Machine Learning Interatomic Potentials Research Articles

Goldene: An Anisotropic Metallic Monolayer with Remarkable Stability and Rigidity and Low Lattice Thermal Conductivity.

In a recent breakthrough in the field of two-dimensional (2D) nanomaterials, the first synthesis of a single-atom-thick gold lattice of goldene has been reported through an innovative wet chemical removal of Ti3C2 from the layered Ti3AuC2. Inspired by this advancement, in this communication and for the first time, a comprehensive first-principles investigation using a combination of density functional theory (DFT) and machine learning interatomic potential (MLIP) calculations has been conducted to delve into the stability, electronic, mechanical and thermal properties of the single-layer and free-standing goldene. The presented results confirm thermal stability at 700 K as well as remarkable dynamical stability of the stress-free and strained goldene monolayer. At the ground state, the elastic modulus and tensile strength of the goldene monolayer are predicted to be over 226 and 12 GPa, respectively. Through validated MLIP-based molecular dynamics calculations, it is found that at room temperature, the goldene nanosheet can exhibit anisotropic tensile strength over 9 GPa and a low lattice thermal conductivity around 10 ± 2 W/(m.K), respectively. We finally show that the native metallic nature of the goldene monolayer stays intact under large tensile strains. The combined insights from DFT and MLIP-based results provide a comprehensive understanding of the stability, mechanical, thermal and electronic properties of goldene nanosheets.

Amorphous MoS2 from a machine learning inter-atomic potential.

Amorphous molybdenum disulfide has shown potential as a hydrogen evolution catalyst, but the origin of its high activity is unclear, as is its atomic structure. Here, we have developed a classical inter-atomic potential using the charge equilibration neural network method, and we have employed it to generate atomic models of amorphous MoS2 by melting and quenching processes. The amorphous phase contains an abundance of molybdenum and sulfur atoms in low coordination. Besides the 6-coordinated molybdenum typical of the crystalline phases, a substantial fraction displays coordinations 4 and 5. The amorphous phase is also characterized by the appearance of direct S-S bonds. Density functional theory shows that the amorphous phase is metallic, with a considerable contribution of the 4-coordinated molybdenum to the density of states at the Fermi level. S-S bonds are related to the reduction of sulfur, with the excess electrons spread over several molybdenum atoms. Moreover, S-S bond formation is associated with a distinctive broadening of the 3s states, which could be exploited for experimental characterization of the amorphous phases. The large variety of local environments and the high density of electronic states at the Fermi level may play a positive role in increasing the electrocatalytic activity of this compound.

First-principles-based machine learning interatomic potential for molecular dynamics simulations of 2D lateral MoS2/WS2 heterostructures

Understanding the mechanical and thermodynamic properties of transition-metal dichalcogenides (TMDs) and their heterostructures is pivotal for advancing the development of flexible semiconductor devices, and molecular dynamics (MD) simulation is widely applied to study these properties. However, current uncertainties persist regarding the efficacy of empirical potentials in MD simulations to accurately describe the intricate performance of complex interfaces within heterostructures. This study addresses these challenges by developing an interatomic potential based on deep neural networks and first-principles calculations. Specifically focusing on MoS2/WS2 heterostructures, our approach aims to predict Young's modulus and thermal conductivities. The potential's effectiveness is demonstrated through the validation of structural features, mechanical properties, and thermodynamic characteristics, revealing close alignment with values derived from first-principles calculations. A noteworthy finding is the substantial influence of the load direction on Young's modulus of heterostructures. Furthermore, our results highlight that the interfacial thermal conductance of the MoS2/WS2 heterostructures is considerably larger than that of graphene-based interfaces. The potential developed in this work facilitates large-scale material simulations, bridging the gap with first-principles calculations. Notably, it outperforms empirical potentials under interface conditions, establishing its significant competitiveness in simulation computations. Our approach not only contributes to a deeper understanding of TMDs and heterostructures but also presents a robust tool for the simulation of their mechanical and thermal behaviors, paving the way for advancements in flexible semiconductor device manufacturing.

