Machine Learning Directed Search for Ultraincompressible, Superhard Materials

How the AI-assisted discovery and synthesis of a ternary oxide highlights capability gaps in materials science

Exploratory synthesis has been the main generator of new inorganic materials for decades. However, our Edisonian and bias-prone processes of synthetic exploration alone are no longer sufficient in an age that demands rapid advances in materials development. In this work, we demonstrate an end-to-end attempt towards systematic, computer-aided discovery and laboratory synthesis of inorganic crystalline compounds as a modern alternative to purely exploratory synthesis. Our approach initializes materials discovery campaigns by autonomously mapping the synthetic feasibility of a chemical system using density functional theory with AI feedback. Following expert-driven down-selection of newly generated phases, we use solid-state synthesis and in situ characterization via hot-stage X-ray diffraction in order to realize new ternary oxide phases experimentally. We applied this strategy in six ternary transition-metal oxide chemistries previously considered well-explored, one of which culminated in the discovery of two novel phases of calcium ruthenates. Detailed characterization using room temperature X-ray powder diffraction, 4D-STEM and SQUID measurements identifies the structure and composition and confirms distinct properties, including distinct defect concentrations, of one of the new phases formed in our experimental campaigns. While the discovery of a new material guided by AI and DFT theory represents a milestone, our procedure and results also highlight a number of critical gaps in the process that can inform future efforts towards the improvement of AI-coupled methodologies.

Coarse-Grained Crystal Graph Neural Networks for Reticular Materials Design

Reticular materials, including metal-organic frameworks and covalent organic frameworks, combine the relative ease of synthesis and an impressive range of applications in various fields from gas storage to biomedicine. Diverse properties arise from the variation of building units─metal centers and organic linkers─in almost infinite chemical space. Such variation substantially complicates the experimental design and promotes the use of computational methods. In particular, the most successful artificial intelligence algorithms for predicting the properties of reticular materials are atomic-level graph neural networks, which optionally incorporate domain knowledge. Nonetheless, the data-driven inverse design involving these models suffers from the incorporation of irrelevant and redundant features such as a full atomistic graph and network topology. In this study, we propose a new way of representing materials, aiming to overcome the limitations of existing methods; the message passing is performed on a coarse-grained crystal graph that comprises molecular building units. To highlight the merits of our approach, we assessed the predictive performance and energy efficiency of neural networks built on different materials representations, including composition-based and crystal-structure-aware models. Coarse-grained crystal graph neural networks showed decent accuracy at low computational costs, making them a valuable alternative to omnipresent atomic-level algorithms. Moreover, the presented models can be successfully integrated into an inverse materials design pipeline as estimators of the objective function. Overall, the coarse-grained crystal graph framework is aimed at challenging the prevailing atom-centric perspective on reticular materials design.

Machine learning guided tuning charge distribution by composition in MOFs for oxygen evolution reaction

Traditional design/optimization of metal-organic frameworks (MOFs) is time-consuming and labor-intensive. In this study, we utilize machine learning (ML) to accelerate the synthesis of MOFs. We have built a library of over 900 MOFs with different metal salts, solvent ratios, reaction durations and temperatures, and utilize zeta potentials as target variables for ML training. A total of four ML models have been used to train the collected dataset and assess their convergence performances, where Random Forest Regression (RFR) and Gradient Boosting Regression (GBR) models show strong correlation and accurate predictions. We then predicted two kinds of MOFs from RFR and GBR models. Remarkably, the experimentally data of the synthesized MOFs closely matched the predicted results, and these MOFs exhibited excellent electrocatalytic performances for oxygen evolution. This study would have general implications in the utilization of machine learning for accelerating the synthesis of MOFs for diverse applications.

Informatics-Driven Design of Superhard B-C-O Compounds

Materials containing B, C, and O, due to the advantages of forming strong covalent bonds, may lead to materials that are superhard, i.e., those with a Vicker's hardness larger than 40 GPa. However, the exploration of this vast chemical, compositional, and configurational space is nontrivial. Here, we leverage a combination of machine learning (ML) and first-principles calculations to enable and accelerate such a targeted search. The ML models first screen for potentially superhard B-C-O compositions from a large hypothetical B-C-O candidate space. Atomic-level structure search using density functional theory (DFT) within those identified compositions, followed by further detailed analyses, unravels on four potentially superhard B-C-O phases exhibiting thermodynamic, mechanical, and dynamic stability.

Predicting Poisson's Ratio: A Study of Semisupervised Anomaly Detection and Supervised Approaches

Auxetics are a rare class of materials that exhibit a negative Poisson's ratio. The existence of these auxetic materials is rare but has a large number of applications in the design of exotic materials. We build a complete machine learning framework to detect Auxetic materials as well as Poisson's ratio of non-auxetic materials. A semisupervised anomaly detection model is presented, which is capable of separating out the auxetics materials (treated as an anomaly) from an unknown database with an average precision of 0.64. Another regression model (supervised) is also created to predict the Poisson's ratio of non-auxetic materials with an R2 of 0.82. Additionally, this regression model helps us to find the optimal features for the anomaly detection model. This methodology can be generalized and used to discover materials with rare physical properties.

