Minority carrier lifetime of lattice-matched CdZnTe alloy grown on InSb substrates
using molecular beam epitaxy
Shi Liu, Xin-Hao Zhao, Calli Campbell, Michael J. DiNezza, Yuan Zhao, and Yong-Hang Zhang
Citation: Journal of Vacuum Science & Technology B 33, 011207 (2015); doi: 10.1116/1.4905289
View online: http://dx.doi.org/10.1116/1.4905289
View Table of Contents: http://scitation.aip.org/content/avs/journal/jvstb/33/1?ver=pdfcov
Published by the AVS: Science & Technology of Materials, Interfaces, and Processing
Articles you may be interested in
Determination of CdTe bulk carrier lifetime and interface recombination velocity of CdTe/MgCdTe double
heterostructures grown by molecular beam epitaxy
Appl. Phys. Lett. 105, 252101 (2014); 10.1063/1.4904993
Impact of dislocations on minority carrier electron and hole lifetimes in GaAs grown on metamorphic SiGe
substrates
Appl. Phys. Lett. 84, 3447 (2004); 10.1063/1.1736318
Effect of dislocations on minority carrier lifetime in HgCdTe
J. Appl. Phys. 95, 2467 (2004); 10.1063/1.1644039
Radiative and nonradiative recombination processes in lattice-matched (Cd,Zn)O/(Mg,Zn)O multiquantum wells
Appl. Phys. Lett. 77, 1632 (2000); 10.1063/1.1308540
N -type doping of lattice-matched ZnCdSe and Zn x Cd y Mg 1−x−y Se epilayers on InP using ZnCl 2
J. Appl. Phys. 84, 1472 (1998); 10.1063/1.368210

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. Download to IP: 209.147.144.12 On: Wed, 28 Jan 2015 18:38:19

Minority carrier lifetime of lattice-matched CdZnTe alloy grown on InSb
substrates using molecular beam epitaxy
Shi Liu
Center for Photonics Innovation and School of Electrical, Computer and Energy Engineering,
Arizona State University, Tempe, Arizona 85287

Xin-Hao Zhao and Calli Campbell
Center for Photonics Innovation and School for Engineering of Matter, Transport and Energy,
Arizona State University, Tempe, Arizona 85287

Michael J. DiNezza, Yuan Zhao, and Yong-Hang Zhanga)
Center for Photonics Innovation and School of Electrical, Computer and Energy Engineering,
Arizona State University, Tempe, Arizona 85287

(Received 2 November 2014; accepted 19 December 2014; published 6 January 2015)
A CdZnTe/MgCdTe double-heterostructure (DH) consisting of a 3 lm thick Cd0.9946Zn0.0054Te
middle layer that is lattice-matched to an InSb substrate has been grown using molecular beam
epitaxy. A long carrier lifetime of 3.4  102 ns has been demonstrated at room temperature, which
is approximately three times as long as that of a CdTe/MgCdTe DH with identical layer thickness.
This substantial improvement is due to the reduction in misfit dislocation density in the CdZnTe
alloy. In contrast, a CdTe/MgCdTe DH with 3 lm thick CdTe layer grown on an InSb substrate
exhibits a strain relaxation of 30%, which leads to a wider x-ray diffraction peak, a weaker
integrated photoluminescence intensity, and a shorter minority carrier lifetime of 1.0  102 ns.
These findings indicate that CdZnTe lattice-matched to InSb has great potential as applied to
high-efficiency solar cells as well as virtual substrates for high-performance large-area HgCdTe
C 2015 American Vacuum Society. [http://dx.doi.org/10.1116/1.4905289]
focal plane arrays. V

