Computational Screening of All Stoichiometric Inorganic Materials

Mapping inorganic crystal chemical space

The combination of elements from the Periodic Table defines a vast chemical space. Only a small fraction of these combinations yields materials that occur naturally or are accessible synthetically. Here, we enumerate binary, ternary, and quaternary element and species combinations to produce an extensive library of over 1010 stoichiometric inorganic compositions. The unique combinations are vectorised using compositional embedding vectors drawn from a variety of published machine-learning models. Dimensionality-reduction techniques are employed to present a two-dimensional representation of inorganic crystal chemical space, which is labelled according to whether the combinations pass standard chemical filters and if they appear in known materials databases.

Crystal Composition Transformer: Self-Learning Neural Language Model for Generative and Tinkering Design of Materials

Self-supervised neural language models have recently achieved unprecedented success from natural language processing to learning the languages of biological sequences and organic molecules. These models have demonstrated superior performance in the generation, structure classification, and functional predictions for proteins and molecules with learned representations. However, most of the masking-based pre-trained language models are not designed for generative design, and their black-box nature makes it difficult to interpret their design logic. Here a Blank-filling Language Model for Materials (BLMM) Crystal Transformer is proposed, a neural network-based probabilistic generative model for generative and tinkering design of inorganic materials. The model is built on the blank-filling language model for text generation and has demonstrated unique advantages in learning the "materials grammars" together with high-quality generation, interpretability, and data efficiency. It can generate chemically valid materials compositions with as high as 89.7% charge neutrality and 84.8% balanced electronegativity, which are more than four and eight times higher compared to a pseudo-random sampling baseline. The probabilistic generation process of BLMM allows it to recommend materials tinkering operations based on learned materials chemistry, which makes it useful for materials doping. The model is applied to discover a set of new materials as validated using the Density Functional Theory (DFT) calculations. This work thus brings the unsupervised transformer language models based generative artificial intelligence to inorganic materials. A user-friendly web app for tinkering materials design has been developed and can be accessed freely at www.materialsatlas.org/blmtinker.

From computational screening to the synthesis of a promising OER catalyst

The search for new materials can be laborious and expensive. Given the challenges that mankind faces today concerning the climate change crisis, the need to accelerate materials discovery for applications like water-splitting could be very relevant for a renewable economy. In this work, we introduce a computational framework to predict the activity of oxygen evolution reaction (OER) catalysts, in order to accelerate the discovery of materials that can facilitate water splitting. We use this framework to screen 6155 ternary-phase spinel oxides and have isolated 33 candidates which are predicted to have potentially high OER activity. We have also trained a machine learning model to predict the binding energies of the *O, *OH and *OOH intermediates calculated within this workflow to gain a deeper understanding of the relationship between electronic structure descriptors and OER activity. Out of the 33 candidates predicted to have high OER activity, we have synthesized three compounds and characterized them using linear sweep voltammetry to gauge their performance in OER. From these three catalyst materials, we have identified a new material, Co2.5Ga0.5O4, that is competitive with benchmark OER catalysts in the literature with a low overpotential of 220 mV at 10 mA cm-2 and a Tafel slope at 56.0 mV dec-1. Given the vast size of chemical space as well as the success of this technique to date, we believe that further application of this computational framework based on the high-throughput virtual screening of materials can lead to the discovery of additional novel, high-performing OER catalysts.

Diffusion models for conditional generation of hypothetical new families of superconductors

Effective computational search holds great potential for aiding the discovery of high-temperature superconductors (HTSs), especially given the lack of systematic methods for their discovery. Recent progress has been made in this area with machine learning, especially with deep generative models, which have been able to outperform traditional manual searches at predicting new superconductors within existing superconductor families but have yet to be able to generate completely new families of superconductors. We address this limitation by implementing conditioning-a method to control the generation process-for our generative model and develop SuperDiff, a denoising diffusion probabilistic model with iterative latent variable refinement conditioning for HTS discovery-the first deep generative model for superconductor discovery with conditioning on reference compounds. With SuperDiff, by being able to control the generation process, we were able to computationally generate completely new families of hypothetical superconductors for the very first time. Given that SuperDiff also has relatively fast training and inference times, it has the potential to be a very powerful tool for accelerating the discovery of new superconductors and enhancing our understanding of them.

Inferring energy-composition relationships with Bayesian optimization enhances exploration of inorganic materials

Computational exploration of the compositional spaces of materials can provide guidance for synthetic research and thus accelerate the discovery of novel materials. Most approaches employ high-throughput sampling and focus on reducing the time for energy evaluation for individual compositions, often at the cost of accuracy. Here, we present an alternative approach focusing on effective sampling of the compositional space. The learning algorithm PhaseBO optimizes the stoichiometry of the potential target material while improving the probability of and accelerating its discovery without compromising the accuracy of energy evaluation.

Dynamic sampling of liquid metal structures for theoretical studies on catalysis

Liquid metals have recently emerged as promising catalysts that can outcompete their solid counterparts for many reactions. Although theoretical modelling is extensively used to improve solid-state catalysts, there is currently no way to capture the interactions of adsorbates with a dynamic liquid metal. We propose a new approach based on ab initio molecular dynamics sampling of an adsorbate on a liquid catalyst. Using this approach, we describe time-resolved structures for formate adsorbed on liquid Ga-In, and for all intermediates in the methanol oxidation pathway on Ga-Pt. This yields a range of accessible adsorption energies that take into account the at-temperature motion of the liquid metal. We find that a previously proposed pathway for methanol oxidation on Ga-Pt results in unstable intermediates on a dynamic liquid surface, and propose that H desorption must occur during the path. The results showcase a more accurate way to treat liquid metal catalysts in this emerging field.

