We have performed a numerical study of the carrier transport properties in gallium-free strained-balanced InAs/InxAs1-xSb type-II superlattices (T2SL) intended for infrared detector applications. The motivation was to understand whether the performance of T2SL detectors is ultimately limited by fundamental or technological processes. We have employed a rigorous one-dimensional quantum mechanical transport model based on the non-equilibrium Green's function (NEGF) formalism that includes a k.p description of the electronic structure. This approach has allowed us to avoid making any a priori assumptions on the physical mechanism (tunneling, sequential tunneling or hopping) dominating the transport. We find that regardless of its nature and cause, the presence of positional and compositional disorder, introduced inherently during materials growth by layer thickness fluctuations, nonuniform antimony composition and segregation throughout the superlattice stack, significantly affects the vertical carrier transport properties. In particular, the minority carrier hole mobility is fundamentally limited by the nonideal disorder. Furthermore, upon reducing the temperature, holes become fully localized and transport occurs by hopping, which explains published measured photodetector data that demonstrates the quantum efficiency exhibiting a very strong temperature dependence that degrades as the temperature is reduced. We also found that the minority carrier electron mobility is largely unaffected by disorder, indicating the p-type absorbing layer as the preferred option.
We have also performed simulations of typical nBn detectors with a simplified NEGF model based on the effective mass approximation and Büttiker-probe self-energies, which also allows us to explore the possibilities offered by quantum-corrected drift-diffusion approaches. We will present simulation examples of representative gallium-containing and gallium-free nBn structures intended for MWIR and LWIR photodetector applications. These examples will include calculations of I-V characteristics, SRH recombination rates and quantum efficiency.
Future and advanced sensor technologies needed for DoD applications will require more efficient semiconductor materials and devices. Pushing sensor device performance beyond present levels requires a deep understanding of the fundamental limiters. Therefore fundamental research is needed to assure transition of technology from demonstration to system deployment. To address this problem, the Army Research Laboratory (ARL) and Boston University (BU) have come together to create a BU led Consortium for semiconductor Modeling of Materials and Devices (CSM). The Consortium brings together government, academia, and industry in a collaborative fashion to continuously push semiconductor research forward to meet DoD needs. The leveraged attributes of the Consortium include combined broad knowledge base in semiconductor modeling, materials growth and device expertise; sharing of computational resources; project continuity; and extension of the bench. Details regarding the Consortium’s first research topic on understanding vertical transport in Type 2 SL will be discussed.
Timely technology transition with minimal risk requires an understanding of fundamental and technology limitations of material synthesis, device operation and design controllable parameters. However, this knowledge-based approach requires substantial investment of resources in the Science and Technology (ST) stage of development. For low volume niche semiconductor technologies of Department of Defense (DoD) relevance, there is little drive for industry to expend their limited resources towards basic research simply because there is no significant return on investment. As a result, technology transition from ST to product development is often delayed, expensive and carries risks. The Army Research Laboratory (ARL) is addressing this problem by establishing a Center for Semiconductor Modeling of Materials and Devices (CSM) that brings together government, academia, and industry in a collaborative fashion to address research opportunities through its Open Campus initiative. This Center leverages combined core competencies of partner organizations, which include a broad knowledge base in modeling, and its validation; sharing of computational, characterization, materials growth and device processing resources; project continuity; and ‘extension of the bench’ via exchange of researchers between affiliated entities. A critical DoD technology is sensing in the infrared (IR) spectrum, where understanding of materials, devices and methods for sensing and processing IR information must continually improve to maintain superiority in combat. In this paper we focus on the historical evolution of IR technology and emphasize the need for understanding of material properties and device operation to accelerate innovation and shorten the cycle time, thereby ensuring timely transition of technology to product development and manufacturing. There are currently two competing IR technologies being pursued, namely the incumbent II-VI Hg1- xCdxTe technology and the III-V Type 2 Superlattices (SLs) technology. A goal of the CSM is to develop physics based models for Type 2 SLs with the capability to timely understand the knowledge gap between what is built and what is designed.
