Our core focus is on fundamental challenges facing Insulatronics. While charges cannot flow through insulators, spins can flow through magnetic insulators.
The coupling between electric currents in metals and spin-excitations in magnetic insulators is via spin-pumping and spin-transfer torques. Spin-pumping is the emission of spin-currents from dynamical spins in insulating or conducting magnetic materials to adjacent metals. The reciprocal process, spin-transfer, transfers spin angular momentum from the itinerant spin-polarized electrons in metals to the magnetic materials' spins. In several articles covering a wide range of ferromagnetic and antiferromagnetic insulators, we have made predictions for the strength, nature, and extent of these couplings. In doing so, we have fulfilled one of the main goals of our projects.
We first predicted that spins could flow across several micrometers in common antiferromagnetic insulators. The long-range transport is feasible both at high temperature via the semi-classical diffusions of spins and at low-temperatures via spin superfluidity. The spin current flow can be controlled by tuning the antiferromagnetic resonance frequency with an external magnetic field. Subsequently, with experimental colleagues, we realized the flow of spin information across tens of micrometers in the common antiferromagnetic insulator hematite, the rust's main ingredient. The results open doors towards the realization of ultra-fast, low-power antiferromagnet-insulator-based spin-logic devices that operate at room temperature.
Experimental colleagues have demonstrated spin-pumping from insulating antiferromagnets into metal, the emission of spin currents from a precessing antiferromagnet. The experimental results are in agreement with our earlier developed theory. Our results open the door to the unprecedented controlled generation of coherent, pure spin currents at terahertz frequencies.
In another collaboration with experimentalists, we have achieved efficient spin-wave propagation in thin films of the antiferromagnetic insulator hematite with large magnetic domains. Zero field spin transport is achieved across micrometers, as required for device integration.
We have also explored the hitherto less known physics related to magnon condensation, superfluidity, and superconductivity mediated by magnons.
Antiferromagnets may exhibit spin superfluidity since the dipole interaction is weak. We investigate nonlocal spin transport in a planar antiferromagnetic insulator with a weak uniaxial anisotropy. The anisotropy hinders spin superfluidity by creating a substantial threshold that the current must overcome. Nevertheless, we show that applying a high magnetic field removes this obstacle near the antiferromagnet spin-flop transition. Notably, the spin superfluidity can then persist across many micrometers, even in dirty samples.
Additionally, we explore routes to realize electrically driven Bose-Einstein condensation of magnons in insulating antiferromagnets. Even in insulating antiferromagnets, the localized spins can get coupled strongly to itinerant spins in adjacent metals via spin-transfer torque and spin pumping. We describe the formation of steady-state magnon condensates controlled by a spin accumulation polarized along the staggered field in an adjacent metal. Two types of magnons, which carry opposite magnetic moments, exist in antiferromagnets. Consequently, and in contrast to ferromagnets, Bose-Einstein condensation can occur for either sign of the spin accumulation. This condensation may occur even at room temperature when the metal interaction is fast compared to the antiferromagnet's relaxation processes. In antiferromagnets, the operating frequencies of the condensate are orders of magnitude higher than in ferromagnets.
We have demonstrated that magnons can mediate superconductivity in thin metal films sandwiched between magnetic insulators. The exchange coupling causes electrons within the metal to interact with magnons in the insulators at the interfaces. This electron- magnon interaction induces electron-electron interactions, which in turn can result in p-wave superconductivity. In yttrium iron garnet (YIG)-Au-YIG trilayers, superconductivity sets in at temperatures somewhere in the interval between 1 and 10 K.