ScGAN: a generative adversarial network to predict hypothetical superconductors

Despite having been discovered more than three decades ago, high temperature superconductors (HTSs) lack both an explanation for their mechanisms and a systematic way to search for them. To aid this search, this project proposes ScGAN, a generative adversarial network (GAN) to efficiently predict new superconductors. ScGAN was trained on compounds in Open Quantum Materials Database and then transfer learned onto the SuperCon database or a subset of it. Once trained, the GAN was used to predict superconducting candidates, and approximately 70% of them were determined to be superconducting by a classification model-a 23-fold increase in discovery rate compared to manual search methods. Furthermore, more than 99% of predictions were novel materials, demonstrating that ScGAN was able to potentially predict completely new superconductors, including several promising HTS candidates. This project presents a novel, efficient way to search for new superconductors, which may be used in technological applications or provide insight into the unsolved problem of high temperature superconductivity.

In Pursuit of the Exceptional: Research Directions for Machine Learning in Chemical and Materials Science

Exceptional molecules and materials with one or more extraordinary properties are both technologically valuable and fundamentally interesting, because they often involve new physical phenomena or new compositions that defy expectations. Historically, exceptionality has been achieved through serendipity, but recently, machine learning (ML) and automated experimentation have been widely proposed to accelerate target identification and synthesis planning. In this Perspective, we argue that the data-driven methods commonly used today are well-suited for optimization but not for the realization of new exceptional materials or molecules. Finding such outliers should be possible using ML, but only by shifting away from using traditional ML approaches that tweak the composition, crystal structure, or reaction pathway. We highlight case studies of high-Tc oxide superconductors and superhard materials to demonstrate the challenges of ML-guided discovery and discuss the limitations of automation for this task. We then provide six recommendations for the development of ML methods capable of exceptional materials discovery: (i) Avoid the tyranny of the middle and focus on extrema; (ii) When data are limited, qualitative predictions that provide direction are more valuable than interpolative accuracy; (iii) Sample what can be made and how to make it and defer optimization; (iv) Create room (and look) for the unexpected while pursuing your goal; (v) Try to fill-in-the-blanks of input and output space; (vi) Do not confuse human understanding with model interpretability. We conclude with a description of how these recommendations can be integrated into automated discovery workflows, which should enable the discovery of exceptional molecules and materials.

Closed-Loop Error-Correction Learning Accelerates Experimental Discovery of Thermoelectric Materials

The exploration of thermoelectric materials is challenging considering the large materials space, combined with added exponential degrees of freedom coming from doping and the diversity of synthetic pathways. Here, historical data is incorporated, and is updated using experimental feedback by employing error-correction learning (ECL). This is achieved by learning from prior datasets and then adapting the model to differences in synthesis and characterization that are otherwise difficult to parameterize. This strategy is thus applied to discovering thermoelectric materials, where synthesis is prioritized at temperatures <300 °C. A previously unexplored chemical family of thermoelectric materials, PbSe:SnSb, is documented, finding that the best candidate in this chemical family, 2 wt% SnSb doped PbSe, exhibits a power factor more than 2× that of PbSe. The investigations herein reveal that a closed-loop experimentation strategy reduces the required number of experiments to find an optimized material by a factor as high as 3× compared to high-throughput searches powered by state-of-the-art machine-learning (ML) models. It is also observed that this improvement is dependent on the accuracy of the ML model in a manner that exhibits diminishing returns: once a certain accuracy is reached, factors that are instead associated with experimental pathways begin to dominate trends.

Design of New Inorganic Crystals with the Desired Composition Using Deep Learning

New solid-state materials have been discovered using various approaches from atom substitution in density functional theory (DFT) to generative models in machine learning. Recently, generative models have shown promising performance in finding new materials. Crystal generation with deep learning has been applied in various methods to discover new crystals. However, most generative models can only be applied to materials with specific elements or generate structures with random compositions. In this work, we developed a model that can generate crystals with desired compositions based on a crystal diffusion variational autoencoder. We generated crystal structures for 14 compositions of three types of materials in different applications. The generated structures were further stabilized using DFT calculations. We found the most stable structures in the existing database for all but one composition, even though eight compositions among them were not in the data set trained in a crystal diffusion variational autoencoder. This substantiates the prospect of the generation of an extensive range of compositions. Finally, 205 unique new crystal materials with energy above hull <100 meV/atom were generated. Moreover, we compared the average formation energy of the crystals generated from five compositions, two of which were hypothetical, with that of traditional methods like atom substitution and a generative model. The generated structures had lower formation energy than those of other models, except for one composition. These results demonstrate that our approach can be applied stably in various fields to design stable inorganic materials based on machine learning.

A computationally-guided non-equilibrium synthesis approach to materials discovery in the SrO-Al2O3-SiO2 phase field

Glass-crystallisation synthesis is coupled to probe structure prediction for the guided discovery of new metastable oxides in the SrO-Al2O3-SiO2 phase field, yielding a new ternary ribbon-silicate, Sr2Si3O8. In principle, this methodology can be applied to a wide range of oxide chemistries by selecting an appropriate non-equilibrium synthesis route.

