A5102733647- Machine Learning in Materials Science
- Electrocatalysts for Energy Conversion
- Physics of Superconductivity and Magnetism
- Nanomaterials for catalytic reactions
- Chalcogenide Semiconductor Thin Films
- Fuel Cells and Related Materials
- Superconductivity in MgB2 and Alloys
- Perovskite Materials and Applications
- Hybrid Renewable Energy Systems
- Advanced Thermoelectric Materials and Devices
- Advanced Photocatalysis Techniques
- Ferroelectric and Piezoelectric Materials
- Catalysis and Hydrodesulfurization Studies
- Iron-based superconductors research
- Machine Learning and ELM
- Electrochemical Analysis and Applications
- Ammonia Synthesis and Nitrogen Reduction
- Inorganic Chemistry and Materials
- Glass properties and applications
- Thermal properties of materials
- Metallic Glasses and Amorphous Alloys
- Asymmetric Hydrogenation and Catalysis
- Material Dynamics and Properties
- Graphene research and applications
- Theoretical and Computational Physics
Harbin Institute of Technology
2022-2025
PRG S&Tech (South Korea)
2024-2025
Guilin University of Technology
2022
Nanjing Normal University
2019-2020
Guangxi Normal University
2019
Superconductivity allows electric conductance with no energy losses when the ambient temperature drops below a critical value (Tc). Currently, machine learning (ML)-based prediction of potential superconductors has been limited to chemical formulas without explicit treatment material structures. Herein, we implement an efficient structural descriptor, smooth overlap atomic position (SOAP), into ML models predict Tc values information. Using data set containing 5713 compounds, our SOAP...
Predictive materials design of high-performance alloy electrocatalysts is a grand challenge in hydrogen production via water electrolysis. The vast combinatorial space element substitutions offers wealth candidate materials, but presents significant terms experimental and computational exploration all possible options. Recent scientific technological developments machine learning (ML) have offered new opportunity to accelerate such electrocatalyst design. Herein, by incorporating both the...
Abstract The pursuit of designing superconductors with high T c has been a long‐standing endeavor. However, the widespread incorporation doping in significantly impacts electronic structure, intricately influencing . complex interplay between structural composition and material performance presents formidable challenge superconductor design. Based on novel generative model, diffusion adaptive representation: three‐channel matrix, we have designed inverse design model called...
Understanding the fundamental relationship between structural information of electrocatalysts and their catalytic activities plays a key role in controlling many important electrochemical processes. Recently, single-atom catalysts (SACs) with so-called MN4 structure, consisting central transition metal quadruply bound to four pyridine nitrogen atoms all situated an extended carbon-based matrix, have attracted intensive scientific attention owing exceptional performance. In this work, we...
The figure of merit (zT) is a key parameter to measure the performance thermoelectric materials. At present, prediction zT values via machine leaning has emerged as promising method for exploring high-performance However, learning-based predictions still suffer from unsatisfactory accuracy, and this related size data set, hyperparameters models, quality data. In work, 5038 pieces materials were selected, several regression models generated predict values. This large set-driven light gradient...
Seawater electrolysis is a promising route for hydrogen production using inexhaustible seawater resource. However, the chloride ion (Cl−)-induced corrosion and limited catalytic activity pose challenges to development of non-noble...
While applying computer simulations to study semiflexible polymers, it is a primary task determine the persistence length that characterizes chain stiffness. One frequently asked question concerns relationship between and bending constant of applied potential. In this paper, theoretical lengths polymers with two different potentials were analyzed examined by using lattice Monte Carlo simulations. We found was consistent predictions only in bond fluctuation model cosine squared angle The...
Identifying new superconductors with high transition temperatures (Tc > 77 K) is a major goal in modern condensed matter physics. The inverse design of Tc relies heavily on an effective representation the superconductor hyperspace due to underlying complexity involving many-body physics, doping chemistry and materials, defect structures. In this study, we propose deep generative model that combines two widely used machine learning algorithms, namely, variational auto-encoder (VAE)...
Abstract Predicting the power conversion efficiency (PCE) using machine learning (ML) can effectively accelerate experimental process of perovskite solar cells (PSCs). In this study, a high‐quality dataset containing 2079 PSCs is established to predict PCE values an accurate ML model, achieving impressive coefficient determination ( R 2 ) value 0.76. 12 validation experiments with PSCs, average absolute error between observed and predicted only 1.6%. Leveraging recommended improvement...
The data-driven machine learning approach has greatly improved the predictive accuracy of T g and D max values. governing rules for GFA have been successfully established through feature significance analysis.
Doping plays a crucial role in determining the critical temperature (Tc) of superconductors, yet accurately predicting its effects remains significant challenge. Here, we introduce novel doping descriptor that captures complex influence dopants on superconductivity. By integrating with elemental and physical features within Mixture Experts (MoE) model, achieve remarkable R2 0.962 for Tc prediction, surpassing all published prediction models. Our approach successfully identifies optimal...
Abstract Systematically manipulating the shape, dimension, and surface structure of PdAu nanocrystals is an active subject because it offers a powerful means to regulate investigate their structure–activity relationship. Meanwhile, still urgent reduce use two‐dimensional precious‐metal‐based nanomaterials. This work demonstrates that with variety shapes/dimensions, including 1D anisotropic nanowires, 2D porous nanosheets, 3D penetrative nanoflowers, can be systematically synthesized by...
The isolated Pd single atoms anchored on graphene demonstrate a catalytic activity that is 21.3 times higher than of Pd/C in the RhB hydrogenation reaction.