Evidence of interface superconductivity in ultrathin FeSe/STO grown by PLD

Dec. 1 16:20-16:35

*Tomoki Kobayashi1, Hiroki Ogawa1, Hiroki Nakagawa1, Fuyuki Nabeshima1, Atsutaka Maeda1
The University of Tokyo, Japan1

Iron chalcogenide superconductor, FeSe, has attracted much attention because monolayer film on SrTiO3 (STO) exhibits significant enhancement of superconducting transition temperature (Tc) from 8K to 40--- –65 K[1,2]. The high Tc in the monolayer FeSe/STO is attributed to the interface effects such as electron doping from STO substrate and the coupling between electrons in FeSe and phonons in STO. Other than in FeSe/STO, such Tc enhancement by the interface effects was reported in different heterostructures, i.e. FeSe/BaTiO3[3], FeSe/TiO2[4,5], FeSe/LaFeO3[6], and FeSe/NdGaO3[7]. The search for higher Tc and the mechanism of the superconductivity can be explored by fabricating heterostructures of FeSe with different oxide materials. Pulsed laser deposition (PLD) has advantages in fabricating various heterostructures at low cost because different materials can be deposited simply by replacing target materials. However, almost all studies on the Tc enhancement in monolayer FeSe/oxide so far were performed using the molecular beam epitaxy (MBE) technique. We have worked on realizing Tc enhancement in ultrathin films of FeSe/STO using PLD. Previously, we successfully realized Tc enhancement in FeSe/STO with thickness d ≥ 7 nm[8], which is quantitatively the same as the results in MBE-grown films[9]. To get these results, preparing atomically flat STO substrate is crucial. The high Tc cannot be explained by the strain effect on FeSe established so far[10]. This suggests the realization of Tc enhancement by the interface effects in PLD-grown FeSe/STO. In the early stage, however, for films with d ≦ 5 nm, superconductivity was not realized possibly because of the sample degradation by the air exposure. This result shows the need for the protection layer to realize superconductivity in thinner films. In this study, we grew ultrathin films of FeSe/STO with capping layers to protect the FeSe films and investigated transport properties under magnetic field. We deposited 10-nm-thick amorphous Si at room temperature as a capping layer. Figure (a) shows the temperature dependence of sheet resistance for capped films with d = 2–4 nm. All films showed superconducting transition at low temperatures. onset Tc (Tconset) of the films are higher than 25 K and much higher than Tc of bulk FeSe (Tc = 9 K), suggesting Tc enhancement probably caused by the interface effects. This also indicates the successful role of the protection layer. Figure (b) shows the temperature dependence of the value of RH/d, where RH is Hall coefficient, for all films. All films exhibited negative values of RH/d at low temperatures, which is in good agreement with the results for ultrathin FeSe/STO grown by MBE[9]. This is in contrast to ultrathin FeSe flakes, where RH has positive values[10]. In addition, positive RH irrespective of the degree of strain was reported for FeSe thin films grown on substrates other than STO[11]. Thus, these results indicate electron doping from STO substrate. Figure (c) shows the temperature dependence of the resistance under magnetic field H of 0–9 T for the 4-nm-thick film. For H ⊥film, Tconset became lower and the superconducting transition became broader. In contrast, for H//film, superconducting transition was almost unchanged, indicating large anisotropy of critical fields. The estimated c-axis coherence length ξc is found to be 0.2 nm which is smaller than c-axis parameter of FeSe. This suggests that superconductivity is confined at the interface of FeSe/STO.

Figure (a). Temperature dependence of sheet resistance for 2-, 3-, and 4-nm-thick films with capping layers.
Figure (b). Temperature dependence of RH/d, for all films.
Figure (c). Temperature dependence of sheet resistance for the 4-nm-thick film perpendicular (top) and parallel (bottom) magnetic field of 0−9 T.

[1]. Q. Y. Wang et al., Chin. Phys. Lett. 29, 037402 (2012).
[2]. S. He et al., Nat. Mater. 12, 605 (2013).
[3]. R. Peng et al., Nat Commun. 5, 5044 (2014).
[4]. H. Ding et al., Phys. Rev. Lett. 117, 067001 (2016).
[5]. S. N. Rebec et al., Phys. Rev. Lett. 118, 067002 (2017).
[6]. Y. Song et al., Nat. Commun. 12, 5926 (2021).
[7]. H. Yang et al., Sci. Bull. 64, 490 (2019).
[8]. T. Kobayashi et al., Supercond. Sci. Technol. 35, 07LT01 (2022).
[9]. Q. Wang et al., 2D Mater. 2, 044012 (2015).
[10]. C. S. Zhu, et al., Phys. Rev. B 104, 024509 (2021).
[11]. F. Nabeshima et al., Jpn. J. Appl. Phys. 57, 120314 (2018).

Keywords: Iron chalcogenide, Pulsed laser deposition, ultrathin film, Interface superconductivity