A Quantitative Exploration of the Effect of Interfacial Phenomena on the Thermomechanical Properties of Polymer Nanocomposites

Polymer nanocomposites (PNC) are complex material systems in which the prevailing length scales, i.e., the particle size, radii of gyration of the polymer and the interparticle spacing, converge. This convergence leads to an increased dominance of the interface polymer over bulk properties, when compared to conventional "microcomposites". The development of fascinating nanoscopic filler materials (C60, nanotubes, graphene, quantum dots) along with this potential gain in interfacial area has fueled the expansion of PNCs. Nanocomposites literature has demonstrated a myriad of potential chemistries and self assembled structures that could significantly impact a diverse range of applications. However, most noteworthy results in this field are serendipitous and/or are outcomes of resource-intensive "trial and error" experiments supplemented by intuition. Intuition suggests, qualitatively, that the properties of PNCs depend on the individual properties of the participating species, the interphase and the spatial distribution of filler particles. However, the individual roles of these parameters are difficult to identify, since they are interrelated due to their co-dependence on the chemical constitution of the filler and matrix. A quantitative unifying picture is yet to emerge and the commercialization of this material class has been severely hampered by the lack of design rules and structure-property constitutive relationships that would aid in the prediction of bulk properties. In this thesis, a quantitative understanding of interfacial phenomena was sought and structure-property relationships between the filler/matrix interface chemistry and the dispersion and thermomechanical properties of PNCs were obtained by systematic experiments on 2 distinct kinds of nanocomposite systems (a) Enthalpic short silane modified fillers and (b) Entropic long polymer chain grafted filler embedded PNCs. In order to quantitatively understand the role of enthalpic compatibility, an array of hybrid systems spanning a wide range of interfacial interactions, were studied. Four different matrices of surface energies varying from polar to non polar (Poly(2-VinylPyridine) (P2VP), Polymethylmethacrylate (PMMA), Polyethylmethacrylate (PEMA) and Polystyrene (PS)), were filled with three monofunctional-silane modifications (Octyldimethylmethoxysilane, Chloropropyldimethylethoxysilane and Aminopropyldimethylethoxysilane) of colloidal silica nanospheres (14 nm). Dispersion was observed to have a strong and moderate dependence on the ratio of the work of adhesion between filler and polymer to the work of adhesion of filler to filler (WPF/WFF) and the relative work of adhesion (deltaWa) respectively. Work of spreading (Ws) in conjunction with the dispersion parameters, was shown to dictate the change in Tg. These correlations led to the development of the first unifying empirical model linking energetics to bulk Tg change. In order to further illustrate the effectiveness of such a unifying model in property prediction, a heuristic informatics-based approach was developed. Materials Quantitative Structure Property Relationships (MQSPR) coupled with physics-based continuum models and experimental validation were demonstrated as being effective in predicting meso-scale morphologies, macroscale Tg and full viscoelastic spectra under quasi-equilibrium processing conditions, based purely on filler and matrix compositions. This protocol has far reaching implications for the virtual design of polymer nanocomposites and other material systems. To understand the role of enthalpic and entropic effects in grafted particle systems, the viscoelastic properties and microstructures of monomodal and novel bimodal PS brush-graft 14 nm silica filled PS composites were studied, by systematically varying loadings, graft densities and matrix molecular weights. The combination of a high graft density short brush and a low graft density long brush was observed to cause enhanced dispersions when compared to single populations of long and short brushes of corresponding graft densities and molecular weights. This enhanced dispersion affected enhanced glassy state properties when compared to monomodal grafted particles with similar wetting behavior. A new quantitative model was developed to understand these results and was found capable of predicting dispersions in grafted nanoparticle composites in the "allophobic dewetting" regime. The bimodal-brush-graft particles were also found to be remarkably well dispersed in an entropically unfavorable higher molecular weight matrix. This facilitated, a study of the role of brush / matrix entanglement behavior on thermomechanical properties of PNCs, isolated from the effects of particle dispersion. The best enhancements in glassy properties result from improved polymer brush / matrix entanglement, attained by lowering the long chain graft density and increasing the long chain to matrix molecular weight ratio.