Electrons, which are the fundamental particle responsible for electrical currents within metals, possess not only electrical charge but also carry an intrinsic spin. In most existing electrical devices, the spin does not play any role. However, in the past decade new generations of technology have begun development in which the electron spins are manipulated to add new functionalities. This type of spin-electronics, or "spintronics", has already achieved widespread use in the form of magnetic-field sensors in hard disk drives, and it is also under intense development to make computer memories in which spin-aligned electrical currents reorient ferromagnetic components to store information. The goal of this research project is to develop new understanding about the interactions between spin-aligned electrons and a ferromagnet or other selected materials, and particularly to understand current-induced torques that can arise from these interactions. The project will include the development of new experimental techniques to achieve accurate measurements of the strength and direction of current-induced spin torques, and to image how the magnetization of a ferromagnet moves in response to the torque. This work will advance basic understanding of electron spin dynamics and will provide information and techniques that will be needed for the development of magnetic memories and other technologies. The graduate and undergraduate students supported by the project will also gain an excellent interdisciplinary training in nanoscience.
The aim of this project is to develop new understanding about spin-transfer torques, which are mechanisms by which spin-polarized electrical currents can transfer their spin angular momentum to ferromagnets and (perhaps) antiferromagnets to reorient their magnetic moments. Already it has been shown that spin transfer can provide much stronger torques per unit current compared to using conventional magnetic fields, and spin torques can efficiently drive magnetic switching and precession. The proposed work will invent new experimental techniques to enable quantitative measurements of the strength and direction of spin torques in ferromagnetic layers. It will advance the technology of ultrafast electrical measurements and time-resolved x-ray microscopy to understand the magnetic dynamics that can be excited by spin torques. The project will also extend the study of spin torques to new classes of materials, specifically magnetic nanoparticles and antiferromagnets, for which interesting effects are predicted but no definitive experiments have been performed. The impact of project will be to advance basic understanding of electron spin dynamics and to provide information and experimental techniques that will be needed for the development of spin-torque-driven magnetic memories, frequency-tunable nanoscale microwave sources, and other technologies. The graduate and undergraduate students supported by the project will also gain an excellent interdisciplinary training in nanoscience.
The purpose of this project has been to develop new fundamental understanding about spin-transfer torques, which are mechanisms by which electrical currents can cause the transfer spin angular momentum to magnetic materials, in this way applying a torque to reorient their magnetic moments. The first commercial product based on the spin-torque mechanism, nonvolatile magnetic random access memory, entered commercial production in 2012. The research pursued within this project has sought to develop new methods to enhance the strength of spin transfer torques and to characterize the magnetic dynamics that they produce. One of the main achievements of the project is that it developed quantitative techniques to measure the strength and direction of the spin transfer torque in the type of magnetic tunnel junctions used for memory devices. These methods allowed the first accurate measurements of spin torque at high voltage bias, the regime of primary interest for magnetic memory applications. The new measurement techniques also led directly to a major discovery (funded primarily by other sources) of a new, much more efficient mechanism for producing spin transfer torques, the so-called spin Hall effect in heavy metals. This new mechanism can provide torques that are at least a factor of 10 stronger per unit current at room temperature than any previous mechanism. Following this discovery, research under this project was refocused in part to investigate what heavy-metal materials can produce the strongest spin Hall torques, and how the spin Hall torques can be used in new ways to manipulate magnetic insulators and antiferromagnetic materials. The project also investigated other new classes of materials that contain strong coupling between the electron spin direction and the direction of electron motion (i.e., spin-orbit coupling), to see if they might provide even better mechanisms for generating current-induced torques on magnetic materials. These exploratory experiments investigated both atomic membrane materials (e.g., MoSe2) and "topological insulators" (e.g., Bi2Se3). An additional major achievement of the project is the discovery that topological insulators can provide current-induced torques for manipulating magnetic devices that are even more efficient than the spin Hall effect in heavy metals. The technical results of this project are of immediate practical use for improving the performance of magnetic memory technologies based on spin transfer torque. As noted above, this technology entered commercial production in 2012, and currently many large electronic companies are working to develop it further in the hope that it can find widespread commercial application as the only type of memory that is nonvolatile, high-density, low-power, fast, and does not wear out, so that it will replace many types of silicon-based memory that do not possess all of these virtues. The research on the spin Hall effect, in particular, shows promise for providing a dramatic improvement in the energy-efficiency of magnetic memory devices while simultaneously improving their reliability, and it has the potential to enable new types of nonvolatile magnetic logic circuits, as well. The project has also provided broad societal benefit by contributing to the scientific research education of five graduate students and six undergraduates, who have gained mastery of state of the art techniques for nanofabricating and characterizing magnetic devices as well as training in how to present their research findings orally and in writing.
