Materials Informatics Screening of Li‐Rich Layered Oxide Cathode Materials with Enhanced Characteristics Using Synthesis Data

Abstract

Lithium-ion batteries (LIBs) are the objects of active research and attract interest as important elements of near-term energy storage technologies. The ever-growing requirements for cathode materials of next-generation LIBs impel the need to screen the materials with high energy and power densities, cycling stability, rate capability, safety and compatibility with other battery elements. This study is focused on Li-rich layered oxide cathode materials as the materials candidates that are characterized by high energy density and large capacity and are therefore attractive as the potential solution for next-generation LIBs whereas moderate structural stability hinders their com-mercialization. Machine learning-assisted analysis of the collected experimental data has been performed for assessing the contribution of lattice doping, composition of compounds, the method and details of synthesis in the electrochemical characteristics. The possibility to relate parameters with the formation of a certain structure type (composite or solid solution), conceivable phase transformations and space charge layer formation are discussed. From this analysis, the features most probably amenable for electrochemical characteristics enhancement are distinguished and the focus area for further research is defined.[a] Dr.

Synthesis–structure relationships in Li- and Mn-rich layered oxides: phase evolution, superstructure ordering and stacking faults

“…[1][2][3] The latter two compounds have been subject to a variety of efforts aimed at optimising their electrochemical performance as a step towards commercialization. [4][5][6][7] Despite the research attention, the crystallographic description of these materials suffers from a distinct lack of consensus, [8][9][10] possibly due to a high susceptibility to synthetic variations. [11][12][13][14][15] Previously, the interdependencies between the synthesis, composition and structure of (non-Li-rich) Li transition metal (TM) oxides have been explored through combinatorial studies.…”

Artificial Intelligence Applied to Battery Research: Hype or Reality?

“…In addition, the authors proposed optimal experimental parameters by conducting inverse design, which they successfully validated with additional experiments. In a more recent study, Kireeva and Pervov also considered experimental data sets to identify synthesis and electrochemical property relationships in Li-rich layered oxide cathodes using a SVM model . In this case, input variables included composition, synthesis method, Li and transition metal sources, Li excess, temperature, and time of calcination and sintering; whereas initial discharge capacity and Coulombic efficiency were set as output variables.…”

Research on Recommendation of College Mental Health Teaching Materials Based on Improved Deep Learning Algorithm

“…The stack noise reduction selfcoder further calculates the collected teaching data affecting college mental health. In the learning process, further extract the sample features, image features, or text data features of the data to be learned [19], so as to further learn the received abstract sample data, and apply the learning results to the nearest neighbor algorithm for further prediction and scoring [20,21]. Through this algorithm, we can actively search and find the information with high user interest from the massive database information and recommend the information to the users.…”