I. INTRODUCTION
There has been growing interest in CdTe solar cells in
recent years due to continuing improvements in device performance and market share. The efficiency record of polycrystalline CdTe solar cells has been frequently broken,
recently reaching 21.0%.1 However, this is still more than
ten absolute percent lower than the Shockley–Queisser limit.
Although the short-circuit current density is already close to
the theoretical limit, there is still significant room for
improvement in open-circuit voltage (VOC) and fill factor,
which are currently below 0.9 V and 80%, respectively, for
most devices.2 Such imperfections are believed to be due to
large defect density, short minority carrier lifetime and so
on. The use of monocrystalline CdTe is anticipated to
improve the VOC dramatically because of its better material
quality and longer carrier lifetime. Several research efforts
have demonstrated long carrier lifetimes in both epitaxial
CdTe on InSb and bulk CdTe, along with a VOC of greater
than 0.9 V in monocrystalline CdTe solar cells.3–9
Regardless of these new achievements, CdTe is not completely lattice-matched to InSb, which may cause material
quality degradation when the CdTe epilayer is beyond a
certain thickness (e.g., 2 lm). This problem can be circumvented by adding a small fraction of Zn into CdTe in order
to form a CdZnTe alloy that is completely lattice-matched to
InSb and exhibits a negligible increase in bandgap. A longer
minority carrier lifetime is thus expected in the CdZnTe
epilayer, especially when the layer is more than 2 lm thick,
a)

Electronic mail: yhzhang@asu.edu

011207-1 J. Vac. Sci. Technol. B 33(1), Jan/Feb 2015

since the misfit dislocation density is lower in this latticematched material system. Moreover, CdZnTe alloys grown
on InSb, with tunable Zn compositions, can provide highquality virtual substrates for HgCdTe growth. The advantages of this approach include: (1) CdZnTe is chemically
compatible with, and lattice-matched to, HgCdTe;10,11 (2)
InSb substrates are of both high quality and low cost which
overcomes the existing cost issues of CdZnTe substrates that
impede their widespread application in large-area focal plane
arrays;10,11 and (3) CdZnTe on InSb offers better material
quality compared to other alternative substrate techniques,
such as CdTe/Si,12 CdTe/Ge,13 CdTe/GaAs,14 CdTe/GaSb,15
and InSb,16 and thus has the feasibility to achieve high
performance in long-wavelength infrared detectors.
II. EXPERIMENT
In this paper, we report the demonstration of a high-quality
Cd0.9946Zn0.0054Te/Mg0.24Cd0.76Te double-heterostructure (DH)
grown on an InSb (100) substrate by molecular beam epitaxy
(MBE) and characterized by high-resolution x-ray diffraction
(XRD), steady-state photoluminescence (PL), and timeresolved photoluminescence (TRPL). Two sets of samples
have been studied here. The first set consists of samples A to C
(growth run numbers are A1654, A1656, and A1659, respectively), which are 1 lm CdZnTe layers grown under various Zn
fluxes, for the purpose of lattice-match calibration. The second
set consists of samples D and E (growth run numbers are
A1667 and A1666, respectively), which are 3 lm CdTe/
Mg0.24Cd0.76Te and Cd0.9946Zn0.0054Te/Mg0.24Cd0.76Te DHs,
for structural and optical properties comparisons. The

2166-2746/2015/33(1)/011207/4/$30.00

C 2015 American Vacuum Society
V

011207-1

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. Download to IP: 209.147.144.12 On: Wed, 28 Jan 2015 18:38:19

