Abstract
The results of a series of high-magnetic-field (up to 22 T) Zeeman-spectroscopy measurements are reported on the Cr-related 0.84-eV luminescence line in GaAs. The observed Zeeman splitting and oscillator strengths can be satisfactorily explained in terms of a model in which both the excited and ground state have C3v symmetry and a twofold-orbital and fivefold-spin (S=2) nature. In addition the model provides an excellent fit to the oscillator strengths and energies of the fine structure observed in the no-phonon line seen at zero magnetic field in both absorption and luminescence. Our result conclusively demonstrates that a recently proposed model based on a ground-state effective Hamiltonian containing terms of only cubic symmetry is incorrect.