A moment tensor potential for lattice thermal conductivity calculations of α and β phases of Ga2O3

Calculations of heat transport in crystalline materials have recently become mainstream, thanks to machine-learned interatomic potentials that allow for significant computational cost reductions while maintaining the accuracy of first-principles calculations. Moment tensor potentials (MTPs) are among the most efficient and accurate models in this regard. In this study, we demonstrate the application of MTP to the calculation of the lattice thermal conductivity of α and β-Ga2O3. Although MTP is commonly employed for lattice thermal conductivity calculations, the advantages of applying the active learning methodology for potential generation are often overlooked. Here, we emphasize its importance and illustrate how it enables the generation of a robust and accurate interatomic potential while maintaining a moderate-sized training dataset.

Efficient calculation of unbiased atomic forces in ab initio variational Monte Carlo

quantum Monte Carlo (QMC) is a state-of-the-art numerical approach for evaluating accurate expectation values of many-body wave functions. However, one of the major drawbacks that still hinders widespread QMC applications is the lack of an affordable scheme to compute unbiased atomic forces. In this study, we propose an efficient method to obtain unbiased atomic forces and pressures in the variational Monte Carlo (VMC) framework with the Jastrow-correlated Slater determinant ansatz or the Jastrow antisymmetrized geminal power ansatz, exploiting the gauge-invariant and locality properties of their geminal representation. We demonstrate the effectiveness of our method for H2 and Cl2 molecules and for the cubic boron nitride crystal. Our framework has a better algorithmic scaling with the system size than the traditional finite-difference method and, in practical applications, is as efficient as single-point VMC calculations. Thus, it paves the way to study dynamical properties of materials, such as phonons, and is beneficial for pursuing more reliable machine-learning interatomic potentials based on unbiased VMC forces. Published by the American Physical Society 2024

Experimental and Computational Study Toward Identifying Active Sites of Supported SnOx Nanoparticles for Electrochemical CO2 Reduction Using Machine-Learned Interatomic Potentials.

SnOx has received great attention as an electrocatalyst for CO2 reduction reaction (CO2RR), however; it still suffers from low activity. Moreover, the atomic-level SnOx structure and the nature of the active sites are still ambiguous due to the dynamism of surface structure and difficulty in structure characterization under electrochemical conditions. Herein, CO2RR performance is enhanced by supporting SnO2 nanoparticles on two common supports, vulcan carbon and TiO2. Then, electrolysis of CO2 at various temperatures in a neutral electrolyte reveals that the application window for this catalyst is between 12 and 30°C. Furthermore, this study introduces a machine learning interatomic potential method for the atomistic simulation to investigate SnO2 reduction and establish a correlation between SnOx structures and their CO2RR performance. In addition, selectivity is analyzed computationally with density functional theory simulations to identify the key differences between the binding energies of *H and *CO2 -, where both are correlated with the presence of oxygen on the nanoparticle surface. This study offers in-depth insights into the rational design and application of SnOx-based electrocatalysts for CO2RR.

Constructing and Evaluating Machine-Learned Interatomic Potentials for Li-Based Disordered Rocksalts.