011207-2 Liu et al.: Minority carrier lifetime of lattice-matched CdZnTe alloy

schematic layer structures of samples D and E are shown in
Fig. 1. The 30 nm MgCdTe barrier layers, with Mg compositions of 24%, prevent photogenerated carriers from reaching
the top surface and bottom buffer/substrate interface to recombine nonradiatively. The Zn composition in the CdZnTe alloy
is designed to be 0.54% so that the alloy is perfectly latticematched to the InSb substrate.
All the samples are grown using a VG V80H dualchamber MBE system with separate III-V and II-VI growth
chambers connected by an ultrahigh vacuum (UHV) transfer
chamber. The oxide removal process of InSb substrates
starts with an initial substrate temperature ramp to a thermocouple reading of 500  C at 20  C/min. Then, the temperature is further ramped up at a slower rate of 5  C/min in 5  C
intervals, with a 3 min pause after each ramp to prevent the
substrate from overheating, until a pyrometer reading of
475–480  C is reached. Such a careful temperature ramp is
necessary because the InSb oxide removal temperature is
very close to the InSb melting point. The substrates are kept
at this temperature for one hour, and the completion of oxide removal is then indicated by streaky pseudo-(1  3)
reflection high energy electron diffraction patterns. Next,
the substrate temperature is brought down to 390  C for
the 500 nm InSb buffer growth, with a growth rate of
9.1 nm/min and a Sb/In flux ratio of 1.5. After the III-V
growth, the samples are transferred to the II-VI chamber
under UHV, which avoids surface oxidation and contamination. The II-VI layers are grown at 265  C with a growth
rate of 9.6 nm/min and a Cd/Te flux ratio of 1.5. Detailed
growth procedures of II-VI layers can be found in the previous publication.3
III. RESULTS AND DISCUSSION
The lattice-match calibration of the CdZnTe alloy to the
InSb substrate is carried out by measuring the XRD patterns
of samples A to C and from there determining the Zn composition. The x/2h scans at the (004) orientation of samples
A to C, which are grown under Zn cell temperatures of 235,
245, and 255  C, respectively, are shown in Fig. 2. It is
observed that there are two separate peaks in samples A and
B that correspond to the strained CdZnTe layer and the InSb

FIG. 1. Schematic layer structures of sample D: CdTe/Mg0.24Cd0.76Te DH,
and sample E: Cd0.9946Zn0.0054Te/Mg0.24Cd0.76Te DH.

011207-2

FIG. 2. High-resolution x-ray diffraction (004) patterns of samples A, B, and
C, which have 1 lm CdZnTe layers on InSb substrates, showing that a complete lattice match is achieved.

substrate. However, in sample C, the two peaks merge together, indicating that a complete lattice match is reached.
The Zn compositions listed in the figure are obtained by fitting the measured diffraction patterns based on the published
structural parameters of ZnTe, CdTe,17 and InSb. The narrow full-width at half-maximums (FWHMs) of the CdZnTe
peaks, which are comparable to those of InSb substrates,
along with the clear Pendell€osung fringes are indicators of
high quality CdZnTe layers and smooth CdZnTe/InSb interfaces. The (004) x/2h scans are also performed for samples
D and E, and the experimental and simulated diffraction
patterns are shown in Fig. 3. The diffraction pattern of sample D clearly shows two separate peaks representing InSb
and CdTe, and a strain relaxation of 30% for the CdTe
layer is determined by XRD pattern simulation. In contrast,
only one peak is present in the diffraction pattern of sample
E, which means that the CdZnTe layer and the InSb substrate
are lattice-matched. A larger FWHM of 22.6 arc sec
observed for sample D compared to that for sample E
(16.6 arc sec) shows degraded material quality due to strain
relaxation. The broad peaks of the MgCdTe barrier layers
and the Pendell€osung fringes are observed in both samples,
indicating the high material quality of these layers and
smooth MgCdTe/Cd(Zn)Te interfaces.
The steady-state PL spectra of samples D and E are measured at room temperature using a 532 nm diode pumped
solid state laser as a pumping source. The laser excitation
power density is kept at 100 mW/cm2 to mimic AM 1.5G
solar radiation and to protect the II-VI layers from thermal
damage by laser heating. As shown in Fig. 4, sample D has a
PL peak wavelength of 825 nm, which corresponds to a
band-to-band transition energy of 1.503 eV, while sample E
has a PL peak wavelength of 824 nm, which corresponds to a
band-to-band transition energy of 1.505 eV. Such a slight

J. Vac. Sci. Technol. B, Vol. 33, No. 1, Jan/Feb 2015

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. Download to IP: 209.147.144.12 On: Wed, 28 Jan 2015 18:38:19

