Unlocking Potential of Pyrochlore in Energy Systems via Soft Voting Ensemble Learning.

In traditional machine learning (ML)-based material design, the defects of low prediction accuracy, overfitting and low generalization ability are mainly caused by the training of a single ML model. Here, a Soft Voting Ensemble Learning (SVEL) approach is proposed to solve the above issues by integrating multiple ML models in the same scene, thus pursuing more stable and reliable prediction. As a case study, SVEL is applied to develop the broad chemical space of novel pyrochlore electrocatalysts with the molecular formula of A2B2O7, to explore promising pyrochlore oxides and accelerate predictions of unknown pyrochlore in the periodic table. The model successfully established the structure-property relationship of pyrochlore, and selected six cost-effective pyrochlore from the periodic table with a high prediction accuracy of 91.7%, all of which showed good electrocatalytic performance. SVEL not only effectively avoids the high costs of experimentation and lengthy computations, but also addresses biases arising from data scarcity in single models. Furthermore, it has significantly reduced the research cycle of pyrochlore by ≈ 22 years, offering broad prospects for accelerating the development of materials genomics. SVEL method is intended to integrate multiple AI models to provide broader model training clues for the AI material design community.

Long-Cycling-Life Sodium-Ion Battery Using Binary Metal Sulfide Hybrid Nanocages as Anode.

Due to the relatively high capacity and lower cost, transition metal sulfides (TMS) as anode show promising potential in sodium-ion batteries (SIBs). Herein, a binary metal sulfide hybrid consisting of carbon encapsulated CoS/Cu2 S nanocages (CoS/Cu2 S@C-NC) is constructed. The interlocked hetero-architecture filled with conductive carbon accelerates the Na+ /e- transfer, thus leading to improved electrochemical kinetics. Also the protective carbon layer can provide better volume accommondation upon charging/discharging. As a result, the battery with CoS/Cu2 S@C-NC as anode displays a high capacity of 435.3 mAh g-1 after 1000 cycles at 2.0 A g-1 (≈3.4 C). Under a higher rate of 10.0 A g-1 (≈17 C), a capacity of as high as 347.2 mAh g-1 is still remained after long 2300 cycles. The capacity decay per cycle is only 0.017%. The battery also exhibits a better temperature tolerance at 50 and -5 °C. A low internal impedance analyzed by X-ray diffraction patterns and galvanostatic intermittent titration technique, narrow band gap, and high density of states obtained by first-principle calculations of the binary sulfides, ensure the rapid Na+ /e- transport. The long-cycling-life SIB using binary metal sulfide hybrid nanocages as anode shows promising applications in versatile electronic devices.

A Pyramidal Metal-Organic Frameworks-Derived FeP@CoP Aluminium-Ion Battery Cathode Displaying Low-Temperature Tolerance and Fast Electron Transfer Kinetics.

Anhui Provincial Engineering Laboratory for Engineering appropriate cathode materials is significant for the development of high-performance aluminum-ion (Al-ion) batteries. Here, a pyramidal metal-organic frameworks (MOFs)-derived FeP@CoP composite was developed as cathode, which exhibits good stability and high capacity. FeP@CoP cathode maintains a high capacity of 168 mAh g-1 after 200 cycles, and displays a stable rate-performance at both room and low temperatures of -10 °C. After three rounds of rate-performance cycling, the FeP@CoP composite recovers 178.2 mAh g-1 at 0.3 A g-1 . Moreover, density functional theory (DFT) calculations verify improved electron-transfer kinetics with narrowed band gap and enhanced density of states. These findings inspire a broad set of studies on MOFs-derived composites for high-performance secondary batteries.

MatGPT: A Vane of Materials Informatics from Past, Present, to Future.

Combining materials science, artificial intelligence (AI), physical chemistry, and other disciplines, materials informatics is continuously accelerating the vigorous development of new materials. The emergence of "GPT (Generative Pre-trained Transformer) AI" shows that the scientific research field has entered the era of intelligent civilization with "data" as the basic factor and "algorithm + computing power" as the core productivity. The continuous innovation of AI will impact the cognitive laws and scientific methods, and reconstruct the knowledge and wisdom system. This leads to think more about materials informatics. Here, a comprehensive discussion of AI models and materials infrastructures is provided, and the advances in the discovery and design of new materials are reviewed. With the rise of new research paradigms triggered by "AI for Science", the vane of materials informatics: "MatGPT", is proposed and the technical path planning from the aspects of data, descriptors, generative models, pretraining models, directed design models, collaborative training, experimental robots, as well as the efforts and preparations needed to develop a new generation of materials informatics, is carried out. Finally, the challenges and constraints faced by materials informatics are discussed, in order to achieve a more digital, intelligent, and automated construction of materials informatics with the joint efforts of more interdisciplinary scientists.

