arXiv subjects

Zheyong Fan

Publications and source records attributed to Zheyong Fan.

At least 19 recordsLinked to original sources

arXiv2026

NEP89: Universal neuroevolution potential for inorganic and organic materials across 89 elements

While machine-learned interatomic potentials offer near-quantum-mechanical accuracy for atomistic simulations, many are material-specific or computationally intensive, limiting their broader use. Here, we introduce NEP89, a foundation model based on neuroevolution potential architecture, delivering near-empirical-potential speed and high accuracy across 89 elements. A compact yet comprehensive training dataset covering inorganic and organic materials was curated through descriptor-space subsampling and iterative refinement across multiple datasets. NEP89 achieves competitive accuracy compared to representative foundation models while being three to four orders of magnitude more computationally efficient, enabling previously impractical large-scale atomistic simulations of inorganic and organic systems. In addition to its out-of-the-box applicability to diverse scenarios, including million-atom-scale compression of compositionally complex alloys, ion diffusion in solid-state electrolytes and water, rocksalt dissolution, methane combustion, and protein-ligand dynamics, NEP89 also supports fine-tuning for rapid adaptation to user-specific applications, such as mechanical, thermal, structural, and spectral properties of two-dimensional materials, metallic glasses, and organic crystals.

cond-mat.mtrl-sci↗

arXiv2026

Benchmarking Chemically Scalable Machine-Learning Interatomic Potentials for Large-Scale Simulations of Multicomponent Alloys

Machine learning interatomic potentials (MLIPs) with broad chemical flexibility are essential for atomistic simulations of compositionally complex alloys, but their deployment in large-scale molecular dynamics requires a balance among accuracy, efficiency, stability, transferability, and uncertainty quantification. Here, we benchmark two chemically scalable MLIP frameworks, neuroevolution potential (NEP) and graph atomic cluster expansion (GRACE), for 16 elemental metals and their multicomponent alloys. GRACE-FS shows higher training efficiency and generally better average accuracy, chemical transferability, and finite-temperature robustness, whereas UNEP-v1 provides substantially higher inference speed and remains competitive in selected stress and large-error metrics. We further show that chemical transferability is closely linked to high-temperature MD stability in highly multicomponent environments and that ensemble-based uncertainty provides a more reliable error indicator than D-optimality for the heterogeneous systems considered here. Finally, three-million-atom shock simulations demonstrate that UNEP-v1, combined with ensemble uncertainty, enables uncertainty-aware simulations under extreme dynamic conditions, yielding robust global spall-strength predictions while revealing model sensitivity in local damage pathways. These results provide practical guidelines for selecting and deploying MLIPs in large-scale simulations of multicomponent alloys.

cond-mat.mtrl-sci↗

arXiv2026

TorchNEP: Ultra-Efficient and Accurate Training of Neuroevolution Potentials

Neuroevolution Potential (NEP) is one of the most efficient machine-learned interatomic potential frameworks for large-scale atomistic simulations. However, its original training strategy remains computationally demanding, limiting systematic exploration of model architectures and training protocols. Here, we present TorchNEP, a PyTorch-based implementation of NEP that combines analytically derived gradients, adaptive optimization, and a two-stage training strategy. TorchNEP accelerates training by more than two orders of magnitude while maintaining full compatibility with existing NEP models. We further show that the improvement in predictive accuracy primarily originates from the two-stage training protocol rather than the optimization algorithm itself. Across diverse benchmark datasets, TorchNEP consistently improves force and stress predictions while maintaining comparable or improved energy accuracy. Benchmark evaluations on elemental and alloy systems demonstrate enhanced predictive performance for both atomic configurations and key materials properties. Furthermore, we show that increasing model complexity does not necessarily improve predictive performance despite reducing training errors. Overall, TorchNEP provides an efficient and flexible training framework for developing more accurate and robust machine-learned interatomic potentials.

physics.comp-ph↗

arXiv2026

Trillion-atom molecular dynamics simulations with ab initio accuracy

Material properties are fundamentally dictated by multiscale phenomena, which often reach mesoscale in size. The μm mesoscale is also the size which can be observed directly under an optical microscope, bridging the atomistic microscopic description with the continuous model macroscopic world. In this work, we report an unprecedented molecular dynamics (MD) simulation comprising 1.62 trillion atoms. Utilizing the neuroevolution potential (NEP) framework, we attained ab initio accuracy on China's New-generation Intelligent Supercomputer. Our implementation achieves a time-to-solution (s/step/atom) 100 times faster than previous state-of-the-art machine learning force field simulations, and 1,000 times faster than the Gordon Bell Prize-winning application from six years ago. Furthermore, we demonstrate an 86.9% weak scaling efficiency from a single GPGPU to 45,000 GPGPUs. These results redefine atomistic simulation boundaries, enabling direct mesoscopic modeling with quantum-level precision.

