HgCdTe quantum wells grown by molecular beam epitaxy
CdxHg₁₋xTe-based (x = 0 – 0.25) quantum wells (QWs) of 8 – 22 nm in thickness were grown on (013) CdTe/ZnTe/GaAs substrates by molecular beam epitaxy. The composition and thickness (d) of wide-gap layers (spacers) were x ∼ 0.7 mol.frac. and d ∼ 35 nm, respectively, at both sides of the quantum we...
Збережено в:
| Дата: | 2007 |
|---|---|
| Автори: | , , , , , , , |
| Формат: | Стаття |
| Мова: | English |
| Опубліковано: |
Інститут фізики напівпровідників імені В.Є. Лашкарьова НАН України
2007
|
| Назва видання: | Semiconductor Physics Quantum Electronics & Optoelectronics |
| Онлайн доступ: | http://dspace.nbuv.gov.ua/handle/123456789/118333 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Цитувати: | HgCdTe quantum wells grown by molecular beam epitaxy / S.A.Dvoretsky, D.G.Ikusov. Z.D.Kvon, N.N.Mikhailov, V.G.Remesnik, R.N.Smirnov, Yu.G.Sidorov, V.A.Shvets // Semiconductor Physics Quantum Electronics & Optoelectronics. — 2007. — Т. 10, № 4. — С. 47-53. — Бібліогр.: 25 назв. — англ. |
Репозитарії
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-118333 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-1183332017-05-30T03:03:07Z HgCdTe quantum wells grown by molecular beam epitaxy Dvoretsky, S.A Ikusov, D.G. Kvon, Z.D. Mikhailov, N.N. Remesnik, V.G. Smirnov, R.N. Sidorov, Yu.G. Shvets, V.A. CdxHg₁₋xTe-based (x = 0 – 0.25) quantum wells (QWs) of 8 – 22 nm in thickness were grown on (013) CdTe/ZnTe/GaAs substrates by molecular beam epitaxy. The composition and thickness (d) of wide-gap layers (spacers) were x ∼ 0.7 mol.frac. and d ∼ 35 nm, respectively, at both sides of the quantum well. The thickness and composition of epilayers during the growth were controlled by ellipsometry in situ. It was shown that the accuracy of thickness and composition were ∆x = ± 0.002, ∆d = ± 0.5 nm. The central part of spacers (10 nm thick) was doped by indium up to a carrier concentration of ∼10¹⁵ cm⁻³ . A CdTe cap layer 40 nm in thickness was grown to protect QW. The compositions of the spacer and QWs were determined by measuring the Е₁ and Е₁+∆₁ peaks in reflection spectra using layer-by-layer chemical etching. The galvanomagnetic investigations (the range of magnetic fields was 0 – 13 T) of the grown QW showed the presence of a 2D electron gas in all the samples. The 2D electron mobility µe = (2.4 – 3.5)×10⁵ cm² /(V·s) for the concentrations N = (1.5 – 3)×10¹¹ cm⁻² (x < 0.11) that confirms a high quality of the grown QWs. 2007 Article HgCdTe quantum wells grown by molecular beam epitaxy / S.A.Dvoretsky, D.G.Ikusov. Z.D.Kvon, N.N.Mikhailov, V.G.Remesnik, R.N.Smirnov, Yu.G.Sidorov, V.A.Shvets // Semiconductor Physics Quantum Electronics & Optoelectronics. — 2007. — Т. 10, № 4. — С. 47-53. — Бібліогр.: 25 назв. — англ. 1560-8034 PACS 07.60.Fs, 73.43.-f, 75.75.+q, 78.30.Fs, 81.05.Dz, 81.15.Hi http://dspace.nbuv.gov.ua/handle/123456789/118333 en Semiconductor Physics Quantum Electronics & Optoelectronics Інститут фізики напівпровідників імені В.Є. Лашкарьова НАН України |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| language |
English |
| description |
CdxHg₁₋xTe-based (x = 0 – 0.25) quantum wells (QWs) of 8 – 22 nm in
thickness were grown on (013) CdTe/ZnTe/GaAs substrates by molecular beam epitaxy.
The composition and thickness (d) of wide-gap layers (spacers) were x ∼ 0.7 mol.frac.
