Computational discovery of fast interstitial oxygen conductors.

New highly oxygen-active materials may enhance many energy-related technologies by enabling efficient oxygen-ion transport at lower temperatures, for example, below ~400 °C. Interstitial oxygen conductors have the potential to realize such performance but have received far less attention than vacancy-mediated conductors. Here we combine physically motivated structure and property descriptors, ab initio simulations and experiments to demonstrate an approach to discover new fast interstitial oxygen conductors. Multiple new families were found, which adopt completely different structures from known oxygen conductors. From these families, we synthesized and studied oxygen kinetics in La4Mn5Si4O22+δ, a representative member of the perrierite/chevkinite family. We found that La4Mn5Si4O22+δ has higher oxygen-ion conductivity than the widely used yttria-stabilized ZrO2, and among the highest surface oxygen exchange rates at the intermediate temperature of known materials. The fast oxygen kinetics is the result of simultaneously active interstitial and interstitialcy diffusion pathways. We propose that the essential features for forming an effective interstitial oxygen conductor are the availability of electrons and structural flexibility, enabling a sufficient accessible volume. This work provides a powerful approach for understanding and discovering new interstitial oxygen conductors.

How close are the classical two-body potentials to ab initio calculations? Insights from linear machine learning based force matching.

In this work, we propose a linear machine learning force matching approach that can directly extract pair atomic interactions from ab initio calculations in amorphous structures. The local feature representation is specifically chosen to make the linear weights a force field as a force/potential function of the atom pair distance. Consequently, this set of functions is the closest representation of the ab initio forces, given the two-body approximation and finite scanning in the configurational space. We validate this approach in amorphous silica. Potentials in the new force field (consisting of tabulated Si-Si, Si-O, and O-O potentials) are significantly different than existing potentials that are commonly used for silica, even though all of them produce the tetrahedral network structure and roughly similar glass properties. This suggests that the commonly used classical force fields do not offer fundamentally accurate representations of the atomic interaction in silica. The new force field furthermore produces a lower glass transition temperature (Tg ∼ 1800 K) and a positive liquid thermal expansion coefficient, suggesting the extraordinarily high Tg and negative liquid thermal expansion of simulated silica could be artifacts of previously developed classical potentials. Overall, the proposed approach provides a fundamental yet intuitive way to evaluate two-body potentials against ab initio calculations, thereby offering an efficient way to guide the development of classical force fields.

Extracting accurate materials data from research papers with conversational language models and prompt engineering.

There has been a growing effort to replace manual extraction of data from research papers with automated data extraction based on natural language processing, language models, and recently, large language models (LLMs). Although these methods enable efficient extraction of data from large sets of research papers, they require a significant amount of up-front effort, expertise, and coding. In this work, we propose the ChatExtract method that can fully automate very accurate data extraction with minimal initial effort and background, using an advanced conversational LLM. ChatExtract consists of a set of engineered prompts applied to a conversational LLM that both identify sentences with data, extract that data, and assure the data's correctness through a series of follow-up questions. These follow-up questions largely overcome known issues with LLMs providing factually inaccurate responses. ChatExtract can be applied with any conversational LLMs and yields very high quality data extraction. In tests on materials data, we find precision and recall both close to 90% from the best conversational LLMs, like GPT-4. We demonstrate that the exceptional performance is enabled by the information retention in a conversational model combined with purposeful redundancy and introducing uncertainty through follow-up prompts. These results suggest that approaches similar to ChatExtract, due to their simplicity, transferability, and accuracy are likely to become powerful tools for data extraction in the near future. Finally, databases for critical cooling rates of metallic glasses and yield strengths of high entropy alloys are developed using ChatExtract.

Deep-Learning-Based Segmentation of Keyhole in In-Situ X-ray Imaging of Laser Powder Bed Fusion.