cond-mat.mtrl-sci↗

arXiv2026

NEPMaker: Active learning of neuroevolution machine learning potential for large cells

Machine learning potentials (MLPs) achieve near first-principles accuracy but often fail for atomic environments outside the training distribution. Active learning can mitigate this limitation; however, its application to large-scale simulations is hindered by the prohibitive cost of labeling entire configurations. Here, we develop a D-optimality-driven active learning framework for the neuroevolution potential (NEP) implemented within the GPUMD package, named NEPMaker. Extrapolative atomic environments are identified on-the-fly and embedded into locally periodic structures, where boundary atoms are optimized to remain close to the training distribution. This strategy enables large-scale simulations to directly contribute to dataset construction, significantly reducing extrapolation errors while improving model robustness and transferability. The proposed framework provides a scalable route for constructing reliable machine learning potentials in complex materials systems, including those involving defects, interfaces, and phase transitions.

physics.comp-ph↗

arXiv2026

Structurally Triggered Breakdown of the Phonon Gas Model in Crystalline Metal-Organic Frameworks

While crystalline materials with glass-like thermal conductivity are fundamentally intriguing, structurally triggering the transition from propagating to diffusive heat transport within a single framework remains a formidable challenge. Here, using extensive machine learning molecular dynamics, we demonstrate a fundamental thermal transport crossover in metal-organic frameworks. We reveal that grafting flexible side chains onto a pristine MOF backbone acts as a structural switch, strongly reducing the thermal conductivity by $\sim$70% (from $\sim 0.7$ to $\sim 0.2\ \text{W m}^{-1}\text{K}^{-1}$ at 300 K). Crucially, the functionalized derivatives exhibit a drastic transition from a classical Peierls $\sim 1/T$ decay to an anomalous, temperature-independent glass-like plateau. Reciprocal- and real-space analyses reveal the microscopic origins: the side chains act as built-in local resonators that trap acoustic energy via strong low-frequency resonant hybridization, while simultaneously inducing extreme steric crowding. Consequently, the heat-carrying phonon modes become critically damped, with their mean free paths strictly confined to the nanometer scale and their lifetimes collapsing to the Ioffe-Regel limit. This work establishes a highly programmable molecular engineering strategy to dismantle the phonon gas model, forcing crystalline frameworks into an extreme diffusive transport regime.

cond-mat.soft↗

arXiv2026

Heat transport in superionic materials via machine-learned molecular dynamics

Precise modeling and understanding of heat transport in the superionic phase are of great interest. Although simulations combining Green-Kubo (GK) molecular dynamics with machine-learned potentials (MLPs) stand as a promising approach, substantial challenges remain due to the crucial impact of atomic diffusion. Here, we first show that the thermal conductivity ($κ$) of superionic materials calculated via conventional GK integral of the energy flux varies notably with the MLP model. Subsequently, we highlight that reliable, model-independent $κ$ values can be obtained by applying Onsager's reciprocal relations to correctly capture the coupled heat and mass transport. Remarkably, an anomalously invariant $κ$ can be observed over a wide temperature range, distinct from the characteristic trends in traditional crystals and glasses. In addition, we illustrate that conventional $κ$ decompositions into kinetic, potential, and cross terms suffer from ambiguities in the physical interpretation, despite their mathematical rigor. Finally, we propose a criterion for the necessity of the Onsager correction and reveal the underlying mechanism as a competition between thermally and chemically driven ion fluxes.

cond-mat.mtrl-sci↗

arXiv2026

GPU-MetaD: Full-Life-Cycle GPU Accelerated Metadynamics with Machine Learning Potentials

Large-scale molecular dynamics simulations with high accuracy have been increasingly popular for their capability to bridge the gap between atomistic modeling and mesoscale phenomena. Both machine learning potentials and enhanced sampling approaches offer substantial improvements in high-accuracy simulation efficiency, which can be further boosted through GPU acceleration. However, an efficient framework combining these advances for extending simulations to large systems and long timescales remains elusive. In this work, we proposed a full-life-cycle GPU accelerated metadynamics simulations package GPU-MetaD. Benchmarking across molecular, interface, and bulk systems demonstrates that GPU-MetaD efficiently handles diverse atomic systems and delivers an order-of-magnitude performance improvement. Building on this demonstrated capability, it enables ab-initio-level rare-event sampling for systems comprising millions of atoms on a typical single GPU. This capability allows us to reveal a previously unknown size-dependent two-step nucleation mechanism in gallium nitride (GaN), highlighting the potential of GPU-MetaD for uncovering complex rare events in realistic large-scale materials systems.

