nBn based infrared detectors using type-II InAs/(In,Ga)Sb superlattices

nBn Based Infrared Detectors Using Type-II
InAs/(In,Ga)Sb Superlattices
E. Plis a , H.S Kim a , J.B. Rodriguez b , G. D. Bishop a , Y. D. Sharma a , A. Khoshakhlagh a ,
L. R. Dawson a , J. Bundas c , R. Cook c , D. Burrows c , R. Dennis c , K. Patnaude c ,
A. Reisinger c , M. Sundaram c S. Krishna a1
a Center for High Technology Materials, Department of Electrical and Computer Engineering,
University of New Mexico, Albuquerque, NM 87106
b Institut d'Electronique du Sud (IES), UMR CNRS 5214, cc 067, Université Montpellier 2, 34095
Montpellier Cedex 05, France
c QmagiQ, LLC, One Tara Boulevard, Suite 102, Nashua, New Hampshire 03062
ABSTRACT
The development of type-II InAs/(In,Ga)Sb superlattice (SL) detectors with nBn design for single-color and
dual-color operation in MWIR and LWIR spectral regions are discussed. First, a 320 x 256 focal plane array (FPA) with
cutoff wavelength of 4.2 µm at 77K with average value of dark current density equal to 1 x 10 -7 A/cm 2 at V b =0.7V (77
K) is reported. FPA reveals NEDT values of 23.8 mK for 16.3 ms integration time and f/4 optics. At 77K, the peak
responsivity and detectivity of FPA are estimated, respectively, to be 1.5 A/W and 6.4 x 10 11 Jones, at 4 µm. Next,
implementation of the nBn concept on design of SL LWIR detectors is presented. The fabrication of single element nBn
based long wave infrared (LWIR ) with λ c ~ 8.0 µm at V b = +0.9 V and T= 100K detectors are reported. The bias
dependent polarity can be exploited to obtain two color response (λ c1 ~ 3.5 µm and λ c2 ~ 8.0 µm) under different polarity
of applied bias. The design and fabrication of this two color detector is presented. The dual band response (λ c1 ~ 4.5 µm
and λ c2 ~ 8 µm) is achieved by changing the polarity of applied bias. The spectral response cutoff wavelength shifts
from MWIR to LWIR when the applied bias voltage varies within a very small bias range (~100 mV).
Keywords: thermal imaging, focal plane array, InAs/(In,Ga)Sb type-II superlattices, single color, multispectral, nBn
detector.
1. INTRODUCTION
Mid-wave-infrared (MWIR) and long-wave-infrared (LWIR) single element detectors and focal plane arrays
(FPAs) are widely used in a variety of fields, including medical diagnostics, climatology and terrestrial pollution
monitoring. Presently, the most used detectors for these applications are mercury cadmium telluride (MCT), InSb or
intersubband quantum well infrared (QWIP) detectors. However, lack of spatial uniformity over a large area still plagues
InSb and MCT devices, while QWIPs have larger dark currents and lower quantum efficiency (QE) compared to the
interband devices.
1 skrishna@chtm.unm.edu, phone (505) 272 7892; fax (505) 272 7801
Infrared Technology and Applications XXXIV, edited by Bjørn F. Andresen, Gabor F. Fulop, Paul R. Norton,
Proc. of SPIE Vol. 6940, 69400E, (2008) · 0277-786X/08/$18 · doi: 10.1117/12.780375
Proc. of SPIE Vol. 6940 69400E-1

Proposed by Smith and Mailhiot 1 in the 1980s, detectors based on InAs/GaSb superlattices (SLs) have attracted
a lot of attention over the past few years as a possible alternative to present-day IR detection systems. The InAs/GaSb
SLs consist of alternating layers of nanoscale materials whose thicknesses vary from 4-20 nm. These heterostructures are
characterized by a broken-gap type-II alignment as illustrated on the Figure 1. The overlap of electron (hole) wave
functions between adjacent InAs (GaSb) layers results in the formation of an electron (hole) minibands in the conduction
(valence) band. Optical transitions between holes localized in GaSb layers and electrons confined in InAs layers are
employed for the detection of infrared radiation. The effective bandgap of the InAs/InGaSb SLs can be tailored from
3µm to 30 µm by varying thickness of constituent layers, thus allowing fabrication of devices with operating
wavelengths spanning the entire IR region
The effective mass of the superlattice is not dependent on the bandgap, unlike the case of bulk semiconductor
material. In contrast with QWIPs, normal incidence absorption is permitted in SLs, resulting in high quantum efficiency.