011207-3 Liu et al.: Minority carrier lifetime of lattice-matched CdZnTe alloy

FIG. 3. (Color online) High-resolution x-ray diffraction (004) patterns for
samples D (3 lm CdTe/MgCdTe DH) and E (3 lm CdZnTe/MgCdTe DH).
The 3 lm CdTe layer has a relaxation of 30%, while the 3 lm CdZnTe
layer is completely lattice-matched to the InSb substrate. They also clearly
show that the CdTe peak has a FWHM of 22.6 arc sec, compared to the
value of 16.6 arc sec for the CdZnTe peak, indicating that strain relaxation
results in degraded material quality.

increase in the transition energy indicates that the bandgap
increase caused by introducing 0.54% Zn in the CdTe binary
is negligible. This is desirable for solar cells since any deviation from the optimal bandgap for single-junction solar cells
should be avoided. The integrated PL intensity of sample E
is at least one order of magnitude higher than that of sample
D. Note that although the lattice mismatch between CdTe

011207-3

FIG. 5. (Color online) Time-resolved photoluminescence decays of samples
D and E at room temperature, showing a significantly improved minority
carrier lifetime, achieved by replacing CdTe with lattice-matched CdZnTe
in the 3 lm DH design. The longest lifetime of 3.4  102 ns is observed.

and InSb is only 0.03%, the optical properties of the CdTe
layer degrade noticeably due to the large density of misfit
dislocations generated as a result of strain relaxation.
Figure 5 shows the room temperature PL decays of samples D and E measured by a time-correlated single photon
counting system. The samples are excited by an ultrafast
Ti:Sapphire laser with a 750 nm emission wavelength, a 130
fs pulse duration, and a 0.8 MHz repetition rate. The luminescence signal is detected by a spectrometer equipped with
a high-speed microchannel plate photomultiplier tube, with
the detection wavelength set at 820 nm. The minority carrier
lifetimes of both samples are determined by fitting the tails
of the decay curves, and are listed in Fig. 5. Sample E has a
much longer measured lifetime (3.4  102 ns) than that of
sample D (1.0  102 ns), which again proves the necessity of
the lattice match to maintain low defect density and long carrier lifetime for a thick epilayer. Table I summarizes the
strain relaxations, XRD FWHMs, integrated PL intensities
and minority carrier lifetimes of samples D and E. These
findings consistently prove that better structural and optical
qualities are obtained in lattice-matched CdZnTe as opposed
to CdTe grown on InSb substrates.

TABLE I. Strain relaxations, XRD FWHMs, integrated PL intensities, and minority carrier lifetimes of sample D: CdTe/MgCdTe DH, and sample E:
CdZnTe/MgCdTe DH.
Sample

FIG. 4. (Color online) Room temperature photoluminescence spectra of samples D and E, showing that the integrated photoluminescence intensity of
the CdZnTe/MgCdTe DH is over one order of magnitude stronger than that
of the CdTe/MgCdTe DH.

Strain relaxation in Cd(Zn)Te layer (%)
XRD FWHM of Cd(Zn)Te layer (arc sec)
Integrated PL intensity (a.u.)
Carrier lifetime (ns)

D

E

30
22.6
2556
1.0  102

0
16.6
29717
3.4  102

JVST B - Nanotechnology and Microelectronics: Materials, Processing, Measurement, and Phenomena

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. Download to IP: 209.147.144.12 On: Wed, 28 Jan 2015 18:38:19

