Long-wave infrared nBn photodetectors based on InAs/InAsSb type-II superlattices
H. S. Kim, O. O. Cellek, Zhi-Yuan Lin, Zhao-Yu He, Xin-Hao Zhao, Shi Liu, H. Li, and Y.-H. Zhang
Citation: Applied Physics Letters 101, 161114 (2012); doi: 10.1063/1.4760260
View online: http://dx.doi.org/10.1063/1.4760260
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/101/16?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Bias-selectable dual-band mid-/long-wavelength infrared photodetectors based on InAs/InAs1−xSbx type-II
superlattices
Appl. Phys. Lett. 106, 011104 (2015); 10.1063/1.4905565
InAs/InAs1−xSbx type-II superlattices for high performance long wavelength infrared detection
Appl. Phys. Lett. 105, 121104 (2014); 10.1063/1.4896271
High performance long-wave type-II superlattice infrared detectors
J. Vac. Sci. Technol. B 31, 03C122 (2013); 10.1116/1.4798485
Performance improvement of longwave infrared photodetector based on type-II InAs/GaSb superlattices using
unipolar current blocking layers
Appl. Phys. Lett. 96, 231107 (2010); 10.1063/1.3446967
Molecular beam epitaxy growth and characterization of type-II InAs/GaSb strained layer superlattices for longwave infrared detection
J. Vac. Sci. Technol. B 28, C3G13 (2010); 10.1116/1.3276429

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
209.147.144.22 On: Wed, 04 Feb 2015 17:14:25

APPLIED PHYSICS LETTERS 101, 161114 (2012)

Long-wave infrared nBn photodetectors based on InAs/InAsSb type-II
superlattices
H. S. Kim, O. O. Cellek, Zhi-Yuan Lin, Zhao-Yu He, Xin-Hao Zhao, Shi Liu, H. Li,
and Y.-H. Zhanga)
Center for Photonics Innovation and School of Electrical, Computer, and Energy Engineering, Arizona State
University, Tempe, Arizona 85287, USA

(Received 27 August 2012; accepted 4 October 2012; published online 18 October 2012)
Long-wave infrared InAs/InAsSb type-II superlattice nBn photodetectors are demonstrated on
GaSb substrates. The typical device consists of a 2.2 lm thick absorber layer and has a 50% cutoff
wavelength of 13.2 lm, a measured dark current density of 5  104 A/cm2 at 77 K under a bias of
0.3 V, a peak responsivity of 0.24 A/W at 12 lm, and a maximum resistance-area product of 300
X cm2 at 77 K. The calculated generation-recombination noise limited specific detectivity (D*) and
experimentally measured D* at 12 lm and 77 K are 1  1010 cm Hz1/2/W and 1  108 cm Hz1/2/W,
C 2012 American Institute of Physics. [http://dx.doi.org/10.1063/1.4760260]
respectively. V
Mid-wave infrared (MWIR) and long-wave infrared
(LWIR) photodetectors have broad applications in environmental monitoring, imaging, and scientific instrumentation.
Currently, many of these infrared applications need to use
high-performance HgCdTe photodetectors which operate
around 200 K or below to reduce the dark current and to provide the necessary performance.1 Due to higher cost limitations of the HgCdTe materials, antimonide based type-II
superlattices (T2SL) have been proposed as an alternative2,3
to HgCdTe with potentially lower manufacturing cost on
large substrates4 and better device performance with low
dark current density due to suppressed Auger recombination
rate and tunneling current.5–7 These advantages of T2SLs
enable them to operate at higher temperatures, which is
highly desirable for photodetector applications.8,9
Although the feasibility of T2SLs for IR photodetectors
has been studied for decades, the T2SL materials properties
and device performance still have not reached the level predicted by theory.6 One of the limiting factors of the most
widely studied LWIR InAs/(In)GaSb T2SLs is the short minority carrier lifetime, which determines the dark current, detectivity, and ultimately the maximum operating temperature. The
typical minority carrier lifetimes in LWIR InAs/GaSb T2SL
measured with time-resolved photoluminescence (PL) and optical modulation response techniques are only 30 ns and 31 ns
at 77 K, respectively.10,11 Recently, the study of another type
of T2SL, namely the “Ga-free” InAs/InAsSb T2SLs grown on
GaSb substrates, has been revisited and revealed very encouraging results.12–15 Time resolved PL measurements of the
InAs/InAs0.72Sb0.28 T2SL structures showed minority carrier
lifetimes greater than 412 ns due to strong reduction of nonradiative recombination in these superlattices.16 Therefore, it is
reasonable to expect that substantially better photodetector performance and higher operating temperature are achievable
with the InAs/InAsSb T2SL materials.
It has been demonstrated lately that the nBn structure
suppresses the dark current in infrared photodetectors effectively through band gap engineering.17 The electron transport
a)

Electronic mail: yhzhang@asu.edu.

