Fusion of reflected/emitted radiation light sensors can provide significant advantages for target identification and detection. The two bands -- 0.6 - 0.9 or 1 - 2 micrometer reflected light and 8 - 12 micrometer emitted radiation -- offer the greatest contrast since those bands have the lowest correlation, hence the greatest amount of combined information for infrared imaging. Data from fused imaging systems is presented for optical overlay as well as digital pixel fusion. Advantages of the digital fusion process are discussed as well as the advantages of having both bands present for military operations. Finally perception tests results are presented that show how color can significantly enhance target detection. A factor of two reduction in minimum resolvable temperature difference is postulated from perception tests in the chromaticity plane. Although initial results do not yet validate this finding, it is expected with the right fusion algorithms and displays that this important result will be proven shortly.

The objectives of the Integrated Imaging Sensors (I2S) Program are rtwofold. The first is to develop and deliver a rifle sight containing a single aperture and optical path for receiving, combining, and viewing radiation from the separate infrared (IR) and visible bands in a single image simultaneously. The second is to develop a sensor array sensitive in the radiation band spanning approximately from 0.4 μm to 1.7 μm by "fusing" indium-gallium-arsenic material onto silicon charge coupled devices. The ability to coincidentally and simultaneously form images from these two separate radiation bands is expected to significantly improve the detection and identification of objects from the case where only one radiation band is employed. Additionally, extending the cutoff of the visible band from 0.9 μm to 1.7 μm is expected to enhance viewing in this band as there is more available light, and further lessons the exacting requirement of desigining nearly noise free detectors.

Development results of photodetective assemblies for 1.5 ÷ 2.5, 3 ÷ 5 and 8 ÷ 14 micrometer spectral ranges achieved last time in RD&P Center 'Orion' are discussed. In thermal imaging and heat location systems with high space and temperature resolution photodetectors with a minimum size of sensitive elements and a maximum signal/noise ratio at the output of readout microelectronics are used. Submatrix photodetectors (MxN) operating in the time delay and integration mode (TDI) as well as 'staring' arrays (N X N) were the most extensively developed. Comparative parameters and characteristics of photodetectors with intermediate and deep cooling based on PbS (2x1536), PbSe (1x256), InSb (384x288), CdHgTe (2x256, 4x128, 128x128, 384x288) with photosensitive element sizes from 100 to 30 micrometer are given. Thermoelectric cooling or microcryogenic Stirling coolers provide operating temperatures. Development results of high-speed photodiodes based on Ge, Si and InGaAsP for optical communication and range finding in 0.8 ÷ 1.7 μm spectral range as well as CdHgTe photodiodes with high speed of up to 1 GHz for laser detection and ranging at 10.6 μm wavelength are given.

This article deal's with fundamental problems of IR optics. Including perspective directions of development of IR systems, IR optical materials, software for designing, modeling and testing IR lens and systems. Some results of development of lens for matrix cooled detectors and kinoform optical systems are concerned.

Results of the development of the series of modular microcryogenic systems (MCSs) to be used as a part of thermal imaging vision devices, both for military and commercial applications, are presented, comprising the three trends of this development.

The main trends of FSPC GIPO activities are complex investigations and developments in IR region. Technologies of obtaining the optical characteristics of objects -- and -- backgrounds radiation, atmosphere transmission, anthropogenic formations in the wide spectral range including laser wavelengths are developed. With this end in view, the special laboratories equipped with multispectral radiometers are created. In simulation modeling the background-target situations, the technology of computer thermo-frames is widely used. Several types of thermal imaging technique of various modifications are developed and produced by great batches. Currently, works on creating the thermal imaging devices of the second and third generations using Russian components are performed and the first samples are tested. The complex of original measuring instruments for carrying out the tests, certification and operation of thermal imagers is developed. The metrologic complex realizing the new trend in metrology of the scanning thermal imaging devices, i.e., the reproduction of normalized values 'radiation temperature difference' and 'radiance difference' is created. In addition, the equipment complex for ecological monitoring the environment is made on the basis of developments of thermal imagers and spectroradiometers. Developments of IR devices are based on the creation of new optical technologies and new optical elements (diffraction, hologram, kinoform, integrated, aspheric ones) as well as optical coatings of all kinds in FSPC GIPO.

The analysis of development of night vision devices (NVD) and image-intensifier tubes (IIT) is conducted, the change of their characteristics from zero up to the third generations surveyed, the appearance IIT of the fourth generation is determined. Alternatively IIT the solid-state image converters (SSIC) surveyed. SSIC essentially changes appearance NVD and their basic performances. The essential change of the characteristics NVD also is reached at the expense of usage of pulse laser illumination with a gating.

A technology was designed and the photodetector modules were manufactured for the 3 - 5 and 8 - 12 μm spectral range based on the Hg_1-xCd_xTe/GaAs heterostructures and GaAs/AlGaAs multiquantum well structures grown by the molecular beam epitaxy method. The photosensitive HgCdTe layers were grown on the GaAs substrates with the intermediate buffer layer of CdZnTe. To decrease the surface influence on the recombination processes graded gap HgCdTe layers with the increased to the surface composition were grown. A silicon multiplexer was designed and manufactured on the CMOS/CCD technology with frame rate 50 Hz. Hybrid assembly of the photodetectors array and the multiplexer was produced by the group cold welding on the indium bumps with control of the connection process. The manufactured 128 X 128 FPAs on the HgCdTe with the cut-off wavelength 6 μm and 8.7 μm had the NEDT value 0.02 K and 0.032 K, correspondingly, at operating temperature 78 K and frame rate 50 Hz. The photosensitive GaAs/AlGaAs multiquantum well structures were manufactured by the MBE method. It is shown that the designed technology allows to produce 128 X 128 photodetector arrays (λ_max = 8 μm) with the NEDT value 0.021 K and 0.06 K at operating temperature 54 K and 65 K, correspondingly.

Main results of theoretical studies of maximum (limited by background fluctuations) temperature sensitivity of thermal imagers and information capacity of electronic-vision systems are given. Spectral dependencies of noise equivalent temperature difference for thermal imagers with quantum and thermal detectors are calculated. It is shown that by this parameter thermal imagers using the best array photodetective assemblies approach the theoretical limit. It is found that by the information (dynamic) potential and the maximum infomration capacity thermal imaging in conditions of the Earth heat balance does not yield to the day-time human vision. Maximum values of a specific information capacity of electronic-vision systems at high levels of radiation are limited by diffraction (rule "word per channel"). At low levels of radiation maximum information capacity is achieved when a number of quantums necessary for transformation of one information bit (rule of "binary channels") fall on a spatial resolution element during the frame time. Maximum information capacity of coloured electronic vision systems is proportional to the number of independent spectral channels.

The ways of development of high sensitive area IR sensors for medical thermovision, conducted at National Research Institute (NRI) 'Electron' (Russia) are presented. Registration of temperature and spectral variation of IR illumination from small area of human's skin (or other biological objects) in wide spectral range requires use of an area IR sensor of large format, operating in signal accumulation mode. It is shown that IR Schottky Barrier (SB) CCDs comply with majority of requirements and are the most suitable for the purpose from cost/quality point of view. The basic parameters of an area IR SB CCDs and IR cameras for 1,5 - 5,5 μm spectral range, developed at 'Electron' NRI and program of development IR cameras for 1,2 - 12 μm for medical applications are considered.

The AHI sensor consists of a long-wave infrared pushbroom hyperspectral imager and a boresighted 3-color visible high resolution CCD linescan camera. The system used a background suppression system to achieve good noise characteristics (less than 1(mu) fl NESR). Work with AHI has shown the utility of the long-wave infrared a variety of applications. The AHI system has been used successfully in the detection of buried land mines using infrared absorption features of disturbed soil. Recently, the AHI has been used to examine the feasibility active and passive hyperspectral imaging under outdoor and laboratory conditions at three ranges. In addition, the AHI was flown over a coral reef ecosystem on the Hawaiian island of Molokai to study fresh water intrusion into coral reef ecosystems. Theoretical calculations have been done propose extensions to the AHI design in order to produce an instrument with a higher signal to noise ratio.

The MUlti Sensor Trial 2000 experiment was a multi-platform remote sensing deployment in Cairns Australia. Included in the deployment were both visible and infrared airborne hyperspectral images. The University of Hawaii's Airborne Hyperspectral Imager represented the thermal infrared portion of the data collect. The ability to discriminate various targets using the thermal infrared was explored. Consequent data processing involved separating targets from clutter using matched filters. In addition, a preliminary atmospheric correction algorithm was developed based on the ISIS algorithm used in SEBASS.

A cooperative effort between the U.S. Air Force Research Laboratory, Nova Research, Inc., the Raytheon Infrared Operations (RIO) and Optics 1, Inc. has successfully produced a miniature infrared camera system that offers significant real-time signal and image processing capabilities by virtue of its modular design. This paper will present an operational overview of the system as well as results from initial testing of the 'Modular Infrared Imaging Applications Development System' (MIRIADS) configured as a missile early-warning detection system. The MIRIADS device can operate virtually any infrared focal plane array (FPA) that currently exists. Programmable on-board logic applies user-defined processing functions to the real-time digital image data for a variety of functions. Daughterboards may be plugged onto the system to expand the digital and analog processing capabilities of the system. A unique full hemispherical infrared fisheye optical system designed and produced by Optics 1, Inc. is utilized by the MIRIADS in a missile warning application to demonstrate the flexibility of the overall system to be applied to a variety of current and future AFRL missions.

