Mechanism and kinetics of the catalytic CO-H2 reaction:: An approach by chemical transients and surface relaxation spectroscopy

The aim of this paper is to demonstrate the importance of providing time-resolved information in catalysis research. Two truly in situ methods will be presented and compared for their merits and drawbacks: chemical transient kinetics (CTK) and pulsed field desorption mass spectrometry (PFDMS). The presentation will be given by way of example choosing the syngas (CO/H-2) reaction over cobalt-based catalysts as a catalytic process. Despite numerous efforts in the past, the mechanism of this reaction is still under debate. In CTK the reaction is studied on a metal-supported catalyst under flow conditions in a pressure range extending from atmospheric pressure down to 100 Pa. Sudden changes in the partial pressures of the reactants then allow following the relaxation to either steady-state conditions ("transients") or cleanoff ("back transients"). In PFDMS short field pulses of several volts per nanometer are applied to a model catalyst which resembles a single metal particle grain (a "tip"). These pulses intervene during the ongoing reaction under flow conditions at pressures ranging from 10(-1) to 10(-5) Pa and cause field desorption of adsorbed species. This method is particularly suited to detect reaction intermediates in a time-dependent manner since the repetition frequency of the pulses can be systematically varied. It is shown that both methods lead to complementary results. While CTK allows conclusions on the mechanism of CO hydrogenation by following the time-dependent formation of hydrocarbon species, PFDMS provides insight into the initial steps leading to adsorbed CxHx species. A quantitative assessment of the CTK data allows the demonstration that the catalyst under working conditions is in an oxidized rather than metallic state. The initial steps to oxidation are also traced by PFDMS. Most importantly, however, CTK results allow formulation of a reaction mechanism that is common for both hydrocarbon and oxygenate formation and is based on the occurrence of a formate-type species as the most abundant surface intermediate.