CEEM: a Chemically Explainable Deep Learning Platform for Identifying Compounds with Low Effective Mass

The semiconductor industry occupies a crucial position in the fields of integrated circuits, energy, and communication systems. Effective mass (mE ), which is closely related to electron transition, thermal excitation, and carrier mobility, is a key performance indicator of semiconductor. However, the highly neglected mE is onerous to measure experimentally, which seriously hinders the evaluation of semiconductor properties and the understanding of the carrier migration mechanisms. Here, a chemically explainable effective mass predictive platform (CEEM) is constructed by deep learning, to identify n-type and p-type semiconductors with low mE . Based on the graph network, a versatile explainable network is innovatively designed that enables CEEM to efficiently predict the mE of any structure, with the area under the curve of 0.904 for n-type semiconductors and 0.896 for p-type semiconductors, and derive the most relevant chemical factors. Using CEEM, the currently largest mE database is built that contains 126 335 entries and screens out 466 semiconductors with low mE for transparent conductive materials, photovoltaic materials, and water-splitting materials. Moreover, a user-friendly and interactive CEEM web is provided that supports query, prediction, and explanation of mE . CEEM's high efficiency, accuracy, flexibility, and explainability open up new avenues for the discovery and design of high-performance semiconductors.

Data-driven discovery of ultraincompressible crystals from a universal correlation between bulk modulus and volumetric cohesive energy

Bulk modulus and cohesive energy are two important quantities of condensed matter. From the interatomic energy landscape, we here derived a correlation between the bulk modulus (B) and the volumetric cohesive energy (ρe), i.e.B= 2(ln2)2ρe/9ϵs2=kρe, whereϵsandkare the strain-to-failure of interatomic bonds and the factor of proportionality, respectively. By analyzing numerous crystals from first principles calculations, it was shown that this correlation is universally applicable to various crystals including simple substances and compounds. Most interestingly, it was found thatϵsof crystals with a similar structure are almost a constant, resulting in a linear relationship betweenBandρe. Furthermore, we found that the value ofkfor any compound can be determined based on the rule of mixtures, i.e.k= ∑xiki, wherexiandkiare the atomic fraction and the factor of proportionality for each element in this compound, respectively. Finally, this correlation was used to predict the bulk moduli for a vast number of crystals with knownρein databases. After first principles verification of the top 50 crystals with the highest predicted bulk modulus, 25 ultraincompressible crystals with a bulk modulus greater than 400 GPa that can rival diamond (436 GPa) were discovered.

Molecular Graph-Based Deep Learning Algorithm Facilitates an Imaging-Based Strategy for Rapid Discovery of Small Molecules Modulating Biomolecular Condensates

Biomolecular condensates are proposed to cause diseases, such as cancer and neurodegeneration, by concentrating proteins at abnormal subcellular loci. Imaging-based compound screens have been used to identify small molecules that reverse or promote biomolecular condensates. However, limitations of conventional imaging-based methods restrict the screening scale. Here, we used a graph convolutional network (GCN)-based computational approach and identified small molecule candidates that reduce the nuclear liquid-liquid phase separation of TAR DNA-binding protein 43 (TDP-43), an essential protein that undergoes phase transition in neurodegenerative diseases. We demonstrated that the GCN-based deep learning algorithm is suitable for spatial information extraction from the molecular graph. Thus, this is a promising method to identify small molecule candidates with novel scaffolds. Furthermore, we validated that these candidates do not affect the normal splicing function of TDP-43. Taken together, a combination of an imaging-based screen and a GCN-based deep learning method dramatically improves the speed and accuracy of the compound screen for biomolecular condensates.

Discovery of all-inorganic lead-free perovskites with high photovoltaic performance via ensemble machine learning

Growing evidence shows that all-inorganic lead-free perovskites hold promise for solving stability and toxicity problems in perovskite solar cells. However, the power conversion efficiency of all-inorganic perovskites cannot match that of hybrid organic-inorganic perovskites. To face the challenges of efficiency, stability and toxicity simultaneously for application in perovskite solar cells, this study conducts a high-throughput materials search via ensemble machine learning for nearly 12 million all-inorganic perovskites to obtain candidates with non-toxicity and excellent photovoltaic performance. Based on experimental data, models for structure identification and band gap classification are established for , and a physics-inspired multi-component neural network is proposed as part of the exploration of the model's logical structure. It is found that extracting key features for input into the model and treating non-key features as supplements make model learning easier and are more effective in reducing the model parameters. Then, based on established ensemble models as well as the new criteria of ion radius difference and the optimization rules of toxicity and cost, over 80 000 candidates are screened. Among the 34 lead-free identified with suitable band gaps and negative formation energies through first principles calculations, 17 candidates have theoretical power conversion efficiencies over 20%. The Debye temperature of 10 lead-free , basically Bi-based compounds, is greater than 350 K, which is advantageous for suppressing nonradiative recombination and thermally induced degradation.

Coexistence of Superhardness and Metal-Like Electrical Conductivity in High-Entropy Dodecaboride Composite with Atomic-Scale Interlocks

High electrical conductivity and super high hardness are two sought-after material properties, but both are contradictory because the effective suppression of dislocation movement generally increases the scattering of conducting electrons. Here we synthesized a high-entropy dodecaboride composite (HEDC) with a large number of atomic-scale interlocking layers. It shows a Vickers hardness of 51.2 ± 3.6 GPa under an applied load of 0.49 N and an electrical resistivity of 44.5 μΩ·cm at room temperature. Such HEDC achieves superhardness by inheriting the high intrinsic hardness of its constituent phases and restricting the dislocation motion to further enhance the extrinsic hardness through forming numerous atom-scale interlocks between different slip systems. Moreover, the HEDC maintains the excellent electrical conductivity of the constituent borides, and the competition between two correlating structures produces the special kind of coherent boundary that minimizes the scattering of conducting electrons and does not largely deteriorate the electrical conductivity.

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.

