Search

Two-dimensional (2D) 1T-VSe$_2$ has prompted significant interest due to the discrepancies regarding alleged ferromagnetism (FM) at room temperature, charge density wave (CDW) states and the interplay between the two. We employed a combined Diffusion Monte Carlo (DMC) and density functional theory (DFT) approach to accurately investigate the magnetic properties and response of strain of monolayer 1T-VSe$_2$. Our calculations show the delicate competition between various phases, revealing critical insights into the relationship between their energetic and structural properties. We went on to perform Classical Monte Carlo simulations informed by our DMC and DFT results, and found the magnetic transition temperature ($T_c$) of the undistorted (non-CDW) FM phase to be 228 K and the distorted (CDW) phase to be 68 K. Additionally, we studied the response of biaxial strain on the energetic stability and magnetic properties of various phases of 2D 1T-VSe$_2$ and found that small amounts of strain can enhance the $T_c$, suggesting a promising route for engineering and enhancing magnetic behavior. Finally, we synthesized 1T-VSe$_2$ and performed Raman spectroscopy measurements, which were in close agreement with our calculated results. Our work emphasizes the role of highly accurate DMC methods in advancing the understanding of monolayer 1T-VSe$_2$ and provides a robust framework for future studies of 2D magnetic materials.

Scientific machine learning (ML) endeavors to develop generalizable models with broad applicability. However, the assessment of generalizability is often based on heuristics. Here, we demonstrate in the materials science setting that heuristics based evaluations lead to substantially biased conclusions of ML generalizability and benefits of neural scaling. We evaluate generalization performance in over 700 out-of-distribution tasks that features new chemistry or structural symmetry not present in the training data. Surprisingly, good performance is found in most tasks and across various ML models including simple boosted trees. Analysis of the materials representation space reveals that most tasks contain test data that lie in regions well covered by training data, while poorly-performing tasks contain mainly test data outside the training domain. For the latter case, increasing training set size or training time has marginal or even adverse effects on the generalization performance, contrary to what the neural scaling paradigm assumes. Our findings show that most heuristically-defined out-of-distribution tests are not genuinely difficult and evaluate only the ability to interpolate. Evaluating on such tasks rather than the truly challenging ones can lead to an overestimation of generalizability and benefits of scaling.

Large language models (LLMs) such as generative pretrained transformers (GPTs) have shown potential for various commercial applications, but their applicability for materials design remains underexplored. In this article, we introduce AtomGPT, a model specifically developed for materials design based on transformer architectures, to demonstrate the capability for both atomistic property prediction and structure generation. We show that a combination of chemical and structural text descriptions can efficiently predict material properties with accuracy comparable to graph neural network models, including formation energies, electronic bandgaps from two different methods and superconducting transition temperatures. Furthermore, we demonstrate that AtomGPT can generate atomic structures for tasks such as designing new superconductors, with the predictions validated through density functional theory calculations. This work paves the way for leveraging LLMs in forward and inverse materials design, offering an efficient approach to the discovery and optimization of materials.

High-entropy materials (HEMs) have recently emerged as a significant category of materials, offering highly tunable properties. However, the scarcity of HEM data in existing density functional theory (DFT) databases, primarily due to computational expense, hinders the development of effective modeling strategies for computational materials discovery. In this study, we introduce an open DFT dataset of alloys and employ machine learning (ML) methods to investigate the material representations needed for HEM modeling. Utilizing high-throughput DFT calculations, we generate a comprehensive dataset of 84k structures, encompassing both ordered and disordered alloys across a spectrum of up to seven components and the entire compositional range. We apply descriptor-based models and graph neural networks to assess how material information is captured across diverse chemical-structural representations. We first evaluate the in-distribution performance of ML models to confirm their predictive accuracy. Subsequently, we demonstrate the capability of ML models to generalize between ordered and disordered structures, between low-order and high-order alloys, and between equimolar and non-equimolar compositions. Our findings suggest that ML models can generalize from cost-effective calculations of simpler systems to more complex scenarios. Additionally, we discuss the influence of dataset size and reveal that the information loss associated with the use of unrelaxed structures could significantly degrade the generalization performance. Overall, this research sheds light on several critical aspects of HEM modeling and offers insights for data-driven atomistic modeling of HEMs., Comment: DFT dataset available at https://doi.org/10.5281/zenodo.10854500

The Open Databases Integration for Materials Design (OPTIMADE) application programming interface (API) empowers users with holistic access to a growing federation of databases, enhancing the accessibility and discoverability of materials and chemical data. Since the first release of the OPTIMADE specification (v1.0), the API has undergone significant development, leading to the upcoming v1.2 release, and has underpinned multiple scientific studies. In this work, we highlight the latest features of the API format, accompanying software tools, and provide an update on the implementation of OPTIMADE in contributing materials databases. We end by providing several use cases that demonstrate the utility of the OPTIMADE API in materials research that continue to drive its ongoing development.

