-In this work, a study on the role of the long-range term of excess Gibbs energy models in the modeling of aqueous systems containing polymers and salts is presented. Four different approaches on how to account for the presence of polymer in the long-range term were considered, and simulations were conducted considering aqueous solutions of three different salts. The analysis of water activity curves showed that, in all cases, a liquid-phase separation may be introduced by the sole presence of the polymer in the long-range term, regardless of how it is taken into account. The results lead to the conclusion that there is no single exact solution for this problem, and that any kind of approach may introduce inconsistencies.

The thermodynamic modeling of phase equilibrium in aqueous two phase systems containing the neutral polymer poly(ethylene glycol) and a salt was studied in this work. The implemented model is based on the Pitzer equation for electrolytes solutions, modified in order to account for the influence of polymer properties in both the medium dielectric constant and the medium density, values which are present in long range term of the excess Gibbs energy. The dependence of the adjustable parameters on the polymer molecule size was also investigated. Experimental data from literature were used to obtain the adjustable parameters of the model, in whose implementation the FORTRAN computer language was used. The changes introduced into the long range term resulted in a shift of the calculated equilibrium compositions, which could be observed by analyzing the deviation between calculated and experimental compositions and tie-line slopes. For some systems the performance of the modified model was superior, but in most cases the changes introduced did not result in a significant improvement in phase equilibrium calculations, and even worsened them. Specific investigations on the long range term were carried out to verify the reason for such behavior. The insertion of ternary parameters increased the correlation capacity of both models. Concerning the hypothesis that the interaction parameters can be directly related to the polymer chain size, which might eventually lead to a predictive model, it was noticed that a direct dependency holds only for larger chain sizes.

The study of protein solutions aiming at the modeling and simulation of downstream processes entails the study of non-ideality (in thermodynamic sense) of these solutions. For systems wherein the protein concentration is low-a situation often encountered in these processes-the most important technique to experimentally evaluate this non-ideality is the determination of the osmotic pressure generated by the protein in solution. Thus, the objectives of this work were: to study the influence of co-solvents on the osmotic pressure of protein solutions, to verify the absence of change in the secondary and tertiary structure of proteins in these solutions, and to thermodynamically model the obtained osmotic pressure data. The osmotic pressure was directly measured through membrane osmometry, using protein-free reference solutions with the same pH and co-solvent concentration. The data of osmotic pressure were obtained as a function of protein concentration for five different proteins (bovine serum albumin, human immunoglobulin G, ovalbumin, βlactoglobulin and lysozyme) in solution with co-solvents such as polyethylene glycol (of various chain sizes) and salts (ammonium sulfate, sodium sulfate and sodium chloride). Each data set was obtained at constant pH and co-solvent concentration. It was observed that the presence of co-solvents do shift the osmotic pressure, but this effect is dependent on the protein, the co-solvent and its concentration, and the solution pH. Measurements of fluorescence and circular dichroism of these proteins confirmed that they maintain their structure unchanged in the media, which corroborates the use of volumetric equations of state with constant parameters. The osmotic pressure data as a function of protein concentration were correlated using a osmotic equation of state comprising a repulsive term of adhesive hard spheres and a zero-order perturbation term (random approximation). The proposed model, though simple, was sufficient to properly correlate the experimental behavior.