We describe uncooled mid-infrared light emitting and negative luminescent diodes made from indium antimonide based IlI-V compounds, and long wavelength devices made from mercury cadmium telluride. The techniques to produce large area positive and negative luminescent sources, and the application of these devices to gas sensing, improved thermal imagers and imager testing is discussed.

We report here the growth and characterization of intersubband quantum well structures based on the GaAs/AlGaAs material system, designed to emit radiation at approximately 7 micrometers. We present transport behavior, infrared photocurrent spectra, and electroluminescence data. First attempts to fabricate a laser structure from this devices encountered difficulties with the electrical properties of the AIGaAs waveguide cladding layers. Thus, we present measurements with different waveguide concepts as doped AlAs cladding layers and doped superlattice cladding structures.

We present an innovative-design heterostructure based on the exploitation of in-the-barrier piezoelectric field for all-optical light modulation. The novel layout allows an efficient light modulation with low power densities (few tens of W/cm2), easily attainable with standard laser diodes. The modulation mechanism relies upon drastic photocarrier separation by the piezoelectric field in the barrier layers. We present room temperature results showing that an optical “control” power of 70 W/cm2 creates in the heart of the structure a space-charge field of about 30 kV/cm, inducing large spectral shifts (∼100 nm) in the photoluminescence spectra of a CdHgTe quantum well in the 1.5 μm range.

We report on the performance of far-infrared hole inversion lasers made from germanium doped with the multivalent acceptors beryllium and copper. Commonly used hole inversion lasers are made from Czochralski-grown Ga-doped Ge single crystals and show emission from 75 to 125 and 170 to 300 μm. The emission gap between 125 and 170 μm, originating from absorption of the far-infrared light due to internal hole transitions in the neutral Ga acceptor, is absent in the new Be and Cu-doped lasers. We also find a mechanism for inversion depopulation through neutral Ga which hinders lasing at low electric fields. This same mechanism is shown to cause population inversion in the Be-doped laser and allows lasing at lower fields. This reduces the power input into the germanium crystal and has allowed us to increase the duty cycle up to 2.5% which is one order of magnitude higher than the maximum duty cycle reported for Ga-doped Ge lasers. These new lasers may offer an opportunity for achieving continuous-wave operation.

In addition we have performed preliminary studies on the effect of uniaxial stress on the lasing in these new materials. We demonstrate that small uniaxial stress increases laser action in Ge:Cu. We propose that this is due to an increased population inversion because under these conditions two separate mechanisms cause heavy holes to enter the light hole band.

Metallic photonic band gap (MPBG) structures are multi-layer metallic meshes imbedded in a dielectric medium. We report the successful design, fabrication, and characterization of infrared band-reject filters using MPBG structures in a flexible polyimide substrate. The metal layers of the MPBG have square grid patterns with short perpendicular cross-arm defects added halfway between each intersection. The transmission characteristics of these filters show a higher order band-reject region in addition to a lower order band gap that extends from zero to particular cut off frequency. The critical frequencies of the filters depend on the spatial periodicity of the metal grids and length of the cross-arm defects. Optical transmission measurements of the bandreject filters show lower edge cutoff frequency of about 2 THz and the higher order bandgap region centered around 4.5THz with attenuation of more than 35 dB in the bandgap region. This is in good agreement with the theoretical calculations. The filters maintain their optical characteristics after repeated bending, demonstrating mechanical robustness of the MPBG structure and have minimal dependence on angle of incidence.

The combined effects of suppressing Auger recombination by band structure engineering strained-layer superlattices (SL), suppressing radiative recombination by photon recycling, and suppressing both Auger and radiative recombination with carrier depletion are calculated quantitatively for long-wavelength and mid-wavelength InAs/ InxGa1-xSb SL-based infrared detectors operating at temperatures between 200 and 300K. The results are compared to their HgCdTe counterparts. The SL performance is better in all cases. However, the carrier concentrations required for typical background limited performance (300K, 2π field of view), ranging between about 1 × 1013 and 4 × 1013 cm−3 at 300K in long-wavelength to mid-wavelength SLs, are seen to be impractically low. The carrier concentration in a 11 μm photon detector yielding equivalent performance to an ideal 300K thermal detector is about 1014 cm−3. Large performance enhancement using carrier depletion therefore appears impractical even in optimized SLs.