and d ∼ 35 nm, respectively, at both sides of the quantum well. The thickness and
composition of epilayers during the growth were controlled by ellipsometry in situ. It
was shown that the accuracy of thickness and composition were ∆x = ± 0.002, ∆d =
± 0.5 nm. The central part of spacers (10 nm thick) was doped by indium up to a carrier
concentration of ∼10¹⁵ cm⁻³
. A CdTe cap layer 40 nm in thickness was grown to protect
QW. The compositions of the spacer and QWs were determined by measuring the Е₁ and
Е₁+∆₁ peaks in reflection spectra using layer-by-layer chemical etching. The galvanomagnetic
investigations (the range of magnetic fields was 0 – 13 T) of the grown QW
showed the presence of a 2D electron gas in all the samples. The 2D electron mobility
µe = (2.4 – 3.5)×10⁵
cm²
/(V·s) for the concentrations N = (1.5 – 3)×10¹¹ cm⁻² (x < 0.11)
that confirms a high quality of the grown QWs. |
| format |
Article |
| author |
Dvoretsky, S.A Ikusov, D.G. Kvon, Z.D. Mikhailov, N.N. Remesnik, V.G. Smirnov, R.N. Sidorov, Yu.G. Shvets, V.A. |
| spellingShingle |
Dvoretsky, S.A Ikusov, D.G. Kvon, Z.D. Mikhailov, N.N. Remesnik, V.G. Smirnov, R.N. Sidorov, Yu.G. Shvets, V.A. HgCdTe quantum wells grown by molecular beam epitaxy Semiconductor Physics Quantum Electronics & Optoelectronics |
| author_facet |
Dvoretsky, S.A Ikusov, D.G. Kvon, Z.D. Mikhailov, N.N. Remesnik, V.G. Smirnov, R.N. Sidorov, Yu.G. Shvets, V.A. |
| author_sort |
Dvoretsky, S.A |
| title |
HgCdTe quantum wells grown by molecular beam epitaxy |
| title_short |
HgCdTe quantum wells grown by molecular beam epitaxy |
| title_full |
HgCdTe quantum wells grown by molecular beam epitaxy |
| title_fullStr |
HgCdTe quantum wells grown by molecular beam epitaxy |
| title_full_unstemmed |
HgCdTe quantum wells grown by molecular beam epitaxy |
| title_sort |
hgcdte quantum wells grown by molecular beam epitaxy |
| publisher |
Інститут фізики напівпровідників імені В.Є. Лашкарьова НАН України |
| publishDate |
2007 |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/118333 |
| citation_txt |
HgCdTe quantum wells grown by molecular beam epitaxy / S.A.Dvoretsky, D.G.Ikusov. Z.D.Kvon, N.N.Mikhailov, V.G.Remesnik, R.N.Smirnov, Yu.G.Sidorov, V.A.Shvets // Semiconductor Physics Quantum Electronics & Optoelectronics. — 2007. — Т. 10, № 4. — С. 47-53. — Бібліогр.: 25 назв. — англ. |
| series |
Semiconductor Physics Quantum Electronics & Optoelectronics |
| work_keys_str_mv |
AT dvoretskysa hgcdtequantumwellsgrownbymolecularbeamepitaxy AT ikusovdg hgcdtequantumwellsgrownbymolecularbeamepitaxy AT kvonzd hgcdtequantumwellsgrownbymolecularbeamepitaxy AT mikhailovnn hgcdtequantumwellsgrownbymolecularbeamepitaxy AT remesnikvg hgcdtequantumwellsgrownbymolecularbeamepitaxy AT smirnovrn hgcdtequantumwellsgrownbymolecularbeamepitaxy AT sidorovyug hgcdtequantumwellsgrownbymolecularbeamepitaxy AT shvetsva hgcdtequantumwellsgrownbymolecularbeamepitaxy |
| first_indexed |
2025-07-08T13:49:56Z |
| last_indexed |
2025-07-08T13:49:56Z |
| _version_ |
1837086904779014144 |
| fulltext |
Semiconductor Physics, Quantum Electronics & Optoelectronics, 2007. V. 10, N 4. P. 47-53.
© 2007, V. Lashkaryov Institute of Semiconductor Physics, National Academy of Sciences of Ukraine
47
PACS 07.60.Fs, 73.43.-f, 75.75.+q, 78.30.Fs, 81.05.Dz, 81.15.Hi
HgCdTe quantum wells grown by molecular beam epitaxy
S.A. Dvoretsky, D.G. Ikusov, Z.D. Kvon, N.N. Mikhailov, V.G. Remesnik,
R.N. Smirnov, Yu.G. Sidorov, V.A. Shvets
A.V. Rzhanov Institute of Semiconductor Physics of the Siberian Branch of Russian Academy of Sciences
13, Acad. Lavrent’ev Ave., 630090 Novosibirsk, Russia
Phone, fax: +7-383-330-49-67.
E-mail: dvor@isp.nsc.ru
Abstract. CdxHg1-xTe-based (x = 0 – 0.25) quantum wells (QWs) of 8 – 22 nm in
thickness were grown on (013) CdTe/ZnTe/GaAs substrates by molecular beam epitaxy.
