High Quality AlGaAs Regrowth on Oxide-Free Al x Ga 1-x As (x=0.26) by Metalorganic Chemical Vapor Deposition
- PDF / 287,415 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 49 Downloads / 202 Views
etching using HCL has also been carried out in MOCVD prior to regrowth[2]. A thin GaAs protecting layer has been grown as the top layer to reduce oxidation [3-4]. In this study, we used a Se-doped AlGaAs as a substrate to regrow the AlGaAs epilayer by MOCVD. This is a continuous study based on previous results of our group, which show that Se will produce a passivation effect to deactivate deep level traps in AIGaAs and improve the AIGaAs surface quality [5-6]. By applying this passivation effect to the regrowth technique, we have observed that Se can delay the oxidation on AlGaAs surface and thus improve the quality of regrown AlGaAs layers. EXPERIMENT A horizontal, atmospheric home-made reactor was used for material growth. Highquality Al-containing materials have been routinely grown using this reactor. Trimethylgallium (TMGa), trimethylaluminum (TMAl) and AsH 3 (in concentration of 100%) were used as precursor species. Diethylzinc (DEZn) and H2Se(100 ppm in ultra high-purity H2) serve as p and n type dopant sources, respectively. The source of alkyls (TMGa, TMAI) and DEZn are held in temperature controlled bath to obtain constant vapor pressure. Pd purified H2 is used as the carrier gas. The materials were grown at 730 TC with a V/III ratio of 40, and a growth rate of 4.0 ýtm/hr. The undoped Al 0 26Ga 0.74As with thickness of 1.5 ptm was grown on n' GaAs 61 Mat. Res. Soc. Symp. Proc. Vol. 405 01996 Materials Research Society
1.5
.8
A
,, S.4 S.2 0
1.05
C
A
.5
-
-1300
-1500
1.5
0
9200
9000
1.2 C
•1.0 A
B3C
.9
AA
.6
.5
.3 ~0
0
2900
3100
L 5100
arc see, Axis 2
5300
arc see, Axis 2
A: AlGaAs substrate layer, B: AIGaAs regrown layer, C: GaAs substrate Figure 1. X-ray rocking curve for the (004) reflection fromAlo. 13Gao.87As grown on (a) #634o, (b) 634f, (c) 635o, and (d) 635f substrates. substrate (sample# 634). Se doped Al 0 .26 Ga0 .74 As (n=3.4x10 7 cm -3) with the same thickness was grown later (sample #635) to avoid memory effect. Both samples were intentionally exposed to air for more than fourteen hours. Then each of them was split into two equal parts (we marked them as 634o, 634f, 635o and 635f). Sample 634o and 635o were cleaned by TCE, ACE then Methanol ("o" means overnight). Sample 634f and 635f were cleaned by TCE, ACE, Methanol, then HCI : 120 = 1 : 1 for ten seconds to remove the native oxide on the surface. (V' means fresh) We put all these four samples into reactor immediately after cleaning. A 1.5 ptm Alo. 13Gao. 87As layer without any doping was regrown on the top of four sample to compare the interface and quality of regrown layers. A high-resolution double crystal X-ray diffraction (Bede Scientific) was used to exam the material quality and to determine the Al composition. Electrochemical C-V profiler was employed to characterize the interface properties. EDTA was used for both etching and forming a Schottky contact to the semiconductor. An 62
(a)
1019 10 18
AIGaAs regrown layer
1017
GaAs
16
1015
0
101 118
. 1
AlGaAs substrate
2
3
40
2
(C)
3
4
Data Loading...
