Quantum Simulations of Radiation Damage in a Molecular Polyethylene Analog

Abstract

An atomic-level understanding of radiation-induced damage in simple polymers like polyethylene is essential for determining how these chemical changes can alter the physical and mechanical properties of important technological materials such as plastics. Ensembles of quantum simulations of radiation damage in a polyethylene analog are performed using the Density Functional Tight Binding method to help bind its radiolysis and subsequent degradation as a function of radiation dose. Chemical degradation products are categorized with a graph theory approach, and occurrence rates of unsaturated carbon bond formation, crosslinking, cycle formation, chain scission reactions, and out-gassing products are computed. Statistical correlations between product pairs show significant correlations between chain scission reactions, unsaturated carbon bond formation, and out-gassing products, though these correlations decrease with increasing atom recoil energy. The results present relatively simple chemical descriptors as possible indications of network rearrangements in the middle range of excitation energies. Ultimately, the work provides a computational framework for determining the coupling between nonequilibrium chemistry in polymers and potential changes to macro-scale properties that can aid in the interpretation of future radiation damage experiments on plastic materials.

Keywords: molecular dynamics; polymer degradation; radiation damage.

Figure 1

a) Heat map of the C–C radial distribution function (RDF) of a single 70 eV PKA simulation as a function of time. The backbone chemistry reaches a steady state within several hundred fs. b) C–C RDF plots for the non‐irradiated system and for systems equilibrated after PKA excitations at different energies. Higher excitation energies result in a monotonic increase of unsaturated carbon bonds.

Figure 2

Distribution of chain reactions following excitation. The graph illustrates the probabilistic outcomes of an excited polymer chain, categorized into chain scission (CS), cross‐linking (CL), and no change (NC), as a function of excitation energy (eV). For each excitation energy level, the sum of CS, CL, and NC probabilities totals 100%, representing the comprehensive breakdown of possible chain reactions.

Figure 3

Categories of chemical products found in our PKA simulations, with carbon atoms colored in teal and hydrogen atoms in white.

Figure 4

Histogram of the average rate of occurrence of each product class per simulation as a function of PKA energy. Error bars were calculated using the bootstrap method and represent a 95% confidence interval.

Figure 5

Highest conditional probabilities among the products of 30 eV PKA excitations. Pairs of products for which both P(A|B) and P(B|A) are high are shown.

Figure 6

Correlation coefficients for 30, 50, and 70 eV simulation for a subset of product pairs with high conditional probabilities.