physics.comp-ph↗

arXiv2026

GPUMDkit: A User-Friendly Toolkit for GPUMD and NEP

Machine-learned interatomic potentials have revolutionized molecular dynamics simulations by providing quantum-mechanical accuracy at empirical-potential speeds. The graphics processing unit molecular dynamics (GPUMD) package, featuring the highly efficient neuroevolution potential (NEP) framework, has emerged as a powerful tool in this domain. However, the complexity of force field development, active learning, and trajectory post-processing often requires extensive manual scripting, imposing a steep learning curve on new users. To address this, we present GPUMDkit, a comprehensive and user-friendly toolkit that streamlines the entire simulation workflow for GPUMD and NEP. GPUMDkit integrates a suite of essential functionalities, including format conversion, structure sampling, property calculation, and data visualization, accessible through both interactive and command-line interfaces. Its modular, extensible architecture ensures accessibility for users of all experience levels while allowing seamless integration of new features. By automating complex tasks and enhancing productivity, GPUMDkit substantially lowers the barrier to using GPUMD and NEP programs. This article describes the program architecture and demonstrates its capabilities through practical applications.

cond-mat.mtrl-sci↗

arXiv2026

Nine-element machine-learned interatomic potentials for multiphase refractory alloys

New refractory alloys are being continuously designed and characterised for applications requiring good high-temperature mechanical properties and stability. Computational design from atomistic simulations is limited by interatomic potentials missing key elements, being too inaccurate, or computationally too slow for large-scale simulations. Here we present development of a refractory alloy database and two computationally efficient and general-purpose machine-learned potentials (tabGAP and NEP). We also design a cross-sampling strategy for effective sampling of training data using predictions from two potentials with completely different underlying architecture. The potentials support arbitrary alloy compositions of elements in groups four to six in the periodic table (Ti, Zr, Hf, V, Nb, Ta, Cr, Mo, W). The database is diverse yet multitargeted to enable simulations of refractory metals and alloys across different pure-metal, solid-solution, intermetallic, and glassy phases. We demonstrate the usefulness of the potentials by reproducing known pressure-, temperature-, and solute-induced phase transitions, grain boundary segregation, and simulations of radiation damage in the WTaCrVHf metallic glass.

cond-mat.mtrl-sci↗

arXiv2026

NEP-CG and NEP-AACG: Efficient coarse-grained and multiscale all-atom-coarse-grained neuroevolution potentials

Machine-learned coarse-grained (CG) models often suffer from noisy training data, limiting their accuracy and transferability. We propose a method to generate low-noise training data based on the potential of mean force by constraining CG beads during atomistic simulations and accumulating time-averaged forces. Implemented within the neuroevolution potential (NEP) framework, our approach achieves training accuracy comparable to atomistic models trained on density functional theory data. For liquid water, the NEP-CG model accurately reproduces densities from 1 bar to 1 GPa, successfully extrapolating beyond the 0.5 GPa training limit, with a virial correction essential for the correct equation of state. For an anisotropic C$_{60}$ monolayer, distinguishing crystallographically distinct bead types reduces stress errors by an order of magnitude and captures directional thermal conductivity. We further introduce a multiscale NEP-AACG model integrating all-atom (AA) and CG degrees of freedom, demonstrated for gold nanowire fracture at an experimentally relevant strain rate. Computational speeds for NEP-CG models reach hundreds to thousands of ns/day using a single consumer-grade GPU. This work provides a robust framework for constructing accurate, transferable, and efficient CG models across diverse systems.

physics.comp-ph↗

arXiv2026

Modular hybrid machine learning and physics-based potentials for scalable modeling of van der Waals heterostructures