By using resets of multiple wavelengths to limit dispersion, we show conceptually that we will be able to steer passive, broadband, electro-optical sensors over wide angles using optical phased array technology. The dispersion associated with optical phased array beam steering can be limited to values of 20 to 80 times the diffraction limit for very large angle beam steering. Much of the remaining factor of 20 - 80 may be correctable using digital techniques. A significant obstacle associated with widespread implementation of optical phased array beam steering is wavelength dispersion associated with resets. We discuss optical phased array beam steering using resets of greater than one wavelength. By using larger resets the unfolded phase front for wavelengths other than the design wavelength can be maintained closer to a prism, thus limiting dispersion. This technique can be implemented with the variable period beam steering approach or the variable blaze beam steering approach. One way to implement the variable blaze beam steering approach is using moveable lenslets. This method of implementation is amenable to large value resets, so it is an attractive method of implementation for this multiple wavelength reset, limited dispersion, beam steering technique.

Here we investigate a novel approach to steering broadband imagery with a Liquid Crystal Optical Phased Array (LCOPA). Our approach overcomes the deleterious blurring and echoing effects inherent in the use of such a device. We develop a model for the LCOPA and formulate a method in which a steered, graybody scene may be restored through the application of a Wiener filter. We also show this approach may be extended to scenes that are not strictly composed of graybodies but instead are only spectrally smooth over an appropriate bandwidth. Experimental results are presented that demonstrate the effectiveness of this approach.

Focal plane arrays (FPA's), used for remote sensing applications, are required to operate at high temperatures and are subject to high terrestrial background fluxes. Typical remote sensing applications like cloud/weather imagery, sea- surface temperature measurements, ocean color characterization, and land-surface vegetation indices also require FPAs that operate from the visible through the LWIR portion of the spectrum. This combination of harsh requirements have driven the design of a unique LWIR FPA, that operates at 80 K under 300 K background conditions, with an operating spectral range from 11.5 micrometers to 12.5 micrometers , and a spectral cutoff of 13.5 micrometers . The FPA consists of 2 side by side arrays of 1 X 60 HgCdTe, (grown by molecular beam epitaxy) photovoltaic, detector arrays bump bonded to a custom CMOS Si readout. The 2 arrays are completely independent, and can be operated as such. The readout unit cell uses two, current-mode, analog building blocks; a Current Conveyor (CC1) and a dynamic current mirror. The CC1 has input impedance below 300 Ohms and an injection efficiency that is independent of the detector characteristics. This combination extracts high performance and excellent sensitivity from detectors whose average RoA values are approximately 1.7 Ohm-cm² at T equals 80 K. The dynamic current mirror is used to subtract high background photocurrent while preserving excellent dynamic range. In addition to the performance enhancing readout, the detectors are manufactured with integral microlenses and operated in reverse bias to take advantage of their increased dynamic impedance. The dark currents associated with reverse bias operation are subtracted along with the background photocurrents by the dynamic current mirror. The expected and measured LWIR FPA performance will be presented. Measurements were performed on an LWIR FPA. Expected and measured FPA results are shown in the table below. The expected data are calculated from FPA models and compared to the measured values.

Research by DERA aimed at unmanned air vehicle (UAV) size reduction and control automation has led to a unique solution for a short range reconnaissance UAV system. Known as OBSERVER, the UAV conventionally carries a lightweight visible band sensor payload producing imagery with a large 40°x90° field of regard (FOR) to maximize spatial awareness and target detection ranges. Images taken from three CCD camera units set at elevations from plan view and up to the near horizon and are 'stitched' together to produce the large contiguous sensor footprint. This paper describes the design of a thermal imaging (TI) sensor which has been developed to be compatible with the OBSERVER UAV system. The sensor is based on UK uncooled thermal imaging technology research and offers a compact and lightweight solution operating in the 8-12 μm waveband without the need for cryogenic cooling. Infra-red radiation is gathered using two lead scandium tantalate (PST) hybrid thermal detectors each with a 384 X 288 pixel resolution, known as the Very Large Array (VLA). The TI system is designed to maintain the imaging format with that of the visible band sensor. In order to practically achieve this with adequate resolution performance, a dual field of view (FOV) optical system is used within a pitchable gimbal. This combines the advantages of a wide angle 40°x30° FOV for target detection and a narrow angle 13°x10° FOV 'foveal patch' to improve target recognition ranges. The gimbal system can be steered in elevation to give the full 90° coverage as with the visible band sensor footprint. The concept of operation is that targets can be detected over the large FOV and then the air vehicle is maneuvered so as to bring the target into the foveal patch view for recognition at an acceptable stand-off range.

HARLID^TM is a digital approach to achieving angular sensitivity in a laser warning system. In this version of the HARLID^TM module, a number of improvements are described which correct for certain problems and limitations of earlier devices. The detector used is a 2-detector assembly, consisting of matching silicon and InGaAs arrays assembled in a sandwich configuration, to achieve spectral sensitivity between 500 and 1700 nm. Systematic angular readout errors observed in previous work have been avoided with the use of a new light-guide in which the optical channels are air instead of glass. Improved response time in the short wavelength end of the spectral range has been achieved with the use of thinner active regions in the elements of the silicon array, and a redesigned digital aperture mask significantly improves accuracy and reduces optical vignetting effects. The design and performance characteristics of a 6-bit HARLID^TM are presented.

Uncooled infrared cameras have made dramatic strides recently. Very low cost, lightweight, low power cameras have been built. Also low cost high performance uncooled cameras have been built. A discussion of this technology to make this happen and the resulting new applications will follow.

Raytheon Infrared Operations (RIO) has achieved a significant technical breakthrough in uncooled FPAs by reducing the pixel size by a factor of two while maintaining state-of-the-art sensitivity. Raytheon has produced high-quality 320 X 240 microbolometer FPAs with 25 μm pitch pixels. The 320 X 240 FPAs have a sensitivity that is comparable to microbolometer FPAs with 50 micrometers pixels. The average NETD value for these FPAs is about 35 mK with an f/1 aperture and operating at 30 Hz frame rates. Good pixel operability and excellent image quality have been demonstrated. Pixel operability is greater than 99% on some FPAs, and uncorrected responsivity nonuniformity is less than 4% (sigma/mean). The microbolometer detectors also have a relatively fast thermal time constant of approximately 10 msec. This state-of-the-art performance has been achieved as a result of an advanced micromachining fabrication process. The process allows maximization of both the thermal isolation and the optical fill-factor. The reduction in pixel size offers several potential benefits for IR systems. For a given system resolution (IFOV) requirement, the 25 μm pixels allow a factor of two reduction in both the focal length and aperture size of the sensor optics. The pixel size reduction facilitates a significant FPA cost reduction since the number of die printed on a wafer can be increased. The pixel size reduction has enabled the development of a large-format 640 X 512 FPA array applicable to wide-field-of-view, long range surveillance and targeting missions, and a 160 X 128 array where applications for miniaturization and temperature invariance are required as well as low cost and low power.

Uncooled thermal infrared sensors require to be operated in an ambient gas pressure of about 50 mTorr or less to avoid sensitivity being reduced by thermal conduction through the gas. Although sealed packages have been developed which can retain a sufficiently low internal pressure for many years, the packaging process (cleaning, assembly, pumping, baking, getter firing, sealing) and materials add significant cost and weight. Lower cost it the major reason for the development of uncooled arrays, and low weight is essential for many applications (e.g. unmanned aerial vehicles, helmet mounted applications). In response to these needs, Honeywell has developed a silicon 'Integrated Vacuum Package' (IVP) process which produces a low-cost lightweight (0.2 gram) compact vacuum package by a wafer-scale process. The IVP process basically consists of bonding a silicon 'topcap' wafer to the array wafer, to produce a bonded double-wafer with multiple arrays protected in individual vacuum packages. The double- wafer may be easily handled without damage to the protected arrays, and diced into individual dies using normal silicon dicing techniques. It has been found helpful to use an etched evacuation via, which allows wafer bonding, pumping, baking and sealing to be performed in separate stages, at their different optimum times and temperatures. The IVP process will be described, and packages suitable for linear and two- dimensional uncooled arrays will be reported, with performance and lifetime measurements.

