The XXIII International Scientific Conference of Young Scientists and Specialists (AYSS-2019)

Computer modeling of disordered crystals

Not scheduled

1h

LIT, JINR

Oral

Speaker

Dr Roman Eremin (Samara Center for Theoretical Materials Science, Samara University/Samara State Technical University)

Description

For several reasons, the title families of compounds are of increasing interest for theoretical investigation of composition-structure-properties relations. For instance, their wide range practical applications and thermodynamic stability are closely related to possible defects in the structure such as doping, vacancies, and atomic site substitutions. Recently, we have developed several approaches intended to perform comprehensive modeling of the properties for cathode materials [1–3], solid electrolytes [4,5], and intermetallics containing Mackay clusters[6]. Within the current research, the machine learning approaches are discussed as a tool for comprehensive description, better interpretation, and further extension of the results obtained by means of density functional theory (DFT) modeling numerous (up to several thousand) entries of the compositional/configurational spaces (CCS) of the studied materials. For each group of substances, sets of the relevant and robust structural-topological descriptors allowing to set/train computational models and, subsequently, to reduce the complexity of the CCS to be modeled within DFT approaches are considered in connection with the experimental results. The authors thank hardware and software facilities of the ‘Zeolite’ supercomputer at the Samara Center for Theoretical Materials Science and the ‘HybriLIT’ heterogeneous platform at the Laboratory of Information Technologies of the Joint Institute for Nuclear Researches (Dubna, Russia). The study was partially funded by Ministry of Science and Higher Education of the Russian Federation (project No.3.6588.2017/9.10) and Russian Foundation for Basic Researches (projects No.18-33-00477 and 18-03-00443). References: [1] R.A. Eremin, P.N. Zolotarev, O.Y. Ivanshina, I.A. Bobrikov, Li(Ni,Co,Al)O2 Cathode Delithiation: A Combination of Topological Analysis, Density Functional Theory, Neutron Diffraction, and Machine Learning Techniques, J. Phys. Chem. C. 121 (2017) 28293−28305. doi:10.1021/acs.jpcc.7b09760. [2] R. Eremin, P. Zolotarev, I. Bobrikov, Delithiated states of layered cathode materials : doping and dispersion interaction effects on the structure, EPJ Web Conf. 177 (2018) 02001. [3] P. Zolotarev, R. Eremin, Comparative analysis of DFT-vdW vs . Coulomb energies for configurational space of layered cathode material at different delithiation levels, EPJ Web Conf. 201 (2019) 02004. [4] R.A. Eremin, N.A. Kabanova, Y.A. Morkhova, A.A. Golov, V.A. Blatov, High-throughput search for potential potassium ion conductors: A combination of geometrical-topological and density functional theory approaches, Solid State Ionics. 326 (2018) 188–199. doi:10.1016/j.ssi.2018.10.009. [5] P. Zolotarev, N. Nekrasova, A. Golov, R. Eremin, A combined DFT / topological analysis approach for modeling disordered solid electrolytes, EPJ Web Conf. 201 (2019) 02005. [6] T.G. Akhmetshina, P. Solokha, R.A. Eremin, V.A. Blatov, D.M. Proserpio, S. De Negri, Sacc, Nanocluster model and its application for crystal structure prediction of complex intermetallics, in: 31st Eur. Crystallogr. Meet. Oviedo, Spain 22-27 August, 2018: p. 57.