The composition and thickness (d) of wide-gap layers (spacers) were x ∼ 0.7 mol.frac.
and d ∼ 35 nm, respectively, at both sides of the quantum well. The thickness and
composition of epilayers during the growth were controlled by ellipsometry in situ. It
was shown that the accuracy of thickness and composition were ∆x = ± 0.002, ∆d =
± 0.5 nm. The central part of spacers (10 nm thick) was doped by indium up to a carrier
concentration of ∼1015 cm−3. A CdTe cap layer 40 nm in thickness was grown to protect
QW. The compositions of the spacer and QWs were determined by measuring the Е1 and
Е1+∆1 peaks in reflection spectra using layer-by-layer chemical etching. The galvano-
magnetic investigations (the range of magnetic fields was 0 – 13 T) of the grown QW
showed the presence of a 2D electron gas in all the samples. The 2D electron mobility
µe = (2.4 – 3.5)×105 cm2/(V·s) for the concentrations N = (1.5 – 3)×1011 cm−2 (x < 0.11)
that confirms a high quality of the grown QWs.
Keywords: molecular beam epitaxy, mercury cadmium telluride HgCdTe, quantum well,
ellipsometry, infrared spectroscopy, quantum Hall effect.
Manuscript received 20.11.07; accepted for publication 19.12.07; published online 31.01.08.
1. Introduction
The interest to nanostructures based on II-VI narrow-gap
semiconductors containing mercury (such as HgTe/CdTe
superlattices) was due to their advantages in comparison
with mercury cadmium telluride (MCT) alloys on the
fabrication of high-quality multiple-unit infrared (IR)
photodetectors firstly from the point of view of uni-
formity of a spectral response [1]. However, the first
growth experimental results did not lead to high-quality
nanostructures [2]. The development of molecular beam
epitaxy (MBE) technique in last years allows one to
grow nanostructures suitable for the development and
fundamental investigations of devices. The authors in [3]
demonstrated the growth of doped HgTe/HgCdTe
superlattices which were used for the fabrication of very
long wavelength IR photodetectors (cut-off wavelength
over 20 µm). Microcavity Cd0.65Hg0.35Te/Cd0.36Hg0.64Te
(350 nm / 250 nm in thickness) structures were grown by
MBE and used to fabricate 3−3.5 µm IR photo-
emitters [4].
The fundamental characteristics of HgTe-based
quantum wells (QWs) were intensively studied (see [5-
8]). It was shown that the electron effective mass in
HgTe-based QWs is lower than that in QWs of III-V
compounds. This leads to a wider splitting of Landau
levels, weaker electron localization, increasing the
amplitude of Shubnikov−de Haas oscillations, and
increasing the Rashba effect [5]. HgTe-based QWs with
thicknesses over than 5 nm are characterized by the
inverted conduction band (the band of light holes becomes
similar to the conduction band) that is undoubtedly
interesting for studying the electron behavior in such a
system. It is interesting also the behavior of electrons in
HgCdTe-based QWs at small Cd mole fractions. From
fundamental considerations, this should lead to decreasing
the band gap (up to a zero-gap semiconductor) with
increase in the Cd content and, as a consequence, to
changing the thickness, at which the band inversion
occurs. So it is of interest to investigate HgCdTe-based
QWs at increasing the Cd content beyond a zero-gap
semiconductor, at which no band inversion occurs.
Semiconductor Physics, Quantum Electronics & Optoelectronics, 2007. V. 10, N 4. P. 47-53.
© 2007, V. Lashkaryov Institute of Semiconductor Physics, National Academy of Sciences of Ukraine
48
Fig. 1. Scheme of a CdxHg1-xTe-based quantum well.
At the growth of nanostructures, the precise control
over the MCT composition and thickness is very
important. However, there are no data in the literature
about providing such a control at the growth of QWs.
The layers thickness control at the growth of QWs
on the basis of III-V compounds in Si-Ge system is
carried out by measuring the reflection high energy
electron diffraction (RHEED) oscillations. However,
using RHEED at the growth of HgCdTe-based QWs is
impossible for two reasons. First, the electron beam with
an energy over than 15 keV essentially influences the
growth process, which leads to variations in the surface
morphology and MCT composition. Second, the
RHEED oscillations are observed only during the
growth on singular (001) orientations. However, because
of technological reasons, the growth of MCT alloys is
usually carried out on the vicinal (112)B, (013)
orientations of substrates.
We showed a very effective use of ellipsometry for
the growth of MCT heteroepitaxial structures (HS). This
technique allows one to provide a high-precision control
in situ of the MCT composition, thickness, and growth
rates. This allows one to grow MCT HS with a special
MCT composition profile throughout the thickness for
different IR detectors [9-11]. The use of ellipsometry for
the control over the nanolayers growth was firstly shown
in [12]. The author in [13] presented the calculations of
ellipsometric parameters changing at the HgTe/CdTe
superlattices growth that ensures their possible effective
use in practice. The ellipsometry control over the growth
of MCT-based potential barriers, wells, and periodical
structures with 100 nm in thickness was carried out in
[14]. There, the possibility to grow nanostructures
ellipsometrically controlled in situ at growth rates up to
3.5 µm/h was demonstrated. The accuracies of the
determination of the MCT composition and thickness
were ∆x ∼ ± 0.002 mol.frac. and ∆d ∼ 1 monolayer,
respectively.
