首页|Investigators from McGill University Have Reported New Data on Machine Learning (Machine Learning for High-throughput Configuration Sampling of Li-la-ti-o Disor dered Solid-state Electrolyte)
Investigators from McGill University Have Reported New Data on Machine Learning (Machine Learning for High-throughput Configuration Sampling of Li-la-ti-o Disor dered Solid-state Electrolyte)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – New research on Machine Learning is th e subject of a report. According to news reporting originating from Montreal, Ca nada, by NewsRx correspondents, research stated, “Most solidstate lithium elect rolytes are disordered ionic crystalline materials that possess crystallographic sites that can be vacant or occupied by different ions. The presence of these p artially occupied sites enables lithium diffusion through their lattice and make s such materials promising for developing all-solid batteries.” Financial supporters for this research include Natural Sciences and Engineering Research Council of Canada (NSERC), Triagency Institutional Programs Secretariat through New Frontiers in Research Fund. Our news editors obtained a quote from the research from McGill University, “Hig h-throughput computational screening of such materials must bypass costly DFT sa mpling of disordered configurations and, therefore, commonly relies on the compu tationally efficient Coulomb approximation to find just a few representative low -energy ionic configurations, for which DFT is then used to quickly predict a nu mber of important materials’ properties, such as the electrochemical stability w indow. This work demonstrates, using the Li-La-Ti-O solid electrolyte (LLTO) as an example, that the Coulomb approximation fails to correctly detect the most st able arrangement of Li and La ions in the LLTO, which has a noticeable impact on the accuracy of subsequent computational prediction of the electrochemical stab ility window of the material. The analysis herein shows that the sampling proble m arises from the relatively modest geometry relaxation of the LLTO lattice. A k ernel ridge regression machine learning (ML) method employing the smooth overlap of atomic positions as a structure descriptor (SOAP-KRR) leads to significant i mprovements in detecting the most stable configurations of the LLTO. The univers al ML potential based on the multiple atomic cluster expansion is also found to be reliable but to a lesser extent than SOAP-KRR.”
MontrealCanadaNorth and Central Amer icaChemicalsCyborgsElectrochemicalsElectrolytesEmerging TechnologiesInorganic ChemicalsMachine LearningMcGill University
NETL
NSTL
