There is a growing interest in exploring the use of electron spin in semiconductors for a variety of applications ranging from devices based on spin polarized transport to coherent quantum information processors. Information can be stored in both spin and orbital states, however spin has the advantage that it is more insulated against information-destroying environmental interactions than orbital states. For example, the homogeneous dephasing time T2 for an electron spin in doped silicon was estimated to be in the millisecond range at low temperatures by spin-echo measurements. Devices based on electron orbital states have coherence times several orders of magnitude shorter than this. Consequently, spin is a promising candidate for storage of quantum information, and can provide a new paradigm and opportunity for information technology with capabilities that are far beyond those of current state-of-the-art systems. This advantage is particularly prominent in the phase coherent regime where entanglement between quantum elements can be utilized. Entanglement provides a truly parallel processing environment and an exponential increase in functionality. Promising applications of quantum information technology include secure quantum communication and networking as well as quantum computing.
Secure quantum communication relies on entanglement of coherent photon states. To realize this opportunity for practical use, one must be able to transmit quantum information over long distances, perform quantum error correction, and retain the information without decoherence. Such rigorous conditions require development of a system that is capable of receiving quantum information in the form of coherent photon states, storing the information, performing the necessary operations, and then retransmitting the photon signal while maintaining quantum coherence throughout the process. It is well known that information in the form of photon polarization can be transferred to electron spin in semiconductors and vice versa in absorption/emission processes. Hence, utilizing the electron spin degree of freedom in solids provides a clear pathway to the development of a practical quantum communication system. Such a system could reliably function as a repeater to transmit quantum information over long distances and would enable a number of goals significant to DOD, including secure data transmission.
Recently a group led by Prof. Yablonovitch at UCLA proposed a promising scheme to achieve this function that is based on nanoscale semiconductor technology. As schematically illustrated in the figure at left they envision a semiconductor quantum communication system consisting of three main components: a receiver, an information-processing unit (IPU), and a transmitter. The receiver converts the incident photon polarization to a spin-polarized electron by a unique absorption process. The generated electron with quantum information stored in the spin degree of freedom is transferred to the IPU for processing and storage. Finally, the transmitter reconverts the electron spin state to a polarized photon. Electrons move between the transmitter/receiver and the IPU to couple to IPU processing/storage elements.
Design of the receiver and transmitter needs to satisfy several demanding conditions simultaneously. Since the device operates connected to optical fibers of a specific wavelength, the energy of principal electron hole transitions must be matched to the photon energy. In order to be able to transfer photon polarization into the electron subsystem without loss of information in the presence of a magnetic field, the electron g-factor should be small so that electron-hole transitions for both spin sublevels are within the bandwidth of the received photon. Both electron states should couple to a single lowest energy hole sublevel and selection rules of the corresponding optical transitions should enable the transfer of photon polarization into the electron spin without loss of information. In order to operate with only a single hole sublevel, the valence band Zeeman splitting should be large in a moderate magnetic field (or it should be a structure with very large hole g-factor in case of weak field). Models are being developed to learn the degree to which these conditions can simultaneously met.