He is also an IBM Fellow (IBM’s highest technical honor) and a Consulting professor in the Dept. of Applied Physics at Stanford University. Until summer 2015 he will continue to be director of the IBM–Stanford Spintronic Science and Applications Center and to manage the Magnetoelectronics group at the IBM Almaden Research Center, San Jose, CA.
Here in Halle, he will continue to further develop and shape the field of material sciences, especially of applied spintronics.
Stuart Parkin is well known for his work on the giant magneto-resistance (GMR) effect, for which he shared the 1994 American Physical Society’s James C. McGroddy Prize for New Materials and the Hewlett-Packard Europhysics Prize (European Physical Society, 1997) with Peter Grünberg and Albert Fert. Drs. Grünberg and Fert went on to win the 2007 Nobel Prize in Physics for their GMR discovery. The Royal Swedish Academy acknowledged the seminal importance of Parkin’s work in their report entitled Scientific Background on the Nobel Prize. The ground breaking work of Parkin was honoured by their citing six of his publications. Parkin translated the GMR effect into devices that revolutionized the field of non-volatile information technology.
Within the IBM Corporation Parkin was the initiator of the commercialization of the GMR effect in the form of hard disk reading heads. This way he is, in large part, responsible for, (a) the dramatically enhanced storage densities of hard disks approaching 1 Tbit/in2 since the introduction of GMR read heads by IBM in 1997, and, (b) one of the best examples of a discovery that arose out of basic science that was taken to market within a short time.
In 1995 Parkin passed another milestone of magnetoelectronics by the patented proposal of a novel non-volatile magnetic random access memory (MRAM). The functional element consisted in the replacement of the metallic spacer layer in a GMR element by an insulating barrier. Thus the device resistance for current flow perpendicular to the multilayer stack could be measured much more easily in the form of a tunneling magnetoresistance (TMR). While the first prototype was demonstrated by Parkin and colleagues in 1999, a few years later he was able to achieve record values of the TMR of over 200% at room temperature (Nat. Mat. 2004). The success was due to Parkin’s outstanding capabilities and expertise in materials science and its transfer into technologically feasible solutions. In that sense he acts as a catalyst mediating between the basic sciences, such as physics, chemistry and materials, and the commercial realization of novel products, such as magnetic sensors or memory devices. This is also reflected in his 9 patents on “spin valve” sensors and more than 30 patents concerning the implementation of magnetic tunneling junctions into MRAMs and magnetic sensors. Another 10 patents cover other areas of his intense research work. Overall Parkin has ~94 issued patents in the USA.
In the field of future spin-electronics Parkin started in 1999 to work on the magnetic tunneling transistor (MTT), a 3-terminal device, which combines spin-dependent tunneling with spin filtering in order to achieve highly spin-polarized ballistic electrons. The spin polarization of the hot electrons in GaAs was detected optically. This device may prove useful for efficient spin injection into semiconductors at room temperature: Parkin and his co-workers subsequently demonstrated more than 80% spin injection efficiency at room temperature in GaAs using a 2 terminal tunnel spin injector with MgO tunnel barriers.
Most recently, Parkin has proposed a radically different approach to building a solid-state non- volatile memory device using the current controlled motion of magnetic domain walls in magnetic nano-wires- what Dr. Parkin has termed “Racetrack Memory”. This memory is what Parkin would describe as being “innately three-dimensional”, in contrast to the inherent two-dimensional structure of both magnetic disk drives (data is stored in a single two-dimensional sheet of magnetic material) and silicon based microelectronics (logic is carried out using a single sheet of transistors fabricated in the surface of a single crystal of silicon). By fabricating nano-wires which are oriented perpendicular to the surface of a silicon wafer, multiple domain walls can be stored in these nano-wires, thereby increasing the memory capacity by one or more orders of magnitude compared to all conventional solid-state memories, both those in production and those which are under investigation.