Target-Driven Design of Deep-UV Nonlinear Optical Materials via Interpretable Machine Learning

The development of a data-driven science paradigm is greatly revolutionizing the process of materials discovery. Particularly, exploring novel nonlinear optical (NLO) materials with the birefringent phase-matching ability to deep-ultraviolet (UV) region is of vital significance for the field of laser technologies. Herein, a target-driven materials design framework combining high-throughput calculations (HTC), crystal structure prediction, and interpretable machine learning (ML) is proposed to accelerate the discovery of deep-UV NLO materials. Using a dataset generated from HTC, an ML regression model for predicting birefringence is developed for the first time, which exhibits a possibility of achieving fast and accurate prediction. Essentially, crystal structures are adopted as the only known input of this model to establish a close structure-property relationship mapping birefringence. Utilizing the ML-predicted birefringence which can affect the shortest phase-matching wavelength, a full list of potential chemical compositions based on an efficient screening strategy is identified. Further, eight structures with good stability are discovered to show potential applications in the deep-UV region, owing to their promising NLO-related properties. This study provides a new insight into the discovery of NLO materials and this design framework can identify desired materials with high performances in the broad chemical space at a low computational cost.

Predicting Synthesizability using Machine Learning on Databases of Existing Inorganic Materials

Defining the metric for synthesizability and predicting new compounds that can be experimentally realized in the realm of data-driven research is a pressing problem in contemporary materials science. The increasing computational power and advancements in machine learning (ML) algorithms provide a new avenue to solve the synthesizability challenge. In this work, using the Inorganic Crystal Structure Database (ICSD) and the Materials Project (MP) database, we represent crystal structures in Fourier-transformed crystal properties (FTCP) representation and use a deep learning model to predict synthesizability in the form of a synthesizability score (SC). Such an SC model, as a synthesizability filter for new materials, enables an efficient and accurate classification to identify promising material candidates. The SC prediction model achieved 82.6/80.6% (precision/recall) overall accuracy in predicting ternary crystal materials. We also trained the SC model by only considering compounds uploaded on the MP before 2015 as the training set and testing on multiple sets of materials uploaded after 2015. In the post-2019 test set, we obtain a high 88.60% true positive rate accuracy, coupled with 9.81% precision, indicating that newly added materials remain unexplored and have high synthesis potential. Further, we provide a list of 100 materials predicted to be synthesizable from this post-2019 dataset (highest SC) for future studies, and our SC model, as a validation filter, is beneficial for future material screening and discovery.

Benzoquinoline Chemical Space: A Helpful Approach in Antibacterial and Anticancer Drug Design

Benzoquinolines are used in many drug design projects as starting molecules subject to derivatization. This computational study aims to characterize e benzoquinone drug space to ease future drug design processes based on these molecules. The drug space is composed of all benzoquinones, which are active on topoisomerase II and ATP synthase. Topological, chemical, and bioactivity spaces are explored using computational methodologies based on virtual screening and scaffold hopping and molecular docking, respectively. Topological space is a geometrical space in which the elements composing it can be defined as a set of neighbors (which satisfy a particular axiom). In such space, a chemical space can be defined as the property space spanned by all possible molecules and chemical compounds adhering to a given set of construction principles and boundary conditions. In this chemical space, the potentially pharmacologically active molecules form the bioactivity space. Results show a poly-morphological chemical space that suggests distinct characteristics. The chemical space is correlated with properties such as steric energy, the number of hydrogen bonds, the presence of halogen atoms, and membrane permeability-related properties. Lastly, novel chemical compounds (such as oxadiazole methybenzamide and floro methylcyclohexane diene) with drug-like potential, active on TOPO II and ATP synthase have been identified.

Emerging Chalcohalide Materials for Energy Applications

Semiconductors with multiple anions currently provide a new materials platform from which improved functionality emerges, posing new challenges and opportunities in material science. This review has endeavored to emphasize the versatility of the emerging family of semiconductors consisting of mixed chalcogen and halogen anions, known as "chalcohalides". As they are multifunctional, these materials are of general interest to the wider research community, ranging from theoretical/computational scientists to experimental materials scientists. This review provides a comprehensive overview of the development of emerging Bi- and Sb-based as well as a new Cu, Sn, Pb, Ag, and hybrid organic-inorganic perovskite-based chalcohalides. We first highlight the high-throughput computational techniques to design and develop these chalcohalide materials. We then proceed to discuss their optoelectronic properties, band structures, stability, and structural chemistry employing theoretical and experimental underpinning toward high-performance devices. Next, we present an overview of recent advancements in the synthesis and their wide range of applications in energy conversion and storage devices. Finally, we conclude the review by outlining the impediments and important aspects in this field as well as offering perspectives on future research directions to further promote the development of chalcohalide materials in practical applications in the future.