This paper reports the development of a low-cost CMOS microbolometer focal plane array with a new temperature coefficient enhancement readout circuit. We have recently reported an uncooled microbolometer detector that uses the CMOS n-well layer as the active material, where the suspended and thermally isolated n-well structure is obtained by silicon bulk micromachining of fabricated CMOS dies. In addition, we have successfully fabricated a 16 X 16 n-well microbolometer FPA. Although n-well is single crystal silicon and has very low 1/f noise, the fabricated array performance was limited due to low TCR of the n-well. The n-well has a TCR of 0.50 - 0.70%/K, which is the highest among the CMOS layers, but lower compared to the state-of-the-art microbolometer materials whose TCR values are about 2 - 3%/K. This paper reports a new n-well microbolometer FPA with a readout circuit that enhances the temperature coefficient (TC) of the microbolometer current, compensating for the low TCR value of the detector. The TC enhancement is achieved by passing the pixel current through a 4^th power taking circuit prior to integration, increasing the pixel current TC four times and resulting in an effective TC of 2.0 - 2.8%/K. A 16 X 16 test array has been designed and fabricated using a 0.8 micrometers standard CMOS process. The chip measures 2.4 X 3.8 mm² and contains 80 micrometers X 80 micrometers microbolometer pixels with 13% fill factor. The measurements and calculations show that the 16 X 16 prototype FPA can provide a responsivity (R) of 2 X 10⁷ V/W, a detectivity (D*) of 1.68 X 10⁹ cm(root)Hz/W, and NETD of 290 mK at a scanning rate of 260 fps. The same NETD value can be obtained for a 128 X 128 pixel array operating at 30 fps. NETD can further be decreased by improving the noise performance of the readout circuit, since the performance is not limited by the n-well microbolometer noise.

LETI LIR has been involved in Amorphous Silicon uncooled microbolometer development for years. This technology is now in production at Sofradir and first delivery have already been done to customers. From our background in modeling and material mastering LETI/LIR concentrate now on performance enhancement. This is a key point for cost reduction due to the fact that signal to noise ratio enhancement will allow us to decrease the pitch. A new approach of packaging is also described in this paper and first results are displayed. A new technological stack of amorphous silicon fully compatible with industrial process is presented. Electro-optical results obtained from an IRCMOS 320 X 240 with 35 μm pitch are presented. NETD close to 35 mK has been obtained with our new embodiment of amorphous silicon microbolometer technology.

Microbolometers respond to variations in impinging infrared energy through changes in the temperature of their thermally isolated detector structure. Thus, variations in the temperature of the substrate give rise to erroneous detector output variations. In order to minimize these effects, microbolometers are thermally stabilized using Peltier- junction coolers or heaters and operated and calibrated at a single temperature set point. This method provides the lowest thermal fluctuation noise at the expense of power consumption, waste heat generation and power-up readiness times often measured in minutes. This paper discusses strategies for providing rapid system readiness through non-thermally stabilized operation over a broad temperature range. A commercially available thermal imaging module which produces analog or digital images within 300 ms of power-up and operates without choppers, shutters or thermal stabilization has been demonstrated.

This paper describes the modeling, design, fabrication and testing of advanced uncooled thermal detectors, based on semiconducting YBaCuO. The aim is to provide NASA with advanced broad-band infrared (IR) detectors to replace the current CERES (Clouds and the Earth's Radiant Energy System) hardware that utilizes three channels, each housing a 1.5 mm X 1.5 mm thermister bolometer with 1 X 4 array of detectors in each of the three channels, thus yielding a total of 12 channels. A double mirror structure is used to obtain uniform spectral response from 0.3-100 μm wavelength. Double absorbers are utilized to further flatten the spectral response and to enhance the absorption of infrared radiation. The devices were fabricated using a polyimide sacrificial layer to achieve thermal isolation of the detector. A low thermal conductivity to the substrate enables the detector to integrate the energy from the incident radiation. An air gap was created by ashing the polyimide sacrificial layer from underneath the thermometer. A passivation layer was used to protect YBaCuO during ashing process and maintain a relatively high temperature coefficient of resistance of around 2.8%. These devices have successfully demonstrated voltage responsivities over 10³ V/W, detectivities above 10⁸ cm Hz^1/2/W, NEP per root Hertz bandwidth less than 4 X 10^-10 W/Hz^1/2 and thermal time constant less than 15 ms. Several specific designs were fabricated and tested. Relatively uniform response in the wavelength range of 0.6 to 15 μm was measured.

In an effort to leverage uncooled microbolometer technology, testing of bolometer performance in various nonimaging applications has been performed. One of these applications makes use of an uncooled microbolometer array as the sensing element for a laser beam analyzer. Results of the characterization of cw CO₂ laser beams with this analyzer are given. A comparison with the results obtained with a commercial laser beam analyzer is made. Various advantages specific to microbolometer arrays for this application are identified. A second application makes use of microbolometers for absolute temperature measurements. The experimental method and results are described. The technique's limitations and possible implementations are discussed. Finally, the third application evaluated is related to the rapidly expanding field of biometry. It consists of using a modified microbolometer array for fingerprint sensing. The basic approach allowing the use of microbolometers for such an application is discussed. The results of a proof-of-principle experiment are described. Globally, the described work illustrates the fact that microbolometer array fabrication technology can be exploited for many important applications other than IR imaging.

Uncooled microbolometer thermal imaging sensor technology has begun to successfully address military, government and commercial applications in the real world. BAE SYSTEMS, located in Lexington MA, has been involved in the design and development of uncooled IR technology since the early 1980s. Our current MicroIR^TM products are based on vanadium oxide (VO_x) microbolometers. Thousands of uncooled microbolometer thermal imaging sensors are now being produced and sold annually. A the same time, applied research and development on the technology continues to improve the basic products and make them suitable for new applications. In this paper we report on the status and improvements achieved in the MicroIR^TM product line, based on 320 X 240 element and 160 X 120 element FPA's with 46 μm pixel pitch. Other near term MicroIR^TM products include 320 X 240 and 640 X 480 FPA's with 28 micrometers pixel pitch and measured sensitivities below 50 mK. In the systems area we discuss development and testing of a Light Thermal Weapon Sight (LTWS) for the U.S. Army, being developed by BAE SYSTEMS in partnership with Thales, based upon our uncooled MicroIR^TM focal plane arrays (FPA) and systems. The LTWS prototypes were based upon our Standard Imaging Module SIM200, which employs our LAM2C, 320 X 240 element, microbolometer FPA. Finally we discuss the 480 X 640 element FPA and its application to the Heavy Thermal Weapon Sight application.

Infrared Components Corporation (ICC) recently announced that the company had acquired rights to uncooled microbolometer technology developed at the Defence Science and Technology Organization (DSTO), Australia. Under the license agreement DSTO is developing a technology transfer package for implementation in a silicon MEMS foundry. ICC has contracted Electro-optic Sensor Design (EOSD) for FPA design and analysis and local co-ordination of the technology transfer program. ICC has also entered into an agreement with SUNY Albany Center for Advanced Thin Film Technology (ACATFT) to transition the DSTO technology to production. In this paper we outline the microbolometer processing technology and discuss the aims and objectives of the ICC program.

This paper describes a newly developed thermoelectric infrared imager having a 48 X 32 element thermoelectric focal plane array (FPA) and an experimental vehicle featuring a blind spot pedestrian warning system, which employs four infrared imagers. The imager measures 100 mm in width, 60 mm in height and 80 mm in depth, weighs 400 g, and has an overall field of view (FOV) of 40 deg X 20 deg. The power consumption of the imager is 3 W. The pedestrian detection program is stored in a CPU chip on a printed circuit board (PCB). The FPA provides high responsivity of 2,100 V/W, a time constant of 25 msec, and a low cost potential. Each element has external dimensions of 190 μm x 190 μm, and consists of six pairs of thermocouples and an Au-black absorber that is precisely patterned by low-pressure evaporation and lift-off technologies. The experimental vehicle is called the Nissan ASV-2 (Advanced Safety Vehicle-2), which incorporates a wide range of integrated technologies aimed at reducing traffic accidents. The blind spot pedestrian warning system alerts the driver to the presence of a pedestrian in a blind spot by detecting the infrared radiation emitted from the person's body. This system also prevents the vehicle from moving in the direction of the pedestrian.

Lithium tantalate (LiTaO₃) is a material that is excellently suited for pyroelectric infrared detectors. Its figures of merit are first rate, it has a good long-time stability and it is available at reasonable cost. Researchers at the Institute for Solid State Electronics have been developing a flexible technology to manufacture LiTaO₃ detectors for the last twelve years. As a result, devices can be produced that can be optimally adjusted to the planned application. This paper describes the design and basic features of linear arrays with up to 256 responsive elements. The arrays are hybrid devices consisting of the pyroelectric chip and a CMOS multiplexer. It is demonstrated that special patterning technologies (ion beam etching) and optical functional layers on the responsive element considerably increase the thermal and spatial resolution. Thus, NEP values smaller than 0.2 nW were obtained at 40 Hz chopper frequency. Main applications of detectors are the contactless temperature measurement and -- increasingly -- the spectrometry in the wavelength range 0.8 ... 25 μm.

The major defence systems companies have been expending considerable effort on the development of thermal imaging and sensing systems technologies for use in the military field for many years. Major efforts have been made to spin-off low cost variants of the technology into the civil market sector with varying degrees of success. IRISYS have approached the civil market opportunity from a different perspective. The company has developed an inherently low cost technology with the objective of achieving a low cost imaging sensor. The achievement of the low price for civil sensing applications is by use of a detector technology that is optimized for the very high volume detection market. The paper reviews the detector technology, design methodologies and implementation of low cost thermal detection and imaging products.