Accurately modeling the structural reconstruction and thermodynamic behavior of van der Waals (vdW) heterostructures remains a significant challenge due to the limitations of conventional force fields in capturing their complex mechanical, thermal, electronic, and tribological properties. To address these limitations, we develop a hybrid framework that combines single-layer machine-learned potential ($s$MLP) with physics-based anisotropic interlayer potential (ILP), effectively decoupling intralayer and interlayer interactions. This $s$MLP+ILP approach modularizes the modeling of vdW heterostructures like assembling LEGOs, reducing the required training configurations by at least an order of magnitude compared to the pure MLP approach, while retaining predictive accuracy and computational efficiency. We validate our framework by accurately reproducing the mechanical and thermal transport properties of graphite and bulk hexagonal boron nitride ($h$-BN), and by resolving intricate Moiré patterns in graphene/$h$-BN bilayer and graphene/graphene/$h$-BN trilayer heterostructures, achieving excellent agreement with experimental observations. Leveraging the developed $s$MLP+ILP approach, we reveal the stacking order-dependent formation of Moir{é} superlattice in trilayer graphene/$h$-BN/MoS$_2$ heterostructures, demonstrating its ability to accurately model large-scale vdW systems comprising hundreds of thousands of atoms with near $ab$ $initio$ precision. These findings demonstrate that the hybrid $s$MLP+ILP framework remarkably outperforms existing pure machine-learned or empirical potentials, offering a scalable and transferable solution for accurately and extensively modeling complex vdW materials across diverse applications, including sliding ferroelectricity, thermal management, resistive switching, and superlubric nanodevices.

physics.comp-ph↗

arXiv2026

qNEP: A highly efficient neuroevolution potential with dynamic charges for large-scale atomistic simulations

Although electrostatics can be incorporated into machine-learned interatomic potentials, existing approaches are computationally very demanding, limiting large-scale, long-time simulations of electrostatics-driven phenomena such as dielectric response, infrared activity, and field-matter coupling. Here, we extend the neuroevolution potential (NEP), a highly efficient machine-learned interatomic potential, to a charge-aware framework (qNEP) by introducing explicit, environment-dependent partial charges. Each ionic partial charge is represented by a neural network as a function of the local descriptor vector, analogous to the NEP site-energy model. This formulation enables the direct prediction of the Born effective charge tensor for each ion and, consequently, the polarization. As a result, dielectric properties, infrared spectra, and coupling to external electric fields can be evaluated within a unified framework. We derive consistent expressions for the forces and virials that explicitly account for the position dependence of the partial charges. The qNEP method has been implemented in the free-and-open-source GPUMD package, with support for both Ewald summation and particle-particle particle-mesh treatments of electrostatics. We demonstrate the accuracy and efficiency of the qNEP approach through representative applications to water, Li7La3Zr2O12, BaTiO3, and a magnesium-water interface. These results show that qNEP enables accurate atomistic simulations with explicit long-range electrostatics, scalable to million-atom systems on nanosecond time scales using consumer-grade GPUs.

physics.comp-ph↗

arXiv2026

GPUTB-2:An efficient E(3) network method for learning high-precision orthogonal Hamiltonian

Although equivariant neural networks have become a cornerstone for learning electronic Hamiltonians, the intrinsic non-orthogonality of linear combinations of atomic orbitals (LCAO) basis sets poses a fundamental challenge. The computational cost of Hamiltonian orthogonalization scales as O(N^3), which severely hinders electronic structure calculations for large-scale systems containing hundreds of thousands to millions of atoms. To address this issue, we develop GPUTB-2, a framework that learns implicitly orthogonality-preserving Hamiltonians by training directly on electronic band structures. Benefiting from an E(3)-equivariant network accelerated by Gaunt tensor product and SO(2) tensor product layers, GPUTB-2 achieves significantly higher accuracy than GPUTB across multiple benchmark systems. Moreover, GPUTB-2 accurately predicts large-scale electronic structures, including transport properties of temperature-perturbed SnSe and the band structures of magic-angle twisted bilayer graphene. By further integrating this framework with the linear-scaling quantum transport (LSQT) method, we investigate the electronic properties of million-atom amorphous graphene and uncover pressure-induced electronic structure transitions in more complex amorphous silicon. Together, these results establish GPUTB-2 as a high-accuracy and scalable approach for predicting orthogonal Hamiltonians.

cond-mat.mtrl-sci↗

arXiv2025

Thermal conductivities of monolayer graphene oxide from machine learning molecular dynamics simulations

Graphene oxide (GO) exhibits rich chemical heterogeneity that strongly influences its structural, thermal, and mechanical properties, yet quantitatively linking reduction chemistry to heat transport remains challenging. In this work, we develop a machine-learned neuroevolution potential (NEP) trained on an existing density functional theory dataset (\textit{Angew.\ Chem.\ Int.\ Ed.}, \textbf{63} , e202410088 (2024)), achieving reasonable accuracy at a computational cost much lower than the existing machine-learned and empirical potentials. Leveraging this potential, we perform large-scale molecular dynamics (MD) simulations to model the thermal reduction of GO across realistic structural domains. Using the homogeneous nonequilibrium MD method with a proper quantum-statistical correction scheme, we find that reduced GO exhibits strongly suppressed thermal conductivities, ranging from a few to tens of Wm$^{-1}$K$^{-1}$, substantially lower than pristine GO without defects and far below graphene. Moreover, the thermal conductivity of reduced GO increases moderately with increasing OH/O ratio, except at the highest oxidation level (O/C=0.5) where this trend inverts, while decreasing significantly with increasing O/C ratio, a trend strongly correlated with the fraction of recovered graphene-like structures. Our work provides a computationally tractable and predictive atomistic machine learning framework for exploring how chemical structure governs heat transport in heterogeneous carbon materials.

