QED Induced Switching in Spin-Valves

Dec. 7, 2012 - PRLog -- Background
Today, spin-valve ferromagnetism is based on theoretical predictions [1,2 ] made over a decade ago. Spin-valves comprise alternating nanoscale layers of FMs separated by a NM spacer as illustrated in the thumbnail. FM stands for ferromagnetic and NM for non-magnetic. Spin-valves produce spin-polarized current by passing un-polarized current through a first FM layer, the polarization unchanged as the current flows through the NM spacer. Upon interaction with the second FM layer, a giant magnetoresistance known as GMR is thought to transfer the spin angular momentum from the first FM to the second FM, the process tending to produce parallel spins that significantly lower the GMR.

The spin-valve configuration based on metal instead of insulator NM spacers evolved from emphasis on spin-transfer as the mechanism by which the GMR is lowered. Because of this, increased photoconductivity in either FM or NM layers from charges created by the photoelectric effect were not considered. Today, the theory [1,2] of spin-valves still does not consider any significant lowering of the GMR from photo carriers created in either the NM or FM layers.

Problem
The problem is the non-parallel spins absorbed in the second FM more likely transfer their angular momentum as a spin-torque to the relatively rigid lattice and not to the electron spins. Moreover, the spin-torque propagates by phonons through the second FM lattice. Phonons respond at frequencies < 10 GHz having delay times of order > 100 ps. However, the response of electron spins is much faster. Laser studies in femtomagnetism [3,4] show demagnetization times on a sub-picosecond time scale (< 350 fs) that suggests spin-valve theory [1,2 ] based on the transfer of angular momentum through the lattice by phonons is not yet optimized. Indeed, light-induced switching by spin-valves is more consistent with sub-picosecond spin response and may very well supersede the current-induced spin-torque mechanism in the future.

Joule Heat in Nanoelectronics and QED Radiation
In nanoelectronics, QM requires the heat capacity of atoms in circuit elements like memristors and PC-RAM devices to vanish thereby precluding conservation of Joule heat by an increase in temperature. Instead, Joule heat is conserved by the QED induced creation of non-thermal EM radiation. QM stands for quantum mechanics, QED for quantum electrodynamics, and EM for electromagnetic. Provided the RI of the circuit element is greater than that of the surroundings, EM radiation at EUV levels is created by the frequency up-conversion of Joule heat to the TIR confinement frequency of the element. RI stands for refractive index, EUV for extreme ultraviolet, and TIR stands for total internal reflection. For example, memristors and PC-RAM devices comprising titanium dioxide and GST films having thicknesses < 10 nm create EUV > 40 eV, and therefore excitons (hole and electron pairs) are readily created by the photoelectric effect, the positive holes of which act as photo carriers that significantly reduce the nominal resistance of the circuit element by the dramatic increase in photoconductivity.

Significant reductions in the GMR by lowering the Curie temperature Tc  near ambient by photo carriers [5,6 ] to allow switching by small temperature increases from Joule heating including thermoelectric control [7] may be ruled out because QM precludes temperature changes, however small  in the FM  layers of spin-valves. Indeed, the QED creation of holes as charge carriers that significantly increase the photoconductivity of FM layers suggests that ferromagnetism may not even be necessary for switching in spin-valves. In this regard, spin-valves find similarity with PC-RAM devices where data is stored and erased with changes in resistance produced by the QED induced creation and neutralization of holes upon intemittant reversals in bias voltage. See http://www.prlog.org/11649780-the-ovshinsky-effect-by-qua... (http://www.prlog.org/11649780-the-ovshinsky-effect-by-qua...) Less similar, but related are QED induced holes in memristors under continuous sinusoidal voltage. See http://www.prlog.org/11642370-electrical-charging-of-memr... (http://www.prlog.org/11642370-electrical-charging-of-memr...)

QED Spin-Valve Mechanism
Like memristors and PC-RAM, the QED induced spin-valve mechanism is based on positive charged holes produced by the photoelectric effect from EUV radiation created from Joule heat by QED. Indeed,QED induced radiation claims the GMR is the usual resistance of the FM layer in the absence of electrical fields while the significant reduction in GMR is caused from holes created by the photoelectric effect from the QED induced conversion of Joule heat to EUV radiation.

Summary and Comments
1. The injection of localized spin-currents in a spin-valve allows precise spatial control in switching not possible with external non-local magnetic fields. Like memristors and PC-RAM devices, QED in spin-valves creates EM radiation at EUV levels from Joule heat that by the photoelectric effect produces holes in the FM that in turn increases the FM photoconductivity and lowers the GMR. Notably, switching in memristors and PC-RAM devices proceeds without invoking ferromagnetism, and therefore spin-valve switching in spin-valves today is most likely caused by the increase in photoconductivity alone from the holes created by QED induced radiation.

2. QED induced EUV is created provided the RI of the FM is greater than that of the surroundings. The RI condition is illustrated for ultrathin Fe nanoislands [8] driven by a spin-polarized current from a STM tip across a vacuum gap on one side and a W substrate on the other, thereby satisfying the QED condition that the RI of Fe nanoislands is greater than the surroundings. However, in design applications, RI data in the EUV for other materials is required but may not be reliable, even if available.  
   Moreover, only current pulses with opposite polarity cause switching reversal. Since Joule heating is independent of current direction, spin-torque and not Joule heat are commonly thought to be the driving force for ferromagnetism reversal. However, the QED interpretation differs. Polarity is indeed important, but reductions in the GMR cannot occur by QED without Joule heat creating the EUV necessary for the dramatic increase in hole-induced photoconductivity.

References
[1] L. Berger, “Emission of spin waves by a magnetic multilayer traversed by a current.” Phys. Rev. B, 54, pp. 9353,1996.
[2] J.C. Slonczewski, “Current-driven excitation of magnetic multilayers,” J. Magn. Magn. Mater., 150, pp. L1–L7, 1996.
[3  J-V Bigot, et al., “A. Ultrafast Magnetization Dynamics,” http://www-ipcms.u-strasbg.fr/spip.php?rubrique631&la... (http://www-ipcms.u-strasbg.fr/spip.php?rubrique631&la...),
[4] U. Bovensiepen, “Femtomagnetism: Magnetism in step with light,” Nature Physics, 5, 461, 2009.
[5] H. Ohno, et al., “Electric-field Control of Ferromagnetism,” Nature, 408, 944, 2000.
[6] T. Dietl, et al., “Zener Model Description of Ferromagnetism in Zinc-Blende Magnetic Semiconductors.” Science, 287, 1019, 2000.
[7] A. M. Kadigrobov, et al.,”Thermoelectrical manipulation of nano-magnets,” J. Appl. Phys., 107, 123706, 2010.
[8] G. Herzog, et al., “Heat assisted spin torque switching of quasistable nanomagnets across a vacuum gap,” App. Phys. Lett., 96, 102505, 2010.