Optimization of the surface morphology of GaAs epitaxial layers grown by close-spaced vapor transport

1993 ◽  
Vol 71 (9-10) ◽  
pp. 462-469 ◽  
Author(s):  
Z. Huang ◽  
N. Guelton ◽  
D. Cossement ◽  
D. Guay ◽  
R.G. Saint-Jacques ◽  
...  

GaAs epitaxial layers were grown on (100) GaAs and vicinal substrates (2° off (100) toward (110)) by close-spaced vapor transport (CSVT) using water vapor as transport agent. We demonstrate that the temperature at which water vapor is injected into the reactor, Tinj, is a criticial parameter for the layer morphology. Furthermore, the optimum Tinj, is a function of water partial pressure, [Formula: see text]. For each [Formula: see text], there is a Tinj range for which specular layers are obtained. This range is defined for [Formula: see text] Torr (1 Torr = 133.3 Pa). The only defects appearing on the specular layers grown on (100) GaAs substrates are oval hillocks (structureless oval defects, and occasionally oval plateaus and oval defects with a faceted central core head). All oval hillocks have their long axis, L1, parallel to < 110 >. By measuring L1 and L2 (the short axis of the defect, perpendicular to L1), lateral growth rates of GaAs are obtained. They vary with [Formula: see text]. At high [Formula: see text], [Formula: see text] is 1 at low water vapor pressure. All oval hillocks result from the contamination of the substrate surface. By using vicinal GaAs substrates, the oval hillock density was decreased to about 500 cm−2, a result similar to that obtained with molecular beam epitaxy.

1994 ◽  
Vol 72 (1-2) ◽  
pp. 44-50
Author(s):  
D. Cossement ◽  
Z. Huang ◽  
G. Perron ◽  
B. Jean ◽  
J. P. Dodelet

In view of developing the close-spaced vapor transport technique (CSVT) to obtain III/V homojunction solar cells, it is necessary to finely control the growth rate of GaAs epitaxial layers. This has been performed either by controlling the water vapor pressure, [Formula: see text] injected in the reactor along with H2, in H2 + H2O ambient, or by controlling the water vapor pressure generated in situ by the reaction of H2 + CO2 in the reactor. For H2 + CO2 ambient, [Formula: see text], controls [Formula: see text] according to the following reaction: [Formula: see text]. The growth rates calculated with a diffusion controlled model are in agreement with the experimental values for both ambients, including the observation of a maximum in the evolution of the growth rate with [Formula: see text], Controlling the growth rate of GaAs by changing [Formula: see text] affects the carrier density (NA–ND) of p-type layers grown from Zn-doped GaAs sources. In both ambients (NA–ND) is a function of [Formula: see text]. Such a behavior is also obtained for the calculated carrier densities. It is the result of the transport of Zn as ZnO in CSVT. In H2 + CO2 ambient, where H2O and C are generated in situ, carbon is not incorporated as a major p-type doping impurity, contrarily to expectations, n-type GaAs layers were also obtained from Te-doped GaAs sources. In that case, the measured NA–ND values are not affected by changes in [Formula: see text] because water is not involved in the transport of Te in CSVT.


1991 ◽  
Vol 69 (3-4) ◽  
pp. 390-406 ◽  
Author(s):  
E. Koskiahde ◽  
D. Cossement ◽  
N. Guelton ◽  
R. Fillit ◽  
R. G. Saint-Jacques ◽  
...  

