(346b) A Predictive Model for the Determination of Mixed Gas Transport and Solubility in Glassy Polymers

The description of the mixed gas transport and sorption in polymers is of fundamental importance for the development gas separation membranes, and for the evaluation of their performances under real conditions [1,2]. However, very few approaches are suitable to that aim, especially for glassy phases, for which the Dual Model Sorption model (DMS) is still largely used, in spite of its limited, if any, predictive ability [3].

The transport model recently proposed by Minelli and Sarti has already proved its ability in the description of the permeability and solubility of pure gases and vapors in glassy polymers, on the basis of a simple fundamental approach [4]. The model considers the diffusion coefficient as the product of the molecular mobility and the thermodynamic factor, related to the concentration dependence of the penetrant chemical potential. The model relies on the thermodynamic description provided by the nonequilibrium thermodynamics for glassy polymers, NET-GP (and the nonequilibrium lattice fluid, NELF, in particular), which provides the thermodynamic behavior of the polymer/penetrant system, in a predictive fashion [5]. In parallel, a simple exponential function of penetrant concentration is used to describe the mobility coefficient behavior, which contains the only two adjustable parameters of the model: the penetrant mobility at infinite dilution and the plasticization factor.

In this work, the transport model has been extended to describe the transport and the solubility of gaseous mixtures in glassy polymeric membranes. In particular, the experimental permeability data of CO₂/CH₄ mixtures in conventional as well as in innovative glassy membranes have been examined, and described by the present model, with no need of any additional parameters with respect to the pure gas case. The NET-GP approach can be properly employed to represent the multicomponent solubility in glassy polymers, as already proved for various systems [6], and the transport model accounts for the presence of a second component through its concentration dependence of the diffusion coefficient. The model is able to describe well the experimental data for all the systems investigated, and the permeability behaviors of both penetrant species are accurately represented in a wide feed pressure ranges.

Interestingly, the model is able to highlight completive sorption phenomena among the two penetrants, and the effects produced on both solubility and diffusivity coefficients. Furthermore, polymer dilation induced by CO₂ (or other swelling agents) can be simply identified, and its plasticization effect in case of real mixtures is quantitatively identified.

References:

[1] B. Freeman, Y. Yampolkii, Membrane Gas Separation John Wiley and Sons, 2010

[2] P. Bernardo, E. Drioli, Ind. Eng. Chem. Res. 48 (2009) 4638.

[3] W.J. Koros et al., J. Poly. Sci. B: Polym. Phys. 19 (1981) 1513.

[4] M. Minelli, G.C. Sarti, J. Membr. Sci. 435 (2013) 176.

[5] F. Doghieri, G.C. Sarti, Macromolecules 29 (1996) 7885.

[6] M. Minelli et al., Macromolecules 44 (2011) 4852.