The Statistical Associated Fluid Theory has been implemented for modeling of phase equilibrium for a binary mixture of polyethylene glycol (PEG) and CO2 . Three different molecular weights of PEG of higher order have been used to aid the process of particle formulation in solid phase for pharmaceutical coating purpose. The results have been compared with previously calculated experimental values for solubility using bubble pressure approach to tune the binary interaction parameter, finding out the closest fit between calculated and experimental pressure values. The phase equilibrium has been examined at a temperature range of 323-373 K and pressure range of 10-30 MPa and for PEG, the molecular weights of 1500, 4000 and 6000 g/mol have been considered. Finally symbolic regression has been performed to include the effect of temperature and molecular weight on binary interaction parameter. The selection of different functions has been analyzed based on genetic algorithm as an outer layer of the algorithm while the coefficients of those functions have been tuned using nonlinear optimization as an inner layer. It has been observed that the model is reliable in determining the polymer-supercritical phase behavior accurately, taking into account the effects of metrices like temperature and molecular weight to make it more dynamic.
