Search

We apply the Green–Kubo (G–K) approach to obtain the thermal conductivity tensor of β‐1,3,5,7‐tetranitro‐1,3,5,7‐tetrazoctane (β‐HMX) as a function of pressure and temperature from equilibrium molecular dynamics (MD) simulations. Direct application of the G–K formula exhibits slow convergence of the integrated thermal conductivity values even for long (120 ns) accumulated simulation times. To partially mitigate this slow convergence, we apply a recently implemented numerical procedure that involves physically justified filtering of the MD‐calculated G–K heat current and fitting the integrated time‐dependent thermal conductivity to a physically motivated double‐exponential function. The thermal conductivity tensor was determined for pressures 0.1 MPa≤P≤ ${\le P\le }$ 30 GPa and temperatures 300 K≤T≤ ${\le T\le }$ 900 K. The thermal conductivity καβT,P ${{\kappa }^{\alpha \beta }\left(T,P\right)}$ increases with increasing pressure, by approximately an order of magnitude over the interval considered, and decreases with increasing temperature. The MD predictions are compared to experimental and other theoretically determined values for the thermal conductivity of β‐HMX. A simple, semi‐empirical fitting form is proposed that captures the behavior of κααT,P ${{\kappa }^{\alpha \alpha }\left(T,P\right)}$ over the pressure and temperature intervals studied. [ABSTRACT FROM AUTHOR]

We present a machine learning framework to train and validate neural networks to predict the anisotropic elastic response of a monoclinic organic molecular crystal known as β$$ \beta $$‐HMX in the geometrical nonlinear regime. A filtered molecular dynamic (MD) simulations database is used to train neural networks with a Sobolev norm that uses the stress measure and a reference configuration to deduce the elastic stored free energy functional. To improve the accuracy of the elasticity tangent predictions originating from the learned stored free energy, a transfer learning technique is used to introduce additional tangential constraints from the data while necessary conditions (e.g., strong ellipticity, crystallographic symmetry) for the correctness of the model are either introduced as additional physical constraints or incorporated in the validation tests. Assessment of the neural networks is based on (1) the accuracy with which they reproduce the bottom‐line constitutive responses predicted by MD, (2) the robustness of the models measured by detailed examination of their stability and uniqueness, and (3) the admissibility of the predicted responses with respect to mechanics principles in the finite‐deformation regime. We compare the training efficiency of the neural networks under different Sobolev constraints and assess the accuracy and robustness of the models against MD benchmarks for β$$ \beta $$‐HMX. [ABSTRACT FROM AUTHOR]

Shock‐induced collapse of an elongated pore in the energetic crystal β‐HMX (β‐1,3,5,7‐tetranitro‐1,3,5,7‐tetrazoctane) is examined using all‐atom molecular dynamics (MD) and continuum mechanics. The continuum simulation employs a recently proposed MD‐guided material model for β‐HMX. Collapse‐induced shear band formation and hotspot‐zone properties are calculated using both MD and continuum mechanics, for nearly identical simulation domains and identical impact conditions. The continuum model predicts shear band patterns and pore collapse behavior in good agreement with MD results; shear localization, plastic heating, and hydrodynamic impact‐generated temperature rise lead to geometrically complicated hotspot zones in the vicinity of the collapse site. This work demonstrates that—for the ≈10 GPa shock pressure studied—isotropic rate‐dependent Johnson‐Cook‐type elastoplastic models for HMX can provide physically consistent pore‐collapse dynamics and hotspot features in comparison to MD, for nontrivial pore shapes. Such physically accurate models are required for reliable predictions of detonation sensitivity and performance for shocked energetic crystals. Opportunities for further model improvement are identified. [ABSTRACT FROM AUTHOR]

