With the increasing commercial applications of solutions containing surfactants and/or polymer blends, a great deal of attention has been focused on developing techniques for predicting and characterizing the miscibility of various components within these systems. In this thesis, both experimental and computer simulation methods have been developed to address the issues of the solubility properties of non-ionic surfactants and miscibility of polymer blends.

As part of this thesis, an experimental technique, based on inverse gas chromatography (IGC), has been developed to measure the three-dimensional solubility parameters of materials with low vapour pressures. This technique was applied to determine the interaction parameters of common non-ionic surfactants, nonyl phenol ethoxylates, with a series of different vapour phase probe molecules at two different temperatures. Assignments of the three-dimensional solubility parameters of these surfactants were made based upon the measured interaction parameters and known three-dimensional solubility parameters of the probe molecules.

On the other hand, we have also applied molecular simulation techniques based upon molecular mechanics models coupled with molecular dynamics to compute the three-dimensional solubility parameters of nonyl phenol ethoxylates and two types of cellulosic. In general, the predicted three-dimensional solubility parameters are in good agreement with those values derived from group contributions, mutual solvent and IGC methods with the exception of electrostatic interactions.

Both techniques seem to provide simple, fast, reliable and objective alternatives to estimate the three-dimensional solubility parameters of various non-ionic surfactant systems. As a result, we expect that when these techniques are used together, they can be further applied to characterizing the solubility properties of other classes of materials.

Another major focus of this thesis is to develop a phenomenological theoretical approach to describe the phase behaviours of polymer blends. In particular, we have employed the molecular mechanics/dynamics techniques to investigate the molecular morphology as well as to calculate the heats of mixing and Flory-Huggins interaction parameters of polyethylene-polypropylene mixtures in the melt state. At temperatures well above the crystal melting points of the homopolymers, local ordered structures were observed and polyethylene and polypropylene were found to be phase separated. Furthermore, it seems that the energy cost of disrupting the order domains formed by these two polymers, especially polyethylene, is relatively high and that mixing on the molecular scale is very unlikely. Based on these simulations, we speculate that the immiscibility of these two non-polar polymers, in the melt state, may be due to differences in local morphologies rather than to their specific interaction energies.

We expect that this approach will eventually become a routine procedure to estimate the phase behaviours of polymer blends with two or more components as powerful workstations and molecular simulation software become available.