FCC catalyst performance evaluation

Abstract

Fluid catalytic cracking (FCC) is currently the most important route to gasoline and the efficiency of this operation is vital to the economics of a modern refinery. Unlike the constant properties required for most industrial catalysts, FCC catalysts are continuously being modified and new systems developed to meet changing demands, e.g. the ability to generate different product distributions, to cope with heavier and contaminated feedstocks, and to resist more severe regeneration conditions. The need to evaluate catalysts in the laboratory is common to catalyst manufacturers (to develop new systems and to generate accurate performance comparisons for promotional purposes) and refiners (to monitor the activity of catalysts currently in use in commercial units under controlled laboratory conditions, and to compare the performance of new catalysts from rival manufacturers). The testing of FCC catalyst performance in the laboratory is difficult, particularly if an attempt is made to mimic a riser cracker operation (500–1000 tonne catalyst inventory) on the small scale (e.g. 5 g). However, if catalysts are first artificially deactivated so as to bring about the same physical and chemical changes that occur during full scale operation, and then tested for activity and product selectivity using catalyst-to-oil ratios, contact times, and temperatures not too far removed from commercial practice, then useful results can be achieved. These methods allow catalysts to be ranked accurately, but do not predict the absolute differences to be expected in full-scale units. Such information, however, can be obtained from much more elaborate test methods or through the use of complex deactivation and cracking models. This paper discusses catalyst requirements in relation to the FCC process, examines feedstocks and possible contaminants, and highlights two key stages in laboratory evaluation, namely catalyst deactivation and activity/selectivity testing. Equipment designs are compared and typical results used to demonstrate the effects of laboratory processes. Finally, the development of models is discussed in broad terms and some results are shown which demonstrate the accuracy of laboratory predictions against commercial unit experience.