(478c) A Critical Evaluation of the Accuracy of Thermoset Models Created Using Atomistic Simulation Methods, Based on Gelation Studies

In recent years, a steadily increasing number of molecular modeling studies have focused on prediction of mechanical and thermophysical property prediction for thermosetting polymeric materials, driven in large part by incorporation of such materials in advanced composites used in aerospace, automotive and electronics applications. Although comparison of property predictions has shown encouraging results (e.g. the successful and quantitative prediction of the effect of epoxy resin precursor molecular architecture on small strain mechanical behavior [1]), detailed examinations of the topological and chemical accuracy of the molecular models used have thus far been limited.

Since model accuracy must ultimately determine the ability of molecular simulations to predict a wide range of properties relevant to industrial applications, we have recently concentrated on extending earlier studies to include more detailed evaluation of model quality for common thermosetting materials. Rather than relying on comparisons on calculated and measured values of properties related indirectlyto network structure (e.g. as is the case with elastic constants), we instead choose to focus on the experimentally observable gel point, which has long been recognized both theoretically and experimentally to depend on chemical functionality of reactive groups and on aspects of the crosslinking process, such as cyclization, which determine when incipient infinite network formation occurs [2].

In this presentation we begin by summarizing both the experimental methods used to measure the gel point together with published experimental data for a variety of network-forming reactions, before proceeding with details of the simulations and gel point predictions obtained therefrom. The types of systems studied in the simulations include the popular amine-cured epoxy systems in addition to polyester containing materials. For selected systems, we consider possible artifacts associated with the relatively small size of the models, and other factors such as the customary neglect of less frequent secondary reactions or unequal reactivities of the various reacting groups. Finally, while for the most part the studies have focused on so-called RA_fa+R'B_fb systems formed by a step growth polymerization mechanism (with fa and fb referring to chemical functionalities of A and B groups), we will briefly consider networks formed by chain growth polymerization.

References

1. 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, eds., Ch. 14, John Wiley & Sons (2014).

2. see, for example, Flory, P.J. "Principles of Polymer Chemistry", Ch. 9, Cornell University Press, Ithaca, New York (1953).