Improving conversion and selectivity of catalytic reactions in bubbling gas-solid fluidized bed reactors by control of the nonlinear bubble dynamics

In this paper a model is presented that is a dynamic extension of the classic two-phase reactor models used to predict conversion and selectivity of fluidized reactors. The most important part of the model is a dynamic discrete bubble model that can correctly predict bubble sizes and also exhibits chaotic dynamics. This bubble model is based on the discrete bubble models presented by Clift and Grace [AIChE Symp. Ser. 66 (105) (1970) 14; 67 (116) (1971) 23; in: J.F. Davidson, R. Clift, D. Harrison (Eds.), Fluidization, Academic Press, London, 1985, p. 73] and Daw and Halow [AIChE Symp. Ser. 88 (289) (1992) 61]. The latter showed that this type of models can exhibit chaotic behavior. By application of an extended version of Pyragas' control algorithm [K. Pyragas, Phys. Lett. A 170 (1992) 421] the bubble dynamics can be changed from chaotic to periodic in a ‘flow'-regime in which the model otherwise would predict chaotic behavior. Pyragas' control algorithm is used to synchronize a chaotic system with one of its periodic solutions using a feedback control loop. This results in smaller bubbles, thus enhancing mass transfer of the reactant gas in the bubbles to the catalyst particles. The model is used to predict the effect of the changed bubble dynamics on a catalytic reaction of industrial importance, viz. the ammoxidation of propylene to acrilnitril (Sohio process). It is shown that both conversion and selectivity are appreciably enhanced.