The spectrum of energy levels of the Ga-vacancy/deuterium complexes in p-GaAs

Abstract

Energy levels of the complexes formed by Ga vacancies binding one or more deuterium (or hydrogen) atoms are investigated by low-temperature photoluminescence. Since the as-grown, nondeuterated material is unintentionally doped p type, the emission is possible because of ‘‘internal’’ recombination from a D donor in an adjacent bond-center position and the ground levels of the different possible vacancy configurations, with zero, one, or more D atoms trapped in its dangling bonds. At 2 K, the transitions occur at 1.14, 1.19–1.22, 1.33–1.30, and 1.34 eV, and the last two have never been observed before. The given energy ranges correspond to laser excitation increasing from the lowest to highest level, for a given D treatment of the material. We explain the new transitions and the multiplet character of some of the bands in terms of the possible different ways of accommodating the D atoms inside the vacancy. The level separation in our data is less than that predicted theoretically for the ‘‘bare’’ vacancy with a fully unrelaxed lattice.

Identification of N–H related acceptor defects in GaAsN grown by chemical beam epitaxy using hydrogen isotopes

“…In addition, N−D related defects are more negatively charged than N−H ones, since they appear at a slightly higher temperature in DLTS signals (Fig. 5(a)) [28][29][30]. This can be explained by the involvement of a negative charge in the structure of the defect, which is more negatively charged when the proton (H, D) has lower vibration amplitude (more distant from the negative charge).…”

A Comparison of Zinc and Carbon Doped on the Atomic Interdiffusion of InGaAs/AlGaAs Quantum Wells Laser Structures After Annealing

“…Hence, we attribute this peak to a point defect complex related to the group-III species. In GaAs, two emission are reported at 77K, 1.32 eV related to Ga 2-As and at 1.284 eV related to Ga -As 22) . Therefore it is possible that the peak observed at 943 nm is related to these antisites.…”

Thermal effects on surface Fermi level for GaAs(001)

“…5 is due to the recombination of electrons on shallow donors ͑Si Ga ͒ with holes on shallow acceptor levels ͑Si As ͒. [19][20][21] The 1.36 eV emission is due to the recombination of an electron bound to an As vacancy with a hole bound to an acceptor level. 19,22 The appearance of the 1.36 eV peak is probably due to the migration of dopant Si atoms and displacement to As vacancy sites.…”

“…23 The 1.24 eV emission band seems to be due to complex defects related to Ga vacancies ͑V Ga ͒. 19,20,22 The low-energy 1.15 eV emission band is due to Si Ga -V Ga 3Ϫ defects. 19,20 In the highly Si-doped substrate, this 1.15 eV emission band is clearer because Ga vacancies related to 1.15 eV are formed by the compensation of Si Ga donors during solidification.…”

“…19,20,22 The low-energy 1.15 eV emission band is due to Si Ga -V Ga 3Ϫ defects. 19,20 In the highly Si-doped substrate, this 1.15 eV emission band is clearer because Ga vacancies related to 1.15 eV are formed by the compensation of Si Ga donors during solidification. 19,24,25 The above experimental results suggest that As and Ga vacancies are formed near the surface by heating in UHV above 450-480°C for the lightly and highly Si-doped substrates, and that the diffusion depth of Ga and As vacancies depends on the doping concentration.…”