(393d) Modeling the Residence Time Distribution in a Continuous Plug-Flow Reactor

In a plug-flow reactor, the residence time distribution (RTD) directly impacts catalyst use, reaction selectivity and reaction yield. The RTD can be measured by tracer studies, but such studies are expensive. We investigate the residence time distribution in a 200 cubic meter industrial-scale plug-flow reactor composed of 13 agitated compartments using both RANS CFD simulations and LES Lattice Boltzmann method (LBM) simulations, and compare the results with tracer measurements. We calculate residence time distributions with mean-age theory in RANS models and with both mean-age theory and particle tracking in LBM models. As the particle based methods have no numerical diffusion, this allows us to quantify the effects of numerical diffusion in mean-age theory. Mean-age theory provides not only the residence time distribution, but also insight into dead-zones; it is not, however, readily measured. Residence time distribution is readily measurable by tracer studies but provides no local information on mixing. Simulation models provide an inexpensive mans of understanding local mixing and overall residence time distributions, as well as providing methods of screening process modifications to enhance the residence time distribution.