Evidence for Two Mechanisms Driving Molecular Weight Dependence of the Glass Transition Temperature in Linear Polymers

Recently, the canonical understanding for the relationship between molecular weight, M, and the glass transition temperature, Tg, has become the focus of renewed attention. Several studies in the past decade have shown that the 70-year-old functional forms used to describe this trend are at best first-order approximations and do not apply to all polymers across the full domain of M (spanning from the monomer to the infinite linear chain). This work analyzes monomer-resolution dynamics of a diverse trio of polymers from molecular dynamics which adds to this growing body of evidence that Fox-Flory-like arguments of chain end effects are not the leading cause of the decrease of Tg with smaller M. To the contrary, we see that chain end effects are negligible in a fully-flexible polymer, and chain end effects present in stiff polymers surprisingly become weaker on cooling towards Tg. Our data suggests M effects on Tg can be separated into two components, wherein enhanced chain end mobility is chemistry-dependent and a separate whole-chain effect is present in all models. Continued work on this fundamental relationship may further guide the long-standing effort for a comprehensive theory describing the glass transition.