Open AccessEffect of Substrate Orientation on the Growth Rate, Surface Morphology and Silicon Incorporation on GaSb Grown by Metal-Organic Vapor Phase Epitaxy
Author(s) -
Jian Yu,
Ishwara B. Bhat
Publication year - 2004
Language(s) - English
Resource type - Reports
DOI - 10.2172/822279
Subject(s) - metalorganic vapour phase epitaxy , silicon , epitaxy , substrate (aquarium) , materials science , dopant , secondary ion mass spectrometry , growth rate , doping , analytical chemistry (journal) , scanning electron microscope , hillock , metal , crystallography , nanotechnology , chemistry , optoelectronics , ion , geometry , composite material , layer (electronics) , metallurgy , organic chemistry , mathematics , geology , oceanography , chromatography
The growth and dopant incorporation during metal-organic vapor phase epitaxy (MOVPE) strongly depends on the orientation of the starting substrates. MOVPE of Si-doped GaSb layers on (100), (111)B and (111)A substrates were used to investigate the effect of orientation on the growth rate, the surface morphology and the silicon incorporation. orientation dependence of growth rates was studied as a function of temperature and V/III ratio. As the V/III ratios increases, the growth rate on the (111)B oriented substrate decreases, whereas that on the (111)A oriented substrate increases. The surface morphology on different substrates was studied by scanning electron microscopy (SEM) and atomic force microscopy (AFM). Flat-top hexagonal hillocks were observed on (111)B surface, and the growth was by step-flow on these facets. A surface kinetic growth model has been proposed to describe the growth features observed on (111)B surfaces. Finally, the orientation dependence of silicon incorporation was studied by secondary ion mass spectrometry (SIMS) and Hall measurements. It was found that the silicon incorporation rate was four times higher on (100) oriented surface than on (111)B oriented substrates. Although no conduction type switch was found on (111)B oriented substrate, throughout the Si doping range studied, a Ga slightly rich growth condition might be one of the factors accounting for the discrepancy between theoretical prediction and experimental results