We report X-ray magnetic circular dichroism (XMCD) experiments in an antiferromagnetic metal Mn3Sn. This material has been recently attracting much attention due to the large anomalous Hall, Nernst, and magnetooptical Kerr effects with vanishingly small net magnetization. For these responses, time reversal symmetry (TRS) breaking is necessary, but the mechanism of TRS breaking was not well understood since the magnetization of Mn3Sn is much smaller than that of ferromagnets. In this study, we have applied XMCD to Mn3Sn to elucidate the microscopic origin of TRS breaking. We found that the anisotropic-magnetic-dipole contribution called Tz term, which is mostly negligible in conventional ferromagnets, is a main origin of observed XMCD signal.
Reflection high-energy electron diffraction (RHEED) has the advantage of being able to observe a wide reciprocal space due to the high-energy electrons. We have developed Weissenberg RHEED method, which is based on the principle of the Weissenberg camera used in X-ray diffraction and can measure a wide range of three-dimensional reciprocal space while rotating the sample. In general, a structural analysis is conducted by using a dynamical scattering theory for the electron diffraction due to the strong effects of multiple scattering. Therefore, the analysis is performed based on the comparison between experimental and simulation results, but it is difficult to apply this method to completely unknown structures. We have attempted to utilize the kinematic analysis by using a huge data set, taking advantage of the ability to measure a large amount of diffraction data in a short time. Structural parameters can be directly obtained by Fourier transform-based analysis in the kinematical analysis. While the kinematical analysis of Weissenberg RHEED has been successfully applied to known surface structures, it has never been succeeded to unveil unknown structures. Recently, we performed structural analysis of the Si (110)3×2-Bi surface, which has been relatively well investigated on Si (110) substrates but the structure was unknown. The new structure, which contains four Bi atoms in the unit lattice and a complex reconstruction of the underlying Si substrate, was obtained by solving Patterson function. The positions of 62 atoms from the surface to the fifth layer have been determined in three-dimensionally with an accuracy of 0.05 Å. The bond lengths and bond angles fall within reasonable ranges expected from the atomic radii and electron orbitals, and the surface is found to be a stable surface structure with complete termination with no dangling bonds. This result could not have been achieved by trial-and-error study of structural models and proves that the kinematic analysis of the W-RHEED is feasible.
Phonon calculations based on density functional theory have been actively used to predict and understand various phonon-related properties, including phase stability, carrier mobility, and superconducting transition temperature. However, the commonly employed harmonic approximation (HA) often fails in functional materials as it yields imaginary phonons. To overcome the limitation of the HA, ab initio self-consistent phonon method, as well as its extension, have recently been developed and employed successfully. In this short review, we show how the advent of these new techniques solves the long-standing issues inherent to the HA and opens up the way for exploring the finite-temperature properties of broader material space from first principles.