Equivalence between polymer nanocomposites and thin polymer films: Effect of processing conditions and molecular origins of observed behavior

Recently we established a quantitative equivalence in thermomechanical properties between polystyrene-silica nanocomposites and planar freestanding polystyrene thin films. This equivalence was quantified by drawing a direct analogy between film thickness and an appropriate experimental particle spacing. Using these findings, here we unequivocally show that the glass transition process in confined geometries is controlled by the mean volume fraction of polymer that is affected by the presence of surfaces. Since separate signatures of the bulk and the surface layers are never found, we can clearly rule out any simple “two layer” model which postulates the existence of surfaces which are dynamically decoupled from the bulk. Rather, we argue that the modification of properties at the surfaces propagates into the bulk through a spatial gradient: macroscopic experimental techniques average over these gradients and yield a broadened signature relative to the bulk polymer. In a second aspect of this paper we focus on the role of processing conditions on the results obtained. We have developed a new method of processing the nanocomposites which results in a better dispersion of the nanoparticles in the matrix. However, these samples did not show the unique glass transition behavior seen in the first set of nanocomposites discussed above. This indicates that processing conditions can profoundly affect the nature of the particle-polymer interface which controls the macroscopic behavior of these important systems.