Aidan Thompson

The final report for a Laboratory Directed Research and Development project entitled, ''From Atom-Picoseconds to Centimeter-Years in Simulation and Experiment'' is presented. In this project, separate modeling methods at the atomic scale were used to bridge gaps in time and space with higher scales. For understanding of continuum mechanics quantities at various scales atomistic simulations that ranged from nanometers to

MD simulations were used to create, deform, and postcure polymer networks to study how the stress is influenced by the coupling between crosslinking and deformation. Virtual stress and permanent set measurements were taken. Our results are consistent with the independent network hypothesis of Tobolsky. For networks crosslinked in the undeformed state, the modulus was found to increase linearly with the crosslink density as expected from rubber elasticity theory. When crosslinks were added to a uniaxially deformed network, the stress remained constant in accordance with the independent network hypothesis. Using the independent network hypothesis and the affine theory of rubber elasticity, we developed a constitutive model that accounts for the effect of the coupling between the crosslink density and strain histories of the network. The permanent set predictions from this model are in approximate (30%) agreement with the MD results.

Several experiments have indicated that the shock sensitivity of single crystal energetic materials can depend on the crystallographic direction. We develop a compress-and-shear modeling approach to study the mechanisms of anisotropic shock sensitivity using the ReaxFF reactive molecular dynamics. ReaxFF is a first-principles based force field capable to reproduce the quantum chemical energies of the reactants, products, intermediates and transition states with functional forms suitable for large-scale molecular dynamics simulations of chemical reactions under extreme conditions. In this presentation we will discuss the results of high-rate shear simulations of uniaxially compressed PETN. We found noticeable differences in the physical and chemical responses of PETN for different combinations of the slip system and compression direction. The simulation results agree well with the experimental shock-initiation sensitivity data and Dick's steric hindrance theory.

ABSTRACT Building the next-generation of extreme-scale distributed systems will require overcoming several challenges related to system resilience. As the number of processors in these systems grow, the failure rate increases proportionally. One of the most common sources of failure in large-scale systems is memory. In this paper, we propose a novel runtime for transparently exploiting memory content similarity to improve system resilience by reducing the rate at which memory errors lead to node failure. We evaluate the viability of this approach by examining memory snapshots collected from eight high-performance computing (HPC) applications and two important HPC operating systems. Based on the characteristics of the similarity uncovered, we conclude that our proposed approach shows promise for addressing system resilience in large-scale systems.

ABSTRACT Building the next-generation of extreme-scale distributed systems will require overcoming several challenges related to system resilience. As the number of processors in these systems grows, the failure rate increases proportionally. One of the most common sources of failure in large-scale systems is memory errors. In this paper, we propose a novel run-time for transparently exploiting memory content similarity to improve system resilience by reducing the rate at which memory errors lead to node failure. We evaluate the feasibility of this approach by examining memory snapshots collected from eight HPC applications. Based on the characteristics of the similarity that we uncover in these applications, we conclude that our proposed approach shows promise for addressing system resilience in large-scale systems.

This report summarizes materials issues associated with advanced micromachines development at Sandia. The intent of this report is to provide a perspective on the scope of the issues and suggest future technical directions, with a focus on computational materials science. ...

ABSTRACT Hydrocarbon foams are versatile materials extensively used in high energy-density physics (HEDP) experiments. However, little data exist above 100 GPa, where knowledge of the behavior is particularly important for designing, analyzing, and optimizing HEDP experiments. The complex internal structure and properties of foam call for a multi-scale modeling effort validated by experimental data. We present results from experiments, classical molecular dynamics simulations, and mesoscale hydrodynamic modeling of poly(4-methyl-1-pentene) (PMP) foams under strong shock compression. Experiments conducted using the Z-machine at Sandia National Laboratories shock compress ∼0.300 g/cm3 density PMP foams to 185 GPa. Molecular dynamics (MD) simulations model shock compressed PMP foam and elucidate behavior of the heterogeneous foams at high pressures. The MD results show quantitative agreement with the experimental data, while providing additional information about local temperature and dissociation. Three-dimensional nm-scale hydrocode simulations of the foam show internal structure of pore collapse as well as provide detailed information on the foam state behind the shock front. Finally, the experimental and MD results are compared to continuum hydrodynamics simulations to assess a potential equation of state model for PMP foams to use in large scale hydrodynamics simulations.

ABSTRACT We report an approach to large-scale atomistic simulations of chemical initiation processes in shocked energetic materials based on parallel implementation of the ReaxFF reactive force field. Here, we present results of reactive molecular dynamics (MD) simulations of shocked Pentaerythritol Tetranitrate (PETN) single crystal, a conventional high explosive. We study a planar wall impact to compare mechanical and chemical response at different speeds. The dominant initiation reactions in both systems lead to the formation of NO2. The lagging secondary reactions lead to a formation of water, nitrogen, and other products. By tracking the position of the shock front as a function of time, we have been able to observe how the shock velocity changes in response to the storage and release of chemical energy behind the shock front. We also investigate the effect of shear along different slip systems on chemical initiation. All calculations are performed with massively parallel MD code GRASP enabling multi-million atom reactive MD simulations of chemical processes in many important stockpile materials.

Density functional theory (DFT) molecular dynamics (MD) and classical MD simulations of the principal shock Hugoniot are presented for two hydrocarbon polymers, polyethylene (PE) and poly(4-methyl-1-pentene) (PMP). DFT results are in excellent agreement with ...