In laser powder bed fusion processes, keyholes are the gaseous cavities formed where laser interacts with metal, and their morphologies play an important role in defect formation and the final product quality. The in-situ X-ray imaging technique can monitor the keyhole dynamics from the side and capture keyhole shapes in the X-ray image stream. Keyhole shapes in X-ray images are then often labeled by humans for analysis, which increasingly involves attempting to correlate keyhole shapes with defects using machine learning. However, such labeling is tedious, time-consuming, error-prone, and cannot be scaled to large data sets. To use keyhole shapes more readily as the input to machine learning methods, an automatic tool to identify keyhole regions is desirable. In this paper, a deep-learning-based computer vision tool that can automatically segment keyhole shapes out of X-ray images is presented. The pipeline contains a filtering method and an implementation of the BASNet deep learning model to semantically segment the keyhole morphologies out of X-ray images. The presented tool shows promising average accuracy of 91.24% for keyhole area, and 92.81% for boundary shape, for a range of test dataset conditions in Al6061 (and one AliSi10Mg) alloys, with 300 training images/labels and 100 testing images for each trial. Prospective users may apply the presently trained tool or a retrained version following the approach used here to automatically label keyhole shapes in large image sets.

Deep Learning Approach for High-accuracy Electron Counting of Monolithic Active Pixel Sensor-type Direct Electron Detectors at Increased Electron Dose.

Electron counting can be performed algorithmically for monolithic active pixel sensor direct electron detectors to eliminate readout noise and Landau noise arising from the variability in the amount of deposited energy for each electron. Errors in existing counting algorithms include mistakenly counting a multielectron strike as a single electron event, and inaccurately locating the incident position of the electron due to lateral spread of deposited energy and dark noise. Here, we report a supervised deep learning (DL) approach based on Faster region-based convolutional neural network (R-CNN) to recognize single electron events at varying electron doses and voltages. The DL approach shows high accuracy according to the near-ideal modulation transfer function (MTF) and detector quantum efficiency for sparse images. It predicts, on average, 0.47 pixel deviation from the incident positions for 200 kV electrons versus 0.59 pixel using the conventional counting method. The DL approach also shows better robustness against coincidence loss as the electron dose increases, maintaining the MTF at half Nyquist frequency above 0.83 as the electron density increases to 0.06 e-/pixel. Thus, the DL model extends the advantages of counting analysis to higher dose rates than conventional methods.

Strong Room-Temperature Ferromagnetism in Ultrathin NiOOH Nanosheets through Surfactant Manipulation.

Two-dimensional (2D) ferromagnetic (FM) materials with nanoscale thickness and spontaneous net magnetization have emerged as a promising class of functional materials for applications in next-generation spintronics, quantum processing, and data storage devices. However, most 2D materials exhibit weak FM even at low temperatures, limiting their potential applications in many technological fields. The fabrication of strong room-temperature FM 2D materials is highly desirable for the development of practical applications. Here, we demonstrate an ionic layer epitaxy strategy to synthesize few-layered NiOOH nanosheets with strong room-temperature FM and a saturation magnetization up to 409.86 emu cm-3 at 300 K. The results are consistent with the ab initio predictions of a stable FM NiOOH nanolayer structure with an FM configuration. The FM strength of the NiOOH nanosheets can be tuned by controlling the surfactant monolayer density and annealing. This work offers a promising strategy for achieving strong high-temperature FM in 2D materials for spintronic applications.

Benchmark Tests of Atom Segmentation Deep Learning Models with a Consistent Dataset.

The information content of atomic-resolution scanning transmission electron microscopy (STEM) images can often be reduced to a handful of parameters describing each atomic column, chief among which is the column position. Neural networks (NNs) are high performance, computationally efficient methods to automatically locate atomic columns in images, which has led to a profusion of NN models and associated training datasets. We have developed a benchmark dataset of simulated and experimental STEM images and used it to evaluate the performance of two sets of recent NN models for atom location in STEM images. Both models exhibit high performance for images of varying quality from several different crystal lattices. However, there are important differences in performance as a function of image quality, and both models perform poorly for images outside the training data, such as interfaces with large difference in background intensity. Both the benchmark dataset and the models are available using the Foundry service for dissemination, discovery, and reuse of machine learning models.

Substantial lifetime enhancement for Si-based photoanodes enabled by amorphous TiO2 coating with improved stoichiometry.

