Abstract
The aim of molecular modeling is to mimic reality by considering approximations appropriate to the scale at which the simulation is carried out. At the atomic level, forcefields that represent average atomic interactions are used. However, the phase space has to be adequately explored in order to compare successfully computed and experimental properties. The procedure exposed in this article considers an initial selection of relevant configurations on which a simulated annealing process is applied using the first generation forcefield OPLS, followed by a uniform hydrostatic compression using the second generation forcefield COMPASS©. The resulting data are fitted by an equation of state, from which density and bulk modulus are determined. The glass transition is then simulated and T gs are computed. Our approach is tested using a series of vinylic polymers, which differ from each other by small variations in atomic interaction combinations. The excellent agreement with experimental data shows the validity of the procedure exposed. Moreover, a clear linear relationship between simulated and experimental T gs is revealed.