(139a) A Group Contribution Based-Model to Estimate Free Volume in Amorphous Polymers for Solvent Diffusion

The transport of organic solvents through polymers plays a crucial role in many different types of polymer-related technological and industrial applications, such as polymerization and membrane separation. Nowadays, the widely accepted approach to describe the transport properties in polymer solutions is the free-volume theory, which emphasizes on the amount of ?empty space? as the dominant factor for diffusion. The model proposed by Vrentas and Duda is a representation of this theory, since most parameters can be obtained from pure component data and no adjustable parameters are used. However, free-volume parameters of polymers have to be determined by viscoelasticity measurements, which usually are cost and time-consuming, and then the free volume in polymers can be estimated. In order to overcome this shortcoming, our previous study has employed equations-of-state (EOSs) to estimate polymer free volume, and obtained good consistence of calculations with experiments. Even though, the characteristic parameters of polymers in the EOSs are limited only to some commonly-used polymers. From the viewpoint of design, a full predictive model is necessary without complicated parameter requirements. In this study, a group contribution-based approach has been established to estimate solvent diffusion coefficients, and the free volume of polymers becomes predictive, without knowledge of any diffusion or viscoelastic data. Predictions of solvent diffusion coefficient were generally in good agreement with experimental data. The positron annihilation lifetime spectroscopy (PALS) results have also been employed to show the reliability of the proposed method in estimating free volume.