The aim of the present paper is to present
the results of the growth of Hg1-хCdхTe-based QWs (x =
0–0.25) with precise ellipsometric control of the MCT
composition and thickness in situ on (013)
CdTe/ZnTeGaAs substrates and their magnetogalvanic
investigation.
2. Experimental results and discussion
An MCT-based QW structure is shown schematically in
Fig. 1. The QW growth was carried out on the MBE
equipment of the “Ob-M”-type [11]. The buffer layers of
ZnTe (0.3 µm) and sequence CdTe (5 – 7 µm) were
grown on the atomic clean surface of (013) GaAs
substrate preparing by chemical etching and thermal
annealing in the As flux in an ultra vacuum chamber
[15]. Then the growth of an MCT-based (x = 0–0.25)
QW with a spacer (x ∼ 0.7) was carried out on the CdTe
buffer layer surface at a temperature of 180-190 °C. The
doping of the central part of a spacer up to the carrier
concentration 1015 cm−3 was realized in situ with the use
of an In conventional source [16]. The growth rate was
0.15 µm/h. The MCT composition during the growth
was varied by a cadmium flux. An ultra fast single-beam
(λ = 0.6328) automatic ellipsometer LEF-755 (UFE)
based on the original static scheme [17] was used for the
control over the epilayer buffer thickness and the MCT
composition and thickness.
At the epitaxial growth of a layer on the substrate
with different optical constants, the decaying oscillations
of the ellipsometric parameters ψ and ∆ were observed.
The number of oscillations depends on the absorption
coefficient of a growing layer and decreases with
increase in its value. The oscillation amplitude is
determined by the difference of the absorption
coefficients of the substrate and the growing layer. One
period of oscillations corresponds to a 100-nm MCT
layer. At the growth of multilayer structures with
thicknesses lower than 100 nm, one should observe
sectionally smooth curves which correspond to the
growth of MCT layers with constant composition. The
length of a sectionally smooth curve corresponds to the
thickness of a growing layer.
The evolution of the ellipsometric parameters ∆
and ψ during the growth of HgTe-based QWs is shown
in Fig. 2. It is represented as a sectionally smooth curve
in the ψ-∆ plane. Smooth sections correspond to the
constant MCT composition of layers. Their length
determines the layer thickness. The initial point O
corresponds to the ellipsometric parameters of the CdTe
surface. The fabrication of a QW begins from the growth
Semiconductor Physics, Quantum Electronics & Optoelectronics, 2007. V. 10, N 4. P. 47-53.
© 2007, V. Lashkaryov Institute of Semiconductor Physics, National Academy of Sciences of Ukraine
49
Fig. 2. Evolution of the ellipsometric parameters ∆ and ψ
during the growth of a HgTe-based QW. The points at
breaks of the curve (O-C) correspond to the initial stage of
growth of the following QW layer. Dots – experimental
data; solid line and circles (through 1 nm) − calculated data.
In the inset: the MCT composition versus the thickness.
Fig. 3. Evolution of ellipsomentric parameters ∆ and ψ at
the wide-gap layer growth on the CdTe surface. Rhombs –
experimental data. Lines 1-3 – calculated MCT composition
(1 − xCdTe = 0.75, 2 − 0.735, and 3 − 0.72) The transverse
line to O-A – lines of equal thicknesses for different MVT
compositions.
of a wide-gap layer (x ∼ 0.7) (curve O-A) after the
opening of the tellurium and cadmium source shutters.
The doping of the central part of this layer is carried out
by indium after the opening at O′ and the closing at A′ of
the indium source shutter. Then the growth of a wide-
gap layer is continued up to point A. In the insert, the
sectors O-O′, O′-A′, and A′-A correspond to the un-
doped wide-gap layer with thickness d1, the In-doped
wide-gap layer with thickness d2, and the undoped wide-
gap layer with thickness d3. After the closing of the
cadmium source shutter, the HgTe layer is growing
between A-B points of the curve with thickness dQW.
The wide-gap layer (see B-C curve) is growing in
manner that is analogous to the growth of the first wide-
gap layer (O-A curve) after the opening of the cadmium
source shutter. The thicknesses d4, d6, and d5 correspond
to undoped and In-doped wide-gap layers. The grown
QW structure is covered by a CdTe cap layer with
dCdTe ∼ 40 nm.
The dots on the curve in Fig. 2 are experimental
data which were measured at 1s-intervals. The circles
correspond to the values of ellipsometric parameters
calculated at 1-nm intervals. The calculations were
carried out by using the one-layer model [18]
analogously to [19]. The calibration curve and the
optical constants for calculations were taken for different
MCT compositions from [20, 21]. To improve the
comparison of the experimental and calculated data, it
was necessary to know the optical constants of the
substrate and the growing layers at incident light
wavelengths and the growth temperature with high
accuracy. The MCT optical constants for different
compositions were obtained from the spectral
measurements at the room or a lower temperature
[22, 23]. But these data cannot be used for calculating
the variation of ellipsometric parameters during the
growth of a QW which occurs at much higher
temperatures. Thus, we used the dependences of the
optical constants [n (x) and k (x)] on the MCT
composition which were measured at room temperature
[20] and then calculated them for temperatures of the
growth using the experimentally determined
thermooptical coefficients. The experimental data are in
good agreement with the results of calculations (Fig. 2)
within this procedure.