Amorphous titanium dioxide (TiO2) film coating by atomic layer deposition (ALD) is a promising strategy to extend the photoelectrode lifetime to meet the industrial standard for solar fuel generation. To realize this promise, the essential structure-property relationship that dictates the protection lifetime needs to be uncovered. In this work, we reveal that in addition to the imbedded crystalline phase, the presence of residual chlorine (Cl) ligands is detrimental to the silicon (Si) photoanode lifetime. We further demonstrate that post-ALD in-situ water treatment can effectively decouple the ALD reaction completeness from crystallization. The as-processed TiO2 film has a much lower residual Cl concentration and thus an improved film stoichiometry, while its uniform amorphous phase is well preserved. As a result, the protected Si photoanode exhibits a substantially improved lifetime to ~600 h at a photocurrent density of more than 30 mA/cm2. This study demonstrates a significant advancement toward sustainable hydrogen generation.

Materials swelling revealed through automated semantic segmentation of cavities in electron microscopy images.

Accurately quantifying swelling of alloys that have undergone irradiation is essential for understanding alloy performance in a nuclear reactor and critical for the safe and reliable operation of reactor facilities. However, typical practice is for radiation-induced defects in electron microscopy images of alloys to be manually quantified by domain-expert researchers. Here, we employ an end-to-end deep learning approach using the Mask Regional Convolutional Neural Network (Mask R-CNN) model to detect and quantify nanoscale cavities in irradiated alloys. We have assembled a database of labeled cavity images which includes 400 images, > 34 k discrete cavities, and numerous alloy compositions and irradiation conditions. We have evaluated both statistical (precision, recall, and F1 scores) and materials property-centric (cavity size, density, and swelling) metrics of model performance, and performed targeted analysis of materials swelling assessments. We find our model gives assessments of material swelling with an average (standard deviation) swelling mean absolute error based on random leave-out cross-validation of 0.30 (0.03) percent swelling. This result demonstrates our approach can accurately provide swelling metrics on a per-image and per-condition basis, which can provide helpful insight into material design (e.g., alloy refinement) and impact of service conditions (e.g., temperature, irradiation dose) on swelling. Finally, we find there are cases of test images with poor statistical metrics, but small errors in swelling, pointing to the need for moving beyond traditional classification-based metrics to evaluate object detection models in the context of materials domain applications.

Experimentally informed structure optimization of amorphous TiO2 films grown by atomic layer deposition.

Amorphous titanium dioxide TiO2 (a-TiO2) has been widely studied, particularly as a protective coating layer on semiconductors to prevent corrosion and promote electron-hole conduction in photoelectrochemical reactions. The stability and longevity of a-TiO2 is strongly affected by the thickness and structural heterogeneity, implying that understanding the structure properties of a-TiO2 is crucial for improving the performance. This study characterized the structural and electronic properties of a-TiO2 thin films (∼17 nm) grown on Si by atomic layer deposition (ALD). Fluctuation spectra V(k) and angular correlation functions were determined with 4-dimensional scanning transmission electron microscopy (4D-STEM), which revealed the distinctive medium-range ordering in the a-TiO2 film. A realistic atomic model of a-TiO2 was established guided by the medium-range ordering and the previously reported short-range ordering of a-TiO2 film, as well as the interatomic potential. The structure was optimized by the StructOpt code using a genetic algorithm that simultaneously minimizes energy and maximizes the match to experimental short- and medium-range ordering. The StructOpt a-TiO2 model presents improved agreements with the medium-range ordering and the k-space location of the dominant 2-fold angular correlations compared with a traditional melt-quenched model. The electronic structure of the StructOpt a-TiO2 model was studied by ab initio calculations and compared to the crystalline phases and experimental results. This work uncovered the medium-range ordering in a-TiO2 thin films and provided a realistic a-TiO2 structure model for further investigation of structure-property relationships and materials design. In addition, the improved multi-objective optimization package StructOpt was provided for structure determination of complex materials guided by experiments and simulations.

Understanding the Fragile-to-Strong Transition in Silica from Microscopic Dynamics.

