Spin-tranfer-Torque (STT) based devices are being actively pursued by many semiconductor manufacturers today as a memory technology for applications beyond CMOS scaling limit. The chief advantages of the so-called spin-torque magnetic random access memory (STT-MRAM) include simple integration topology, non-volatility in data retention, and potentials for processor-embedded applications. Integration of advanced metal-oxide-metal tunnel junctions into the backend of silicon technologies presents a set of materials and processing challenges that are being aggressively addressed industry-wide. In this talk I review the basic device physics of STT-based nanomagnetic switching, using the characteristics of the two-terminal STT-based tunnel junction as an example. As an example of real-world device and materials complexity, I’ll discuss the dependence of tunnel device STT switching characterstics on junction resistance-area product, and the likely role hot-electron spin-flip scattering plays in these processes. The demand for high area-density circuit for cost-competitiveness, the need for highly reliable switching characteristics for computation, and the nature of finite temperature nanomagnet dynamics combine to generate needs for significantly more spin-current density. For fast, nonvolatile and deterministic manipulation of nanomagnets, and based on present-day device physics understanding, this requires new sources of spin-current to be considered, such as thermal magnonic or spin-orbit-derived spin-currents. I will briefly review recent advances with these new sources of spin currents, and the likely common challenges they will give rise to, in terms of materials and device design and development.

Bio：Jonathan Sun is a Research Staff Member at the IBM T. J. Watson Research Center. He’s been with IBM Research for over 22 years, and has focused his work over the last decade and more on the device and materials physics of spin-torque and related phenomena and magneto-dynamics. Jonathan got his BS in Physics from Fudan University in Shanghai, China, and his PhD degree in Applied Physics from Stanford University. After a three-year stint at Superconductor Technologies, a start-up in Santa Barbara, CA, he came to IBM Yorktown, first studying device physics related to magnetic field sensing with Josephson junctions and SQUIDs of high-temperature superconductors, then on spin-dependent transport physics, materials, and dynamics. Jonathan holds some of the very early patents on spin-torque switched junction device concepts. He is one of the first to grasp the concept and implications of spin-torque in transport physics and in dynamics, to experimentally demonstrate STT switching, and to formulate the basic device- and materials-physics understanding of the STT phenomena that is the science behind the
STT-technology today. Jonathan is a Fellow of the American Physical Society, and a senior member of the IEEE Magnetics Chapter.