We introduce a computational framework (InterMat) to predict band offsets of semiconductor interfaces using density functional theory (DFT) and graph neural networks (GNN). As a first step, we benchmark OptB88vdW generalized gradient approximation (GGA) work functions and electron affinities for surfaces against experimental data with accuracies of 0.29 eV and 0.39 eV, respectively. Similarly, we evaluate band offset values using independent unit (IU) and alternate slab junction (ASJ) models leading to accuracies of 0.45 eV and 0.22 eV, respectively. We use bulk band structure calculations with the TBmBJ meta-GGA functional to correct for band gap underestimation when predicting conduction band properties. During ASJ structure generation, we use Zur algorithm along with a unified GNN force-field to tackle the conformation challenges of interface design. At present, we have 607 surface work functions calculated with DFT, from which we can compute 183921 IU band offsets as well as 593 directly calculated ASJ band offsets. Finally, as the space of all possible heterojunctions is too large to simulate with DFT, we develop generalized GNN models to quickly predict bulk band edges with an accuracy of 0.26 eV. We show how these models can be used to predict relevant quantities including ionization potentials, electron affinities, and IU-based band offsets. We establish simple rules using the above models to pre-screen potential semiconductor devices from a vast pool of nearly 1.4 trillion candidate interfaces. InterMat is available at website: https://github.com/usnistgov/intermat

The observation of superconductivity in hydride-based materials under ultrahigh pressures (for example, H$_3$S and LaH$_{10}$) has fueled the interest in a more data-driven approach to discovering new high-pressure hydride superconductors. In this work, we performed density functional theory (DFT) calculations to predict the critical temperature ($T_c$) of over 900 hydride materials under a pressure range of (0 to 500) GPa, where we found 122 dynamically stable structures with a $T_c$ above MgB$_2$ (39 K). To accelerate screening, we trained a graph neural network (GNN) model to predict $T_c$ and demonstrated that a universal machine learned force-field can be used to relax hydride structures under arbitrary pressures, with significantly reduced cost. By combining DFT and GNNs, we can establish a more complete map of hydrides under pressure.

The development of large databases of material properties, together with the availability of powerful computers, has allowed machine learning (ML) modeling to become a widely used tool for predicting material performances. While confidence intervals are commonly reported for such ML models, prediction intervals, i.e., the uncertainty on each prediction, are not as frequently available. Here, we investigate three easy-to-implement approaches to determine such individual uncertainty, comparing them across ten ML quantities spanning energetics, mechanical, electronic, optical, and spectral properties. Specifically, we focused on the Quantile approach, the direct machine learning of the prediction intervals and Ensemble methods., Comment: arXiv admin note: substantial text overlap with arXiv:2107.07997

Here, we develop a framework for the prediction and screening of native defects and functional impurities in a chemical space of Group IV, III-V, and II-VI zinc blende (ZB) semiconductors, powered by crystal Graph-based Neural Networks (GNNs) trained on high-throughput density functional theory (DFT) data. Using an innovative approach of sampling partially optimized defect configurations from DFT calculations, we generate one of the largest computational defect datasets to date, containing many types of vacancies, self-interstitials, anti-site substitutions, impurity interstitials and substitutions, as well as some defect complexes. We applied three types of established GNN techniques, namely Crystal Graph Convolutional Neural Network (CGCNN), Materials Graph Network (MEGNET), and Atomistic Line Graph Neural Network (ALIGNN), to rigorously train models for predicting defect formation energy (DFE) in multiple charge states and chemical potential conditions. We find that ALIGNN yields the best DFE predictions with root mean square errors around 0.3 eV, which represents a prediction accuracy of 98 % given the range of values within the dataset, improving significantly on the state-of-the-art. Models are tested for different defect types as well as for defect charge transition levels. We further show that GNN-based defective structure optimization can take us close to DFT-optimized geometries at a fraction of the cost of full DFT. DFT-GNN models enable prediction and screening across thousands of hypothetical defects based on both unoptimized and partially-optimized defective structures, helping identify electronically active defects in technologically-important semiconductors.

