EXPERIMENTAL STUDY AND THEORETICAL MODELING OF CATALYST DEACTIVATION (COKE)

The decay of catalyst activity in solid-catalyzed reactions is of great commercial importance, since the catalyst deactivation rate has a major influence on the economic design and operation of industrial catalytic reactors. Therefore, catalyst deactivation has been widely investigated, both experimentally and theoretically. In the present program, simple mathematical models have been developed to describe the kinetics of the primary and deactivation reactions. Experimental deactivation data from a model system, NO reduction by NH(,3), were used to model deactivation by a pore-filling mechanism. Literature data were used to model coke-related deactivation at high coke levels, (up to 20%), greatly exceeding monolayer coverage. Experimentally, the effect of SO(,2) on the reaction rate for the selective catalytic reduction of NO by NH(,3) over a vanadia-alumina catalyst was examined in a flow reactor. A two parameter model (based on two simultaneous and competitive reactions) has been developed for the analysis of the data. The activation energies for both fresh and deactivated catalysts appear similar. The sulfur content of the catalyst as well as its surface area appears to be a dominant deactivation parameter, analogous to coke-induced deactivation. Deactivation in this reaction system appears to be due to pore-filling and plugging by the deactivating agent, aluminum sulfate. Pore size distributions and thermal gravimetric analyses were employed to draw these inferences. For the theoretical modeling of coking, a mathematical model has been developed which provides a relationship between activity and coke content, based upon the concept of multilayer coke formation. The model provides a single functional relationship capable of describing the empirically observed linear, exponential and hyperbolic relationships. Extension of this analysis to deactivation of naphtha reforming catalyst is discussed. A mechanistic model is presented which may be useful in elucidating reasons for the high coke tolerance of bimetallic reforming catalysts; it is based upon a hydrogen equilibration step which can reduce the concentration of coke precursor on the catalyst. Laboratory program considerations are presented for modeling deactivation associated with a commercial constant conversion reactor operating policy. The consistency of primary and deactivation models has also been discussed for the mechanistic modeling of deactivating systems