Anaerobiosis favors biosynthesis of single and multi-element nanostructures

Herein we report the use of an environmental multimetal(loid)-resistant strain, MF05, to biosynthesize single- or multi-element nanostructures under anaerobic conditions. Inorganic nanostructure synthesis typically requires methodologies and conditions that are harsh and environmentally hazardous. Thus, green/eco-friendly procedures are desirable, where the use of microorganisms and their extracts as bionanofactories is a reliable strategy. First, MF05 was entirely sequenced and identified as an Escherichia coli-related strain with some genetic differences from the traditional BW25113. Secondly, we compared the CdS nanostructure biosynthesis by whole-cell in a design defined minimal culture medium containing sulfite as the only sulfur source to obtain sulfide reduction from a low-cost chalcogen reactant. Under anaerobic conditions, this process was greatly favored, and irregular CdS (ex. 370 nm; em. 520-530 nm) was obtained. When other chalcogenites were tested (selenite and tellurite), only spherical Se0 and elongated Te0 nanostructures were observed by TEM and analyzed by SEM-EDX. In addition, enzymatic-mediated chalcogenite (sulfite, selenite, and tellurite) reduction was assessed by using MF05 crude extracts in anaerobiosis; similar results for nanostructures were obtained; however Se0 and Te0 formation were more regular in shape and cleaner (with less background). Finally, the in vitro nanostructure biosynthesis was assessed with salts of Ag, Au, Cd, and Li alone or in combination with chalcogenites. Several single or binary nanostructures were detected. Our results showed that MF05 is a versatile anaerobic bionanofactory for different types of inorganic NS. synthesis.

Rapid discovery of stable materials by coordinate-free coarse graining

A fundamental challenge in materials science pertains to elucidating the relationship between stoichiometry, stability, structure, and property. Recent advances have shown that machine learning can be used to learn such relationships, allowing the stability and functional properties of materials to be accurately predicted. However, most of these approaches use atomic coordinates as input and are thus bottlenecked by crystal structure identification when investigating previously unidentified materials. Our approach solves this bottleneck by coarse-graining the infinite search space of atomic coordinates into a combinatorially enumerable search space. The key idea is to use Wyckoff representations, coordinate-free sets of symmetry-related positions in a crystal, as the input to a machine learning model. Our model demonstrates exceptionally high precision in finding unknown theoretically stable materials, identifying 1569 materials that lie below the known convex hull of previously calculated materials from just 5675 ab initio calculations. Our approach opens up fundamental advances in computational materials discovery.

TCSP: a Template-Based Crystal Structure Prediction Algorithm for Materials Discovery

Fast and accurate crystal structure prediction (CSP) algorithms and web servers are highly desirable for the exploration and discovery of new materials out of the infinite chemical design space. However, currently, the computationally expensive first-principles calculation-based CSP algorithms are applicable to relatively small systems and are out of reach of most materials researchers. Several teams have used an element substitution approach for generating or predicting new structures, but usually in an ad hoc way. Here we develop a template-based crystal structure prediction (TCSP) algorithm and its companion web server, which makes this tool accessible to all materials researchers. Our algorithm uses elemental/chemical similarity and oxidation states to guide the selection of template structures and then rank them based on the substitution compatibility and can return multiple predictions with ranking scores in a few minutes. A benchmark study on the 98290 formulas of the Materials Project database using leave-one-out evaluation shows that our algorithm can achieve high accuracy (for 13145 target structures, TCSP predicted their structures with root-mean-square deviation < 0.1) for a large portion of the formulas. We have also used TCSP to discover new materials of the Ga-B-N system, showing its potential for high-throughput materials discovery. Our user-friendly web app TCSP can be accessed freely at www.materialsatlas.org/crystalstructure on our MaterialsAtlas.org web app platform.

Environmental Stability of Crystals: A Greedy Screening

Discovering materials that are environmentally stable and also exhibit the necessary collection of properties required for a particular application is a perennial challenge in materials science. Herein, we present an algorithm to rapidly screen materials for their thermodynamic stability in a given environment, using a greedy approach. The performance was tested against the standard energy above the hull stability metric for inert conditions. Using data of 126 320 crystals, the greedy algorithm was shown to estimate the driving force for decomposition with a mean absolute error of 39.5 meV/atom, giving it sufficient resolution to identify stable materials. To demonstrate the utility outside of a vacuum, the in-oxygen stability of 39 654 materials was tested. The enthalpy of oxidation was found to be largely exothermic. Further analysis showed that 1438 of these materials fall into the range required for self-passivation based on the Pilling-Bedworth ratio.

Highly accurate machine learning prediction of crystal point groups for ternary materials from chemical formula

One of the most challenging problems in condensed matter physics is to predict crystal structure just from the chemical formula of the material. In this work, we present a robust machine learning (ML) predictor for the crystal point group of ternary materials (A\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$_l$$\end{document}lB\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$_m$$\end{document}mC\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$_n$$\end{document}n) - as first step to predict the structure - with very small set of ionic and positional fundamental features. From ML perspective, the problem is strenuous due to multi-labelity, multi-class, and data imbalance. The resulted prediction is very reliable as high balanced accuracies are obtained by different ML methods. Many similarity-based approaches resulted in a balanced accuracy above 95% indicating that the physics is well captured by the reduced set of features; namely, stoichiometry, ionic radii, ionization energies, and oxidation states for each of the three elements in the ternary compound. The accuracy is not limited by the approach; but rather by the limited data points and we should expect higher accuracy prediction by having more reliable data.