Unleashing the Power of Artificial Intelligence in Materials Design

The integration of artificial intelligence (AI) algorithms in materials design is revolutionizing the field of materials engineering thanks to their power to predict material properties, design de novo materials with enhanced features, and discover new mechanisms beyond intuition. In addition, they can be used to infer complex design principles and identify high-quality candidates more rapidly than trial-and-error experimentation. From this perspective, herein we describe how these tools can enable the acceleration and enrichment of each stage of the discovery cycle of novel materials with optimized properties. We begin by outlining the state-of-the-art AI models in materials design, including machine learning (ML), deep learning, and materials informatics tools. These methodologies enable the extraction of meaningful information from vast amounts of data, enabling researchers to uncover complex correlations and patterns within material properties, structures, and compositions. Next, a comprehensive overview of AI-driven materials design is provided and its potential future prospects are highlighted. By leveraging such AI algorithms, researchers can efficiently search and analyze databases containing a wide range of material properties, enabling the identification of promising candidates for specific applications. This capability has profound implications across various industries, from drug development to energy storage, where materials performance is crucial. Ultimately, AI-based approaches are poised to revolutionize our understanding and design of materials, ushering in a new era of accelerated innovation and advancement.

Atomic stiffness for bulk modulus prediction and high-throughput screening of ultraincompressible crystals

Determining bulk moduli is central to high-throughput screening of ultraincompressible materials. However, existing approaches are either too inaccurate or too expensive for general applications, or they are limited to narrow chemistries. Here we define a microscopic quantity to measure the atomic stiffness for each element in the periodic table. Based on this quantity, we derive an analytic formula for bulk modulus prediction. By analyzing numerous crystals from first-principles calculations, this formula shows superior accuracy, efficiency, universality, and interpretability compared to previous empirical/semiempirical formulae and machine learning models. Directed by our formula predictions and verified by first-principles calculations, 47 ultraincompressible crystals rivaling diamond are identified from over one million material candidates, which extends the family of known ultraincompressible crystals. Finally, treasure maps of possible elemental combinations for ultraincompressible crystals are created from our theory. This theory and insights provide guidelines for designing and discovering ultraincompressible crystals of the future.

Converting Nanotoxicity Data to Information Using Artificial Intelligence and Simulation

Decades of nanotoxicology research have generated extensive and diverse data sets. However, data is not equal to information. The question is how to extract critical information buried in vast data streams. Here we show that artificial intelligence (AI) and molecular simulation play key roles in transforming nanotoxicity data into critical information, i.e., constructing the quantitative nanostructure (physicochemical properties)-toxicity relationships, and elucidating the toxicity-related molecular mechanisms. For AI and molecular simulation to realize their full impacts in this mission, several obstacles must be overcome. These include the paucity of high-quality nanomaterials (NMs) and standardized nanotoxicity data, the lack of model-friendly databases, the scarcity of specific and universal nanodescriptors, and the inability to simulate NMs at realistic spatial and temporal scales. This review provides a comprehensive and representative, but not exhaustive, summary of the current capability gaps and tools required to fill these formidable gaps. Specifically, we discuss the applications of AI and molecular simulation, which can address the large-scale data challenge for nanotoxicology research. The need for model-friendly nanotoxicity databases, powerful nanodescriptors, new modeling approaches, molecular mechanism analysis, and design of the next-generation NMs are also critically discussed. Finally, we provide a perspective on future trends and challenges.

Accelerating the Discovery of Metastable IrO2 for the Oxygen Evolution Reaction by the Self-Learning-Input Graph Neural Network

The discovery of active and stable catalysts for the oxygen evolution reaction (OER) is vital to improve water electrolysis. To date, rutile iridium dioxide IrO₂ is the only known OER catalyst in the acidic solution, while its poor activity restricts its practical viability. Herein, we propose a universal graph neural network, namely, CrystalGNN, and introduce a dynamic embedding layer to self-update atomic inputs during the training process. Based on this framework, we train a model to accurately predict the formation energies of 10,500 IrO₂ configurations and discover 8 unreported metastable phases, among which C2/m-IrO₂ and P62-IrO₂ are identified as excellent electrocatalysts to reach the theoretical OER overpotential limit at their most stable surfaces. Our self-learning-input CrystalGNN framework exhibits reliable accuracy, generalization, and transferring ability and successfully accelerates the bottom-up catalyst design of novel metastable IrO₂ to boost the OER activity.

Machine Learning Full NMR Chemical Shift Tensors of Silicon Oxides with Equivariant Graph Neural Networks

The nuclear magnetic resonance (NMR) chemical shift tensor is a highly sensitive probe of the electronic structure of an atom and furthermore its local structure. Recently, machine learning has been applied to NMR in the prediction of isotropic chemical shifts from a structure. Current machine learning models, however, often ignore the full chemical shift tensor for the easier-to-predict isotropic chemical shift, effectively ignoring a multitude of structural information available in the NMR chemical shift tensor. Here we use an equivariant graph neural network (GNN) to predict full ²⁹Si chemical shift tensors in silicate materials. The equivariant GNN model predicts full tensors to a mean absolute error of 1.05 ppm and is able to accurately determine the magnitude, anisotropy, and tensor orientation in a diverse set of silicon oxide local structures. When compared with other models, the equivariant GNN model outperforms the state-of-the-art machine learning models by 53%. The equivariant GNN model also outperforms historic analytical models by 57% for isotropic chemical shift and 91% for anisotropy. The software is available as a simple-to-use open-source repository, allowing similar models to be created and trained with ease.

Materials informatics for developing new restorative dental materials: A narrative review

Materials informatics involves the application of computational methodologies to process and interpret scientific and engineering data concerning materials. Although this concept has been well established in the fields of biology, drug discovery, and classic materials research, its application in the field of dental materials is still in its infancy. This narrative review comprehensively summarizes the advantages, limitations, and future perspectives of materials informatics from 2003 to 2022 for exploring the optimum compositions in developing new materials using artificial intelligence. The findings indicate that materials informatics, which is a recognized and established concept in the materials science field, will accelerate the process of restorative materials development and contribute to produce new insights into dental materials research.

