Session S17: Matter in Extreme Environments: Theoretical Methods and Applications III

Many-body molecular dynamics force fields for chemistry at extreme conditions

Molecular materials at high pressures and temperatures, such as those produced by a shock wave or detonation, often undergo rapid chemical reactions. There has been significant progress in understanding chemical processes at extreme conditions through simulations based on Kohn-Sham density function theory (DFT). DFT simulations, however, are typically limited in size to less than 1000 atoms and in time to less than 100 ps. We have developed a reactive molecular dynamics force field, called ChIMES, that retains much of the accuracy of density functional theory, while allowing application to much larger systems and longer time scales. The ChIMES approach has been recently extended to treat 4 body interactions and charge fluctuations. We find that these additions significantly enhance the accuracy of the force field, but require more sophisticated calibration methods to maintain accuracy and stability for a wide range of conditions. Effective calibration methods will be presented based on techniques adapted from the machine learning community. Applications to liquid carbon, H₂O, and N₃H will be discussed.
  

Modeling High Strain Rate Plasticity in BCC Lead

High-energy lasers have enabled the determination of constitutive properties of metals at very high pressures and strain rates. Here we consider the strength (flow stress) of lead in the high-pressure body-centered cubic (bcc) phase. There were two models of high-strain-rate lead strength available previously. Both models were constructed using data from the low-pressure, face-centered cubic phase of lead. Plasticity in bcc and fcc crystals can be very different. Experiments conducted at the National Ignition Facility have used ramp-compression to drive Rayleigh-Taylor instability and measured the ripple growth to infer lead strength in the bcc phase. We have developed an Improved Steinberg-Guinan model for bcc lead strength [1] using ab initio calculations of the shear modulus at pressure. We compare the predictions of the model with those from the two previous models and results from experiment. We also discuss the effect of alloying.   

Role of “soft” wall confinement in the particle dynamics of supercritical fluids

Supercritical fluids (SCF), having a heterogeneous phase behavior ("liquid-like" and "gas-like") across the Frenkel line (FL) [1] , seem to be a promising candidate for the study of the effect of wall softness on the particle dynamics. Recently, structural features of SCF have been found to vary significantly as the walls became “softer”[2].

Molecular dynamics simulations are carried out to understand the effect of softness of the walls on the particle dynamics of a supercritical fluid using mean-squared displacement (MSD) and the velocity autocorrelation function (VACF). Walls are systematically made “softer” by lowering the stiffness coefficient of the springs attached to each of the wall particles. The existence of non-diffusive modes, and the effect of wall softness on it, is investigated through the vibrational density of states (DoS). We discuss the contrasting trends observed in self-diffusion parallel to the walls, on the either side of the FL, as a function of the wall-stiffness. The effect of the “softness” of the walls to support or disrupt collective motion will also be discussed.


[1] V. V. Brazhkin, et al., Phys. Rev. Lett., 111, 145901, 2013.
[2] K. Ghosh, et al., Phys. Rev. E., 97, 012131, 2018.   

Maximum initial growth-rate of strong-shock-driven Richtmyer-Meshkov instability

We focus on the classical problem of the dependence on the initial conditions of the initial growth-rate of strong shock driven Richtmyer-Meshkov instability (RMI) by developing a novel empirical model and by employing rigorous theories and Smoothed Particle Hydrodynamics simulations to describe the simulation data with statistical confidence in a broad parameter regime. For the given values of the shock strength, fluid density ratio, and wavelength of the initial perturbation of the fluid interface, we find the maximum value of the RMI initial growth-rate, the corresponding amplitude scale of the initial perturbation, and the maximum fraction of interfacial energy. This amplitude scale is independent of the shock strength and density ratio and is characteristic quantity of RMI dynamics. We discover the exponential decay of the ratio of the initial and linear growth-rates of RMI with the initial perturbation amplitude that excellently agrees with available data.   

Water and hydrocarbon desorption from rapidly-heated metal oxide surfaces

Understanding and controlling water and hydrocarbon desorption from steel surfaces under vacuum are crucial for high-voltage pulsed power applications. Ohmic heating in Sandia’s Z-machine conductors can drive temperature rises of 1000 K over nanoseconds, leading to plasma formation and current loss. We apply reactive and non-reactive classical molecular dynamics (MD) and grand canonical Monte Carlo (GCMC) methods to study the thermodynamics and kinetics of fast desorption from hematite Fe2O3 surfaces. MD simulations are conducted using thermodynamically consistent coverages deduced from GCMC. For water, the resulting time- and temperature-dependent desorption profiles on the Fe2O3 (0001) and (1-102) surfaces show cooperative behavior. Results are in reasonable agreement with simple Temkin isotherm model estimates. Similar reduced models will be discussed for hydrocarbons desorption. The effect of external electric field on desorption profiles will also be discussed.

Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International, Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA-0003525.   

Sweeping wave impact on tantalum plate

The tensile plasticity (TEPLA) model is extended to study sweeping wave impact, where the material undergoing large deformation, pore growth, and failure. We start with the ensemble phase averaging method to derive averaged equations for inhomogeneous material. For a given time and spatial position, it only averages over the material configurations in which the given spacetime is occupied by the specified material. The averaging method leads to the decomposition of the velocity gradient into intrinsic part representing the gradient experienced by the solid material and the plastic part caused by deformations related to slipping planes, void growth, and sliding on the crack surfaces.

To compute large plastic deformation, while accurately tracking the material history Dual Domain Material Point (DDMP) method is employed. The DDMP method uses both Eulerian and Lagrangian descriptions. We use the Eulerian capability to compute the large deformation of the solid caused by a swiping shock wave and the Lagrangian capability to track history dependent plastic pore growth. Numerical results are compared to an experiment to show the needed improvements on both the physical models and numerical techniques.   

Effects of alloying elements on defect production and evolution in Fe-based alloys

Fe-based alloys are important structural and cladding materials for current and future nuclear reactors. Radiation-induced defect production and subsequent defect evolution play important roles in microstructural evolution in these alloys. In this work, the effects of alloying elements on defect production and evolution in Fe-based alloys have been investigated using molecular dynamics simulations. Primary damage simulations have been conducted for pure Fe and Fe-based alloys, including Fe-Cr, Fe-Cu, and Fe-W. It is found that Cr or Cu do not affect the total number of produced Frenkel pairs, while oversized W increases the total number of Frenkel pairs. In addition, Cr interstitials are over-produced while Cu and W interstitials are under-produced comparing to their solute concentrations. Both dislocation loops and C-15 clusters have been found in these alloys but their population is affected by the alloying elements. Interstitial cluster evolution is studied in these alloys. It is found that Cr and W suppress the growth of dislocation loops. Finally, the defect energetics are calculated to interpret these simulation results.   

Dynamic recrystallization in adiabatic shear banding: an entropic, effective-temperature model

Dynamic recrystallization (DRX) is often observed in conjunction with adiabatic shear banding (ASB) in polycrystalline materials. The recrystallized nanograins in the shear band have few dislocations compared to the material outside of the shear band. We reformulate the recently developed Langer-Bouchbinder-Lookman (LBL) continuum theory of polycrystalline plasticity and include the creation of grain boundaries. While the shear-banding instability emerges because thermal heating is faster than heat dissipation, recrystallization is interpreted as an entropic effect arising from the competition between dislocation creation and grain boundary formation. We show that our theory closely matches recent experimental results in sheared ultrafine-grained titanium and in compressed AISI 316L stainless steel. The theory thus provides a thermodynamically consistent way to systematically describe the formation of shear bands and recrystallized grains therein.   

Crack deflection-penetration at a nanoscale interface of graphene/hBN heterostructure

We investigate crack deflection-penetration behavior at an interface of two elastically dissimilar 2D nanomaterial taking an interface between graphene and hexagonal boron nitride (hBN) as an example test case. Growth of an edge crack from graphene towards hBN and similar growth of crack from hBN towards graphene are considered in two separate cases. Crack deflection-penetration (DP) criterion for continuum study is redefined in terms of cohesive energies of individual solids in this MD study. In case of graphene/hBN heterostructure, crack typically prefers to penetrate either material over deflection when it reaches the interface. Ratios of cohesive energies of interface and material ahead of the crack are 0.67 (when crack in graphene) and 1.0 (when crack in hBN). As these ratios lie in the crack penetration zone of the DP criterion, crack penetration is the only logical choice. However, in some cases slight deflection has been observed. This phenomenon occurs when crack bifurcates into several branches and the mother crack interacts with defects formed at the interface due to lattice constant mismatch.This initial study indicates that at nanoscale crack growth, DP criterion developed for continuum studies cannot be directly applied due to the role of crystallography itself.   

