We have performed neutron diffraction measurement on a single crystal of parent compound of iron-based superconductor, BaFe$_2$As$_2$ at 12~K. In order to investigate in-plane anisotropy of magnetic form factor in the antiferromagnetic phase, the detwinned single crystal is used in the measurement. The magnetic structure factor and magnetic form factor are well explained by the spin densities consisting of $3d_{yz}$ electrons with a fraction of about 40~% and the electrons in the other four $3d$ orbitals with each fraction of about 15~%. Such anisotropic magnetic form factor is qualitatively consistent with the anisotropic magnetic behaviors observed in the antiferromagnetic phase of the parent compound of iron-based superconductor.
Understanding magnetic interactions in the parent compounds of high-temperature superconductors forms the basis for determining their role for the mechanism of superconductivity. For parent compounds of iron pnictide superconductors such as $A$Fe$_2$As$_2$ ($A=$ Ba, Ca, Sr), although spin excitations have been mapped out throughout the entire Brillouin zone (BZ), measurements were carried out on twinned samples and did not allow for a conclusive determination of the spin dynamics. Here we use inelastic neutron scattering to completely map out spin excitations of $sim$100% detwinned BaFe$_2$As$_2$. By comparing observed spectra with theoretical calculations, we conclude that the spin excitations can be well described by an itinerant model with important contributions from electronic correlations.
Neutron diffraction measurements have been carried out to investigate the magnetic form factor of the parent SrFe2As2 system of the iron-based superconductors. The general feature is that the form factor is approximately isotropic in wave vector, indicating that multiple d-orbitals of the iron atoms are occupied as expected based on band theory. Inversion of the diffraction data suggests that there is some elongation of the spin density toward the As atoms. We have also extended the diffraction measurements to investigate a possible jump in the c-axis lattice parameter at the structural phase transition, but find no detectable change within the experimental uncertainties.
We show that the Fermi surface (FS) in the antiferromagnetic phase of BaFe$_2$As$_2$ is composed of one hole and two electron pockets, all of which are three dimensional and closed, in sharp contrast to the FS observed by angle-resolved photoemission spectroscopy. Considerations on the carrier compensation and Sommerfeld coefficient rule out existence of unobserved FS pockets of significant sizes. A standard band structure calculation reasonably accounts for the observed FS, despite the overestimated ordered moment. The mass enhancement, the ratio of the effective mass to the band mass, is 2--3.
Ternary Ba-Fe-As system has been studied to determine a primary solidification field of the BaFe$_2$As$_2$ phase. We found that the BaFe$_2$As$_2$ phase most likely melts congruently and primarily solidifies either in the FeAs excess or Ba$_{x}$As$_{100-x}$ excess liquid. Knowing the primary solidification field, we have performed the vertical Bridgman growth using the starting liquid composition of Ba$_{15}$Fe$_{42.5}$As$_{42.5}$. Large single crystals of the typical size 10x4x2 mm$^3$ were obtained and their quality was confirmed by X-ray Laue and neutron diffraction.
We performed angle resolved photoelectron spectroscopy (ARPES) studies on mechanically detwinned BaFe2As2. We observe clear band dispersions and the shapes and characters of the Fermi surfaces are identified. Shapes of the two hole pockets around the {Gamma}-point are found to be consistent with the Fermi surface topology predicted in the orbital ordered states. Dirac-cone like band dispersions near the {Gamma}-point are clearly identified as theoretically predicted. At the X-point, split bands remain intact in spite of detwinning, barring twinning origin of the bands. The observed band dispersions are compared with calculated band structures. With a magnetic moment of 0.2 ?B per iron atom, there is a good agreement between the calculation and experiment.