physics.chem-ph↗

arXiv2025

Anisotropic and isotropic elasticity and thermal transport in monolayer C$_{24}$ networks from machine-learning molecular dynamics

Two-dimensional fullerene networks have recently attracted increasing interest due to their diverse bonding topologies and mechanically robust architectures. In this work, we develop an accurate machine-learned potential NEP-C$_{24}$ for both the quasi-hexagonal phase (qHP) and the quasi-tetragonal phase (qTP) C$_{24}$ monolayers, based on the neuroevolution potential (NEP) framework. Using this NEP-C$_{24}$ model, we systematically investigate the elastic and thermal transport properties. Compared with C$_{60}$ monolayers, both C$_{24}$ phases exhibit markedly enhanced stiffness, arising from the combination of reduced molecular size and increased density of covalent bonds. The qTP C$_{24}$ monolayer shows nearly isotropic elastic properties and thermal conductivities along its two principal axes owing to its four-fold symmetry, whereas the chain-like, misaligned bonding topology of the qHP C$_{24}$ monolayer leads to pronounced in-plane anisotropy. Homogeneous nonequilibrium molecular dynamics and spectral decomposition analyses reveal that low-frequency ($<5$ THz) acoustic phonons dominate heat transport, with directional variations in phonon group velocity and mean free path governing the anisotropic response in qHP C$_{24}$. Real-space heat flow visualizations further show that, in these fullerene networks, phonon transport is dominated by strong inter-fullerene covalent bonds rather than weak van der Waals interactions. These findings establish a direct link between intermolecular bonding topology and phonon-mediated heat transport, providing guidance for the rational design of fullerene-based two-dimensional materials with tunable mechanical and thermal properties.

cond-mat.mtrl-sci↗

arXiv2025

Achieving Empirical Potential Efficiency with DFT Accuracy: A Neuroevolution Potential for the $α$-Fe--C--H System

A neuroevolution potential (NEP) for the ternary $α$-Fe--C--H system was developed based on a database generated from spin-polarized density functional theory (DFT) calculations, achieving empirical potential efficiency with DFT accuracy. At the same power consumption, simulation speeds using NEP are comparable to, or even faster than, those with bond order potentials. The NEP achieves DFT-level accuracy across a wide range of scenarios commonly encountered in studies of $α$-Fe- and $α$-Fe--C under hydrogen environments. The NEP enables large-scale atomistic simulations with DFT-level accuracy at the cost of empirical potentials, offering a practical tool to study hydrogen embrittlement in steel.

cond-mat.mtrl-sci↗

arXiv2025

PYSED: A tool for extracting kinetic-energy-weighted phonon dispersion and lifetime from molecular dynamics simulations

Machine learning potential-driven molecular dynamics (MD) simulations have significantly enhanced the predictive accuracy of thermal transport properties across diverse materials. However, extracting phonon-mode-resolved insights from these simulations remains a critical challenge. Here, we introduce PYSED, a Python-based package built on the spectral energy density (SED) method, designed to efficiently compute kinetic-energy-weighted phonon dispersion and extract phonon lifetime from large-scale MD simulation trajectories. By integrating high-accuracy machine-learned neuroevolution potential (NEP) models, we validate and showcase the effectiveness of the implemented SED method across systems of varying dimensionalities. Specifically, the NEP-driven MD-SED accurately reveals how phonon modes are affected by strain in carbon nanotubes, as well as by interlayer coupling strengths and the twist angles in two-dimensional molybdenum disulfide. For three-dimensional systems, the SED method effectively establishes the thermal transport regime diagram for metal-organic frameworks, distinguishing between particlelike and wavelike propagation regions. Moreover, using bulk silicon as an example, we show that phonon SED can efficiently capture quantum dynamics based on path-integral trajectories. The PYSED package bridges MD simulations with detailed phonon-mode insights, delivering a robust tool for investigating thermal transport properties with detailed mechanisms across various materials.

physics.comp-ph↗