The larger effective mass in SLs leads to a reduction of tunneling currents. Large splitting between heavy-hole and lighthole
valence subbands due to strain in the SLs contributes to the suppression of Auger recombination. Commercial
availability of low defect density substrates and a high degree of uniformity for III-V processing over a large area also
offers technological advantages for the SLs. This makes detectors based on SLs an attractive technology for the
realization of high performance single element detectors and FPAs.
Presently all SLs detectors are based on a photodiode (p-on-n or n-on-p) design. In this case, the optically active
area of the photodiode is defined by an etched mesa as shown on the Figure 2 (a). During the mesa isolation process, the
periodic nature of the idealized crystal structure ends abruptly at the mesa lateral surface. Disturbance of the periodic
potential function due to a broken crystal lattice leads to allowed electronic quantum states within the energy band gap of
SLs, resulting in large surface leakage currents. The suppression of these currents is the most demanding challenge for
present day SLs technology, especially for LWIR and VLWIR spectral regions, since the dimensions of SLS pixels have
to be scaled to ~ 20 µm in FPAs.
In order to overcome the limitation imposed by surface leakage currents, a stable surface passivation layer is
needed. So far, various approaches have been proposed such as the deposition of polyimide layer 2 , overgrowth of wide
band gap material 3 , deposition of passivation sulphur coating electrochemically 4 and post etch treatment in chemical
solutions 5 . However, these passivation methods are either sensitive to the cut-off wavelength of the device or complicates
the fabrication process of detectors and FPAs.
InAs
InAs
Electron
miniband E
Cl
Hole —
miniband
Yl
iHHn1n1r
— I— I — I —
__________________
$1
I I
Absorption
GaSb
Figure 1. Schematic bandgap alignment of Type II InAs/GaSb superlattices (SLs).
Recently 6 , a new heterostructure design with extremely low surface currents has been proposed. This socalled
nBn structure consists of a n-type narrow band-gap contact and absorber layers separated by a 50-100 nm thick
wide band-gap barrier layer. A 100K increase in the background limited infrared photodetector (BLIP) temperature has
been demonstrated. Implementation of the nBn design for InAs/GaSb MWIR SLs single element detectors has been
reported 7,8 . Unlike a conventional photodiode processing, the size of the device with nBn design is not defined by the
Proc. of SPIE Vol. 6940 69400E-2

etch dimensions but, rather by the lateral diffusion length of minority carriers (holes), as illustrated in Figure 2(b).
SLs-based detectors with nBn design and special processing scheme showed a dark current reduction of two orders of
magnitude (at 77K) in comparison with conventional photodiode processing and values of quantum efficiency and
shot-noise-limited specific detectivity comparable to the state-of-the-art p-i-n diodes based on the SLs. However, no
FPAs based on InAs/GaSb SLs detectors with nBn design have been reported yet.
In the present work we report on our efforts to fabricate FPAs based on InAs/GaSb SL nBn-design detectors
for the MWIR spectral region. Also, we extend the nBn-design concept using InAs/InGaSb SLs in order to achieve
LWIR detection.
