Tetragonal SnOFeSe: A possible parent compound of the FeSe-based superconductor

Recent experiments have reported that inserting metal atoms or small molecules in between the FeSe layers of $\beta$-FeSe can significantly enhance the superconducting transition temperature. Here, based on first-principles electronic structure calculations, we propose a stable compound SnOFeSe by alternatively stacking the SnO and $\beta$-FeSe layers. The predicted SnOFeSe has the same tetragonal structure as the well-known FeAs-based compound LaOFeAs, meanwhile their electronic structures in the nonmagnetic state are quite similar. The magnetic ground state of SnOFeSe is predicted to be the dimer antiferromagnetic (AFM) state, which is energetically only 2.77 (2.15) meV/Fe lower than the trimer (dimer-trimer-dimer-trimer) AFM state, indicating that strong magnetic fluctuations might be induced via slight modulation. Interestingly, SnOFeSe is at the verge of metal-insulator transition in these low-energy magnetic states, hence bridging the metallic parent compounds of iron-based superconductors and the insulating ones of cuprate superconductors. With the reduced dimensionality, monolayer SnOFeSe also shows great similarities in the electronic and magnetic properties to its bulk phase. Given that SnOFeSe is adjacent to magnetic frustration and resembles LaOFeAs in both crystal and electronic structures, we suggest that SnOFeSe is a possible superconductor parent compound, which may provide a promising platform to study the interplay between magnetism and unconventional superconductivity in FeSe-derived materials.

Figure 1

(a) Crystal structures of the tetragonal-phase SnO and $\beta$-FeSe. (b) Six typical stacking sites of the SnO layer on top of the FeSe layer, labeled by the red crosses, dot, and triangles. (c) Brillouin zone (BZ) and (d) crystal structure of bulk SnOFeSe. The high-symmetry $k$ points in the BZ are indicated by the red dots.

Figure 2

(a) Phonon dispersion of bulk SnOFeSe in the NM state along the high-symmetry paths of the BZ [Fig. 1]. (b) One-dimensional (1D) and (c) three-dimensional (3D) differential charge densities [$\mathrm{\Delta }\rho$ = ${\rho }_{\text{SnOFeSe}}-{\rho }_{\text{SnO}}-{\rho }_{\text{FeSe}}$]. The atomic positions in panel (b) are marked by the color bars on the right axis. The yellow and cyan isosurfaces in panel (c) represent the electron accumulation and depletion areas, respectively. The isosurface value is set to $6.14\times {10}^{-4}e/{\text{Å}}^3$.

Figure 3

(a) Band structure along the high-symmetry BZ paths, (b) PDOS, and (c) five Fermi surface sheets for bulk SnOFeSe in the NM state. The Fermi energy is set to zero.

Figure 4

(a) Real and (b) imaginary parts of the electron susceptibility $\chi$ plotted in the $q_{\text{z}}$ = 0 plane for bulk SnOFeSe in the NM state.

Figure 5

Sketches of six typical spin configurations for the Fe lattice in SnOFeSe: (a) FM state, (b) checkerboard AFM $\mathrm{N}\acute{\text{e}}\mathrm{el}$ state, (c) single-stripe (collinear) AFM state, (d) dimer AFM state, (e) trimer AFM state, and (f) tetramer AFM state. Here, the solid rectangles represent the supercells, while the purple and yellow balls represent the spin-up and spin-down Fe atoms, respectively.

Figure 6

Total DOS and PDOS for (a) the dimer AFM and (b) the trimer AFM states of bulk SnOFeSe. The Fermi energy is set to zero.

Figure 7

(a) Band structure along the high-symmetry paths of the BZ and (b) PDOS for monolayer SnOFeSe in the NM state. The Fermi energy is set to zero.