||Oscillatory and re-entrant phenomena in superconductor-ferromagnet- nanostructured bilayers and applications to spin-valves
||Project DFG (Germany)
||Ghitu Institute of Electronic Engineering and Nanotechnologies,ASM
||Sidorenko Anatolie dr. hab., prof. univ.
||Laboratory of Cryogenics
||Superconductivity, nanostructures, spintronics, spin-switch
The so-called Fulde-Ferrell-Larkin-Ovchinnikov (FFLO) state can be realized in artificially layered superconductor-ferromagnet (S/F) nanostructures, i.e. in S/F layers with film thickness in the nanometer range. Because of the finite pairing momentum the Cooper-pair wave function oscillates in the ferromagnetic layer. Upon changing the thickness of the flat F layer the pairing function flux through the S/F interface becomes modulated due to a change of the interference conditions for the incident and reflected wave function at the interface. As a result, the superconducting critical temperature Tc oscillates with increasing thickness of the ferromagnetic layer.
The most spectacular effect predicted by the theory for quite some time, is the reentrance of the superconducting state in S/F bilayers which could be convincingly detected only very recently. Two ferromagnetic layers offer a further control of superconductivity in the S layer sandwiched in between, if one involves a rotation of the magnetization of one of the layers in a F/S/F trilayer with respect to the other. For a proper choice of materials of S and F layers, and adjustment of their thicknesses, L. Tagirov suggested that the supercurrent through the S layer can be switched on and off as the mutual alignment of magnetizations of the F layers changes from antiparallel (AP) to parallel (P), respectively. The effect was called superconducting spin-valve in analogy with the conventional giant magnetoresistance spin-valves by B.Dieny. The maximum value of ΔTc=TcAP–TcP, yielding the optimal condition for the spin-valve effect, is in direct dependence on the scale of the Tc oscillations, and one achieves a maximum of ΔTc in a properly designed F/S/F trilayer exhibiting the superconductivity reentrance phenomenon. It is worth to mention that from the point of view of the existing theory the F/S/F trilayer is a stack of 2 bilayers (F/S and S/F), each of the half-thickness for the S-layer of the trilayer (see below). Thus, these F/S and S/F bilayers are building blocks of the F/2S/F trilayer, however, a new degree of freedom to align the magnetizations in the F/2S/F trilayer parallel or antiparallel gives rise to a new physics and functionality of the structure.
Several tough conditions have to be fulfilled simultaneously: (1) a flat superconducting layer with a thickness of about the superconducting coherence length (~10 nm); (2) weak ferromagnetic and homogeneous material for the F layer; (3) clean and transparent interfaces between F and S materials; (4) adjustment of the optimal growth conditions for F/S and S/F sequences; (5) exchange biasing of one of the F layers without disturbing the boundary conditions for the pairing function. Thus, all experiments up to date have given controversial results: either very small spin-valve effect ΔTc ~ 10-100 mK, or also small but an inverse effect of ΔTc ~ -(10-30) mK (TcAP < TcP).
We have already fabricated successfully the first building block of the superconducting spin-valve, a S/F (Nb/CuNi) bilayer, fulfilling the conditions (1) to (3) for the S → F sequence. Further work towards the spin-valve is a main goal of the proposal. It can be reviewed as follows: (1) apply experience, already collected upon fabrication of S/F bilayers, to prepare optimized F/S bilayers which demonstrate reentrant superconductivity (the bilayer is not necessarily specular-symmetric because of different growth conditions and the different sequence of technological operations; (2) prepare optimized F/2S/F core trilayers as a stack of F/S and S/F bilayers, demonstrating reentrant superconductivity or deep Tc oscillations of the structure; (3) prepare CuNi layers with a pinning CoO layer, or on a surface-reconstructed substrate to produce exchange bias or additional coercivity in the material, study the hysteresis properties of the system; (4) incorporate the bias into the S/F or/and F/S structure and provide reentrant or deeply oscillating behavior of the superconducting Tc; (5) prepare optimized F/2S/F trilayers with a pinned F-layer to keep the magnetization fixed, demonstrate the spin-valve effect upon rotating the magnetization of the second F layer by an external magnetic field.
These investigations will result in deep understanding of the interplay between superconductivity and ferromagnetism, and provide a functional core structure which utilizes the unconventional physics realized in the S/F-proximity systems.