Large Data Set-Driven Machine Learning Models for Accurate Prediction of the Thermoelectric Figure of Merit

The figure of merit (zT) is a key parameter to measure the performance of thermoelectric materials. At present, the prediction of zT values via machine leaning has emerged as a promising method for exploring high-performance materials. However, the machine learning-based predictions still suffer from unsatisfactory accuracy, and this is related to the size of the data set, the hyperparameters of models, and the quality of the data. In this work, 5038 pieces of data of thermoelectric materials were selected, and several regression models were generated to predict zT values. This large data set-driven light gradient boosting (LGB) model with 57 features performed with an excellent accuracy, achieving a coefficient of determination (R²) value of 0.959, a root mean squared error (RMSE) of 0.094, a mean absolute error (MAE) of 0.057, and a correlation coefficient (R) of 0.979. Owing to the large size of the data set, the prediction accuracy exceeds that of most reported zT predictions via machine learning. The "M_E Lattice Parameter" was verified as the most important feature in the zT prediction. Furthermore, nine potential candidates were screened out from among one million pieces of data. This study solves the problem of the data set size, adjusts the hyperparameters of the models, uses feature engineering to improve data quality, and provides an efficient strategy to perform wide-ranging screening for promising materials.

Treating Superhard Materials as Anomalies

Superhard materials are among the most scarce functional inorganic solids in existence. Indeed, recent research suggested that less than 0.1% of all known materials are likely to have a Vickers hardness ≥40 GPa. Here, an anomaly detection framework is created to treat these materials as rare occurrences by encoding and reconstructing the input composition and crystal structure information without supervision. The resulting model can quantitatively identify outliers from "normal" behaving materials, leading to the discovery of materials with exceptional properties such as a superhard response. Moreover, examining the difference between the encoded and decoded crystal structure provides fundamental insights into the crystal-chemical origin of hardness. The presented methodology is ultimately generalizable, enabling the design of other outlier materials with rare and unexpected physical properties.

Machine learning-based inverse design for electrochemically controlled microscopic gradients of O2 and H2O2

In microbiology, extracellular oxygen (O₂) and reactive oxygen species (ROS) are spatiotemporally heterogenous, ubiquitously, at macroscopic level. Such spatiotemporal heterogeneities are critical to microorganisms, yet a well-defined method of studying such heterogenous microenvironments is lacking. This work develops a machine learning–based inverse design strategy that builds an electrochemical platform for achieving spatiotemporal control of O₂ and ROS microenvironments relevant to microbiology. The inverse design strategy not only demonstrates the power of machine learning to design concentration profiles in electrochemistry but also accelerates the development of custom microenvironments for specific microbial systems and allows researchers to better study how microenvironments affect microorganisms in myriads of environmental, biomedical, and sustainability-related applications.

A fundamental understanding of extracellular microenvironments of O₂ and reactive oxygen species (ROS) such as H₂O₂, ubiquitous in microbiology, demands high-throughput methods of mimicking, controlling, and perturbing gradients of O₂ and H₂O₂ at microscopic scale with high spatiotemporal precision. However, there is a paucity of high-throughput strategies of microenvironment design, and it remains challenging to achieve O₂ and H₂O₂ heterogeneities with microbiologically desirable spatiotemporal resolutions. Here, we report the inverse design, based on machine learning (ML), of electrochemically generated microscopic O₂ and H₂O₂ profiles relevant for microbiology. Microwire arrays with suitably designed electrochemical catalysts enable the independent control of O₂ and H₂O₂ profiles with spatial resolution of ∼10¹ μm and temporal resolution of ∼10° s. Neural networks aided by data augmentation inversely design the experimental conditions needed for targeted O₂ and H₂O₂ microenvironments while being two orders of magnitude faster than experimental explorations. Interfacing ML-based inverse design with electrochemically controlled concentration heterogeneity creates a viable fast-response platform toward better understanding the extracellular space with desirable spatiotemporal control.

Discovering Superhard B-N-O Compounds by Iterative Machine Learning and Evolutionary Structure Predictions

We searched for new superhard B-N-O compounds with an iterative machine learning (ML) procedure, where ML models are trained using sample crystal structures from an evolutionary algorithm. We first used cohesive energy to evaluate the thermodynamic stability of varying B _x N _y O _z compositions and then gradually focused on compositional regions with high cohesive energy and high hardness. The results converged quickly after a few iterations. Our resulting ML models show that B _x+2N _x O₃ compounds with x ≥ 3 (like B₅N₃O₃, B₆N₄O₃, etc.) are potentially superhard and thermodynamically favorable. Our meta-GGA density functional theory calculations indicate that these materials are also wide bandgap (≥4.4 eV) insulators, with the valence band maximum related to the p-orbitals of nitrogen atoms near vacant sites. This study demonstrates that an iterative method combining ML and ab initio simulations provides a powerful tool for discovering novel materials.

Graph Convolutional Network-Based Screening Strategy for Rapid Identification of SARS-CoV-2 Cell-Entry Inhibitors

The cell entry of SARS-CoV-2 has emerged as an attractive drug development target. We previously reported that the entry of SARS-CoV-2 depends on the cell surface heparan sulfate proteoglycan (HSPG) and the cortex actin, which can be targeted by therapeutic agents identified by conventional drug repurposing screens. However, this drug identification strategy requires laborious library screening, which is time consuming, and often limited number of compounds can be screened. As an alternative approach, we developed and trained a graph convolutional network (GCN)-based classification model using information extracted from experimentally identified HSPG and actin inhibitors. This method allowed us to virtually screen 170,000 compounds, resulting in ∼2000 potential hits. A hit confirmation assay with the uptake of a fluorescently labeled HSPG cargo further shortlisted 256 active compounds. Among them, 16 compounds had modest to strong inhibitory activities against the entry of SARS-CoV-2 pseudotyped particles into Vero E6 cells. These results establish a GCN-based virtual screen workflow for rapid identification of new small molecule inhibitors against validated drug targets.

