Nano‑Micromechanics Workshop

Wear in high-pressure lubrication is a critical engineering issue. Recently, researchers have used C60 endohedral fullerene-based nanoparticles to reduce wear in high-pressure lubrication. Here, we present our nano-mechanics modeling of atomic transport processes in nanoparticle-mediated wear reduction, informed by our multi-scale experimental characterization of surfaces subjected to 52100-steel pin-on-disk testing under high nominal contact pressure. The base lubricant was PAO4, with or without C60-modified nanoparticles. The experimental macroscopic friction coefficient remained comparable between the two conditions; in contrast, the additive reduced pin material loss by approximately 33.3%, indicating a significant improvement in wear resistance, but at the cost of high-frequency lubricant-temperature fluctuations. AFM topography further shows that the wear region with the additive has a smoother surface than the additive-free case, suggesting reduced surface damage. The additive-containing wear region also shows noticeably lower nanoscale friction.

We demonstrate that grain boundaries (GBs) exhibit direction-dependent mobilities and unidirectional motion under oscillatory driving forces or cyclic thermal annealing. We observed these phenomena for nearly all nonsymmetric GBs but not for symmetric ones. Our observations build on molecular dynamics and phase-field crystal simulations for a wide range of GB types and driving forces in both bicrystal and polycrystalline microstructures. We corroborate these simulation results through in situ experimental observations. We analyze these results with a Markov chain model and explore the implications of GB ratchet behavior for materials processing and microstructure tailoring.

Tribological contacts account for a significant fraction of global primary energy consumption through friction and wear. Much of this energy is dissipated within the material beneath the two sliding surfaces, where severe shear deformation can produce layered subsurface microstructures with refined grain sizes. Systematic model experiments are used to elucidate the elementary mechanisms governing the evolution of these microstructures.

Understanding how hydrogen affects defect kinetics in metals is essential for clarifying the mechanisms of hydrogen-induced damage accumulation and embrittlement. However, the elementary kinetic processes involved remain insufficiently understood. Atomistic simulations provide a powerful means of directly investigating these processes, although their predictive capability depends critically on the accuracy of the interatomic interactions used for metal–hydrogen systems. Recent advances in artificial neural network (ANN) interatomic potentials, trained on density functional theory (DFT) datasets of atomic energies, forces, and configurations, have enabled simulations that combine the computational efficiency of empirical potentials with near-DFT accuracy and transferability. These potentials allow quantitative analyses of hydrogen-dependent defect kinetics, including vacancy migration, surface diffusion, and dislocation motion, under prescribed hydrogen chemical potential, temperature, and applied stress. In this presentation, we introduce recent atomistic simulation results and a unified theoretical framework for describing defect kinetics in hydrogen environments.

We develop a theory for the plastic shear rate on each slip system of face-centered cubic (FCC) single crystals, by integrating the shear rate contributions from dislocation links of length l. A generalized Orowan equation expresses the length-resolved shear rate in terms of the link density n_i(l) and the average link velocity v_i(l). This model accurately reproduces the discrete dislocation dynamics (DDD) shear-rate data for all 12 slip systems across 99 loading orientations. In particular, it explains why the plastic shear rate depends exponentially on the resolved shear stress τ_i, as reported previously, in the absence of any thermally activated mechanism. This dependence arises from the exponential tail of n_i(l) filtered through the threshold-like velocity function v_i(l). Furthermore, v_i(l) is well approximated by a single two-dimensional function v(τ_i, l) for all these slip systems. The link velocity is therefore governed mainly by τ_i and l, with only a minor effect from the complex interactions between slip systems. These interactions instead manifest themselves mainly through n_i(l). As a demonstration of the effect of the link-length distribution on the slip rate, we show that even when τ_i is limited to a narrow window, the shear rates still span two orders of magnitude. The supply of long links is therefore the dominant factor controlling slip-system activity, outweighing the direct effect of τ_i on the link velocity.

Over the last 30–35 years myself and my collaborators at MIT, LLNL and elsewhere have been working to harness the power of extreme HPC for predictive simulations of materials dynamics. Rather than giving a summary of our findings and predictions, I would like to name and address a few challenges that extreme supercomputing poses for multi-scale materials simulations. Chief among them is the immense amounts of data generated in such simulations and our limited ability to learn from this data.

Discrete dislocation dynamics has mostly been applied to problems it can pose on its own — a loaded micropillar or thin film, a periodic cell, a single slip system. Most problems that matter are boundary value problems, and coupling DD to finite element analysis opens a class of questions the DD community has only begun to explore. This talk gives an overview of the Defect Dynamics Model (DDM), a concurrent framework in which three-dimensional DD and FEM are solved together, so that complex geometries, free surfaces, interfaces and realistic loading can be treated while plasticity remains an outcome of line mechanics rather than a constitutive fit. I will describe what the framework can now do and the atomistic inputs it takes — mobility, nucleation barriers, interfacial properties, and increasingly mobility laws learned from molecular dynamics — then turn to recent applications: size-dependent plasticity and passivation in thin films and micropillars; slip transmission and dislocation avalanches near grain boundaries under nanoindentation; dislocation interaction with nanoprecipitates; and the hardening of graphene-coated nickel foams. A growing part of this work addresses reliability in semiconductor devices, where failure is rarely a single mechanism. Coupling lets them be treated together: DD sets where and when damage begins from the actual dislocation structure, while the finite element side carries interfacial delamination, crack growth through XFEM, and phase-field fracture. Assessing device reliability means putting microstructure and macroscopic failure in the same calculation, and this is what the coupled model makes possible.

Materials deformation is ultimately controlled by physical processes at the atomic scale. In many cases, these processes are controlled by fluctuations characterized by highly stochastic behavior. While we have learned to subsume this stochasticity into continuum laws that describe average behavior, as we push the boundaries of what is observable in terms of time and length scales, the mean-field approach becomes limited. Important phenomena in metals deformation such as creep, dynamic strain aging, solute hardening, and deformation processes in chemically complex alloys cannot be properly studied without capturing atomic-level fluctuations and their effect on meso/macroscopic behavior. In this presentation, I will discuss the development of models that take into account thermal and compositional fluctuations explicitly and how their results can be extended into larger length and longer time scales. Results for refractory transition metals, ferritic materials, and high-entropy alloys will be shown.

While the path to fusion energy has seen extraordinary progress, the lack of structural and first-wall materials capable of withstanding extreme irradiation conditions remains a critical bottleneck. In this talk I will discuss recent progress we have made to develop a multiscale modeling framework for predicting irradiation-induced microstructure evolution in refractory complex alloys. The framework integrates neutronics, machine-learning interatomic potentials (MLIP), molecular dynamics, and kinetic Monte Carlo simulations. I will show how our newly developed atomic cluster expansion-based MLIP for the W–Ta–Hf–Cr–V–Ti–He system enables us to perform primary cascade and diffusion calculations with accuracy comparable to that of the largest existing MLIPs but at a fraction of the cost. These atomistic calculations then parameterize our generalized first-passage kinetic Monte Carlo approach capable of handling complex diffusion behaviors. I will also discuss how we plan to leverage AI agents to orchestrate the work and help navigate the design space in search for unique compositions that satisfy design requirements. Altogether, the objective of our alloy-design workflow is to enable atomistically informed predictions of irradiated microstructure evolution over reactor time scales, thereby helping identify compositions and mechanisms that minimize radiation-induced hardening, swelling, and embrittlement.