Metallization
Metallization
P-Contact
N-Contact
Barrier
J surface
Absorber
DL
DL
J bulk
N-Contact
Substrate
Absorber
(a)
(b)
Figure 2. Schematic of (a) conventionally defined mesa (b) shallow etched isolation nBn device. In the latter case
active area is defined by the by diffusion length (DL) of the minority carriers (holes)
2. NBN InAs/GaSb SLS MWIR FPA IMAGERS
N-Contact
Substrate
Substrate
The devices presented in this work were grown on Te-doped epi-ready (100) GaSb substrates using a solid
source molecular beam epitaxy VG-80 system. The system was equipped with SUMO ® cells for gallium and indium, a
standard effusion cell for aluminum and cracker cells for antimony and arsenic. Growth rates were calibrated by
monitoring the intensity oscillations in the reflected high-energy electron diffraction (RHEED) patterns and confirmed
by growth of calibration SL samples with different thickness of SL period. More details about the growth procedure can
be found elsewhere 9 . Growths were performed on 2” diameter unintentionally doped epi-ready n-type GaSb (001)
substrates. The detector structure consists of a 100 nm thick Al 0.2 Ga 0.8 Sb etch stop layer followed by a 360 nm bottom
contact layer formed by 8 monolayers (MLs) InAs:Si (n = 4 x 10 18 cm -3 )/8 MLs GaSb SLS. Then a 2.4 µm thick nonintentionally
doped (n.i.d.) absorber formed by 8 MLs InAs/ 8 MLs GaSb SLS was grown followed by a 100nm thick
Al 0.2 GaSb barrier layer. The structure was terminated by a 100 nm n-type (n = 4 x 10 18 cm -3 ) top contact layer with the
same composition as the bottom contact layer. The heterostructure schematic is shown in Figure 3.
J bulk
Proc. of SPIE Vol. 6940 69400E-3

SL (n) 100 nm
Contact
Al 0.2 GaSb 100 nm
Barrier
SL nid 2.4 µm
Absorber
SL (n) 360 nm
Contact
Al 0.2 GaSb 100 nm
Etch stop layer
GaSb:Te 2”
Substrate
Figure 3. Heterostructure schematic of InAs/GaSb SLs MWIR detector
Each processed FPA die consists of 320 x 256 pixels with a 30 µm pitch. Processing was initiated by defining
24 µm x 24 µm squares with standard UV photolithography and a shallow wet chemical etch using phosphoric acid was
performed. It is to be noted that the depth of the shallow etch was equal to 0.15 nm, which corresponds to the middle of
the barrier layer. Thus the active absorber layer underneath is untouched. Then an inductively coupled plasma (ICP) dry
etch to the middle of the bottom contact layer on the three outermost rows and columns of the FPA was undertaken. A
scanning electron microscope (SEM) image of a part of a fully processed FPA is shown in Figure 4 and illustrates the
two steps of the etching process. Top and bottom contacts were then deposited using an electron beam metal evaporation
system. We used Ti\Pt\Au (500Å\500Å\3000Å) as contact metals for both top and bottom ohmic contact metallization.
Finally, to enable well defined Indium bumps, an under bump metal (UBM) deposition was conducted using Ti\Ni\Au
(300Å\1500Å\500Å). Indium bumps with a thickness ~3 µm were thermally evaporated on the UBM metal pads.
Following this, the FPAs were hybridized to ISC0209 read-out integrated circuits (ROICs) made by Indigo.
To reduce the free carrier absorption in the substrate and to minimize the thermal stress between the FPA and
the ROIC under cool-down conditions, the backside of the FPAs were thinned by a combination of mechanical polishing
and wet chemical etching. Hybridization and characterization of the FPAs were undertaken at QmagiQ, LLC. Hybrid
FPAs were tested with a CamIRa infrared FPA evaluation system made by SE-IR Corp. with a Ge window for the
imaging performance tests. For the tests, the background was a 300K scene under f/4 illumination. The applied bias was
equal to 0.7V and is defined as a positive voltage applied to the bottom contact of the FPA.
The noise equivalent temperature difference (NETD) was measured by acquiring a sequence of 25 image scans
with the focal plane exposed to a uniform blackbody target at 30°C and extracting the temporal noise for each pixel. A
histogram of the NETD distribution is shown in Figure 5 (a) for an integration time of 16.3 ms. Good spatial uniformity
is obtained over the array, resulting in a median NETD value of 23.8 mK with a standard deviation of 10 mK at a FPA
temperature of 77K. Pixel operability was equal to 79% with predominant single pixel faults without large clusters.
Pixel outages probably originated from the weak connection between the ROIC and FPA since the thickness of indium
bumps was equal only 3 µm and no reflow process was utilized.