The procedure of determination of the MCT
composition and the layer thickness for the growth of a
wide-gap layer on the CdTe surface is shown in more
details in Fig. 3. The point O corresponds to the
ellipsometric parameters of the CdTe surface. After the
start of the growth, a smooth variation of the
ellipsometric parameters ψ and ∆ is observed (O-A
sector). After point A, the growth of a HgTe layer is
running. Lines 1-3 are the calculated data for the growth
of MCT layers with different MCT compositions
x = 0.75 (1), 0.735 (2), and 0.72 (3) and the following
parameters: the angle of incident light – 67.9°±0.05°;
nCdTe = 3.003; kCdTe = 0.204; nHgTe = 4.08, and
kHgTe = 1.16. These parameters were determined by
measuring the ellipsometric parameters during the
growth of an epilayer with constant MCT composition.
The thickness of the growing wide-gap layer is
determined by the length of О-А curve. The thick
transverse lines to O-A curve mean the line of equal
thickness for different MCT compositions. It is clear that
the accuracy of the MCT composition and the layer
thickness is determined by the coincidence of
experimental and calculated curves and the accuracy of
3
Semiconductor Physics, Quantum Electronics & Optoelectronics, 2007. V. 10, N 4. P. 47-53.
© 2007, V. Lashkaryov Institute of Semiconductor Physics, National Academy of Sciences of Ukraine
50
the position of the inflection point. For the data in Fig. 2,
the MCT composition was determined using the above-
mentioned procedure as x = 0.735 ± 0.002. The upper
bound of the thickness determination accuracy does not
exceed 0.5 nm, i.e., 1 monolayer. In an analogous
manner, we performed the analysis of other parts of the
curve in Fig. 2.
Thus, we determined the MCT compositions and
the thickneses of a HgTe-base QW. The determination
accuracies of the MCT composition and the thickness
were ∆x ∼ 0.002 and 0.5 nm, respectively.
The MCT compositions and thicknesses of layers
of seven HgCdTe-based QWs are given in Table.
The MCT composition distribution throughout the
thickness of a HgCdTe-based QW was measured by
reflection spectra in the range 1.5 – 6 eV with layer-by-
layer chemical etching in a 0.05 % Br-HBr solution.
The thickness of a layer removed by chemical
etching was determined by the shifts of the interference
maxima of transmission spectra in the range 1600 –
2300 cm−1. The measurements of the transmission
spectra were carried out using an IR Fourier-
spectrometer “Infralum FT-801” (2 – 25 µm). The
positions of interference maxima are described by the
relation 2nd = m×104/ν, where n – refraction index, d –
thickness in µm, m – maximum order, and ν –
wavenumber in cm−1. It was easy to evaluate that a 1-nm
variation of the CdTe thickness leads to a 0.3-cm−1 shift
of interferential maxima for 5.5-µm CdTe, and the
refraction index nCdTe = 2.8 (ν = 1780 cm−1). The
accuracy of the determination of a position of inter-
ference maxima with the use of an “Infralum FT-801” is
lower than 0.01 cm−1. Fig. 4 shows the shifts of
interference maxima during the removal of a layer of
5486.8 nm in thickness at QW N 221 (see Table) under
chemical etching. After the removal of 31.9-nm layers,
the interference maxima shift to thicknesses of 5454.9
and 5430.0 nm, respectively. The thickness deter-
mination accuracy is lower than 0.1 nm.
The measurement of the reflection spectra of
narrow-gap semiconductors (1.5-6 eV range) allows one
to determine the shape of E1 and E1+∆1 peaks and their
energy positions. These peaks are related to the critical
points of the density of states for direct band gap
transitions at energies over the band gap. For MCT
compounds, the positions of peaks depend on the MCT
composition, which allows one to determine x with
accuracy ∼ 0.5 % [24]. Some formulas for the
dependence of E1 and E1+∆1 peaks on the MCT
composition are available in the literature [24, 25]. In the
present work, we have measured the reflection spectra at
the room temperature to determine the position of the E1
and E1+∆1 peaks for MCT epitaxial layers grown by
MBE with a constant MCT composition (x = 0 – 1).
These data are described by the following expressions:
E1 = 2.0944 + 0.5869x + 0.6295x2, (1)
E1 + ∆1 = 2.7441 + 0.4895x + 0.6895x2 . (2)
Fig. 4. Position of inference maxima during chemical etching
of QW N 221 with an initial thickness of 5486.8 nm. The
numbers show the thicknesses measured after a sequence of
chemical etchings.
These dependences were used for the structure
composition determination.