Machine learning (ML) is widely used to explore crystal materials and predict their properties. However, the training is time-consuming for deep-learning models, and the regression process is a black box that is hard to interpret. Also, the preprocess to transfer a crystal structure into the input of ML, called descriptor, needs to be designed carefully. To efficiently predict important properties of materials, we propose an approach based on ensemble learning consisting of regression trees to predict formation energy and elastic constants based on small-size datasets of carbon allotropes as an example. Without using any descriptor, the inputs are the properties calculated by molecular dynamics with 9 different classical interatomic potentials. Overall, the results from ensemble learning are more accurate than those from classical interatomic potentials, and ensemble learning can capture the relatively accurate properties from the 9 classical potentials as criteria for predicting the final properties.

Lack of rigorous reproducibility and validation are major hurdles for scientific development across many fields. Materials science in particular encompasses a variety of experimental and theoretical approaches that require careful benchmarking. Leaderboard efforts have been developed previously to mitigate these issues. However, a comprehensive comparison and benchmarking on an integrated platform with multiple data modalities with both perfect and defect materials data is still lacking. This work introduces JARVIS-Leaderboard, an open-source and community-driven platform that facilitates benchmarking and enhances reproducibility. The platform allows users to set up benchmarks with custom tasks and enables contributions in the form of dataset, code, and meta-data submissions. We cover the following materials design categories: Artificial Intelligence (AI), Electronic Structure (ES), Force-fields (FF), Quantum Computation (QC) and Experiments (EXP). For AI, we cover several types of input data, including atomic structures, atomistic images, spectra, and text. For ES, we consider multiple ES approaches, software packages, pseudopotentials, materials, and properties, comparing results to experiment. For FF, we compare multiple approaches for material property predictions. For QC, we benchmark Hamiltonian simulations using various quantum algorithms and circuits. Finally, for experiments, we use the inter-laboratory approach to establish benchmarks. There are 1281 contributions to 274 benchmarks using 152 methods with more than 8 million data-points, and the leaderboard is continuously expanding. The JARVIS-Leaderboard is available at the website: https://pages.nist.gov/jarvis_leaderboard

Large-language models (LLMs) such as GPT-4 caught the interest of many scientists. Recent studies suggested that these models could be useful in chemistry and materials science. To explore these possibilities, we organized a hackathon. This article chronicles the projects built as part of this hackathon. Participants employed LLMs for various applications, including predicting properties of molecules and materials, designing novel interfaces for tools, extracting knowledge from unstructured data, and developing new educational applications. The diverse topics and the fact that working prototypes could be generated in less than two days highlight that LLMs will profoundly impact the future of our fields. The rich collection of ideas and projects also indicates that the applications of LLMs are not limited to materials science and chemistry but offer potential benefits to a wide range of scientific disciplines.

The Joint Automated Repository for Various Integrated Simulations (JARVIS) infrastructure at the National Institute of Standards and Technology (NIST) is a large-scale collection of curated datasets and tools with more than 80000 materials and millions of properties. JARVIS uses a combination of electronic structure, artificial intelligence (AI), advanced computation and experimental methods to accelerate materials design. Here we report some of the new features that were recently included in the infrastructure such as: 1) doubling the number of materials in the database since its first release, 2) including more accurate electronic structure methods such as Quantum Monte Carlo, 3) including graph neural network-based materials design, 4) development of unified force-field, 5) development of a universal tight-binding model, 6) addition of computer-vision tools for advanced microscopy applications, 7) development of a natural language processing tool for text-generation and analysis, 8) debuting a large-scale benchmarking endeavor, 9) including quantum computing algorithms for solids, 10) integrating several experimental datasets and 11) staging several community engagement and outreach events. New classes of materials, properties, and workflows added to the database include superconductors, two-dimensional (2D) magnets, magnetic topological materials, metal-organic frameworks, defects, and interface systems. The rich and reliable datasets, tools, documentation, and tutorials make JARVIS a unique platform for modern materials design. JARVIS ensures openness of data and tools to enhance reproducibility and transparency and to promote a healthy and collaborative scientific environment.