Proc. of SPIE Vol. 6940 69400E-4

fln 1 na
Efl 1 fl
smai
nna I
nwrinn
winfln nfl
a
ncnn
r ri Etn
30µ
m
Figure 4. Scanning Electron Microscopy (SEM) image showing part of fully processed FPA with deposited indium
bumps. The two-step etch process is illustrated.
The measured average dark current density is presented in Figure 5(b). Extrapolated to an operating
temperature of 77K, the average dark current density was as low as 1 x 10 -7 A/cm 2 (V b =0.7V). This value is comparable
to the state-of-the-art reports for MWIR detectors based on a conventional photodiode design with a suitable passivation
scheme.
The external quantum efficiency (QE) was obtained using the spectral response curve from the test diode on the
FPA chip with 380µm x 380µm optical area. Spectral measurements were performed using a Fourier Transform Infrared
Spectrometer (FTIR). Under 0.7V bias and at 77K the FPA exhibits a QE as high as 52% at 3.8 µm (Fig. 6). Peak
responsivity and detectivity of FPA were estimated, respectively, to be 1.6 A/W and 6.7 x 10 11 Jones, at 3.8 µm and 77K
(V b =0.7V). Values of specific detectivity, QE and responsivity as a function of wavelength are presented in Figure 7.
The responsivity was calculated assuming no photoconductive gain.
ZOS1SVINfl - mea juaiin Apsua
6000.
5000.
Mean vaIue23.8mK
3
yTT TTT TT
u000
3000.
a)
E
2000.
z
1000.
0.
iL25 50 75 100
a
a
a
C
a
L09
9L
09
T V
99
YVY
TV
VT
06
ainwiadwaj. j)
96
OO
9O
NEdT(mK)
Figure 5. (a) Histogram of the noise equivalent temperature difference (NETD) distribution in the FPA for an
integration time of 16.3 ms with f/4 optics (b) Average dark current density of FPA.
A thermal image taken with 320 x 256 MWIR FPA camera based on InAs/GaSb SL at a detector temperature of
77 K and integration time of 16.3 ms is shown in Figure 7. A two point non-uniformity correction (NUC) was used for
the imaging. Temperatures of 20°C and 40°C were utilized as the low and high temperature for the NUC correction
Proc. of SPIE Vol. 6940 69400E-5

algorithm. The bright areas of the image represent warmer regions whereas the dark areas exhibit colder regions. In the
figure, the facial imprint of the fingerprints after touching the cold can are clearly visible, demonstrating the good
imaging quality of thermal imager.
1.6
1.6
Responsivity, D*
Quantum efficiency
T = 77K
V b
= 0.7V
7x10 11
6x10 11
Quantum efficiency
1.2
0.8
Responsivity (A/W)
1.2
0.8
0.4 0.4
5x10 11
4x10 11
3x10 11
2x10 11
1x10 11
D* (Jones)
0.0
0.0
1 2 3 4 5 6
Wavelength (µm)
0
Figure 6. Spectral detectivity D*, responsivity and quantum efficiency for 4.2 µm cut-off wavelength FPA
Figure 7. Thermal image taken with 320x256 InAs/GaSb SLs nBn camera at a detector temperature of 77 K and
integration time of 16.3 ms. A two-point non-uniformity correction was utilized.
3. NBN InAs/GaSb SLS LWIR SINGLE-ELEMENT DETECTORS
The structure of the nBn InAs/GaSb SLs LWIR detector is similar to that described in the previous section except for
the composition of the absorber SL. The structure consists of a 100 nm thick Al 0.2 Ga 0.8 Sb etch stop layer followed by a
480 nm bottom contact layer formed by 8 MLs InAs:GaTe (n = 4 x 10 18 cm -3 )/8 MLs GaSb:GaTe SLS. Then a 1.9 µm
thick non-intentionally doped (n.i.d.) absorber formed by 9 MLs InAs/ 5 MLs GaIn 0.25 Sb SLS was grown, followed by a
100nm thick Al 0.2 GaSb barrier layer. A 100 nm n-type top contact layer with the same composition and doping level as
the bottom contact layer capped the structure. The heterostructure schematic of the LWIR nBn detector is presented in
Figure 8 (a) .