The reflection spectra of HgCdTe-based QWs were
measured at room temperature using a “SPECORD UV-
VIS” equipped by reflection attachments alternating with
layer-by-layer chemical etching in a 0.05 % Br-HBr
solution. The reflection from an Al mirror was used as
the reference signal. Fig. 5 shows the reflection spectrum
of QW N 221 of the initial surface (curve 1) and that
after three steps of chemical etching (curves 2-4). The
peak at 3.31 eV (curve 1) corresponds to the E1
transition in CdTe that is in agreement with [24] and
with data for bulk CdTe (curve 1′). It seems that the
peak near 2.2 eV is related to the interference of the
incident light and the light reflected from QW interfaces.
After removing a 32-nm layer (curve 2), the peaks at
2.75 and 3.37 eV correspond to the E1 and E1+∆1
transitions for the MCT composition x ∼ 0.67. Then,
after removing a 57-nm layer (curve 3), the reflection
spectrum is essentially changed. The peak at the 3.0 –
3.5-eV interval disappears. Apparently, the peaks at 2.11
and 2.76 eV correspond to the E1 and E1+∆1 transitions
for the MCT composition x ∼ 0.03. Finally, after
removing a 89-nm layer (curve 4), peaks at 2.66 and
3.27 eV correspond to the E1 and E1+∆1 transitions for
the MCT composition x ∼ 0.67 that is similar to that
observed on curve 2. After removing the QW, the
reflection spectrum corresponds to the buffer CdTe layer
(curve 1′).
The results of measurements of the MCT
composition by ellipsometry in situ and by reflection
spectra with layer-by-layer chemical etching are in a
good agreement. Really, we need to take into account
the influence of thin layers with different compositions
on the position of peaks because of the penetration of
incident light through these layers and the interference
effects.
For all the grown samples (Table, except for
N 219), the magnetransport measurements were carried
out at temperatures of 1.6 – 4.2 K at a magnetic field up
to 13 T using standard Hall bars (50 µm in width, and
Semiconductor Physics, Quantum Electronics & Optoelectronics, 2007. V. 10, N 4. P. 47-53.
© 2007, V. Lashkaryov Institute of Semiconductor Physics, National Academy of Sciences of Ukraine
51
Table. Parameters of grown QW samples.
MCT composition of a
wide-gap layer, xCdTe
MCT
composition
of QW, x
MCT composition of a
wide-gap layer, xe
Thickness
of a
CdTe cap
layer
N
d1,
nm
d2, nm d3, nm DQW,
nm
d4, nm d5, nm d6, nm dCdTe, nm
MCT
composition
of QW, x,
from
reflection
spectra
Carrier
concentra-
tion,
1011 cm−2
Mobility,
cm2/(V⋅s)
0.73 0.06 0.63 218
9 12.3 10.2 20.3 9.1 10.2 8.2
34
−
2
24
0.79 0.24 0.80 219
8.4 14.3 8.8 12.5 7.5 14.7 4.8
34
0.05
−
−
0.70 0.16 0.7 220
8.8 10.9 9.8 10.5 10 12 7.9
38
−
1.6
8.6
0.72 0.11 0.69 221
6.1 13.9 11.4 16.2 12.4 15.6 5.0
44
0.03
3
30
0.68 0.03 0.66 330
7.1 6.4 11.2 18.4 8.6 9.5 5.0
46
0.00
2.2
35
0.7 0.06 0.65 331
8.6 12.8 11.3 21.7 8.0 10.0 4.8
39
−
1.5
24
the 100-µm interval between potentiometric contacts)
fabricated by means of photolithography. Indium ohmic
contacts were formed by thermal compression. The
electron concentrations and mobilities determined from
these measurements at 4.2 K are presented in Table. The
high value of mobility µe = (2.4−3.5)×105 cm2/(V·s) for
electron concentrations Ns = (1.5−3)×1011 cm−2 reveals a
high quality of grown QWs. The investigations of Hall
structures made on different areas of the plate show
identical results, which testifies to a high 2D uniformity
of samples. The results obtained for samples N 218 and
331 confirm a high reproducibility of the QW growth
technology. The identical mobility (µe =
2.4×105 cm2/(V·s)) and close concentrations (Ns =
2×1011 cm−2 and Ns = 1.5×1011 cm−2) were obtained for
QWs with similar compositions and thicknesses
(x = 0.06 and d ≈ 21 nm). The samples were grown on
the same MBE system with a time interval of several
months.
Fig. 5. Longitudinal (ρxx) and Hall (ρxy) components of
magnetoresistance versus the magnetic field (B) for 2DEG in a
21-nm QW.
In the quantum Hall effect regime, the well-
pronounced plateaus and minima are observed in all the
examined structures. As an example, the longitudinal
dissipative (ρxx) and Hall (ρxy) components of
magnetoresistance versus the magnetic field (B) are
presented in Fig. 5 for the sample N 330. The wide
plateaus in ρxy(B) and the corresponding wide minima in
ρxx(B) are clearly observed even at 1.6 K. It is due to a
small value of the effective mass of electrons in the two-
dimensional electron gas (2DEG) in a HgTe QW and,
respectively, a large distance between Landau levels.