Discovery of Lead-Free Perovskites for High-Performance Solar Cells via Machine Learning: Ultrabroadband Absorption, Low Radiative Combination, and Enhanced Thermal Conductivities

Exploring lead-free candidates and improving efficiency and stability remain the obstacle of hybrid organic-inorganic perovskite-based devices commercialization. Traditional trial-and-error methods seriously restrict the discovery especially for large search space, complex crystal structure and multi-objective properties. Here, the authors propose a multi-step and multi-stage screening scheme to accelerate the discovery of hybrid organic-inorganic perovskites A2 BB'X6 from a large number of candidates through combining machine learning with high-throughput calculations for pursuing excellent efficiency and thermal stability in solar cells. Followed by a series of screenings, the structure-property relationships mapping A2 BB'X6 properties are built and the predictions are close to reported experimental results. Successfully, four experimental-feasibly candidates with good stability, high Debye temperature and suitable band gap are screened out and further verified by density-functional theory calculations, in which the predicted efficiency for three lead-free candidates ((CH3 NH3 )2 AgGaBr6 , (CH3 NH3 )2 AgInBr6 and (C2 NH6 )2 AgInBr6 ) achieves 20.6%, 19.9% and 27.6% due to ultrabroadband absorption region ranging from UVC to IRC with excitonic radiative combination rates as low as 10 ps, large or intermediate polarons form with properties similar to CH3 NH3 PbI3 and the calculated thermal conductivities are 5.04, 4.39 and 5.16 Wm-1 K-1 , respectively, with Debye temperatures larger than 500 K, beneficial for suppression of both nonradiative combination and heat-induced degradation.

Machine Learning Driven Synthesis of Few-Layered WTe2 with Geometrical Control

Reducing the lateral scale of two-dimensional (2D) materials to one-dimensional (1D) has attracted substantial research interest not only to achieve competitive electronic applications but also for the exploration of fundamental physical properties. Controllable synthesis of high-quality 1D nanoribbons (NRs) is thus highly desirable and essential for further study. Here, we report the implementation of supervised machine learning (ML) for the chemical vapor deposition (CVD) synthesis of high-quality quasi-1D few-layered WTe₂ NRs. Feature importance analysis indicates that H₂ gas flow rate has a profound influence on the formation of WTe₂, and the source ratio governs the sample morphology. Notably, the growth mechanism of 1T' few-layered WTe₂ NRs is further proposed, which provides new insights for the growth of intriguing 2D and 1D tellurides and may inspire the growth strategies for other 1D nanostructures. Our findings suggest the effectiveness and capability of ML in guiding the synthesis of 1D nanostructures, opening up new opportunities for intelligent materials development.

Materials Informatics for Mechanical Deformation: A Review of Applications and Challenges

In the design and development of novel materials that have excellent mechanical properties, classification and regression methods have been diversely used across mechanical deformation simulations or experiments. The use of materials informatics methods on large data that originate in experiments or/and multiscale modeling simulations may accelerate materials' discovery or develop new understanding of materials' behavior. In this fast-growing field, we focus on reviewing advances at the intersection of data science with mechanical deformation simulations and experiments, with a particular focus on studies of metals and alloys. We discuss examples of applications, as well as identify challenges and prospects.

Machine Learning Approach to Analyze the Surface Properties of Biological Materials

Similar to how CRISPR has revolutionized the field of molecular biology, machine learning may drastically boost research in the area of materials science. Machine learning is a fast-evolving method that allows for analyzing big data and unveiling correlations that otherwise would remain undiscovered. It may hold invaluable potential to engineer novel functional materials with desired properties, a field, which is currently limited by time-consuming trial and error approaches and our limited understanding of how different material properties depend on each other. Here, we apply machine learning algorithms to classify complex biological materials based on their microtopography. With this approach, the surfaces of different variants of biofilms and plant leaves can not only be distinguished but also correctly classified according to their wettability. Furthermore, an importance ranking provided by one of the algorithms allows us to identify those surface features that are critical for a successful sample classification. Our study exemplifies how machine learning can contribute to the analysis and categorization of complex surfaces, a tool, which can be highly useful for other areas of materials science, such as damage assessment as well as adhesion or friction studies.

Machine Learning in Chemical Product Engineering: The State of the Art and a Guide for Newcomers

Chemical Product Engineering (CPE) is marked by numerous challenges, such as the complexity of the properties–structure–ingredients–process relationship of the different products and the necessity to discover and develop constantly and quickly new molecules and materials with tailor-made properties. In recent years, artificial intelligence (AI) and machine learning (ML) methods have gained increasing attention due to their performance in tackling particularly complex problems in various areas, such as computer vision and natural language processing. As such, they present a specific interest in addressing the complex challenges of CPE. This article provides an updated review of the state of the art regarding the implementation of ML techniques in different types of CPE problems with a particular focus on four specific domains, namely the design and discovery of new molecules and materials, the modeling of processes, the prediction of chemical reactions/retrosynthesis and the support for sensorial analysis. This review is further completed by general guidelines for the selection of an appropriate ML technique given the characteristics of each problem and by a critical discussion of several key issues associated with the development of ML modeling approaches. Accordingly, this paper may serve both the experienced researcher in the field as well as the newcomer.

