Prediction of ABX compounds (X — As, Sn, Sb, Pb or Bi) with the MgAgAs structure type
and their crystal lattice parameters
N. N. Kiselyova, V. A. Dudarev, A. V. Stolyarenko, O. V. Senko,
A. A. Dokukin, Yu. O. Kuznetsova
Using machine learning programs, the prediction of not yet obtained 250 compounds of composition ABX (where A and B are different chemical elements, and X are As, Sn, Sb, Pb or Bi) with a crystal structure of the MgAgAs type was carried out and the values of their crystal lattice parameter were estimated. Using the cross-validation method, the best machine learning algorithms were selected for subsequent predicting. When predicting compounds that have not yet been synthesized, the most accurate programs were based on neural network training algorithms, support vector machines and k-nearest neighbors, for which the accuracy was determined to be 88.5, 91.0 and 88.4 %, respectively. When predicting the value of the crystal lattice parameter of the predicted compounds, the best results were obtained using programs based on the Bayesian Ridge methods (coefficient of determination R2 = 0.959, mean absolute error MAE = 0.0370, mean square error MSE = 0.0030), ARD Regression (R2 = 0.950, MAE = 0.0401, MSE = 0.0036) and Ridge (R2 = 0.959, MAE = 0.0368, MSE = 0.0029), i.e. the deviation of the calculated values from the experimental ones was in the range of 0.0368 – 0.0401 Å. When predicting new compounds and estimating their crystal lattice parameters, only the values of the properties of the chemical elements included in their composition were used.
Keywords: MgAgAs, lattice parameter, prediction, machine learning.
DOI: 10.30791/1028-978X-2025-3-5-14
Kiselyova Nadezhda — Federal State Institution of Science A.A. Baikov Institute of Metallurgy and Materials Sciences RAS (119334 Moscow, Russia, Leninskii Prospect, 49), Dr Sci (Chem), chief researcher, specialist in the application of information tech-nologies (IT) to chemistry and materials science. E-mail: kis@imet.ac.ru.
Dudarev Victor — Ruhr-Universität Bochum (Universitätsstraβe 150; 44801 Bochum), PhD (Eng), researcher, IT specialist. E-mail: vic-dudarev@mail.ru.
Stolyarenko Andrey — Federal State Institution of Science A.A. Baikov Institute of Metallurgy and Materials Sciences RAS (119334 Moscow, Russia, Leninskii Prospect, 49), PhD (Eng), researcher, IT specialist. E-mail: stol-drew@yandex.ru.
Senko Oleg — Federal Research Center “Computer Science and Control” RAS (119333 Moscow, Russia, ul.Vavilova, 40), Dr Sci (Phys-Math), professor, leading researcher, machine learning specialist. E-mail: senkoov@mail.ru.
Dokukin Aleksandr — Federal Research Center “Computer Science and Control” RAS (119333 Moscow, Russia, ul.Vavilova, 40), PhD (Phys-Math), senior re-searcher; Federal State Institution of Science A.A. Baikov Institute of Metallurgy and Materials Sciences RAS (119334 Moscow, Russia, Leninskii Prospect, 49), senior researcher, machine learning specialist. E-mail: dalex@ccas.ru.
Kuznetsova Yuliana — Federal State Institution of Science A.A. Baikov Institute of Metallurgy and Materials Sciences RAS (119334 Moscow, Russia, Leninskii Prospect, 49), engineer-researcher, IT specialist. E-mail: jul98@yandex.ru
Reference citing:
Kiselyova N.N., Dudarev V.A., Stolyarenko A.V., Senko O.V., Dokukin A.A., Kuznetsova Yu.O Prognozirovanie soedinenij sostava ABX (X — As, Sn, Sb, Pb ili Bi) so strukturoj tipa MgAgAs i parametrov ih kristallicheskoj reshetki [Prediction of ABX compounds (X — As, Sn, Sb, Pb or Bi) with the MgAgAs structure type and their crystal lattice parameters]. Perspektivnye Materialy [Advanced Materials] (in Russ), 2025, no. 3, pp. 5 – 14. DOI: 10.30791/1028-978X-2025-3-5-14