In this paper, we developed 'Optically Readable BM Infrared Detector' of 55μm x 55μm pixel pitch and 266 X 194 array and evaluated the characteristics of this detector. In the infrared imaging test, we succeeded in obtaining infrared image of a person by adding an offset correction to the raw image taken through CCD camera, whereby we made a great progress to embody an infrared camera with this method. In the infrared imaging testing without correction, we succeeded in obtaining an infrared image of an electric heater and we confirmed the feasibility of the direct viewing in this method. This result offers us a hope to explore new application fields as an infrared camera of low power consumption under this method.

Narrow gap IV-VI (lead chalcogenides like PbSnSe and PbTe) layers grown epitaxially on Si(111)-substrates by molecular beam epitaxy exhibit high quality despite the large lattice and thermal expansion mismatch. Line arrays in IV-VI on Si layers have so far been realized and described. We present the first realization of a 2-d narrow gap IR-FPA on a Si-substrate containing the active addressing electronics: A 96 X 128 array with 75 μm pitch for row-by-row electronic scanning and parallel read-out of the line addressed. Each pixel contains a bare Si-area onto which epitaxial growth occurs, and an access transistor. A MWIR PbTe layer is grown by MBE onto complete Si-read-out substrates at temperatures below 450°C (because of the Al-metallization). Photovoltaic sensors are then delineated in the layers. Each pixel is connected to the Si read-out by sputtered Al-stripes. Yield in completely fabricated arrays was up to above 97%, with quantum efficiencies around 60% and differential resistances at zero bias of several 100 kOhms.

Based on the development of an IR seeker in 1992 and its extensive captive flight testing, the seeker of the trilateral (Germany, France, Italy) Polyphem missile was developed and tested in captive and free flights. It is based on a 640 X 486 element FPA and a cardanic stabilization and pointing system with an on-gimbal rate gyro. The digitized output of the FPA is sent to the ground via an optical fiber where it is observed by the operator and processed for cueing of targets and tracking for the final approach. After general seeker design considerations its design and performance and an in- flight video of a live firing will be presented. On the basis of the prior development of an IR seeker module set consisting of a 256 X 256 element InSb FPA and its electronics, the seeker for the German KEPD350 missile has been in development since 1997. Recently it has been qualified successfully and 25 of it will be delivered in 2001. Its mechanics consist of a miniaturized cardanic system which uses the data of the missile's Inertial Measurement Unit (IMU) for strap-down pointing and stabilization. Its image processor is designed for navigation during cruise and for tracking of the target in final approach. The seeker's features and performance will be presented.

Several junction formation methods are known to make HgCdTe photovoltaic devices. Ion implantation is the most popular process, but it needs additional thermal annealing process. In-situ junction formation by several epitaxy techniques is the advanced process, but is still hard to fabricate. In this paper, for the first time, hydrogenation technique for p-to-n type conversion in HgCdTe has been studied to fabricate HgCdTe photovoltaic infrared detector. H₂ plasma generated in an inductively coupled plasma (ICP) system was used to hydrogenate p-type HgCdTe wafer. Using the ICP system, damages given to the HgCdTe wafer could be minimized. Junction depth measured by differential Hall measurement was able to be adjusted from 2μm to 20μm. Hydrogen atom profile was measured by secondary ion mass spectroscopy (SIMS) and doping profile was measured by differential Hall measurement. Similar depth profile was found between the hydrogen profile and doping profile. It suggests the diffused hydrogen atom is the source of the type conversion. Several experiments were also taken with vacancy-doped and gold-doped p-type HgCdTe wafers. Type conversion was observed only in vacancy doped HgCdTe wafer, not in gold-doped HgCdTe wafer. This means that junction formation by hydrogenation is not due to the damage by the hydrogen plasma, but due to the diffusion of the hydrogen atoms. By applying the hydrogenation process to vacancy-doped wafers, LWIR diodes were successfully fabricated. Current-voltage (I-V) characteristics of hydrogenated Hg_0.79Cd_0.21Te diodes were also measured. Average RoA products of these diodes were about 50 Ω cm². Device uniformity and stability were also tested. The characteristics of the hydrogenated devices did not changed under the baking condition of 80°C over 10 days.

The increasing proliferation of infrared technology, including domestic and international product development, is bringing very high performance systems into the commercial market. Raytheon Infrared Operations (RIO) programs have produced a variety of products that are economically viable for the commercial market and retain very high performance. These products include both cooled and uncooled sensors. Examples of these products range from high-resolution camera engines to high-performance focal planes. These sensors are available as commodity products directly from RIO, a merchant supplier.

Raytheon has consolidated the products and expertise of the former Hughes Mahwah (Magnavox) and Torrance cryocooler operations to the Raytheon Infrared Operations (RIO) located in Goleta, CA (formerly SBRC). Co-location of the cryocooler operations with the detector/dewar operations yields infrared systems with reduced cost. This paper describes the current capabilities of the linear and rotary cryocooler products as well as developments underway and planned. Development goals include cost reduction, high performance while operating in extreme environmental conditions (> 90°C skin temperatures), and long life (> 20,000 hrs). Technologies developed by a Raytheon sister division for space cryocoolers are now being applied to tactical cryocoolers at RIO. Data, specifications, and a technology roadmap for the product-line cryocoolers encompassing cooling capacities including 0.2-, 0.35-, 0.75-, 1.0- and 1.75-watt ranges will be shown.

A tremendous developmental effort in the field of infrared detectors during the last decade in Israel has resulted in a variety of InSb and HgCdTe infrared detectors. Additional and significant R&D effort associated with other IR components, have also been done in Israel, in order to integrate the detectors into advanced Detector-Dewar-Cooler assemblies (DDCs). This R&D effort included notable activities in the field of materials, signal processors, dewars and cryocoolers. These activities are presented together with the status of infrared detector work in Israel. Several two-dimensional InSb staring detectors and DDCs are demonstrated. This includes two versions of the classical 256 X 256 detectors and DDCs, improved 640 X 480 InSb detectors and DDC, and a 2000- element detector with high TDI level. SADA II type HgCdTe detectors are also presented. Considerations regarding the course of future detector work are also described. The classical DDC requirement list which traditionally included demands for high D*, low NETD and high resolution is widened to include cost related issues such as higher reliability, lower maintenance, smaller volume, lower power consumption and higher operation temperature.

Characterization of variable area InSb photodiode arrays as well as electron beam induced current (EBIC) measurements of InSb p-n junction have been carried out. The lateral collection length of the diode was deduced to be 35-38 μm for InSb substrate with electron concentration of 0.5 - 1x10¹⁵ cm^-3. EBIC linescans of the diode cross- section yielded same values. Simulations of EBIC profiles of plan view scans at different electron beam voltages across the diode edge yielded diffusion length of 32-38 μm for different diodes. These results show that for FPA design and fabrication the reasonable value of diffusion length in n-InSb at 80 K should be 32-38μm.

We fabricated the GaAs/AlGaAs Quantum Well Infrared Photo detector (QWIP) focal plane array with selectively re-grown N- GaAs interconnection plugs and demonstrated its device operation, in order to establish the technology to obtain both complex device functions and device manufacturability. MBE (Molecular Beam Epitaxy) grown QWIP MQW wafers were covered with SiON and SiN_x mask films to obtain selectivity of the re-growth process. N-GaAs plugs were re-grown selectively with low-pressure MOCVD (Metal-Organic Chemical Vapor Deposition) with AsH₃ and Dimethylgalliumchloride as precursors, only on the bottom surfaces of the holes for the interconnection to extract the electrodes from the underlying epilayer. Cross- sectional SEM observation revealed that the feature of the re- grown N-GaAs plugs was triangular, rather than rectangular as expected. The reason for this discrepancy is not yet clear. The electrical contact between the epilayer and re-grown N- GaAs plug was 'ohmic-like,' without any trace of interfacial barrier. The Current-Voltage characteristics of the fabricated QWIP device showed no tangible leakage current between the N- GaAs plug and device structure, indicating that electrical insulation between the N-GaAs plugs and device structure was sufficient. Fabricated devices were successfully operated as a hybrid focal plane array, indicating the selective re-growth was a promising technique to realize complex QWIP based devices.

Advanced thermal imaging infrared cameras have been a cost effective and reliable method to obtain the temperature of objects. Quantum Well Infrared Photodetector (QWIP) based thermal imaging systems have advanced the state-of-the-art and are the most sensitive commercially available thermal systems. QWIP Technologies LLC, under exclusive agreement with Caltech University, is currently manufacturing the QWIP-Chip^TM, a 320 X 256 element, bound-to-quasibound QWIP FPA. The camera performance falls within the long-wave IR band, spectrally peaked at 8.5 μm. The camera is equipped with a 32-bit floating-point digital signal processor combined with multi- tasking software, delivering a digital acquisition resolution of 12-bits using nominal power consumption of less than 50 Watts. With a variety of video interface options, remote control capability via an RS-232 connection, and an integrated control driver circuit to support motorized zoom and focus- compatible lenses, this camera design has excellent application in both the military and commercial sector. In the area of remote sensing, high-performance QWIP systems can be used for high-resolution, target recognition as part of a new system of airborne platforms (including UAVs). Such systems also have direct application in law enforcement, surveillance, industrial monitoring and road hazard detection systems. This presentation will cover the current performance of the commercial QWIP cameras, conceptual platform systems and advanced image processing for use in both military remote sensing and civilian applications currently being developed in road hazard monitoring.

