Abstract

The cost of membrane separation processes for removing CO 2 and H 2S from low-quality natural gas can be reduced for some concentration ranges of CO 2 and H 2S by utilizing concurrently two different types of polymer membranes, one with a high CO 2/CH 4 selectivity and the other with a high H 2S/CH 4 selectivity. The polymers considered in this exploratory study were 6FDA-HAB polyimide for the removal of CO 2 and [poly(ether urethane urea)] (PEUU) for the removal of H 2S. It was required that the concentrations of CO 2 and H 2S in low-quality natural gas be reduced to US pipeline specifications (≤2 mol% CO 2 and ≤4 ppm H 2S). Low-quality natural gas was simulated in this study by CH 4/CO 2/H 2S mixtures containing up to 40 mol% CO 2 and 10 mol% H 2S. Twenty-seven membrane process configurations (PCs) were examined by computer simulations and optimized in order to determine the most economical configurations. Part I of this study considered only PCs without recycle streams [J. Hao, P.A. Rice, S.A. Stern, Upgrading low-quality natural gas with H 2S- and CO 2-selective polymer membranes. Part I. Process design and economics of membrane stages without recycle streams, J. Membr. Sci. 209 (2002) 177–206]. In Part II, reported below, the study was extended to two- and three-stage PCs with various recycle options. A sensitivity analysis was also made to determine the effects of variations in feed flow rate, feed pressure, membrane module cost, and wellhead price of natural gas on process economics. The economically optimal PCs were found to be either two membrane stages connected in series with or without recycle streams or single stages without recycle, depending on feed composition and selected operating conditions. The optimal two-stage PCs with recycle streams would utilize the H 2S/CH 4-selective membranes in the first stage and either the CO 2/CH 4 or the H 2S/CH 4-selective membranes, or both, in the second stage. Three-stage membrane PCs were not found to be economically competitive under the conditions assumed in this study.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call