SnS2 quantum dots-coated VO2@carbon nanorods for secondary battery displaying high capacity and rate-performance.

Electrode materials optimization is one of the keys to improving the energy storage characteristics of secondary batteries. Herein, a VO2@carbon@SnS2 composite is developed by coating SnS2 quantum dots (QDs) on lamellar VO2@carbon nanorods, yielding a high-performance aluminum-ion battery cathode. SnS2 QDs embedded in VO2@carbon accelerate electron transport, while the in situ coating of carbon improves cycling stability. When cycling at 0.5 A g-1, capacity is maintained at 157.6 mA h g-1 after 200 cycles. Even at 1.0 A g-1, the cathode can be stably cycled 1000 times. Capacity remains at 176.3 mA h g-1 and coulombic efficiency is 99.1% at temperatures below -10 °C after 100 cycles. These findings provide new ideas for the development of QD-modified composites for application in secondary batteries.

RotNet: A Rotationally Invariant Graph Neural Network for Quantum Mechanical Calculations.

Deep learning has proven promising in biological and chemical applications, aiding in accurate predictions of properties such as atomic forces, energies, and material band gaps. Traditional methods with rotational invariance, one of the most crucial physical laws for predictions made by machine learning, have relied on Fourier transforms or specialized convolution filters, leading to complex model design and reduced accuracy and efficiency. However, models without rotational invariance exhibit poor generalization ability across datasets. Addressing this contradiction, this work proposes a rotationally invariant graph neural network, named RotNet, for accurate and accelerated quantum mechanical calculations that can overcome the generalization deficiency caused by rotations of molecules. RotNet ensures rotational invariance through an effective transformation and learns distance and angular information from atomic coordinates. Benchmark experiments on three datasets (protein fragments, electronic materials, and QM9) demonstrate that the proposed RotNet framework outperforms popular baselines and generalizes well to spatial data with varying rotations. The high accuracy, efficiency, and fast convergence of RotNet suggest that it has tremendous potential to significantly facilitate studies of protein dynamics simulation and materials engineering while maintaining physical plausibility.

High-Entropy Alloy Electrocatalysts Bidirectionally Promote Lithium Polysulfide Conversions for Long-Cycle-Life Lithium-Sulfur Batteries.

High-entropy alloys (HEAs) have attracted considerable attention, owing to their exceptional characteristics and high configurational entropy. Recent findings demonstrated that incorporating HEAs into sulfur cathodes can alleviate the shuttling effect of lithium polysulfides (LiPSs) and accelerate their redox reactions. Herein, we synthesized nano Pt0.25Cu0.25Fe0.15Co0.15Ni0.2 HEAs on hollow carbons (HCs; denoted as HEA/HC) by a facile pyrolysis strategy. The HEA/HC nanostructures were further integrated into hypha carbon nanobelts (HCNBs). The solid-solution phase formed by the uniform mixture of the five metal elements, i.e., Pt0.25Cu0.25Fe0.15Co0.15Ni0.2 HEAs, gave rise to a strong interaction between neighboring atoms in different metals, resulting in their adsorption energy transformation across a wide, multipeak, and nearly continuous spectrum. Meanwhile, the HEAs exhibited numerous active sites on their surface, which is beneficial to catalyzing the cascade conversion of LiPSs. Combining density functional theory (DFT) calculations with detailed experimental investigations, the prepared HEAs bidirectionally catalyze the cascade reactions of LiPSs and boost their conversion reaction rates. S/HEA@HC/HCNB cathodes achieved a low 0.034% decay rate for 2000 cycles at 1.0 C. Notably, the S/HEA@HC/HCNB cathode delivered a high initial areal capacity of 10.2 mAh cm-2 with a sulfur loading of 9 mg cm-2 at 0.1 C. The assembled pouch cell exhibited a capacity of 1077.9 mAh g-1 at the first discharge at 0.1 C. The capacity declined to 71.3% after 43 cycles at 0.1 C. In this work, we propose to utilize HEAs as catalysts not only to improve the cycling stability of lithium-sulfur batteries, but also to promote HEAs in energy storage applications.