In this paper, we report on our group’s efforts to improve the coercivity of Sm₂Fe₁₇N₃ powder, especially on reducing the particle size to submicron scale, smoothing the particle surface, and suppression of the formation of coarse particles by developing a new reduction-diffusion process. During the course of a series of these works, it was revealed that the washing step, which is performed to remove excess Ca, supplied hydrogen into the Sm₂Fe₁₇N₃ crystal structure, and induced unfavorable elongation of the crystal structure along the c-axis. To avoid this problem, the powders were subjected to dehydrogenation treatment, demonstrating reasonably high coercivity values that we expect from the known relationship between particle size and coercivity. It was also found that the conventional dissolution and the removal of impurities by acetic acid were roughening the particle surfaces. Thus, development of an alternative process to acetic acid cleaning prevented the surface roughening and showed the further improvement of the coercivity. Finally, the development of a new uniform reduction-diffusion reaction using a rotary furnace brought about a breakthrough for further improvement of coercivity by suppressing the formation of coarse particles. As a result, we succeeded in synthesizing Sm₂Fe₁₇N₃ anisotropic powder with an ultra-high coercivity (i.e. the current world record) of 31.7 kOe, and also showed that the powder can maintain a coercivity higher than 10 kOe at 200°C.

We developed a neural network-based model to predict the Born effective charges from atomic structures. By combining forces due to an applied electric field, expressed as a product of the Born effective charge and the electric field, and forces evaluated by a neural network potential (NNP), a simulation scheme of ion dynamics under an electric field was proposed. Taking Li₃PO₄ as a prototype, we demonstrated the validity of our computation scheme. Using the constructed model of the Born effective charge predictor and NNP based on density functional (perturbation) theory calculation data, molecular dynamics (MD) simulations under a uniform electric field of 0.1 V/Å were performed. We obtained an enhanced mean square displacement of Li along the electric field, which seems physically reasonable. In addition, we found that the external forces along the direction perpendicular to the electric field, which originated from the off-diagonal components of the Born effective charges, had a non-negligible effect on the Li motion. Furthermore, we observed a more susceptive response of Li to the electric field in an amorphous structure.

Usually, expected values for various physical quantities, such as the number of electrons occupying certain states or the Coulomb interaction between different states of electrons, can be expressed in terms of integrals. In contrast, our method, based on differential forms, shows that expected values can be obtained by averaging over time. To confirm the validity of our method, we prepare the two cases: one is a very simple case with no many-body interaction, and the other is the case where the many-body term is included (the simplest Anderson Hamiltonian). Regarding the simple case without inclusion of many-body term, we prove analytically that the number of electrons occupying any state derived from our method is equivalent to the analytical one evaluated from the Greenʼs function method. When the many-body term is included, our results show good numerical agreement with the analytical ones derived from the Greenʼs function method. By the two cases, the calculation of expected values based on our method is considered valid.