To increase Soldier readiness and enhance situational understanding in ever-changing and complex environments, there is a need for rapid development and deployment of Army technologies utilizing sensors, photonics, and electronics. Fundamental aspects of these technologies include the research and development of semiconductor materials and devices which are ubiquitous in numerous applications. Since many Army technologies are considered niche, there is a lack of significant industry investment in the fundamental research and understanding of semiconductor technologies relevant to the Army. To address this issue, the US Army Research Laboratory is establishing a Center for Semiconductor Materials and Device Modeling and seeks to leverage expertise and resources across academia, government and industry. Several key research areas—highlighted and addressed in this paper—have been identified by ARL and external partners and will be pursued in a collaborative fashion by this Center. This paper will also address the mechanisms by which the Center is being established and will operate.
The US Army’s future operating concept will rely heavily on sensors, nano-electronics and photonics technologies to rapidly develop situational understanding in challenging and complex environments. Recent technology breakthroughs in integrated 3D multiscale semiconductor modeling (from atoms-to-sensors), combined with ARL’s Open Campus business model for collaborative research provide a unique opportunity to accelerate the adoption of new technology for reduced size, weight, power, and cost of Army equipment. This paper presents recent research efforts on multi-scale modeling at the US Army Research Laboratory (ARL) and proposes the establishment of a modeling consortium or center for semiconductor materials modeling. ARL’s proposed Center for Semiconductor Materials Modeling brings together government, academia, and industry in a collaborative fashion to continuously push semiconductor research forward for the mutual benefit of all Army partners.
Since the late '60's Teledyne Imaging Sensors (TIS-formerly Rockwell Science Center) has developed IR
sensor technology and produced IR sensors for both military and commercial applications. In the late '70's,
after excursions into the Pb-salts and InAsSb alloys, TIS began to study HgCdTe and has pursued this
materials system aggressively ever since. Beginning with Te-corner liquid phase epitaxy (LPE) by dipping,
tipping, and sliding, Teledyne migrated through metal organic chemical vapor deposition (MOCVD)-a
very challenging growth technique-to molecular beam epitaxy (MBE), where we have found a reliable
and flexible technique suited to the most advanced architectures. We used substrates from Cd(Zn)Te to
sapphire, GaAs, and silicon. Ion implantation and planar diode architectures have allowed high density
device geometries exploited in our double layer planar heterostructure (DLPH) single color diodes and our
simultaneous multispectral integrated technology (SUMIT) two color diodes. The performance of these
devices equals or exceeds that of all baseline MCT devices reported by other techniques. These devices
have dark currents that are readily characterized over 13 orders of magnitude by a simple heuristic, "Rule
07," for a wide range of temperature and wavelength.
Teledyne Imaging Sensors develops and produces high performance silicon-based CMOS image sensors, with associated
electronics and packaging for astronomy and civil space. Teledyne's silicon detector sensors use two technologies:
monolithic CMOS, and silicon PIN hybrid CMOS. Teledyne's monolithic CMOS sensors are large (up to 59 million
pixels), low noise (2.8 e- readout noise demonstrated, 1-2 e- noise in development), low dark current (<10 pA/cm2 at
295K) and can provide in-pixel snapshot shuttering with >103 extinction and microsecond time resolution. The QE
limitation of frontside-illuminated CMOS is being addressed with specialized microlenses and backside illumination. A
monolithic CMOS imager is under development for laser guide star wavefront sensing. Teledyne's hybrid silicon PIN
CMOS sensors, called HyViSITM, provide high QE for the
x-ray through near IR spectral range and large arrays
(2K×2K, 4K×4K) are being produced with >99.9% operability. HyViSI dark current is 5-10 nA/cm2 (298K), and further reduction is expected from ongoing development. HyViSI presently achieves <10 e- readout noise, and new high speed
HyViSI arrays being produced in 2008 should achieve <4 e- readout noise at 900 Hz frame rate. A Teledyne 640×480
pixel HyViSI array is operating in the Mars Reconnaissance Orbiter, a 1K×1K HyViSI array will be launched in 2008 in
the Orbiting Carbon Observatory, and HyViSI arrays are under test at several astronomical observatories. The
advantages of CMOS in comparison to CCD include programmable readout modes, faster readout, lower power,
radiation hardness, and the ability to put specialized processing within each pixel. We present one example of in-pixel
processing: event driven readout that is optimal for lightning detection and x-ray imaging.
