PC8-4

Nernst effect in ultrathin FeSe/STO grown by PLD

16:30-16:45 Dec.5

*Tomoki Kobayashi, Ryo Ogawa, Atsutaka Maeda
Dept. of Basic Science, the University of Tokyo, 3-8-1 Komaba, Meguro-ku, Tokyo 153-8902, Japan
Abstract Body

The iron-chalcogenide superconductor FeSe has garnered significant attention because the monolayer film on oxides shows a large enhancement of superconducting transition temperature (Tc) [1]. The enhancement is attributed to interface effects, including electron transfer from the oxide and possible electron-phonon coupling via the interface. In FeSe/SrTiO3 (STO), superconducting transitions with the onset Tc around 40 K and zero resistivity below Tczero (=10–29 K) was observed by the resistivity measurements [1], which are higher than Tc of 9 K in bulk form. The enhanced superconductivity has been realized through molecular beam epitaxy [1] and pulsed laser deposition [2]. Furthermore, gap opening around Fermi energy below 65 K, which may suggest superconducting transition, was reported using angle-resolved photoemission spectroscopy (ARPES) [3]. However, recent studies of in-situ resistivity measurement reported that the Tczero and Tconset was still lower than gap opening temperatures (Tg) even in the same sample [4,5]. The discrepancy between resistive Tc and Tg is believed to be attributed to pseudogap phase due to the superconducting fluctuations enhanced by the two-dimensional(2D) nature of the superconductivity, however, the pseudogap can also result from a density wave state. Thus, other probes are needed to elucidate the existence of superconducting fluctuations well above Tc. One promising technique for detecting superconducting fluctuations with high sensitivity is the Nernst effect which is the generation of transverse thermoelectric voltage by the thermal gradient in the presence of perpendicular magnetic field [6]. In superconductors, the movement of vortices induced by a thermal gradient produce the Nernst signals. In addition, the Nernst effect caused by superconducting fluctuation is observed above Tc. In amorphous superconductors, Nernst signals was detected far above Tc, and the magnetic field and temperature dependence are good agreement with the theory based on amplitude fluctuations of superconducting order parameter [7]. The Nernst effect has also been employed to investigate potential phase fluctuations in high Tc superconductors [8].

In this study, we performed the Nernst effect measurements on ultrathin FeSe/STO. We fabricated 4- and 9-layer films using the Pulsed laser deposition. The Nernst effect measurement was performed using one-heater-two-thermometer configuration under 0.4–9 T. Figure (a) shows temperature dependence of sheet resistance Rsq (upper panel) and \( \scriptsize{\alpha} \begin{subarray}{rl} \scriptsize{2D} \\ \small{xy} \end{subarray} \ /\!B \) (lower panel), where \( \scriptsize{\alpha} \begin{subarray}{rl} \scriptsize{2D} \\ \small{xy} \end{subarray} \ /\!B \) is the transverse thermoelectric coefficient and B is the magnetic field, for the 4-layer film. The 4-layer film showed superconducting transition with Tconset of 29 K and zero resistance below 8.0 K at 0 T. As B increases, resistive transition became broader. Notably, in this temperature regime, \( \scriptsize{\alpha} \begin{subarray}{rl} \scriptsize{2D} \\ \small{xy} \end{subarray} \ /\!B \) started to increases at T* = 30 K, and peak structure was observed at all magnetic fields. Furthermore, B dependence of \( \scriptsize{\alpha} \begin{subarray}{rl} \scriptsize{2D} \\ \small{xy} \end{subarray} \ /\!B \) changed below T*. Above T*, \( \scriptsize{\alpha} \begin{subarray}{rl} \scriptsize{2D} \\ \small{xy} \end{subarray} \ /\!B \) is independent of B while it varies with B at lower temperatures. These results suggest that the existence of superconducting fluctuations just below T* and vortex movement in lower temperatures. The similar behavior was observed in the 9-layer film as shown in Fig. (b). While the 9-layer films exhibited Tconset = 23 K and Tczero=7.3 K, \( \scriptsize{\alpha} \begin{subarray}{rl} \scriptsize{2D} \\ \small{xy} \end{subarray} \ /\!B \) started increasing at T*= 28 K. Finally, let us discuss the relationship between \( \scriptsize{\alpha} \begin{subarray}{rl} \scriptsize{2D} \\ \small{xy} \end{subarray} \ /\!B \) as a function of T and fundamental properties: resistance and diamagnetism which is measured by the two-coil mutual inductance technique. Figure (c) shows temperature dependence of Rsq in 0 and 1 T(upper panel), real part of mutual inductance M1 between drive and pick up coils in 0 T (middle panel), and \( \scriptsize{\alpha} \begin{subarray}{rl} \scriptsize{2D} \\ \small{xy} \end{subarray} \ /\!B \) in 1 T (lower panel). Although Rsq decreased around Tconset, the Meissner effect was observed only below 8 K which is close to Tczero. This result is similar to the previous report [9] while a singular study reported observation of the Meissner effect below 65 K [10]. Above Tczero, the vortex movement was detected, and \( \scriptsize{\alpha} \begin{subarray}{rl} \scriptsize{2D} \\ \small{xy} \end{subarray} \ /\!B \) caused by superconducting fluctuations persist up to T.*, which is about 1.2 times Tconset. These results are similar to other iron-chalcogenide thin films [11].

References

[1] Q. Y. Wang et al., Chin. Phys. Lett. 29, 037402 (2012).
[2] T. Kobayashi et al., Supercond. Sci. Technol. 35, 07LT01 (2022).
[3] S. He et al., Nat. Mater. 12, 605 (2013).
[4] A. K. Penderson et al., Phys.Rev. Lett. 124, 227002 (2020).
[5] B. D. Faeth et al., Phys. Rev. X 11, 021054 (2021).
[6] A. Pourret et al., New J. Phys. 11, 055071 (2009).
[7] I. Ussishkin, S. L. Sondhi, and D. A. Huse Phys. Rev. Lett. 89, 28 (2002).
[8] Y. Wang, L. Li, and N. P. Ong Phys. Rev. B 73, 024510 (2006).
[9] Z. W. Hao et al., CPL 31, 017401 (2014).
[10] Z. Zhang et al., Sci. Bull. 60, 1301 (2015).
[11] F. Nabeshima et al., Phys. Rev. B 97, 024504 (2018).

pict

(a),(b)Temperature dependence of sheet resistance (upper panel) and \( \scriptsize{\alpha} \begin{subarray}{rl} \scriptsize{2D} \\ \small{xy} \end{subarray} \ /\!B \) (lower panel) for the 4-layer(a) and 9-layer films, respectively. (c) Temperature dependence of Rsq in 0 and 1 T (upper), M1 in 0 T, and \( \scriptsize{\alpha} \begin{subarray}{rl} \scriptsize{2D} \\ \small{xy} \end{subarray} \ /\!B \)  in 1 T.

Keywords: Iron-based superconductor, Interface superconductivity, Nernst effect, Superconducting fluctuations