0003-6951/2012/101(16)/161114/3/$30.00

in nBn structures is blocked by a conduction band barrier
while the hole can easily transport to the contact. Recently, an
MWIR focal plane array based on an unpassivated nBn structure with an AlGaSb barrier and an InAs/GaSb T2SL n-type
absorber has been reported.18 In this paper, we demonstrate
LWIR nBn photodetectors with an InAs/InAsSb T2SL
absorber having a cutoff wavelength of 13.2 lm.
Figure 1 shows the bandedge diagram of the device
design, which consists of an n-type InAs/InAs0.62Sb0.38
T2SL as the top contact layer, lattice-matched InAs/
Al0.8Ga0.2.As0.01Sb0.99 T2SL as the barrier for electrons, and
an InAs/InAs0.62Sb0.38 T2SL as the absorber. A three-band
Kronig-Penney model is used to calculate the electron and
hole miniband energy levels in the InAs/Al0.8Ga0.2As0.01Sb0.99
T2SL barrier layer. The calculated conduction band barrier
height is 409 meV. The Al and As mole fractions of the barrier
are carefully chosen to align the hole miniband in the InAs/
InAsSb T2SL layer with that in the barrier layer to ensure
optimized hole transport.
The designed nBn device is grown on a Te-doped (100) 2
in. GaSb substrate using a VG V80 solid-source MBE system.

FIG. 1. Designed energy band diagram of T2SL nBn photodetector using
the three-band Kronig-Penney model calculation. The conduction band offset between the barrier and the absorber is 0.4 eV.
101, 161114-1

C 2012 American Institute of Physics
V

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
209.147.144.22 On: Wed, 04 Feb 2015 17:14:25

161114-2

Kim et al.

The growth starts with a 500 nm GaSb buffer at 510  C followed by the bottom contact layer consisting of 15 periods of
13.9 nm InAs:Si (1  1018 cm3)/4.5 nm InAs0.62Sb0.38 T2SL
layers. Then, a 120 periods un-intentionally doped 13.9 nm
InAs/4.5 nm InAs0.62Sb0.38 T2SL absorber layer is grown, followed by a barrier consisting of 17 periods of 2.8 nm InAs/
2.5 nm Al0.8Ga0.2As0.01Sb0.99 T2SL layers. Finally, the n-type
top contact layer is grown, which consists of 8 periods of
13.9 nm InAs:Si (1  1018 cm3)/4.5 nm InAs0.62Sb0.38 T2SL
layers and a 13.9 nm InAs:Si (1  1018 cm3) top layer. The
growth temperature is kept at 415  C during the growth of all
superlattice layers. The high-resolution x-ray diffraction
(XRD) shows that the actual T2SL barrier consists of InAs/
Al0.8Ga0.2As0.03Sb0.97, where the increase of As composition
compared to the designed value is due to the incorporation of
background As in the MBE chamber. The calculated hole
miniband energy level in the barrier layer indicates that the
sample possesses a hole barrier of 38 meV in the valance
band between the valance bandedge of barrier and the valance
bandedge of absorber layer.
Device processing starts with photolithography to define
the mesas (410 lm  410 lm), which have open optical apertures ranging from 50 lm to 200 lm diameter. Wet etching is
done with citric acid solution (1 g citric acid:1 ml water):
H2O2:DI water solution with ratio of 1:1:10. The etching is
stopped at the middle of the n-type bottom contact, resulting
in an etch depth of 2.53 lm. The bottom and top metal contacts (50 nm Ti/50 nm Pt/300 nm Au) are deposited using an
e-beam evaporator. The front-side illuminated device does
not have any anti-reflecting coating and passivation. The fabricated devices are wire bonded to a ceramic package with
the bottom n-type InAs/InAsSb T2SL layer connected as a
common ground, and external bias is applied to the top contact of each mesa.
The spectral response is measured using a Fourier transform infrared spectrometer with a KBr beam splitter and
SRS-570 low-noise current preamplifier. The LWIR spectral
response is measured at 0.5 V, while the detector temperature is swept from 8 K to 77 K (Fig. 2). The cutoff wavelength at 8 K is 12.5 lm, which red-shifts to 13.2 lm at 77 K.

FIG. 2. Front side illuminated single pass spectral responsivity spectrum
for an nBn photodetector under the applied bias of 0.5 V, measured at 8 K
and 77 K.

Appl. Phys. Lett. 101, 161114 (2012)

FIG. 3. Measured responsivity and specific detectivity as a function of
applied bias at 50 K and 180 field of view conditions.