OCELOT: An infrastructure for data-driven research to discover and design crystalline organic semiconductors

Materials design and discovery are often hampered by the slow pace and materials and human costs associated with Edisonian trial-and-error screening approaches. Recent advances in computational power, theoretical methods, and data science techniques, however, are being manifest in a convergence of these tools to enable in silico materials discovery. Here, we present the development and deployment of computational materials data and data analytic approaches for crystalline organic semiconductors. The OCELOT (Organic Crystals in Electronic and Light-Oriented Technologies) infrastructure, consisting of a Python-based OCELOT application programming interface and OCELOT database, is designed to enable rapid materials exploration. The database contains a descriptor-based schema for high-throughput calculations that have been implemented on more than 56 000 experimental crystal structures derived from 47 000 distinct molecular structures. OCELOT is open-access and accessible via a web-user interface at https://oscar.as.uky.edu.

A general QSPR protocol for the prediction of atomic/inter-atomic properties: a fragment based graph convolutional neural network (F-GCN)

In this study, a general quantitative structure-property relationship (QSPR) protocol, fragment based graph convolutional neural network (F-GCN), was developed for the prediction of atomic/inter-atomic properties. We applied this novel artificial intelligence (AI) tool in predictions of NMR chemical shifts and bond dissociation energies (BDEs). The obtained results were comparable to experimental measurements, while the computational cost was substantially reduced, with respect to pure density functional theory (DFT) calculations. The two important features of F-GCN can be summarised as: first, it could utilise different levels of molecular fragments for atomic/inter-atomic information extraction; second, the designed architecture is also open to include additional descriptors for a more accurate solution of the local environment at atomic level, making itself more efficient for structural solutions. And during our test, the averaged prediction error of 1H NMR chemical shifts is as small as 0.32 ppm, and the error of C-H BDE estimation is 2.7 kcal mol-1. Moreover, we further demonstrated the applicability of this developed F-GCN model via several challenging structural assignments. The success of the F-GCN in atomic and inter-atomic predictions also indicates an essential improvement of computational chemistry with the assistance of AI tools.

AB2X6 Compounds and the Stabilization of Trirutile Oxides

The properties of crystalline materials tend to be strongly correlated with their structures, and the prediction of crystal structure from only the composition is a coveted goal in the field of inorganic materials. However, even for the simplest compositions, such prediction relies on a complex network of interactions, including atomic or ionic radii, ionicity, electronegativity, position in the periodic table, and magnetism, to name only a few important parameters. We focus here on the AB₂X₆ (AB₂O₆ and AB₂F₆) composition space with the specific goal of finding new oxide compounds in the trirutile family, which is known for unusual one-dimensional (1D) antiferromagnetic behavior. Through machine learning methods, we develop an understanding of how geometric and bonding constraints determine the crystallization of compounds in the trirutile structure as opposed to other ternary structures in this space. In combination with density functional theory (DFT) calculations, we predict 16 previously unreported candidate trirutile oxides. We successfully prepare one of these and show it forms in the disordered rutile structure, under the preparation conditions adopted here.

Artificial intelligence and machine learning in design of mechanical materials

Artificial intelligence, especially machine learning (ML) and deep learning (DL) algorithms, is becoming an important tool in the fields of materials and mechanical engineering, attributed to its power to predict materials properties, design de novo materials and discover new mechanisms beyond intuitions. As the structural complexity of novel materials soars, the material design problem to optimize mechanical behaviors can involve massive design spaces that are intractable for conventional methods. Addressing this challenge, ML models trained from large material datasets that relate structure, properties and function at multiple hierarchical levels have offered new avenues for fast exploration of the design spaces. The performance of a ML-based materials design approach relies on the collection or generation of a large dataset that is properly preprocessed using the domain knowledge of materials science underlying chemical and physical concepts, and a suitable selection of the applied ML model. Recent breakthroughs in ML techniques have created vast opportunities for not only overcoming long-standing mechanics problems but also for developing unprecedented materials design strategies. In this review, we first present a brief introduction of state-of-the-art ML models, algorithms and structures. Then, we discuss the importance of data collection, generation and preprocessing. The applications in mechanical property prediction, materials design and computational methods using ML-based approaches are summarized, followed by perspectives on opportunities and open challenges in this emerging and exciting field.

Enhanced Hardness in Transition-Metal Monocarbides via Optimal Occupancy of Bonding Orbitals

An efficient strategy that can guide the synthesis of materials with superior mechanical properties is important for advanced material/device design. Here, we report a feasible way to enhance hardness in transition-metal monocarbides (TMCs) by optimally filling the bonding orbitals of valence electrons. We demonstrate that the intrinsic hardness of the NaCl- and WC-type TMCs maximizes at valence electron concentrations of about 9 and 10.25 electrons per cell, respectively; any deviation from such optimal values will reduce the hardness. Using the spark plasma sintering technique, a number of W_1-xRe_xC (x = 0-0.5) have been successfully synthesized, and powder X-ray diffractions show that they adopt the hexagonal WC-type structure. Subsequent nanoindentation and Vickers hardness measurements corroborate that the newly developed W_1-xRe_xC samples (x = 0.1-0.3) are much harder than their parent phase (i.e., WC), marking them as the hardest TMCs for practical applications. Furthermore, the hardness enhancement can be well rationalized by the balanced occupancy of bonding and antibonding states. Our findings not only elucidate the unique hardening mechanism in a large class of TMCs but also offer a guide for the design of other hard and superhard compounds such as borides and nitrides.

