Modeling and optimization of membrane reactors for carbon capture in IGCC units

According to DOE projections, carbon dioxide (CO ₂) emissions from the combustion of fossil fuels will exceed six billion metric tons by 2035. About one-third of these emissions originate from coal-fired electricity generation [1]. These emissions will need to be mitigated in order to reduce the impact of projected climate change within this century [2]. Thus, there is a need to develop new technologies for economical production of electricity from coal that minimize the release of CO ₂ to the atmosphere. Integrated Gasification Combined Cycle (IGCC) power plants are a promising technology that can achieve higher efficiencies than conventional pulverized coal (PC)-fired plants. IGCC units also enable CO ₂ capture with lower penalties in energy efficiency and cost of electricity than their PC counterparts [3]. In this presentation, we investigate the alternative of pre-combustion capture of CO ₂ from IGCC plants using membrane reactors equipped with H2-selective molecular sieve (zeolite) membranes for the water gas shift (WGS) reaction. A challenge with using H2-selective membranes in the WGS section of coal-based gasification plants is their stability under high pressure and temperature conditions, and in the presence of steam and possibly other traces components such as hydrogen sulfide (H ₂S). Typical membrane materials used or proposed for H2 separations as well as the issues associated with each group of materials under WGS conditions [4] are: (i) metals (typically Pd-based): high cost, stability in the presence of contaminants (H ₂S) and H2 embrittlement; (ii) polymers: thermal degradation; (iii) amorphous silica: hydrothermal stability. Zeolite-based, molecular sieve membranes are one promising alternative for this application, as they are hydrothermally stable and have potential for high selectivity and flux [5]. The objective of this work is to develop a membrane reactor model for the WGS reaction using zeolite membranes. The developed model will be used for stand-alone simulation and optimization studies, and will ultimately be integrated into an IGCC system model. These studies aim to determine the membrane characteristics necessary to achieve the U.S. DOE R&D goal of 90% CO ₂ capture [6] and to obtain desired H ₂ recovery and CO conversion values. The desired targets should be reached for an optimal membrane use and without violating constraints in the reactor outlet streams, such as the retentate stream (rich in CO ₂) for capture and sequestration and the permeate stream (rich in H ₂) for power generation. Regarding the modeling task, we have developed a one-dimensional and isothermal shell and tube membrane reactor model for the WGS reaction. The model assumes the catalyst is packed in the tube side, a thin membrane layer is placed on the interior surface of the tube wall and the sweep gas flows in the shell side. This reactor model was simulated considering co-current and counter-current flow configurations to obtain steady-state compositions for primary species (CO, H ₂O, CO ₂, H ₂ and N ₂) present in the retentate and permeate streams. Several case studies have been performed assuming different membrane characteristics (permeance and selectivity). For each case study, we calculated the values of the membrane reactor parameters, such as CO conversion, H ₂ recovery/productivity and CO ₂ capture; and computed stream purities, such as the CO ₂ purity in the retentate and H2 purity in the permeate. Target values for these parameters as well as constraints for the reactor streams were defined based on data reported by the DOE [7]. The simulation set up considers WGS reactor operating conditions that are taken from the literature and are consistent with IGCC units. Simulation results showed good agreement with published simulation data [8] and a better performance for the counter-current configuration when compared to the co-current mode. Regarding the optimization task, we formulated and solved a novel optimization problem using the developed membrane reactor model to guide the selection of the optimal reactor design among typical scenarios of operation, including: (i) pre-shift, membrane separator, WGS reactor; (ii) pre-shift, WGS membrane reactor; (iii) WGS reactor, membrane separator; (iv) stand-alone WGS membrane reactor. To address all of these design alternatives in one formulation, the decision variables considered in the problem were specified as the lengths associated with the reaction and permeation zones. The problem was solved with the objectives of maximizing the CO conversion and H2 recovery, while minimizing the cost of membrane used as a function of its surface area required. This problem was also subjected to the target specifications in the membrane reactor parameters and constraints in the retentate and permeate streams that were mentioned above. Optimization results indicated as optimal solution the reactor design with a pre-shift followed by a WGS membrane reactor and potential savings in membrane material when compared to the original membrane reactor design.