Effect of vacancy creation and annihilation on grain boundary motion.

Interaction of vacancies with grain boundaries (GBs) is involved in many processes occurring in materials, including radiation damage healing, diffusional creep, and solid-state sintering. We analyze a model describing a set of processes occurring at a GB in the presence of a non-equilibrium, non-homogeneous vacancy concentration. Such processes include vacancy diffusion toward, away from, and across the GB, vacancy generation and absorption at the GB, and GB migration. Numerical calculations within this model reveal that the coupling among the different processes gives rise to interesting phenomena, such as vacancy-driven GB motion and accelerated vacancy generation/absorption due to GB motion. The key combinations of the model parameters that control the kinetic regimes of the vacancy-GB interactions are identified via a linear stability analysis. Possible applications and extensions of the model are discussed.

Phases, phase equilibria, and phase rules in low-dimensional systems.

We present a unified approach to thermodynamic description of one, two, and three dimensional phases and phase transformations among them. The approach is based on a rigorous definition of a phase applicable to thermodynamic systems of any dimensionality. Within this approach, the same thermodynamic formalism can be applied for the description of phase transformations in bulk systems, interfaces, and line defects separating interface phases. For both lines and interfaces, we rigorously derive an adsorption equation, the phase coexistence equations, and other thermodynamic relations expressed in terms of generalized line and interface excess quantities. As a generalization of the Gibbs phase rule for bulk phases, we derive phase rules for lines and interfaces and predict the maximum number of phases than may coexist in systems of the respective dimensionality.

Effect of interface phase transformations on diffusion and segregation in high-angle grain boundaries.

Recent experimental measurements of Ag impurity diffusion in the Σ5(310) grain boundary (GB) in Cu revealed an unusual non-Arrhenius behavior suggestive of a possible structural transformation Divinski et al., [Phys. Rev. B 85, 144104 (2012)]. On the other hand, atomistic computer simulations have recently discovered phase transformations in high-angle GBs in metals Frolov et al., [Nat. Commun. 4, 1899 (2013)]. In this Letter we report on atomistic simulations of Ag diffusion and segregation in two different structural phases of the Cu Σ5(310) GB which transform to each other with temperature. The obtained excellent agreement with the experimental data validates the hypothesis that the unusual diffusion behavior seen in the experiment was caused by a phase transformation. The simulations also predict that the low-temperature GB phase exhibits a monolayer segregation pattern while the high-temperature phase features a bilayer segregation. Together, the simulations and experiment provide the first convincing evidence for the existence of structural phase transformations in high-angle metallic GBs and demonstrate the possibility of their detection by GB diffusion measurements and atomistic simulations.

Liquid nucleation at superheated grain boundaries.

Grain boundaries with relatively low energies can be superheated above the melting temperature and eventually melt by heterogeneous nucleation of liquid droplets. We propose a thermodynamic model of this process based on the sharp-interface approximation with a disjoining potential. The distinct feature of the model is its ability to predict the shape and size of the critical nucleus by using a variational approach. The model reduces to the classical nucleation theory in the limit of large nuclei but is more general and remains valid for small nuclei. Contrary to the classical nucleation theory, the model predicts the existence of a critical temperature of superheating and offers a simple formula for its calculation. The model is tested against molecular dynamic simulations in which liquid nuclei at a superheated boundary were obtained by an adiabatic trapping procedure. The simulation results demonstrate a reassuring consistency with the model.

The Role of Grain Boundary Diffusion in the Solute Drag Effect.

Molecular dynamics (MD) simulations are applied to study solute drag by curvature-driven grain boundaries (GBs) in Cu-Ag solid solution. Although lattice diffusion is frozen on the MD timescale, the GB significantly accelerates the solute diffusion and alters the state of short-range order in lattice regions swept by its motion. The accelerated diffusion produces a nonuniform redistribution of the solute atoms in the form of GB clusters enhancing the solute drag by the Zener pinning mechanism. This finding points to an important role of lateral GB diffusion in the solute drag effect. A 1.5 at.%Ag alloying reduces the GB free energy by 10-20% while reducing the GB mobility coefficients by more than an order of magnitude. Given the greater impact of alloying on the GB mobility than on the capillary driving force, kinetic stabilization of nanomaterials against grain growth is likely to be more effective than thermodynamic stabilization aiming to reduce the GB free energy.

Stable nanocolloidal structures in metallic systems.

Molecular dynamics simulations with a semiempirical interatomic potential predict the existence of a thermodynamically stable colloidal structure with nanometer-size Ta particles suspended in liquid Cu. The thermodynamic stability of this structure against coarsening, coalescence or transformation to a homogeneous solution is explained by a negative and strongly curvature-dependent tension of the Cu/Ta interfaces. Simulations with this potential provide a simple computational model for studies of generic thermodynamic properties of nanodisperse systems with a negative interface free energy.

Thermal conductivity and its relation to atomic structure for symmetrical tilt grain boundaries in silicon.

We perform a systematic study of thermal resistance/conductance of tilt grain boundaries (GBs) in Si using classical molecular dynamics. The GBs studied are naturally divided into three groups according to the structural units forming the GB core. We find that, within each group, the GB thermal conductivity strongly correlates with the excess GB energy. All three groups predict nearly the same GB conductivity extrapolated to the high-energy limit. This limiting value is close to the thermal conductivity of amorphous Si, suggesting similar heat transport mechanisms. While the lattice thermal conductivity decreases with temperature, the GB conductivity slightly increases. However, at high temperatures it turns over and starts decreasing if the GB structure undergoes a premelting transformation. Analysis of vibrational spectra of GBs resolved along different directions sheds light on the mechanisms of their thermal resistance. The existence of alternating tensile and compressive atomic environments in the GB core gives rise to localized vibrational modes, frequency gaps creating acoustic mismatch with lattice phonons, and anharmonic vibrations of loosely-bound atoms residing in open atomic environments.

