Answers are grounded in this paper and cite their sources.
Experimental work demonstrates effective H2 separation from syngas using membranes, suggesting a feasible industrial solution.
H 2 separation and recovery play a key role in the industrial field due to the variety of pure H 2 applications and its relevance as clean fuel, energy carrier and feed material for many processes (e.g., fuel cells, ammonia and methanol production, oil refining etc.). Conventional gas separation methods include PSA, liquid absorption and cryogenic; however, membrane technology is proving to be a valid alternative that is increasingly developing in these last years. In this work, hollow fibre polymer membranes were used to concentrate H 2 from a syngas mixture, while also ensuring high recovery levels. Experiments, which were carried out considering pure gases (i.e., H 2 , CO 2 , CO and CH 4 ) and mixtures, reveal high H 2 /CO and H 2 /CH 4 ideal selectivity (e.g., 145 and 222, respectively), whereas a moderate H 2 /CO 2 value of about 4 was measured. In binary mixtures, H 2 permeation was improved by CO 2 and slowed down by the other species, showing a s behaviour similar to that of CO 2. Therefore, H 2 /gas selectivity resulted lower than the ideal value. A quaternary gas mixture representing a syngas stream was then analysed, achieving a maximum H 2 concentration of 93% paired with a recovery of 92%, when membrane process operates in two stages. A three-stage membrane system for H 2 and CO concentration from syngas was simulated, being able to achieve a H 2 purity of about 90% with a recovery higher than 98%. This process also allowed for a CO concentration from 13 up to about 71% in the final retentate product, where it was assumed that all CH 4 is recovered, since it was not experimentally detected in the permeate stream. An economic analysis showed that raw material was the main cost (68 – 74%), followed by compression (24 – 30%), whereas membrane cost was negligible (just 2% of the total).
No takes yet. Share an insight, caveat, or question.
Zito et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: