanism. The reaction of LMoO 2 (SPh) with phosphines led Hall and co-workers 6 to examine the reaction of MoO 2 (NH 3 ) 2 - (SH) 2 with PMe 3 by computational methods. In the first step of the reaction, nucleophilic attack of PMe 3 on a π* ModO orbital perpendicular to the MoO 2 unit and at an Mo-O‚‚‚P angle of ca. 130° takes place. This results in a transition state with a weakened Mo-O bond (1.83 Å), an O-P interaction (2.43 Å), and a P-O- ModO torsion angle of 89.7°; the remaining ModO bond becomes stronger consistent with a “spectator oxo” function. 11 The OPMe 3 ligand then rotates about the Mo-O bond, breaking the Mo-O π interaction to generate an intermediate with Mod O ) 1.67 Å, Mo-O ) 2.18 Å, O-P ) 1.53 Å, and P-O-Mod O torsion ) 0.5°. The intermediate was 68.9 kcal‚mol -1 lower in energy than the reactants. At this stage replacement of OPMe 3 by water is predicted to take place by an associatiVe mechanism. Here, we report the detection of oxo(phosphine oxide) inter- mediates in the OAT reactions of LMoO 2 X and PPh 3 by fast atom bombardment mass spectrometry (FABMS) and the use of this technique to assess the stability of the intermediates and examine the kinetics of decay for unstable species. 12 Also, we report the isolation and characterization of L Pr Mo IV O(OPh)(OPEt 3 ) (L Pr ) hydrotris(3-isopropylpyrazol-1-yl)borate), the first stable oxo- (phosphine oxide) complex to be synthesized by incomplete OAT in a molybdoenzyme model system. Intermediates in the reactions of LMoO 2 X (X ) Cl - , Br - , OPh - , and SPh - ) 8,13 with PPh 3 were detected by FABMS. 14 In each case, a peak cluster indicative of the initial formation of LMoOX(OPPh 3 ) was observed. Figure 1 shows the parent ion of the intermediate formed in the reaction of LMoO 2 Cl and PPh 3 ; the base peak at m/z 724 and isotope pattern match those expected for [LMoOCl(OPPh 3 )] + ([M] + ). Related species were detected for X ) Br - ([M] + ) and X ) SPh - and OPh - ([M + H] + ) complexes. In blanks containing no added PPh 3 only a strong parent ion due to [LMoO 2 X] + was detected. The [M] + peak intensity of the intermediate LMoOCl(OPPh 3 ) monitored as a function of time showed an exponential decay with a first-order constant of 0.038 s -1 (Figure 1). Under similar experimental conditions, [LMoO(SPh)(OPPh 3 ) + H] + decayed with a rate constant of 0.0097 s -1 while no decay of [LMoO(OPh)(OPPh 3 ) + H] + was observed. These observations are consistent with the (1) (a) University of Melbourne. E-mail: c.g.young@chemistry. unimelb.edu.au. (b) PPG Industries Inc. (c) Duquesne University. E-mail: basu@duq.edu. (2) (a) Enemark, J. H.; Young, C. G. AdV. Inorg. Chem. 1993, 40,1-88. (b) Hille, R. Chem. ReV. 1996, 96, 2757-2816. (c) Pilato, R. S.; Stiefel, E. I. In Inorganic Catalysis, 2nd ed.; Reedijk, J., Bouwman, E., Eds.; Marcel Dekker: New York, 1999; pp 81-152. (3) (a) Holm, R. H. Coord. Chem. ReV. 1990, 100, 183-221. (b) Young, C. G. In Biomimetic Oxidations Catalyzed by Transition Metal Complexes; Meunier, B., Ed., Imperial College Press: London, 2000; Chapter 9, pp 415- 459. (4) Schultz, B. E.; Hille, R.; Holm, R. H. J. Am. Chem. Soc. 1995, 117, 827-828. (5) McAlpine, A. S.; McEwan, A. G.; Bailey, S. J. Mol. Biol. 1998, 275, 613-623. (6) (a) Pietsch, M. A.; Couty, M.; Hall, M. B. J. Phys. Chem. 1995, 99, 16315-16319. (b) Pietsch, M. A.; Hall, M. B. Inorg. Chem. 1996, 35, 1273- 1278. (c) Zaric, S.; Hall, M. B. In Molecular Modeling and Dynamics of Bioinorganic Systems; Banci, L., Comba, P., Eds.; Kluwer: Dordrecht, 1997; pp 255-277. Parameters quoted in the text vary with the level of theory. (7) Thapper, A.; Deeth, R. J.; Nordlander, E. Inorg. Chem. 1999, 38, 1015- 1018. (8) Roberts, S. A.; Young, C. G.; Kipke, C. A.; Cleland, W. E., Jr.; Yamanouchi, K.; Carducci, M. D.; Enemark, J. H. Inorg. Chem. 1990, 29, 3650-3656. (9) (a) Roberts, S. A.; Young, C. G.; Cleland, W. E., Jr.; Ortega, R. B.; Enemark, J. H. Inorg. Chem. 1988, 27, 3044-3051. (b) Laughlin, L. J.; Young, C. G. Inorg. Chem. 1996, 35, 1050-1058. (10) (a) Xiao, Z.; Young, C. G.; Enemark, J. H.; Wedd, A. G. J. Am. Chem. Soc. 1992, 114, 9194-9195. (b) Xiao, Z.; Bruck, M. A.; Enemark, J. H.; Young, C. G.; Wedd, A. G. Inorg. Chem. 1996, 35, 7508-7515. (11) (a) Rappe ´, A. K.; Goddard, W. A., III. J. Am. Chem. Soc. 1982, 104, 448-456. (b) Rappe ´, A. K.; Goddard, W. A., III. J. Am. Chem. Soc. 1982, 104, 3287-3294. (12) Caprioli, R. M. ACS Symp. Ser. 1985, 291, 209-216. (13) Xiao, Z.; Bruck, M. A.; Doyle, C.; Enemark, J. H.; Grittini, C.; Gable, R. W.; Wedd, A. G.; Young, C. G. Inorg. Chem. 1995, 34, 5950-5962. (14) In a typical experiment, a dichloromethane solution of the complex (1-2 μM) was mixed with an excess (>10 equiv) of PPh3 directly on an FAB probe containing ca. 1 μL of m-nitrobenzoic acid (mNBA). The final volume of the reaction mixtures was less that 3 μL. Once mixed, the probe was quickly placed inside a Micromass Autospec-EQ spectrometer (OPUS operating system) and bombarded with a CsI gun; the average time for this operation was ca. 30-45 s. The probe was maintained at a constant temperature of ca. 65 °C. Positive ion FABMS were collected. Figure 1. Left panel: The FABMS parent ion of LMoOCl(OPPh3), formed in the reaction of LMoO2Cl and PPh3 in mNBA. The experimental and calculated spectra are plotted using hatched and solid bars, respec- tively. Right panel: A plot of the normalized intensity of the m/z 724 peak of LMoOCl(OPPh3) vs time. 9298 J. Am. Chem. Soc. 2000, 122, 9298-9299 10.1021/ja0008362 CCC: $19.00 © 2000 American Chemical Society Published on Web 09/13/2000