Using molecular simulation to explore protein and colloidal phase behavior in bulk, confinement, and mixtures
Rosch, Thomas W.
MetadataShow full item record
Because of the ubiquity of colloidal solutions in everyday industrial applications such as papermaking and coatings there is a need to be able to efficiently design and manufacture these substances. A related issue concerns the connection between many physiological diseases and heath defects and the stability and phase behavior of certain proteins. It is imperative to understand the physical mechanisms that cause proteins to change their normal solution characteristics. To design colloidal solutions for specific applications as well as to produce preventative medicines and therapies an intimate knowledge of the connection between particle interactions and overall physical properties of the solution is needed. To probe this issue four types of systems are examined. In each system solution conditions are altered affecting the nature and strength of the particle interactions. Our goal is to understand the physics behind the evolution of fluid properties that occurs because of changes in microscopic interactions. The method we employ in this pursuit is grand canonical transition matrix Monte Carlo. We examine an embedded point charge protein model of lysozyme in bulk, mixed with polymer, as well as in confinement. We find that in bulk the model is able to capture qualitatively experimental trends for changes in critical temperature and evolution of the fluid phase diagram with changing solution conditions such as salt concentration and pH. Quantitatively the model predicts a relatively narrow coexistence curve compared to experimental values. It is found that the osmotic second virial coefficient remains relatively constant over a broad range of solutions conditions suggesting a universal magnitude of attraction needed to induce phase separation. We examine a simple system consisting of hard sphere colloids with added Gaussian core polymers. Decreasing the size of the polymers relative to colloids as well as increasing the energetic repulsion between polymers upon overlap results in an overall stabilization of the mixture. Unlike bulk solutions containing molecules of the Carlsson et al. lysozyme model, the osmotic second virial coefficient at the critical point for model colloid-polymer mixtures is not constant but depends on polymer size and interaction. Increasing polymer size or decreasing polymer repulsion results in a larger negative value. Overall the model fails to capture the experimental behavior of polymer excluded volume interactions because its inability to describe the polymers capability of deformation around the colloid. We extend our analysis to a mixture containing the embedded charge model for lysozyme and Gaussian core polymers. Overall, the system exhibited a strong dependence on pH and salt concentration that qualitatively followed experimental trends. Increase of salt concentration or decrease in protein charge decreases the number of polymers needed to induce phase separation. This trend was not sensitive to the size of the polymer relative to the protein. Finally we examine the effect surface interactions have on the phase behavior for the lysozyme model as well as a simple square well model. Both systems exhibited a distinctly non-monotonic variation of its critical temperature as a function of fluid-wall interaction strength. A maximum occurs at an intermediate strength. We introduce two metrics that enable one to predict the location of this maximum. The first is related to the contact angle a fluid makes with the confining substrate while the second is based upon virial coefficient information. Because similar trends are exhibited in both systems we believe that the results should be general in nature.