Proc. of SPIE Vol. 6940 69400E-6

SL (n) 100 nm
MWIR Contact
Al 0.2 GaSb 100 nm
Barrier
SL nid 2.4 µm
LWIR Absorber
SL (n) 480 nm
MWIR Contact
Counts/s
SL 0
SL LWIR
10 4 0
GaSb
(004)
SL MWIR
10 4 10
GaSb
+1
3
-1
10 2
-2
10 3
Ω/2θ ( ο )
-3
+2
10 2
-4
+3
10 1
+4
Counts/s
30.6 30.9 31.2
0
Al 0.2 GaSb 100 nm
Etch stop layer
GaSb:Te 2”
Substrate
10 0
26 28 30 32 34 36
Ω/2θ ( ο )
(a)
(b)
Figure 8. (a) The heterostructure schematic and (b) (004) XRD scan of LWIR SLs detector
Symmetric (004) X-ray scan was performed on the sample with a Philips double-crystal X-ray diffractometer
using the Cu-K α1 line and is presented in Figure 8 (b). Peaks from both SLs forming the contacts (MWIR) and absorbing
layer (LWIR) are visible. The LWIR superlattice is slightly compressively strained, with a lattice mismatch to the GaSb
substrate equal to 0.2%. The full width at half maximum (FWHM) of the 1 st satellite peak of LWIR SL is equal to 32
arcsec, thus indicating the good crystalline quality of the material. The overall periodicity of the structure, measured by
the fringe spacing of the superlattice peaks, is found to be 39.2 Å and 48.4 Å for the LWIR and MWIR SLs, respectively.
These values are in a good agreement with the nominal ones.
Normalized spectral response (a.u.)
1.0
0.8
0.6
0.4
0.2
x 50
V b
= -0.7 volts
V b
= 0.9 volts
0.0
1 2 3 4 5 6 7 8 9 10 11
Wavelength (µm)
100 K
150 K
Top
contact
Top contact
Barrier
(b)
Barrier
LWIR
absorber
LWIR
absorber
(a)
(c)
Figure 9. (a) Normalized spectral response of nBn LWIR detector at different temperatures under different polarities
of applied bias; Schematic band diagram for the nBn LWIR detector under (b) forward and (c) reverse bias
Proc. of SPIE Vol. 6940 69400E-7

Normal incidence single pixel photodiodes were fabricated using standard lithography with apertures ranging
from 25-300 µm in diameter. Processing was initiated with the formation of ohmic contacts on the n-type top contact
layer followed by dry etching of the device to the top of the barrier (etch depth 100 nm) for the mesa definition. Then the
wafer was patterned with the photoresist, and a deep dry etch (etch depth ~2 µm) to the middle of the bottom n-type
contact layer was performed. Finally, an ohmic contact was evaporated on the bottom contact layer. Ti (500 Å) / Pt (500
Å) /Au (3000 Å) was used as n-contact metal for both top and bottom contacts. Devices were then wire bonded to a
leadless chip carrier for further characterization.
Spectral measurements were performed using a FTIR spectrometer and a Keithley 428 preamplifier. Figure 9
shows the normalized spectral response (obtained by dividing the photocurrent of the SLS detector with that obtained
using a pyroelectric detector) for a 300 µm diameter device for two different temperatures at the different polarities of
applied bias. For the nBn detector structure, forward bias is defined as a negative voltage applied to the top contact of
the detector. As depicted in Figure 9, the nBn LWIR detector structure demonstrated two color response (λ c1 ~ 3.5 µm
and λ c2 ~ 8.0 µm) under different polarity of applied bias.
Schematic band diagrams for the nBn LWIR detector under forward (negative voltage on the top) and reverse
(positive voltage on the top) biases are shown in Figure 9(b) and Figure 9(c), respectively. Under forward bias, the photo
carriers are collected from the absorber. When the device is under reverse bias, the photo carriers from the heavily doped
(n-type) contact layer are collected, while those from the absorber are blocked by the barrier. Heavily doped InAs in the
top contact layer results in a larger optical bandgap due to the Moss-Burstein effect and is the source of the MWIR
signal. Thus, the two color response come from the detector absorber (LWIR signal) and top contact layer (MWIR
signal).
