Described here is a QSPR equation for calculating glass transition temperatures for acrylate and methacrylate polymers, especially those with bulky ester substituents. This approach is based on molecular mechanics calculations and exclusively involves a force field to describe a particular polymer system; i.e., no group additivity values are required. Results from two different force fields yielded similar results, indicating that this model is not dependent on a particular force field parameter set but rather on the atomic properties that the force field describes. The molecular mechanics calculation results (energy term), the repeat unit mass, and a measure of the volume surrounding the polymer segment (TSSV) were used to determine an energy density function that is related to experimental T g values. This energy density function is important because it illustrates that the glass transition temperature of an amorphous polymer is related not only to the volume surrounding the polymer segment but also to its conformational energy. Limitations of other QSPR approaches (stemming from not having a particular group or bond connectivity described within the given model) are not present in this approach.

No abstract

We have previously described an original model called the EVM (energy, volume, and mass) model, which uses only three descriptors, the energy of a polymer segment conformation using molecular mechanics and molecular dynamics, its volume (the occupied space by the atoms as well the unoccupied space between them), and the repeat unit molar mass, to calculate successfully the glass transition temperatures (T g) of aliphatic acrylate and methacrylate polymers. We report here the application of this model to a series of various polystyrenes. The EVM model allows correct description of the substituent position effect on the ring as well as on the backbone, for various alkyl group and halides. The estimated T g values with the EVM model are in excellent agreement with literature values. We compare this model with two other methods (Bicerano's model and Porter's approach) and find better or comparable correlations.

No abstract

No abstract