Génération optique d'ondes de spin pour le transport d'information – GOSPININFO

This ambitious project proposes an alternate solution to the pure spin current generation. This question raised recently in the scientific community. One key feature of such current is that they could be dissipation-less while being able to carry information. Our approach is to take advantage of the spin flip wave, a well defined collective mode of a spin polarized two dimensional electron gas. A mode which naturally propagates by rotation of electron spins without displacement of their charge. The electron system we will use is a paramagnetic electron gas embedded in a semimagnetic quantum well of CdMnTe. The growth of such heterostructure is well understood and high mobility electron gas have been obtained. We will generate the spin flip wave with a stimulated Raman mechanism induced by a 100fs pulsed pump. Kerr rotation measurements of a linearly polarized retarded probe will give us the oscillation frequency of the induced spin polarization. If the frequency matches the spin flip wave frequency determined by spontaneous Raman measurements, this will demonstrate the spin wave generation. A second tilted pump beam will be used to transfer a finite wavevector to the wave. The probe will be diffracted by spatial periodic spin polarization and the detection of the diffracted beam will demonstrate the wavevector transfer occurring through stimulated Raman photon exchange between the two pump beams. By varying the tilt angle of the two pump beams we will extract a dispersion curve of the spin flip wave. Comparison with the dispersion obtained by spontaneous Raman will again identify the spin flip wave as the diffraction source. We will further focus on understanding the conditions to optimize (1) the generation mechanism (2) the spin flip wave life time. As we already know that the localized spin system of the Mn impurities will act as the main spin relaxation source, we will vary the Mn concentration in order to (a) generate the mixed spin wave resulting from the coupling between the localized and the itinerant spin systems or (b) further reduction of Mn concentration. We expect to achieve lifetimes of 100ps, a time which will allow us to investigate the spin flip wave propagation over 10 µm distances. Indeed, we plan to excite and detect locally the wave, by using, for example, a lithographed mask with two µm-scale slits separated by a few µm. We expect to measure a Kerr signal retarded by the propagation time over the distance between the two slits. The wave velocity estimed to be a few millions of cm/s will be detremined. We will further explore the coherent phase control of the spin wave whose phase could be used to carry information. The spin flip wave will then act as a phase modulated carrier. Beyond the new concepts that our results will show, we have chosen the model system CdMnTe of semi-magnetic compounds in a future perspective of spin based electronics using room temperature ferromagnetic semiconductors (like ZnMnO for example). We will be in position to anticipate the dynamics of spin flip wave in spin polarized systems where both localized and itinerant spin systems interact.