Lattice dynamics of ultrathin FeSe films on ${\mathrm{SrTiO}}_3$

Charge transfer and electron-phonon coupling (EPC) are proposed to be two important constituents associated with enhanced superconductivity in the single unit cell FeSe films on oxide surfaces. Using high-resolution electron energy loss spectroscopy combined with first-principles calculations, we have explored the lattice dynamics of ultrathin FeSe films grown on ${\mathrm{SrTiO}}_3$. We show that, despite the significant effect from the substrate on the electronic structure and superconductivity of the system, the FeSe phonons in the films are unaffected. The energy dispersion and linewidth associated with the Fe- and Se-derived vibrational modes are thickness and temperature independent. Theoretical calculations indicate the crucial role of antiferromagnetic correlation in FeSe to reproduce the experimental phonon dispersion. Importantly, the only detectable change due to the growth of FeSe films is the broadening of the Fuchs-Kliewer (F-K) phonons associated with the lattice vibrations of ${\mathrm{SrTiO}}_3$(001) substrate. If EPC plays any role in the enhancement of film superconductivity, it must be the interfacial coupling between the electrons in FeSe film and the F-K phonons from substrate rather than the phonons of FeSe.

Figure 1

(a) STM topography of 1 uc-FeSe/STO sample ($400\times 400$ nm, 5.0 V/50 pA). Inset: Atomically resolved STM topography of 1 uc-FeSe/STO sample ($5\times 5$ nm, 0.4 V/100 pA). (b) LEED patterns of 1 uc-FeSe/STO sample taken at 35 K with the primary energy of 80 eV. Red dashed lines represent the first surface Brillouin zone. (c) ARPES spectrum of 1 uc-FeSe/STO, taken at 35 K along the $\overline{\mathrm{\Gamma }M}$ direction and centered at $\overline{M}$. The photon energy is 21.2 eV. The black line represents the band fitting by a tight binding model [5]. (d) Plot of the evolution of the ARPES symmetrized energy distribution curves (EDCs) near $k_F$ as a function of temperature, indicating the gap closes between 63 K and 73 K.

Figure 2

(a) Schematic structure of the FeSe film on STO substrate. (b) Illustration of ionic vibrations of F-K phonons in STO. (c) Illustration of ionic vibrations of phonons in FeSe at $\overline{\mathrm{\Gamma }}$ point. (d) Energy-momentum mapping of 2D HREELS measurements of 1 uc-FeSe/STO samples along $\overline{\mathrm{\Gamma }X}$ direction at 35 K, where red solid lines are guides to the eye. Orange stars label the $A_{1g}$ mode (22.6 meV) and $B_{1g}$ mode (25.6 meV) measured by Raman scattering at 7 K [43]. (e) Second derivative image of (d).

Figure 3

(a) Energy-momentum mapping (0–40 meV) of 2D HREELS measurements of 1uc-FeSe/STO samples along $\overline{\mathrm{\Gamma }X}$ direction at 35 K. (b) Second derivative image of (a). (c) Momentum-dependent EDCs of (a), where black lines are only guides. (d)–(f) Corresponding results along $\overline{\mathrm{\Gamma }M}$ direction.

Figure 4

Phonon dispersions of low-energy FeSe phonons of (a) 1 uc-FeSe/STO, (b) 3 uc-FeSe/STO, and (c) 10 uc-FeSe/STO samples at 35 K, where the colored images are the second derivative energy-momentum mappings of 2D HREELS measurements, and red solid lines are guides to eyes. EDCs from different samples are compared at different momentum points: (d) $q=0.3{\AA }^{-1}$ at 35 K, (e) $q=0.8{\AA }^{-1}$ at 35 K, and (f) $q=0.8{\AA }^{-1}$ at 300 K.

Figure 5

(a) LEED I/V spectra at 35 K of the (00) spot for STO (pink), 1 uc-FeSe/STO (purple), and 10 uc-FeSe/STO (orange). (b) LEED patterns for 1 uc-FeSe/STO at 35 K and (c) 300 K, for a primary energy of 140 eV. The red circles indicate the (00) spot. (d) The intensity of the (00) spot of 1 uc-FeSe/STO as a function of temperature with the primary energy of 220 eV.

Figure 6

(a) EDCs of phonon spectra at $\overline{\mathrm{\Gamma }}$ point of clean STO(001) (without Nb) at 35K. Red data are expanded by 15 times from the raw data in black. (b) Temperature-dependent EDCs of treated STO at $\overline{\mathrm{\Gamma }}$ point. (c) Temperature-dependent EDCs of 1 uc-FeSe/STO at $\overline{\mathrm{\Gamma }}$ point. (d) Plot of the energies of several modes as a function of temperature for treated STO and 1 uc-FeSe/STO at $\overline{\mathrm{\Gamma }}$ point.

Figure 7

(a) Comparison of EDCs of phonon spectra at $\overline{\mathrm{\Gamma }}$ point of clean STO(001) (without Nb) and clean STO(001). (b) Comparison of EDCs of phonon spectra at $\overline{\mathrm{\Gamma }}$ point of 1 uc, 3 uc, and 10 uc-FeSe/STO. (c) Energy loss spectra at $\overline{\mathrm{\Gamma }}$ point of several typical oxide surfaces: rutile ${\mathrm{TiO}}_2$(110) (without Nb), ${\text{BaTiO}}_3$ (without Nb), STO(110) $2\times 8$, STO(110) $4\times 1$, and clean STO, all at room temperature.

Figure 8

Comparison between experimental data and theoretical calculations for single-layer FeSe films, (a) with checkerboard AFM spin configuration on Fe lattice, and (b) without magnetic structure on Fe lattice. Orange stars label the $A_{1g}$ mode (22.6 meV) and $B_{1g}$ mode (25.6 meV) measured by Raman scattering at 7 K [43]. (c) Comparison between theoretical results for single layer and strained bulk FeSe, both with AFM spin configuration on Fe lattice with lattice constant $a=3.905\text{Å}$ (d) Comparison between theoretical results for bulk FeSe with the lattice constants of 1 uc, 3 uc, and 10 uc-FeSe films respectively, both with AFM spin configuration on Fe lattice. Lattice constants used in theoretical calculations are extracted from Ref. [26].