Glass transition temperature of a polymer thin film: statistical and fitting uncertainties

The polymer glass transition is an important property in a wide variety of applications. The glass transition temperature of a polymer composite or confined thin film can be significantly different to the pure polymer. Molecular dynamics simulations are useful for providing molecular level insight and prediction, particularly at interfaces, that are not easily observable experimentally. However, there are significant methodological uncertainties in calculating the polymer glass transition temperature using molecular dynamics simulations. In this work we investigate how the cooling method, fitting range and statistical variation affects the calculated glass transition temperature of polyethylene. We found that it is necessary to perform multiple independent simulations to obtain statistically significant results, and that appropriate fitting ranges must be chosen. The methodological findings were used to investigate the difference in glass transition temperature between pure polyethylene and a polyethylene film confined between graphene surfaces. It was found that the glass transition temperature of a 9 nm thick confined film was higher than bulk polyethylene by approximately 15 K.