Spin-based electronics, commonly referred to as spintronics, exploits the spin of the electron in addition to its charge to provide more functional and energy-efficient devices. In recent years the concept of spin-transfer torque has gained much attention since it offers high density in computer memory and logic devices based on spintonics. However, existing electronic spin-torque efficiency is low, and, as a result, requires high current for information storage in devices. This proposal from the university-industry team at the University of Alabama and IBM Research Center puts forward an investigation into novel spin-based electronic device structures that utilizes a potentially extremely efficient means of generating spin-transfer torque - by spin currents generated by a temperature difference in magnetic multilayer structures. Understanding the spin transfer properties at the interfaces and successful demonstration of the proposed concept can lead to considerable improvement in the energy efficiency of spin-based devices that will open up new applications. The proposed project requires a multidisciplinary effort that will make significant contributions to scientific knowledge, education outreach and infrastructure. Project personnel will play a key role in several ongoing education and outreach activities related to the proposed research. These include collaboration with local schools to facilitate participation by high school students in research, public tours and demonstrations.
The proposed project will investigate a novel spintronic device concept that addresses the issue of high write currents required for present spin-transfer torque devices. The physical principle behind the project goals is based on a recent theoretical prediction that outlines a thermal route of enhancing the spin torque efficiency required to switch an ultra-thin magnetic layer. The key material is a magnetic insulator which generates spin waves (or magnons) when thermally excited. The net result is amplification in the spin-polarized current when the magnons transfer spin information to the conduction electrons in an adjoining conducting spacer layer. The proposed thermagnonic method has the potential for achieving one to two orders-of-magnitude greater quantum yield in spin-transfer torque devices. The research collaboration will develop fundamentally new knowledge and understanding of the theoretically predicted thermal spin-torque effect from magnons generated within oxide-based magnetic insulators. Fundamental investigations of spin-torque and spin-pumping behavior in candidate heterostructures shall be undertaken. Ferromagnetic resonance, magneto-optic techniques, and magnetic microscopy of patterned nanostructures shall elucidate the thermal spin-transfer torque effect and quantify the thermal-to-spin-current conversion efficiency. Unique synthetic strategies shall be pursued that exploit low temperature growth processes for magnetic spinel oxide and garnet films, which is expected to be crucial for integrating a disparate set of materials.
