Inorganic filler material in the form of oxides of silicon, aluminum, calcium and zinc are often added to crosslinked epoxy materials both to reduce cost and to enhance selected mechanical, thermal and electrical insulating properties. In recent studies, we have focused on using atomistic simulation methods to perform accurate and precise prediction of elastic constants of polymer glasses and to examine the mechanical properties of densely-crosslinked epoxy resins as a function of the molecular architecture of the epoxy resin component . Furthermore, we have also reported preliminary studies of the mechanical properties of crosslinked epoxy interacting with a hydroxyl-coated silica substrate .
In most studies of this type, the molecular forcefields used were originally developed for studying both low molecular weight organic compounds and polymers, with inorganic material incorporated only infrequently. Thus, in some studies it has been customary simply to treat inorganic substrates as completely rigid, thereby limiting or completely avoiding the need to parameterize potential functions to describe distortion of the substrate geometry. Moreover, while other research such as our preliminary study of epoxy-coated silica have attempted to incorporate a flexible inorganic phase, the parameters employed for this phase were developed originally to describe the structure of zeolites , with little if any emphasis placed on accurate description of other properties such as elastic constants. One of the main conclusions of the preliminary study was that additional parameter refinement for the inorganic phase is required in order to enable accurate property predictions for systems containing silica, and other oxide, filler material.
The current presentation accordingly focuses on two primary topics as follows. First, we have employed software developed recently in our laboratory, which uses the VASP ab initio and LAMMPS simulation programs [4, 5] within the MedeA® simulation environment , to parameterize the inorganic substrate interactions. The approach uses the same 'class II' forcefield functional form employed and previously optimized for the organic phase , and emphasizes the ability to reproduce both structure and mechanical behavior. Since composite systems often contain primer molecules chemically bonded to both the substrate and to the epoxy, the necessary parameters for describing the coupling of the two material types are also examined. Secondly, the combined forcefield has then been applied to compute two categories of properties for pure organic and inorganic components and systems in which the two combine to form an interface. Specifically, we investigate mechanical properties of the systems in addition to the thermal conductivity, where the latter has not been explicitly taken into account during the parameterization, thereby providing a useful test of the parameterization.
 Rigby, D., Saxe, P.W., Freeman, C.M and Leblanc, B., Computational Prediction of Mechanical Properties of Glassy Polymer Blends and Thermosets, in Advanced Composites for Aerospace, Marine, and Land Applications, T. Sano, T.S. Srivatsan, and M.W. Paretti, Editors. 2014, John Wiley & sons. p. 157-171.
 Rigby, D. and Saxe, P.W., Atomistic Simulations of Structure and Mechanical Behavior of Primed and Unprimed Epoxy-Silica Interfaces, presented at 2014 AIChE Annual Meeting, Atlanta, Georgia, Nov 17, 2014.
 pcff.frc forcefield ref 10 (1992), "Parameters for Zeolites". For a general discussion of PCFF, see, Sun, H., Mumby, S.J., Maple, J.R., and Hagler, A.T., An ab Initio CFF93 All-Atom Force Field for Polycarbonates, J. Am. Chem. Soc. 116, 2978–2987 (1994).
 see, for example, Kresse, G. and Hafner, J. Ab initio molecular dynamics for liquid metals. Phys. Rev. B, 47, 558 (1993).
 Plimpton, S., Fast Parallel Algorithms for Short-Range Molecular Dynamics, J. Comp. Phys., 117, 1-19 (1995).
 MedeA: Materials Exploration and Design Analysis, © 1998-2015, Materials Design, Inc.
 PCFF+ Force Field, Materials Design, Inc., in preparation.
See more of this Group/Topical: Computational Molecular Science and Engineering Forum