Sour gas transport in poly(ether-b-amide) membranes for natural gas separations

Sour gas transport in poly(ether-b-amide) membranes for natural gas separations

Carbon dioxide

Separation

Gas industry

Sulfur compounds

Hydrogen bonds

Gas hazards

Well testing

Gas permeable membranes

Inert gases

Ethers

Sour gas

Permeation

Amides

Natural gas well production

Health hazards

Natural gas wells

Bonding

Innovations in membranes for natural gas processing frequently consider ideal separation performance as the defining success metric for removing carbon dioxide (CO2) from methane (CH4). However, realistic natural gas feeds are complex mixtures containing additional constituents of varying concentrations, including hydrocarbons, inert gases, and hydrogen sulfide (H2S). As natural gas demand spurs production from sub-quality sour natural gas wells, simultaneous CO2 and H2S removal becomes an increasingly relevant separation challenge. Although CO2 separations are well studied, health and environmental hazards of H2S often impede mixed gas permeation testing of membranes with H2S-containing feeds. This work aims to further sour gas membrane research by studying the impact of feed compositions on gas transport properties of membranes formed from five different grades of commercial polyether-b-amide (Pebax). High pressure mixed gas permeation testing confirmed the importance of membrane stability: CO2/CH4 selectivity values for each membrane dropped by 2442% under mixed sour gas feeds compared to ideal selectivities. Membrane stability correlates well with extents of hydrogen bonding, but overall performance depends largely on block type and relative block content. When exposed to a 20% H2S-containing feed, Pebax 1074 and 1657 retained the highest selectivities of 12.012.8 for CO2/CH4 and 79.980.6 for H2S/CH4 at 800 psi. The changes in membrane performance with variations in gas content and pressure highlight the importance of screening membranes with realistic gas feeds. 2019

595

Journal of Membrane Science

2020

Harrigan, Daniel J.

Yang, John

Sundell, Benjamin J.

Lawrence, John A.

O'Brien, Jeremy T.

Ostraat, Michele L.

journalArticle

3767388

10.1016/j.memsci.2019.117497