Lithium-based disordered rocksalts (LDRs), which are an important class of positive electrode materials that can increase the energy density of current Li-ion batteries, represent a significantly complex chemical and configurational space for conventional density functional theory (DFT)-based high-throughput screening approaches. Notably, atom-centered machine-learned interatomic potentials (MLIPs) are a promising pathway to accurately model the potential energy surface of highly disordered chemical spaces, such as LDRs, where the performance of such MLIPs has not been rigorously explored yet. Here, we represent a comprehensive evaluation of the accuracy, transferability, and ease of training of five atom-centered MLIPs, including the artificial neural network potentials developed by the atomic energy network (AENET), the Gaussian approximation potential (GAP), the spectral neighbor analysis potential (SNAP) and its quadratic extension (qSNAP), and the moment tensor potential (MTP), in modeling a 11-component LDR chemical space. Specifically, we generate a DFT-calculated data set of 10,842 configurations of disordered LiTMO2 and TMO2 compositions, where TM = Sc, Ti, V, Cr, Mn, Fe, Co, Ni, and/or Cu. To provide a point-of-comparison on the performance of atom-centered MLIPs, we also trained the neural equivariant interatomic potential (NequIP) on a subset of our data. Importantly, we find AENET to be the best potential in terms of accuracy and transferability for energy predictions, while MTP is the best for atomic forces. While AENET is the fastest to train among the MLIPs considered at low number of epochs (300), the training time increases significantly as epochs increase (3300), with a corresponding reduction in training errors (∼60%). Note that AENET and GAP tend to overfit in small data sets, with the extent of overfitting reducing with larger data sets. Finally, we observe AENET to provide reasonable predictions of average Li-intercalation voltages in layered, single-TM LiTMO2 frameworks, compared to DFT (∼10% error on average). Our study should pave the way both for discovering novel disordered rocksalt electrodes and for modeling other configurationally complex systems, such as high-entropy ceramics and alloys.

In-plane thermal conductivity of hexagonal boron nitride from 2D to 3D

The in-plane thermal conductivity of hexagonal boron nitride (h-BN) with varying thicknesses is a key property that affects the performance of various applications from electronics to optoelectronics. However, the transition of the thermal conductivity from two-dimensional (2D) to three-dimensional (3D) h-BN remains elusive. To answer this question, we have developed a machine learning interatomic potential within the neuroevolution potential (NEP) framework for h-BN, achieving a high accuracy akin to ab initio calculations in predicting its thermal conductivity and phonon transport from monolayer to multilayers and bulk. Utilizing molecular dynamics simulations based on the NEP, we predict the thermal conductivity of h-BN with a thickness up to ∼100 nm, demonstrating that its thermal conductivity quickly decreases from the monolayer and saturates to the bulk value above four layers. The saturation of its thermal conductivity is attributed to the little change in phonon group velocity and lifetime as the thickness increases beyond four layers. In particular, the weak thickness dependence of phonon lifetime in h-BN with a nanoscale thickness results from its extremely high phonon focusing along the in-plane direction. This research bridges the knowledge gap of phonon transport between 2D and 3D h-BN and will benefit the thermal design and performance optimization of relevant applications.

Thermal transport in disordered wurtzite ScAlN alloys using machine learning interatomic potentials

Disordered wurtzite ScAlN alloy is an emerging material with a strong piezoelectric coefficient that promotes its application in 5 G communication devices. To extend the service lifetime of the device, it is an urgent task to understand the thermal transport properties of this alloy for thermal management. Yet a thorough understanding of the thermal transport in the disordered ScAlN alloy is hindered by the inaccuracy of empirical interatomic potentials adopted for atomistic simulations. Herein, we developed an accurate and efficient neuro-evolution potential and combined it with Allen and Feldman theory and molecular dynamics to investigate thermal transport properties of disordered ScAlN alloys with varying ratios. This potential is shown to have the same computational accuracy as first principles by comparing the phonon spectrum and radial distribution function. Based on the homogeneous non-equilibrium molecular dynamics simulations, the lattice thermal conductivity of the disordered alloys reaches its maximum value (less than 12 W/m·K) at room temperature. Further investigation of the temperature effects on the vibrational modes shows that it affects the mode lifetimes for the propagons and the mode heat capacities for the diffusions. In addition, it was found that both mode lifetime and diffusivity decrease with increasing Sc concentration. This work aims to train a universal machine learning interatomic potential of disordered ScAlN alloys, seeking to accelerate the thermal management design of related devices.

Experiment-Driven Atomistic Materials Modeling: A Case Study Combining X-Ray Photoelectron Spectroscopy and Machine Learning Potentials to Infer the Structure of Oxygen-Rich Amorphous Carbon.