Performance of HgCdTe detector technology surpasses all others in the mid-wave and long-wave infrared spectrum. This technology is relatively mature with current effort focused on improving uniformity, and demonstrating increased focal plane array (FPA) functionality. Type-II superlattice (InAs-GaSb and related alloys) detector technology has seen rapid progress over the past few years. The merits of the superlattice material system rest on predictions of even higher performance than HgCdTe and of engineering advantages. While no one has demonstrated Type-II superlattice detectors with performance superior to HgCdTe detectors, the difference in performance between these two technologies is decreasing. In this paper, we review the status and highlight relative merits of both HgCdTe and Type-II superlattice based detector technologies.
Silicon-based hybrid CMOS visible focal plane array technology is emerging as a viable high performance alternative to scientific CCDs. The progress is attributed to the rapid advances in CMOS technology, mature precision flip-chip hybridization of large size and fine pixel arrays, and detector array performance improvements. Its technology readiness level (TRL) for space applications is being enhanced by relevant environmental tests and in-depth characterization of sensor performance. In this paper, we present recent results of Rockwell Scientific's hybrid CMOS silicon focal plane array technology, including large format arrays up to 2048x2048, broadband QE, sensor noise improvement, high radiation hardness, and the higher degree of system integration through on-chip ADCs and companion ASICs.
An attempt is made to connect the material parameters of Hg1-xCdxTe layer growth to the parameters measured following photovoltaic detector fabrication. We found that the Cd composition X value extracted from spectral response measurements on detectors at 78 K are lower than the X values obtained from the room temperature transmission measurements, or the X value used to fit the measured material minority carrier lifetime versus temperature data. The lateral collection length Lc that determines the thermally generated carriers that contribute to the diffusion current and Lopt extracted from the "flood-illuminated" to "focused-spot" photocurrent ratio are in excellent agreement. Devices exhibit near theoretical RoA uniformity at 77K for MWIR, LWIR and VLWIR. RoAopt was also found to be uniform throughout the range of detector dimensions measured such as 8 μm diameter circular to 250 μm x 250 μm square. Median RoAopt values are 1266, 66 and 0.75 ohm-cm2 for the 9.7, 11.3 and 15.4 μm cutoff wavelengths respectively. The uniformity in RoAopt confirms that the detector performance is limited by the bulk properties of the material, and not by surface effects.
Silicon-based hybrid CMOS focal plane array technology offers many advantages needed for both ground-based and space imaging applications. These advantages include enhanced UV and NIR sensitivity, extensive on-chip readout capability, inherent radiation hardness, flexible imaging readout and the ability to provide extremely low noise at high video rates. For infrared imaging applications that involve UV-through visible channels, the readout electronics commonality facilitates a great simplification to system designs. In this paper, Rockwell Scientific CMOS-based hybrid silicon FPA technology and the recent progress are presented. The hybrid FPAs developed include 640x480, 1024x1024 and 2048x2048 formats with pixel sizes ranging from 27μm to 18μm square, featuring a high optical fill factor (~100%), broad-band response (200nm to 1000nm) with high quantum efficiency, and low read noise (<6e-) that approaches astronomy CCDs at 100KHz video rate and surpasses astronomy CCDs at 1MHz rate. Other performance parameters, such as spatial uniformity, dark current, pixel crosstalk/MTF and CMOS features are also discussed.
