Abstract
Recently we proposed a new approach for analyzing and understanding dynamics and feasibility of combined reaction separation processes (Grüner and Kienle, 2004). The apprach is based on assumption of simultaneous phase and reaction equilibrium and can be viewed as an extension of the well-known equilibrium theory developed by Rhee et.al. (1986, 1989) among other non-reactive separation processes. It makes use of transformed concentration variables, which were first indtroduces by Doherty and co-worker (1987) for the steady state design of reactive distillation processes. It was shown that these transformed concentration variables can be directly generalized to the corresponding dynamic problem. Further, they can also be applied to other reactive separation processes, like fixed-bed as well as moving bed chromatographic reactors, for example. The approach provides profound insight into the dynamic behavior of these processes reveals inherent bounds feasible operation caused by reactive azeotropy.
In the present contribution the methods are applied to ternary reaction systems of type 2A <-> B + C. It is shown that in the first case total conversion is only possible if reactant A has intermediate affinity to the solid phase. If A has highest or lowest affinity to the solid phase conversion, is limited due to reactive azeotropy for any feed concentration. In contrast to this in the second case the behavior depends on the feed concentration. If the feed concentration is sufficiently low, total conversion is possible for any order of affinities to the solid phase. Instead for high feed concentrations similar limitations occur as in the first case. As a practical example for the second reaction scheme, some ester hydrolysis reactions of methyl and ethyl formate, as well as methyl and ethyl acetate are discussed by Seidel-Morgenstern and co-workers (2004). Water is taken as a solvent in large excess and therefore has approximately constant concentration. In this application, the ester has always the highest affinity to the solid phase. It is shown that in the present case no limitations occur within the feasible range of feed concentration below the solubility boundary. The theoretical predictions are in agreement with simulation results from a rigorous model and with experimental observations.
References:
Barbosa, D., M. F. Doherty, Pro. R. Soc. Lond. A413 (1987) 459.
Grüner, S., A. Kienle, Chem. Eng. Sci. 59 (2004) 901.
Mai, P. T., T. D. Vu, K. X. Mai, A. Seidel-Morgenstern, Ind. Eng. Chem. Res. 43 (2004) 4691.
Rhee, H.-K., R. Aris, N. R. Amundson, First-Order Partial Differential Equations: Volume I - Theory and Application of Single Equations, Prentice Hall, New Jersey, 1st Ed. (1986).
Rhee, H.-K., R. Aris, N. R. Amundson, First-Order Partial Differential Equations: Volume II - Theory and Application of Hyperbolic Systems of Quasilinear Equations, Prentice Hall, New Jersey, 1st Ed. (1989).
