Search

A remarkable property of certain covalent glasses and their melts is intermediate range order, manifested as the first sharp diffraction peak (FSDP) in neutron-scattering experiments, as was exhaustively investigated by Price, Saboungi, and collaborators. Atomistic simulations thus far have relied on either quantum molecular dynamics (QMD), with systems too small to resolve FSDP, or classical molecular dynamics, without quantum-mechanical accuracy. We investigate prototypical FSDP in GeSe2 glass and melt using neural-network quantum molecular dynamics (NNQMD) based on machine learning, which allows large simulation sizes with validated quantum mechanical accuracy to make quantitative comparisons with neutron data. The system-size dependence of the FSDP height is determined by comparing QMD and NNQMD simulations with experimental data. Partial pair distribution functions, bond-angle distributions, partial and neutron structure factors, and ring-size distributions are presented. Calculated FSDP heights agree quantitatively with neutron scattering data for GeSe2 glass at 10 K and melt at 1100 K.

Despite the growing success of machine learning for predicting structure-property relationships in molecules and materials, such as predicting the dielectric properties of polymers, it is still in its infancy. We report on the effectiveness of solving structure-property relationships for a computer-generated database of dielectric polymers using recurrent neural network (RNN) models. The implementation of a series of optimization strategies was crucial to achieving high learning speeds and sufficient accuracy: (1) binary and nonbinary representations of SMILES (Simplified Molecular Input Line System) fingerprints and (2) backpropagation with affine transformation of the input sequence (ATransformedBP) and resilient backpropagation with initial weight update parameter optimizations (iRPROP- optimized). For the investigated database of polymers, the binary SMILES representation was found to be superior to the decimal representation with respect to the training and prediction performance. All developed and optimized Elman-type RNN algorithms outperformed nonoptimized RNN models in the efficient prediction of nonlinear structure-activity relationships. The average relative standard deviation (RSD) remained well below 5%, and the maximum RSD did not exceed 30%. Moreover, we provide a C++ codebase as a testbed for a new generation of open programming languages that target increasingly diverse computer architectures.

The static dielectric constant ϵ_{0} and its temperature dependence for liquid water is investigated using neural network quantum molecular dynamics (NNQMD). We compute the exact dielectric constant in canonical ensemble from NNQMD trajectories using fluctuations in macroscopic polarization computed from maximally localized Wannier functions (MLWF). Two deep neural networks are constructed. The first, NNQMD, is trained on QMD configurations for liquid water under a variety of temperature and density conditions to learn potential energy surface and forces and then perform molecular dynamics simulations. The second network, NNMLWF, is trained to predict locations of MLWF of individual molecules using the atomic configurations from NNQMD. Training data for both the neural networks is produced using a highly accurate quantum-mechanical method, DFT-SCAN that yields an excellent description of liquid water. We produce 280×10^{6} configurations of water at 7 temperatures using NNQMD and predict MLWF centers using NNMLWF to compute the polarization fluctuations. The length of trajectories needed for a converged value of the dielectric constant at 0°C is found to be 20 ns (40×10^{6} configurations with 0.5 fs time step). The computed dielectric constants for 0, 15, 30, 45, 60, 75, and 90°C are in good agreement with experiments. Our scalable scheme to compute dielectric constants with quantum accuracy is also applicable to other polar molecular liquids.

Confluence of extreme-scale quantum dynamics simulations (i.e. quantum@scale) and cutting-edge x-ray free-electron laser experiments are revolutionizing materials science. An archetypal example is the exciting concept of using picosecond light pulses to control emergent material properties on demand in atomically-thin layered materials. This paper describes efforts to scale our quantum molecular dynamics engine toward the United States' first exaflop/s computer, under an Aurora Early Science Program project named "Metascalable layered material genome". Key algorithmic and computing techniques incorporated are: (1) globally-scalable and locally-fast solvers within a linear-scaling divide-conquer-recombine algorithmic framework; (2) algebraic 'BLASification' of computational kernels; and (3) data alignment and loop restructuring, along with register and cache blocking, for enhanced vectorization and efficient memory access. The resulting weak-scaling parallel efficiency was 0.93 on 131,072 Intel Xeon Phi cores for a 56.6 million atom (or 169 million valence-electron) system, whereas the various code transformations achieved 5-fold speedup. The optimized simulation engine allowed us for the first time to establish a significant effect of substrate on the dynamics of layered material upon electronic excitation.