011207-4 Liu et al.: Minority carrier lifetime of lattice-matched CdZnTe alloy

IV. SUMMARY
In summary, a CdZnTe ternary alloy completely latticematched to an InSb substrate has been demonstrated with
accurate composition control using MBE. The Zn composition is determined to be 0.54% using XRD. Structural and
optical properties comparisons have been carried out for
3 lm CdTe/MgCdTe DH and CdZnTe/MgCdTe DH samples
grown on InSb substrates. Despite the fact that the lattice
mismatch between CdTe and InSb is only 0.03%, the 3 lm
CdTe layer has a relaxation of 30%, and its (004) XRD
peak is wider than that of a 3 lm lattice-matched CdZnTe
layer. It has also been observed that the CdZnTe/MgCdTe
DH sample shows an integrated PL intensity one order of
magnitude higher than, and a carrier lifetime three times as
long as, those of the CdTe/MgCdTe DH sample. The longest
carrier lifetime is measured as 3.4  102 ns for the CdZnTe/
MgCdTe DH. These comparisons show the excellent material quality of a lattice-matched CdZnTe alloy grown on an
InSb substrate, and manifest its great potential in highefficiency solar cells and large-area focal plane arrays.
ACKNOWLEDGMENTS
The authors would like to thank Su Lin for assistance
with TRPL measurements, and Martyn Fisher for manuscript
revision. This work was partially supported by AFOSR
(Grant No. FA9550-12-1-0444), and National Science
Foundation Graduate Research Fellowship (Grant No. DGE0802261). The authors gratefully acknowledge the use of

011207-4

facilities in the LeRoy Eyring Center for Solid State Science
at Arizona State University.
1

First Solar press release, 5 Aug 2014. First Solar achieves world’s highest
CdTe solar cell efficiency of 21.0%, see http://investor.firstsolar.com/
releasedetail.cfm?ReleaseID¼864426.
2
T. A. Gessert et al., Sol. Energy Mater. Sol. Cells 119, 149 (2013).
3
M. J. DiNezza, X.-H. Zhao, S. Liu, A. P. Kirk, and Y.-H. Zhang, Appl.
Phys. Lett. 103, 193901 (2013).
4
A. P. Kirk, M. J. DiNezza, S. Liu, X.-H. Zhao, and Y.-H. Zhang,
Proceedings of the 39th IEEE Photovoltaic Specialists Conference,
Tampa, FL, 16–21 June 2013, pp. 2512–2517.
5
X.-H. Zhao, M. J. DiNezza, S. Liu, S. Lin, Y. Zhao, and Y.-H. Zhang,
J. Vac. Sci. Technol., B 32, 040601 (2014).
6
X.-H. Zhao, M. J. DiNezza, S. Liu, C. Campbell, Y. Zhao, and Y.-H.
Zhang, Appl. Phys. Lett. 105, 252101 (2014).
7
D. Kuciauskas, A. Kanevce, J. M. Burst, J. N. Duenow, R. Dhere, D. S. Albin,
D. H. Levi, and R. K. Ahrenkiel, IEEE J. Photovoltaics 3, 1319 (2013).
8
D. Kuciauskas, A. Kanevce, P. Dippo, S. Seyedmohammadi, and R.
Malik, IEEE J. Photovoltaics 5, 366 (2014).
9
J. N. Duenow, J. M. Burst, D. S. Albin, D. Kuciauskas, S. W. Johnston, R.
C. Reedy, and W. K. Metzger, Appl. Phys. Lett. 105, 053903 (2014).
10
A. Rogalski, Rep. Prog. Phys. 68, 2267 (2005).
11
O. Gravarnd et al., J. Electron. Mater. 42, 3349 (2013).
12
R. Sporken, S. Sivananthan, K. K. Mahavadi, G. Monfroy, M. Boukerche,
and J. P. Faurie, Appl. Phys. Lett. 55, 1879 (1989).
13
J. P. Zanatta et al., J. Electron. Mater. 35, 1231 (2006).
14
R. N. Jacobs et al., J. Electron. Mater. 41, 2707 (2012).
15
W. Lei, R. J. Gu, J. Antoszewski, J. Dell, and L. Faraone, J. Electron.
Mater. 43, 2788 (2014).
16
T. J. de Lyon et al., J. Electron. Mater. 39, 1058 (2010).
17
S. Adachi, Handbook on Physical Properties of Semiconductors, II-VI
Compound Semiconductors (Kluwer Academic, Boston, MA, 2004), Vol.
III, pp. 212 and 360.

J. Vac. Sci. Technol. B, Vol. 33, No. 1, Jan/Feb 2015

Redistribution subject to AVS license or copyright; see http://scitation.aip.org/termsconditions. Download to IP: 209.147.144.12 On: Wed, 28 Jan 2015 18:38:19