Understanding the reentrant superconducting phase diagram of the iron pnictide Ca4Al2O6Fe2(As1−xPx)2: First-principles calculations

Abstract

Recently, a very rich phase diagram has been obtained for an iron-based superconductor Ca4Al2O6Fe2(As1−xPx)2. It has been revealed that nodeless (x∼0) and nodal (x=1) superconductivity are separated by an antiferromagnetic phase. Here we study the origin of this peculiar phase diagram using a five orbital model constructed from first-principles band calculation, and applying the fluctuation exchange approximation assuming spin-fluctuation-mediated pairing. At x=1, there are three hole Fermi surfaces, but the most inner one around the wave vector (0,0) has strong dX2−Y2 orbital character, unlike in LaFeAsO, where the most inner Fermi surface has dXZ/YZ character. Since the Fermi surfaces around (0,0), (π,0), and (π,π) all have dX2−Y2 orbital character, the repulsive pairing interaction mediated by the spin fluctuations gives rise to a frustration in momentum space, thereby degrading superconductivity despite the bond angle being close to the regular tetrahedron angle. As x decreases and the bond angle is reduced, the inner hole Fermi surface disappears, but the frustration effect still remains because the top of the band with dX2−Y2 character lies close to the Fermi level. On the other hand, the loss of the Fermi surface itself gives rise to a very good nesting of the Fermi surface because the number of electron and hole Fermi surfaces are now the same. The pairing interaction frustration and the good nesting combined favors antiferromagnetism over superconductivity. Finally for x close to 0, the band sinks far below the Fermi level, reducing the frustration effect, so that superconductivity is enhanced. There, the Fermi surface nesting is also lost to some extent, once again favoring superconductivity over antiferromagnetism. To see whether the present theoretical scenario is consistent with the actual nature of the competition between superconductivity and antiferromagnetism, we also perform hydrostatic pressure experiment for Ca4Al2O6Fe2(As1−xPx)2. In the intermediate x regime where antiferromagnetism occurs at ambient pressure, applying hydrostatic pressure smears out the antiferromagnetic transition, but superconductivity does not take place. This supports our scenario that superconductivity is suppressed by the momentum space frustration in the intermediate x regime, apart from the presence of the antiferromangnetism.

Source-reported materials — not catalogue approval

Similar papers