Using shock‐induced quasi‐2D pore collapse in β‐HMX as a specific case study, we address three practical questions that arise when designing and interpreting molecular dynamics (MD) simulations of explicit shock wave propagation in crystals. How sensitive are the overall results to sample thickness perpendicular to the (quasi) plane of the problem? For impacts on a given crystal surface, how much does the transverse orientation of the sample matter? And, for a given sample size and orientation, how much run‐to‐run variability exists among results for independent but statistically equivalent realizations of the shock? The first and second questions are interrelated but pertain individually to assessing the roles of finite‐size and crystal anisotropy effects, respectively, on the simulated collapse mechanisms and associated local thermo‐mechanical states in the sample during and after collapse. The third addresses the confidence with which the results of individual simulations can be regarded as representative. All three questions become particularly important if the MD predictions are intended to serve as "ground truth" for validation of continuum mesoscale models. Here, quasi‐2D samples of (010)‐oriented single‐crystal β‐HMX (β‐1,3,5,7‐tetranitro‐1,3,5,7‐tetrazocane) containing a cylindrical pore were subjected to reverse‐ballistic impacts, resulting in explicit, supported shock propagation initially along [010], for impact speeds up =1.0 and 0.5 km s−1. The individual samples differed in either the thickness perpendicular to the sample plane or the transverse crystal orientation normal to [010] in the quasi‐2D cell. Three independent realizations of the shock were performed for a selected case to assess run‐to‐run variability. Comparisons of qualitative features of the collapse, temperature and pressure distributions in the samples, and time scales for pore collapse suggest that there is little sensitivity to sample thickness for the same crystal orientation and moderate sensitivity to transverse crystal orientation for samples of the same thickness. Run‐to‐run variability is evident to the eye in side‐by‐side system observations. However, overall mechanisms of collapse, distributions for temperature and pressure in the samples, and time scales for collapse are in near‐quantitative agreement among the realizations. [ABSTRACT FROM AUTHOR]

All‐atom molecular dynamics (MD) and Eulerian continuum simulations, performed using the LAMMPS and SCIMITAR3D codes, respectively, were used to study thermo‐mechanical aspects of the shock‐induced collapse of an initially cylindrical 50 nm diameter pore in single crystals of 1,3,5‐triamino‐2,4,6‐trinitrobenzene (TATB). Three impact speeds, 0.5 km s−1, 1.0 km s−1 and 2.0 km s−1, were used to generate the shocks. These impact conditions are expected to yield collapse mechanisms ranging from predominantly visco‐plastic to hydrodynamic. For the MD studies, three crystal orientations (relative to shock‐propagation direction) were examined that span the limiting cases with respect to the crystalline anisotropy in TATB. An isotropic constitutive model was used for the continuum simulations, thus crystal anisotropy is absent. The evolution of spatiotemporally resolved quantities during collapse is reported including local pressure, temperature, pore size and shape, and material flow. Multiple models for the melting curve and specific heat were studied. Within the isotropic elastic/perfectly plastic continuum framework and for the range of impact conditions studied, the specific heat and melting curve sub‐models are shown to have modest effects on the continuum hotspot predictions in the present inert calculation. Treating the MD predictions as 'ground truth', albeit with a classical rather than quantum‐like heat capacity, it is clear that – at a minimum – an extension of the constitutive model to account for crystal plasticity and anisotropic strength will be required for a high‐fidelity continuum description. [ABSTRACT FROM AUTHOR]

Abstract: The second‐order elastic stiffness tensor and isotropic moduli of β‐octahydro‐1,3,5,7‐tetranitro‐1,3,5,7‐tetrazocine (β‐HMX, P21/n space group setting) on the 0 K isotherm are presented for hydrostatic pressures between 10−4 GPa and 30 GPa. The results were obtained from molecular statics simulations using a validated all‐atom flexible‐molecule force field. Comparisons to previous experimental and computational determinations are provided. [ABSTRACT FROM AUTHOR]

The cover image is based on the Original Article I Molecular dynamics inferred transfer learning models for finite-strain hyperelasticity of monoclinic crystals: Sobolev training and validations against physical constraints i by Nikolaos Vlassis et al., https://doi.org/10.1002/nme.6992. [Extracted from the article]

There are many compelling scientific and technological reasons for understanding the processes leading to burn ignition and detonation initiation in propellants and explosives. Multi-physics continuum simulations have been successfully applied to the crystal scale, accompanied by the development of more sophisticated constitutive models. [Extracted from the article]