4. NBN INAS/GASB SLS DUAL COLOR (MWIR/LWIR) DETECTORS
On the basis of the results obtained from the MWIR and LWIR nBn detectors, we designed a two color nBn
structure 10 . The heterostructure schematic of such structure is presented in Figure 10 (a). The growth procedure started
with the deposition of a 480 nm bottom contact layer formed by 8 ML InAs:GaTe (n=4 x 10 18 cm −3 )/8 ML GaSb SLS.
Then a 1.8 µm thick LWIR absorber consisting of unintentionally doped 9 ML InAs/5 ML In 0.25 Ga 0.75 Sb SLS was grown
followed by a 1.5 µm thick MWIR absorber composed of 8 ML InAs/8 ML GaSb SLS. The thicknesses of the LWIR and
MWIR absorbers were designed to be approximately the same. A 100 nm Al0.2Ga0.8Sb barrier separated the two
absorbers. The structure was capped with a ~ 0.1 µm top contact layer with the same composition and doping level as the
bottom contact layer.
SL (n) 97 nm
MWIR Contact
SL nid 1.5 µm
MWIR Absorber
Al 0.2 GaSb 100 nm
Barrier
SL nid 1.8 µm
LWIR Absorber
SL (n) 480 nm
MWIR Contact
GaSb:Te 2”
Substrate
Counts/s
GaSb
10 4 SLMWIR SL LWIR
0
0
+1 LWIR
10 3
10 2
10 1
+1 MWIR
-1 LWIR
-2 LWIR -1 MWIR
-2 MWIR +2 MWIR
+2 LWIR
AlGaSb Barrier
10 0
27 28 29 30 31 32 33 34
Ω/2θ ( ο )
(a)
(b)
Figure 10. (a) The heterostructure schematic and (b) (004) XRD scan of multispectral (LWIR/MWIR) nBn SLs
detector
Proc. of SPIE Vol. 6940 69400E-8

Symmetric (004) X-ray scan was performed on the multispectral nBn SLs sample and is presented in Figure 8
(b). The lattice mismatch of LWIR and MWIR SLs to the GaSb substrate is 0.13% (compressive strain) and –0.04%
(tensile strain), respectively. The full width at half maximum (FWHM) of the 1 st satellite peak of the LWIR SL is equal
to 25 arcsec (45 arcsec for the MWIR SLs). The periods of LWIR and MWIR SLs were measured from the fringe
spacing of the SLs peaks and were found to be 42.2 Å and 48.2 Å for the LWIR and MWIR SLs, respectively.
The detector material was processed to operate at normal incidence as single pixel photodiodes with apertures
ranging from 25 to 300 µm. The processing procedure was similar to that described earlier for nBn LWIR single element
detectors in Section 2.
The relative spectral responses of the bicolor structure are shown in Fig. 11 (a). The data clearly show cutoff
wavelengths λ c1 ~ 4.5 µm and λ c2 ~ 8 µm under different polarities of applied bias. When a forward bias is applied, the
carriers from the LW absorber are collected, and when a reverse bias is applied, the carriers from the MW absorber are
collected. It is to be noted that there will be barriers between unintentionally doped LWIR layer and the bottom n-contact
that can impede the transport.
An interesting feature in the detector’s responsivity vs. wavelength behavior was noticed while the spectral
response of the multispectral nBn detector within the range of applied bias from 0 to + 0.1V was being investigated
(Figure 11 (b)). The data clearly indicate that the response cutoff wavelength shifts from MWIR to LWIR when the
applied bias voltage varies between 0 and +0.1 V. As the applied bias voltage increases from 0 to +0.1 V, the intensity of
the MWIR response decreases whereas the intensity of LWIR response increases. Due to this bias dependent feature of
the spectral response, a single detector can be operated at multiple biases sequentially, whereby the detector’s spectral
response changes in accordance with the applied bias. Consequently, one of the advantages of nBn design is that it does
not require multiple contacts per pixel, which reduces the cost and complexity associated with the fabrication process.