Towards Predictive Synthesis of Inorganic Materials Using Network Science

Accelerating materials discovery is the cornerstone of modern technological competitiveness. Yet, the inorganic synthesis of new compounds is often an important bottleneck in this quest. Well-established quantum chemistry and experimental synthesis methods combined with consolidated network science approaches might provide revolutionary knowledge to tackle this challenge. Recent pioneering studies in this direction have shown that the topological analysis of material networks hold great potential to effectively explore the synthesizability of inorganic compounds. In this Perspective we discuss the most exciting work in this area, in particular emerging new physicochemical insights and general concepts on how network science can significantly help reduce the timescales required to discover new materials and find synthetic routes for their fabrication. We also provide a perspective on outstanding problems, challenges and open questions.

Ionic Liquids and Deep Eutectics as a Transformative Platform for the Synthesis of Nanomaterials

Ionic liquids (ILs) are becoming a revolutionary synthesis medium for inorganic nanomaterials, permitting more efficient, safer and environmentally benign preparation of high quality products. A smart combination of ILs and...

John P, Gorai P. Predicting energy and stability of known and hypothetical crystals using graph neural network

The discovery of new inorganic materials in unexplored chemical spaces necessitates calculating total energy quickly and with sufficient accuracy. Machine learning models that provide such a capability for both ground-state (GS) and higher-energy structures would be instrumental in accelerated screening. Here, we demonstrate the importance of a balanced training dataset of GS and higher-energy structures to accurately predict total energies using a generic graph neural network architecture. Using ∼ 16,500 density functional theory calculations from the National Renewable Energy Laboratory (NREL) Materials Database and ∼ 11,000 calculations for hypothetical structures as our training database, we demonstrate that our model satisfactorily ranks the structures in the correct order of total energies for a given composition. Furthermore, we present a thorough error analysis to explain failure modes of the model, including both prediction outliers and occasional inconsistencies in the training data. By examining intermediate layers of the model, we analyze how the model represents learned structures and properties.

Entropy-based active learning of graph neural network surrogate models for materials properties

Graph neural networks trained on experimental or calculated data are becoming an increasingly important tool in computational materials science. Networks once trained are able to make highly accurate predictions at a fraction of the cost of experiments or first-principles calculations of comparable accuracy. However, these networks typically rely on large databases of labeled experiments to train the model. In scenarios where data are scarce or expensive to obtain, this can be prohibitive. By building a neural network that provides confidence on the predicted properties, we are able to develop an active learning scheme that can reduce the amount of labeled data required by identifying the areas of chemical space where the model is most uncertain. We present a scheme for coupling a graph neural network with a Gaussian process to featurize solid-state materials and predict properties including a measure of confidence in the prediction. We then demonstrate that this scheme can be used in an active learning context to speed up the training of the model by selecting the optimal next experiment for obtaining a data label. Our active learning scheme can double the rate at which the performance of the model on a test dataset improves with additional data compared to choosing the next sample at random. This type of uncertainty quantification and active learning has the potential to open up new areas of materials science, where data are scarce and expensive to obtain, to the transformative power of graph neural networks.

Thermodynamic exploration of xenon/krypton separation based on a high-throughput screening

Nanoporous framework materials are a promising class of materials for energy-efficient technology of xenon/krypton separation by physisorption. Many studies on Xe/Kr separation by adsorption have focused on the determination of structure/property relationships, the description of theoretical limits of performance, and the identification of top-performing materials. Here, we provide a study based on a high-throughput screening of the adsorption of Xe, Kr, and Xe/Kr mixtures in 12 020 experimental MOF materials, to provide a better comprehension of the thermodynamics behind Xe/Kr separation in nanoporous materials and the microscopic origins of Xe/Kr selectivity at both low and ambient pressure.

Synthetic accessibility and stability rules of NASICONs

In this paper we develop the stability rules for NASICON-structured materials, as an example of compounds with complex bond topology and composition. By first-principles high-throughput computation of 3881 potential NASICON phases, we have developed guiding stability rules of NASICON and validated the ab initio predictive capability through the synthesis of six attempted materials, five of which were successful. A simple two-dimensional descriptor for predicting NASICON stability was extracted with sure independence screening and machine learned ranking, which classifies NASICON phases in terms of their synthetic accessibility. This machine-learned tolerance factor is based on the Na content, elemental radii and electronegativities, and the Madelung energy and can offer reasonable accuracy for separating stable and unstable NASICONs. This work will not only provide tools to understand the synthetic accessibility of NASICON-type materials, but also demonstrates an efficient paradigm for discovering new materials with complicated composition and atomic structure.

Use of metamodels for rapid discovery of narrow bandgap oxide photocatalysts

New photocatalysts are traditionally identified through trial-and-error methods. Machine learning has shown considerable promise for improving the efficiency of photocatalyst discovery from a large potential pool. Here, we describe a multi-step, target-driven consensus method using a stacking meta-learning algorithm that robustly predicts bandgaps and H2 evolution activities of photocatalysts. Trained on small datasets, these models can rapidly screen a large space (>10 million materials) to identify promising, non-toxic compounds as candidate water splitting photocatalysts. Two effective compounds and two controls possessing optimal bandgap values (∼2 eV) but not photoactivity as predicted by the models were synthesized. Their experimentally measured bandgaps and H2 evolution activities were consistent with the predictions. Conspicuously, the two compounds with strong photoactivities under UV and visible light are promising visible-light-driven water splitting photocatalysts. This study demonstrates the power of machine learning and the potential of big data to accelerate discovery of next-generation photocatalysts.