The past 2 to 3 years has been a period of explosive growth in technology development for imaging sensors at Rockwell Scientific Co. (RSC). The state of the art has been advanced significantly, resulting in a number of unique advanced imaging sensor products. A few key examples are: 2048 x 2048 sensor chip assemblies (SCA) for ground and space-based applications, 4096 x 4096 mosaic close-butted mosaic FPA assemblies, a very high performance 10 x 1024 hybridized linear SCA for optical network monitoring and other applications, the revolutionary CMOS ProCam-HD imaging system-on-a-chip for high definition television (HDTV), and RSC's near-infrared emission microscope camera for VLSI defect detection/analysis. This paper provides selected updates of these products and thereby provides an overview of the ongoing highly fertile period of technology and product development at Rockwell Scientific. A view into future directions for advanced imaging sensors is also provided.
State-of-the-art large area photovoltaic detectors fabricated in HgCdTe grown by Molecular Beam Epitaxy have been demonstrated for the Crosstrack Infrared Sounder instrument. Large area devices (1 mm in diameter) yielded excellent electrical and optical performance operating at 81K for LWIR band and at 98K for MW and SWIR bands. LWIR and MWIR detectors have near-theoretical electrical performance, and AR-coated quantum efficiency is greater than 0.70. Measured average RoA at 98K is 2.0E7 W-cm2 and near-theoretical quantum efficiencies greater than 0.90 were obtained on SWIR detectors. These state-of-the-art large area photovoltaic detector results reflect high quality HgCdTe grown by Molecular Beam Epitaxy on CdZnTe substrates in all three spectral bands of interest.
The National Polar-orbiting Operational Environmental Satellite System (NPOESS) Cross-track Infrared Sounder (CrIS) is an interferometric sensor that measures earth radiances at high spectral resolution, using the data to provide pressure, temperature and moisture profiles of the atmosphere. The pressure, temperature and moisture sounding data are used in weather prediction models that track storms, predict levels of precipitation etc. The CrIS instrument contains SWIR ((λc approximately 5 μm at 98K), MWIR (λc approximately 9 μm at 98K) LWIR (λc approximately 16 μm at 81K) Focal Plane Array (FPA) modules. A critical CrIS design selection was the use of photovoltaic (PV) detectors in all three spectral bands. PV detectors have the important benefits of high sensitivity and linearity. Each FPA modules consists of nine large (1000 μm diameter) photovoltaic detectors with accompanying cold preamplifiers. This paper describes the performance for all the modules forming the CrIS Detector Preamplifier Module (DPM). Molecular Beam Epitaxy (MBE) is used to grow the appropriate bandgap n-type Hg1-xCdxTe on lattice matched CdZnTe. SWIR, MWIR and LWIR 1000 μm diameter detectors have been manufactured using the Lateral Collection Diode (LCD) architecture. Custom pre-amplifiers have been designed to interface with the large SWIR, MWIR and LWIR detectors. The operating temperature is above 78K, permitting the use of passive radiators in spacecraft to cool the detectors. Recently fabricated 1000 micrometers diameter photovoltaic detectors have the measured performance parameters listed in the Table below. Expected D* performance from the detector/pre-amplifier models are also listed in the table. The D* values are calculated at the CrIS program peak wavelength specified for each spectral band.
