Spinning electrons could lead to an electronics revolution. Today's semiconductor devices rely on the precise control of electronic charge. However, electrons have spin as well as charge leading towards new device ideas such as spin transistors, spin memory devices, or even spin quantum computers. Most essential requirements for such devices are, first, efficient injection of spin polarized electrons into the semiconductor, that is all electron spins must point into the same direction. Second, spin transport, the direction of the spins must be preserved as the electrons drift or diffuse through the semiconductor; in other words, the number of ''spin flips'' must be negligible.

The interest in spin-electronics has increased rapidly since the discovery of the giant magnetoresistance and in the meantime metallic electron-spin devices are commercially available, e.g., in reader heads of computer hard discs. However, in our opinion a major breakthrough of spin electronics and spin-controlled optoelectronics will only be possible in conjunction with semiconductors.

The efficient injection of spin polarized electrons into a semiconductor is still a problem. The injection of polarized electrons from a ferromagnetic tip through a vacuum barrier into GaAs has been demonstrated in 1992, but this approach is less, if at all, applicable in devices. Large efforts have been dedicated to demonstrate spin injection out of ferromagnetic contacts into semiconductors but only very recently Hammar et al. reported the first ferrromagnet-semiconductor interfacial current polarization of the order of 20 %. [1] The origin of the difficulities concerning this concept is still under discussion. A very efficient spin injection was demonstrated by our group using a dilute magnetic semiconductor as spin aligner. However, this concept works so far only at low temperatures. [2]

To study spin transport in semiconductors, the problem of electrical spin injection can be circumvented by optically creating spin-oriented electrons. Last year, our group in Marburg made the first measurement of electron-spin transport in GaAs in high electric fields. [3] We observed no loss of spin coherence for electric fields up to 6 kV cm^{-1} and drift lengths of 4 \mu m. This year, Kikkawy and Awschalom demonstrated spin transport over several tens of micrometers in low electric fields of some Vcm^{-1}. [4] All these measurements werde made at low temperatures. However, spin electronics is not just low temperature physics. While the coherence of the spatial part of the wave function decreases with increasing temperature, we measure, e. g. in ZnSe quantum wells, an increase of the coherence of the spin part of the wave function over much more than one order of magnitude when the temperature is increased from 5K to room temperature.

References

[1] P. R. Hammar, B. R. Bennett, M. J. Yang, and M. Johnson, Phys. Rev. Lett. 83, 203 (1999).
[2] M. Oestreich, J. Hübner, D. Hägele, P. J. Klar, W. Heimbrodt, W. W. Rühle, D. E. Ashenford, and B. Lunn, Appl. Phys. Lett. 74, 1251 (1999).
[3] D. Hägele, M. Oestreich, W. W. Rühle, N. Nestle, and K. Eberl, Appl. Phys. Lett. 73, 1580 (1998).
[4] J. M. Kikkawa and D. D. Awschalom, Nature 397, 139 (1999).