Extensive efforts to gather materials data have largely overlooked potential data redundancy. In this study, we present evidence of a significant degree of redundancy across multiple large datasets for various material properties, by revealing that up to 95 % of data can be safely removed from machine learning training with little impact on in-distribution prediction performance. The redundant data is related to over-represented material types and does not mitigate the severe performance degradation on out-of-distribution samples. In addition, we show that uncertainty-based active learning algorithms can construct much smaller but equally informative datasets. We discuss the effectiveness of informative data in improving prediction performance and robustness and provide insights into efficient data acquisition and machine learning training. This work challenges the "bigger is better" mentality and calls for attention to the information richness of materials data rather than a narrow emphasis on data volume., Comment: Main text: 9 pages, 2 tables, 6 figures. Supplemental information: 31 pages, 1 table, 24 figures

Finding new superconductors with a high critical temperature ($T_c$) has been a challenging task due to computational and experimental costs. We present a diffusion model inspired by the computer vision community to generate new superconductors with unique structures and chemical compositions. Specifically, we used a crystal diffusion variational autoencoder (CDVAE) along with atomistic line graph neural network (ALIGNN) pretrained models and the Joint Automated Repository for Various Integrated Simulations (JARVIS) superconducting database of density functional theory (DFT) calculations to generate new superconductors with a high success rate. We started with a DFT dataset of $\approx$1000 superconducting materials to train the diffusion model. We used the model to generate 3000 new structures, which along with pre-trained ALIGNN screening results in 61 candidates. For the top candidates, we performed DFT calculations for validation. Such approaches go beyond the funnel-like materials design approaches and allow for the inverse design of next-generation materials.

Computer vision techniques have immense potential for materials design applications. In this work, we introduce an integrated and general-purpose AtomVision library that can be used to generate, curate scanning tunneling microscopy (STM) and scanning transmission electron microscopy (STEM) datasets and apply machine learning techniques. To demonstrate the applicability of this library, we 1) generate and curate an atomistic image dataset of about 10000 materials, 2) develop and compare convolutional and graph neural network models to classify the Bravais lattices, 3) develop fully convolutional neural network using U-Net architecture to pixelwise classify atom vs background, 4) use generative adversarial network for super-resolution, 5) curate a natural language processing based image dataset using open-access arXiv dataset, and 6) integrate the computational framework with experimental microscopy tools. AtomVision library is available at https://github.com/usnistgov/atomvision.

High-throughput density functional theory (DFT) calculations allow for a systematic search for conventional superconductors. With the recent interest in two-dimensional (2D) superconductors, we used a high-throughput workflow to screen over 1,000 2D materials in the JARVIS-DFT database and performed electron-phonon coupling calculations, using the McMillan-Allen-Dynes formula to calculate the superconducting transition temperature ($T_c$) for 165 of them. Of these 165 materials, we identify 34 dynamically stable structures with transition temperatures above 5 K, including materials such as W$_2$N$_3$, NbO$_2$, ZrBrO, TiClO, NaSn$_2$S$_4$, Mg$_2$B$_4$C$_2$ and the previously unreported Mg$_2$B$_4$N$_2$ ($T_c$ = 21.8 K). Finally, we performed experiments to determine the $T_c$ of selected layered superconductors (2H-NbSe$_2$, 2H-NbS$_2$, ZrSiS, FeSe) and discuss the measured results within the context of our DFT results. We aim that the outcome of this workflow can guide future computational and experimental studies of new and emerging 2D superconductors by providing a roadmap of high-throughput DFT data.

Recent advances in machine learning (ML) methods have led to substantial improvement in materials property prediction against community benchmarks, but an excellent benchmark score may not imply good generalization of performance. Here we show that ML models trained on the Materials Project 2018 (MP18) dataset can have severely degraded prediction performance on new compounds in the Materials Project 2021 (MP21) dataset. We document performance degradation in graph neural networks and traditional descriptor-based ML models for both quantitative and qualitative predictions. We find the source of the predictive degradation is due to the distribution shift between the MP18 and MP21 versions. This is revealed by the uniform manifold approximation and projection (UMAP) of the feature space. We then show that the performance degradation issue can be foreseen using a few simple tools. Firstly, the UMAP can be used to investigate the connectivity and relative proximity of the training and test data within feature space. Secondly, the disagreement between multiple ML models on the test data can illuminate out-of-distribution samples. We demonstrate that the simple yet efficient UMAP-guided and query-by-committee acquisition strategies can greatly improve prediction accuracy through adding only 1~\% of the test data. We believe this work provides valuable insights for building materials databases and ML models that enable better prediction robustness and generalizability.

