(561f) A Multiscale Strategy for Predicting Radiation Damage in Polymers

Polymers are routinely subjected to ionizing radiation as a means for sterilization, as part of planned operation conditions, and as a driver to accelerate aging. Chemical reactions resulting from this exposure can alter polymer networks, leading to undesirable macroscale degradation such as permanent set. A primary mode for radiation damage arises from ballistic electrons that induce electronic excitations, but subsequent chemical mechanisms are poorly understood. We develop a multiscale modeling strategy to predict this chemistry starting from subatomic scattering calculations. Ensembles of nonadiabatic molecular dynamics simulations based on time-dependent density functional theory are used to sample initial bond-breaking events following the most likely excitations. These excited state configurations in turn feed into semiempirical quantum-based simulations of the approach towards chemical equilibrium. Application of our approach to polyethylene shows that local backbone conformation plays a significant role in the initial steps of radiolysis, providing a plausible explanation for experimental observations of a morphology dependence in network crosslinking. Statistical and graph-analysis techniques are described that cast quantum simulations as an easy-to-implement test of degradative chemical reaction schemes.

This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. Approved for unlimited release, LLNL-ABS-833130.