DEFECT STUDIES OF MBE GROWN ALGAAS/GAAS AND INGAAS/GAAS MATERIALS

Abstract

In this project, the DX center in n-AIGaAs and dislocation induced traps in pInGaAs/GaAs are studied. Molecular Beam Epitaxy (MBE) has been employed in this work to grow high quality AIGaAs/GaAs and InGaAs/GaAs materials. The DX center, which is one of the defects studied in this work, is a very important defect in n-type III-V or even II-VI compound semiconductors. Recently, we (in collaboration with U. C. Berkeley) and other groups independently given the experimental evidences that the DX center is a negative-U center which traps two electrons at the ground state. Based on the negative-U theory, DX centers can cause carrier saturation effect. The carrier concentration saturation effect is an important subject to study as it is a general effect in n-type compound semiconductors and it will affect the performance of optoelectronic and microelectron devices significantly. However, the study of this effect has been hindered by the lack of experimental proof of its presence owing to many complexities in this matter. In this project, much effort was invested on the experimental work and the difficulties were finally overcome. After carefully designing the sample structures to minimize influences which can cause confusions, the carrier saturation effect due to Si DX centers in AIGaAs was experimentally examined and verified for the first time. In the process, the understanding of the carrier saturation effect has also been enhanced. The dislocation induced trap is the second subject in this thesis. InGaAs/GaAs is one of the widely used materials in the lattice mismatched semiconductor devices. Owing to the few studies in dislocation induced hole traps, the p-InGaAs/GaAs is used in this work to study the dislocation related traps in these materials. In the design of the sample structures, the misfit dislocation distribution in heterostructures and the requirements to perform capacitance Deep Level Transient Spectroscopy (DLTS) measurement on the samples are the principal considerations. The fact that we are able to differentiate the threading dislocation induced traps from the misfit dislocation induced traps for the first time is a result of thoughtful characterizations of the sample with carefully designed sample structures. Dislocations and traps in the pInGaAs/GaAs lattice-mismatched heterostructures are investigated by cross-section Transmission Electron Microscopy (XTEM), DLTS and Photoluminescence (PL) in the samples. The amount and types of dislocations observed by XTEM in the samples generally agree with Dodson-Tsao' s plastic flow model. The XTEM work has confirmed that all misfit dislocations in these samples are 60° dislocations. By comparing the XTEM and DLTS results, we unambiguously identify that the threading dislocations in bulk layers introduce three hole traps, H1, H2 and H5 with activation energies 0.32eV, 0.40eV and 0.88eV, respectively, and one electron trap, E1 with an activation energy 0.54 eV. The misfit dislocations in relaxed InGaAs/GaAs interface induce a hole trap, H4 with an activation energy 0.67~0.73 eV. The PL spectra of GaAs and InGaAs layer in our samples show that these epilayers are of very high quality if there is no misfit dislocation. If all the traps introduced by dislocations are nonradiative recombination centers, then the PL spectra of these samples can be easily explained by the presence of these traps. The excellent agreement among the different techniques is a strong indication of the correctness of our results. We believe that these results will benefit future studies of dislocations in compound semiconductors.