Standard GaAs/AlGaAs QWIPs are now well established for LWIR detection. The main advantage of this technology is the duality with the technology of commercial GaAs devices. The realization of large FPAs (up to 640 X 480) drawing on the standard III-V technological process has already been demonstrated. The second advantage widely claimed for QWIPs is the so-called band-gap engineering, allowing the custom design of the quantum structure to fulfill the requirements of specific applications such as multispectral detection. QWIP technology has been growing up over the last ten years and now reaches an undeniable level of maturity. As with all quantum detectors, the operating temperature of QWIPs is limited by the thermal current, particularly in the LWIR range. It is very crucial to achieve an operating temperature as high as possible and at least above 77 K in order to reduce volume and power consumption and to improve the reliability of the detection module. This thermal current offset has three detrimental effects: noise increase, storage capacitor saturation and high sensitivity of FPAs to fluctuations in operating temperature. For LWIR FPAs, large cryocoolers are required, which means volume and power consumption unsuitable for handheld systems. The understanding of detection mechanisms has led us to design and realize high performance 'standard' QWIPS working near 77 K. Furthermore, a new in situ skimmed architecture accommodating this offset has already been demonstrated. In this paper we summarize the contribution of THALES Research & Technology to this progress. We present the current status of QWIPs in France, including the latest performances achieved with both standard and skimmed architectures. We illustrate the development of our QWIPs by results on FPAs.

We report on results of laboratory and field tests of dual- band MWIR/LWIR focal plane arrays (FPAs) produced under the Army Research Laboratory's Multidomain Smart Sensor Federated Laboratory program. The FPAs were made by DRS Infrared Technologies using the HgCdTe material system and by BAE Systems using QWIP technology. The HgCdTe array used the DRS HDVIP^TM process to bond two single-color detector structures to a 640 X 480-pixel single-color read-out integrated circuit (ROIC) to produce a dual-band 320 X 240 pixel array. The MWIR and LWIR pixels are co-located and have a high fill factor. The images from each band may be read out either sequentially (alternating frames) or simultaneously. The alternating frame approach must be used to produce optimal imagery in both bands under normal background conditions. The QWIP FPA was produced using MBE-grown III-V materials. The LWIR section consisted of GaAs quantum wells and AlGaAs barriers and the MWIR section used InGaAs quantum wells with AlGaAs barriers. The detector arrays were processed with three ohmic contacts for each pixel allowing for independent bias control over both the MWIR and LWIR sections. The arrays were indium bump-bonded to an ROIC (specifically designed for two color operation) which puts out the imagery from both bands simultaneously. The ROIC has variable gain and windowing capabilities. Both FPAs were tested under similar ambient conditions with similar optical components. The FPAs were subjected to a standard series of laboratory performance tests. The relative advantages and disadvantages of the two material systems for producing medium-format dual-band FPAs are discussed.

According to the common understanding, the 3rd generation of infrared (IR) detection modules is expected to provide advanced functionalities like more pixels, multicolor or multiband capability, higher frame rates and better thermal resolution. This paper is intended to present the present status at AIM on such technologies. A high speed device with 256 X 256 pixels in a 40μm pitch is designed to provide up to 800 Hz full frame rate with pixel rates as high as 80 Mpixels/s. The read out circuit is designed to stare while scan in a flash integration mode to allow nearly full frame integration for even 800 Hz frame rate. A miniaturized command and control electronics with 14 Bit deep digital output and a non uniformity correction board capable to take into account non linear self learning scene based correction models are developed together with the integrated detector cooler assembly (IDCA). As working horse for dual color/band capabilities, AIM has developed a sequential multi color module to provide customers with a flexible tool to analyze the pros and cons of spectral selective detection. The module is based on a 384x288 mercury cadmium telluride (MCT) detector available in the mid wave (MWIR) or long wave spectral band (LWIR). A rotating wheel with 4 facets for filters or microscanner plates provides spectral selectivity. AIM's programmable MVIP image processing is used for controlling the detector and for non uniformity correction. The MVIP allows set the integration time and NUC coefficients individually for each filter position for comparable performance to accurately evaluate the pay off of spectral selectivity in the IR. In parallel, a dual color detector FPA is under development. The FPA is realized as a MCT MWIR device, LWIR, however, is also doable. Dual color macro cells are realized with 192x192 pixels in a pitch of effectively 56 μm. The cell design provides, that both colors detect radiation from target points identical within the limited resolution of the optics to ensure coincident detection plus compensates the significant variation in photon flux of the different colors to output the analog signal at approximately the same level for good thermal resolution and correctability. The photovoltaic device is realized using AIM's mature liquid phase epitaxy. Since quantum well (QWIP) technology has proven state of the art results based on a well established material system, AIM is heading for QWIP devices for most affordable solutions in the MWIR/LWIR dual band applications. A summary of state of the art results achieved so far as basis for a QWIP dual band detector is presented.

The objectives of the Low Cost Microsensors (LCMS) Program are twofold. The first is to develop and deliver a long-range infrared (IR) sensor built upon an uncooled vanadium oxide (VOx) 640 X 512 format focal plane array (FPA) engine. The second is to develop an expendable microsensor built upon a VOx 160 X 128 format FPA engine. The 640 X 480 sensor is applicable to long-range surveillance and targeting missions and is a reusable asset. The 160 X 120 sensor is designed for applications where miniaturization is required as well as low cost and low power. The 160 X 120 is also intended for expendable military applications. The intent of this DUS&T effort is to further reduce the cost, weight, and power of uncooled IR sensors, and to increase the capability of these sensors, thereby expanding their applicability to military and commercial markets never before addressed by thermal imaging.

The HgCdTe high-density vertically integrated photodiode (HDVIP^TM) concept developed at DRS Infrared Technologies is described. This technology is currently in production in both large-area scanning and staring focal plane array (FPA) formats. Detector models are presented and compared to performance data from scanning and staring FPAs. Performance data from 256 X 256 and 640 X 480 LWIR and MWIR staring FPAs, in keeping with these models, is presented with responsivity and D* operabilities in excess of 99.9%. Third generation system requirements mandate megapixel FPA operation at high temperatures, with multi-color capability, and high frame rates. To this end operation of 640 X 480 MWIR HgCdTe FPAs has been demonstrated at temperatures in excess of 150 K, and the push to these higher operating temperatures, with its effect on system cost, is discussed. The technology has also been extended into the realm of simultaneous two-color detection with large area formats, and this effort is described.

AIM has developed a sequential multicolor thermal imager to provide customers with a test system to realize real-time spectral selective thermal imaging. In contrast to existing PC based laboratory units, the system is miniaturized with integrated signal processing like non-uniformity correction and post processing functions such as image subtraction of different colors to allow field tests in military applications like detection of missile plumes or camouflaged targets as well as commercial applications like detection of chemical agents, pollution control, etc. The detection module used is a 384x288 mercury cadmium telluride (MCT) focal plane array (FPA) available in the mid wave (MWIR) or long wave spectral band LWIR). A compact command and control electronics (CCE) provides clock and voltage supply for the detector as well as 14 bit deep digital conversion of the analog detector output. A continuous rotating wheel with four facets for filters provides spectral selectivity. The customer can choose between various types of filter characteristics, e.g. a 4.2 μm bandpass filter for CO₂ detection in the MWIR band. The rotating wheel can be synchronized to an external source giving the rotation speed, typical 25 l/s. A position sensor generates the four frame start signals for synchronous operation of the detector -- 100 Hz framerate for the four frames per rotation. The rotating wheel is exchangeable for different configurations and also plates for a microscanner operation to improve geometrical resolution are available instead of a multicolor operation. AIM's programmable MVIP image processing unit is used for signal processing like non- uniformity correction and controlling the detector parameters. The MVIP allows to output the four subsequent images as four quarters of the video screen to prior to any observation task set the integration time for each color individually for comparable performance in each spectral color and after that also to determine separate NUC coefficients for each filter position. This procedure allows to really evaluate the pay off of spectral selectivity in the IR. The display part of the MVIP allows linear look up tables (LUT) for dynamic reduction as well as histogram equalization for automatic LUT optimization. Parallel to the video output a digital interface is provided for digital recording of the 14 bit corrected detector data. The architecture of the thermal imager with its components is presented in this paper together with some aspects on multicolor thermal imaging.

This paper summarizes our results of detecting composite delamination using infrared camera. Rivets in airframe are usually made of composite materials. It is very difficult to detect rivet delamination as conventional optical methods can not identify the delaminations. Here a neural net-based image- processing tool was developed by Intelligent Automation, Inc. (IAI) to process the infrared images. The tool consists of Fast Fourier Transform (FFT), Principal Component Analysis (PCA), and Fuzzy CMAC (Cerebellar Model Arithmetic Computer) neural networks. Results show that our tool can accurately pinpoint those delaminated rivet heads.

