Interdiffusion in GaAs(1-x)Sbx/GaAs superlattices studied with high-resolution x-ray diffraction and secondary ion mass spectroscopy

Abstract

Nominally undoped GaAsSb/GaAs superlattices were annealed at temperatures between 900 and 1100°C in a closed quartz ampoule. A strong dependence of the interdiffusion coefficients in the GaAs/GaAsSb superlattices on the arsenic vapor pressure was observed by two independent methods: secondary ion mass spectroscopy, and high-resolution x-ray diffraction using dynamic calculations to extract interdiffusion coefficients. The interdiffusion coefficient was low in the Ga-rich regime where an Arrhenius like dependence with an activation energy of 1.5Ϯ0.4 eV and a preexponential factor of 7.1ϫ10 Ϫ12 cm 2 s Ϫ1 was found. For the As-rich regime the activation energy was 2.0Ϯ0.1 eV and the preexponential factor 7.8ϫ10 Ϫ9 cm 2 s Ϫ1 .

Interdiffusion induced In(Ga)NAs films growth on GaAs substrates by low-pressure metalorganic chemical vapor deposition

“…It is well known that diffusion process plays a crucial role in the growth of semiconductor materials and interdiffusion or intermixing is inherent in the heteroepitaxial growth [12][13][14]. There have been a number of experimental reports about the diffusion and interdiffusion effect in III-V compound semiconductor, such as carbon diffusion in GaAs [12], N/As atomic intermixing in the GaInNAs/GaAs quantum-well [13] and Sb interdiffusion in GaAsSb/GaAs superlattices [14].…”

“…There have been a number of experimental reports about the diffusion and interdiffusion effect in III-V compound semiconductor, such as carbon diffusion in GaAs [12], N/As atomic intermixing in the GaInNAs/GaAs quantum-well [13] and Sb interdiffusion in GaAsSb/GaAs superlattices [14].…”

Interdiffusion effect on GaAsSbN/GaAs quantum well structure studied by 10-band k•p model

“…[14]. Two types of interdiffusion are possible in the GaAsSbN/GaAs QW: (1) the interdiffusion between As and Sb atoms and (2) the interdiffusion between N and As atoms.…”

Hybrid Neural Network Modeling of Anion Exchange at the Interfaces of Mixed Anion III–V Heterostructures Grown by Molecular Beam Epitaxy

“…The local gradients for the neural network component are used to derive the gradient of the error with respect to the previous weights ( ). This gradient, which is required to construct the weight correction mechanism, is expressed as (16) The partial derivatives of the kinetic model account for the contribution of each kinetic parameter to the network error. The delta rule of the standard BP algorithm is used for the hybrid neural network, and the synaptic weights are updated such that the predicted responses approach the desired responses.…”