Abstract
Anisotropy of superconductivity is one of the fundamental physical parameters for understanding layered iron-based superconductors (IBSs). Here we investigated the anisotropic response of resistive transition as a function of thickness in iron selenide (FeSe) based electric-double-layer transistors (EDLTs) on , which exhibit superconducting transition temperatures as high as 40 K below . According to the analyses of the in-plane and out-of-plane upper critical fields and the magnetic field angle dependence of the resistance in ultrathin condition, we found that the anisotropy factor is 7.4 in the thin limit of , which is larger than that of bulk IBSs. In addition, we observed the shorter out-of-plane coherence length of 0.19 nm compared to the -axis lattice constant, which implies the confinement of the order parameter in the one unit cell FeSe. These findings suggest that high- superconductivity in the ultrathin FeSe-EDLT exhibits an anisotropic three-dimensional (3D) or quasi-two-dimensional (2D) nature rather than the pure 2D one, leading to the robust superconductivity. Moreover, we carried out the systematic evaluation of the anisotropic against thickness reduction in the FeSe channel. The in-plane as a function of normalized temperature is almost independent of until the thin limit condition. On the other hand, the out-of-plane near decreases with increasing , resulting in the increase of at around to 32.0 at the thick condition of , which is also confirmed by measurements. The counterintuitive behavior can be attributed to the degree of coupling strength between two electron-rich layers possessing a high superconducting order parameter induced by electrostatic gating at the top interface and charge transfer from substrates at the bottom interface. Besides a large for exceeding 20 T even at , we observe the decoupling crossover of the two superconducting layers at low temperature, which is a unique feature for the high- FeSe-EDLT on .
- Received 11 December 2017
- Revised 5 April 2018
DOI:https://doi.org/10.1103/PhysRevB.97.174520
©2018 American Physical Society