Normalized spectral response (a.u.)
1.0
0.8
0.6
0.4
0.2
0.0
V b
= + 0.5 V
V b
= -0.6 V
3 4 5 6 7 8 9 10
Wavelength (µm)
100 K
150 K
Spectral response (a.u)
0.06
0.05
0.04
0.03
0.02
0.01
0.00
2 3 4 5 6 7 8 9 10
Wavelength(µm)
0 V
+ 0.01 V
+ 0.05 V
+ 0.09 V
+ 0.1 V
Figure 11. (a) Normalized spectral response of the multispectral nBn detector at different temperatures and different
polarity of applied bias (b) Spectral response of the multispectral nBn detector within a range of applied bias from 0 to
+0.1V
5. SUMMARY AND CONCLUSION
The development of nBn based detectors for MWIR and LWIR spectral regions using type-II InAs/(In,Ga)Sb
SL has been presented. First, we demonstrated a 320 x 256 FPA with cutoff wavelength of 4.2 µm at 77K. The average
value of the dark current density (1 x 10 -7 A/cm 2 at 77K (V b =0.7V) proved comparable to that reported for the state-ofthe-art
MWIR SLs photodiodes utilizing some passivation schemes. At 77K, FPA reveals an average NEDT of 23.8 mK
Proc. of SPIE Vol. 6940 69400E-9

for 16.3 ms integration time. The peak responsivity and detectivity of FPA were estimated to be 1.6 A/W and 6.7 x 10 11
Jones, respectively, at 3.8 µm and 77K.
We then described our implementation of the nBn concept to design SL LWIR detectors. Within addition to the
expected long-wave cut off wavelength, the nBn LWIR detector structure demonstrated two color response (λ c1 ~ 3.5 µm
and λ c2 ~ 8.0 µm) under different polarity of applied bias.
Finally, a multispectral nBn detector structure has been presented. The dual band response (λ c1 ~ 4.5 µm and
λ c2 ~ 8 µm) was achieved by changing the polarity of applied bias. The spectral response shows a significant change in
the LWIR to MWIR ratio within a very small bias range (100 mV), making it compatible with commercially available
readout integrated circuits.
6. ACKNOWLEDGEMENT
Support from AFRL Grant FA9453-07-C-0171 and HOT MWIR program is acknowledged.
1. D.L. Smith and C. Mailhiot, J. Appl. Phys., 62, 2545 (1987).
2. A. Hood, P.-Y. Delaunay, D. Hoffman, B.-M. Nguyen, Y. Wei and M. Razeghi, Appl. Phys. Lett. 90, 233513 (2007).
3. R. Rehm, M. Walther, F. Fuchs, J. Schmitz and J. Fleissner, Appl. Phys. Lett., 86,173501 (2005).
4. E. Plis, J.B. Rodriguez, S.J. Lee and S. Krishna, Electron. Lett., 42, 1248 (2006).
5. A. Gin, Y. Wei, A. Hood, A. Bajowala, V. Yazdanpanah and M. Razeghi, Appl. Phys. Lett., 84, 2037 (2004).
6 . Maimon and G. W. Wicks, Appl. Phys. Lett. 89, 151109 (2006).
7 J.B. Rodriguez, E. Plis, G. Bishop, Y.D. Sharma, H. Kim, L.R. Dawson and S. Krishna, Appl. Phys. Lett. 91, 043514
(2007).
8 G. Bishop, E. Plis, J.B. Rodriguez, Y. D. Sharma, H.S. Kim, L. R. Dawson, and S. Krishna, J. Vac. Sci. Technol., B 26
(3), 1 (2008)
9. E. Plis, S. Annamalai, K. T. Posani, S. Krishna, R. A. Rupani, and S. Ghosh, J. Appl. Phys., 100, 014510 (2006).
10. A. Khoshakhlagh, J. B. Rodriguez, E. Plis, G. D. Bishop, Y. D. Sharma, H. S. Kim, L. R. Dawson, and S. Krishna,
Appl. Phys. Lett. 91, 263504 (2007).
Proc. of SPIE Vol. 6940 69400E-10