Machine Learning Screening of Metal-Ion Battery Electrode Materials

Data-Driven Battery Characterization and Prognosis: Recent Progress, Challenges, and Prospects

Battery characterization and prognosis are essential for analyzing underlying electrochemical mechanisms and ensuring safe operation, especially with the assistance of superior data-driven artificial intelligence systems. This review provides a unique perspective on recent progress in data-driven battery characterization and prognosis methods. First, recent informative image characterization and impedance spectrum as well as high-throughput screening approaches on revealing battery electrochemical mechanisms at multiple scales are summarized. Thereafter, battery prognosis tasks and strategies are described, with the comparison of various physics-informed modeling strategies. Considering unlocking mechanisms from tremendous battery data, the dominant role of physics-informed interpretable learning in accelerating energy device development is presented. Finally, challenges and prospects on data-driven characterization and prognosis are discussed toward accelerating energy device development with much-enhanced electrochemical transparency and generalization. This review is hoped to supply new ideas and inspirations to the next-generation battery development.

Employing Graph Neural Networks for Predicting Electrode Average Voltages and Screening High-Voltage Sodium Cathode Materials

For many years, humans have been relentlessly focused on enhancing battery longevity and boosting energy storage capacities. The performance and durability of a battery depend significantly on the material used for its electrodes. In this context, merging machine learning with density functional theory (DFT) calculations has emerged as a pivotal approach to advancing the exploration of battery crystal structures. We present a new method that combines a graph convolutional neural network (GNN) with a Transformer convolutional layer, which we call Transformer-GNN. To underscore its efficacy, we benchmarked Transformer-GNN against three established statistical machine learning models: Support Vector Machine, Random Forest, and XGBoost. We also developed a standard GNN, which we refer to as Basic-GNN. Additionally, we compared Basic-GNN with Transformer-GNN to highlight the improvements brought about by incorporating the Transformer convolutional layer. The Transformer-GNN model outperforms the other models, achieving the highest R2 value of 0.82 and the lowest mean squared error of 0.3161. Our findings demonstrate that the Transformer-GNN can profoundly understand battery crystal structures, thus forging the path toward more sophisticated and durable battery systems. Leveraging the GNN model's voltage predictions in tandem with the capacity data sourced from the database, we screened and pinpointed Na(NiO2)2 as a high-voltage (higher than 5 V), high-capacity sodium cathode material. We conducted DFT calculations on Na(NiO2)2 and revealed the migration mechanism of the Na ions.

Exploring the large chemical space in search of thermodynamically stable and mechanically robust MXenes via machine learning

To effectively utilize MXenes, a family of two-dimensional materials, in various applications that include thermoelectric devices, semiconductors, and transistors, their thermodynamic and mechanical properties, which are closely related to their stability, must be understood. However, exploring the large chemical space of MXenes and verifying their stability using first-principles calculations are computationally expensive and inefficient. Therefore, this study proposes a machine learning (ML)-based high-throughput MXene screening framework to identify thermodynamically stable MXenes and determine their mechanical properties. A dataset of 23 857 MXenes with various compositions was used to validate this framework, and 48 MXenes were predicted to be stable by ML models in terms of heat of formation and energy above the convex hull. Among them, 45 MXenes were validated using density functional theory calculations, of which 23 MXenes, including Ti2CClBr and Zr2NCl2, have not been previously known for their stability, confirming the effectiveness of this framework. The in-plane stiffness, shear moduli, and Poisson's ratio of the 45 MXenes were observed to vary widely according to their constituent elements, ranging from 90.11 to 198.02 N m-1, 64.00 to 163.40 N m-1, and 0.19 to 0.58, respectively. MXenes with Group-4 transition metals and halogen surface terminations were shown to be both thermodynamically stable and mechanically robust, highlighting the importance of electronegativity difference between constituent elements. Structurally, a smaller volume per atom and minimum bond length were determined to be preferable for obtaining mechanically robust MXenes. The proposed framework, along with an analysis of these two properties of MXenes, demonstrates immense potential for expediting the discovery of stable and robust MXenes.

Unlocking the Potential of Dual-Ion Batteries: Identifying Polycyclic Aromatic Hydrocarbon Cathodes and Intercalating Salt Combinations through Machine Learning

Dual-ion batteries (DIBs) represent a promising energy storage technology, offering a cost-effective safe solution with impressive electrochemical performance. The large combinatorial configuration space of the electrode-electrolyte leads to design challenges. We present a machine learning (ML) approach for accurately predicting the voltage and volume changes of polycyclic aromatic hydrocarbon (PAH) cathodes upon intercalation with a variety of DIB salts following different mechanisms. Gradient Boosting and XGBoost Regression models trained on the data set demonstrate exceptional performance in voltage and volume change prediction, respectively. The models are further cross-validated and utilized to predict the properties for ∼700 combinations of PAH and DIB salt intercalations, a subset of which is further validated by density functional theory. Using average voltage and volume change for all combinations of PAHs and salts, preferable combinations for high/low voltage requirements along with long-term stability are obtained. Overall, the study shows the applicability of PAHs in DIBs exhibiting good electrochemical performance with low volume change compared to graphite indicative of its potential to overcome the cycling stability issues of DIBs. This research establishes a reliable and broadly applicable ML-based workflow for efficient screening and accelerated design of advanced PAH cathodes and salts, thus driving progress in the field of DIBs.