Mercury Cadmium Telluride (HgCdTe) material growth, detector array fabrication and read out integrated circuit (ROIC) design and fabrication technologies have continued to advance and have led to the demonstration of high resolution, low noise and large format reliable hybrid IR Focal Plane Arrays (IRFPAs). MBE HgCdTe-based p-on-n planar heterostructure device technology has matured to a point that high performance IRFPAs are being fabricated routinely for applications in the 1-16micrometers spectral region. Control and flexibility have proven to be distinct advantages of MBE. Rapid advances in the commercial submicron Si-CMOS process continue to allow increasing functionality on ROICs. Hybrid focal pane arrays, formed by cold welding of indium columns deposited on the detector and the ROIC, are being fabricated to suit a broad range of military, civilian and scientific applications. High performance HgCdTe/CdZnTe 256 by 256, 640 by 480 and 1024 by 1024 focal plane arrays operating over a broad range of wavelengths, temperatures, and background radiation flux, have been produced. To mitigate issues associated with the thermal expansion coefficient mismatch between Si ROIC and CdZnTe substrate, growth of HgCdTe on alternate substrates, such as Si and sapphire, has been developed for large, 1024 by 1024 and 2048 by 2048 HgCdTe FPAs operating in the Short Wavelength IR (SWIR) 0.9-2.5 micrometers and mid wavelength IR 2.5-5.5 micrometers spectral bands. Simultaneous two-color IR imaging has been proven feasible, suing MBE in situ grow multilayer structures. Hybrid visible silicon imager, where detectors are processed on silicon and hybridized to the same ROIC fabricated originally for HgCdTe devices, is emerging as a competitive technology for imagin in the 0.3-1.05 micrometers spectral region. This paper provides an overview of the status of HgCdTe materials, detectors and FPA technologies at Rockwell Science Center.
Short wavelength, middle wavelength, mid-long wavelength, long wavelength, and very long wavelength focal plane arrays (FPAs) are required for remote sensing applications. Advances in the Molecular Beam Epitaxy (MBE) growth of Mercury Cadmium Telluride (HgCdTe) and detector architectures utilized, have resulted in high performance detectors being fabricated in the 1 micrometers to 16 micrometers spectral range Custom Read Out Integrated Circuits (ROICs) are designed and fabricated to interface the HgCdTe detector arrays. The hybrid focal pane array is made up of the HgCdTe detector array and the CMOS-based ROIC. Hybrid FPAs performance parameters are presented. The HgCdTe detector material is used are grown by MBE on lattice matched CdZnTe substrates. Custom ROICs are fabricated in a commercial CMOS foundry. FPA D* performance values have been obtained for a multitude of spectral ranges and configurations that include; (i) (lambda) c equals 1.8 micrometers , 12 X 256 arrays operating at 295 K with median D* approximately 1.4 X 1012 cm Hz1/2/W, (ii) (lambda) c equals 10.5 micrometers , 256 X 256 arrays operating at 85 K with medina D* equals 3.9 X 1011 cm Hz1/2/W at a background flux (phi) b equals 7.82 X 1015 ph/cm2-2 and (iii) (lambda) c equals 15.8 micrometers at 65K, 128 X 128 array operating at 40K with peak D* of 2.76 X 1011 cm Hz1/2/W at a background flux (phi) b equals 8.0 X 1015 ph/cm2- s. The performance of these FPAs will be presented.
Rockwell is developing the world's largest HgCdTe IR focal plane array (FPA) for astronomy and low background applications. The format of the device is a hybrid 2048 X 2048 with a unit cell size of 18 micrometers X 18 micrometers . SWIR detectors with a spectral response of 0.85 micrometers to 2.5 micrometers have been processed on liquid phase epitaxy (LPE) HgCdTe on sapphire substrates. The MWIR detectors with a spectral response of 0.4 micrometers to 5 micrometers will be processed on molecular beam epitaxy HgCdTe on CdZnTe substrates. The multiplexer has been designed and fabricated at Conexant. Room temperature probing shows that the device is functional with excellent yield. Novel hybrid fabrication techniques will be used to demonstrate the FPA. This HAWAII-2 device is based on the highly successful HAWAII 1024 X 1024 device and the performance will be similar. The ultimate performance expected from the array is: dark currents of < 0.01 3-/s, quantum efficiency of > 75 percent across the spectral band, and noise levels of < 3 e- for the SWIR and < 10 e- for the MWIR band using Fowler sampling. We expected to achieve these performance levels at 77K for the SWIR and > 40K for the MWIR band. The status of the 2048 X 2048 detector arrays and FPAs are discussed.
