Spin-based nanoelectronic devices for mobile Information-Communication Technology

Perhaps the best known (or most successfully implemented) spin-based device is the hard-disk read-head. Indeed, the discovery of giant magnetoresistance enabled a paradigm shift in the miniaturization of magnetic storage technology, which was disruptive enough to earn a Nobel for the two researchers who carried out the initial studies [1]. In a nutshell, giant magnetoresistance refers to the fact that the electrical properties of a multilayer containing at least two magnetic layers depend on the orientation of their magnetic moment. For instance, if the magnetic layers are cobalt, iron or nickel (or their alloys), the resistance of the structure is maximum when the magnetic moments are antiparallel to each other, and minimum when they are parallel.

More recently, it has been demonstrated the inverse phenomenon can also be observed: the relative orientation of the magnetic moments of two ferromagnetic layers can be manipulated by applying an electrical bias (i.e. a current or a voltage) across the structure. This is a consequence of spin-momentum transfer between the conduction electrons and the magnetization of the layer they are travelling across, which effectively induces a torque on the magnetization, the so-called ‘spin-transfer torque’ or ‘spin-torque’ [2-6]. Two main effects can be induced exploiting this torque: the magnetic moment of a given layer can be switched to a chosen direction – for instance, from parallel to antiparallel to the magnetization of the second layer – or it can be induced to gyrate around a given direction for as long as the electrical bias is applied.

Spin-transfer switching as the first spin-transfer induced phenomenon to be demonstrated experimentally, with the first report published at the end of 2000 [5]. Today, spin-transfer switching is the write scheme for non-volatile, ultra-fast Spin-Transfer Torque Random Access Memory (STT-RAM) bit devices. STT-RAM can be designed so that they can scale down to more than one fifth of all other available technologies, including SRAM [7,8]. Spin-transfer driven precession, first demonstrated in 2003 [6], has been suggested as working principle for other spin-based nanoelectronics devices currently under consideration, which range from tuneable, low input power radio-frequency oscillators wireless communication, to magnetic field sensors, negative resistors, amplifiers, write heads and random number generators. Indeed, the frequency of such devices can be adjusted simply by changing the applied bias, and they provide sufficient power [9] while at the same time being about 50 times smaller than present devices used in mobile telecommunication [10]. Moreover, novel materials hold the promise of pushing the frequency limit beyond what present-day technology can achieve [11]. Possible applications include anti-collision systems for cars, remote hospitals and immersive audio-video entertainment systems.

[1] http://www.nobelprize.org/nobel_prizes/physics/laureates/2007/index.html
[2] J. C. Slonczewski, J. Magn. Magn. Mater. 159, L1 (1996).
[3] L. Berger, Phys. Rev. B 54, 9353 (1996).
[4] M. D. Stiles, A. Zangwill, Phys. Rev. B 66, 014407 (2002).
[5] J. A. Katine, F. J. Albert, R. A. Buhrman et al., Phys. Rev. Lett. 84, 3149 (2000).
[6] S. I. Kiselev, J. C. Sankey, I. N. Krivorotov et al., Nature (London) 425, 380 (2003).
[7] http://www.avalanche-technology.com/technology/ram
[8] http://www.everspin.com/
[9] A. Deac, A. Fukushima, H. Kubota, et al., Nature Phys. 4, 803 (2008).
[10] P. Villard, U. Ebels, D. Houssameddine, et al., IEEE J. Solid-State Circuits 45, 214
(2010).
[11] S. Mizukami, F. Wu, A. Sakuma, et al., Phys. Rev. Lett. 106, 117201 (2011).