The dynamics of polymers at the macros-scale using rheological properties has been long investigated and, to a large extent, is now well understood. On the other hand there is significant interest in the dynamics of polymers at the nanometer size scale where there are few methods of obtaining the molecular dynamic information, especially for nano-meter thick polymer films. This is a problem of importance because of the observation that the glass transition temperature can decrease signficantly in ultrathin polymer films. Furthermore, there is interest in the surface dynamics because some works consider that the surface of the nanometer thick films contributes substantially to, or is even the cause of, the observed reductions in the glass transition temperature. Yet, direct measurements of the surface dynamics are extremely difficult to obtain. For the most part, the works relating to thin films and surfaces have been obtained by indirect or what we refer to as "pseudo-thermodynamic" measurements. In the present work we review investigations from our laboratories that directly measure the viscoelastic response of freely standing films using a "nano-bubble" inflation method and show that the results are not consistent with a simple free surface explanation of the behavior. Furthermore, we make a strong case that the observation of the mechanical response of the material in the "rubbery plateau" regime, is related to the segmental relaxation dynamics in the macroscopic state as suggested by Ngai et al in a recent work. We also find that the behavior is related to the macroscopic dynamic fragility index m. These results will be discussed in detail.
For the surface dynamics, we have expanded upon the particle embedment method originally proposed by Forrest's group and use it to examine the behavior of stacked polymer films in a way that addresses gradients in the glass transition that are similar to the pseudo-thermodynamic measurements made by Torkelson's group. In our work, we see little change in the glass transition temperature determined from silica nanoparticle embedment kinetics when very thin films of polystyrene are placed on different thickness polystyrene substrates overlaying a Silica surface. This differs quantitatively from the findings of Torkelson's group and suggests that pseudo-thermodynamic measurements are probing a different aspect of the confined material behavior (surface or gradient effects) than is the more direct particle embedment method. We relate these results to other findings in the literature.
Finally, we have built upon the film dewetting from a liquid method developed by Bodiguel and Fretigny to investigate further the dynamics of polymers on a mobile substrate. Here, by watching the films dewet from glycerol we are able to obtain both the conventional glass transition defined by, e.g., a 1000 s retardation time, or we can obtain a glass transition temperature using a temperature-step method that allows for more rapid through-put to determine the Tg vs. film thickness from measurements on a single film of a given initial thickness. Our findings from making measurements on different architecture materials show that simple surface effects are extremely difficult to reconcile with all of the observations of Tg vs. film thickness. in particular, it is important to remark that eight-arm star polystyrene materials show non-monotonic changes in Tg with decreasing film thickness, i.e., Tg initially increases before decreasing as film thickness decreases unlike most materials that seem to exhibit a monotonic decrease in Tg with decreasing thickness.