A wide variety of two-dimensional layered materials are synthesized by liquid-phase exfoliation. Here we examine exfoliation of MoS2 into nanosheets in a mixture of water and isopropanol (IPA) containing cavitation bubbles. Using force fields optimized with experimental data on interfacial energies between MoS2 and the solvent, multimillion-atom molecular dynamics simulations are performed in conjunction with experiments to examine shock-induced collapse of cavitation bubbles and the resulting exfoliation of MoS2. The collapse of cavitation bubbles generates high-speed nanojets and shock waves in the solvent. Large shear stresses due to the nanojet impact on MoS2 surfaces initiate exfoliation, and shock waves reflected from MoS2 surfaces enhance exfoliation. Structural correlations in the solvent indicate that shock induces an ice VII like motif in the first solvation shell of water.

Electrical and optoelectronic properties of two-dimensional (2D) transition metal dichalcogenides (TMDCs) can be tuned by exploiting their structural phase transitions. Here semiconducting (2H) to metallic (1T) phase transition is investigated in a strained ${\mathrm{MoWSe}}_{2}$ monolayer using molecular dynamics (MD) simulations. Novel intermediate structures called $\ensuremath{\alpha}$ and $\ensuremath{\beta}$ are found between the 2H and 1T phases. These intermediate structures are similar to those observed in a 2D $\mathrm{Mo}{\mathrm{S}}_{2}$ by scanning transmission electron microscopy. A deep generative model, namely the variational autoencoder (VAE) trained by MD data, is used to generate novel heterostructures with $\ensuremath{\alpha}$ and $\ensuremath{\beta}$ interfaces. Quantum simulations based on density functional theory show that these heterostructures are stable and suitable for novel nanoelectronics applications.

QXMD is a scalable, parallel program for Quantum Molecular Dynamics simulations with various eXtensions. Its simulation engine is based on (time-dependent) density functional theory using pseudopotentials and a plane-wave basis set, while extensions include nonadiabatic electron–nuclei dynamics and multiscale shock technique. QXMD serves as a community-development platform for new methods and algorithms, a research platform on high-end parallel supercomputers, and an educational platform for hands-on training. Keywords: Nonadiabatic quantum molecular dynamics, Parallel computing, Hands-on training

The mechanical response of ultralight kagome structures consisting of hollow nickel (Ni) nanotubes and solid Ni nanorods to compression is studied using molecular dynamics simulations. In both kagome architectures, 1 6 [ 112 ] Shockley partial dislocations and twin formation are observed under compression. The structure made from solid nanorods shows deformation near both the nodes and beams of the kagome lattice. The hollow kagome architecture has a higher yield point than the solid kagome structure. The deformation in the hollow nanotube structure is mostly localized in the nodal region for strains less than 11%. At higher strains, the deformation sets in all the struts and nodes of the hollow kagome lattice. Owing to this two-stage deformation mechanism, the hollow Ni nanotube kagome structure shows less bending and greater toughness than the solid Ni nanorod kagome architecture.

This study uses ab initio quantum molecular dynamics (QMD) simulations to validate multimillion-atom reactive molecular dynamics (RMD) simulations, and predicts unexpected condensation of carbon atoms during high-temperature oxidation of silicon-carbide nanoparticles (nSiC). For the validation process, a small nSiC in oxygen environment is chosen to perform QMD simulation. The QMD results provide the number of Si-O and C-O bonds as a function of time. RMD simulation is then performed under the identical condition. The time evolutions of different bonds are compared between the QMD and RMD simulations. We observe the condensation of large number of C-cluster nuclei into larger C clusters in both simulations, thereby validating RMD. Furthermore, we use the QMD simulation results as an input to a multi-objective genetic algorithm to train the RMD force-field parameters. The resulting force field far better reproduces the ground-truth QMD simulation results.

