Challenges and emerging directions in spintronics

Summary form only given. Spintronics is a new field of research exploiting the influence of the electron spin on the electrical conduction. It is mainly known for the “giant magnetoresistance” (GMR)1-2 and the large increase of the hard disc capacity obtained with the read heads based on GMR. But the research on spintronics has also revealed many other interesting effects and is now developing along promising novel directions. The physical basis of spintronics is the influence of the electron spin orientation on the electrical conduction in magnetic materials: the conductivity can be much larger for electrons having, for example, their spin aligned with the magnetization 3-4. The GMR1,2,5 exploits this spin dependence in magnetic multilayers composed by a stack of ultrathin layers (a few nm) with, alternately, layers of magnetic (iron for example) and nonmagnetic metals (copper or chromium for example). A magnetic field, by aligning the magnetizations of all the magnetic layers, makes that there is an electrical short-circuit by half of the electrons which have the “good” spin orientation in ALL the magnetic layers. Without magnetic ordering by an applied field, the short-circuit effect does not exist and the electrical resistance is much larger. The large reduction of the electrical resistance by a magnetic field has been called “giant magnetoresistance” or GMR. The GMR is used to read the magnetic inscriptions on the hard discs of today 6-7 and the possibility of reading smaller inscriptions has led to a considerable increase (three orders of magnitude) of the capacity of the hard discs 6-7. The discovery of the GMR in 1988 kicked off an intense search of other phenomena also related to the influence of the electron spin on the electrical conduction. New effects have been found and this domain of research is now called spintronics8, sometimes described as a new type of electronics exploiting both the charge and the spin of the electrons. An example of very active field of research is the study of the spin transfer phenomena 9-11. In a spin transfer experiment one manipulates the magnetization orientation of a magnet without applying any magnetic field - the usual way - but by a transfusion of spin angular momentum from a spin-polarized current. This can be used, for example, to reverse the magnetization10 and this will be used soon in the next generation of magnetic memories called STT-RAM. The STT-RAM, in contrast with the semiconductor RAM of today, are nonvolatile, they do not need any electrical power to maintain the memory alive7. This will probably lead to a significant reduction of the energy consumption by the computers ad servers. In another regime the spin transfer can be used to generate oscillation in the radio wave frequency range11. The spin transfer oscillators (STO) are very promising of applications in telecommunications. The research in spintronics extends today in many promising directions. Spintronics with semiconductors aims at combining the potential of conventional semiconductors with the potential of spintronics8. Spintronics with graphene, carbon nanotubes or organic molecules has revealed the advantage of carbon-based materials on metals and semiconductors in term of long spin life time and long spin diffusion length12. The recent results on graphene are promising for the relay of conventional electronics in the so-called “beyond-CMOS” perspective and open the road, for example, to “spin only logic circuits” for a novel type of computer technology. Emerging directions are also single electron spintronics, one of the way to quantum computing, and neuromorphic spintronics in the direction of bio-inspired computers.