An important yet challenging aspect of atomistic materials modeling is reconciling experimental and computational results. Conventional approaches involve generating numerous configurations through molecular dynamics or Monte Carlo structure optimization and selecting the one with the closest match to experiment. However, this inefficient process is not guaranteed to succeed. We introduce a general method to combine atomistic machine learning (ML) with experimental observables that produces atomistic structures compatible with experiment by design. We use this approach in combination with grand-canonical Monte Carlo within a modified Hamiltonian formalism, to generate configurations that agree with experimental data and are chemically sound (low in energy). We apply our approach to understand the atomistic structure of oxygenated amorphous carbon (a-COx), an intriguing carbon-based material, to answer the question of how much oxygen can be added to carbon before it fully decomposes into CO and CO2. Utilizing an ML-based X-ray photoelectron spectroscopy (XPS) model trained from GW and density functional theory (DFT) data, in conjunction with an ML interatomic potential, we identify a-COx structures compliant with experimental XPS predictions that are also energetically favorable with respect to DFT. Employing a network analysis, we accurately deconvolve the XPS spectrum into motif contributions, both revealing the inaccuracies inherent to experimental XPS interpretation and granting us atomistic insight into the structure of a-COx. This method generalizes to multiple experimental observables and allows for the elucidation of the atomistic structure of materials directly from experimental data, thereby enabling experiment-driven materials modeling with a degree of realism previously out of reach.

Interatomic Interaction Models for Magnetic Materials: Recent Advances

Abstract Atomistic modeling is a widely employed theoretical method of computational materials science. It has found particular utility in the study of magnetic materials. Initially, magnetic empirical interatomic potentials or spin-polarized density functional theory (DFT) served as the primary models for describing interatomic interactions in atomistic simulations of magnetic systems. Furthermore, in recent years, a new class of interatomic potentials known as magnetic machine-learning interatomic potentials (magnetic MLIPs) has emerged. These MLIPs combine the computational efficiency, in terms of CPU time, of empirical potentials with the accuracy of DFT calculations. In this review, our focus lies on providing a comprehensive summary of the interatomic interaction models developed specifically for investigating magnetic materials. We also delve into the various problem classes to which these models can be applied. Finally, we offer insights into the future prospects of interatomic interaction model development for the exploration of magnetic materials.

Two-Step Growth Mechanism of the Solid Electrolyte Interphase in Argyrodyte/Li-Metal Contacts.

The structure and growth of the solid electrolyte interphase (SEI) region between an electrolyte and an electrode is one of the most fundamental yet less well-understood phenomena in solid-state batteries. We present an atomistic simulation of the SEI growth for one of the currently promising solid electrolytes (Li6PS5Cl), based on ab initio-trained machine learning interatomic potentials, for over 30,000 atoms during 10 ns, well beyond the capabilities of conventional molecular dynamics. This unveils a two-step growth mechanism: a Li-argyrodite chemical reaction leading to the formation of an amorphous phase, followed by a kinetically slower crystallization of the reaction products into a 5Li2S·Li3P·LiCl solid solution. The simulation results support the recent, experimentally founded hypothesis of an indirect pathway of electrolyte reduction. These findings shed light on the intricate processes governing SEI evolution, providing a valuable foundation for the design and optimization of next-generation solid-state batteries.

Machine-Learning-Driven High-Throughput Screening for High-Energy Density and Stable NASICON Cathodes.

The Na super ionic conductor (NASICON), which has outstanding structural stability and a high operating voltage, is an appealing material for overcoming the limits of low specific energy and larger volume distortion of sodium-ion batteries. In this study, to discover ideal NASICON cathode materials, a screening platform based on density functional theory (DFT) calculations and machine learning (ML) is developed. A training database was generated utilizing the previous 124 545 electrode databases, and a test set of 3126 potential NASICON structures [NaxMyM'1-y(PO4)3] with 27 dopants at the metal site and 6 dopants at the polyanion central site was constructed. The developed ML surrogate model identifies 796 materials that satisfy the following criteria: formation energy of <0.0 eV/atom, energy above hull of ≤0.025 eV/atom, volume change of ≤4%, and theoretical capacity of ≥50 mAh/g. The thermodynamically stable configurations of doped NASICON structures were then selected using machine learning interatomic potential (MLIP), enabling rapid consideration of various dopant site configurations. DFT calculations are followed on 796 screened materials to obtain energy density, average voltage, and volume change. Finally, 50 candidates with an average voltage of ≥3.5 V are identified. The suggested platform accelerates the exploration for optimal NASICON materials by narrowing the focus on materials with desired properties, saving considerable resources.

