Search

Abstract We report a novel deep learning (DL) method for classifying inorganic compounds using 3D electron density data. We transform Density Functional Theory (DFT)-derived CHGCAR files from the Materials Project (MP) and experimental data from the Inorganic Crystal Structure Database (ICSD) into point clouds and sparse tensors, optimized for use in DL models such as PointNet and Sparse 3D CNN. This approach effectively overcomes the limitations of handling the dense 3D data, a common challenge in DL. Contrasting with traditional 1D or 2D X-ray diffraction (XRD) patterns that necessitate complex reciprocal space analysis, our method utilizes 3D density data for direct interpretation in real lattice space. This shift significantly enhances classification accuracy, outperforming traditional XRD-driven DL methods. We achieve accuracies of 97.28%, 90.77%, and 90.10% for crystal system, extinction group, and space group classifications, respectively. Our 3D electron density-based DL approach not only showcases improved accuracy but also contributes a more intuitive and effective framework for materials discovery.

Abstract Mechanoluminescence (ML) and digital image correlation (DIC) have emerged as promising optical methods to visualize and measure deformation fields. In this study, a dual‐modal sensing skin, called the ML‐DIC skin is introduced, that is capable of emitting ML and facilitating DIC measurements under various lighting conditions, including daylight, night or darkness, and UV irradiation. Four ML‐DIC skins are fabricated with or without carbon nanotubes (CNTs) using a composite powder consisting of SrAl2O4: Eu,Dy (SAO), and acrylic resin, with CNT milling times of 48, 72, and 96 h for three of four skins, respectively. DIC measurements are performed under multiple lighting conditions for measuring photoluminescence, persistence luminescence, and reflection. Uniaxial tension tests demonstrate the superior performance of ML‐DIC skins with CNTs compared with pristine SAO skins, with the skin subjected to 48 h of CNT dispersion exhibiting optimal performance. Further investigations focus on ML emission and DIC measurements near the crack‐tip vicinity of static and propagating cracks as well as on surfaces above subsurface cracks. The integration of ML and DIC techniques offers a versatile approach for comprehensive deformation analysis applicable to diverse environments, with implications for materials science, engineering, and structural health monitoring.

On‐site inspection of invisible subsurface defects in multiscale structural materials by conventional nondestructive testing (NDT) methods, such as X‐ray and ultrasound, requires complex sample preparation and data acquisition processes. Moreover, the inspected area is very small. Herein, a simple, inexpensive, and ultrasensitive NDT method for identifying and classifying the geometries of subsurface defects using commercial cameras, digital image correlation software, and object detection (OD) algorithms is developed. Three OD algorithms—Faster region‐based convolutional neural network (Faster R‐CNN), Mask R‐CNN, and you‐only‐look‐once (YOLO)v3—are evaluated for their ability to locate defects and identify defect geometries. Specifically, bounding boxes of two sizes (large and small) are applied to the regions of defect‐induced perturbations in strain tensors, which serve as virtual representatives of invisible subsurface defects. The performance of the proposed approach is validated on test datasets of known and unknown defect types. The experimental results confirm that the proposed approach can effectively utilize the surface deformation field information to accurately and reliably locate and identify subsurface defects. The method is nondestructive and low cost, enables real‐time detection, is robust against noise‐dominated deformation fields, and can be applied to various structural deformations. The method is therefore suitable for multiscale structural health monitoring and characterization of internal defects in materials.

A deep learning (DL)‐based approach for analysis is proposed. Using synthetic XRD data for a DL approach is inevitable due to the lack of real‐world XRD data. There are two main challenges when conducting a DL‐based XRD analysis: generating realistic XRD data including all possible perturbations, such as peak shift, broadening, texture, and noisy background, and generalizing the DL model applicability to all ICSD entries. To address both the perturbation and generalizability issues, a large‐scale computation is required because it would be infeasible with typical lab‐scale computation. Cloud computing infrastructures are leveraged for parallel computations and to obtain symmetry classification test accuracies of 98.95%, 97.18%, and 96.03% for the crystal system, extinction group, and space group, respectively. A stricter individual compound‐based train and test dataset‐splitting scheme also produces reasonable test accuracies of 92.25%, 87.34%, and 84.39%, which are still state‐of‐the‐art records. Crucially, the DL model trained on synthetic data is assessed using real‐world experimental XRD datasets to ensure its practical applicability. When tested on the real‐world experimental XRD dataset, the model achieves a test accuracy of 90.38% in predicting crystal systems.

