Light-Activated Nuclear Spin Polarization in Dilute Ferromagnetic (Ga,Mn)As

We study light-induced nuclear spin-polarization in a thin film of Ga1-xMnxAs (x 0.04), a dilute ferromagnetic semiconductor, grown on a GaAs substrate. High-field inductively-detected Ga-71 NMR was performed with samples immersed in superfluid He to investigate the effects of continuous-wave near band-edge optical illumination on lattice nuclear spins in the ferromagnetic phase. The photon energy dependence of the light-induced NMR signals for GaAs and the GaMnAs film samples were recorded using circularly polarized light. Interpretation of the data was guided by electronic band structure calculations using the k.p method in the presence of an external magnetic field using the modified 8-band Pidgeon-Brown model. The photon energy dependence of the NMR transition intensity exhibited a shift of the absorption band edge; invariance with respect to the sense of helicity of the exciting light; and an absence of oscillations in the photon energy dependence, all of which are consistent with theoretical predictions. The dynamics of the optically activated NMR experiments was investigated by variable optical intensity studies and light/dark modulated optical pumping experiments. This is because doping with Mn (a p-type dopant) can push the Fermi level deep into the valence bands and block the optical transitions (Burstein-Moss effect) needed to create spin polarized electrons. Additionally, the calculated enhancement of the conduction electron g-factor by over two orders of magnitude is expected to quench the electron-nuclear spin angular moment transfer, which impedes the hyperpolarization of lattice nuclei. Experiments with variable light intensity and optical gating reveal a mechanism consistent with light-induced quadrupolar relaxation, a process that will certainly interfere with the optical transfer and storage of quantum information in the lattice nuclear spin states in this material.
Comment: Assignment of the experimental optically pumped NMR signals to the Ga(Mn)As film layer, as reported in this manuscript, is probably incorrect and a revised interpretation is currently under consideration