Element selection for crystalline inorganic solid discovery guided by unsupervised machine learning of experimentally explored chemistry

The selection of the elements to combine delimits the possible outcomes of synthetic chemistry because it determines the range of compositions and structures, and thus properties, that can arise. For example, in the solid state, the elemental components of a phase field will determine the likelihood of finding a new crystalline material. Researchers make these choices based on their understanding of chemical structure and bonding. Extensive data are available on those element combinations that produce synthetically isolable materials, but it is difficult to assimilate the scale of this information to guide selection from the diversity of potential new chemistries. Here, we show that unsupervised machine learning captures the complex patterns of similarity between element combinations that afford reported crystalline inorganic materials. This model guides prioritisation of quaternary phase fields containing two anions for synthetic exploration to identify lithium solid electrolytes in a collaborative workflow that leads to the discovery of Li3.3SnS3.3Cl0.7. The interstitial site occupancy combination in this defect stuffed wurtzite enables a low-barrier ion transport pathway in hexagonal close-packing.

Microbial biomanufacture of metal/metallic nanomaterials and metabolic engineering: design strategies, fundamental mechanisms, and future opportunities

Biomanufacturing metal/metallic nanomaterials with ordered micro/nanostructures and controllable functions is of great importance in both fundamental studies and practical applications due to their low toxicity, lower pollution production, and energy conservation. Microorganisms, as efficient biofactories, have a significant ability to biomineralize and bioreduce metal ions that can be obtained as nanocrystals of varying morphologies and sizes. The development of nanoparticle biosynthesis maximizes the safety and sustainability of the nanoparticle preparation. Significant efforts and progress have been made to develop new green and environmentally friendly methods for biocompatible metal/metallic nanomaterials. In this review, we mainly focus on the microbial biomanufacture of different metal/metallic nanomaterials due to their unique advantages of wide availability, environmental acceptability, low cost, and circular sustainability. Specifically, we summarize recent and important advances in the synthesis strategies and mechanisms for different types of metal/metallic nanomaterials using different microorganisms. Finally, we highlight the current challenges and future research directions in this growing multidisciplinary field of biomaterials science, nanoscience, and nanobiotechnology.

Discovery of a Low Thermal Conductivity Oxide Guided by Probe Structure Prediction and Machine Learning

We report the aperiodic titanate Ba10 Y6 Ti4 O27 with a room-temperature thermal conductivity that equals the lowest reported for an oxide. The structure is characterised by discontinuous occupancy modulation of each of the sites and can be considered as a quasicrystal. The resulting localisation of lattice vibrations suppresses phonon transport of heat. This new lead material for low-thermal-conductivity oxides is metastable and located within a quaternary phase field that has been previously explored. Its isolation thus requires a precisely defined synthetic protocol. The necessary narrowing of the search space for experimental investigation was achieved by evaluation of titanate crystal chemistry, prediction of unexplored structural motifs that would favour synthetically accessible new compositions, and assessment of their properties with machine-learning models.

Perovskite-inspired materials for photovoltaics and beyond-from design to devices

Lead-halide perovskites have demonstrated astonishing increases in power conversion efficiency in photovoltaics over the last decade. The most efficient perovskite devices now outperform industry-standard multi-crystalline silicon solar cells, despite the fact that perovskites are typically grown at low temperature using simple solution-based methods. However, the toxicity of lead and its ready solubility in water are concerns for widespread implementation. These challenges, alongside the many successes of the perovskites, have motivated significant efforts across multiple disciplines to find lead-free and stable alternatives which could mimic the ability of the perovskites to achieve high performance with low temperature, facile fabrication methods. This Review discusses the computational and experimental approaches that have been taken to discover lead-free perovskite-inspired materials, and the recent successes and challenges in synthesizing these compounds. The atomistic origins of the extraordinary performance exhibited by lead-halide perovskites in photovoltaic devices is discussed, alongside the key challenges in engineering such high-performance in alternative, next-generation materials. Beyond photovoltaics, this Review discusses the impact perovskite-inspired materials have had in spurring efforts to apply new materials in other optoelectronic applications, namely light-emitting diodes, photocatalysts, radiation detectors, thin film transistors and memristors. Finally, the prospects and key challenges faced by the field in advancing the development of perovskite-inspired materials towards realization in commercial devices is discussed.

Identification of Potential Optoelectronic Applications for Metal Thiophosphates

Metal thiophosphates are a large family of compounds that received far less attention than conventional chalcogenides. Recently, however, metal thiophosphates arouse research interest in regard of energy harvesting and conversion due to their structural and chemical diversity. Nevertheless, there remain many unexplored metal thiophosphates. Here, we performed a comprehensive investigation on the electronic and optoelectronic properties of a series of metal thiophosphates using first-principles calculations and identified several highly promising compounds as p-type transparent conductors, photovoltaic absorbers, and single visible-light-driven photocatalysts for water splitting. Our investigation reveals the intrinsic features of a series of typical metal thiophosphates, identifies their new optoelectronic applications, and validates that metal thiophosphates are promising materials deserving exploration.

Predicting materials properties without crystal structure: deep representation learning from stoichiometry

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.