In this work, we revisit the fragile-to-strong transition (FTS) in the simulated BKS silica from the perspective of microscopic dynamics in an effort to elucidate the dynamical behaviors of fragile and strong glass-forming liquids. Softness, which is a machine-learned feature from local atomic structures, is used to predict the microscopic activation energetics and long-term dynamics. The FTS is found to originate from a change in the temperature dependence of the microscopic activation energetics. Furthermore, results suggest there are two diffusion channels with different energy barriers in BKS silica. The fast dynamics at high temperatures is dominated by the channel with small energy barriers (<∼1 eV), which is controlled by the short-range order. The rapid closing of this diffusion channel when lowering temperature leads to the fragile behavior. On the other hand, the slow dynamics at low temperatures is dominated by the channel with large energy barriers controlled by the medium-range order. This slow diffusion channel changes only subtly with temperature, leading to the strong behavior. The distributions of barriers in the two channels show different temperature dependences, causing a crossover at ∼3100 K. This transition temperature in microscopic dynamics is consistent with the inflection point in the configurational entropy, suggesting there is a fundamental correlation between microscopic dynamics and thermodynamics.

CO Oxidation Catalytic Effects of Intrinsic Surface Defects in Rhombohedral LaMnO3.

There is an ongoing effort to replace rare and expensive noble-element catalysts with more abundant and less expensive transition metal oxides. With this goal in mind, the intrinsic defects of a rhombohedral perovskite-like structure of LaMnO3 and their implications on CO catalytic properties were studied. Surface thermodynamic stability as a function of pressure (P) and temperature (T) were calculated to find the most stable surface under reaction conditions (P=0.2 atm, T=323 K to 673 K). Crystallographic planes (100), (111), (110), and (211) were evaluated and it was found that (110) with MnO2 termination was the most stable under reaction conditions. Adsorption energies of O2 and CO on (110) as well as the effect of intrinsic defects such as Mn and O vacancies were also calculated. It was found that O vacancies favor the interaction of CO on the surface, whereas Mn vacancies can favor the formation of carbonate species.

Surface Diffusion Is Controlled by Bulk Fragility across All Glass Types.

Surface diffusion is vastly faster than bulk diffusion in some glasses, but only moderately enhanced in others. We show that this variation is closely linked to bulk fragility, a common measure of how quickly dynamics is excited when a glass is heated to become a liquid. In fragile molecular glasses, surface diffusion can be a factor of 10^{8} faster than bulk diffusion at the glass transition temperature, while in the strong system SiO_{2}, the enhancement is a factor of 10. Between these two extremes lie systems of intermediate fragility, including metallic glasses and amorphous selenium and silicon. This indicates that stronger liquids have greater resistance to dynamic excitation from bulk to surface and enables prediction of surface diffusion, surface crystallization, and formation of stable glasses by vapor deposition.

Machine learning for impurity charge-state transition levels in semiconductors from elemental properties using multi-fidelity datasets.

Quantifying charge-state transition energy levels of impurities in semiconductors is critical to understanding and engineering their optoelectronic properties for applications ranging from solar photovoltaics to infrared lasers. While these transition levels can be measured and calculated accurately, such efforts are time-consuming and more rapid prediction methods would be beneficial. Here, we significantly reduce the time typically required to predict impurity transition levels using multi-fidelity datasets and a machine learning approach employing features based on elemental properties and impurity positions. We use transition levels obtained from low-fidelity (i.e., local-density approximation or generalized gradient approximation) density functional theory (DFT) calculations, corrected using a recently proposed modified band alignment scheme, which well-approximates transition levels from high-fidelity DFT (i.e., hybrid HSE06). The model fit to the large multi-fidelity database shows improved accuracy compared to the models trained on the more limited high-fidelity values. Crucially, in our approach, when using the multi-fidelity data, high-fidelity values are not required for model training, significantly reducing the computational cost required for training the model. Our machine learning model of transition levels has a root mean squared (mean absolute) error of 0.36 (0.27) eV vs high-fidelity hybrid functional values when averaged over 14 semiconductor systems from the II-VI and III-V families. As a guide for use on other systems, we assessed the model on simulated data to show the expected accuracy level as a function of bandgap for new materials of interest. Finally, we use the model to predict a complete space of impurity charge-state transition levels in all zinc blende III-V and II-VI systems.

Electronic Structure-Based Descriptors for Oxide Properties and Functions.