We report the investigation of free-carrier absorption characteristics for epitaxially grown p-type thin films in the far-infrared region (50 ∼ 200 μm), where homojunction interfacial workfunction internal photoemission (HIWIP) detectors are employed. Five Si and three GaAs thin films were grown by MBE over a range of carrier concentrations, and the experimental absorption data were compared with calculated results. The freehole absorption is found to be almost independent of the wavelength. A linear regression relationship between the absorption coefficient and the carrier concentration, in agreement with theory, has been obtained and employed to calculate the photon absorption probability in HIWIP detectors. The detector responsivity follows the quantum efficiency predicted by concentration dependence of the free carrier absorption coefficient.

A monolithic quantum well infrared photodetector (QWIP) structure has been presented that is suitable for dual bands in the two atmospheric transmission windows of 3 – 5.3 μm and 7.5 – 14μm, respectively. The proposed structure employs dual stacked, strain InGaAs/AlGaAs and latticematched GaAs/AlGaAs quantum well infrared photodetector for mid wavelength and long wavelength detection. The response peak of the strain InGaAs/AlGaAs quantum well is at 4.9 μm and the lattice-matched GaAs/AlGaAs is at 10.5μm; their peak sensitivities are in the spectral regions of 3 – 5.3mu;m and 7.5 – 14μm. The peak responsivity when the dual-band QWIP is biased at 5 Volts is ∼0.065A/W at 4.9μm and ∼0.006A/W at 10.5μm; at this voltage the dual-band QWIP is more sensitive at the shorter wavelengths due to its larger impedance thus exhibiting wavelength tunability characteristics with bias. Additionally, single colored 4.9 and 10.5μm QWIPs were fabricated from the dual-band QWIP structure to study the bias-dependent behavior and also to understand the effects of growing the strain layer InGaAs/AlGaAs QWIP on top of the lattice-matched GaAs/AlGaAs QWIP. In summary, two stack dual-band QWIPs using GaAs/AlGaAs and strained InGaAs/AlGaAs multiquantum wells have been demonstrated with peak spectral sensitivities in the spectral region of 3 – 5.3μm and 7.5 – 14μm. Also, the voltage tunable dual-band detection have been realized for this kind of QWIP structure.

We report on the development of Germanium Blocked Impurity Band (BIB) photoconductors for long wavelength infrared detection in the 100 to 250.μm region. Liquid Phase Epitaxy (LPE) was used to grow the high purity blocking layer, and in some cases, the heavily doped infrared absorbing layer that comprise theses detectors. To achieve the stringent demands on purity and crystalline perfection we have developed a high purity LPE process which can be used for the growth of high purity as well as purely doped Ge epilayers. The low melting point, high purity metal, Pb, was used as a solvent. Pb has a negligible solubility <1017 cm−3 in Ge at 650°C and is isoelectronic with Ge. We have identified the residual impurities Bi, P, and Sb in the Ge epilayers and have determined that the Pb solvent is the source. Experiments are in progress to purify the Pb. The first tests of BIB structures with the purely doped absorbing layer grown on high purity substrates look very promising. The detectors exhibit extended wavelength cutoff when compared to standard Ge:Ga photoconductors (155 μm vs. 120 μm) and show the expected asymmetric current-voltage dependencies. We are currently optimizing doping and layer thickness to achieve the optimum responsivity, Noise Equivalent Power (NEP), and dark current in our devices.