RXMD is a parallel reactive molecular dynamics (RMD) simulator designed to perform large-scale RMD simulations on commodity laptop computers to supercomputer platforms. With extensive Time-to-Solution (TtoS) optimization techniques and scalable algorithms implemented, RXMD enables researchers to explorer materials design space that requires vast spatial-extent and long-time atomistic events with highly accurate chemistry. Extensible and modular software architecture allows to implement innovative optimization techniques. RXMD is one of core simulation engines at Material Genome Innovation for Computational Software (MAGICS) center, and is freely distributed as an open-source software from MAGICS website and Github. RXMD is also used as computational materials courseware to provide basic training of parallel RMD simulations for computational researchers and community.

Reactive molecular dynamics simulations are performed to study self-healing of cracks in Al2O3 containing core/shell SiC/SiO2 nanoparticles. These simulations are carried out in a precracked Al2O3 under mode 1 strain at 1426 °C. The nanoparticles are embedded ahead of the precrack in the Al2O3 matrix. When the crack begins to propagate at a strain of 2%, the nanoparticles closest to the advancing crack distort to create nanochannels through which silica flows toward the crack and stops its growth. At this strain, the Al2O3 matrix at the interface of SiC/SiO2 nanoparticles forms facets along the prismatic (A) ⟨2110⟩ and prismatic (M) ⟨1010⟩ planes. These facets act as nucleation sites for the growth of multiple secondary amorphous grains in the Al2O3 matrix. These grains grow with an increase in the applied strain. Voids and nanocracks form in the grain boundaries but are again healed by diffusion of silica from the nanoparticles.

Quantification of lattice thermal conductivity of two-dimensional semiconductors like MoS2 is necessary for the design of electronic and thermoelectric devices, but direct experimental measurements on free-standing samples is challenging. Molecular dynamics simulations, with appropriate corrections, can provide a reference value for thermal conductivity for these material systems. Here, we construct a new empirical forcefield of the Stillinger-Weber form, parameterized to phonon dispersion relations, lattice constants and elastic moduli and we use it to compute a material-intrinsic thermal conductivity of 38.1 W/m-K at room temperature and estimate a maximum thermal conductivity of 85.4 W/m-K at T = 200 K. We also identify that phonon scattering by the large isotopic mass distribution of Mo and S contributes a significant correction (>45%) to the thermal conductivity at low temperatures.

Anisotropic frictional response and corresponding heating in cyclotrimethylene-trinitramine molecular crystals are studied using molecular dynamics simulations. The nature of damage and temperature rise due to frictional forces is monitored along different sliding directions on the primary slip plane, (010), and on non-slip planes, (100) and (001). Correlations between the friction coefficient, deformation, and frictional heating are established. We find that the friction coefficients on slip planes are smaller than those on non-slip planes. In response to sliding on a slip plane, the crystal deforms easily via dislocation generation and shows less heating. On non-slip planes, due to the inability of the crystal to deform via dislocation generation, a large damage zone is formed just below the contact area, accompanied by the change in the molecular ring conformation from chair to boat/half-boat. This in turn leads to a large temperature rise below the contact area.Anisotropic frictional response and corresponding heating in cyclotrimethylene-trinitramine molecular crystals are studied using molecular dynamics simulations. The nature of damage and temperature rise due to frictional forces is monitored along different sliding directions on the primary slip plane, (010), and on non-slip planes, (100) and (001). Correlations between the friction coefficient, deformation, and frictional heating are established. We find that the friction coefficients on slip planes are smaller than those on non-slip planes. In response to sliding on a slip plane, the crystal deforms easily via dislocation generation and shows less heating. On non-slip planes, due to the inability of the crystal to deform via dislocation generation, a large damage zone is formed just below the contact area, accompanied by the change in the molecular ring conformation from chair to boat/half-boat. This in turn leads to a large temperature rise below the contact area.