Tunneling anisotropic magnetoresistance driven by magnetic phase transition

Abstract

The independent control of two magnetic electrodes and spin-coherent transport in magnetic tunnel junctions are strictly required for tunneling magnetoresistance, while junctions with only one ferromagnetic electrode exhibit tunneling anisotropic magnetoresistance dependent on the anisotropic density of states with no room temperature performance so far. Here, we report an alternative approach to obtaining tunneling anisotropic magnetoresistance in α′-FeRh-based junctions driven by the magnetic phase transition of α′-FeRh and resultantly large variation of the density of states in the vicinity of MgO tunneling barrier, referred to as phase transition tunneling anisotropic magnetoresistance. The junctions with only one α′-FeRh magnetic electrode show a magnetoresistance ratio up to 20% at room temperature. Both the polarity and magnitude of the phase transition tunneling anisotropic magnetoresistance can be modulated by interfacial engineering at the α′-FeRh/MgO interface. Besides the fundamental significance, our finding might add a different dimension to magnetic random access memory and antiferromagnet spintronics.

Giant Tunnel Magnetoresistance with a Single Magnetic Phase-Transition Electrode

“…As shown in Fig.3 (b), after the insertion, the transmission of GAFM-FeRh/MgO/Cu has been drastically suppressed and results in a significant drop of MPT-TMR to 16% (see Table 1 for details). The MPT-TMR now is qualitatively comparable with experimental value in '-FeRh/γ-FeRh/MgO/γ-FeRh tunnel junction [26]. This proves that the direct interface contact of '-FeRh with MgO barrier is essential for achieving giant MPT-TMR.…”

“…However, different from Fe/MgO/Fe MTJs where the lattice mismatch is as high as 4%, the lattice mismatch between '-FeRh and MgO is quite small and less than 1%. This suggests that the proposed '-FeRh/MgO/Cu junction may have better epitaxial quality which is confirmed by recent experiments [26]. Nevertheless, below we will briefly discuss the effect of some structural disorder especially the interface imperfection on MPT-TMR.…”

“…Despite of the fact that from previous theoretical calculations [25], the ground magnetic phase of '-FeRh is ferromagnetic with Rh termination and GAFM with Fe termination in FeRh/MgO interface. However, experiments always indicate that the magnetic order in '-FeRh/MgO interface is GAFM [26]. This suggests the Rh termination interface may not be a favorable interface compared with Fe termination interface.…”

“…The MPT-TMR can be hundreds of percent which is much larger than TAMR [11][12][13][14][15] and comparable with TMR in Fe/MgO/Fe-MTTs [3][4][5][6][7][8]. This huge MPT-TMR in '-FeRh/MgO/Cu make this tunnel junction structure promising for device application at room temperature [26]. Please note that the MPT-TMR for 5ML-MgO is +918% which is slightly smaller than the theoretical value reported +1161% in reference [26].…”

Spin density wave instability in a ferromagnet

“…More interesting is the idea that triplet superconductivity in FM UGe 2 is related to coupled CDW-ISDW fluctuations that emerge with pressure within the FM phase 31 . Because the unusual transition between ISDW and FM states in Fe 3 Ga 4 occurs just above room temperature, it has been suggested as a possible material for a magnetic tunneling junction device 32 . This suggestion takes advantage of the ability to drive a phase transition between the ISDW and spin-polarized phases with moderately sized magnetic fields and the ability to tune the phase transition temperature via pressure and disorder.…”