Assessing the Impact of Strength Model Parameters on Simulated Cerium Flyer Plate Behavior

Cerium flyer plate experiments were simulated using the LANL Lagrangian hydrodynamics code FLAG and results were compared to experimental results. The range of experiments examined includes shocking through the cerium γ-α phase transition. Several sets of parameters for the Preston-Tonks-Wallace strength model have been produced through fits to experimental Hopkinson bar data. We will present simulation results showing the effect of varying the strength parameter sets including simulations with no strength. This process may aid in differentiating the various parameter sets by testing them under conditions different from the calibration experiments.   

Update on the Development of a Multiscale Friction Model

Two sets of experiments designed to produce dry sliding of metal-on-metal resulting in normal pressures up to tens of GPa and sliding velocities up to hundreds of meters per second are simulated numerically for the purpose of evaluating a new multiscale friction model. These experiments involve the impact of a cylindrical copper flyer onto a composite cylindrical target composed of an aluminum inner core and a stainless steel circumferential confinement. The primary diagnostic in
these experiments is a measurement of free-surface velocity. Numerical simulations are conducted using the Los Alamos finite volume continuum mechanics code FLAG. The primary metric of evaluation is a comparison of predicted free-surface velocities to those measured. The importance of accounting for friction in the simulation of these experiments is clearly demonstrated. It is shown that the FLAG implementation of the new multiscale friction model provides capabilities that are essential in the modeling of dry sliding friction.   

Uncertainty Quantification of Velocimetry-Based Load Current Inferences for 100 ns Multi-Mega-Amp Pulsed Power Experiments

Accurate determination of the electric current delivered to the load region is critical to the success of various multi-mega-amp pulsed power experiments. Magnetic inductance (B-dot) probes often fail over ten mega-amps, and when successful can only infer current centimeters from the load region, before major current loss mechanisms kick in. This has prompted interest in velocimetry-based load current inferences. The present work is part of a systematic uncertainty quantification effort. Here we apply Bayesian statistics to a series of synthetic experiments. Tens of thousands of multiphysics simulations were run with perturbed drive currents to determine correlations between input current and output surface velocity (our experimental measurable). We find that for very short pulses (100 ns rise time) relevant to cylindrical platforms significant uncertainties, not typical of planar dynamic materials experiments, can be introduced around peak current if the velocimetry measurement is not well matched to the current pulse. All regions of enhanced uncertainty are explained physically, and compensating strategies are developed in order to increase confidence and certainty in load current inferences.   

Numerical modeling of the phase transition kinetics for the sub-microsecond solidification of water under dynamic compression

Several landmark experimental studies on the solidification of liquid water to the high-pressure ice VII phase under multiple-shock and ramp dynamic compression have been carried out over the past two decades, yet modeling this rapid phase transition has proven challenging. The application of classical nucleation theory (CNT)-based approaches to rapid phase transition kinetics occurring under extreme temperatures and pressures presents a variety of new opportunities for predictive computational modeling. This work attempts to model the liquid water-ice VII phase transformation using a numerical discretization scheme to solve the Zel’dovich-Frenkel partial differential equation, an underlying CNT-based kinetic equation describing the statistical time-dependent behavior of solid cluster formation. One major result of this research is that the Zel’dovich-Frenkel equation is able to accurately determine---without the need for empirical scaling parameters---the duration of the induction time prior to the onset of the phase transformation.   

Li-ion battery material under high pressure: amorphization and enhanced conductivity of Li4Ti5O12

Li₄Ti₅O₁₂ (LTO), a “zero-strain” anode material for lithium-ion batteries, exhibits excellent cycling performance. However, its poor conductivity highly limits its applications. Here, the structural stability and conductivity of LTO were studied using in situ high-pressure measurements and first-principles calculations. LTO underwent a pressure-induced amorphization (PIA) at 26.9 GPa. The impedance spectroscopy revealed that the conductivity of LTO improved significantly after amorphization compared with its starting phase. Furthermore, our calculations demonstrated that the different compressibility of the LiO₆ and TiO₆ octahedra in the structure was crucial for the pressure-induced amorphization. The amorphous phase promotes Li⁺ diffusion and enhances its ionic conductivity by providing defects for ion migration. Our results not only provide an insight into the pressure depended structural properties of a spinel-like material but also facilitate exploration of the interplay between pressure-induced amorphization and conductivity.