Fueled by a broad range of government needs and funding, HgCdTe materials and device technology has matured significantly over the last two decades. Also in this same time period, we have come to understand better the phenomenology which limits imager performance. As a result of these developments, it appears that HgCdTe arrays may be tailored in wavelength to outperform GaAs-based image intensifier devices in sensitivity and to compete with bolometric and pyroelectric imaging arrays in NEDT at temperatures at or near room temperature (250K–295K). These benefits can be fully realized, however, only if HgCdTe can be brought to a level of maturity where the material and detectors made from it are limited by fundamental mechanisms. We will discuss the state of HgCdTe near room temperature performance and the practical and theoretical limits which constrain it.

Recent results on MOVPE growth of multilayer two-color HgCdTe detectors, for simultaneous and independent detection of medium wavelength (MW, 3–5 μm) and long wavelength (LW, 8–12 μm) bands, are reported. The structures are grown in situ on lattice matched (100) CdZnTe in the double-heterojunction p-n-N-P configuration. A barrier layer is placed between the LW and MW absorber layers to prevent diffusion of MW photocarriers into the LW junction and thereby eliminate spectral crosstalk. X-ray double crystal rocking curve widths are ∼ 45 arc-secs, indicating good epitaxial quality. SIMS depth profile measurements of these 28 μm thick structures show well-defined alloy compositions, and arsenic and iodine doping. SIMS data on a series of thirteen films show that good run-to-run repeatability is obtained on thicknesses, compositions, and dopant levels with values close to the device design targets. Depth profile of etch pits through the thickness of the films show etch pit densities in the range of 8×105-5×106 cm−2.

Zinc composition variation and gross structural defects including grain and tilt boundaries, twins, and mechanical cracks in high pressure Bridgman Cd1−xZnxTe are characterized and correlated to various detector-related responses. Triple axis x-ray diffraction, double crystal x-ray topography, infrared microscopy, and etch pit density measurements are used to reveal and quantify the spatial distribution and the nature of the structural defects. Mechanical cracks in the material are found to act as conductive “shorting paths”, indicated by excessive leakage currents and reduced charge (electron) collection measured along these cracks. Reduced charge collection is also obtained across grain boundaries and in regions with poor crystallinity, indicating that they serve as carrier recombination sites. Finally, the effects of the zinc composition variation on the measured leakage current and the amount of electrons collected are found to be masked by gross structural defects. These characterization techniques provide a wealth of information which can be used not only to study the relationship between the structural and device properties of CdZnTe but also to screen production material for subsequent device fabrication.

Gtain boundaries and twin boundaries in commercial Cd1−xZnxTe, which is prepared by a high-pressure Bridgeman technique, have been investigated with transmission electron microscopy, scanning electron microscopy, infrared-light microscopy and visible-light microscopy. Boundaries inside these materials were found to be decorated with Te precipitates. The shape and local density of the precipitates were found to depend on the particular boundary. For precipitates that decorate grain boundaries, their microstructure was found to consist of a single, saucer-shaped grain of hexagonal Te (space group P3,2 1). Analysis of a Te precipate by selected-area diffraction revealed the Te to be aligned with the surrounding Cd1−xZnxTe grains. This alignment was found to match the (111) Cd1−xZxTe planes with the (0 111) planes of hexagonal Te. Crystallographic alignments between the Cd1−xZnxTe grains were also observed for a high-angle grain boundary. The structures of the grain boundaries and the Te/Cd1−xZnxTe interface are discussed.

Cadmium Zinc Telluride (CZT) shows great promise as a semiconductor radiation detector material. CZT possesses advantageous material properties over other radiation detector materials in use today, such as a high intrinsic resistivity and a high cross-section for x and γ-rays. However, presently available CZT is not without limitations. The hole transport properties severely limit the performance of these detectors, and the yield of material possessing adequate electron transport properties is currently much lower than desired. The result of these material deficiencies is a lack of inexpensive CZT crystals of large volume for several radiation detector applications. One approach to help alleviate this problem is to measure the spatial distribution (or map) the electrical properties of large area CZT wafers prior to device fabrication. This mapping can accomplish two goals: identify regions of the wafers suitable for detector fabrication and correlate the distribution of crystalline defects with the detector performance. The results of this characterization can then be used by the crystal manufacturers to optimize their growth processes. In this work, we discuss the design and performance of apparatus for measuring the electrical characteristics of entire CZT wafers (up to 10 cm × 10 cm). The data acquisition and manipulation will be discussed and some typical data will be presented.

