The effect of zeolite type on the hydrocracking of long n-paraffins

Master Thesis


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

University of Cape Town

Although it is debatable as to the lifetime of the planet’s crude oil reserves, it is indisputable that they are finite and will, eventually, become exhausted. As such it is desired to devise methods whereby currently available non-crude oil derived hydrocarbon feedstocks may be utilised for the production of clean, high quality liquid fuels (particularly middle distillate fuels such as diesel and jet fuel), aiding in alleviating the demand on crude oil reserves. One technique whereby this may be achieved involves the conversion of the non-crude oil derived hydrocarbon feedstock (for example stranded gas, remote natural gas, coal or biomass) to syngas (a mixture of CO and H2). This syngas is subsequently converted to paraffinic wax by Fischer-Tropsch Synthesis, and this wax selectively hydrocracked down to the desired distillate fuels fraction. This hydrocracking may be conducted utilising either monofunctional or bifunctional catalysts. Monofunctional catalysts, such as the supported sulphided base metals or metal oxides utilised in many crude oil refineries, yield a product with minimal additional branching, as desired, though would contaminate the otherwise clean Fischer-Tropsch wax with sulphur. Bifunctional catalysts, utilising a metal on an acidic support, yield a product with significant branching, yet do not contaminate the product. Furthermore, the acid supports utilised in oil refinery applications are limited to either amorphous silica-alumina, or H-USY (a thermally treated large pore zeolite) due to the presence of large, bulky polycyclic and highly branched molecules in the feedstock. Fischer-Tropsch wax exhibits only minimal branching, and it was hence the aim of this investigation to determine whether zeolites with pores smaller than those utilised in the hydrocracking of crude oil derived feedstocks, specifically medium pore zeolites such as H-MFI, may be utilised in the bifunctional hydrocracking of this wax to impart shape selectivity upon the reaction, thereby limiting the extent to which branching may occur. In this regard, four different zeolites (H-MFI, H-BEA, H-USY and H-MOR) were tested under the same, industrially relevant conditions, and the results collated so as to quantify the effects of the each zeolite’s unique properties, in particular their pore geometries (in terms of pore size, shape and channel inter-connectivity), the performance of each catalyst in terms of its activity (the overall conversion of the feedstock) and selectivity (with regards to both carbon number distribution and the degree of branching). Furthermore, it was desired to determine the extent to which the anticipated transition state shape selectivity of some of the zeolites affected the stability (on-stream lifetime) of the catalyst through a reduction in coke formation. The results of this investigation indicated that medium pore zeolites show significant potential for use in the selective hydrocracking of a Fischer-Tropsch wax feedstock. It was found that those zeolites possessing medium sized pores (specifically H-MFI and H-BEA) exhibited a significantly higher activity than did those with only larger pores (H-USY and H-MOR), a phenomenon theorised to be due to the more orderly and efficient configuration of the adsorbed molecules within the medium pores promoting contact with active acid sites. Furthermore, it was found that H-MFI, with a porous network comprised entirely of medium pores, showed an improved selectivity towards the desired linear products, whilst zeolites with only wide pores (H-USY and H-MOR) and those with intersecting wide and medium pores (H-BEA) showed branched product selectivities roughly equivalent to one another, all greatly favouring the production of mono-methyl branched species. Unfortunately, due primarily to the large variations in the observed activity between the zeolites tested, the results of the deactivation analysis were inconclusive.

Includes synopsis.

Includes bibliographical references (leaves 141-144).