Infrared (IR) remote sensing imaging applications require high-performance Focal Plane Arrays (FPAs) operating in all ranges of the IR spectrum. Short wavelength (SWIR; 1 to 3 micrometer), middle wavelength (MWIR; 3 to 5 micrometer), mid- long wavelength (MLWIR; 6 to 8 micrometer), long wavelength (LWIR; 8 to 14 micrometer), and very long wavelength (VLWIR; greater than 14 micrometer). These diverse spectral bands require high performance detectors and Read Out Integrated Circuits (ROICs) to perform the multi-spectral mission requirements. Significant progress in the design and fabrication of HgCdTe detector arrays and Read Out Integrated Circuits (ROICs) over the past few years has led to the demonstration of high resolution, low noise and large format reliable FPAs. Hybrid FPAs have been measured and their performance parameters are presented. Focal Plane Array D* performance values have been obtained for a multitude of spectral ranges and configurations that include; (1) (lambda) c equals 1.8 micrometer, 12 X 256 arrays operating at 295 K with median D* approximately 1.4 X 1012 cm Hz1/2/W, (2) (lambda) c equals 2.4 micrometer, 12 X 256 arrays operating at 250 K with median D* equals 1.6 X 1012 cm Hz1/2/W, detectors used are grown by MBE on lattice matched CdZnTe, (3) PACE-1 detectors with (lambda) c equals 2.5 micrometer, 1024 X 1024 arrays operating at 115 K with peak D* of 2.3 X 1013 cm Hz1/2/W at a background flux (phi) b equals 1.2 X 1011 ph/cm2- s, (4) MBE HgCdTe on Silicon MWIR detectors have been fabricated and the detector RoA performance for (lambda) co approximately 5.0 micrometer is in the 106 to 107 ohm-cm2 range at 78 K. (5) MBE HgCdTe on CdZnTe detectors, ((lambda) c equals 15.8 micrometer at 65 K), 128 X 128 array operating at 40 K with peak D* of 2.76 X 1011 cm Hz1/2/W at a background flux (phi) b equals 8.0 X 1015 ph/cm2-s. High performance 640 X 480 arrays imaging in the MWIR band have been fabricated on CdZnTe and PACE-1 substrates. The performance of these and additional FPAs will be presented.
The nucleation of CdTe onto basal plane sapphire and the subsequent growth of a CdTe buffer layer has been studied using in-situ laser reflectance (probe wavelength 633 nm, HeNe laser). The production of midwave infrared focal plane arrays requires the growth of typically 10 micrometer of CdTe (111)B buffer layer in order to grow out problems due to stacking faults, dislocation clusters and twinning. A-face and B-face growth of CdTe is seen to produce different reflectance 'signatures' within the first 6000 angstroms of growth, so enabling the early identification of problems with the growth process. Laser reflectance was also successfully demonstrated to predict the thickness of the buffer layer. Oscillations in the laser reflectance are attenuated due to absorption by the film at the probe wavelength used after approximately 6000 angstrom. However by the on-line calculation of the growth rate at every half wavelength oscillation, it is possible to extrapolate a film thickness for the total growth time. This extrapolated value is seen to be in good agreement with the thickness calculated ex-situ by beta-back scattering. The dependence on the buffer layer growth on the nucleation conditions was also investigated. The determination of whether the buffer layer grows A-face or B-face is seen to be more influenced by the II:VI ratio than the temperature during nucleation. For a nucleation temperature of 400 degrees Celsius, with a II:VI ratio of 6:1 the growth of the buffer layer is seen to be 100% A-face. As the II:VI ratio is increased the degree of A-face growth is seen to decline and the material becomes dominated by B-face growth. At a II:VI ratio of 60:1 the material is entirely B-face and predominantly untwinned. The difference in the two growth modes is manifested in the laser reflectance. Greater scattering of the laser light occurs during B-face growth due to the increased roughening compared to A-face growth. Consequently the reflectance signal in the B-face signature is seen to fall away more rapidly than is the case with A-face growth.
