Natural gas that contains larger amounts of hydrogen sulfide (H(2)S) and carbon dioxide (CO(2)) is termed sour gas. Before it can enter a pipeline, it must be "sweetened" by removal of its acidic impurities. Through fine tuning of the ratios of two molecular components, it is possible to produce tailored polyimide membranes that can purify sour gas with a wide range of compositions, as reported by researchers in the journal Angewandte Chemie.
The main component of natural gas is methane (CH(4)). The H(2)S and CO(2) in sour gas react acidically with moisture, making them highly corrosive. In addition, H(2)S is highly toxic and poses a safety risk. Today, sweetening is usually achieved through very energy-intensive chemical scrubbing, which is not economically viable for gas with high concentrations of H(2)S and CO(2). In addition, this process requires a large, complex apparatus that is impossible to use in remote or offshore facilities. Scalable, economical membrane separations represent an excellent alternative.
Membranes based on glassy polyimide polymers made of a special nitrogen- and oxygen-containing group demonstrate good separation efficiency. However, a fundamental understanding of the relationships between the structures of polyimides and their gas-transport properties in the presence of H(2)S has been lacking, impeding the design of advanced membranes. A team led by William J. Koros at the Georgia Institute of Technology (Atlanta, USA) has now taken on this subject.
Membrane separations are based on the fact that gases with higher solubility pass more easily through membrane materials; however, smaller gas molecules can also diffuse through membranes more easily. The challenge for sweetening lies in the fact that the separation of CO(2) relies primarily on a size difference (CO(2) is smaller than CH(4)), while the separation of the similarly sized H(2)S and CH(4) depends on differences in solubility. In addition, glassy polyimide membranes begin to soften as they absorb more dissolved gas. This is favorable for the separation of H(2)S but unfavorable for the separation of CO(2).
For their experiments, the researchers produced polyimides based on 6FDA (4,4'-(hexafluoroisopropylidene) diphthalic anhydride. They used two different 6FDA building blocks, which they polymerized in a variety of ratios. One building block (DAM) introduces a bulky trimethyl benzene group, which prevents the polymer chains from being densely packed. This increases both the gas permeability and the tendency to soften. The other building block (DABA) contains a polar benzoic acid group. This tightens the packing of the chains, decreasing permeability, but increases H(2)S solubility.
Higher proportions of DAM increase the permeability toward CO(2), but also CH(4), which decreases selectivity. In contrast, the selectivity with regard to H(2)S is barely affected. The more DAM included, the more the polymer softens, which is unfavorable for CO(2) but favorable for H(2)S. By carefully adjusting the relative amounts of the building blocks, the packing of the polymer chains and the tendency to plasticize can be balanced to produce membranes that simultaneously and efficiently separate out both H(2)S and CO(2). This makes it possible to tailor membranes for different natural gas compositions.
About the Author
William J. Koros is the Roberto Goizueta Chair and Georgia Research Alliance Eminent Scholar in Membranes at Georgia Tech. He was elected to the National Academy of Engineering in 2000 and is a Fellow of the AIChE and AAAS. His research focuses on developing advanced membranes for challenging gas separation.