Accurately predicting the spin Hall conductivity (SHC) is crucial for designing novel spintronic devices that leverage the spin Hall effect. First-principles calculations of SHCs are computationally intensive and unsuitable for quick hight-hroughput screening. Here, we have developed a residual crystal graph convolutional neural network (Res-CGCNN) deep learning model to classify and predict SHCs solely based on the structural and compositional information. This is enabled by having access to 9249 instances of SHCs data and incorporating extra residual networks into the standard CGCNN framework. We found that Res-CGCNN surpasses CGCNN, achieving a mean absolute error of 115.4 (ħ/e) (S/cm) for regression and an area under the receiver operating characteristic curve of 0.86 for classification. Additionally, we utilized Res-CGCNN to conduct hight-hroughput screenings of materials in the Materials Project database that were absent in the training set. This led to the prediction of several previously unreported materials displaying large SHCs exceeding 1000 (ħ/e) (S/cm), which were validated through first-principles calculations. This study represents the inaugural endeavor to construct a machine learning model capable of effectively capturing the intricate nonlinear relationship between SHCs and crystal structure and composition, serving as a useful tool for the efficient screening and design of materials exhibiting high SHCs.
Prediction of creep rupture life of high-temperature titanium alloys is crucial for their practical applications. The efficient representations (features) of the information encoded in the data are essential to achieve an accurate prediction model. Here, using convolutional neural networks (CNN) enhanced features, we obtain largely improved prediction models for creep rupture life. Comparison of CNN-based features with the original features in describing different samples reveals that the former, by assigning more individualized labels, outperforms the latter and underpins improved prediction models. This work suggests that beyond images, CNN is also suitable for numerical data to obtain enhanced features and surrogate models.
This paper reviews various inverse analysis models used in steel material design, with a focus on integrating process, microstructure, and properties through advanced machine learning techniques. The study underscores the importance of establishing comprehensive models that effectively link these elements for enhanced materials engineering. Key models discussed include the convolutional neural network–artificial neural network-coupled model, which employs convolutional neural networks for feature extraction; the Bayesian-optimized generative adversarial network–conditional generative adversarial network model, which generates diverse virtual microstructures; the multi-objective optimization model, which concentrates on process–property relationships; and the microstructure–process parallelization model, which correlates microstructural features with process conditions. Each model is assessed for its strengths and limitations, influencing its practical applicability in material design. The paper concludes by advocating for continued improvements in model accuracy and versatility, with the ultimate goal of enhancing steel properties and expanding the scope of data-driven material development.
The design of advanced materials for applications in areas of photovoltaics, energy storage, and structural engineering has made significant strides. However, the rapid proliferation of candidate materials—characterized by structural complexity that complicates the relationships between features—presents substantial challenges in manufacturing, fabrication, and characterization. This review introduces a comprehensive methodology for materials design using cutting-edge quantum computing, with a particular focus on quadratic unconstrained binary optimization (QUBO) and quantum machine learning (QML). We introduce the loop framework for QUBO-empowered materials design, including constructing high-quality datasets that capture critical material properties, employing tailored computational methods for precise material modeling, developing advanced figures of merit to evaluate performance metrics, and utilizing quantum optimization algorithms to discover optimal materials. In addition, we delve into the core principles of QML and illustrate its transformative potential in accelerating material discovery through a range of quantum simulations and innovative adaptations. The review also highlights advanced active learning strategies that integrate quantum artificial intelligence, offering a more efficient pathway to explore the vast, complex material design space. Finally, we discuss the key challenges and future opportunities for QML in material design, emphasizing their potential to revolutionize the field and facilitate groundbreaking innovations.
Although the power conversion efficiency of organic solar cells (OSCs) has been rapidly improved, there is still a lot of room for designing and developing new materials and their combinations to approach the efficiency limit. In this work, we establish a database of ∼100 bulk heterojunction OSCs composed of representative donors and acceptors reported in the literature, and train machine learning models to identify the efficiency potential of donor-acceptor combinations. We find that the fully connected neural network achieves a Pearson coefficient of up to 0.88 for predicting the efficiency of OSCs with different combinations of donors and acceptors. We use sure independence screening and sparsifying method with feature analysis to analyze and evaluate the performance of OSCs. To prove the reliability and viability of the predictive model, we introduce the theoretical efficiency limits and confidence tests into the process, which provides a simple but reliable solution to quickly analyze and evaluate the potential of OSC materials and material combinations.
