Browsing by Author "Shi, SQ"

Now showing 1 - 6 of 6
  • Thumbnail Image
    Item
    A customized strategy to design intercalation-type Li-free cathodes for all-solid-state batteries
    (Oxford University Press, 2023-01-10) Wang, D; Yu, J; Yin, X; Shao, S; Li, Q; Wang, YC; Avdeev, M; Chen, LQ; Shi, SQ
    Pairing Li-free transition-metal-based cathodes (MX) with Li-metal anodes is an emerging trend to overcome the energy-density limitation of current rechargeable Li-ion technology. However, the development of practical Li-free MX cathodes is plagued by the existing notion of low voltage due to the long-term overlooked voltage-tuning/phase-stability competition. Here, we propose a p-type alloying strategy involving three voltage/phase-evolution stages, of which each of the varying trends are quantitated by two improved ligand-field descriptors to balance the above contradiction. Following this, an intercalation-type 2H-V1.75Cr0.25S4 cathode tuned from layered MX2 family is successfully designed, which possesses an energy density of 554.3 Wh kg−1 at the electrode level accompanied by interfacial compatibility with sulfide solid-state electrolyte. The proposal of this class of materials is expected to break free from scarce or high-cost transition-metal (e.g. Co and Ni) reliance in current commercial cathodes. Our experiments further confirm the voltage and energy-density gains of 2H-V1.75Cr0.25S4. This strategy is not limited to specific Li-free cathodes and offers a solution to achieve high voltage and phase stability simultaneously. TheAuthor(s) 2023. Published byOxfordUniversity Press on behalf of China Science Publishing&Media Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License
  • Thumbnail Image
    Item
    EMFDTW: an automated crystallographic identification tool supporting multiple comparison criteria
    (American Chemical Society, 2024-07-13) He, B; Meng, Y; Gong, Z; Wang, K; Jiang, Z; Avdeev, M; Shi, SQ
    Identification of the same and similar crystal structures assists in searching for duplicate materials data and discovering prototype structures. Although several structure identification methods exist, their requirements for the input information limit their ability to accurately and automatically process structures within big materials databases and especially distinguish disordered ion conductor structures due to the site occupancy uncertainty of migration ions. Here, we introduce an automated crystal structure identification method called EMFDTW, in which a set of eigen-subspace modular functions (EMFs) is derived from a distance matrix incorporating site type identifiers, and then the similarity between them is measured through dynamic time warping (DTW). In this way, not only the conventional spatial sites in the crystal structure but also the atomic attributes (type, occupancy, oxidation state, magnetic moment, etc.) on the sites can be considered as the comparative features. Furthermore, by conducting a skeleton similarity analysis on 113,586 crystal structures sourced from the crystallography open database and the inorganic crystal structure database, we establish a database of 17,340 skeleton prototypes, which paves the way for searching potential ionic conductors. Our work provides an easy-to-use tool to analyze complex crystal structures, providing new insights for the discovery and design of new materials. © 2024 American Chemical Society.
  • No Thumbnail Available
    Item
    Feature selection method reducing correlations among features by embedding domain knowledge
    (Elsevier, 2022-10-01) Liu, Y; Zou, X; Ma, S; Avdeev, M; Shi, SQ
    Selecting proper descriptors, also known as features, is one of the key problems in modeling for materials properties using machine learning models. Redundant features reduce accuracy of machine learning modeling, and results of purely data-driven feature selection methods are often inconsistent with materials domain knowledge. Herein, a feature selection method embedded with materials domain knowledge named NCOR-FS is proposed to select higher quality features. The method translates materials domain knowledge about highly correlated features into Non-Co-Occurrence Rules (NCORs), which allows to quantify the degree to which NCORs are violated by feature subsets and to design optimization process for FS method based on swarm intelligence algorithm. Experiments on seven datasets show that compared with multiple other FS methods commonly used in materials, NCOR-FS selects the feature subset with more appropriate number of highly correlated features, which improves the prediction accuracy and interpretability of the ML model. NCOR-FS can be applied to any materials systems, and the idea of embedding domain knowledge into data-driven algorithm is expected to facilitate constructing extensive machine learning models embedded with materials domain knowledge. © 2022 Acta Materialia Inc. Published by Elsevier Ltd.
