High-throughput bonding analysis of magnetic materials Katharina Ueltzen 1 , Janine George 1,2 1 Federal Institute for Materials Research and Testing, Department Materials Chemistry, Unter den Eichen 87, 12205 Berlin, Germany. 2 Friedrich Schiller University Jena, Institute of Condensed Matter Theory and Solid-State Optics, Max-Wien-Platz 1, 07743 Jena, Germany. Environmental and availability issues of conventional ferromagnets have put great interest in rare-earth-free alternatives [1] . However, exploring the chemical whitespace experimentally as well as by quantum-chemical methods is highly time- and resource-consuming. Computational determination of the magnetic ground state often involves calculating all likely magnetic orderings, spending most of the computational resources on irrelevant excited magnetic states. An alternative magnetism theory was proposed [2] which relies on quantum- chemical bonding analysis and on the computation of crystal orbital Hamilton populations (COHP, bond-weighted density of states) [3,4] . It states that antibonding states at the Fermi level in the COHP of the non-spin-polarized calculation that are reduced upon spin polarization present the origin of (itinerant) ferromagnetism [2] . We revisit this theory and present an approach to predict ferromagnetic ground states solely from the hypothetical, non- spin-polarized model. This is accomplished by using a recently developed bonding analysis workflow [5] on the 3 d metals and 150 transition-metal oxides of a computational magnetism database [6] . Multiple approaches for the numerical implementation of the qualitative criterion are developed and evaluated. The proposed criterion clearly distinguishes non-ferromagnetic from ferromagnetic 3 d metals. For the oxides, significant deviations from the underlying computational database were observed. Evaluation of a control group of materials with experimentally observed magnetic ground states showed that this was due to prediction errors of the database, while our approach showed very good accordance with experimental values with an accuracy of 0.85. The criterion based on COHPs holds great potential as an input feature in machine learning studies of new ferromagnets. References 1. Z. Shao, S. Ren, Nanoscale Adv. 2020 , 2 (10), 4341-4349. 2. G. A. Landrum, R. Dronskowski, Angew. Chem. Int. Ed. 2000 , 39 (9), 1560–1585. 3. R. Dronskowski, P. E. Bloechl , J. Phys. Chem. 1993 , 97 (33), 8617–8624. 4. R. Nelson, C. Ertural, J. George, V. L. Deringer, G. Hautier, R. Dronskowski, J. Comput. Chem. 2020 , 41 , 1931–1940. 5. J. George, G. Petretto, A. Naik, M. Esters, A. J. Jackson, R. Nelson, R. Dronskowski, G. Rignanese, G. Hautier, ChemPlusChem 2022 , 87 (11). 6. N. C. Frey, M. K. Horton, J. M. Munro, S. M. Griffin, K. A. Persson, V. B. Shenoy, Sci. Adv. 2020 , 6 (50).
P53DD
Made with FlippingBook Learn more on our blog