Neural Network with Optimal Neuron Activation Functions Based on Additive Gaussian Process Regression

Feed-forward neural networks (NNs) are a staple machine learning method widely used in many areas of science and technology, including physical chemistry, computational chemistry, and materials informatics. While even a single-hidden-layer NN is a universal approximator, its expressive power is limited by the use of simple neuron activation functions (such as sigmoid functions) that are typically the same for all neurons. More flexible neuron activation functions would allow the use of fewer neurons and layers and thereby save computational cost and improve expressive power. We show that additive Gaussian process regression (GPR) can be used to construct optimal neuron activation functions that are individual to each neuron. An approach is also introduced that avoids nonlinear fitting of neural network parameters by defining them with rules. The resulting method combines the advantage of robustness of a linear regression with the higher expressive power of an NN. We demonstrate the approach by fitting the potential energy surfaces of the water molecule and formaldehyde. Without requiring any nonlinear optimization, the additive-GPR-based approach outperforms a conventional NN in the high-accuracy regime, where a conventional NN suffers more from overfitting.

Recent Advancements in Devising Computational Strategies for Dual-Ion Batteries

Dual-ion batteries (DIBs) have been considered a viable alternative to increasingly costly and hazard-prone lithium-ion batteries (LIBs), which have reached a level of saturation. DIBs differ from LIBs in the way that the cations and anions originate from the electrolyte, thus signifying the active role played by electrolyte. In this Review, the major developments in research in the field of DIBs are summarized with a major emphasis on computational approaches in this direction. The various computational methods for understanding and designing electrodes are discussed. The advancements in electrode and electrolyte design for efficient DIBs are highlighted. Further, the ways to investigate solid-electrolyte interphase formation through simulations to comprehend the role of various components are discussed. Finally, directions are given on which future computational research can be carried out to design futuristic DIBs to provide useful guidelines to the researchers to understand and design DIBs.

Machine Learning-Aided Design of Gold Core-Shell Nanocatalysts toward Enhanced and Selective Photooxygenation

We demonstrate the use of the machine learning (ML) tools to rapidly and accurately predict the electric field as a guide for designing core-shell Au-silica nanoparticles to enhance ¹O₂ sensitization and selectivity of organic synthesis. Based on the feature importance analysis, obtained from a deep neural network algorithm, we found a general and linear dependent descriptor (θ ∝ aD^0.25t^-1, where a, D, and t are the shape constant, size of metal nanoparticles, and distance from the metal surface) for the electric field around the core-shell plasmonic nanoparticle. Directed by the new descriptor, we synthesized gold-silica nanoparticles and validated their plasmonic intensity using scanning transmission electron microscopy-electron energy loss spectroscopy (STEM-EELS) mapping. The nanoparticles with θ = 0.40 demonstrate an ∼3-fold increase in the reaction rate of photooxygenation of anthracene and 4% increase in the selectivity of photooxygenation of dihydroartemisinic acid (DHAA), a long-standing goal in organic synthesis. In addition, the combination of ML and experimental investigations shows the synergetic effect of plasmonic enhancement and fluorescence quenching, leading to enhancement for ¹O₂ generation. Our results from time-dependent density functional theory (TD-DFT) calculations suggest that the presence of an electric field can favor intersystem crossing (ISC) of methylene blue to enhance ¹O₂ generation. The strategy reported here provides a data-driven catalyst preparation method that can significantly reduce experimental cost while paving the way for designing photocatalysts for organic drug synthesis.

Accurate Prediction of Voltage of Battery Electrode Materials Using Attention-Based Graph Neural Networks

Performing first-principles calculations to discover electrodes' properties in the large chemical space is a challenging task. While machine learning (ML) has been applied to effectively accelerate those discoveries, most of the applied methods ignore the materials' spatial information and only use predefined features: based only on chemical compositions. We propose two attention-based graph convolutional neural network techniques to learn the average voltage of electrodes. Our proposed methods, which combine both atomic composition and atomic coordinates in 3D-space, improve the accuracy in voltage prediction significantly when compared to composition-based ML models. The first model directly learns the chemical reaction of electrodes and metal ions to predict their average voltage, whereas the second model combines electrodes' ML predicted formation energy (Eform) to compute their average voltage. Our Eform-based model demonstrates improved accuracy in transferability from our subset of learned Li ions to Na ions. Moreover, we predicted the theoretical voltage of 10 NaxMPO4F (M = Ti, Cr, Fe, Cu, Mn, Co, and Ni) fluorophosphate battery frameworks, which are unavailable in the Material Project database. It could be shown that we can expect average voltages higher than 3.1 V from those Na battery frameworks except from the NaTiPO4F and TiPO4F pair of electrodes, which offer an average voltage of 1.32 V.