The responsivity and detectivity are measured under 180
field of view using an 800  C blackbody source with a 1 in.
aperture, chopped at 130 Hz, at a distance of 91 cm, with a
7.3 lm long-pass IR filter with 90% transmission. The peak
responsivity at 12 lm wavelength is measured as 0.3 A/W
at 0.5 V bias at 77 K. The responsivity drops to 0.2 A/W
at 8 K which is attributed to reduced mobility of the holes at
lower temperatures and shorter carrier lifetime.16
Figure 3 shows the bias dependent responsivity and specific detectivity (D*) at 12 lm. The responsivity increases with
bias and reaches 0.36 A/W at 50 K at a bias of 0.8 V. The
peak D* is 3  108 cm Hz1/2/W under a bias of 0.3 V. Figure
4 shows the temperature dependence of the peak responsivity
and D*, which is 0.24 A/W and 1  108 cm Hz1/2/W, respectively, at 77 K and 12 lm under the fixed bias of 0.3 V with
180 field of view. The quantum efficiency (QE) under these
conditions is 2.5%.
To compare with the theoretical limit, theoretical D* is
calculated using the following equation:
pffiffiffi
q
g
A
E
c
;
(1)
D ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ð4kT=RÞ þ 4qI

FIG. 4. Measured responsivity and specific detectivity as a function of device temperature under the applied bias of 0.3 V and 180 field of view
conditions.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
209.147.144.22 On: Wed, 04 Feb 2015 17:14:25

161114-3

Kim et al.

FIG. 5. Dark current density as a function of applied bias ranging from 12 K
to 100 K, the solid color line shows the total current under 290 K background
with f/2 optics at 50 K with 28% optical aperture fill-factor. The inset shows
dynamic dark RA product at 77 K.

where q is the electron charge, Ec is the photon energy, g is
the quantum efficiency, A is the mesa area, k is the Boltzmann constant, T is the temperature, R is the AC resistance
of the mesa, and I is the total current passing through the
mesa. The 4kT/R term represents the Johnson noise power
and the 4qI term represents the generation-recombination
(GR) noise power with assumption of unity gain. GR noise
limited D* at 12 lm is 1  1010 cm Hz1/2/W at 77 K. We partially attribute measured lower D* to the unpassivated mesa
sidewalls which may create excess noise and dark current. In
addition, further optimization of the valance band edge
energy position and doping profile will lower dark current
and improve D*.
Figure 5 shows the measured dark current density as a
function of applied bias for temperatures ranging from 12 K
to 100 K with a cold-shield cover. At 77 K, the dark current
density is 5  104 A/cm2 at an operating point of 0.3 V
bias. The inset shows that the dynamic dark resistance-area
(RA) product is 330 X cm2 under 0.35 V applied bias at
77 K. The room temperature background (f/2, 290 K) current
of the nBn device is also shown in Figure 5, which has an
aperture of 100 lm diameter defined by metal contact and a
28% fill factor of optically active area. The photocurrent
shows weak dependence on temperature between 50 K and
77 K. The definition of Itotal/Idark > 5 is used as the background limited photodetection (BLIP) condition.9 When the
optically active area fill factor is 100%, the ratio Itotal/
Idark ¼ 2 at 77 K, and 13 at 70 K; the latter is claimed as the
f/2 BLIP temperature for this 13.2 lm cutoff photodetector.
It is worth noting that the measured dark current density
5  104 A/cm2 of the devices with a cutoff wavelength of
13.2 lm at 77 K is lower than that given by the well-known
HgCdTe Rule-07,19 although it should be pointed out that
the QE of the detector reported here is 2.5%, while the Rule07 is defined for HgCdTe photodetectors with 60% QE.20
The suppression of the dark current is due to blocking of the
majority carriers, long minority carrier lifetime and low hole

Appl. Phys. Lett. 101, 161114 (2012)

mobility in the InAs/InAsSb absorber.16,17 But the device
performance is still inferior to that of HgCdTe devices with
similar cutoff wavelength in terms of D* and QE due to limited absorber thickness.20,21
In summary, nBn long-wave infrared photodetectors
based on InAs/InAs0.62Sb0.38 type-II strained-layer superlattice have been demonstrated. The unpassivated nBn detectors have a 50% cutoff wavelength of 13.2 lm, a dark
current density of 5  104 A/cm2, and a dark current GR
noise limited D* of 1  1010 cm Hz1/2/W at 77 K. Although
the D* and QE (2.5%) are still lower than those of HgCdTe
devices with similar cutoff wavelength, the measured device
dark current density is below the Rule-07 proposed for
HgCdTe photodetectors and is the lowest value reported so
far for any T2SL at similar wavelengths.21 These results
indicate that higher operating temperature of type-II superlattice LWIR photodetectors can be demonstrated experimentally in the near future.
The authors acknowledge funding by the AFOSR under
Grant No. FA9550-10-1-0129 and U.S. Army Research
Office MURI program under Grant No. W911NF-10-1-0524.
1

