Different response of the crystal structure to isoelectronic doping in BaFe2[As(1-x)P(x)]2 and [Ba(1-x)Sr(x)]Fe2As2

Marianne Rotter,Christine Hieke,Dirk Johrendt
DOI: https://doi.org/10.48550/arXiv.1005.1411
2010-05-09
Superconductivity
Abstract:Superconductivity up to 30 K in charge neutrally doped BaFe2[As(1-x)P(x)]2 has been ascribed to chemical pressure, caused by the shrinking unit cell. But the latter induces no superconductivity in [Ba(1-x)Sr(x)]Fe2As2 in spite of the same volume range. We show that the spin-density-wave (SDW) state of BaFe2As2 becomes suppressed in BaFe2[As(1-x)P(x)]2 by a subtle reorganization of the crystal structure, where arsenic and phosphorus are located at different coordinates z(As), z(P). High-resolution X-ray diffraction experiments with BaFe2[As(1-x)P(x)]2 single crystals reveal almost unchanged Fe-P bonds, but a contraction of the Fe-As bonds, which remain nearly unchanged in [Ba(1-x)Sr(x)]Fe2As2. Since the Fe-As bond length is a gauge for the magnetic moment, our results show why the SDW is suppressed by P-, but not by Sr-doping. Only the Fe-P interaction increases the width of the iron 3d bands, which destabilizes the magnetic SDW ground state. The simultaneous contraction of the Fe{As bonds is rather a consequence of the vanishing magnetism. Ordered structure models of BaFe2[As(1-x)P(x)]2 obtained by DFT calculations agree perfectly with the single-crystal X-ray structure determinations. The contraction of the Fe-As bonds saturates at doping levels above x = 0.3, which corrects the unreasonable linear decrease of the so-called pnictide height.
What problem does this paper attempt to address?