Engineered two-dimensional nanomaterials: an emerging paradigm for water purification and monitoring

Water scarcity has become an increasingly complex challenge with the growth of the global population, economic expansion, and climate change, highlighting the demand for advanced water treatment technologies that can provide clean water in a scalable, reliable, affordable, and sustainable manner. Recent advancements on 2D nanomaterials (2DM) open a new pathway for addressing the grand challenge of water treatment owing to their unique structures and superior properties. Emerging 2D nanostructures such as graphene, MoS₂, MXene, h-BN, g-C₃N₄, and black phosphorus have demonstrated an unprecedented surface-to-volume ratio, which promises ultralow material use, ultrafast processing time, and ultrahigh treatment efficiency for water cleaning/monitoring. In this review, we provide a state-of-the-art account on engineered 2D nanomaterials and their applications in emerging water technologies, involving separation, adsorption, photocatalysis, and pollutant detection. The fundamental design strategies of 2DM are discussed with emphasis on their physicochemical properties, underlying mechanism and targeted applications in different scenarios. This review concludes with a perspective on the pressing challenges and emerging opportunities in 2DM-enabled wastewater treatment and water-quality monitoring. This review can help to elaborate the structure-processing-property relationship of 2DM, and aims to guide the design of next-generation 2DM systems for the development of selective, multifunctional, programmable, and even intelligent water technologies. The global significance of clean water for future generations sheds new light and much inspiration in this rising field to enhance the efficiency and affordability of water treatment and secure a global water supply in a growing portion of the world.

On the crystal chemistry of inorganic nitrides: crystal-chemical parameters, bonding behavior, and opportunities in the exploration of their compositional space

The scarcity of nitrogen in Earth's crust, combined with challenging synthesis, have made inorganic nitrides a relatively unexplored class of compounds compared to their naturally abundant oxide counterparts. To facilitate exploration of their compositional space via a priori modeling, and to help a posteriori structure verification not limited to inferring the oxidation state of redox-active cations, we derive a suite of bond-valence parameters and Lewis acid strength values for 76 cations observed bonding to N3-, and further outline a baseline statistical knowledge of bond lengths for these compounds. Examination of structural and electronic effects responsible for the functional properties and anomalous bonding behavior of inorganic nitrides shows that many mechanisms of bond-length variation ubiquitous to oxide and oxysalt compounds (e.g., lone-pair stereoactivity, the Jahn-Teller and pseudo Jahn-Teller effects) are similarly pervasive in inorganic nitrides, and are occasionally observed to result in greater distortion magnitude than their oxide counterparts. We identify promising functional units for exploring uncharted chemical spaces of inorganic nitrides, e.g. multiple-bond metal centers with promise regarding the development of a post-Haber-Bosch process proceeding at milder reaction conditions, and promote an atomistic understanding of chemical bonding in nitrides relevant to such pursuits as the development of a model of ion substitution in solids, a problem of great relevance to semiconductor doping whose solution would fast-track the development of compound solar cells, battery materials, electronics, and more.

Finding the Next Superhard Material through Ensemble Learning

An ensemble machine-learning method is demonstrated to be capable of finding superhard materials by directly predicting the load-dependent Vickers hardness based only on the chemical composition. A total of 1062 experimentally measured load-dependent Vickers hardness data are extracted from the literature and used to train a supervised machine-learning algorithm utilizing boosting, achieving excellent accuracy (R²  = 0.97). This new model is then tested by synthesizing and measuring the load-dependent hardness of several unreported disilicides and analyzing the predicted hardness of several classic superhard materials. The trained ensemble method is then employed to screen for superhard materials by examining more than 66 000 compounds in crystal structure databases, which show that 68 known materials have a Vickers hardness ≥40 GPa at 0.5 N (applied force) and only 10 exceed this mark at 5 N. The hardness model is then combined with the data-driven phase diagram generation tool to expand the limited number of reported high hardness compounds. Eleven ternary borocarbide phase spaces are studied, and more than ten thermodynamically favorable compositions with a hardness above 40 GPa (at 0.5 N) are identified, proving this ensemble model's ability to find previously unknown materials with outstanding mechanical properties.

Discovery of Low-Modulus Ti-Nb-Zr Alloys Based on Machine Learning and First-Principles Calculations

The discovery of low-modulus Ti alloys for biomedical applications is challenging due to a vast number of compositions and available solute contents. In this work, machine learning (ML) methods are employed for the prediction of the bulk modulus (K) and the shear modulus (G) of optimized ternary alloys. As a starting point, the elasticity data of more than 1800 compounds from the Materials Project fed linear models, random forest regressors, and artificial neural networks (NN), with the aims of training predictive models for K and G based on compositional features. The models were then used to predict the resultant Young modulus (E) for all possible compositions in the Ti-Nb-Zr system, with variations in the composition of 2 at. %. Random forest (RF) predictions of E deviate from the NN predictions by less than 4 GPa, which is within the expected variance from the ML training phase. RF regressors seem to generate the most reliable models, given the selected target variables and descriptors. Optimal compositions identified by the ML models were later investigated with the aid of special quasi-random structures (SQSs) and density functional theory (DFT). According to a combined analysis, alloys with 22 Zr (at. %) are promising structural materials to the biomedical field, given their low elastic modulus and elevated beta-phase stability. In alloys with Nb content higher than 14.8 (at. %), the beta phase has lower energy than omega, which may be enough to avoid the formation of omega, a high-modulus phase, during manufacturing.

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.

Toward Accurate Predictions of Atomic Properties via Quantum Mechanics Descriptors Augmented Graph Convolutional Neural Network: Application of This Novel Approach in NMR Chemical Shifts Predictions

In this study, an augmented Graph Convolutional Network (GCN) with quantum mechanics (QM) descriptors was reported for its accurate predictions of NMR chemical shifts with respect to experimental values. The prediction errors of ¹³C/¹H NMR chemical shifts can be as small as 2.14/0.11 ppm. There are two crucial characteristics for this modified GCN: in one aspect, such a novel neural network could efficiently extract the overall molecule structure information; in another aspect, it could accurately solve the chemical environment of the target atom. As there exists an imperfect linear regression between the experimental NMR chemical shifts (δ) and the density functional theory (DFT) calculated isotropic shielding constants (σ), the inclusion of QM descriptors within GCN can largely improve its performance. Moreover, few-shot learning also becomes feasible with these descriptors. The success of this novel GCN in chemical shifts predictions also indicates its potential applicability for other computational studies.