A gated multi-cycle integrator (GMCI) is presented to recover weak repetitive image signal from strong background. The GMCI could operate in several modes such as Capacitive- transimpedance amplifier (CTIA), gated integration (GI) and background-cancellation integration (BCI). When GMCI operates at BCI mode, the storage well of a pixel is mainly used for signal integration even there exists strong background or large dark current. Thus the signal-to-noise ratio (SNR), dynamic range and the sensitivity of detection are greatly improved. In addition, the transmission windows of BCI peak at odd harmonics of the modulation frequency. Therefore the detector's 1/f and other low frequency noises can be attenuated. A switched capacitor integrator (SCI) was designed to carry out the performance of GMCI. The switch induced fixed pattern can be obliterated by taking the differentia of two multi-cycle integrated signals with 180° phase difference. Preliminary chip test shows that the GMCI can read out modulated signal that is five orders less than the background.

To help system designers to evaluate the performances of a 128 * 128 cooled IRFPA, a behavioral model has been realized. It simulates the characteristics of the component in term of temporal and fixed pattern noise. This model was designed to be incorporated in a complete simulation of an optronic system including IR scene generation, optics modeling, post-treatment and corrections of the simulated video signal delivered by the detector. By this way, costs and duration of system evaluation for different kinds of scenario were strongly reduced. The architecture of the model was based on electrical simulations of the circuit. Its parameters were fitted with measurements on a real IRCMOS demonstrator in order to reflect with precision its behavior. In this paper we present the architecture of the IRFPA model. Comparisons between modeling and measurements are also presented exhibiting a real good agreement in term of video signal uniformity and noise. This demonstrates the advantages of this modeling approach for optronic system evaluations.

Two types of focal plane image processing chips are presented. They address the two extremes of the application spectrum: general purpose and application specific designs. They both exploit the promise of focal-plane computation offered by CMOS technology. The general-purpose computational sensor, a 16x16 pixels prototype (easily scalable to larger arrays), has been fabricated in a standard 1.2 μ CMOS process, and its spatio-temporal filtering capabilities have been successfully tested. An array larger than 300x300 array will use only 0.5% of the chip area for the processing unit while providing multiple spatio-temporally processed images in parallel. The 16x16 chip performs 1 GOPS/mW (5.5-bit scale-accumulate) while computing four spatio-temporal images in parallel. The application specific system realizes a hybrid imaging system by combining a 120 X 36 low-noise active pixel sensor (APS) array with a 60x36 current mode motion detection and centroid localization array. These two arrays are spatially interleaved. The APS array, which integrates photo-generated charges on a capacitor in each pixel, includes column parallel correlated double sampling for fixed pattern noise reduction. The current mode array operates in continuous time, however, the programmable motion detection circuit indicates if the intensity of light at pixel is time varying. The centroid, x and y position, of all time varying pixels is computed using circuits located at the edges of the array. Clocked at greater than 60 fps, the chip consumes less than 2 mW.

MOSAD©, Multiplexed OverSample Analog to Digital conversion, is a low power on focal plane analog to digital, A/D, process that places an oversample A/D at each pixel site. Two designs for a visible light staring array were developed with this approach. One used a silicon photo diode and the other used a photo gate for detection. The array was designed with a 320 X 240 format with the pixels placed on 16 micron centers. There are a total of 76,800 A/D's on the chip. The device is a monolithic integrated circuit that includes the sensors, A/D's and readout circuitry. A production 1.2 micron CCD/CMOS process was used in it construction. The A/D uses charge well switching at the pixel to convert the accumulated analog signal to digital data. There was negligible impact on the pixel area due to the A/D such that a fill factor of 73% was achieved with front side illumination for both approaches. At 400 samples per second, measured on chip power consumption is under 10 milliwatts. Noise measurements at sample rates from 400 samples per second to 1,600 samples per second were taken for both parts. It was found that the photo gate noise performance was four times better than the photo diode. At a nominal 28 times oversample, the photo diode obtained 8 to 9 bits performance and the photo gate achieved 10 to 11 bits. Nonuniformity variation was below the noise floor. No explanation for the difference in noise performance has yet been determined. This development was sponsored by NASA under a SBIR program.

Raytheon's Infrared Operations (RIO) has invented and developed a new type of focal plane array with 'Image Processing on-the-chip' named the Adaptive IR Sensor (AIRS). The AIRS FPA is based upon the human retina in that it performs signal processing near the photoreceptors. The AIRS FPA has been reduced to practice and adaptively removes detector and optic temperature drift and 1/f induced fixed pattern noise. This 3rd-generation multi-mode IRFPA, also called a Smart FPA, is a 256 X 256-array format capable of operation in four modes: (1) Direct Injection (DI), (2) Adaptive Non-uniformity Correction (NUC), (3) Motion/Edge Detection, and (4) Subframe Averaging. Nova Research has developed a Modular IR Application Development System (MIRIADS) which is a compact single board camera system that is highly integrated with the dewar assembly. The MIRIADS system coupled with the AIRS Smart FPA results in a very high performance wide field of view IR Sensor and processing system with integrated in one of the smallest packages to date.

A midwave infrared (IR) imager with silicon lenses and a laminated plastic surface has been made. It has a focal length of 50 mm and f/# of 2.6. Typically this type of lens system would be made out of silicon, germanium, or some other materials transparent in the midwave infrared regime. The design of the lens system would include aspheric and/or diffractive surfaces that require the use of a single point diamond turning machine to make it. In this system aspheric/diffractive silicon lenses have been replaced with a flat piece of silicon coated with a thin layer of plastic that has an aspheric/diffractive surface.

Unlike visible optics, infrared optics and optical systems pose some unique challenges. From the materials typically used for IR systems to cold stops to thermal behavior, many differences exist for the designer, manufacturer, and user of IR components and systems. Even testing IR systems can be an issue, since the eye is not sensitive to these wavelengths of electromagnetic radiation. This paper will attempt an overview of some of the unique issues associated with infrared systems and is not intended as a complete reference for infrared optical systems. The sited references contain substantial additional information on these and other subjects related to optical and infrared systems.

Infra-red (IR) seekers on missiles at high Mach-numbers in the lower tier air defence often suffer from degradation in performance due to aerothermodynamic effects. The kind and rate of degradation depends on the geometric design (shape) and location of the IR-window. Optimal design may reduce those effects but still misses to totally withstand the imposed thermal stresses (thermal shock). Proper thermal protection systems and/or window cooling systems will be needed. The first part of this paper deals particularly with passive IR- window design features to reduce the thermal stresses. A series of wind-tunnel testings focused on the thermal shock behavior of different IR-window shapes under critical flight conditions. The variation of typical design parameters demonstrates the available features to reduce thermal shock by passive ways. The second part presents active thermal stress reduction devices, e.g. an active cooling system. Among others the most efficient reduction of thermal heating is based on three components: A partial coverage of the IR-dome to protect most parts against heating effects, a rotating system bearing the IR-dome and a liquid spray-cooling system in the gap between the cover and the IR-dome. The hemispherical or pyramidal dome can be located either midways in the missile nose section or sideways on the structure. The liquid spray cooling system combines both, a heat exchange by fluid evaporation and a heat transfer by fluid and gas cross flow (convection), causing a low fluid consumption. Such a cooling system along with their driving parameters and the resulting analytical performance will be presented.

The development of a compact and high performance MWIR zoom camera based on the 320 X 240 staring focal plane array (FPA) is described. The camera has 20 magnification zoom optics with the maximum 40°x 30° of super wide field of view and 7.6 cycles/mrad of resolving power with the operation of attached microscanning system. In order to correct non-uniformities of detector arrays, we have proposed the multi-point correction method using the defocusing technique of the optics and acquired the highest quality images. The MRTD of our system shows good results below 0.05 °K at spatial frequency 1 cycles/mrad at narrow field of view.

High quality extended-wavelength InGaAs (80% In) has been grown heteroepitaxially on 3'-GaAs substrates by means of Molecular Beam Epitaxy (MBE). In contrast growth on InP, sensor manufacturing on GaAs substrates can be readily scaled to 6'. An additional benefit is the larger cut-off of the 80%- In material (2.5μm vs. 1.7μm). The photodiode layer design was optimized for maximum improvement of dark current with temperature. A high-uniformity and high- yield process was developed to manufacture bondable 128 pixel 50 μm pitch linear array sensors. To this end, a dedicated mesa-type process was developed assuring high uniformity in dark current and photoresponse. The sensors were designed for top-illumination. Sensors have been characterized electrically and optically at the intended thermo-electrical operation temperature (250K). The homogeneity of dark current and photoresponse was also evaluated at 250 K. The improvement in dark current as a function of temperature has been studied in more detail, yielding a zero bias resistance of approximately 2.5 MΩcm² at 77K. A linear array has been bonded to a commercial CMOS read-out circuit and the functionality is shown. The sensor is suitable for spectroscopic and imaging applications in the short wavelength infrared (SWIR) region. Work is underway to develop sensors with smaller pitch and up to 512 pixels.

The need for higher performance short wave length infrared (SWIR) cameras for the telecom industry is becoming increasingly important. This paper will discuss the measured performance and capabilities of Indigo System's near infrared cameras. These cameras are ideally suited for telecom applications because of their high quantum efficiency, uniformity, and linearity over the 1.0 to 1.65 μm wavelength band. The cameras' high dynamic range and high damage threshold to laser light also make them ideally suited for testing and characterizing S-Band, C-band, and L-band fiber optic systems used in DWDM. Measurements of the camera's sensitivity, linearity, uniformity, and dynamic range will be presented and compared to what is currently considered the industry standard, pyroelectric vidicon cameras. The performance improvement resulting from Indigo's approach for reducing the effects of the sensor package window when imaging coherent light will be demonstrated. In addition, an analysis of the camera's tolerance to high levels of laser light will be discussed.