Uncertainty-biased molecular dynamics for learning uniformly accurate interatomic potentials

Efficiently creating a concise but comprehensive data set for training machine-learned interatomic potentials (MLIPs) is an under-explored problem. Active learning, which uses biased or unbiased molecular dynamics (MD) to generate candidate pools, aims to address this objective. Existing biased and unbiased MD-simulation methods, however, are prone to miss either rare events or extrapolative regions—areas of the configurational space where unreliable predictions are made. This work demonstrates that MD, when biased by the MLIP’s energy uncertainty, simultaneously captures extrapolative regions and rare events, which is crucial for developing uniformly accurate MLIPs. Furthermore, exploiting automatic differentiation, we enhance bias-forces-driven MD with the concept of bias stress. We employ calibrated gradient-based uncertainties to yield MLIPs with similar or, sometimes, better accuracy than ensemble-based methods at a lower computational cost. Finally, we apply uncertainty-biased MD to alanine dipeptide and MIL-53(Al), generating MLIPs that represent both configurational spaces more accurately than models trained with conventional MD.

Plutonium oxide melt structure and covalency.

Advances in nuclear power reactors include the use of mixed oxide fuel, containing uranium and plutonium oxides. The high-temperature behaviour and structure of PuO2-x above 1,800 K remain largely unexplored, and these conditions must be considered for reactor design and planning for the mitigation of severe accidents. Here, we measure the atomic structure of PuO2-x through the melting transition up to 3,000 ± 50 K using X-ray scattering of aerodynamically levitated and laser-beam-heated samples, with O/Pu ranging from 1.57 to 1.76. Liquid structural models consistent with the X-ray data are developed using machine-learned interatomic potentials and density functional theory. Molten PuO1.76 contains some degree of covalent Pu-O bonding, signalled by the degeneracy of Pu 5f and O 2p orbitals. The liquid is isomorphous with molten CeO1.75, demonstrating the latter as a non-radioactive, non-toxic, structural surrogate when differences in the oxidation potentials of Pu and Ce are accounted for. These characterizations provide essential constraints for modelling pertinent to reactor safety design.

Global machine learning potentials for molecular crystals.

Molecular crystals are difficult to model with accurate first-principles methods due to large unit cells. On the other hand, accurate modeling is required as polymorphs often differ by only 1kJ/mol. Machine learning interatomic potentials promise to provide accuracy of the baseline first-principles methods with a cost lower by orders of magnitude. Using the existing databases of the density functional theory calculations for molecular crystals and molecules, we train global machine learning interatomic potentials, usable for any molecular crystal. We test the performance of the potentials on experimental benchmarks and show that they perform better than classical force fields and, in some cases, are comparable to the density functional theory calculations.

Do we really need machine learning interatomic potentials for modeling amorphous metal oxides? Case study on amorphous alumina by recycling an existing ab initio database

In this study, we critically evaluate the performance of various interatomic potentials/force fields against a benchmark ab initio database for bulk amorphous alumina. The interatomic potentials tested in this work include all major fixed charge and variable charge models developed to date for alumina. Additionally, we introduce a novel machine learning interatomic potential constructed using the NequIP framework based on graph neural networks. Our findings reveal that the fixed-charge potential developed by Matsui and coworkers offers the most optimal balance between computational efficiency and agreement with ab initio data for stoichiometric alumina. Such balance cannot be provided by machine learning potentials when comparing performance with Matsui potential on the same computing infrastructure using a single Graphical Processing Unit. For non-stoichiometric alumina, the variable charge potentials, in particular ReaxFF, exhibit an impressive concordance with density functional theory calculations. However, our NequIP potentials trained on a small fraction of the ab initio database easily surpass ReaxFF in terms of both accuracy and computational performance. This is achieved without large overhead in terms of potential fitting and fine-tuning, often associated with the classical potential development process as well as training of standard deep neural network potentials, thus advocating for the use of data-efficient machine learning potentials like NequIP for complex cases of non-stoichiometric amorphous oxides.

