Chapter 3. Physical Aging of Membranes for Gas Separations

Abstract

Most gas separation membranes are formed from glassy polymers because of their exceptional permeability–selectivity properties. However, glassy polymers are non-equilibrium materials that will spontaneously, but usually slowly, change over time towards an equilibrium state by a process known as physical aging. Gas separation membranes, in asymmetric hollow fibers or composite membrane form, require ultra-thin active layers (i.e. on the order of 100 nm or less) to produce the high flux required to compete with other separation technologies. While it was previously assumed that the behavior of the thin films used in gas separation membranes would match that of bulk films, the physical aging rate can become orders of magnitude more rapid when film thickness is decreased below 1 μm, indicating a strong influence of the free surface on aging behavior. This phenomenon is an intrinsically fascinating scientific issue, and understanding physical aging has broad impacts in predicting the performance of gas separation materials.Early evidence of accelerated aging in thin gas separation membranes was reported for asymmetric hollow fibers. Quantifying the influence of film thickness on aging rate was difficult, however, because of the inability to accurately measure thickness in these materials. As a step towards understanding the aging behavior complex, asymmetric hollow fiber gas separation membranes, studies have focused on well-defined free-standing films of known thickness in the relevant thickness range. By studying physical aging in these films, insight has been gained regarding the mechanisms causing accelerated aging in thin films and a better understanding of the influence of free surfaces and confinement on polymer films has been developed. This chapter discusses how aging behavior evolves with thickness, across a broad range of thicknesses from bulk films to length scales approaching the polymer coil size.This chapter also presents the development of specialized techniques for studying physical aging in thin and ultra-thin glassy polymer films as gas separation membranes. Because of the fragility of these delicate films, specialized techniques for preparing and handling samples are required. One major roadblock to studying gas permeability in ultra-thin films is the presence of microscopic pinhole defects, which form with increasing frequency as film thickness is reduced. While this issue may not compromise results from studies using ellipsometry, fluorescence, and other techniques, these trans-membrane defects destroy selectivity and mask permeability of a material under study, thus rendering the sample useless for gas transport studies. A novel coating technique to circumvent this problem was recently developed, based on a similar technique that enabled the industrial development of hollow fiber membranes for gas separation.Since previous history has a strong influence on glassy polymer properties, including physical aging, it is important that aging studies employ samples with identified and reproducible histories. Typically, a controlled quench from above the glass transition temperature, Tg, is used to define an experimental starting time; however, actual applications of these materials involve much more complex histories. Because of the complex and ill-defined history of glassy polymers in industrial applications, it is critical to understand how aging behavior changes as a function of the material's previous history to predict long-term performance. This chapter includes discussion on the influence of these effects on gas separation membrane performance and physical aging.The rate of physical aging depends on the ratio of the driving force, i.e. the displacement of the specific volume from its equilibrium value, and the relaxation time for the sample, which is a function of temperature and the material's current free volume state. Mechanisms to describe physical aging have been proposed, including lattice contraction and the diffusion of free volume from the system. Although these models have been used to describe a variety of aging data in a phenomenological manner, the mechanisms that cause accelerated aging remain unclear. Recent experimental advances have promoted the development of predictive models regarding the long-term permeation properties of gas separation membranes.Finally, some of the remaining questions and suggestions for future experimental design to better understand the deviations from bulk behavior in thin glassy films as gas separation membranes are presented.