The more extensive investigation of the quantum Hall
effect in the described samples demonstrates the
presence of critical points which are some additional
evidence for a high quality of the grown HgTe QW.
Such a behavior of magnetoresistances is typical of
2DEG.
3. Conclusion
The narrow-gap CdxHg1-xTe-based (x = 0–0.25) QWs of
10.5−22 nm in thickness were grown on GaAs (013)
substrates with precise in situ ellipsometric control over
the MCT composition and thickness. The determination
accuracies of the MCT composition and thickness were
∼ ±0.002 mol. frac. and ∆d ∼ 0.5 nm, respectively.
The measurements of reflection spectra with layer-
by-layer chemical etching qualitatively confirmed the
results of in situ ellipsometric measurements.
We have demonstrated experimentally the
reproducible process of growth of HgCdTe-based QWs.
We have obtained 2DEG with high values of the
electron mobility µe = (2.4−3.5)×105 cm2/(V·s) for con-
centrations Ns = (1.5−3)×1011 cm−2 observed for all the
grown QW structures, which indicates a high quality of
Semiconductor Physics, Quantum Electronics & Optoelectronics, 2007. V. 10, N 4. P. 47-53.
© 2007, V. Lashkaryov Institute of Semiconductor Physics, National Academy of Sciences of Ukraine
52
QW. The width of dissipative resistance minima and the
extensive Hall quantization plateaus testify to a low
effective mass of electrons in 2DEG.
Acknowlegements
The work is partially supported by a complex integration
project of SB RAS N 3.20.
The authors are very grateful to L.D. Burdina for
carrying out the GaAs substrate preparation and
chemical layer-by-layer etching, V.A. Kartashov and
I.N. Uzhakov for the growth of buffer layers, and
T.I. Zakharyash for the fabricaton of Hall structures.
References
1. J.N. Schulman, T.C. McGill, The CdTe/HgTe
superlattice: Proposal for a new infrared material //
Appl. Phys. Lett. 34, p. 663-665 (1979).
2. M.W. Goodwin, M.A. Kinch, R.J. Koestner,
Metal–insulator–semiconductor properties of
HgTe–CdTe superlattices // J. Vac. Sci. Technol. A
6, p. 2685-2692 (1988).
3. Y.D. Zhou, C.R. Becker, Y. Selament, Y. Chang,
R. Ashokan, R.T. Boreiko, T. Aoki, D.J. Smith,
A.L. Betz, S. Sivananthan, Far-infrared detector
based on HgTe/HgCdTe superlattices // J. Electron.
Mater. 32(7), p. 608-614 (2003).
4. J.P. Zanatta, F. Noel, P. Ballet, N. Hidadach,
A. Million, G. Destefanis, E. Mottin, C. Kopp,
E. Picard, E. Hadji, HgCdTe MBE material for
microcavity light emitters: application to gas
detection in the 2-6 µm range// J. Electron. Mater.
32(7), p. 602-607 (2003).
5. C.R. Becker, X.C. Zhang, A. Pfeuffer-Jeschke,
K. Ortner, V. Hock, G. Landwehr, L.W. Molen-
kamp, Very large Rashba spin-orbit splitting in
HgTe quantum wells // Journal of Supercon-
ductivity: Incorporating Novel Magnetism 16(4),
p. 625-634 (2003).
6. G. Landwehr, J. Gerschutz, S. Oehling, A. Pfeuf-
fer-Jeschke, V. Latussek, C.R. Becker, Quantum
transport in n-type modulation-doped mercury
telluride quantum well // Physica E 6, p. 713-717
(2000).
7. C.R. Becker, X.C. Zhang, K. Ortner, J. Schmidt,
A. Pfeuffer-Jeschke, V. Latussek, Y.S. Gui,
V. Daumer, J. Liu, H. Buhmann, G. Landwehr,
L.W. Molenkamp, MBE growth and characte-
rization of based Hg compounds and heterostruc-
tures // Thin Solid Films 412, p. 129-138 (2002).
8. C.R. Becker, K. Ortner, X.C. Zhang, A. Pfeuffer-
Jeschke, V. Latussek, Y.S. Gui, V. Daumer,
H. Buhmann, G. Landwehr, L.W. Molenkamp,
Growth and studies of Hg1-xCdxTe based low
dimensional structures // Physica E 20, p. 436-443
(2004).
9. N.N. Mikhailov, S.V. Rychlitsky, E.V. Spesivtsev,
S.A. Dulin, N.I. Nazarov, Yu.G. Sidorov,
S.A. Dvoretsky, Integrated analytical equipment for
film growth in MBE technology // Mater. Sci. &
Eng. B 80, p. 41-45 (2001).
