(453e) Nonlinear Differential Algebraic Model of a Water-Gas Shift Membrane Reactor for Low-Carbon Hydrogen Production | AIChE

(453e) Nonlinear Differential Algebraic Model of a Water-Gas Shift Membrane Reactor for Low-Carbon Hydrogen Production

Authors 

Agi, D. - Presenter, University of Notre Dame
Dowling, A., University of Notre Dame
Water-gas shift membrane reactors (WGS-MR) are a promising pathway to low-cost blue hydrogen production from renewable sources, including biomass [1]. WGS-MRs enable the simultaneous removal of hydrogen as it forms, thereby shifting CO conversions beyond equilibrium limitations, resulting in CAPEX/OPEX cost savings [2, 3]. Pd-based membranes have been extensively studied in WGS-MRs due to their permselectivity towards hydrogen and suitability for high-temperature operation [1, 4]. To exploit their cost benefits for blue hydrogen production, WG-MRs performance needs to be optimized, which includes tuning the complex mix of operating conditions (temperature, pressure, gas hourly specific velocity (GHSV), sweep), performance parameters (permeance and recovery), spanning material (permeability), and process scales.

In this work, we developed a mathematical modeling framework for a Pd-based WGS-MR that supports process simulation and optimization. The steady-state WGS-MR nonlinear differential algebraic model, which accounts for reaction kinetics and transport of H2 through Pd-based membranes based on solution/diffusion theory, is implemented in Pyomo, an open-source algebraic modeling language [5]. Using Ipopt numerical solver [6], the Pyomo model reliably converges in only a few CPU-seconds over a wide range of parameters, including feed pressures of 500 to 3500 kPa, reactor temperatures of 625 to 825 K, and gas hourly space velocities of 500 to 3500 hr-1. The model is qualitatively validated by comparison with laboratory-scale data and trends from the literature. Our modeling approach facilitates complex optimization applications, such as material property targets, and supports integration into open-source process flowsheet modeling environments, such as IDAES-PSE.

We demonstrated the optimization capabilities of our WGS-MR model by examining multi-objective functions to maximize hydrogen recovery and minimize production cost per unit feed, taking GHSV and reactor temperature as the decision variables. Further, we evaluated the economic viability of the WGS-MR for application in the production of blue hydrogen from biomass through techno-economic analysis (TEA) benchmarked on the Hydrogen Production Model (H2A) developed at the National Renewable Energy Laboratory [7]. The TEA showed that replacing the two-step WGS and pressure swing absorption units in the H2A process with WGS-MR boosts hydrogen recovery and results to cost savings and heat integration opportunities.

The WGS-MR model developed in this work serves as a tool for design and performance optimization to inform the integration of Pd-based WGS-MRs in blue hydrogen production and could be adapted to other membrane reactors for process development.