A theoretical model of the ethane/deuterium exchange reaction catalysed by platinum. The nature of the αβ process

Abstract

The bimodal initial product distributions in ethane exchange with deuterium observed using several transition-metal catalysts were quantitatively reproduced by a theoretical model in the classical work of Anderson and Kemball in 1954. They showed that the only mechanistic process required is the interconversion of adsorbed ethyl and 1,2-diadsorbed ethene (the αβ process) but two independent types of site seemed necessary. This important problem has now been re-examined in the light of many developments since the early fifties in organometallic chemistry, heterogeneous catalysis and surface science. Sets of steady-state equations are derived based on the original mathematical model, but much more sophisticated descriptions of the αβ process are employed using the following basic assumptions. A small concentration of defects of a fundamental type is responsible for alkane activation but only generates a limited αβ process. The main αβ process is now described as interconversion of σ-bonded alkyl and π-bonded alkene and is associated with numerous sites of a different kind which are populated by transfer of intermediates from the first type of site rather than by direct alkane activation. Using these postulates the U-shaped distributions reported in the literature including that for a Pt single crystal (111 face) are quantitatively modelled. The general philosophical importance of these basic concepts for an understanding of alkane reactions on transition-metal catalysts is discussed.