The reaction between carbon monoxide and hydrogen can be catalysed to give a variety of products including methane, methanol, acetic acid and waxes.  The mechanisms involved are rarely simple and using ¹³C, ¹⁸O, and D₂ as tracers has helped to determine reaction paths.  Often a combination of isotopes is needed to fully specify the mechanism. In this talk we will look at complex systems where there is the potential for multiple products and where understanding the mechanism is critical to catalyst optimisation.

The synthesis of methanol from carbon monoxide-hydrogen and carbon monoxide carbon dioxide-hydrogen over a copper/zinc oxide/alumina catalyst was studied using transient isotope tracing techniques.  The non-steady-state period immediately after start-up has been studied and the product distribution and time delay could be explained with reference to the amount of residual oxygen on the catalyst after reduction.  Using [^l8O]carbon monoxide and [¹⁸O]carbon dioxide it was shown the label is not detected in the methanol product for 0.3 h.  From the steady-state activities and a residence time of 0.3 h the size of the surface reservoir of methanol precursor was calculated for both carbon monoxide-hydrogen and carbon monoxide-carbon dioxide/hydrogen feedstreams.  This figure was in good agreement with the amounts of methanol removed from the catalyst when the feedstream was switched from carbon monoxide-hydrogen or carbon monoxide-carbon dioxide-hydrogen to hydrogen alone.

The production of oxygenate species over rhodium catalysts can under the correct conditions produce a range of oxygenate species including methanol, ethanol, acetaldehyde and acetic acid.  Using [¹³C]CO, [¹⁸O]CO, and D₂ the mechanism of formation of each species was determined.  When labeled carbon monoxide was introduced, neither methane nor ethanol nor methanol showed any incorporation; however, the labels were rapidly incorporated into the aldehydic function of ethanal.  Labeled water, produced from the hydrogenation of [¹⁸O]CO, took more than 0.3 h to desorb.  The results suggest that (i) ethanol and ethanal are produced independently with no common intermediate, (ii) the formation of alcohols is slow (taking over 0.5 h), and (iii) carbon monoxide is not hydrogenated directly to methane but goes through the hydrocarbonaceous residue present on the surface.  The carbonaceous residue was found to play a central role in the mechanism, supplying in effect both hydrogen and CH2 units to other reactive surface species.

The details of these and other studies will be reported showing the unrivalled access to mechanistic information that isotopes can give in the field of catalysis.