The main activity of our laboratory is based on the design and the fabrication of thermal microsensors whose infrared radiometers are a significant part. These microsensors work by converting infrared radiation into spatial periodical temperature gradients patterned on substrate. An array of plated bimetallic microthermocouples is designed to produce the output voltage from the maximum generated temperature differences. Until now, planar sensors featuring a metering area from 3 to 100 mm² have been realized on glass or kapton (submitted for SPIE 2001) substrates. To take advantages of silicon technology, it is necessary to adapt the structure of the sensors to such a material. Indeed, the sensitivity of this kind of sensor is approximately proportional to the thermal resistance between hot and cold junctions. Thus, taking into account the low thermal resistivity of silicon, membranes must be recessed what makes it possible to strongly increase the thermal resistance of the zone located under the junctions of the thermocouples. However, to preserve the advantage of immunity against convection, these membranes must present quite particular properties of symmetry and periodicity compared to the thermocouples. In this case, the microthermopile is constituted of alternatively doped N and P polysilicon layers. Using a simple model, operation principle is analyzed and the microradiometer sensitivity can be computed. The microsensors are manufactured within the framework of a national project (CNRS Interlab) and the successive steps of the technological process are described. Lastly, the first experimental results are presented.

In this paper the last results obtained in the field of design and realization of planar infrared microsensors are presented. These sensors can be considered as a pattern of cells electrically serialized. For each cell, the incident radiation is converted into heat transfer by way of alternatively absorbent and reflecting areas. The center of each area is crossed with a thin microthermocouple whose hot and cold junctions are submitted to the superficial thermal field. By using micromachining, cell dimensions can be shrinked and 5 X 5 cm² microsensors have been manufactured with more than 3000 cells. Kapton^TM is used as substrate and a liquid resin polyimide intended to constitute the infrared absorbing layer. To determine the intrinsic absorption spectrum of this resin, processing a membrane of some cm² was needed. In this case, the spectral transmittance of this membrane was measured with an infrared spectrometer (Perkin - Elmer) and absorptivity can be mathematically deduced. The sensitivity represented by the ratio between the voltage delivered by the sensor and the absorptive heat flux is calculated by the way of a monodimensional analytical model dependent on a parameter representing the penetration depth of heat in the monolithic substrate. This parameter is computed from 2D finite elements modeling and takes into account the geometrical characteristics of each basic cell constituting the sensor. Finally, by multiplying the absorption spectrum with the sensitivity, the curves of sensors spectral sensitivities in the range 5 - 20 μm can be deduced.

In this paper we present a new membrane transfer bonding technology for fabrication of uncooled infrared focal plane arrays (IRFPAs). The technology consists only of low temperature processes, thus, it is compatible with standard integrated circuits (ICs). In the future this technology may allow infrared detectors with high temperature annealed, high performance thermistor materials to be integrated in CMOS based uncooled IRFPAs. The infrared detectors and the ICs are processed and optimized on different wafers. The wafer with the detectors (sacrificial detector-wafer) is bonded to the IC wafer (target wafer) using low temperature adhesive bonding. The detector-wafer is sacrificially removed by etching or by a combination of grinding and etching, while the detectors remain on the target wafer. The detectors are mechanically and electrically contacted to the target wafer. Finally, the adhesive bonding material is sacrificially removed. One of the unique advantages of this technology is the ability to integrate small, high temperature annealed detectors and ICs. We have applied membrane transfer bonding to the fabrication of arrays of infrared bolometers with polycrystalline silicon thermistors. In principle, membrane transfer bonding can be applied to the fabrication of any type of free-standing transducer including bolometers, ferroelectric detectors and movable micro-mirrors.

Optimization of optronics sensors requires models that compute the effective solid angle of a cold screen as seen from any point in the focal plane. Until now, the problem has been theoretically solved for circular screen, and some works solved the problem for a linear detector arrays. This article presents the exact mathematical formula of the effective solid angle at any point in the focal plane, thus preventing digital integral computation on each pixel of a detector matrix.

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 ((λ_c approximately 5 μm at 98K), MWIR (λ_c approximately 9 μm at 98K) LWIR (λ_c 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 Hg_1-xCd_xTe 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.

Thermopiles are uncooled, broadband detectors that require no chopper or temperature stabilizer. Their wide operating- temperature range, lack of temperature stabilization, and radiometric accuracy make thermopiles well suited for some space-based scientific imaging applications. These detectors may also offer advantages over bolometers for night vision. Previous work at JPL has produced thermopile linear arrays with D* values over 10⁹ cmHz^1/2/W by combining high- performance thermoelectric materials Bi-Te and Bi-Sb-Te with bulk micromachining processes. To date, however, 2-D thermopile arrays have demonstrated only moderate performance. The purpose of the present work is to improve thermopile 2-D arrays substantially by combining Bi-Te and Bi-Sb-Te thermoelectric materials with a unique pixel structure and low-noise readout circuitry. The initial goal is a 128 X 128 array with a single multiplexed analog output stream, with system D* values (including readout noise) of 10⁹ cmHz^1/2/W, and with a focal-plane power dissipation of 20 mW. 100 micrometers square detectors have been demonstrated with D* values of 2 X 10⁸ cmHz^1/2/W and response times of 4 ms. Models predict D* values well over 10⁹ cmHz^1/2/W for optimized detectors. Modeling of a preliminary readout design shows that, for the expected detector resistance of 100 kΩ, the total noise will be 50% higher than the detector Johnson noise. CMOS test chips containing front-end circuits presently display a noise about 2.5 times higher than modeled and a power dissipation of 0.6 μW per pixel.

Flexible film electrostatic MEMS actuators can be used as micromachined IR choppers for pyroelectric and microbolometer sensors. The flexible actuators act as tightly curled shutters, providing transmission of IR radiation to the sensor elements when open and reflection of the IR when closed. These actuators consist of a polymer/metal film stack which is microfabricated and released from a substrate. Thermal and mechanical stress in the film stack causes the actuator to curl when released, and the film can be uncurled by applying an electric field between the curled film and the substrate. Tightly curled actuators in the range of 50 μm to 1 mm square have been fabricated, and arrays have been produced and operated. Operating voltage is in the range of 50 - 300 V with frequencies > 5 kHz. The performance of these actuators is presented, and their applicability to IR choppers is discussed.

Quantum wells, especially those made of GaAs and InP related compounds, have enabled several unique infrared devices. A very successful example is the quantum well infrared photodetector (QWIP). Thermal imaging using focal plane arrays (FPAs) based on QWIPs is the main established existing application. In a different direction, the intrinsic short carrier lifetime (approximately 5 ps) makes the QWIP well suited for high speed and high frequency applications. In such cases, since lasers are commonly used, a high dark current can be tolerated. The most important parameter is then the absorption efficiency. For system simplicity and potential wide use, near room temperature operation is desirable. An optimization study is carried out and reported here, using GaAs/AlGaAs QWIP structures. High absorption (approximately 100%) and up to room temperature operation are achieved in devices having high doping densities and 100 quantum wells.

Front sector optronics is an optronical sensor, which provides surveillance, tracking and identification of multiple targets to the aircraft system. This sensor is a multifunction sensor, which is located in the nose of the aircraft. It will be on the Rafale F2 standard, which will be delivered in 2005. The paper will present the main operational advantages of FSO compared with the other sensors of the Rafale F2. Technical characteristics of the FSO will be described and the key points of the design of the equipment (technology and software) in relation with the operational requirements will be presented. The functions of the FSO will then be described, in particular the way the system or the crew, in association with other sensors uses them in order to designate the target to the weapons. A summary of the latest flight tests on Rafale will be presented.

This paper explains the technologies used for high-performance long linear arrays based on HgCdTe/CMOS hybrid multiplexers with bidirectional Time Delay and Integration (TDI) functions, and it describes the development of the first high-resolution Forward Looking Infrared (FLIR) system with the SXGA format. Long-wavelength Infrared (LWIR) photodiode arrays are fabricated using liquid-phase epitaxially grown HgCdTe on a CdZnTe substrate. Each photodiode array consists of 480x8-element n+/n-on-p diodes formed by B+ implantation. Each photodiode is surrounded by a crosswise drain diode to define the detection area. The diodes with a 10.3-μm cutoff wavelength had a typical zero-bias resistance of 10 MΩ and a shunt resistance of 1 GΩ. Four CMOS Read Out Integrated Circuits (ROICs) were used for bidirectional TDI and multiplex operations where each ROIC summed up and multiplexed eight signals from 120 channels. The ROIC also includes pixel deselection and gain control circuits along with the corresponding memory and writing means. The Infrared Focal Plane Arrays (IRFPAs) had a typical Noise Equivalent Temperature Difference (NETD) of 18 mK after TDI with F/1.55 optics and 10-μs integration. The FLIR system using the 480x8 IRFPA demonstrated a high spatial resolution of 1280 horizontal lines by 960 vertical lines (SXGA format) and NETD of less than 30 mK. The unique algorithm for image enhancement was successfully confirmed to be efficient.

