Datensatz

Zusammenfassung

In the field of chemical reaction engineering intensive research is devoted to the development of new processes in order to improve selectivity and yields of intermediate product. Important example reactions are e.g. catalytic partial oxidations (CPO) or selective hydrogenations (Dixon, 2003). For such reactions improved integral reactor performance can be achieved by optimised stage-wise dosing of one or several reactants (Hamel et al., 2003). Adjusted dosing profiles can be realised e.g. by feeding reactants at certain discrete positions in a fixed-bed reactor or by dosing them separately through permeable reactor walls, e.g. through tubular ceramic membranes (Fig. 1). This presentation delivers a deeper insight into various aspects of multi-satge dosing concepts based on numerical simulations and experimental investigations. For this aim the partial oxidation of ethane to ethylene on a VO_x/Al₂O₃ catalyst was considered as a model reaction. For the experimental study a laboratory scale set-up has been constructed with a three-stage conventional fixed bed (FBR3) and a three-stage packed bed membrane reactor cascade (PBMR3). The reaction kinetics and the properties of the asymmetric alumina membranes were quantified in prior studies (Klose et al. 2003, Thomas et al. 2001). One dimensional reactor models have been used to identify the operation window of both rectors. More detailed models based on the Maxwell-Steffan approach were included to examine mass transfer through the membrane. Based on the theoretical results, an experimental study was carried out in a temperature range between 580 °C and 660 °C. The molar O₂/C₂H₆ ratio was varied between 0.5 and 4, whereby the ethane concentration was kept below 2 vol. %. The tube to shell side flow ratio (TS/SS) in the membrane reactor was varied between 1:4 and 1:9. In the three-stage membrane reactor, experiments with different o dosing profiles were performed, e.g. incerasing (17-33-50), uniform (33-33-33), decreasing (50-33-17). Unexpectedly for the conditions studied the shape of the axial dosing profile hardly affected the reactor performance (Fig. 2.) A possible explanation for this observation is back diffusion of ethane and the products to the shell side emerges, which was not adequatly predicted by the simulations. Additional mea-surements confirmed this hypothesis. To avoid undesired back diffusion, the shell side flow rates should be maintained above a certain limit, which narrows down the operation window of such kind of membrane reactor. This turned out to be one of the most important limitations of the PBMR3 concept compared to FBR3 on this laboratory scale level. However, his problematic issue should not be so critical for larger reactor scales, where trans-membrane flow rates can be considerably higher. References: Dixon, A.G., Int. J. Chem. React. Eng., Vol. 1, 2003, R6 Hamel C., Thomas S., Schädlich K., Seidel-Morgenstern A., Chem. Eng. Sci., 58, 2003, 4483 Klose F., Wolff T., Thomas S., Seidel-Morgenstern A., Catal. Today, 82, 2003, 25 Thomas S., Schäfer R., Caro J., Seidel-Morgenstern A., Catal. Today, 67, 2001, 205