Explore the vast range of titles available.

A simultaneously accurate and computationally efficient parametrization of the potential energy surface of molecules and materials is a long-standing goal in the natural sciences. While atom-centered message passing neural networks (MPNNs) have shown remarkable accuracy, their information propagation has limited the accessible length-scales. Local methods, conversely, scale to large simulations but have suffered from inferior accuracy. This work introduces Allegro, a strictly local equivariant deep neural network interatomic potential architecture that simultaneously exhibits excellent accuracy and scalability. Allegro represents a many-body potential using iterated tensor products of learned equivariant representations without atom-centered message passing. Allegro obtains improvements over state-of-the-art methods on QM9 and revMD17. A single tensor product layer outperforms existing deep MPNNs and transformers on QM9. Furthermore, Allegro displays remarkable generalization to out-of-distribution data. Molecular simulations using Allegro recover structural and kinetic properties of an amorphous electrolyte in excellent agreement with ab-initio simulations. Finally, we demonstrate parallelization with a simulation of 100 million atoms. The paper presents a method that allows scaling machine learning interatomic potentials to extremely large systems, while at the same time retaining the remarkable accuracy and learning efficiency of deep equivariant models. This is obtained with an E(3)- equivariant neural network architecture that combines the high accuracy of equivariant neural networks with the scalability of local methods.

Although historically materials discovery has been driven by a laborious trial-and-error process, knowledge-driven materials design can now be enabled by the rational combination of Machine Learning methods and materials databases. Here, data from the AFLOW repository for ab initio calculations is combined with Quantitative Materials Structure-Property Relationship models to predict important properties: metal/insulator classification, band gap energy, bulk/shear moduli, Debye temperature and heat capacities. The prediction’s accuracy compares well with the quality of the training data for virtually any stoichiometric inorganic crystalline material, reciprocating the available thermomechanical experimental data. The universality of the approach is attributed to the construction of the descriptors: Property-Labelled Materials Fragments. The representations require only minimal structural input allowing straightforward implementations of simple heuristic design rules. Machine learning methods can be useful for materials discovery; however certain properties remain difficult to predict. Here, the authors present a universal machine learning approach for modelling the properties of inorganic crystals, which is validated for eight electronic and thermomechanical properties.

Computational material discovery is under intense study owing to its ability to explore the vast space of chemical systems. Neural network potentials (NNPs) have been shown to be particularly effective in conducting atomistic simulations for such purposes. However, existing NNPs are generally designed for narrow target materials, making them unsuitable for broader applications in material discovery. Here we report a development of universal NNP called PreFerred Potential (PFP), which is able to handle any combination of 45 elements. Particular emphasis is placed on the datasets, which include a diverse set of virtual structures used to attain the universality. We demonstrated the applicability of PFP in selected domains: lithium diffusion in LiFeSO 4 F, molecular adsorption in metal-organic frameworks, an order–disorder transition of Cu-Au alloys, and material discovery for a Fischer–Tropsch catalyst. They showcase the power of PFP, and this technology provides a highly useful tool for material discovery. Existing neural network potentials are generally designed for narrow target materials. Here the authors develop a neural network potential which is able to handle any combination of 45 elements and show its applicability in multiple domains.

Machine learning potentials have become an important tool for atomistic simulations in many fields, from chemistry via molecular biology to materials science. Most of the established methods, however, rely on local properties and are thus unable to take global changes in the electronic structure into account, which result from long-range charge transfer or different charge states. In this work we overcome this limitation by introducing a fourth-generation high-dimensional neural network potential that combines a charge equilibration scheme employing environment-dependent atomic electronegativities with accurate atomic energies. The method, which is able to correctly describe global charge distributions in arbitrary systems, yields much improved energies and substantially extends the applicability of modern machine learning potentials. This is demonstrated for a series of systems representing typical scenarios in chemistry and materials science that are incorrectly described by current methods, while the fourth-generation neural network potential is in excellent agreement with electronic structure calculations. Machine learning potentials do not account for long-range charge transfer. Here the authors introduce a fourth-generation high-dimensional neural network potential including non-local information of charge populations that is able to provide forces, charges and energies in excellent agreement with DFT data.

High-entropy materials have attracted considerable interest due to the combination of useful properties and promising applications. Predicting their formation remains the major hindrance to the discovery of new systems. Here we propose a descriptor—entropy forming ability—for addressing synthesizability from first principles. The formalism, based on the energy distribution spectrum of randomized calculations, captures the accessibility of equally-sampled states near the ground state and quantifies configurational disorder capable of stabilizing high-entropy homogeneous phases. The methodology is applied to disordered refractory 5-metal carbides—promising candidates for high-hardness applications. The descriptor correctly predicts the ease with which compositions can be experimentally synthesized as rock-salt high-entropy homogeneous phases, validating the ansatz, and in some cases, going beyond intuition. Several of these materials exhibit hardness up to 50% higher than rule of mixtures estimations. The entropy descriptor method has the potential to accelerate the search for high-entropy systems by rationally combining first principles with experimental synthesis and characterization. The overwhelming number of possible high-entropy materials represents a big challenge for predicting their existence. Here, the authors introduce an entropy-forming-ability descriptor capturing the synthesizability of these systems, and apply the model to the discovery of new refractory metal carbides.