Because the semiconducting YBaCuO films which are fabricated by sputtering have a temperature coefficient of resistance (TCR) over 3%/K at room temperature, they are considered to be candidates for bolometer materials of uncooled infrared (IR) detectors. There is a problem, however, in that the resistivity of the films is over 10 (Omega) cm, which is two orders of magnitude higher than that of conventional VO_X bolometer films. To decrease the resistance of the bolometers, we researched sputtering conditions of the YBaCuO films and combined them with comb-shaped electrodes. When the YBaCuO film was deposited on these electrodes by RF magnetron sputtering at room temperature in an atmosphere of 2% O₂ and 98% Ar, it showed a resistivity of 90 Ωcm and a TCR of -3.2%/K; ultimately the YBaCuO bolometer resistance became 82 k(Omega) using the comb-shaped electrodes. The YBaCuO bolometer detector that contains an infrared absorbing membrane achieved a high fill factor of 90% and high infrared absorptance of 79%. Moreover, the detector showed a thermal conductance of 1.3x10^-7 W/K and a responsivity of 6.8x10⁵ V/W in a vacuum. The YBaCuO microbolometer FPA which we have developed has an array format of 320x240 pixels and a pixel pitch of 40 μm. The FPA showed a noise equivalent temperature difference (NETD) of 0.08 K with a prototype camera and f/1.0 optics.

Optronics Mast Systems (US: Photonics Mast Systems) are electro-optical devices which enable a submarine crew to observe the scenery above water during dive. Unlike classical submarine periscopes they are non-hull-penetrating and therefore have no direct viewing capability. Typically they have electro-optical cameras both for the visual and for an IR spectral band with panoramic view and a stabilized line of sight. They can optionally be equipped with laser range- finders, antennas, etc. The brand name ATTICA (Advanced Two- dimensional Thermal Imager with CMOS-Array) characterizes a family of thermal cameras using focal-plane-array (FPA) detectors which can be tailored to a variety of requirements. The modular design of the ATTICA components allows the use of various detectors (InSb, CMT 3...5 μm , CMT 7...11 μm ) for specific applications. By means of a microscanner ATTICA cameras achieve full standard TV resolution using detectors with only 288 X 384 (US:240 X 320) detector elements. A typical requirement for Optronics-Mast Systems is a Quick- Look-Around capability. For FPA cameras this implies the need for a 'descan' module which can be incorporated in the ATTICA cameras without complications.

A pair of thin prisms is used to deviate the light beam without changing the image orientation in a vectorial shearing interferometer. The relative angle between prisms determines the position of the wave front and its tilt. This system is employed to control the displacement of a sheared wave front as a vector quantity and to introduce a controlled amount of tilt in a novel interferometric shearing system. The predicted performance of this wave-front director is confirmed experimentally.

A 2 stack two color Quantum Well Infrared Photodetector (QWIP) for simultaneously detection of Long Wavelength InfraRed (LWIR) and Near InfraRed (NIR) radiation is demonstrated. The LWIR sensor is based on a regular GaAs/AlGaAs QWIP, while the NIR sensor is based on a strained InGaAs/GaAs multi quantum wells structure. Optical and electrical measurements of the NIR detector and the inclusive device are presented and discussed. Aspects of the monolithic thermal imaging and see spot system are considered. System performance prediction is calculated based on the FLIR92 method, using MATLAB^R as a platform to the numerical calculations. The system calculations are based on a conversion method of the laser radiation at the target to a thermal like radiation, enabling calculation of the whole system parameters. This method allows one to predict the maximum range of the observing system from the target that still allows detection of the laser spot over the thermal picture.

The development of low-cost uncooled thermal LWIR FPAs is resulting in the emergence of a new generation of infrared sensors for applications where affordability is the prerequisite for volume production. Both ferroelectric detector arrays and silicon-based microbolometers are finding numerous applications from gun sights to automotive FLIRs. There would be significant interest in a similar uncooled offering in the MWIR, but to date, thermal detectors have lacked sufficient sensitivity. The existing uncooled MWIR photon detector technology, based on polycrystalline lead salts, has been relegated to single-element detectors and relatively small linear arrays due to the high dark current and the stigma of being a 50-year-old technology.

Mercury Cadmium Telluride focal plane arrays with well over 1000 pixels have been fabricated for a number of years. These FPA's have been built as large two-dimensional arrays of HgCdTe p-n junction diodes on a single CdTe or CdZnTe substrate. Sensitivity of each pixel to impinging radiation is one of the most important quality factors for these devices. However, material parameters, which give diode high sensitivity, are the same as those that cause cross talk between adjacent diodes in the array. This cross talk causes a blurred image and in general is a detrimental factor for the FPA system. The cross talk modeling is done in a two- dimensional simulation format to achieve high accuracy. In addition, the output information can be generated as a statistical function of the material and design parameter variations. Actual heterojunction FPA devices have been fabricated and tested for cross talk. In the paper, this data is compared to the simulation results. This design method and its algorithms are encapsulated in a software program called IRSIM. This physics-based simulator allows the engineer to use versatile geometries and material concentrations.

InAs-based alloys has been the subject of intense studies over the past few years, due to its importance as a material for photovoltaic detectors that can be tuned to optimum performance at any wavelength within 1 - 3.6 μm spectral range. We report here new results concerning electrical and photoelectrical characteristics of epitaxial InAs photovoltaic detectors optically immersed to GaAs microlenses and operating at near room temperatures. The devices are based on epitaxial heterostructures grown directly on thick GaAs substrates by molecular beam epitaxy. The performance of the detectors has been significantly improved by the use of multilayer heterostructures with optimized doping and band gap profiles, improved crystalline perfection of the heterostructure, improved processing resulting in reduction of parasitic series resistance and increase of shunt resistance of the device. Additionally, monolithic optical immersion to GaAs substrates has been applied. The devices have been characterized by current and capacitance measurements performed at steady-state and time dependent conditions at temperatures 77-300 K. The measurements provide an indication that tunneling and defect- assisted rather than band-to-band processes limit performance of the devices. The measurements of capacitance and series resistance of devices show picosecond RC time constant even for devices with relatively large apparent area (1 mm²). This is due to the reduction of junction capacitance by two orders of magnitude in comparison with non-immersed devices of the same optical area. The unique properties make possible more widespread applications of the optically immersed InAs devices in various infrared systems.

HgCdTe (MCT) infrared detectors offer the possibility to tune the material cut-off wavelength between 1 μm and 14 μm in order to adapt the detector detection waveband to the application need. At the same time the detector operating temperature can be tuned as well as the cut-off wavelength between 77 K and more than 230 K in order to offer high performance. Within this range of wavelengths, the SWIR sub- band (1 μm to 3 μm) presents a growing interest for military, commercial and space applications such as non destructive test, spectrometry, SWIR hyperspectral instruments for space applications, ... In addition to the interest of the waveband, these detectors present the advantage of a combination of high performance with the use of thermoelectrical cooler (TEC). Thus, in order to answer the market need for this type of application, Sofradir has developed infrared detectors in the SWIR waveband based on an adaptation of its MCT material to the required waveband and using low noise readout circuits adapted to operate with low photonic current. This paper presents Sofradir MCT detectors results in this waveband including possible trade-offs between the detector performances and operating temperature in function of the application. Then, the SWIR detectors developments and production at Sofradir are presented. In particular, a 320x256 SWIR focal plane array packaged with a TEC and its performances are shown and a particular attention will be paid to the development in progress of a large SWIR focal plane array with high performances between 150 K and 180 K for space application. Finally, trends for future developments in SWIR waveband at Sofradir will be presented.

It has long been known that 1/f noise is the bane of staring thermal imaging systems because the detector pixels are sampled at a low data rate. In the frequency range between about one Hertz and an upper limit set by the eye integration time, 1/f noise is quite distracting and detrimental to real-world system performance to an extent greater than would be expected from its impact on NETD. For very low frequencies 1/f noise amounts to drift. Because the noise in individual pixels is uncorrelated with noise in other pixels, this appears as a time-varying spatial noise, which degrades system performance more seriously than temporal noise. This paper quantifies these effects and shows that the electrical integration time is a critical factor in the rate at which 1/f noise degrades MRT after calibration. Use of a chopper reduces or eliminates the spatial noise, but with a sensitivity penalty. These results are particularly apropos to uncooled thermal imaging.

A 10-16 μm cutoff large format broadband quantum well infrared photodetector (QWIP) focal plane array (FPA) has been demonstrated. The size of the FPA is 640 X 512 and its pixel pitch is 25 microns. The highest operating temperature of the FPA is 45 K, and it was determined by the charge storage capacity and the other features of the particular readout multiplexer used in this demonstration. Excellent imagery, with a noise equivalent differential temperature (NE(Delta) T) of 55 mK has been achieved. In this presentation, we will discuss the development of this large format broadband infrared FPA based on a GaAs/AlGaAs materials system and its performance in quantum efficiency, NE(Delta) T, uniformity, and operability.