Carboxylic Solvents and O-Donor Ligand Effects on CH Activation by Pt(II) Vadim R. Ziatdinov, ² Jonas Oxgaard, ‡ Oleg A. Mironov, ² Kenneth J. H. Young, ² William A. Goddard III,* ,‡ and Roy A. Periana* ,² Department of Chemistry, Loker Hydrocarbon Research Institute, UniVersity of Southern California, Los Angeles, California 90089, and Department of Chemistry, Beckman Institute, California Institute of Technology, Pasadena, California 91125 Received February 9, 2006; E-mail: wag@wag.caltech.edu Organometallic coordination catalysts based on the C-H activa- tion reaction currently show the greatest potential for the develop- ment of new, selective, hydrocarbon hydroxylation chemistry. 1 Previously, we reported on a platinum catalyst, Pt(bpym)Cl 2 (bpym ) η 2 -{2,2′-bipyrimidyl}), that converts methane to methanol in concentrated sulfuric acid with a 70% one-pass yield. 2g The practicality of this system is limited as the average catalyst rate, TOF ∼ 10 -3 s -1 , is below the commercially viable target of ∼1 s -1 . This is due to severe inhibition of the CH activation step (which is the rate-determining step for oxidation in the presence of 1 M water or methanol) by the reaction products, resulting in a maximum methanol concentration to ∼1 M rather than the ∼5 M required for efficient separation. On the basis of our understanding of the mechanism of the Pt- (bpym)Cl 2 /H 2 SO 4 system, 1h,2 we identified two key steps that contribute to the overall CH activation barrier of ∼32 kcal/mol: (A) ∼27 kcal/mol for coordination of the hydrocarbon (RH) and (B) ∼5 kcal/mol for cleavage of the coordinated CH bond. In designing improved catalysts, it is important that both steps be explicitly considered to ensure that catalyst modifications that decrease the energy requirements for one step do not proportionately increase the other. This should be possible since the bonding in the ground state and cleavage transition state should be different. 1h We now report that using an interplay between computational and experimental methods an efficient system has been designed which catalyzes H/D exchange magnitudes faster than the Pt(bpym)Cl 2 system. This improvement results from a reduction in the energetics for coordination without a corresponding increase in the energetics for cleavage. We postulated that replacing the neutral bipyrimidine in Pt- (bpym)(X) 2 with a monoanionic ligand, such as η 2 -N,O-picolinate (pic), would yield a complex which would have reduced energetics for RH coordination due to the increased electron density at the metal center. Furthermore, calculations predicted that use of bidentate TFA ligands in lieu of Cl ligands could facilitate CH cleavage via a six-membered cyclic transition state that is not proportionately higher in energy relative to that for the Pt(bpym)- (X) 2 system, resulting in a predicted lower overall barrier for CH activation for the K[Pt(pic)(TFA) 2 ] system. Similar six-membered, cyclic transition states have previously been postulated by Ryabov 3 and in a recent computational study for cyclometalation of dimethylbenzylamine. 4 All calculations utilized the Jaguar 6.5 suite, using B3LYP/ LACVP**++ with ZPE and solvent corrections for the TFA medium (Poisson-Boltzmann continuum solvent, with ǫ ) 8.42 and radius probe ) 2.479). 5,6 Diffuse functions were not used for solvents. Furthermore, due to inaccuracies of free energy calcula- tions in solvents, no ∆G results have been included. The picolinate ligand was chosen because it is a commercially available monoanionic, chelating N- and O-donor ligand that is stable to protic, thermal, and oxidizing conditions. We have recently shown that O-donor ligands can facilitate the C-H activation reaction as well as selective functionalization of the M-R intermediate, 7 and computational screening indicated that C/H activation should be facile using this ligand. Solutions of 1 were prepared by treatment of the known K[Pt- (pic)Cl 2 ] with silver acetate in TFA (Scheme 1) and has been characterized by 1 H and 13 C NMR, as well as high-resolution electrospray ionization mass spectrometry. Due to the anionic character of 1, we were unable to isolate it as a pure material without contamination from AgOAc, either by recrystallization or extraction. However, the neutral derivative, Pt(pic)(Me)(Et 2 S), has been prepared and fully characterized by 1 H and 13 C NMR as well as elemental analysis. Furthermore, in situ NMR confirms that all acetate is present as free HOAc in the TFA solution. In situ NMR studies of 1 in TFA also showed that 1 was thermally stable in CF 3 CO 2 H at 70 °C for several days in air. At 100 °C, a slow reaction occurs to quantitatively generate Pt(pic) 2 and a black precipitate (believed to be Pt black), with a t 1/2 of ∼10 days. As reactions with methane were found to require temperatures above 100 °C, the CH activation studies were conducted with benzene where reaction could be readily observed at 70 °C. Upon heating a mixture of benzene (0.1 mL) and CF 3 CO 2 D (1 mL) containing 4 mM of 1 to 70 °C, catalytic incorporation of deuterium into benzene was observed. Analysis by GC/MS shows 43 turnovers after 1 h (TOF of ∼1.2 × 10 -2 s -1 ). On the basis of the temperature dependence, a ∆G q of ∼23 kcal/mol was estimated. Control experiments without catalyst did not lead to any observable H/D exchange (<0.5%) under the same reaction conditions. To compare the catalytic activity of 1 to the Pt(bpym) system, Pt(bpym)TFA 2 was prepared and characterized by elemental analysis, high- resolution mass spectroscopy, and 1 H and 13 C NMR. H/D exchange with Pt(bpym)TFA 2 shows that this complex is 300 times less active (TON of 4.6 after 20 h, TOF ∼ 6.4 × 10 -5 s -1 ) than 1 under similar conditions. Unfortunately, direct comparison with (bpym)PtCl 2 is not possible as it is not soluble in TFA. Our theoretical calculations of CH activation with these systems (Figure 1) are consistent with the experimental results. According ² University of Southern California. ‡ California Institute of Technology. Scheme 1 Published on Web 05/17/2006 7404 9 J. AM. CHEM. SOC. 2006, 128, 7404-7405 10.1021/ja060973k CCC: $33.50 © 2006 American Chemical Society