View the related article [J. Phys. Soc. Jpn. 77 (2008) 053709]

High-Temperature Superconductivity in Another Beautiful Crystal Structure [May 12, 2008]
Hidetoshi Fukuyama
(Faculty of Science and Research Institute of Science and Technology, Tokyo University of Science)

  Discoveries of new superconductors, especially those with high critical temperatures, have always surprised people and have disclosed the fact that crystalline materials have infinite possibility with combinations of atomic elements (or molecular elements) and their spatial arrangement in the unit cell. The newly found Fe oxypnictides La(O,F)FeAs [1], whose crystal structure is shown in Fig. 1, is no exception. This family has a layered structure and doped carriers (electrons introduced by the substitution of O with F) move in layers comprising strongly bonded Fe and As atoms, i.e., the dopant and the conducting layers are separated as in cuprates. The fact that superconductivity is realized in materials containing the atoms of a typical (actually the most familiar!) magnetic element, Fe, is itself a great surprise, but the realization of the highest critical temperature next only to cuprates in Fe oxypnictides (around 30 K in the original report [1] and above 50 K in Sm(O,F)FeAs [2] and oxygen-deficient ReO1-xFeAs [3,4] [Re = rare earth element]) is truly astonishing. The question that naturally arises is, “What will be the electronic states that support such superconductivity?”



Fig. 1: Crystal structure of LaOFeAs. (Courtesy: H. Hosono)

  The paper of Ishibashi et al. [5] addresses this question by very carefully studying the band structures and focusing on the parent (undoped) compound LaOFeAs. Understanding the electronic states of the parent compounds at this stage of exploration is very important, and it provides a firm foundation to identify future research targets, as was evidenced in the case of cuprates. In the case of cuprates it turned out that the parent compounds were Mott insulators, which could not be described by band calculations and whose carriers are doped mainly at the oxygen (not copper) sites. The study of Ishibashi et al. [5] has disclosed the fact that all the states near the Fermi energy are basically due to the d states of Fe and that the ground state is antiferromagnetic and of stripe type with relatively large magnetic moments (around 2 Bohr magnetons). (A similar magnetic ordering has also been reported [6]). This theoretical finding is consistent with the results of neutron scattering [7,8] for the spatial pattern at least in the layer (the agreement between theory and experiment regarding the possible type of ordering in the stacking direction is not clear yet) and the magnitude of magnetic moment [8], though the estimated values of the magnetic moments are somewhat different from each other; this difference in the values may possibly be due to the difficulty in controlling the carrier concentration in the experiment. When closely examined, the band dispersion near the Fermi energy, as observed in Fig. 2 (taken from Fig. 7 of Ishibashi et al. [5]), indicates the very intriguing feature that the system is semimetallic and the crossing of energy bands occurs in the energy region very close to the Fermi energy; the latter appears to be similar to that in molecular solids, αET2I3 [9,10], and shows strong enhancement of mobility toward low temperatures, as observed in ref. 7. The fact that the magnitude of the magnetic moment is not small may indicate that electrons in the parent compound are also strongly correlated and may not be considered as spin density wave (SDW) state due to nesting [6], as stressed in ref. 5. Understanding the effects of doping and the reasons for the difference between LaOFeP [11] (Tc being 4 K even in the nondoped state and insensitive to carrier doping) and LaOFeAs both experimentally and theoretically to explore the possibility of even higher critical temperatures (closer to room temperature!) in this family of materials is a scientific challenge.



Fig. 2: Band dispersion near the Fermi energy in the antiferromagnetic stripe state. (After ref. 5)
  Very stimulating discussions with the authors of ref. 5 are gratefully acknowledged.

References
[1] Y. Kamihara et al.: J. Am. Chem. Soc. 130 (2008) 3296.
[2] Z.-A. Ren et al.: condmat: 0804.2053.
[3] Z.-A. Ren et al.: condmat: 0804.2582.
[4] H. Kito et al.: to be published in J. Phys. Soc. Jpn. 77 (2008) No.6.
[5] S. Ishibashi et al.: J. Phys. Soc. Jpn. 77 (2008) 053709.
[6] J. Dong et al.: condmat: 0803.3426.
[7] M. A. McGuire et al.: condmat: 0804.0796.
[8] C. de la Cruz et al.: condmat: 0804.0795.
[9] N. Tajima et al.: J. Phys. Soc. Jpn. 75 (2006) 051010.
[10] S. Katayama et al.: J. Phys. Soc. Jpn. 75 (2006) 054705.
[11] Y. Kamihara et al.: J. Am. Chem. Soc. 128 (2006) 10012.


The above article should be referred as “H. Fukuyama: JPSJ Online-News and Comments [May 12, 2008]” when citing.

Copyright © the Physical Society of Japan.