Abstract The current status of 2D organic–inorganic hybrid perovskites for use in photovoltaic (PV) and light-emitting diode (LED) applications lags far behind their 3D counterparts. Here, we propose a computational strategy for discovering novel perovskites with as few computing resources as possible. A tandem optimization algorithm consisting of an elitism-reinforced nondominated sorting genetic algorithm (NSGA-II) and a multiobjective Bayesian optimization (MOBO) algorithm was used for density functional theory (DFT) calculations. The DFT-calculated band gap and effective mass were taken as objective functions to be optimized, and the constituent molecules and elements of a Ruddlesden–Popper (RP) structure (n = 2) were taken as decision variables. Fourteen previously unknown RP perovskite candidates for PV and LED applications were discovered as a result of the NSGA-II/MOBO algorithm. Thereafter, more accurate DFT calculations based on the HSE06 exchange correlation functional and ab initio molecular dynamics (AIMD) were conducted for the discovered 2D perovskites to ensure their validity.

Summary: Extensive changes in the legal, commercial and technical requirements in engineering fields have necessitated automated real-time structural health monitoring (SHM) and instantaneous verification. An integrated system with mechanoluminescence (ML) and dual artificial intelligence (AI) modules with subsidiary finite element method (FEM) simulation is designed for in situ SHM and instantaneous verification. The ML module detects the exact position of a crack tip and evaluates the significance of existing cracks with a plastic stress-intensity factor (PSIF; KP). ML fields and their corresponding KpML values are referenced and verified using the FEM simulation and bidirectional generative adversarial network (GAN). Well-trained forward and backward GANs create fake FEM and ML images that appear authentic to observers; a convolutional neural network is used to postulate precise PSIFs from fake images. Finally, the reliability of the proposed system to satisfy existing commercial requirements is validated in terms of tension, compact tension, AI, and instrumentation.

Abstract A tandem (two‐step) particle swarm optimization (PSO) algorithm is implemented in the argyrodite‐based multidimensional composition space for the discovery of an optimal argyrodite composition, i.e., with the highest ionic conductivity (7.78 mS cm−1). To enhance the industrial adaptability, an elaborate pellet preparation procedure is not used. The optimal composition (Li5.5PS4.5Cl0.89Br0.61) is fine‐tuned to enhance its practical viability by incorporating oxygen in a stepwise manner. The final composition (Li5.5PS4.23O0.27Cl0.89Br0.61), which exhibits an ionic conductivity (σion) of 6.70 mS cm−1 and an activation barrier of 0.27 eV, is further characterized by analyzing both its moisture and electrochemical stability. Relative to the other compositions, the exposure of Li5.5PS4.23O0.27Cl0.89Br0.61 to a humid atmosphere results in the least amount of H2S released and a negligible change in structure. The improvement in the interfacial stability between the Li(Ni0.9Co0.05Mn0.05)O2 cathode and Li5.5PS4.23O0.27Cl0.89Br0.61 also results in greater specific capacity during fast charge/discharge. The structural and chemical features of Li5.5PS4.5Cl0.89Br0.61 and Li5.5PS4.23O0.27Cl0.89Br0.61 argyrodites are characterized using synchrotron X‐ray diffraction, Raman spectroscopy, and X‐ray photoelectron spectroscopy. This work presents a novel argyrodite composition with favorably balanced properties while providing broad insights into material discovery methodologies with applications for battery development.

Abstract Predicting mechanical properties such as yield strength (YS) and ultimate tensile strength (UTS) is an intricate undertaking in practice, notwithstanding a plethora of well-established theoretical and empirical models. A data-driven approach should be a fundamental exercise when making YS/UTS predictions. For this study, we collected 16 descriptors (attributes) that implicate the compositional and processing information and the corresponding YS/UTS values for 5473 thermo-mechanically controlled processed (TMCP) steel alloys. We set up an integrated machine-learning (ML) platform consisting of 16 ML algorithms to predict the YS/UTS based on the descriptors. The integrated ML platform involved regularization-based linear regression algorithms, ensemble ML algorithms, and some non-linear ML algorithms. Despite the dirty nature of most real-world industry data, we obtained acceptable holdout dataset test results such as R2 > 0.6 and MSE