The search for two-dimensional (2D) magnetic materials has attracted a great deal of attention because of the experimental synthesis of 2D CrI$_3$, which has a measured Curie temperature of 45 K. Often times, these monolayers have a higher degree of electron correlation and require more sophisticated methods beyond density functional theory (DFT). Diffusion Monte Carlo (DMC) is a correlated electronic structure method that has been demonstrated successful for calculating the electronic and magnetic properties of a wide variety of 2D and bulk systems, since it has a weaker dependence on the Hubbard parameter (U) and density functional. In this study we designed a workflow that combines DFT+U and DMC in order to treat 2D correlated magnetic systems. We chose monolayer CrX$_3$ (X = I, Br, Cl, F), with a stronger focus on CrI$_3$ and CrBr$_3$, as a case study due to the fact that they have been experimentally realized and have a finite critical temperature. With this DFT+U and DMC workflow and the analytical method of Torelli and Olsen, we estimated an upper bound of 43.56 K for the T$_c$ of CrI$_3$ and 20.78 K for the T$_c$ of CrBr$_3$, in addition to analyzing the spin densities and magnetic properties with DMC and DFT+U. We expect that running this workflow for a well-known material class will aid in the future discovery and characterization of lesser known and more complex correlated 2D magnetic materials.

In this work, we present the ChemNLP library that can be used for 1) curating open access datasets for materials and chemistry literature, developing and comparing traditional machine learning, transformers and graph neural network models for 2) classifying and clustering texts, 3) named entity recognition for large-scale text-mining, 4) abstractive summarization for generating titles of articles from abstracts, 5) text generation for suggesting abstracts from titles, 6) integration with density functional theory dataset for identifying potential candidate materials such as superconductors, and 7) web-interface development for text and reference query. We primarily use the publicly available arXiv and Pubchem datasets but the tools can be used for other datasets as well. Moreover, as new models are developed, they can be easily integrated in the library. ChemNLP is available at the websites: https://github.com/usnistgov/chemnlp and https://jarvis.nist.gov/jarvischemnlp.

Classical force fields (FF) based on machine learning (ML) methods show great potential for large scale simulations of materials. MLFFs have hitherto largely been designed and fitted for specific systems and are not usually transferable to chemistries beyond the specific training set. We develop a unified atomisitic line graph neural network-based FF (ALIGNN-FF) that can model both structurally and chemically diverse materials with any combination of 89 elements from the periodic table. To train the ALIGNN-FF model, we use the JARVIS-DFT dataset which contains around 75000 materials and 4 million energy-force entries, out of which 307113 are used in the training. We demonstrate the applicability of this method for fast optimization of atomic structures in the crystallography open database and by predicting accurate crystal structures using genetic algorithm for alloys.

The phonon density-of-states (DOS) summarizes the lattice vibrational modes supported by a structure, and gives access to rich information about the material's stability, thermodynamic constants, and thermal transport coefficients. Here, we present an atomistic line graph neural network (ALIGNN) model for the prediction of the phonon density of states and the derived thermal and thermodynamic properties. The model is trained on a database of over 14,000 phonon spectra included in the JARVIS-DFT (Joint Automated Repository for Various Integrated Simulations: Density Functional Theory) database. The model predictions are shown to capture the spectral features of the phonon density-of-states, effectively categorize dynamical stability, and lead to accurate predictions of DOS-derived thermal and thermodynamic properties, including heat capacity $C_{\mathrm{V}}$, vibrational entropy $S_{\mathrm{vib}}$, and the isotopic phonon scattering rate $\tau^{-1}_{\mathrm{i}}$. The DOS-mediated ALIGNN model provides superior predictions when compared to a direct deep-learning prediction of these material properties as well as predictions based on analytic simplifications of the phonon DOS, including the Debye or Born-von Karman models. Finally, the ALIGNN model is used to predict the phonon spectra and properties for about 40,000 additional materials listed in the JARVIS-DFT database, which are validated as far as possible against other open-sourced high-throughput DFT phonon databases.

We envision an autonomous sorbent materials foundry (SMF) for rapidly evaluating materials for direct air capture of carbon dioxide (CO2), specifically targeting novel metal organic framework materials. Our proposed SMF is hierarchical, simultaneously addressing the most critical gaps in the inter-related space of sorbent material synthesis, processing, properties, and performance. The ability to collect these critical data streams in an agile, coordinated, and automated fashion will enable efficient end-to-end sorbent materials design through machine learning driven research framework.