M. Edward Motamedi, William Tennant, Haluk Sankur, Robert Melendes, Natalie Gluck, Sangtae Park, Jose Arias, Jagmohan Bajaj, John Pasko, William McLevige, Majid Zandian, Randolph Hall, Karla Steckbauer, Patti Richardson
To achieve the DoD objective of low cost high performance infrared focal plane arrays a manufacturing technique is required which is intrinsically flexible with respect to device configuration and cutoff wavelength and easily scaleable with respect to volume requirements. The approach adopted is to fully develop the technology of molecular beam epitaxy (MBE) to a level where detector array wafers with a variety of configurations can be fabricated with first pass success at a reduced cost. As a vapor phase process, MBE lends itself directly to: (1) the inclusion of real-time monitoring and process control, (2) a single or multiple wafer growth mode, (3) nearly instantaneous changes in growth parameters. A team has been assembled to carry out the program. It is composed of four industrial organizations -- Rockwell International, Hughes Aircraft Company, Texas Instruments, and Lockheed-Martin, and a university -- Georgia Tech Research Institute. Since team members are committed suppliers and users of IRFPAs, technology transfer among team members is accomplished in real-time. The technical approach has been focused on optimizing the processes necessary to fabricate p-on-n HgCdTe double layer heterostructure focal plane arrays, reducing process variance, and on documenting flexibility with respect to cutoff wavelength. Two device structures have been investigated and fabricated -- a 480 by 4 and a 128 by 128.
M. Edward Motamedi, William Tennant, Robert Melendes, Natalie Gluck, Sangtae Park, Jose Arias, Jagmohan Bajaj, John Pasko, William McLevige, Majid Zandian, Randolph Hall, Karla Steckbauer, Patti Richardson, Donald Cooper
The relatively large detector size of conventional focal plane arrays often acts as a limiting source of noise currents and requires these devices to run at undesirably low temperatures. To reduce the detector size without reducing the detector's quantum efficiency, we have developed efficient on-focal plane collection optics consisting of arrays of thin film binary optic microlenses on the back surface of hybrid detector array structures. P/n polarity photodiodes of an unusual `planar-mesa' geometry were fabricated in epitaxial HgCdTe deposited by molecular beam epitaxy on the `front' side of a CdZnTe substrate. Diffractive (8 - 16 phase level) Ge microlenses were deposited on 48 micrometers centers in a registered fashion (using an IR mask aligner and appropriate marks on the front surface of the CdZnTe) on the back side of the substrate using a lifting process. The lifting process circumvents some of the process limitations of the more conventional chemical etching method to diffractive microlens processing, allowing them to approach more closely their theoretical efficiency limit of > 95%. Prior to microlens deposition, but after diode fabrication, the test structures were flip- chip bonded or `hybridized' using indium interconnections to metallic strip lines which had been photolithographically deposited on sapphire dice (a process equally compatible with a silicon integrated circuit readout). After hybridization, the CdZnTe was thinned to equal the focal length of the lenses in the CdZnTe material. Optical characterization has demonstrated that the microlenses combined with the detector mesas concentrate light sufficiently to increase the effective collection area. The optical size of the mesa detectors being larger than the theoretical diffraction limit of the microlenses precludes determining whether the lenses themselves produce the theoretical diffraction-limited gain, but they clearly decrease required detector area by at least 3 - 6X. To our knowledge, this is the first successful demonstration of IR detectors and binary optics microlens integration.
