Effects of Crystallite size and water partial pressure on the activity and selectivity of low temperature iron-based fischer-tropsch catalysts
Doctoral Thesis
2009
Permanent link to this Item
Authors
Supervisors
Journal Title
Link to Journal
Journal ISSN
Volume Title
Publisher
Publisher
University of Cape Town
Department
License
Series
Abstract
Fischer-Tropsch synthesis is a reaction between hydrogen and carbon monoxide to produce long-chained hydrocarbons and water. It has been stated that the Fischer-Tropsch synthesis reaction is a surface phenomenon, thus for optimum catalyst performance, maximum metal usage must be achieved. Therefore it is expected that the smaller the crystallite size of the metal, the more active the catalyst will be under test conditions. This work investigates the influence of iron crystallite size and water co-feed on the activity and selectivity of low temperature Fischer-Tropsch synthesis. In heterogeneous catalysis it is well known that with smaller crystallites, more active surface area is exposed allowing for higher overall activities. However the subject of size sensitivity remains an issue as research has shown that nano-meter sized crystallites below a certain point (ie. 6 nm) can behave differently to larger crystallites in terms of activity and selectivity in Fischer-Tropsch synthesis. It has been suggested that this behaviour may be due to the effect of crystallite size on iron phase changes such as oxidation by product water during the Fischer-Tropsch (FT) reaction. Alternatively, effects of structure sensitivity may play a role. This work can essentially be divided into 3 sections; firstly, the preparation and characterisation of a model alumina supported iron catalyst with iron crystallites in the nano-meter range between 2 and 16 nm. Secondly, Fischer-Tropsch testing in a fixed bed reactor of model catalyst systems, testing the effect of differing crystallite size on the activity and selectivity of Fischer-Tropsch synthesis, followed by the characterisation of spent catalyst samples. Finally, Fischer-Tropsch testing of model catalyst systems under different conditions of water addition simulating conditions under high syniii thesis gas conversion, again followed by the characterisation of spent samples. The first major objective of this work was the preparation of a model catalyst sample, where a narrow size distribution was required as well as a good distribution of the metal crystallites onto the support material. A narrow size distribution was successfully achieved through the utilisation of the reverse micelle technique. Size control can be achieved through variation of the water to surfactant ratio, with a high ratio leading to larger crystallites being formed. An even distribution of metal crystallites onto the support material was harder to achieve. Four different support addition methods were tested as well as different support materials, drying conditions and calcination conditions. It was shown that the support addition method previously used by Mabaso (2005) was the one that achieved the best dispersion. Furthermore, the variation of support, drying and calcination conditions can have a large impact on the final catalyst with different supports and conditions leading to increased clustering and sintering. Alumina was found to be the best support material and a threshold temperature of 300 C existed for calcination past which severe sintering took place. Model catalysts were successfully prepared with a narrow size distribution and a good dispersion of metal crystallites onto the support material. Six catalyst samples were prepared with crystallite sizes ranging between 2 and 16 nm with a metal loading of 13 wt%. Examination of the synthesis gas conversion showed that the specific Fischer-Tropsch rate shows a decrease with decreasing crystallite size. This change in rate has been theorized to be due to either the thermodynamically simpler oxidation of smaller crystallites or the lack of ensembles of atoms required for Fischer-Tropsch synthesis on smaller crystallites. Through the characterisation of spent samples, it has been shown that the âEnsemble Effectsâ theory to be the more likely one as bulk oxidation was not observed. In terms of product formation, smaller crystallites showed a higher inclination for the production of methane. This is further support for the âEnsemble effectsâ theory, as this result leads to the conclusion that less chain growth sites and more methanation sites may be available on smaller crystallites leading to an increased methane selectivity on small crystallites. iv Chapter 0. Synopsis Other product selectivities such as olefin and oxygenate formation did not follow the trend of hydrogen richer products obtained on smaller crystallites, instead it was the catalyst samples in the middle metal crystallite size range that showed increased propensity toward secondary reactions. This was believed to be due to either an extension of the âEnsemble Effectsâ theory or the electronic effects between readsorbing olefins and the metal surface. A conversion level of below 10% was chosen for this work in order to fully and directly compare the activity and selectivity results of the various catalyst samples. However this condition means that the effect of higher conversion levels are not shown. In order to overcome this problem the water partial pressure was increased, where a higher water partial pressure simulates the conditions of higher conversion via addition of water. The âbasecaseâ water partial pressure was set at 0 bar, while water addition conditions had water partial pressures of 3 and 6 bar. In terms of specific Fischer-Tropsch rate it was shown that the addition of water leads to deactivation of the catalyst irrespective of crystallite size. This deactivation has been theorized by previous work to be either due to oxidation or sintering. Again characterisation results, including an in-situ method, show that oxidation is not the likely cause of the deactivation, instead the clustering and agglomeration of metal crystallites show that sintering is the more likely candidate. Product formation results show that the addition of water leads to a decrease in methane selectivity and an increase in olefin production. It is theorized that these selectivity results are caused by water inhibiting desorption and readsorption mechanisms leading to increased chain growth and decreased secondary reactions respectively.
Description
Reference:
Cheang, V. 2009. Effects of Crystallite size and water partial pressure on the activity and selectivity of low temperature iron-based fischer-tropsch catalysts. University of Cape Town.