ELECTRONIC SPECTROSCOPY STUDIES OF A P2S5NH4OH TREATED GAS(100) SURFACE

Microscopic characteristics of the GaAs(100) surface treated with P2S5/NH4OH solution has been investigated by using Auger-electron spectroscopy (AES) and x-ray photoemission spectroscopy (XPS). AES reveals that only phosphorus and sulfur, but not oxygen, are contained in the interface between passivation film and GaAs substrate. Using XPS it is found that both Ga2O3 and As2O3 are removed from the GaAs surface by the P2S5/NH4OH treatment; instead, gallium sulfide and arsenic sulfide are formed. The passivation film results in a reduction of the density of states of the surface electrons and an improvement of the electronic and optical properties of the GaAs surface.

PHOTOREFLECTANCE SPECTROSCOPY OF SEMIINSULATING GAAS

Clear observations of photoreflectance (PR) spectra due to excitonic transitions in semi-insulating GaAs bulk materials are reported. The modulation mechanism is attributed to the electromodulation induced by the Dember effect. This study indicates that the PR spectroscopy provides an important method for characterizing the crystal quality of high-resistivity GaAs.

INVESTIGATION OF NEGATIVE TRANSIENT CURRENT OF ARGON-IMPLANTED GAAS USING PHOTOINDUCED TRANSIENT-CURRENT SPECTROSCOPY

The investigation of deep levels of argon-implanted LEC-grown semi-insulating GaAs with implantation dosages ranging from 1 x 10(11) to 1 x 10(15) cm-2 has been performed. Using a photoinduced transient-current spectroscopy (PITCS) it was demonstrated that, for implantation dosages below 1 X 10(13) cm-2, a negative peak or negative transient current (NTC) was observed in the temperature range from 330 to 350 K. The magnitude of this negative peak increased with dosage up to a level of 1 X 10(12) cm-2, beyond which it decreased with dosage. The dosage dependence of the EL3 peak height and the resistance of the specimen have also been investigated. It was observed that the variation of the EL3 peak height with dosage was similar to the variation of the magnitude of the negative peak, that is the EL3 peak height likewise increased with dosage up to 1 X 10(12) cm-2, and then decreased. The resistance of the original high-resistivity specimen dropped abruptly when the dosage reached 1 X 10(12) cm-2. This critical dosage (1 X 10(12) cm-2) was found to be a threshold for the generation of a highly disordered state.

Reflectance-difference spectroscopy study of the Fermi-level position of low-temperature-grown GaAs

The results of a reflectance-difference spectroscopy study of GaAs grown on (100) GaAs substrates by low-temperature molecular-beam epitaxy (LT-GaAs) are presented. In-plane optical anisotropy resonances which come from the linear electro-optic effect produced by the surface electric field are observed. The RDS line shape of the resonances clearly shows that the depletion region of LT-GaAs is indeed extremely narrow (much less than 200 Angstrom). The surface potential is obtained from the RDS resonance amplitude without the knowledge of space-charge density. The change of the surface potential with post-growth annealing temperatures reflects a complicated movement of the Fermi level in LT-GaAs. The Fermi level still moves for samples annealed at above 600 degrees C, instead of being pinned to the As precipitates. This behavior can be explained by the dynamic properties of defects in the annealing process.

DETERMINATION OF THE X-CONDUCTION-SUBBAND ENERGIES IN TYPE-II GAAS/ALAS/GAAS QUANTUM-WELL BY DEEP-LEVEL TRANSIENT SPECTROSCOPY

Using deep level transient spectroscopy (DLTS) the X conduction-subband energy levels in an AlAs well sandwiched by double GaAs layers were determined. Calculation gives eight subbands in the well with well width of 50 Angstrom. Among them, five levels and the other three remainders are determined by using the large longitudinal electron effective mass m(1)(1.1m(0)) and transverse electron effective mass m(t)(0.19m(0)) at X valley, respectively. Two subbands with the height energies were hardly detectable and the other six ones with lower energies are active in the present DLTS study. Because these six subbands are close to each other, we divided them into three groups. Experimentally, we observed three signals induced from the three groups. A good agreement between the calculation and experiment was obtained. (C) 1995 American Institute of Physics.