The presence of point defects such as vacancies plays an important role in material design. Here, we demonstrate that a graph neural network (GNN) model trained only on perfect materials can also be used to predict vacancy formation energies ($E_{vac}$) of defect structures without the need for additional training data. Such GNN-based predictions are considerably faster than density functional theory (DFT) calculations with reasonable accuracy and show the potential that GNNs are able to capture a functional form for energy predictions. To test this strategy, we developed a DFT dataset of 508 $E_{vac}$ consisting of 3D elemental solids, alloys, oxides, nitrides, and 2D monolayer materials. We analyzed and discussed the applicability of such direct and fast predictions. We applied the model to predict 192494 $E_{vac}$ for 55723 materials in the JARVIS-DFT database.

We develop a multi-step workflow for the discovery of conventional superconductors, starting with a Bardeen Cooper Schrieffer inspired pre-screening of 1736 materials with high Debye temperature and electronic density of states. Next, we perform electron-phonon coupling calculations for 1058 of them to establish a large and systematic database of BCS superconducting properties. Using the McMillan-Allen-Dynes formula, we identify 105 dynamically stable materials with transition temperatures, Tc>5 K. Additionally, we analyze trends in our dataset and individual materials including MoN, VC, VTe, KB6, Ru3NbC, V3Pt, ScN, LaN2, RuO2, and TaC. We demonstrate that deep-learning(DL) models can predict superconductor properties faster than direct first principles computations. Notably, we find that by predicting the Eliashberg function as an intermediate quantity, we can improve model performance versus a direct DL prediction of Tc. We apply the trained models on the crystallographic open database and pre-screen candidates for further DFT calculations.

Machine learning (ML) based models have greatly enhanced the traditional materials discovery and design pipeline. Specifically, in recent years, surrogate ML models for material property prediction have demonstrated success in predicting discrete scalar-valued target properties to within reasonable accuracy of their DFT-computed values. However, accurate prediction of spectral targets such as the electron Density of States (DOS) poses a much more challenging problem due to the complexity of the target, and the limited amount of available training data. In this study, we present an extension of the recently developed Atomistic Line Graph Neural Network (ALIGNN) to accurately predict DOS of a large set of material unit cell structures, trained to the publicly available JARVIS-DFT dataset. Furthermore, we evaluate two methods of representation of the target quantity - a direct discretized spectrum, and a compressed low-dimensional representation obtained using an autoencoder. Through this work, we demonstrate the utility of graph-based featurization and modeling methods in the prediction of complex targets that depend on both chemistry and directional characteristics of material structures.

Parameterized tight-binding models fit to first principles calculations can provide an efficient and accurate quantum mechanical method for predicting properties of molecules and solids. However, well-tested parameter sets are generally only available for a limited number of atom combinations, making routine use of this method difficult. Furthermore, most previous models consider only simple two-body interactions, which limits accuracy. To tackle these challenges, we develop a density functional theory database of nearly one million materials, which we use to fit a universal set of tight-binding parameters for 65 elements and their binary combinations. We include both two-body and three-body effective interaction terms in our model, plus self-consistent charge transfer, enabling our model to work for metallic, covalent, and ionic bonds with the same parameter set. To ensure predictive power, we adopt a learning framework where we repeatedly test the model on new low energy crystal structures and then add them to the fitting dataset, iterating until predictions improve. We distribute the materials database and tools developed in this work publicly., Comment: updated version

The increasing CO2 level is a critical concern and suitable materials are needed to capture such gases from the environment. While experimental and conventional computational methods are useful in finding such materials, they are usually slow and there is a need to expedite such processes. We use Atomistic Line Graph Neural Network (ALIGNN) method to predict CO2 adsorption in metal organic frameworks (MOF), which are known for their high functional tunability. We train ALIGNN models for hypothetical MOF (hMOF) database with 137953 MOFs with grand canonical Monte Carlo (GCMC) based CO2 adsorption isotherms. We develop high accuracy and fast models for pre-screening applications. We apply the trained model on CoREMOF database and computationally rank them for experimental synthesis. In addition to the CO2 adsorption isotherm, we also train models for electronic bandgaps, surface area, void fraction, lowest cavity diameter, and pore limiting diameter, and illustrate the strength and limitation of such graph neural network models. For a few candidate MOFs we carry out GCMC calculations to evaluate the deep-learning (DL) predictions.