10. V.S. Varavin, V.V. Vasiliev, S.A. Dvoretsky,
N.N. Mikhailov, V.N. Ovsyuk, Yu.G. Sidorov,
A.O. Suslyakov, M.V. Yakushev, A.L. Aseev,
HgCdTe epilayers on GaAs: growth and devices //
Opto-electronics Review 11(2), p. 99-111 (2003).
11. Yu.G. Sidorov, S.A. Dvoretsky, N.N. Mikhailov,
M.V. Yakushev, V.S. Varavin, A.P. Antsiferov,
Molecular beam epitaxy of CdxHg1-xTe com-
pounds: technology and equipment // Optical
Journal 67(1), p. 39-45 (2000).
12. Y. Demay, J.P. Galliard, P. Medina, In situ
spectroscopic ellipsometry of mercury cadmium
telluride MBE layers // J. Cryst. Growth 81, p. 97-
100 (1987).
13. A.V. Rzhanov, K.K. Svitashev, A.S. Mardezhov,
V.A. Shvets, In situ superlattices characterization
by ellipsometry // Doklady Akademii Nauk 297(3),
p. 604-607 (1987) (in Russian).
14. N.N. Mikhailov, R.N. Smirnov, S.A. Dvoretsky,
Yu.G. Sidorov, V.A. Shvets, E.V. Spesivtsev,
S.V. Rykhlitski, Growth of Hg1–xCdxTe nano-
structures by molecular beam epitaxy with
ellipsometric control // Intern. J. Nanotechnology
3(1), p. 120-131 (2006).
15. D.G. Ikusov, S.A. Dvoretsky, V.A. Kartashov,
V.A. Kolesnikov, N.N. Mikhailov, I.V. Sabinina,
Yu.G. Sidorov, Structural perfection and mor-
phology of the CdTe epitaxial layers // Abstracts of
the IX Conference on Semiconductor, Dielectric
and Magnetic Materials Physics, p. 176 (2005).
16. P.A. Bakhtin, S.A. Dvoretsky, V.S. Varavin,
A.P. Korobkin, N.N. Mikhailov, Yu.G. Sidorov,
Conductivity and Hall coefficient to magnetic field
dependences in CdxHg1-xTe films grown by mole-
cular beam epitaxy // Fizika Tekhnika Poluprovod-
nikov 38(10), p. 1203-1206 (2004) (in Russian).
17. E.V. Spesivtsev, S.V. Rykhlitski, Useful model
“Ellipsometer.” License No. 16314 RF // Bulletin
“Useful models. Industrial samples” 35
(20.12.2000).
18. R.M.A. Azzam, N.M. Bashara, Ellipsometry and
Polarized Light. North-Holland, Amsterdam, 1977.
19. V.A. Shvets, S.V. Rykhlitski, E.V. Spesivtsev,
N.A. Aulchenko, N.N. Mikhailov, S.A. Dvoretsky,
Yu.G. Sidorov, R.N. Smirnov, In situ ellipsometry
for control of Hg1-xCdxTe nanolayer structures and
inhomogeneous during MBE growth // Тhin Solid
Films 455-456, p. 688-694 (2004).
20. K.K. Svitashev, S.A. Dvoretsky, Yu.G. Sidorov,
V.A. Shvets et al., The growth of high-quality
MCT films by MBE using in situ ellipsometry //
Cryst. Res. Technol. 29(7), p. 931-937 (1994).
Semiconductor Physics, Quantum Electronics & Optoelectronics, 2007. V. 10, N 4. P. 47-53.
© 2007, V. Lashkaryov Institute of Semiconductor Physics, National Academy of Sciences of Ukraine
53
21. V.A. Shvets, N.N. Mikhailov, M.V. Yakushev,
E.V. Spesivtsev, Ellipsometric measurements of the
optical constants of solids under impulse heating //
Proc. SPIE 4900, p. 46-52 (2002).
22. H. Arwin, D.E. Aspnes, Nondestructive analysis of
Hg1-xCdxTe (x = 0.00, 0.2, 0.29, and 1.00) by spec-
troscopic ellipsometry. Pt. 2. Substrate, oxide and
interface properties // J. Vac. Sci. Technol. A 2,
p. 1316-1323 (1984).
23. L. Vina, C. Umbach, M. Cardona, L. Vodopjanov,
Ellipsometric studies of electronic interband
transitions in CdxHg1-xTe // Phys. Rev. B 29,
p. 6752-6760 (1984).
24. N.A. Kornyushkin, T.I. Kurdina, A.S. Mardezhov
et al., Optical and electrical properties of HgTe
films grown by molecular beam epitaxy //
Poverkhnost’. Fizika, khimiya, mekhanika 8, p. 94-
100 (1991) (in Russian).
25. L. Vina, C. Umbach, M. Cardona, L. Vodopyanov,
Ellipsometric studies of electronic interband
transitions in CdxHg1-xTe // Phys. Rev. B 29(12),
p. 6752-6760 (1984).
|