Spin torque MRAM — Challenges and prospects

Summary form only given. The ability of a spin-polarized current to reversibly switch the orientation of the magnetic moment of a nanomagnet, and/or to excite it into microwave precession, by the torque exerted through the transfer of spin angular momentum from the incident conduction electrons, have catalyzed what is now a quite broad and very active area of spin torque research. There has been very substantial progress in advancing the fundamental understanding of this new spintronics phenomenon, and in successfully moving it towards technological implementations, particularly spin-torque magnetic random access memory (ST-MRAM) and possibly spin-torque excited, nanoscale microwave oscillators. Major objectives of these research efforts are to reduce the current amplitude and the pulse width required to reliably switch a thin film nanomagnet, and to develop spin torque devices where the active element is incorporated in a reliable and high performance magnetic tunnel junction that is impedance matched to highly scaled CMOS transistors. In this presentation I will discuss some recent work that has sought to contribute to the rapidly advancing spin-torque research effort, by advancing the quantitative understanding of the details of the phenomenon, and by developing materials systems and magnetic nanostructures that enhance the efficiency and applicability of the effect. The latter includes tuning the magnetic anisotropy of the nanomagnet, which can reduce the onset current for reversal into the 100 microamp range, and the use of a perpendicular anisotropy reference layer to produce a spin current that is polarized in the direction orthogonal to the plane of the nanomagnet which has yielded reliable reversal behavior for pulse widths as short as 100 psec for an energy pulse of less than 0.2 pj. I will also report on initial results with a three terminal spin valve device where a low impedance spin-valve contact is employed to drive the nanomagnet reversal while a high impedance tunnel junction contact senses the nanomagnet's magnetic orientation. I will conclude by briefly discussing some of challenges that remain to be overcome before spin-torque based technologies, particularly ST-MRAM, can be successfully implemented.