This paper investigates the dynamic recrystallization characteristics of SAE52100 large section bearing steel under hot compression, focusing on both the center and surface. Using data from thermal simulation experiments the physical models were developed. Four machine learning algorithms including support vector regression, k-nearest neighbors, random forest, and extreme gradient boosting were then employed to develop dynamic recrystallization prediction models based on the experimental data and inferred values from the physical model. The results show that the machine learning methods provide a better numerical description of the model, provided these are fed with extensive data. To enhance the scope of application, we obtained data from the dynamic recrystallization models for both the center and surface of SAE52100 steel in the as-cast state, as well as extrapolated values from the literature regarding the hot-rolled condition. When the SHAP method was introduced to reveal the mechanism of the influence of each input feature on the prediction results of the machine learning model, it was found that the test results of the Cr element did not match the theory, mainly because of the small scale of Cr elemental data and the strong dependence on grain size and secondary dendrite spacing.
Machine learning (ML) techniques have made enormous progress in the field of materials science. However, many conventional ML algorithms operate as “blackboxes”, lacking transparency in revealing explicit relationships between material features and target properties. To address this, the development of interpretable ML models is essential to drive further advancements in AI-driven materials discovery. In this study, we present an interpretable framework that combines traditional machine learning with symbolic regression, using Janus III–VI vdW heterostructures as a case study. This approach enables fast and accurate predictions of stability and electronic structure. Our results demonstrate that the prediction accuracy using the classification model for stability, based on formation energy, reaches 0.960. On the other hand, the R2, MAE, and RMSE value using the regression model for electronic structure prediction, based on band gap, achieves 0.927, 0.113, and 0.141 on the testing set, respectively. Additionally, we identify a universal interpretable descriptor comprising five simple parameters that reveals the underlying physical relationships between the candidate heterostructures and their band gaps. This descriptor not only delivers high accuracy in band gap prediction but also provides explicit physical insight into the material properties.
Magnesium alloys, known for their lightweight advantages, are increasingly in demand across a range of applications, from aerospace to the automotive industry. With rising requirements for strength and corrosion resistance, the development of new magnesium alloy systems has become critical. Phase diagrams play a crucial role in guiding the magnesium alloy design by providing key insights into phase stability, composition, and temperature ranges, enabling the optimization of alloy properties and processing conditions. However, accessing and interpreting phase diagram data with thermodynamic calculation software can be complex and time-consuming, often requiring intricate calculations and iterative refinement based on thermodynamic models. To address this challenge, we introduce PDGPT, a ChatGPT-based large language model designed to streamline the acquisition of magnesium alloys Phase Diagram information with high efficiency and accuracy. Enhanced by prompt-engineering, supervised fine-tuning and retrieval-augmented generation, PDGPT leverages the predictive and reasoning capabilities of large language models along with computational phase diagram data. By combining large language models with traditional phase diagram research tools, PDGPT not only improves the accessibility of critical phase diagram information but also sets the stage for future advancements in applying large language models to materials science.
Solution styrene-butadiene rubber (SSBR) finds wide applications in high performance tire design and various other fields. This study aims to create a quantitative structure–property relationship (QSPR) model linking SSBR’s glass transition temperature (Tg) to its structural properties. A dataset of 68 sets of data from published literature was compiled to develop a predictive machine learning model for SSBR’s structural design and synthesis using small sample sizes. To tackle small sample sizes, a framework combining generative adversarial networks (GAN) and the Tree-based Pipeline Optimization Tool (TPOT) is proposed. GAN is first used to generate additional samples that mirror the original dataset’s distribution, expanding the dataset. The TPOT is then applied to automatically find the best model and parameter combinations, creating an optimal predictive model for the mixed dataset. Experimental results show that using GAN to enlarge the dataset and TPOT regression models significantly enhances model performance, increasing the R2 value from 0.745 to 0.985 and decreasing the RMSE from 7.676 to 1.569. The proposed GAN–TPOT framework demonstrates the potential of combining generative models with automated machine learning to improve materials science research. This combination accelerates research and development processes, enhances prediction and design accuracy, and introduces new perspectives and possibilities for the field.
Metamaterial design, encompassing both microstructure topology selection and geometric parameter optimization, constitutes a high-dimensional optimization problem, with computationally expensive and time-consuming design evaluations. Bayesian optimization (BO) offers a promising approach for black-box optimization involved in various material designs, and this work presents several advanced techniques to adapt BO to address the challenges associated with metamaterial design. First, variational autoencoders (VAEs) are employed for efficient dimensionality reduction, mapping complex, high-dimensional metamaterial microstructures into a compact latent space. Second, mutual information maximization is incorporated into the VAE to enhance the quality of the learned latent space, ensuring that the most relevant features for optimization are retained. Third, trust region-based Bayesian optimization (TuRBO) dynamically adjusts local search regions, ensuring stability and convergence in high-dimensional spaces. The proposed techniques are well incorporated with conventional Gaussian processes (GP)-based BO framework. We applied the proposed method for the design of electromagnetic metamaterial microstructures. Experimental results show that we achieve a significantly high probability of finding the ground-truth topology types and their geometric parameters, leading to high accuracy in matching the design target. Moreover, our approach demonstrates significant time efficiency compared with traditional design methods.