doi:10.1016/j.mejo.2005.02.127 
Copyright © 2005 Elsevier Ltd All rights reserved.
In depth study of the compensation in annealed heavily carbon doped GaAs
A. Rebey
, 
, W. Fathallah and B. El Jani
Faculté des Sciences, Unité de Recherche sur l'Hétéroépitaxie et Applications, 5000 Monastir, Tunisia
Received 19 October 2004;
revised 16 February 2005;
accepted 19 February 2005.
Available online 3 May 2005.
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
Abstract
Heavily C-doped GaAs epitaxial layers with holes concentrations ranging from 1019 to 1.6×1020 cm−3 have been grown by metal organic vapour phase epitaxy (MOVPE) using CCl4 as C-growth precursor. The carbon doping characteristics of GaAs epilayers have been investigated by optimizing the V/III ratio and the growth temperature. Additional informations have been extracted from the evolution of the in situ reflectivity signal during the growth of GaAs:C. The appearance of discernible oscillations in the reflectivity response indicates the high carbon incorporation and the good surface quality in spite of the CCl4 etching effect. The hole concentration tends to saturate at about 1.5×1020 cm−3. The comparison between Hall effect measurements realized on sets of as grown and annealed layers, and theoretical calculations of the mobility lead to the determination of the compensation ratio of the samples.
The lattice matching conditions were systematically investigated by using high X-ray diffraction measurements from (004) and (115) planes. A comparison between the experimental mismatch and the one calculated with the Vegard's law allows the estimation of the possible origin of the compensation. Secondary ion mass spectrometry, scanning electron microscopy and atomic force microscopy have been used as complementary tools to characterize the films.
Keywords: C-doped GaAs; Compensation; MOVPE
PACS: 71.55.Eq; 81.15.Gh; 82.60.Cx
Fig. 1. Reflectivity signals recorded during the heteroepitaxy of AlAs/GaAs:C for two growth temperatures TG=550 and 580 °C and for identical otherwise experimental conditions. We also show the SEM image of the layer grown at TG=580 °C.
Fig. 2. In the left, the signal reflectivity normalized on substrate recorder during homoepitaxy of GaAs/GaAs:C for different partial pressures of CCl4. The top and the middle curves correspond to layers having hole concentrations of 1020 and 1.3×1020 cm−3, respectively, the bottom curve corresponds to sample G. A theoretical attempt to reproduce the experimental curves and to deduce some physical parameters of epitaxial layers is made. 
R
is the average value of normalized reflectivity of GaAs:C. In the right, AFM topogram of the high carbon doped GaAs (sample G).
Fig. 3. The hole concentration as a function of the partial pressure of CCl4. For a fixed temperature TG=550 °C and when varying the V/III ratio, or for fixed V/III ratio (24.7) and three temperatures (decrease of temperature is in the sense of row) 562, 550 and 535 °C. Solid line corresponds to linear variation with slope equal 1. The dotted lines correspond to fitting function pH=psat(PCCl4/(kPAsH3+PCCl4)) for psat=1.5×1020 cm−3 and three values of k.
Fig. 4. The theoretical hole mobility as a function of the hole concentration calculated for different compensation ratio when considering μlat=435 cm2 V−1 s−1 (dot line) and μlat (solid line) described by the empirical relation in the text. For the two calculations, the compensation ratio is varied from 0 to 0.5 by 0.1 step. We also show bibliographic values having a known compensation ratio (about zero) used to calibrate the calculation. In the same figure, we observe the values of our work.
Fig. 5. (004) peak separation ΔΘ(004) as a function of the azimuthal angle for G, C, D and A layers. The solid line corresponds to sinusoidal fit. The dotted line corresponds to the average values sinusoidal fit ΔΘavr(004).
Table 1.
The dependence of μlat as a function of the hole concentrations after various references [22], [26], [29] and [31]
Table 2.
Basic characteristics of our samples and the results of experimental measurements (Hall effect and HR-XRD) and semi-theoretical calculations
θH is the compensation ratio deduced from Hall mobility. (−Δa/a)
(calcd) is the calculated mismatch by using different origins of compensation. (−Δa/a) (exp) is the experimental mismatch deduced from HR-XRD measurements. Θ* is the experimental compensation ratio deduced from hole concentration and SIMS measurements.
Abstract
Purchase PDF (261 K)
E-mail Article
Add to my Quick Links
Cited By in Scopus (0)
1020 cm−3) the ratio 
Purchase PDF (378 K)