This work presents Neural Equivariant Interatomic Potentials (NequIP), an E(3)-equivariant neural network approach for learning interatomic potentials from ab-initio calculations for molecular dynamics simulations. While most contemporary symmetry-aware models use invariant convolutions and only act on scalars, NequIP employs E(3)-equivariant convolutions for interactions of geometric tensors, resulting in a more information-rich and faithful representation of atomic environments. The method achieves state-of-the-art accuracy on a challenging and diverse set of molecules and materials while exhibiting remarkable data efficiency. NequIP outperforms existing models with up to three orders of magnitude fewer training data, challenging the widely held belief that deep neural networks require massive training sets. The high data efficiency of the method allows for the construction of accurate potentials using high-order quantum chemical level of theory as reference and enables high-fidelity molecular dynamics simulations over long time scales. An E(3)-equivariant deep learning interatomic potential is introduced for accelerating molecular dynamics simulations. The method obtains state-of-the-art accuracy, can faithfully describe dynamics of complex systems with remarkable sample efficiency.

ChatMOF is an artificial intelligence (AI) system that is built to predict and generate metal-organic frameworks (MOFs). By leveraging a large-scale language model (GPT-4, GPT-3.5-turbo, and GPT-3.5-turbo-16k), ChatMOF extracts key details from textual inputs and delivers appropriate responses, thus eliminating the necessity for rigid and formal structured queries. The system is comprised of three core components (i.e., an agent, a toolkit, and an evaluator) and it forms a robust pipeline that manages a variety of tasks, including data retrieval, property prediction, and structure generations. ChatMOF shows high accuracy rates of 96.9% for searching, 95.7% for predicting, and 87.5% for generating tasks with GPT-4. Additionally, it successfully creates materials with user-desired properties from natural language. The study further explores the merits and constraints of utilizing large language models (LLMs) in combination with database and machine learning in material sciences and showcases its transformative potential for future advancements. LLMs can be augmented with tools to increase their capabilities. Here, authors have developed an artificial intelligence system called ChatMOF combining LLMs and specialised libraries and utilities to predict and generate metal-organic frameworks.

Machine-learned potentials (MLPs) have exhibited remarkable accuracy, yet the lack of general-purpose MLPs for a broad spectrum of elements and their alloys limits their applicability. Here, we present a promising approach for constructing a unified general-purpose MLP for numerous elements, demonstrated through a model (UNEP-v1) for 16 elemental metals and their alloys. To achieve a complete representation of the chemical space, we show, via principal component analysis and diverse test datasets, that employing one-component and two-component systems suffices. Our unified UNEP-v1 model exhibits superior performance across various physical properties compared to a widely used embedded-atom method potential, while maintaining remarkable efficiency. We demonstrate our approach’s effectiveness through reproducing experimentally observed chemical order and stable phases, and large-scale simulations of plasticity and primary radiation damage in MoTaVW alloys. Machine-learned potentials are accurate but often lack broad applicability. Here, authors develop a general-purpose neuroevolution potential for 16 metals and their alloys, achieving efficient and accurate predictions of various physical properties.

Symbolic regression (SR) is an approach of interpretable machine learning for building mathematical formulas that best fit certain datasets. In this work, SR is used to guide the design of new oxide perovskite catalysts with improved oxygen evolution reaction (OER) activities. A simple descriptor, μ / t , where μ and t are the octahedral and tolerance factors, respectively, is identified, which accelerates the discovery of a series of new oxide perovskite catalysts with improved OER activity. We successfully synthesise five new oxide perovskites and characterise their OER activities. Remarkably, four of them, Cs 0.4 La 0.6 Mn 0.25 Co 0.75 O 3 , Cs 0.3 La 0.7 NiO 3 , SrNi 0.75 Co 0.25 O 3 , and Sr 0.25 Ba 0.75 NiO 3 , are among the oxide perovskite catalysts with the highest intrinsic activities. Our results demonstrate the potential of SR for accelerating the data-driven design and discovery of new materials with improved properties. Symbolic regression holds big promise for guiding materials design, yet its application in materials science is still limited. Here the authors use symbolic regression to introduce an activity descriptor predicting new oxide perovskites with improved oxygen evolution activity as corroborated by experimental validation.

Machine learning has the potential to accelerate materials discovery by accurately predicting materials properties at a low computational cost. However, the model inputs remain a key stumbling block. Current methods typically use descriptors constructed from knowledge of either the full crystal structure — therefore only applicable to materials with already characterised structures — or structure-agnostic fixed-length representations hand-engineered from the stoichiometry. We develop a machine learning approach that takes only the stoichiometry as input and automatically learns appropriate and systematically improvable descriptors from data. Our key insight is to treat the stoichiometric formula as a dense weighted graph between elements. Compared to the state of the art for structure-agnostic methods, our approach achieves lower errors with less data. Predicting the structure of unknown materials’ compositions represents a challenge for high-throughput computational approaches. Here the authors introduce a new stoichiometry-based machine learning approach for predicting the properties of inorganic materials from their elemental compositions.