S. Nakagawa1,2, S. Ishida1, J. Kato1,3, T. Nishio3, H. Nakao4, T. Kashiwagi2, H Eisaki1

1Research Institute for Advanced Electronics and Photonics, National Institute of Advanced Industrial Science and Technology (AIST), 1-1-1 Umezono, Tsukuba, Ibaraki 305-8568, Japan

2Graduate School of Pure & Applied Sciences, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 305-8573, Japan

3Department of Physics, Tokyo University of Science, 1-3 Kagurazaka, Shinjuku,

Tokyo 162-8601, Japan

4Photon Factory, Institute of Materials Structure Science, High Energy Accelerator Research

Organization (KEK), Tsukuba, Ibaraki 305-0801, Japan

High Tc cuprate superconductors exhibit a complex electronic phase diagram in which various electronic phases coexist or compete depending on the carrier numbers within the CuO2 planes [1]. In Bi2Sr2CaCu2O8+δ (Bi2212), the carrier numbers are thought to be controlled by either reducing (oxidizing) the as-grown samples, which make the samples under- (over-) doped. On the other hand, the change in the oxygen content (d) should also lead to a change in the crystal structure and an introduction of chemical disorder, but these effects have not been seriously examined. In addition, existing Bi2212 samples usually contain cation non-stoichiometry (e.g., mixing of Bi and Sr) and/or chemical substitutions (e.g., substitution of Pb with Bi) [2]. It is likely that this chemical inhomogeneity also affects the electronic phase diagram of real materials.

In this study, we synthesized Bi2212 single crystals with various cation compositions and studied how Tc and the crystal structures change with changing d in detail. It was found that the δ-dependence of the crystal structure is qualitatively different between the over- and under-doped regions at a certain doping level (pkink). The boundary pkink depends on the cation composition and corresponds to δ required for the formation of the superstructure in the BiO layers (~5 a0 for Pb-free: d=1/5=0.2, ~8a0 for Pb-doped: d=1/8=0.12). This result suggests that the location of the excess/defect oxygens are different between under-doped and over-doped region, which is consistent with the STM results [3] which suggest that the apical oxygen defects in under-doped samples. 

1. [1] B. Keimer et al., Nature 518, 179 (2015).
2. [2] H. Eisaki et al., Phys. Rev. B 69, 064512 (2004).
3. [3] I. Zeljkovic et al., Science 337, 320 (2012).