Ultralow power and ultra-wideband spintronics near thermodynamic limits

ended

Moore’s Law drove the technology revolution for more than five decades and left no aspect of our lives untouched. State-of-the-art computation relies on transistors, whose dimensions or power consumption could no longer be reduced. Nevertheless, growing need for information processing, battery-constrained internet-of-things devices and wireless connectivity necessitates discoveries of nanoelectronic building blocks with novel physics. Thus, fundamental breakthroughs are needed in highly power-efficient non-volatile computational elements that meet the speed, bandwidth and scalability requirements of microelectronics industry. Using electronic spins for non-volatile computation could offer very diverse new device physics and architectures to meet these requirements. In SKYNOLIMIT project, I aim to experimentally demonstrate ultra-wideband, ultralow-power and non-volatile logic circuit architectures that operate based on nanoscale spins called magnetic skyrmions. Skyrmions are nanoscale spin structures that allow for room temperature computation and memory functions near thermodynamic limits while being robust against fabrication imperfections and stray magnetic fields. In this project, (1) I first computationally model, fabricate and test the novel functional nanomaterials with giant spin-orbit coupling and low damping to achieve all-electric generation/detection and processing of skyrmions using multilayers of topological insulators and/or 2D transition metal dichalcogenides on insulating rare earth iron garnet films. Second, (2) I plan to experimentally demonstrate skyrmion processors including signal generators, logic gates, registers, and fast Fourier transformers. Third, (3) I plan to experimentally implement neural network hardware using skyrmionics. Thus, high-speed and ultra-wideband 2D skyrmionics could help reduce power consumption, extend mobile battery life by a few orders of magnitude and help spintronics become a part of mainstream electronics.