Parasitic Reactions between Alkyls and Ammonia in OMVPE
- PDF / 348,031 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 28 Downloads / 179 Views
thickness measurements have been confirmed by cross-sectional scanning electron microscopy. X-ray measurement was done using a QCI double crystal x-ray diffractometer using a wide area detector. The x-ray source is Cu. The primary crystal is GaAs. The measured X-ray peaks are (0004) peak for Wurtzite structures. Results This horizontal reactor has been used to grow GaN and other related nitride materials. The x-ray rocking curve's FWHM of several pim thick GaN epilayers varies between 150 arcsec to 400 arc-sec. The electron background carrier concentration typically is in the low 1016 cm 3 . The highest electron mobility is about 600 cm2/V-sec at room temperature and the bound exciton FWHM at 2K PL measurement is 2.4 meV. Indium concentration as high as 25% has also been obtained in this reactor. These results indicate that the material quality grown using this reactor is high and the observations to be discussed later represent what is actually happening in a real OMVPE reactor (rather than a reactor designed to do decomposition study, for example). Atmospheric growth of GaN has been done easily using TMGa and NH 3 . FigI shows the growth efficiency [3] as a function growth temperature at two reactor pressure. The growth efficiencies show the typical kinetically limited regime at low growth temperature and the diffusion-limited regime at moderate temperature [8]. At growth temperatures above 1000 'C, GaN growth at atmospheric pressure becomes extremely difficult and no thickness data can be obtained. This is probably due to the strong thermal convection at atmospheric pressure and high growth temperature, At a low growth pressure of 85 torr, however, GaN can be easily grown with growth temperature up to 1075 'C (the highest temperature achievable by the RF generator) with high quality. The slight drop in growth efficiency at high temperature for GaN grown at 85 torr is likely due to pre-deposition of GaN on the reactor wall at such a high growth temperature [9].
0 E
1000
( 500 I-aN z-0 300
at-780torr GaN at 85 torr
400
500
600
700
800
900
1000
1100
Tg (°C) Fig. 1: Growth efficiencies for GaN as a function of growth temperature. The reactor pressure is either 780 torr or 85 torrt
104
Atmospheric pressure AIN growth has been attempted at 500 'C, 600 'C, and 700 'C. No epilayer was grown even after using a long growth time and high TMAI flow rate. The growth efficiency of AIN at atmospheric pressure is estimated to be less than 100 ýtm/mole. This low level of unmeasurable growth efficiency indicates that TMAI and NH 3 might undergo parasitic reaction at atmospheric pressure to deplete the reactants. Indeed, white deposits are observed near the reactor entrance where TMAI and NH3 are mixed. The growth of AIN is possible only at low reactor pressure. Fig.2 shows the GaN and AIN growth efficiency dependences as a function of reactor pressure. In the experiments, the mass flow rates of the gases are fixed and only the reactor pressure is changed. This means that the flow velocity and reactant concentrati
Data Loading...
