Alumina-modified cobalt catalysts for the Fischer-Tropsch synthesis

Doctoral Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

In the Fischer-Tropsch process, valuable hydrocarbons are produced using the basic starting materials hydrogen and carbon monoxide, which can be derived from alternative carbon sources such as coal, gas or biomass [1]. Although this process has been studied for almost a century, the effects of the support material on activity, selectivity and stability of the catalyst remain obscure. This study aims to gain fundamental insights into the effect of metal-support interactions in cobalt alumina based Fischer-Tropsch catalysts. To accomplish this, the effects of metal-support interactions have to be isolated from possible convoluting effects of the metal crystallite size and support porosity. This is achieved by preparing inverse-model catalysts, in which the support is deposited onto the metal, in contrast to conventional supported catalysts, in which the metal phase is deposited onto a porous support [2]. Cobalt alumina inverse-model catalysts were prepared by incipient wetness impregnation of cobalt oxide with aluminium sec-butoxide. The alumina loading was varied systematically between 0 and 2.5 wt% Al. The catalysts were characterised by X-ray diffraction (XRD), Transmission electron microscopy (TEM), H2 -chemisorption, and X-ray absorption near edge spectroscopy (XANES). The catalyst reducibility was studied by temperature programmed reduction (TPR), in situ (XRD) and in situ (XANES) experiments. The catalytic performance for the Fischer-Tropsch synthesis was studied in a slurry reactor under industrially relevant conditions. The alumina modification was found to prevent sintering and decrease the reducibility of the catalysts. With increasing alumina loading, and increasing calcination temperature, reduction peaks shifted to higher temperatures and peaks with maxima above 400 ˝C appeared in the TPR. The kinetic evaluation showed that the decreased reducibility was due to a decrease in the pre-exponential factor, which suggests that the alumina modification hindered hydrogen activation and/or nucleation of reduced cobalt phases. The activity of the catalysts for the FT reaction was found to increase with increasing alumina loading. This was likely an effect of the increase in metal dispersion upon alumina modification. Furthermore, alumina-modified catalysts had a higher C5+ and olefin selectivity, and lower methane selectivity. Pyridine-TPD experiments showed that the alumina modification introduced Lewis acid sites to the cobalt catalysts. Lewis acid sites may interact with adsorbed CO thereby weakening the C-O bond and facilitating CO dissociation. This was supported by CO-TPR experiments, which revealed that alumina-modified catalysts had an increased activity for the surface catalysed Boudouard reaction. It is concluded that the alumina modification increased the rate of CO dissociation on metallic cobalt. An increased rate of CO dissociation may lead to coverage of the metal surface with carbon thereby decreasing hydrogenation and shifting the product selectivity towards high molecular weight products. Hence, alumina may promote the selectivity of cobalt catalysts via a synergistic effect.