Lattice vibration frequencies are related to many important materials properties such as thermal and electrical conductivity as well as superconductivity. However, computational calculation of vibration frequencies using density functional theory (DFT) methods is too computationally demanding for a large number of samples in materials screening. Here we propose a deep graph neural network-based algorithm for predicting crystal vibration frequencies from crystal structures with high accuracy. Our algorithm addresses the variable dimension of vibration frequency spectrum using the zero padding scheme. Benchmark studies on two data sets with 15,000 and 35,552 samples show that the aggregated $R^2$ scores of the prediction reaches 0.554 and 0.724 respectively. Our work demonstrates the capability of deep graph neural networks to learn to predict phonon spectrum properties of crystal structures in addition to phonon density of states (DOS) and electronic DOS in which the output dimension is constant., Comment: 9 pages

Deep learning (DL) is one of the fastest growing topics in materials data science, with rapidly emerging applications spanning atomistic, image-based, spectral, and textual data modalities. DL allows analysis of unstructured data and automated identification of features. Recent development of large materials databases has fueled the application of DL methods in atomistic prediction in particular. In contrast, advances in image and spectral data have largely leveraged synthetic data enabled by high quality forward models as well as by generative unsupervised DL methods. In this article, we present a high-level overview of deep-learning methods followed by a detailed discussion of recent developments of deep learning in atomistic simulation, materials imaging, spectral analysis, and natural language processing. For each modality we discuss applications involving both theoretical and experimental data, typical modeling approaches with their strengths and limitations, and relevant publicly available software and datasets. We conclude the review with a discussion of recent cross-cutting work related to uncertainty quantification in this field and a brief perspective on limitations, challenges, and potential growth areas for DL methods in materials science. The application of DL methods in materials science presents an exciting avenue for future materials discovery and design.

Uncertainty quantification in Artificial Intelligence (AI)-based predictions of material properties is of immense importance for the success and reliability of AI applications in material science. While confidence intervals are commonly reported for machine learning (ML) models, prediction intervals, i.e., the evaluation of the uncertainty on each prediction, are seldomly available. In this work we compare 3 different approaches to obtain such individual uncertainty, testing them on 12 ML-physical properties. Specifically, we investigated using the Quantile loss function, machine learning the prediction intervals directly and using Gaussian Processes. We identify each approachs advantages and disadvantages and end up slightly favoring the modeling of the individual uncertainties directly, as it is the easiest to fit and, in most cases, minimizes over-and under-estimation of the predicted errors. All data for training and testing were taken from the publicly available JARVIS-DFT database, and the codes developed for computing the prediction intervals are available through JARVIS-Tools.

Graph neural networks (GNN) have been shown to provide substantial performance improvements for atomistic material representation and modeling compared with descriptor-based machine learning models. While most existing GNN models for atomistic predictions are based on atomic distance information, they do not explicitly incorporate bond angles, which are critical for distinguishing many atomic structures. Furthermore, many material properties are known to be sensitive to slight changes in bond angles. We present an Atomistic Line Graph Neural Network (ALIGNN), a GNN architecture that performs message passing on both the interatomic bond graph and its line graph corresponding to bond angles. We demonstrate that angle information can be explicitly and efficiently included, leading to improved performance on multiple atomistic prediction tasks. We ALIGNN models for predicting 52 solid-state and molecular properties available in the JARVIS-DFT, Materials project, and QM9 databases. ALIGNN can outperform some previously reported GNN models on atomistic prediction tasks by up to 85% in accuracy with better or comparable model training speed.

In this study, we present a novel approach along with the needed computational strategies for efficient and scalable feature engineering of the crystal structure in compounds of different chemical compositions. This approach utilizes a versatile and extensible framework for the quantification of a three-dimensional (3-D) voxelized crystal structure in the form of 2-point spatial correlations of multiple atomic attributes and performs principal component analysis to extract the low-dimensional features that could be used to build surrogate models for material properties of interest. An application of the proposed feature engineering framework is demonstrated on a case study involving the prediction of the formation energies of crystalline compounds using two vastly different surrogate model building strategies - local Gaussian process regression and neural networks. Specifically, it is shown that the top 25 features (i.e., principal component scores) identified by the proposed framework serve as good regressors for the formation energy of the crystalline substance for both model building strategies.