ConspectusThe transition from fossil fuels to renewable energy requires the development of efficient and cost-effective energy storage technologies. A promising way forward is to harness the energy of intermittent renewable sources, such as solar and wind, to perform (electro)catalytic reactions to generate fuels, thus storing energy in the form of chemical bonds. However, current catalysts rely on the use of expensive, rare, or geographically localized elements, such as platinum. Widespread adoption of new (electro)catalytic technologies hinges on the discovery and development of materials containing earth-abundant elements, which can efficiently catalyze an array of (electro)chemical reactions.In the context of catalysis, descriptors provide correlations between fundamental physical properties, such as the electronic structure, and the resulting catalytic activity. The use of easily accessible descriptors has proven to be a powerful method to advance and accelerate discovery and design of new catalyst materials. The position of the oxygen electronic 2p band center has been proposed to capture the basic physical properties of oxides, including oxygen vacancy formation energy, diffusion barrier of oxygen ions, and work function. Moreover, the adsorption strength of relevant reaction intermediates at the surface of oxides can be strongly correlated with the energy of the oxygen 2p states, which affects the catalytic activity of reactions, such as oxygen electrocatalysis, and oxidative dehydrogenation of organic molecules. Such descriptors for catalytic activity can be used to predict the activity of new catalysts and understand trends and behavior among different catalysts.In this Account, we discuss how the energy of the oxygen 2p states can be used as a descriptor for oxide bulk and surface chemical properties. We show how the oxide redox properties vary linearly with the position of the oxygen 2p band center with respect to the Fermi level, and we discuss how this descriptor can be expanded across different materials and structural families, including possible generalizations to compounds outside oxides. We highlight the power of the oxygen 2p band center to predict the catalytic activity of oxides. We conclude with an outlook examining under which conditions this descriptor can be applied to predict oxide properties and possible opportunities for further refining and accelerating property predictions of oxides by leveraging material databases and machine learning.

Machine learning principles applied to CT radiomics to predict mucinous pancreatic cysts.

PURPOSE: Current diagnostic and treatment modalities for pancreatic cysts (PCs) are invasive and are associated with patient morbidity. The purpose of this study is to develop and evaluate machine learning algorithms to delineate mucinous from non-mucinous PCs using non-invasive CT-based radiomics. METHODS: A retrospective, single-institution analysis of patients with non-pseudocystic PCs, contrast-enhanced computed tomography scans within 1 year of resection, and available surgical pathology were included. A quantitative imaging software platform was used to extract radiomics. An extreme gradient boosting (XGBoost) machine learning algorithm was used to create mucinous classifiers using texture features only, or radiomic/radiologic and clinical combined models. Classifiers were compared using performance scoring metrics. Shapely additive explanation (SHAP) analyses were conducted to identify variables most important in model construction. RESULTS: Overall, 99 patients and 103 PCs were included in the analyses. Eighty (78%) patients had mucinous PCs on surgical pathology. Using multiple fivefold cross validations, the texture features only and combined XGBoost mucinous classifiers demonstrated an area under the curve of 0.72 ± 0.14 and 0.73 ± 0.14, respectively. By SHAP analysis, root mean square, mean attenuation, and kurtosis were the most predictive features in the texture features only model. Root mean square, cyst location, and mean attenuation were the most predictive features in the combined model. CONCLUSION: Machine learning principles can be applied to PC texture features to create a mucinous phenotype classifier. Model performance did not improve with the combined model. However, specific radiomic, radiologic, and clinical features most predictive in our models can be identified using SHAP analysis.

Graph network based deep learning of bandgaps.

Recent machine learning models for bandgap prediction that explicitly encode the structure information to the model feature set significantly improve the model accuracy compared to both traditional machine learning and non-graph-based deep learning methods. The ongoing rapid growth of open-access bandgap databases can benefit such model construction not only by expanding their domain of applicability but also by requiring constant updating of the model. Here, we build a new state-of-the-art multi-fidelity graph network model for bandgap prediction of crystalline compounds from a large bandgap database of experimental and density functional theory (DFT) computed bandgaps with over 806 600 entries (1500 experimental, 775 700 low-fidelity DFT, and 29 400 high-fidelity DFT). The model predicts bandgaps with a 0.23 eV mean absolute error in cross validation for high-fidelity data, and including the mixed data from all different fidelities improves the prediction of the high-fidelity data. The prediction error is smaller for high-symmetry crystals than for low symmetry crystals. Our data are published through a new cloud-based computing environment, called the "Foundry," which supports easy creation and revision of standardized data structures and will enable cloud accessible containerized models, allowing for continuous model development and data accumulation in the future.