Learning Representations of Inorganic Materials from Generative Adversarial Networks

The two most important aspects of material research using deep learning (DL) or machine learning (ML) are the characteristics of materials data and learning algorithms, where the proper characterization of materials data is essential for generating accurate models. At present, the characterization of materials based on the molecular composition includes some methods based on feature engineering, such as Magpie and One-hot. Although these characterization methods have achieved significant results in materials research, these methods based on feature engineering cannot guarantee the integrity of materials characterization. One possible approach is to learn the materials characterization via neural networks using the chemical knowledge and implicit composition rules shown in large-scale known materials. This article chooses an adversarial method to learn the composition of atoms using the Generative Adversarial Network (GAN), which makes sense for data symmetry. The total loss value of the discriminator on the test set is reduced from 4.1e13 to 0.3194, indicating that the designed GAN network can well capture the combination of atoms in real materials. We then use the trained discriminator weights for material characterization and predict bandgap, formation energy, critical temperature (Tc) of superconductors on the Open Quantum Materials Database (OQMD), Materials Project (MP), and SuperCond datasets. Experiments show that when using the same predictive model, our proposed method performs better than One-hot and Magpie. This article provides an effective method for characterizing materials based on molecular composition in addition to Magpie, One-hot, etc. In addition, the generator learned in this study generates hypothetical materials with the same distribution as known materials, and these hypotheses can be used as a source for new material discovery.

Biosynthesis of inorganic nanomaterials using microbial cells and bacteriophages

Inorganic nanomaterials are widely used in chemical, electronics, photonics, energy and medical industries. Preparing a nanomaterial (NM) typically requires physical and/or chemical methods that involve harsh and environmentally hazardous conditions. Recently, wild-type and genetically engineered microorganisms have been harnessed for the biosynthesis of inorganic NMs under mild and environmentally friendly conditions. Microorganisms such as microalgae, fungi and bacteria, as well as bacteriophages, can be used as biofactories to produce single-element and multi-element inorganic NMs. This Review describes the emerging area of inorganic NM biosynthesis, emphasizing the mechanisms of inorganic-ion reduction and detoxification, while also highlighting the proteins and peptides involved. We show how analysing a Pourbaix diagram can help us devise strategies for the predictive biosynthesis of NMs with high producibility and crystallinity and also describe how to control the size and morphology of the product. Here, we survey biosynthetic inorganic NMs of 55 elements and their applications in catalysis, energy harvesting and storage, electronics, antimicrobials and biomedical therapy. Furthermore, a step-by-step flow chart is presented to aid the design and biosynthesis of inorganic NMs employing microbial cells. Future research in this area will add to the diversity of available inorganic NMs but should also address scalability and purity.

Data-Mining Element Charges in Inorganic Materials

Oxidation states are well-established in chemical science teaching and research. We data-mine more than 168 000 crystallographic reports to find an optimal allocation of oxidation states to each element. In doing so, we uncover discrepancies between textbook chemistry and reported charge states observed in materials. We go on to show how the oxidation states we recommend can significantly facilitate materials discovery and the heuristic design of novel inorganic compounds.

Machine Learning-Based Prediction of Crystal Systems and Space Groups from Inorganic Materials Compositions

Structural information of materials such as the crystal systems and space groups are highly useful for analyzing their physical properties. However, the enormous composition space of materials makes experimental X-ray diffraction (XRD) or first-principle-based structure determination methods infeasible for large-scale material screening in the composition space. Herein, we propose and evaluate machine-learning algorithms for determining the structure type of materials, given only their compositions. We couple random forest (RF) and multiple layer perceptron (MLP) neural network models with three types of features: Magpie, atom vector, and one-hot encoding (atom frequency) for the crystal system and space group prediction of materials. Four types of models for predicting crystal systems and space groups are proposed, trained, and evaluated including one-versus-all binary classifiers, multiclass classifiers, polymorphism predictors, and multilabel classifiers. The synthetic minority over-sampling technique (SMOTE) is conducted to mitigate the effects of imbalanced data sets. Our results demonstrate that RF with Magpie features generally outperforms other algorithms for binary and multiclass prediction of crystal systems and space groups, while MLP with atom frequency features is the best one for structural polymorphism prediction. For multilabel prediction, MLP with atom frequency and binary relevance with Magpie models are the best for predicting crystal systems and space groups, respectively. Our analysis of the related descriptors identifies a few key contributing features for structural-type prediction such as electronegativity, covalent radius, and Mendeleev number. Our work thus paves a way for fast composition-based structural screening of inorganic materials via predicted material structural properties.

Descriptors for Electron and Hole Charge Carriers in Metal Oxides

Metal oxides can act as insulators, semiconductors, or metals depending on their chemical composition and crystal structure. Metal oxide semiconductors, which support equilibrium populations of electron and hole charge carriers, have widespread applications including batteries, solar cells, and display technologies. It is often difficult to predict in advance whether these materials will exhibit localized or delocalized charge carriers upon oxidation or reduction. We combine data from first-principles calculations of the electronic structure and dielectric response of 214 metal oxides to predict the energetic driving force for carrier localization and transport. We assess descriptors based on the carrier effective mass, static polaron binding energy, and Fröhlich electron-phonon coupling. Numerical analysis allows us to assign p- and n-type transport of a metal oxide to three classes: (i) band transport with high mobility; (ii) small polaron transport with low mobility; and (iii) intermediate behavior. The results of this classification agree with observations regarding carrier dynamics and lifetimes and are used to predict 10 candidate p-type oxides.

