Synergistic computational and experimental discovery of novel magnetic materials

Abstract

We review recent results on discoveries of new magnetic compounds by combining experiments, adaptive genetic algorithm searches, and advanced electronic-structure computational methods.

3d transition metal substituted Fe3Se4: prediction of potential permanent magnet

“…Concerns have been raised regarding the availability of RE elements such as Nd, Sm, and Dy, as well as environmental implications and economical limitations associated with their extraction. Significant efforts have been made in recent years to address potential supply issues and particularly focus on entirely RE-free alternative magnets in order to meet the growing demand for PMs [2][3][4][5][6]. Although RE elements are necessary to obtain significant amount of K u , however for application purpose a large K u is anticipated without such constituent atoms.…”

Machine Learning in Magnetic Materials

“…[24] Later, the development in computational materials science and the application of ab initio theoretical calculations have contributed significantly to the overall scientific effort, in conjuction with experimental synthesis and characterization. [25][26][27][28][29] The application of advanced computational tools enhances the relevant research in the field by screening out phases with possibly unsuitable properties, allowing costly and time-consuming experimental research to focus on the most promising systems. In additon, the feedback of experimental results to theoretical research may lead to the improvement of the latter, by fine-tuning basic parameters and boundary conditions with limited-up to the present-knowledge.…”

“…Kovacs and co‐workers used ML to optimize the structure of hard magnetic materials with no rare‐earth content, such as Fe 3 Sn 0.75 Sb 0.25 , FeNi, and MnAl, to calculate the best possible coercivity that could be obtained with proper processing. [ 75 ] Balasubramanian et al also studied hard magnetic materials without expensive rare‐earth content but rather based on transition metals such as Fe and Co. [ 29 ] The feasibility of using materials genomics and ML is also underlined by J. M. D. Coey for the modification of already known rare‐earth intermetallic systems; in this area the experimental research for new phases and the optimization potential of existing ones is time and money intensive and computational methods may induce a large impact if proven succesful. [ 5 ] Exl et al have used an ML approach for a micromagnetic study of the microstructural dependence of coercivity in Nd 2 Fe 14 B.…”

Accelerating the discovery of novel magnetic materials using machine learning–guided adaptive feedback

“…In the past few decades, several efficient computational algorithms including an adaptive genetic algorithm (AGA) have been developed to predict stable crystal structures based on given chemical compositions ( 1 – 4 ). These algorithms and methods are very useful in guiding materials discovery ( 5 – 7 ). However, the number of possible combinations of different elements with different compositions, and the crystal structures that they may adopt, is enormous, especially for compounds with three or more chemical species.…”

“…The synthesis of bulk Fe-Co-N compounds remains challenging and has yet to be realized. Binary or ternary compounds formed by combining Fe, Co, or FeCo with transition metals such as Zr, Hf, Ti, and Nb also exhibit appreciable magnetocrystalline anisotropies, but their saturation magnetizations are significantly reduced compared to those of Fe and Co ( 5 , 7 ). For the Fe-Co-B system, no stable ternary structures with desired magnetic properties for permanent-magnet applications have been discovered.…”