Herein, data‐driven symmetry identification, property prediction, and low‐dimensional embedding from powder X‐Ray diffraction (XRD) patterns of inorganic crystal structure database (ICSD) and materials project (MP) entries are reported. For this purpose, a fully convolutional neural network (FCN), transformer encoder (T‐encoder), and variational autoencoder (VAE) are used. The results are compared to those obtained from a well‐established crystal graph convolutional neural network (CGCNN). A task‐specified small dataset that focuses on a narrow material system, knowledge (rule)‐based descriptor extraction, and significant data dimension reduction are not the main focus of this study. Conventional powder XRD patterns, which are most widely used in materials research, can be used as a significantly informative material descriptor for deep learning. Both the FCN and T‐encoder outperform the CGCNN for symmetry classification. For property prediction, the performance of the FCN concatenated with multilayer perceptron reaches the performance level of CGCNN. Machine‐learning‐driven material property prediction from the powder XRD pattern deserves appreciation because no such attempts have been made despite common XRD‐driven symmetry (and lattice size) prediction and phase identification. The ICSD and MP data are embedded in the 2D (or 3D) latent space through the VAE, and well‐separated clustering according to the symmetry and property is observed.

Increasing ubiquitous collaborative intelligence between humans and machines requires human–machine communication (HMC) that is more human and less machine‐like to accomplish given tasks. Although speech signals are considered the best modes of communication in HMC, background noise often interferes with these signals. Therefore, research focused on integrating lip‐reading technology into HMC has gained significant attention. However, lip‐reading functions effectively only in well‐lit environments. In contrast, HMC may occur daily in dark environments owing to potential energy shortages, increased exploration in darkness, nighttime emergencies, etc. Herein, a possible method for HMC in the dark mode is presented, which is realized based on deep learning motion patterns of persistent luminescence (PL) of the skin surrounding the lips. An ultrasoft PL–polymer composite patch is used to record the motion pattern of the skin during speech in the dark. It is found that visual geometric group network (VGGNET‐5) and residual neural network (ResNet‐34) could predict spoken words in darkness with test accuracies of 98.5% and 98.75%, respectively. Furthermore, these models could effectively distinguish similar‐sounding words such as “around” and “ground.” Dark‐mode communication can allow a wide range of people, including disabled people with limited dexterity and voice tremors, to communicate with artificial intelligence machines.

Abstract Most data-driven machine learning (ML) approaches established in metallurgy research fields are focused on a build-up of reliable quantitative models that predict a material property from a given set of material conditions. In general, the input feature dimension (the number of material condition variables) is much higher than the output feature dimension (the number of material properties of concern). Rather than such a forward-prediction ML model, it is necessary to develop so-called inverse-design modeling, wherein required material conditions could be deduced from a set of desired material properties. Here we report a novel inverse design strategy that employs two independent approaches: a metaheuristics-assisted inverse reading of conventional forward ML models and an atypical inverse ML model based on a modified variational autoencoder. These two unprecedented approaches were successful and led to overlapped results, from which we pinpointed several novel thermo-mechanically controlled processed (TMCP) steel alloy candidates that were validated by a rule-based thermodynamic calculation tool (Thermo-Calc.). We also suggested a practical protocol to elucidate how to treat engineering data collected from industry, which is not prepared as independent and identically distributed (IID) random data.

Abstract Monitoring structural health using mechanoluminescent (ML) effects is widely considered as a potential full‐field and direct visualizing optical method with high spatial and temporal resolution and simple setup in a noncontact manner. The challenges and uncertainties in the mapping of ML field to effective strain field, however, tend to limit significant commercial ML applications for structural health monitoring systems. Here, however, quantification problems are resolved using the digital image correlation (DIC) method. Specifically, an image containing mechanically induced photon information is processed using a DIC algorithm to measure the strain field components, which enables the establishment of a calibration curve when the ML field is mapped onto the effective strain field using pixel level information. The results show a linear relationship between effective strain and ML intensity despite the plastic flow in ML skin. Furthermore, the calibration curve allows for easy conversion of ML field to effective‐strain field at the crack‐tip plastic zone of the alloy structure, retaining its spatial resolution. The compatibility of ML skin with the DIC algorithm not only enables the quantification of the ML effects of several organic/inorganic ML materials, but may also be useful in elucidating the fundamentals of the trap‐controlled mechanism.

