Bulk separation of hydrogen and carbon monoxide by one-column pressure swing adsorption: Experimental and statistical analysis

This work evaluates separation of CO/H-2 mixture (50:50vol%) on a commercial zeolite 5A by using 1-column pressure swing adsorption process. Statistical analysis using response surface methodology was employed to evaluate the combined effect of 2 independent variables, namely, adsorption step time and purge-to-feed ratioon the hydrogen purity, recovery, and productivity as responses. These responses were efficiently modeled by 3 second-order polynomial equations. Optimization of the response function was done based on the desirability. The optimum conditions were investigated at different values of responses. For instance, when hydrogen purity was set at maximum value, recovery and productivity were set at in range and the model predicted 99.85% hydrogen purity, 47.69%, and 1.26x 10-5 hydrogen recovery and productivity, respectively. In this condition, the optimum values of purge-to-feed ratio and adsorption time were 0.24 and 7.35minutes. Validation of models was carried out by practical experiments in predicted optimum conditions, and the results confirmed the accuracy of models for predicted values. This work evaluated modeling and optimization of H-2/CO separation process (pressure swing adsoprtion process) by response surface methodology (based on the central composite design methodology), and the acquired data were modeled by second-order polynomial equations. Optimum purge-to-feed ratio and adsorption time were obtained depending on the values of purity, recovery, and productivity (maximum or in range). To confirm the accuracy of models for predicted values, experiments were done in optimum conditions, and results showed that the models are acceptable.