(619a) Methane Oxidative Coupling: Synthesis of Membrane Reactor Networks

In this work, attainable performance of the methane oxidative coupling process based on different alternative reactor structures including fixed-bed reactor (FBR), two different feeding-structures of porous packed bed membrane reactor (PBMR), and possible conceptual networks as a combination of them are analyzed in a comprehensive model-based study. In this context, a contour-based graphical visualizing method accompanied by multi-scenario generation and synthesis approach is exploited so as to determine the highest achievable process performance and its corresponding specifications in the resulting network.

As performance indicators, several objective measures including the highest achievable values for methane conversion, yield and selectivity, are applied over a high number of design scenarios. Thereby, several sets of structural and operating parameters are evaluated. The investigated parameters are: temperature, membrane thicknesses, types of catalysts (kinetic characteristics), amount of inert packing in the catalyst bed, the flow rate of oxygen rich stream entering the reactor system (total methane to oxygen ratio), distribution of the oxygen rich stream through the reactor blocks (local methane to oxygen ratio), distribution of the total methane-rich feed stream into the reactor blocks and the contact time represented by the reactor length. A recent work (Godini et al., Ind. Eng. Chem. Res. 2010, 49, 3544?3552) provides detailed information about the concept, performance and characteristics of such a network.

The initial superstructure including all potential scenarios is sequentially discarded and the scenarios are evolutionarily screened to keep the best scenarios with the highest achievable performance. In consequence, the few remained scenarios can be experimentally examined later using a flexible experimental set-up. As an improvement, faster screening and wider coverage of potential scenarios are achieved using the proposed approach in comparison to the approaches, which are only based on either a full-scenario examination or conceptual optimization procedure.

Moreover, the one and two-dimensional mathematical models used for this analysis address the effects of non-isothermal behavior and the effects of gas phase reactions in the process as well. In this regard, crucial practical limitations with regard to the required reactor length of each design-scenario and the hot-spot formation can be tackled as an inequality constraint in such an analysis. As a result, the specifications of the optimum scenarios for the reactor structure are determined considering several objective measures and practical aspects. Generally each individual reactor concept stands for one performance indicator. However, it can be concluded that the optimum network of reactors can satisfy more than one objective measure simultaneously.

Acknowledgment The authors acknowledge support from the Cluster of Excellence ?Unifying Concepts in Catalysis? coordinated by the Berlin Institute of Technology and funded by the German Research Foundation - Deutsche Forschungsgemeinschaft.