It was found that the weak 2x2 reconstruction remained once the Ga shutter was closed, and becomes progressively brighter under NH_3 flux with decreasing substrate temperature as shown in Figure 26, reaching the same brightness as the integral streaks. Then, within a couple of degrees, the reconstruction disappears and the pattern becomes 1x1. It was found that the substrate temperature at which the transition occurred was a function of jet nozzle temperature as shown in Figure 27, but it is unclear at this point if it was due to the NH_3 molecule energy or changing NH_3 flux. The data was taken early on in our GaN investigation with the original jet when it was still poorly characterized using a sample provided by Honeywell Corporation. More detailed studies are planned with our own samples.
It was also noticed that once the substrate temperature was dropped below the transition temperature, reheating of the sample did not bring back the 2x2 reconstruction. GaN growth had to be performed long enough first to get the weak 2x2 reconstruction back to observe the transition. From the growth time needed it appeared that at least 1 monolayer of GaN had to be deposited, which was around 5 min. with the original jet. This observation was one of the factors that lead to the conclusion that the low fluxes obtained with the original jet are impractical for GaN growth.
It was further observed that GaN samples left overnight in the growth chamber became occasionally 3x3 reconstructed, but no RHEED pictures were taken. No explanation has been found so far, but the possibility of contaminations reacting with the surface can not be excluded (a likely candidate would be As since our system is also used for GaAs growth). It should also be noted that Molnar et. al. (Ref. 40) observed the same reconstructions after their samples were cooled after growth in their MBE system.
Figure 25: 112(bar)0 and 11(bar)00 RHEED pattern of GaN on c-plane sapphire (early growth attempts). Notice the 2x2 reconstruction.
Figure 26 (below): 2x2 to 1x1 Transition under NH_3 Flux (112(bar)0 and 11(bar)00 direction)
Growth
Before Transition
After Transition
Figure 26: 2x2 to 1x1 Transition under NH_3 Flux (112(bar)0 and 11(bar)00 direction)
Figure 27: Transition Temperature versus Nozzle Temperature
Table of Contents