Solid-liquid interface free energy in binary systems: theory and atomistic calculations for the (110) Cu-Ag interface.

We analyze thermodynamics of solid-liquid interfaces in binary systems when the solid is in a nonhydrostatic state of stress. The difficulty lies in the fact that chemical potential of at least one of the chemical components in a nonhydrostatic solid is an undefined quantity. We show, nevertheless, that the interface free energy gamma can be defined as excess of an appropriate thermodynamic potential that depends on the chemical potentials in the liquid phase. We derive different forms of the adsorption equation for solid-liquid interfaces, with differential coefficients representing excesses of extensive properties. This leads, in particular, to the formulation of interface stress tau(ij) as an appropriate excess over nonhydrostatic bulk stresses. The interface stress is not unique unless the solid is in a hydrostatic state of stress. We also derive Gibbs-Helmholtz type equations that can be applied for thermodynamic integration of gamma. All thermodynamic relations derived here are presented in forms suitable for atomistic simulations. In particular, the excess quantities can be computed without constructing interface profiles. As an application, we perform semigrand canonical Monte Carlo simulations of the (110) solid-liquid interface in the Cu-Ag system. We show that gamma computed by thermodynamic integration along a coexistence path decreases with increasing composition difference between the phases. At the same time, tau(ij) remains negative (i.e., the interface is in a state of compression), drastically increases in magnitude, and becomes highly anisotropic. Some of the interface excess properties are computed by different methods and demonstrate accurate agreement with each other, confirming the correctness of our analysis.

Physically informed artificial neural networks for atomistic modeling of materials.

Large-scale atomistic computer simulations of materials heavily rely on interatomic potentials predicting the energy and Newtonian forces on atoms. Traditional interatomic potentials are based on physical intuition but contain few adjustable parameters and are usually not accurate. The emerging machine-learning (ML) potentials achieve highly accurate interpolation within a large DFT database but, being purely mathematical constructions, suffer from poor transferability to unknown structures. We propose a new approach that can drastically improve the transferability of ML potentials by informing them of the physical nature of interatomic bonding. This is achieved by combining a rather general physics-based model (analytical bond-order potential) with a neural-network regression. This approach, called the physically informed neural network (PINN) potential, is demonstrated by developing a general-purpose PINN potential for Al. We suggest that the development of physics-based ML potentials is the most effective way forward in the field of atomistic simulations.

Unraveling the dislocation core structure at a van der Waals gap in bismuth telluride.

Tetradymite-structured chalcogenides such as bismuth telluride (Bi2Te3) are of significant interest for thermoelectric energy conversion and as topological insulators. Dislocations play a critical role during synthesis and processing of such materials and can strongly affect their functional properties. The dislocations between quintuple layers present special interest since their core structure is controlled by the van der Waals interactions between the layers. In this work, using atomic-resolution electron microscopy, we resolve the basal dislocation core structure in Bi2Te3, quantifying the disregistry of the atomic planes across the core. We show that, despite the existence of a stable stacking fault in the basal plane gamma surface, the dislocation core spreading is mainly due to the weak bonding between the layers, which leads to a small energy penalty for layer sliding parallel to the van der Waals gap. Calculations within a semidiscrete variational Peierls-Nabarro model informed by first-principles calculations support our experimental findings.

Nickel nanoparticles set a new record of strength.

Material objects with micrometer or nanometer dimensions can exhibit much higher strength than macroscopic objects, but this strength rarely approaches the maximum theoretical strength of the material. Here, we demonstrate that faceted single-crystalline nickel (Ni) nanoparticles exhibit an ultrahigh compressive strength (up to 34 GPa) unprecedented for metallic materials. This strength matches the available estimates of Ni theoretical strength. Three factors are responsible for this record-high strength: the large Ni shear modulus, the smooth edges and corners of the nanoparticles, and the thin oxide layer on the particle surface. This finding is supported by molecular dynamics simulations that closely mimic the experimental conditions, which show that the mechanical failure of the strongest particles is triggered by homogeneous nucleation of dislocation loops inside the particle. The nucleation of a stable loop is preceded by multiple nucleation attempts accompanied by unusually large local atomic displacements caused by thermal fluctuations.

Energy spectrum of a Langevin oscillator.

We derive analytical solutions for the autocorrelation and cross-correlation functions of the kinetic, potential, and total energy of a Langevin oscillator. These functions are presented in both the time and frequency domains and validated by independent numerical simulations. The results are applied to address the long-standing issue of temperature fluctuations in canonical systems.

Molecular dynamics simulation of the martensitic phase transformation in NiAl alloys.

Using molecular dynamics simulations with an embedded-atom interatomic potential, we study the effect of chemical composition and uniaxial mechanical stresses on the martensitic phase transformation in Ni-rich NiAl alloys. The martensitic phase has a tetragonal crystal structure and can contain multiple twins arranged in domains and plates. The transformation is reversible and is characterized by a significant temperature hysteresis. The magnitude of the hysteresis depends on the chemical composition and stress. We show that applied compressive and tensile stresses reduce and can even eliminate the hysteresis. Crystalline defects such as free surfaces, dislocations and anti-phase boundaries reduce the martensitic transformation temperature and affect the microstructure of the martensite. Their effect can be explained by heterogeneous nucleation of the new phase in defected regions.