ABSTRACT
Solidification by the formation of a glass below the glass transition temperature Tg is one of the most important material properties for the processing and applications of polymeric systems. In particular, Tg can be strongly altered in ultrathin polymer films, and there are many reviews emphasizing experimental aspects of the dynamics of thin polymer films [1-5]. While much progress has been made toward understanding dynamical changes in thin films, a theoretical framework to describe and predict the effects of confinement on glass formation remains a significant scientific challenge. A substantial body of research has established that glassforming (GF) liquids in general are “dynamically heterogeneous” [6-8], exhibiting both spatial and temporal fluctuations in local mobility. The notion of heterogeneity, in the form of cooperative rearragement, underlies the Adam-Gibbs (AG) theory [9] and the conceptually related random first-order transition theory (RFOT)
8.1 Introduction 267
8.2 Influence of film thickness and substrate properties on glass formation 269
8.2.1 Summary of Tg and fragility changes 269
8.2.2 Gibbs-Thomson-inspired model for the thickness dependence of Tg 270
8.2.3 Local structure and dynamics of supported films 272
8.3 String-like collective rearrangements as an organizing principle for dynamics 278
8.3.1 String-like collective motion in thin films 278
8.3.2 Adam-Gibbs-inspired string model of relaxation 280
8.3.3 Relationship of the enthalpy and entropy of activation 283
8.4 Living polymerization model of strings 285
8.4.1 Theoretical formulation 286
8.4.2 Validation of the living polymerization model 287
8.4.3 Predictions for glassy polymer films 289
8.5 Relation between the scale of interfacial mobility and string size 290
8.6 Conclusions and prospects 293
8.7 Appendix: Modeling and simulation 294
References 296
[10,11], and these theories provide a useful framework for understanding the dynamics of GF liquids in general. However, both the AG and RFOT models involve many heuristic assumptions that require validation by either simulations or measurements. In this regard, simulations can play a leading role, since fluid properties such as the scale of cooperative rearrangements and the configurational entropy are not readily estimated from experiments. In particular, the actual form of cooperative motion in glass-forming liquids, the hypothetical “cooperatively rearranging regions” (CRR) of AG, or the “entropic droplets” of RFOT, has been an object of much speculation. The central goal of the present review is to develop and investigate a model for glass formation in which the CRR are quantified by string-like cooperative motions [8-23]. Specifically, we consider the ability of this “string model” to rationalize the diverse changes to glass formation in thin supported polymer films. We systematically explore the influence of film thickness, polymer-substrate interaction strength, and substrate stiffness, since all of these parameters can significantly alter thin film dynamics.
