(453g) Membrane Steam Methane Reforming: A Multiobjective Bayesian Optimization Strategy for Low Carbon Emissions

The development of hydrogen production is indispensable for shifting to sustainable energy. However, most of the hydrogen is produced from natural gas, which has to be better intensified to lower carbon emissions. The exploitation of current immense natural gas reserves for hydrogen production will also help in the hydrogen energy infrastructure that will be used for green hydrogen. Also, methane can be used for hydrogen storage with a gravimetric capacity of 25 wt. %. In this study, a CFD model was developed using FORTRAN language and an optimization approach was introduced using Python to analyze the membrane steam methane reforming (MSMR) over Ni/Al2O3 catalyst in a rectangular channel with the use of a Pd membrane. A multiobjective Bayesian optimization was applied seeking the optimal membrane distribution for high hydrogen recovery to membrane length ratio and low CO2 emission, which rise via membrane presence that promotes water gas shift reaction (WGS). The optimization results showed a significant decrease in CO2 selectivity of 36.6% compared to the traditional case. However, with an offset in the hydrogen recovery. This methodology can be successfully used for process control to lower CO2 emissions.