The revolutionary concept of creating a large‐area tactile sensor by training a simple, bulky material through deep learning (DL) is proposed. This enables the replacement of the conventional tactile sensor comprising a patterned array structure with a super‐simple, single‐layer, large‐area tactile sensor pad. A crude carbon nanotube‐dispersed polydimethylsiloxane pad—with a bias applied to the center and the resultant piezoresistive current detected at several electrodes located on the pad edge—plays a smart sensory role without the need for complicated fabrication of microengineered structures. The piezoresistive current while recording the indented location and the pressure thereon is measured, and then various DL models (a multimodel arrangement is necessary due to the viscoelasticity of the pad) using the collected data are trained. The proposed concept is realized using a tandem model comprising a combination of algorithms selected from deep neural networks, convolutional neural networks, long short‐term memory networks, and 16 state‐of‐the‐art machine learning algorithms. The hold‐out dataset test accuracy for the indented location identification reaches 98.89%, and the goodness of fit for pressure prediction is evaluated with mean squared error of 2.5 × 10−3 and coefficient of determination of 98.05%.

Identifying the composition of multiphase inorganic compounds from XRD patterns is challenging. Here the authors use a convolutional neural network to identify phases in unknown multiphase mixed inorganic powder samples with an accuracy of nearly 90%.

Abstract The prediction of macro‐scale materials properties from microstructures, and vice versa, should be a key part in modeling quantitative microstructure‐physical property relationships. It would be helpful if the microstructural input and output were in the form of visual images rather than parameterized descriptors. However, only a typical supervised learning technique would be insufficient to build up a model with real‐image‐output. A generative adversarial network (GAN) is required to treat visual images as output for a promising PMPR model. Recently developed deep‐learning‐based GAN techniques such as a deep convolutional GAN (DCGAN), a cycle‐consistent GAN (Cycle GAN), and a conditional GAN‐based image to image translation (Pix2Pix) could be of great help via the creation of realistic microstructures. In this regard, we generated virtual micrographs for various types of steels using a DCGAN, a Cycle GAN, and a Pix2Pix and confirmed the generated micrographs are qualitatively indistinguishable from the ground truth.

A deep machine-learning technique based on a convolutional neural network (CNN) is introduced. It has been used for the classification of powder X-ray diffraction (XRD) patterns in terms of crystal system, extinction group and space group. About 150 000 powder XRD patterns were collected and used as input for the CNN with no handcrafted engineering involved, and thereby an appropriate CNN architecture was obtained that allowed determination of the crystal system, extinction group and space group. In sharp contrast with the traditional use of powder XRD pattern analysis, the CNN never treats powder XRD patterns as a deconvoluted and discrete peak position or as intensity data, but instead the XRD patterns are regarded as nothing but a pattern similar to a picture. The CNN interprets features that humans cannot recognize in a powder XRD pattern. As a result, accuracy levels of 81.14, 83.83 and 94.99% were achieved for the space-group, extinction-group and crystal-system classifications, respectively. The well trained CNN was then used for symmetry identification of unknown novel inorganic compounds.

Abstract Herein, graphite is proposed as a reliable Ca2+‐intercalation anode in tetraglyme (G4). When charged (reduced), graphite accommodates solvated Ca2+‐ions (Ca‐G4) and delivers a reversible capacity of 62 mAh g−1 that signifies the formation of a ternary intercalation compound, Ca‐G4·C72. Mass/volume changes during Ca‐G4 intercalation and the evolution of in operando X‐ray diffraction studies both suggest that Ca‐G4 intercalation results in the formation of an intermediate phase between stage‐III and stage‐II with a gallery height of 11.41 Å. Density functional theory calculations also reveal that the most stable conformation of Ca‐G4 has a planar structure with Ca2+ surrounded by G4, which eventually forms a double stack that aligns with graphene layers after intercalation. Despite large dimensional changes during charge/discharge (C/D), both rate performance and cyclic stability are excellent. Graphite retains a substantial capacity at high C/D rates (e.g., 47 mAh g−1 at 1.0 A g−1 s vs 62 mAh g−1 at 0.05 A g−1) and shows no capacity decay during as many as 2000 C/D cycles. As the first Ca2+‐shuttling calcium‐ion batteries with a graphite anode, a full‐cell is constructed by coupling with an organic cathode and its electrochemical performance is presented.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

The conventional mechanoluminescence (ML) mechanism of phosphors such as SrAl2O4:Eu and ZnS:Mn is known to utilize carrier trapping at shallow traps followed by stress (or strain)-induced detrapping, which leads to activator recombination in association with local piezoelectric fields. However, such a conventional ML mechanism was found to be invalid for the ZnS:Cu-embedded polydimethylsiloxane (PDMS) composite, due to the absence of luminescence with a rigid matrix and a negligibly small value of the piezoelectric coefficient (d33) of the composite. An alternative mechanism, namely, the triboelectricity-induced luminescence has been proposed for the mechanically driven luminescence of a ZnS:Cu-PDMS composite.