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.

The Role of Machine Learning in the Understanding and Design of Materials

Developing algorithmic approaches for the rational design and discovery of materials can enable us to systematically find novel materials, which can have huge technological and social impact. However, such rational design requires a holistic perspective over the full multistage design process, which involves exploring immense materials spaces, their properties, and process design and engineering as well as a techno-economic assessment. The complexity of exploring all of these options using conventional scientific approaches seems intractable. Instead, novel tools from the field of machine learning can potentially solve some of our challenges on the way to rational materials design. Here we review some of the chief advancements of these methods and their applications in rational materials design, followed by a discussion on some of the main challenges and opportunities we currently face together with our perspective on the future of rational materials design and discovery.

Machine learning approach for elucidating and predicting the role of synthesis parameters on the shape and size of TiO2 nanoparticles

In the present work a series of design rules are developed in order to tune the morphology of TiO₂ nanoparticles through hydrothermal process. Through a careful experimental design, the influence of relevant process parameters on the synthesis outcome are studied, reaching to the develop predictive models by using Machine Learning methods. The models, after the validation and training, are able to predict with high accuracy the synthesis outcome in terms of nanoparticle size, polydispersity and aspect ratio. Furthermore, they are implemented by reverse engineering approach to do the inverse process, i.e. obtain the optimal synthesis parameters given a specific product characteristic. For the first time, it is presented a synthesis method that allows continuous and precise control of NPs morphology with the possibility to tune the aspect ratio over a large range from 1.4 (perfect truncated bipyramids) to 6 (elongated nanoparticles) and the length from 20 to 140 nm.

Accurate predictions of aqueous solubility of drug molecules via the multilevel graph convolutional network (MGCN) and SchNet architectures

Deep learning based methods have been widely applied to predict various kinds of molecular properties in the pharmaceutical industry with increasingly more success. In this study, we propose two novel models for aqueous solubility predictions, based on the Multilevel Graph Convolutional Network (MGCN) and SchNet architectures, respectively. The advantage of the MGCN lies in the fact that it could extract the graph features of the target molecules directly from the (3D) structural information; therefore, it doesn't need to rely on a lot of intra-molecular descriptors to learn the features, which are of significance for accurate predictions of the molecular properties. The SchNet performs well in modelling the interatomic interactions inside a molecule, and such a deep learning architecture is also capable of extracting structural information and further predicting the related properties. The actual accuracy of these two novel approaches was systematically benchmarked with four different independent datasets. We found that both the MGCN and SchNet models performed well for aqueous solubility predictions. In the future, we believe such promising predictive models will be applicable to enhancing the efficiency of the screening, crystallization and delivery of drug molecules, essentially as a useful tool to promote the development of molecular pharmaceutics.

Element Code from Pseudopotential as Efficient Descriptors for a Machine Learning Model to Explore Potential Lead-Free Halide Perovskites

The rapid development of machine learning has proven its potential in material science. To acquire an accurate and promising result, the choice of descriptor plays an essential role in dictating the model performance. In this work, we introduce a set of novel descriptors, Element Code, which is generated from pseudopotential. Using a variational autoencoder to perform unsupervised learning, the produced Element Code is verified to contain representative information on elements. Attributed to the successful extraction of information from pseudopotential, Element Code can serve as the primary descriptor for the machine learning model. We construct a model using Element Code as the sole descriptor to predict the bandgap of a lead-free double halide perovskite, and an accuracy of 0.951 and mean absolute error of 0.266 eV are achieved. We believe our work can offer insights into selecting lead-free halide perovskites and establish a paradigm of exploring new materials.

Data-driven studies of magnetic two-dimensional materials

We use a data-driven approach to study the magnetic and thermodynamic properties of van der Waals (vdW) layered materials. We investigate monolayers of the form \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {A}_2\hbox {B}_2\hbox {X}_6$$\end{document}A2B2X6, based on the known material \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {Cr}_2\hbox {Ge}_2\hbox {Te}_6$$\end{document}Cr2Ge2Te6, using density functional theory (DFT) calculations and machine learning methods to determine their magnetic properties, such as magnetic order and magnetic moment. We also examine formation energies and use them as a proxy for chemical stability. We show that machine learning tools, combined with DFT calculations, can provide a computationally efficient means to predict properties of such two-dimensional (2D) magnetic materials. Our data analytics approach provides insights into the microscopic origins of magnetic ordering in these systems. For instance, we find that the X site strongly affects the magnetic coupling between neighboring A sites, which drives the magnetic ordering. Our approach opens new ways for rapid discovery of chemically stable vdW materials that exhibit magnetic behavior.

Quantum-mechanical transition-state model combined with machine learning provides catalyst design features for selective Cr olefin oligomerization

The use of data science tools to provide the emergence of non-trivial chemical features for catalyst design is an important goal in catalysis science. Additionally, there is currently no general strategy for computational homogeneous, molecular catalyst design. Here, we report the unique combination of an experimentally verified DFT-transition-state model with a random forest machine learning model in a campaign to design new molecular Cr phosphine imine (Cr(P,N)) catalysts for selective ethylene oligomerization, specifically to increase 1-octene selectivity. This involved the calculation of 1-hexene : 1-octene transition-state selectivity for 105 (P,N) ligands and the harvesting of 14 descriptors, which were then used to build a random forest regression model. This model showed the emergence of several key design features, such as Cr-N distance, Cr-α distance, and Cr distance out of pocket, which were then used to rapidly design a new generation of Cr(P,N) catalyst ligands that are predicted to give >95% selectivity for 1-octene.