Generating Minimal Training Sets for Machine Learned Potentials.

This Letter presents a novel approach for identifying uncorrelated atomic configurations from extensive datasets with a nonstandard neural network workflow known as random network distillation (RND) for training machine-learned interatomic potentials (MLPs). This method is coupled with a DFT workflow wherein initial data are generated with cheaper classical methods before only the minimal subset is passed to a more computationally expensive abinitio calculation. This benefits training not only by reducing the number of expensive DFT calculations required but also by providing a pathway to the use of more accurate quantum mechanical calculations. The method's efficacy is demonstrated by constructing machine-learned interatomic potentials for the molten salts KCl and NaCl. Our RND method allows accurate models to be fit on minimal datasets, as small as 32 configurations, reducing the required structures by at least 1 order of magnitude compared to alternative methods. This reduction in dataset sizes not only substantially reduces computational overhead for training data generation but also provides a more comprehensive starting point for active-learning procedures.

Thermodynamic stability and ionic conductivity in lithium–germanium binary system

Lithium–germanium binary compounds are promising anode materials for secondary lithium-ion batteries due to their high capacity, low operating voltage, and high electronic conductivity of lithiated Ge. For their successful application in batteries, it is essential to know the temperature stability of different Li–Ge phases and the variation of their ionic conductivity depending on the operating temperatures of the batteries. This work aims to comprehensively study the thermodynamic stability and ionic conductivity in Li–Ge binary compounds using a combination of first-principle computations and machine-learning interatomic potentials. We calculated convex hulls of the Li–Ge system at various temperatures and a temperature–composition phase diagram was obtained, delineating stability fields of each phase. Our calculations show that at temperatures higher than 590 K, LiGe undergoes a I41/a–P4/mmm transition, which leads to a change in the ionic conductivity. We show that all stable and metastable Li–Ge compounds have high ionic conductivity, but LiGe and Li7Ge12 have the lowest lithium diffusion. Trajectories of diffusion and Ge arrangements depend on lithium concentration. Based on advanced theoretical approaches, this study provides insights for the development of Li–Ge materials in lithium-ion and lithium-metal battery applications.

Machine-Learned Potential Energy Surfaces for Free Sodium Clusters with Density Functional Accuracy: Applications to Melting

Gaussian Process Regression based Gaussian Approximation Potential has been used to develop machine learned interatomic potentials having density functional accuracy, for free sodium clusters. The training data was generated from a large sample of over 100,000 data points computed for clusters in the size range of N = 40–200, where N denotes the number of atoms in the cluster, using the density functional method as implemented in the VASP code. Two models have been developed, model M1 using data for N = 55 only, and model M2 using additional data from larger clusters. The models are intended for computing thermodynamic properties using molecular dynamics. Hence, particular attention has been paid to improve the fitting of the forces. Although it was possible to obtain a good fit using the data of Na55 only, additional data points from larger clusters were needed to get better accuracies in energies and forces for larger sizes. Surprisingly, the model M1 could be significantly improved by adding about 50 data points per cluster from the larger sizes. Thus, it turns out that the best fit can be obtained by carefully selecting a small number of data points viz. 1,900 and 1,300 configurations for the two models M1 and M2, respectively. These two models have been deployed to compute the heat capacities of Na55. The heat capacities of Na147 and about 40 isomers for larger clusters of sizes N = 147, 200, 201, and 252, have been obtained using the final model M2. There is an excellent agreement between the computed and experimentally measured melting temperatures. The geometries of these isomers have been further optimized by density functional theory. The mean absolute error with respect to DFT energies is found to be about 7 meV/atom or less. The errors in the lowest interatomic bond lengths are below 2% in almost all the cases.