Modelling of framework materials at multiple scales: current practices and open questions

The last decade has seen an explosion of the family of framework materials and their study, from both the experimental and computational points of view. We propose here a short highlight of the current state of methodologies for modelling framework materials at multiple scales, putting together a brief review of new methods and recent endeavours in this area, as well as outlining some of the open challenges in this field. We will detail advances in atomistic simulation methods, the development of material databases and the growing use of machine learning for the prediction of properties. This article is part of the theme issue 'Mineralomimesis: natural and synthetic frameworks in science and technology'.

Machine learning for molecular and materials science

Here we summarize recent progress in machine learning for the chemical sciences. We outline machine-learning techniques that are suitable for addressing research questions in this domain, as well as future directions for the field. We envisage a future in which the design, synthesis, characterization and application of molecules and materials is accelerated by artificial intelligence.

Recombinant Escherichia coli as a biofactory for various single- and multi-element nanomaterials

Nanomaterials (NMs) are mostly synthesized by chemical and physical methods, but biological synthesis is also receiving great attention. However, the mechanisms for biological producibility of NMs, crystalline versus amorphous, are not yet understood. Here we report biosynthesis of 60 different NMs by employing a recombinant Escherichia coli strain coexpressing metallothionein, a metal-binding protein, and phytochelatin synthase that synthesizes a metal-binding peptide phytochelatin. Both an in vivo method employing live cells and an in vitro method employing the cell extract are used to synthesize NMs. The periodic table is scanned to select 35 suitable elements, followed by biosynthesis of their NMs. Nine crystalline single-elements of Mn3O4, Fe3O4, Cu2O, Mo, Ag, In(OH)3, SnO2, Te, and Au are synthesized, while the other 16 elements result in biosynthesis of amorphous NMs or no NM synthesis. Producibility and crystallinity of the NMs are analyzed using a Pourbaix diagram that predicts the stable chemical species of each element for NM biosynthesis by varying reduction potential and pH. Based on the analyses, the initial pH of reactions is changed from 6.5 to 7.5, resulting in biosynthesis of various crystalline NMs of those previously amorphous or not-synthesized ones. This strategy is extended to biosynthesize multi-element NMs including CoFe2O4, NiFe2O4, ZnMn2O4, ZnFe2O4, Ag2S, Ag2TeO3, Ag2WO4, Hg3TeO6, PbMoO4, PbWO4, and Pb5(VO4)3OH NMs. The strategy described here allows biosynthesis of NMs with various properties, providing a platform for manufacturing various NMs in an environmentally friendly manner.

Materials discovery by chemical analogy: role of oxidation states in structure prediction

The likelihood of an element to adopt a specific oxidation state in a solid, given a certain set of neighbours, might often be obvious to a trained chemist. However, encoding this information for use in high-throughput searches presents a significant challenge. We carry out a statistical analysis of the occurrence of oxidation states in 16 735 ordered, inorganic compounds and show that a large number of cations are only likely to exhibit certain oxidation states in combination with particular anions. We use this data to build a model that ascribes probabilities to the formation of hypothetical compounds, given the proposed oxidation states of their constituent species. The model is then used as part of a high-throughput materials design process, which significantly narrows down the vast compositional search space for new ternary metal halide compounds. Finally, we employ a machine learning analysis of existing compounds to suggest likely structures for a small subset of the candidate compositions. We predict two new compounds, MnZnBr4 and YSnF7, that are thermodynamically stable according to density functional theory, as well as four compounds, MnCdBr4, MnRu2Br8, ScZnF5 and ZnCoBr4, which lie within the window of metastability.

Computer-aided design of metal chalcohalide semiconductors: from chemical composition to crystal structure

The standard paradigm in computational materials science is INPUT: Structure; OUTPUT: Properties, which has yielded many successes but is ill-suited for exploring large areas of chemical and configurational hyperspace.

The standard paradigm in computational materials science is INPUT: Structure; OUTPUT: Properties, which has yielded many successes but is ill-suited for exploring large areas of chemical and configurational hyperspace. We report a high-throughput screening procedure that uses compositional descriptors to search for new photoactive semiconducting compounds. We show how feeding high-ranking element combinations to structure prediction algorithms can constitute a pragmatic computer-aided materials design approach. Techniques based on structural analogy (data mining of known lattice types) and global searches (direct optimisation using evolutionary algorithms) are combined for translating between chemical composition and crystal structure. The properties of four novel chalcohalides (Sn₅S₄Cl₂, Sn₄SF₆, Cd₅S₄Cl₂ and Cd₄SF₆) are predicted, of which two are calculated to have bandgaps in the visible range of the electromagnetic spectrum.

Grand Challenges and Future Opportunities for Metal-Organic Frameworks

Metal-organic frameworks (MOFs) allow compositional and structural diversity beyond conventional solid-state materials. Continued interest in the field is justified by potential applications of exceptional breadth, ranging from gas storage and separation, which takes advantage of the inherent pores and their volume, to electronic applications, which requires precise control of electronic structure. In this Outlook we present some of the pertinent challenges that MOFs face in their conventional implementations, as well as opportunities in less traditional areas. Here the aim is to discuss select design concepts and future research goals that emphasize nuances relevant to this class of materials as a whole. Particular emphasis is placed on synthetic aspects, as they influence the potential for MOFs in gas separation, electrical conductivity, and catalytic applications.