Epitaxial layers of GaAs on (100) GaAs substrates can be grown by close-spaced vapor transport using water vapor as the transporting agent. The parameters for the transport are [Formula: see text], ΔT′ = 45 °C, and δ = 0.03 cm (where [Formula: see text] is the temperature of the graphite heating the substrate; ΔT′, the temperature difference between the graphite heating the source and the one heating the substrate; and δ, the thickness of the spacer separating the GaAs source and the substrate). Mirrorlike epitaxial layers of GaAs are obtained with these parameters when water vapor, at a partial pressure of 4.58 Torr (1 Torr = 133.3 Pa), is introduced with H2 at the beginning of the temperature rise of the reactor. The dimensions of the epitaxial layer are only limited by the size of the reactor. Using the same growth conditions, it is not possible to obtain mirrorlike films of GaAs on (100) Ge substrates. Instead, the layers are dull grey (sample no. 1). It is however not a polycrystalline deposition since the pole figures, obtained by X-ray diffraction, reveal only four crystallographic orientations; {100} the main one, {221} the secondary one, and {021} + {112} two minor contributions. Mirrorlike films of GaAs on (100) Ge substrates of less than 1 cm2 have been obtained with [Formula: see text], ΔT′ = 25 °C, and δ = 0.03 cm. With these conditions, the growth rate is 0.25 ± 0.08 μm min−1. The time evolution of [Formula: see text] and ΔT′, from room temperature up to the equilibrium temperature also influences the surface morphology of GaAs films on Ge while this was not the case for GaAs films on GaAs substrates. When the Ge substrate is larger than 1 cm2, the centre of the film becomes textured but the edges remain mirrorlike (sample no. 2). Pole figures obtained for the center and the edges of sample no. 2 are similar. They are characterized by one large diffraction due to the {100} orientation. A few random crystallographic orientations and sometimes the {221} orientation, however, bearly emerge from the background of these pole figures. Also transmission electron microscopy does not reveal any major difference between the center and the edges of sample no. 2. The density of threading dislocations is the same for both regions, varying from 108 cm−2, close (2–3 μm) to the interface, to 107 cm−2 in the thickness of the film. No misfit dislocations were observed. Antiphase boundaries are present in both regions as well. The only difference between the centre and the edges of sample no. 2 involves microtwin bundles: in the center region, there are two microtwin bundles per micrometre of interface, extending up to 6 μm in the GaAs film while on the edges, there is one bundle per micrometre with an extension of only one micrometre into the epitaxial layer. Mirrorlike GaAs films can be obtained on (100) Ge substrates of at least 1 in (1 in = 2.5 cm) in diameter by increasing δ to 0.2 cm and by injecting water vapor in the reactor only when [Formula: see text] reached 650 °C; the other deposition parameters remain the same as for sample no. 2. In these conditions, the growth rate of GaAs is 0.075 ± 0.020 μm min−1. By using a transport model based on thermodynamics, it is demonstrated that the flux intensity of GaAs transported from the source to the substrate, as well as the eventual presence of GeO as a nucleation site for GaAs on Ge, are both important for the morphology of the epitaxial layer.


MAUSAM ◽  
2021 ◽  
Vol 68 (2) ◽  
pp. 335-348
Author(s):  
YOUNES KHOSRAVI ◽  
HASAN LASHKARI ◽  
HOSEIN ASAKEREH

Recognitionanddetectionofclimaticparameters inhave animportant role inclimate change monitoring. In this study, the analysis of oneofthe most importantparameters, water vapor pressure (WVP), was investigated. For this purpose, two non-parametric techniques, Mann-Kendall and Sen's Slope Estimator, were used to analyze the WVP trend and to determine the magnitude of the trends, respectively. To analyze these tests, ground station observations [10 stations for period of 44 years (1967-2010)] and gridded data [pixels with the dimension of 9 × 9 km over a 30-year period (1981-2010)] in South and SouthwestofIran were used. By programming in MATLAB software, the monthly, seasonal and annual WVP time series were extracted and MK and Sen's slope estimator tests were done. The results of monthly MK test on ground station observations showed that the significant downward trends are more considerable than significant upward trends. It also showed that the WVP highest frequency was more in warm months, April to September and the highest frequency of significant trends slope was in February and May. The spatial distribution of MK test of monthly gridded WVP time series showed that the upward trends were detected mostly in western zone and near the Persian Gulf in August. On the other hand, the downward trends through months. The maximum and minimum values of positive trends slope occurred in warm months and cold months, respectively. The analysis of the MK test of the annual WVP time series indicated the upward significant trends in the southeast and southwest zones of study area.  


2009 ◽  
Vol 131 (3) ◽  
Author(s):  
Pamela L. Dickrell ◽  
N. Argibay ◽  
Osman L. Eryilmaz ◽  
Ali Erdemir ◽  
W. Gregory Sawyer

Microtribological measurements of a hydrogenated diamondlike carbon film in controlled gaseous environments show that water vapor plays a significant role in the friction coefficient. These experiments reveal an initial high friction transient behavior that does not reoccur even after extended periods of exposure to low partial pressures of H2O and O2. Experiments varying both water vapor pressure and sample temperature show trends of a decreasing friction coefficient as a function of both the decreasing water vapor pressure and the increasing substrate temperature. Theses trends are examined with regard to first order gas-surface interactions. Model fits give activation energies on the order of 40 kJ/mol, which is consistent with water vapor desorption.


Sign in / Sign up

Export Citation Format

Share Document