Extensive material and device statistics of performance and reproducibility are presented to show the maturity of this technology. The demonstration vehicles to monitor yields during this demonstration were long-wavelength infrared (LWIR) HgCdTe multilayer wafers with 128 X 128 detector arrays. The heterostructure photodetectors were of the p-on-n planar configuration. Device data show that MBE LWIR diode test structures have performance that equals that of p-on-n double heterostructure photodiodes made by LPE. Due to the special attention given to understanding and reducing epilayer growth-induced defects, we have achieved improvements in FPA operability values from 92% to 98%. These improvements have resulted in the demonstration of a 128 X 128 FPA hybrid that had detectivity (D*) background limited performance when operating at 80 K in a tactical background environment. Mean D* was 1.28 X 1011 cmHz1/2/W. The corresponding mean NE(Delta) T was an excellent 5.9 mK.
Deep-level transient spectroscopy (DLTS) measurements were performed on midwavelength p+n HgCdTe mesa diodes grown by LPE and p+n HgCdTe planar heterostructures grown on CdZnTe wafers by MBE. The DLTS spectra of the MBE samples showed the presence of at least two hole traps, while at least two shallower electron traps were observed from the LPE samples. In addition to DLTS measurements, the diodes were also studied using double correlation DLTS (DDLTS) and current transient spectroscopy (CTS). The DDLTS spectra showed sharper peaks than the corresponding DLTS spectra. CTS measurements complement the DLTS data in the temperature range where the diode's leakage current is too high for capacitance measurements. These measurements revealed the presence of additional trapping centers in the LPE samples studied. This paper discusses the interpretation of these results and the suitability of the DLTS technique for the study of point defects in HgCdTe.
Recent progress for in situ monitoring of MCT growth is reviewed with particular reference to the need for improved control and reproducibility of the vapor phase growth methods. RHEED continues to be the main in situ monitor for MBE growth, giving both surface structure and growth rate data. However, this is now becoming supplemented by optical in situ monitors such as ellipsometry and new techniques for measuring the substrate temperature which critically affects the growth quality on the preferred (211)B substrate orientation. The lack of in situ monitoring techniques is no longer a disadvantage for MOVPE with reflection difference spectroscopy and surface photo-adsorption spectroscopy for surface characterization and spectroscopic ellipsometry and reflectometry for layer characterization. Organometallic concentration monitoring has been achieved using Epison ultrasonic monitors and has proved to be a vital part of the growth system monitoring to control critical parameters such as VI/II ratio. Recent results on feedback control of both IMP and direct alloy growth show that improved control is now possible in the complex MOVPE growth environment.
Deep levels in p+n Hg0.73Cd0.27Te/Hg0.68Cd0.32Te planar heterostructure diodes, grown by molecular beam epitaxy on CdZnTe wafers, were studied using deep level transient spectroscopy (DLTS). The DLTS spectra showed the presence of at least two hole traps with activation energies close to midgap. The activation energy obtained from the Arrhenius plots showed a strong dependence on the aplied bias, making it difficult to obtain a precise value.
In this paper we present p-on-n heterostructure HgCdTe photovoltaic device data that illustrates the high performance and flexibility in band gap control of the molecular beam epitaxy (MBE) technology. This flexibility demonstration was carried out by growing material for operation in the following cut-off wavelength ((lambda) co) ranges of interest: LWIR [(lambda) co(77 K) equals 9-11 micrometers ], MLWIR [(lambda) co(77 K) equals 6-7 micrometers ], and VLWIR [(lambda) co(40 K) equals 20 micrometers ]. Detailed analyses of the current-voltage characteristics of these diodes as a function of temperature show that their dark currents are diffusion-limited down to 80 K, 50 K, and 30 K for the MLWIR, LWIR, and VLWIR photodiodes, respectively. In general, the RoA device values were uniform for the three band gap ranges when operating under diffusion limited conditions. The planar MBE HgCdTe technology has been further validated with the successful fabrication and operation of 64 X 64 hybrid FPAs.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.