  • No Thumbnail Available
    Item
    Generative artificial intelligence and its applications in materials science: current situation and future perspectives
    (Elsevier, 2023-07) Liu, Y; Yang, ZW; Yu, ZY; Liu, Z; Liu, D; Lin, H; Li, MQ; Ma, S; Avdeev, M; Shi, SQ
    Generative Artificial Intelligence (GAI) is attracting the increasing attention of materials community for its excellent capability of generating required contents. With the introduction of Prompt paradigm and reinforcement learning from human feedback (RLHF), GAI shifts from the task-specific to general pattern gradually, enabling to tackle multiple complicated tasks involved in resolving the structure-activity relationships. Here, we review the development status of GAI comprehensively and analyze pros and cons of various generative models in the view of methodology. The applications of task-specific generative models involving materials inverse design and data augmentation are also dissected. Taking ChatGPT as an example, we explore the potential applications of general GAI in generating multiple materials content, solving differential equation as well as querying materials FAQs. Furthermore, we summarize six challenges encountered for the use of GAI in materials science and provide the corresponding solutions. This work paves the way for providing effective and explainable materials data generation and analysis approaches to accelerate the materials research and development. © 2023 The Authors. Published by Elsevier B.V. on behalf of The Chinese Ceramic Society. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
  • Thumbnail Image
    Item
    Multi‐layer feature selection incorporating weighted score‐based expert knowledge toward modeling materials with targeted properties
    (Wiley, 2020-01-15) Liu, Y; Wu, JM; Avdeev, M; Shi, SQ
    Selecting proper descriptors or features is one of the central problems in exploring structure–activity relationships of materials using machine learning models. The current feature selection algorithms usually require tedious hyperparameter tuning and do not actively consider the prior knowledge of domain experts about the features. Here, this work proposes a data‐driven multi‐layer feature selection method incorporating domain expert knowledge named DML‐FSdek, which is automated, with users entering training data without manual tuning of the hyperparameters. The domain expert knowledge is quantified by means of weighted scoring and integrated into the selection process to eliminate the risk of crucial features being removed. The test studies on ten material properties datasets demonstrate the potential of the approach to automatically search for a reduced feature set with lower root mean square errors than those for the initial feature set. Essentially, the most relevant material features, the number of which is much smaller than that in the original feature set, are automatically selected to establish a closer and more accurate structure–activity relationship for the materials of interest. As a result, the method represents the targeted properties of materials with a smaller and more interpretable set of features while ensuring equal or better prediction accuracy. © 2020 The Authors. Published by WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim
  • No Thumbnail Available
    Item
    The origin of solvent deprotonation in LiI‐added aprotic electrolytes for Li‐O2 batteries
    (Wiley, 2023-03-27) Wang, AP; Wu, XH; Zou, Z; Qiao, Y; Wang, D; Xing, L; Chen, Y; Lin, Y; Avdeev, M; Shi, SQ
    LiI and LiBr have been employed as soluble redox mediators (RMs) in electrolytes to address the sluggish oxygen evolution reaction kinetics during charging in aprotic Li‐O2 batteries. Compared to LiBr, LiI exhibits a redox potential closer to the theoretical one of discharge products, indicating a higher energy efficiency. However, the reason for the occurrence of solvent deprotonation in LiI‐added electrolytes remains unclear. Here, by combining ab initio calculations and experimental validation, we find that it is the nucleophile that triggers the solvent deprotonation and LiOH formation via nucleophilic attack, rather than the increased solvent acidity or the elongated C−H bond as previously suggested. As a comparison, the formation of in LiBr‐added electrolytes is found to be thermodynamically unfavorable, explaining the absence of LiOH formation. These findings provide important insight into the solvent deprotonation and pave the way for the practical application of LiI RM in aprotic Li‐O2 batteries. © 1999-2024 John Wiley & Sons, Inc