Improved methods for the determination of Hansen's solubility parameters and the estimation of solvent uptake for lightly crosslinked polymers

Two new methods for determining Hansen's three-dimensional solubility parameters (3DSP) for lightly crosslinked polymers are presented and evaluated using commercial glove materials composed of butyl, natural, nitrile, and neoprene rubber. In the first method, the polymer 3DSPs are determined from a weighted average of the 3DSPs of a range of organic solvents, where the weighting factor is the product of the solvent molar volume and the fractional uptake of the solvent measured by immersion testing. The resultant polymer 3DSPs were used to model solvent uptake via the Hildebrand-Scott and modified Flory-Rehner expressions for the Flory interaction parameter. After grouping solvents of similar structure and applying an additional weighting factor, estimates of equilibrium solubility within a factor of two of experimental values were obtained for 176 of the 212 (83%) solvent-polymer combinations examined. In the second method of estimating polymer 3DSPs, the Flory interaction parameters for all solvent-polymer pairs are determined at the outset via the modified Flory-Rehner equation and used to solve for the polymer 3DSP values by multiple nonlinear regression. Solubility estimates are then back-calculated. This method provided different values of the 3DSPs for the polymers, including negative delta(p) values for the butyl and natural rubber. However, the accuracy of solubility estimation was comparable to that obtained using the weighted-average method without the need for any adjustable factors. The advantages of these alternative methods over the graphical method for obtaining reliable estimates of the 3DSPs of lightly crosslinked polymers are discussed. This is the first demonstration of a general 3DSP-based approach to quantitative solubility modeling in polymer-solvent systems. (C) 1996 John Wiley & Sons, Inc.