Significant improvements of a previously reported etching process [1] for Hg1−xCdxTe have been achieved with respect to etch rate, surface morphology and surface stoichiometry by optimization of the process parameters. The gas phase and surface reactions driving the etching process have been analyzed by combined optical and electrical characterization of the plasma and surface analyses of the samples. A reaction scheme is suggested which allows to model and upscale the process in a consistent manner.

We are developing modular arrays of CdZnTe radiation detectors for high-resolution nuclear medicine imaging. Each detector is delineated into a 64×64 array of pixels; the pixel pitch is 380 ptm. Each pixel is connected to a corresponding pad on a multiplexer readout circuit. The imaging system is controlled by a personal computer. We obtained images of standard nuclear medicine phantoms in which the spatial resolution of approximately 1.5 mm was limited by the collimator that was used. Significant improvements in spatial resolution should be possible with different collimator designs. These results are promising for high-resolution nuclear medicine imaging.

We present the performance characteristics of CdZnTe radiation detectors with a new P-I-N design and their unique advantages over metal-semiconductor-metal (M-S-M) devices. In M-S-M CdZnTe detectors the bulk resistivity of the substrate largely determines the leakage current. High leakage current is a dominant noise factor for CdZnTe detector arrays, coplanar detectors, and detectors used for low X-ray energy applications. P-I-N devices provide low leakage currents. Early CdZnTe detectors exhibited polarization, were limited to small detection volumes, and some required high deposition temperatures. We have developed a new heterojunction design which can be deposited at low temperatures so that even high-pressure Bridgman CdZnTe can be used. Using the P-I-N design, CdZnTe detectors with high detection volumes (>200 mm3) were fabricated and exhibited low leakage current, good energy resolution, and no polarization. These detectors have significant advantages over M-S-M detectors in three specific areas. First, X-ray fluorescence studies require detectors with low leakage currents to provide less spectral broadening due to electronic noise. Second, less expensive vertical Bridgman CdZnTe material can be used for imaging applications since it normally possesses too low of a bulk resistivity to be useful as a M-S-M detector. Third, leakage currents across the anode grid in large volume coplanar detectors can be significantly reduced.

A CdZnTe strip detector large area array (∼ 60 cm2 with 36 detectors) with capabilities for high resolution imaging and spectroscopy has been built as a prototype for a space flight gamma ray burst instrument. The detector array also has applications in nuclear medical imaging. Two dimensional orthogonal strip detectors with 100 gm pitch have been fabricated and tested. Details for the array design, fabrication and evaluation of the detectors will be presented.

Semiconductor multiple-electrode detectors have been developed for the purpose of reducing effects of hole trapping in room-temperature radiation detectors.1,2 Some reported geometries maintain a nearly-uniform electric field inside the detector, but others generate an electric field that is very non-uniform and highly-concentrated at the anode. This paper reports the results of mapping such a detector (having a non-uniform electric field) with a finely collimated gamma-ray beam to determine the detector response as a function of position.

Doping of the lead telluride and related alloys with the group III impurities results in an appearance of the unique physical features of a material, such as such as persistent photoresponse, enhanced responsive quantum efficiency (up to 100 photoelectrons/incident photon), radiation hardness and many others. We review the physical principles of operation of the photodetecting devices based on the group ifi-doped IV-VI including the possibilities of a fast quenching of the persistent photoresponse, construction of the focal-plane array, new readout technique, and others. The advantages of infrared photodetecting systems based on the group IIIdoped IV-VI in comparison with the modem photodetectors are summarized.