P. Norton, Opto-Electron. Rev. 10, 159 (2002).
D. L. Smith and C. Mailhiot, Appl. Phys. Lett. 62, 2545 (1987).
3
G. C. Osbourn, J. Appl. Phys. 53, 1586 (1982).
4
P. Manurkar, S. Ramezani-Darvish, B.-M. Nguyen, M. Razeghi, and J.
Hubbs, Appl. Phys. Lett. 97, 193505 (2010).
5
E. R. Youngdale, J. R. Meyer, C. A. Hoffman, F. J. Bartoli, C. H. Grein, P.
M. Young, H. Ehrenreich, R. H. Miles, and D. H. Chow, Appl. Phys. Lett.
64, 3160 (1994).
6
C. H. Grein, P. M. Young, M. E. Flatte, and H. Ehrenheich, J. Appl. Phys.
78, 7143 (1995).
7
J. R. Meyer, C. L. Felix, W. W. Bewley, I. Vurgaftman, E. H. Aifer, L. J.
Olafsen, J. R. Lindle, C. A. Hoffman, M.-J. Yang, B. R. Bennett, B. V.
Shanabrook, H. Lee, C.-H. Lin, S. S. Pei, and R. H. Miles, Appl. Phys.
Lett. 73, 2857 (1998).
8
J. L. Johnson, L. A. Samoska, A. C. Cossard, J. L. Merz, M. D. Jack, G. R.
Chapman, B. A. Baumgratz, K. Kosai, and S. M. Johnson, J. Appl. Phys.
80, 1116 (1996).
9
D. Z. Ting, C. J. Hill, A. Soibel, S. A. Keo, J. M. Mumolo, J. Nguyen, and
S. D. Gunapala, Appl. Phys. Lett. 95, 023508 (2009).
10
B. C. Connelly, G. D. Metcalfe, H. Shen, and M. Wraback, Appl. Phys.
Lett. 97, 251117 (2010).
11
D. Donetsky, G. Belenky, S. Svensson, and S. Suchalkin, Appl. Phys. Lett.
97, 052108 (2010).
12
D. Lackner, O. J. Pitts, M. Steger, A. Yang, M. L. W. Thewalt, and S. P.
Watkins, Appl. Phys. Lett. 95, 081906 (2009).
13
E. H. Steenbergen, Y. Huang, J.-H. Ryou, L. Ouyang, J.-J. Li, D. J. Smith,
R. D. Dupuis, and Y.-H. Zhang, Appl. Phys. Lett. 99, 071111 (2011).
14
E. H. Steenbergen, K. Nunna, L. Ouyang, B. Ullrich, D. L. Huffaker, D. J.
Smith, and Y.-H. Zhang, J. Vac. Sci. Technol. B 30, 02B107 (2012).
15
O. O. Cellek, H. Li, X.-M. Shen, Z. Y. Lin, E. H. Steenbergen, D. Ding, S.
Liu, Q. Zhang, H. S. Kim, J. Fan, M. J. DiNezza, W. H. G. Dettlaff, P. T.
Webster, Z.-Y. He, J.-J. Li, S. R. Johnson, D. J. Smith, and Y.-H. Zhang,
Proc. SPIE 8353, 83533F (2012).
16
E. H. Steenbergen, B. C. Connelly, G. D. Metcalfe, H. Shen, M. Wraback,
D. Lubyshev, Y. Qiu, J. M. Fastenau, A. W. K. Liu, S. Elhamri, O. O. Cellek, and Y.-H. Zhang, Appl. Phys. Lett. 99, 251110 (2011).
17
S. Maimon and G. W. Wicks, Appl. Phys. Lett. 89, 151109 (2006).
18
H. S. Kim, E. Plis, J. B. Rodriguez, G. D. Bishop, Y. D. Sharma, L. R.
Dawson, S. Krishna, J. Bundas, R. Cook, D. Burrows, R. Dennis, K.
Patnaude, A. Reisinger, and M. Sundaram, Appl. Phys. Lett. 92, 183502
(2008).
19
W. E. Tennant, D. Lee, M. Zandian, E. Piquette, and M. Carmody, J. Electron.
Mater. 37, 1406 (2008).
20
W. E. Tennant, J. Electron. Mater. 39, 1030 (2010).
21
D. R. Rhiger, J. Electron. Mater. 40, 1815 (2011).
2

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
209.147.144.22 On: Wed, 04 Feb 2015 17:14:25