The Open Databases Integration for Materials Design (OPTIMADE) consortium has designed a universal application programming interface (API) to make materials databases accessible and interoperable. We outline the first stable release of the specification, v1.0, which is already supported by many leading databases and several software packages. We illustrate the advantages of the OPTIMADE API through worked examples on each of the public materials databases that support the full API specification., Comment: 11 pages, 1 table

Quantum chemistry is one of the most promising near-term applications of quantum computers. Quantum algorithms such as variational quantum eigen solver (VQE) and variational quantum deflation (VQD) algorithms have been mainly applied for molecular systems and there is a need to implement such methods for periodic solids. Using Wannier tight-binding Hamiltonian (WTBH) approaches, we demonstrate the application of VQE and VQD to accurately predict both electronic and phonon bandstructure properties of several elemental as well as multi-component solid-state materials. We apply VQE-VQD calculations for 307 spin-orbit coupling based electronic WTBHs and 933 finite-difference based phonon WTBHs. Also, we discuss a workflow for using VQD with lattice Greens function that can be used for solving dynamical mean-field theory problems. The WTBH model solvers can be used for testing other quantum algorithms and models also.

Magnetic topological insulators and semi-metals have a variety of properties that make them attractive for applications including spintronics and quantum computation, but very few high-quality candidate materials are known. In this work, we use systematic high-throughput density functional theory calculations to identify magnetic topological materials from 40000 three-dimensional materials in the JARVIS-DFT database (https://jarvis.nist.gov/jarvisdft). First, we screen materials with net magnetic moment > 0.5 {\mu}B and spin-orbit spillage > 0.25, resulting in 25 insulating and 564 metallic candidates. The spillage acts as a signature of spin-orbit induced band-inversion. Then, we carry out calculations of Wannier charge centers, Chern numbers, anomalous Hall conductivities, surface bandstructures, and Fermi-surfaces to determine interesting topological characteristics of the screened compounds. We also train machine learning models for predicting the spillage, bandgaps, and magnetic moments of new compounds, to further accelerate the screening process. We experimentally synthesize and characterize a few candidate materials to support our theoretical predictions.

Materials with van der Waals-bonding are known to exhibit quantum confinement effect, in which the electronic bandgap of the three-dimensional (3D) realization of a material is lower than that of its two-dimensional (2D) counterpart. However, the possibility of an anomalous quantum confinement effect (AQCE) exists, where the bandgap trend is reversed. In this work, we computationally identify materials for which such AQCE occurs. Using density functional theory (DFT), we compute ~1000 OptB88vdW (semi-local functional), ~50 HSE06 and ~50 PBE0 (hybrid functional) bandgaps for bulk and their corresponding monolayers in the JARVIS-DFT database. OptB88vdW identifies 65 AQCE materials, but the hybrid functionals only confirm such finding in 14 cases. Some of the AQCE systems identified through HSE06 and PBE0 are: hydroxides or oxide hydroxide compounds (AlOH2, Mg(OH)2, Mg2H2O3, Ni(OH)2, SrH2O3) as well as Sb-halogen-chalcogenide compounds (SbSBr, SbSeI) and alkali-chalcogenides (RbLiS and RbLiSe). A detailed electronic structure analysis, based on band-structure and projected density of states, shows AQCE is often characterized by lowering of the conduction band in the monolayer and corresponding changes in the pz electronic orbital contribution, with z being the non-periodic direction in the 2D case. We believe our computational results would spur the effort to validate the results experimentally and will have impact on bandgap engineering applications based on low-dimensional materials.

The Joint Automated Repository for Various Integrated Simulations (JARVIS) is an integrated infrastructure to accelerate materials discovery and design using density functional theory (DFT), classical force-fields (FF), and machine learning (ML) techniques. JARVIS is motivated by the Materials Genome Initiative (MGI) principles of developing open-access databases and tools to reduce the cost and development time of materials discovery, optimization, and deployment. The major features of JARVIS are: JARVIS-DFT, JARVIS-FF, JARVIS-ML, and JARVIS-Tools. To date, JARVIS consists of 40,000 materials and 1 million calculated properties in JARVIS-DFT, 1,500 materials and 110 force-fields in JARVIS-FF, and 25 ML models for material-property predictions in JARVIS-ML, all of which are continuously expanding. JARVIS-Tools provides scripts and workflows for running and analyzing various simulations. We compare our computational data to experiments or high-fidelity computational methods wherever applicable to evaluate error/uncertainty in predictions. In addition to the existing workflows, the infrastructure can support a wide variety of other technologically important applications as part of the data-driven materials design paradigm. The JARVIS datasets and tools are publicly available at the website: https://jarvis.nist.gov .