Asymmetric hydrogenation catalysed by platinum and iridium
The enantioselective hydrogenation of α-ketoesters over supported platinum metal catalysts which have been modified by chiral molecules has been studied. The aim of this thesis was (a) to gain a greater understanding of the kinetics of asymmetric hydrogenation of methyl pyruvate (MeCOCOOMe) to methyl lactate (MeCH(OH)COOMe) over cinchonidine modified EUROPT-1, (b) to diversify the reaction, by variation of the metal, modifier and reactant and (c) to test and develop a mechanism to explain the reaction by both mathematical and molecular modelling. It was found that only very small quantities of cinchonidine modifier were required to render EUROPT-1, a 6.3 % Pt/silica catalyst, suitable for asymmetric hydrogenation of methyl pyruvate. The plots of the variation of with reaction rate and of optical yield with modifier loadings reached plateau values at a loading of ca. 0.8 mg cinchonidine per 100 mg EUROPT-1. The shape of the curves has been fitted to a mathematical model which gives evidence that the mechanism is a result of a 1: 1 interaction between modifier and reactant. Modification of catalyst by mixtures of cinchonidine and cinchonine, or of quinine and quinidine has been performed and results obtained as to the effect on reaction rate and optical yield. The variation of the latter with the mole fraction of modifier has been fitted to a mathematical model which supports the view that the mechanism is a result of a 1: 1 interaction. It has been discovered that Iridium, especially when supported on calcium carbonate, is an effective catalyst for both racemic and enantioselective hydrogenation of methyl pyruvate. When Ir/CaCO³ was reduced and modified by cinchonidine, hydrogenation of methyl pyruvate at 273 K and 10 bar in ethanol resulted in an optical yield of 39 %. The reaction is approximately zero order in reactant and first order in hydrogen, being first order overall. The apparent activation energy was 22 ±2 kJ mol-1. The unmodified reaction proceeded at a very fast rate with an apparent activation energy of 11 ±2 kJ mol- 1. The enhanced rate was explained in terms of stabilisation of the halfhydrogenated state by the carbonate support. Rhodium, Osmium and Rhenium were found not to be effective for asymmetric hydrogenation, which in the case of rhodium is contrary to the literature. Diversification of the modifier was found to be possible, the best alternative being by the use of 10,1 1-dihydroquinine 4-methyl-2-quinoyl ether which gave an optical yield of 22 %. The amino acids tryptophan and histidine were found to give poor but reproducible enantiomeric excesses. Ephedrines and benzyl pyrrolidine methanol were found not to be effective modifiers which is contrary to the literature. A molecular modelling study resulted in a new proposal for the asymmetric hydrogenation site which explained the mechanism in terms of a 1:1 interaction between modifier and reactant at the catalyst surface. The proposal successfully accounts for the observed enantioselectivity in terms of steric hindrance to the unwanted product. The proposed mechanism is in agreement with experimental data in the literature.