Crystal Defects In GaN on (0001) Sapphire
- PDF / 3,382,805 Bytes
- 5 Pages / 414.72 x 648 pts Page_size
- 30 Downloads / 234 Views
a time of 15 min. After deposition of the AIN initiation layer, growth of the GaN was performed. During the growth, the substrate temperature was 790'C and the NH 3 pressure was 1E-5 Torr. A Ga flux of 1.2 ML/s was used, which after 4.0 hours resulted in a film that was approximately 2 pm in thickness. Under these specific conditions growth is occurring in the excess Ga regime. After growth, the substrate were cut into small segments in order to produce samples which were of suitable size for analysis in the TEM. Samples were prepared for TEM analysis in both plan-view and cross-section. The plan-view samples were prepared by grinding then dimpling to a thickness of -10 pm followed by ion milling with argon ions to perforation from the dimpled side. Cross-section samples were prepared by using a 'sandwich' technique. Two samples with the same orientation were glued to one another using M-bond 610 epoxy. The sample was then dimpled from one side followed by single-sector ion milling in a Gatan precision ion polishing system (PIPS) using beam energies ranging from 5.0-2.0 kV. The resulting sample was then dimpled from the opposite direction to a thickness of -10 gm and once again single sector ion milled in the PIPS until perforation occurred. The TEM samples were then analyzed in a Philips CM-30 operating at 300 kV. RESULTS AND DISCUSSION Figure 1 and Figure 2 are two weak-beam dark-field plan-view TEM images which have been taken from the same area near the [2113] zone of GaN. Figure 1 was formed with g = 101 I while figure 2 was formed with g = 0110. In figure 1 fringe contrast occurs from individual domains as they are inclined to the beam at this zone. These domains are randomly located throughout the sample and range in size from 20-110 nm. The boundaries between the domain and matrix appear not to be located on any particular crystallographic planes, though it is possible that these boundaries could be faceted on a fine scale. Similar contrast
from domains have been described and shown by other authors.[4,10] However, the explanation of the origin of these defects varies in the literature. Figure 2, imaged with a different reflection does not show the fringe contrast from the domains that were visible in figure 1. In this type of material, faults are out of contrast when g.R = (0, n) and in contrast when g.R n, where n is any integer.[l 1] An additional weak-beam dark-field Figure 1. Plan-view weak-beam dark-field TEM image of a image formed with GaN film grown on (0001) a-A120 3 . The image was formed g = 1101 showed similar contrast with g = 10 11, near the [2113] zone of the GaN. from the domains that were visible in figure 1. g.R analysis indicates that the displacement of these domains is parallel to the c-direction, as the fringes were not visible when imaging with (hkit), I = 0. Additionally, when the sample was imaged near the (0001) zone it was apparent that the interface between the domain and surrounding matrix was nearly parallel to the c-direction, as they could be viewed almost edge on with no f
Data Loading...
