Our group has previously developed simple fluorescence-based methods to characterize how polymer-substrate interfaces, free surfaces (polymer-air interfaces), and nanoscale confinement itself can alter the glass transition temperature (Tg) and physical aging (relaxation toward equilibrium in the non-equilibrium glassy state) of ultrathin polymeric films from their bulk response. Here, we extend these studies to cases in which there is either covalent attachment between polymer and nanofiller or attractive, hydrogen-bonding interactions between polymer and nanofiller. Using a 'grafting to' approach involving simple condensation-type reactions, we produce 99/1 wt% poly(methyl methacrylate) (PMMA)/carbon nanotube (CNT) nanocomposites in which there is effective covalent bonding between PMMA and the CNT surface. Depending on the size and flexibility of the unit employed in attaching the ester side group of PMMA to the CNT, the PMMA Tg can be enhanced by 8-32 K relative to neat PMMA. Accompanying these increases in Tg are major reductions in physical aging. Based on short-term physical aging monitored via fluorescence, we find that the physical aging (defined on a rate of change relative to logarithmic rather than linear time) can be reduced by up to an order of magnitude or more by the covalent attachment of polymer to the CNT. This dramatic reduction in physical aging rate is further supported by long-term physical aging studies conducted via enthalpy relaxation measurements. For example, in the 99/1 wt% PMMA-CNT sample with the longest, most flexible linking unit between the polymer and CNT and thereby exhibiting the least effect on Tg and physical aging rate, we observed the same amount of enthalpy relaxation in the nanocomposite after 100 days of physical aging at room temperature as we observe after only 4 days of physical aging at room temperature with neat PMMA. This suggests that a potentially important application of polymer nanocomposites can be in producing glassy polymeric materials which exhibit little or no physical aging during in-use application.

We have obtained related Tg and physical aging data in polymer nanocomposites in which polystyrene is covalently attached at low levels to silica nanofiller and in cases in which silica or alumina nanofillers undergo attractive hydrogen bonding interactions with the polymer (PMMA or poly(2-vinyl pyridine). In contrast, polymer nanocomposites that have wetted interfaces with only van der Waals interactions between polymer and nanofiller (e.g. PS/silica nanocomposites with no covalent bonding) exhibit Tg values and physical aging rates that are within error identical to that of the neat polymer. The underlying causes of the effects of nanoconfinement and interactions on Tg and physical aging will be discussed.