Catalysis Letters 65 (2000) 61–66 61 Asymmetric hydrogenation of methyl pyruvate using platinum carbonyl cluster supported on an anion exchanger as the catalyst Sumit Bhaduri a, , Goutam Kumar Lahiri b, , Pradip Munshi b and Doble Mukesh c a Reliance Industries Limited, Maker Chambers IV, 222, Nariman Point, Mumbai 400021, India E-mail: sumit bhaduri@ril.com b Department of Chemistry, Indian Institute of Technology, Powai, Mumbai 400076, India E-mail: lahiri@ether.chem.iitb.ernet.in c ICI India R&T Centre, PO Box 155, Thane 400601, India Received 21 September 1999; accepted 14 January 2000 Anionic platinum carbonyl clusters supported on quaternary amine functionalized cross-linked polystyrene, i.e., anion exchangers, are effective precatalysts for the hydrogenation of methyl pyruvate to methyl lactate. Kinetic data has been obtained for anion exchangers with different quaternary groups. Highest observed rate constant and enantioselectivity are obtained with cinchonine funtionalized resin. The kinetic data also indicates saturation kinetics. The decarbonylated used catalyst could be recarbonylated to give a material spectroscopically (infrared) equivalent to freshly anchored cluster. Keywords: asymmetric, polymer-supported, heterogeneous, hydrogenation, catalysis 1. Introduction Hydrogenation of methyl pyruvate ester, reaction (1), with cinchona-alkaloid-modified platinum catalysts is known to give methyl lactate with high enantioselectivity [1–3]: MeCOCO 2 Me + H 2 MeCH(OH)CO 2 Me (1) This reaction is called the Orito reaction after the name of its inventor, and in recent years has been studied in consid- erable detail [1,3–5]. We had shown that anionic carbonyl clusters could be supported on cross-linked polystyrene resin functionalized with cinchona alkaloids [6]. After de- carbonylation supported platinum carbonyl clusters were bound to be effective catalysts for the Orito reaction. Here we report kinetic and other studies on these catalysts to ad- dress two questions. First, whether on the basis of an em- pirically derived rate law an insight into the mechanism of the reaction could be obtained. Second, could a trend in the kinetic data be observed on variation of the cation, i.e., the ligand environment, and if so, could this trend be correlated to the mechanism of enantioselection? Finally, we also wanted to see if on the basis of kinetic and IR studies some light could be thrown on the nature of the active catalysts. 2. Experimental 2.1. Physical measurements and materials Chloroplatinic acid was purchased from Johnson Mathey, London. All the amines and the chloromethylated divinyl To whom correspondence should be addressed. benzene cross-linked (20%) polystyrene were purchased from Fluka, USA. Infrared spectra were recorded on a Nico- let FT spectrophotometer. Atomic absorption measurements were carried out on an IL 751 spectrophotometer. Microanalyses were carried out with a Carlo–Erba 1106 model C, H, N analyzer. The anionic carbonyl cluster [Pt 15 (CO) 30 ] 2 and the amine 3,4- dihydroxy-N-benzylpyrrolidine were synthesized according to literature reported procedures [7,8]. Hydrogenation ex- periments were carried out in a Parr autoclave, USA. Con- version of methyl pyruvate to methyl lactate was monitored by gas chromatography (GC) using a SC-30 (mesh, 1000– 2000) GC column and FID detector. Optical rotations were measured by using a JASCO polarimeter, model CT-10, Japan. All the solvents and organic compounds were thor- oughly dried and distilled under nitrogen before use. For the simulation studies, the differential equations were numerically integrated using fourth-order Runge– Kutta method [9]. The rate constants for different models were estimated by fitting the simulation results with exper- imental kinetic data. The fitting was done by adjusting the rate constants so that the sum of squares of the differences between the model predictions and the experimental data was minimum. 2.2. Functionalization of the polymer A mixture of chloromethylated 20% DVB cross-linked polystyrene beads (1.0 g) and cinchona alkaloid (1.0 g) was stirred in a mixture of toluene (40 ml) and ethanol (40 ml) at 100 C for 24 h. The polymer beads were filtered off, and washed successively with ethanol, chlo- roform, dichloromethane and ether and then dried under J.C. Baltzer AG, Science Publishers