The Ultraviolet Limb Imaging Experiment (UVLIM) is part of the Air Force Space Test Program's STP-1 shuttle payload. The science objectives of UVLIM are to determine the electron density profile in the F2 region ionosphere and to measure its diurnal and global variation. The UVLIM scientific instruments include an extreme ultraviolet imaging spectrograph and a far ultraviolet 0.25 m scanning grating spectrograph. Dayglow emissions from H, He, N, 0+, 0 and N2 will be measured. The payload is supported during flight by the NASA Hitchhiker carrier which provides a complete real-time link between the UVLIM ground support equipment and the payload. Orbiter maneuvers align the experiment field-of-view with the earth's limb and sweep the line-of-sight vertically to provide limb scans.

Data measured by the Polar BEAR/AIRS UV Imager Experiment were processed to extract spatial radiance characteristics. Photometer mode 1304 Å dayside data measured on Julian Day 219, 1987, near mid-day were analyzed. The spatial structure of thermospheric dayside radiance at 1304Å appeared to be controlled principally by turbulence over spatial scales of 107 - 102 meters, with modifications imposed by Rayleigh scattering effects and magnetospherically forced phenomena. Spatial structure can be adequately modeled fractally, using dimensions based on Kolmogorov formalism modified by the Rayleigh scattering phase function. Mean radiance can be modeled using existing models of radiant intensity, resonance scattering, and absorption combined with thermospheric composition and general circulation models, such as MSIS-83, scaled to the mean and RMS intensities measured by Polar BEAR. The results can be incorporated in a background radiance simulation model that will provide a means for testing and refining phenomenological models of the structured earth background. This will be important not only for improving physical and chemical models of atmospheric features and processes, but it will allow parametric predictions of spatial structure and clutter to be developed for sensor modeling applications.

There is continuing need for information about the earth background clutter at ultraviolet wavelengths. This paper describes the methodology and the results obtained at 1304 Å wavelength from an analysis of the AFGL Polar Bear experiment. The basic measurement equipment provided data of a spatial resolution of 20 km over a large portion of the earth. The instrumentation also provided sampled outputs as the footprint scanned along the measurement track. The combination of the fine scanning and large area coverage provided opportunity for a spatial power spectral analysis that in turn provided a means for extrapolation to finer-spatial scale; a companion paper discusses the physical basis for this extrapolation.

Atmospheric radiance measurements and model calculations in the 200-290 nm wavelength region are found to be in good agreement. The calculations are made using the recently released LOWTRAN 7 transmission and radiance code. The measurements were made in 1978 from the 53-4 satellite. The model is in excellent agreement in regard to radiance level and to spectral variability in this region for the daytime, mid-latitude measurements. Under twilight conditions, the lack of airglow and fluorescence emission sources in the code lead to differences. Additions to the model to provide better agreement are under consideration, but at the present time, the code can be used for a wide range of atmospheric problems.

This paper presents a brief description of some recent work interpreting and analyzing data obtained by the AIRS sensor on the Polar BEAR satellite.

Recent satellite observations of the airglow in the 180-320 nm regime have provided detailed profiles of the noon, midnight, dawn and dusk limbs. Pitch maneuvers allowed spectrographs and imagers to scan the earth limbs through tangent altitudes from -100 km to +250 km. A UV imager observed the atmospheric emissions with an angular resolution of 10-4 radians/pixel, while a spectrograph simultaneously obtained UV spectra with a spectral resolution of 1 nm. The combination of these instruments allowed the vertical and spectral characterization of the earth limb emissions. Caused by solar fluorescence, the NO 7 band emissions dominate the noon, dusk, and dawn spectra, while the 02 Herzberg I emissions dominate the midnight limb. The peak in these emissions varies with solar zenith angle, with the peak emssion increasing in altitude with solar zenith angle. The measured intensities vary from 4 R at midnight at 200 km to 101 R at noon at 50 km.

A first-principles optical backgrounds model is applied to the analysis of UV backgrounds data obtained in 1988 from low earth orbit. The basic forms of the data are spectral matrices versus wavelength and tangent altitude (point of closest approach to the Earth's surface along the line of sight). Matrices were obtained in the far UV (FUV), middle UV (MUV), near UV (NUV) and visible spectral regions from about 250 to 0 km tangent altitude. Observing periods include noon, dusk, dawn and midnight. This paper will examine selected MUV and NUV data from the noon period for which the solar zenith angle was about 50°. We have two different interests in the analysis of these data. The first is to better understand optical backgrounds for discrimination purposes. In this case, our interest is on overall approximate behavior of backgrounds and how well our model describes this behavior. This will be the subject of the present paper. Our second interest is a scientific one where the emphasis is on what can be learned from the data about excitation mechanisms and composition. Analysis along these lines will be reported at a later time. From the present analysis, we generally observe agreement to better than 50% between theory and data for Rayleigh scattering of sunlight in both the MUV and NUV regions. This indicates good calibration of the instruments and that the model can be used with confidence to predict this component of the background. Less satisfactory agreement is achieved for the dayglow. While good overall agreement is seen at MUV wavelengths in spectral structure, the data frequently lie above the calculations by factors of two or more. Agreement is worse at NUV wavelengths. We are presently investigating possible sources of the disagreement which include model uncertainties and imperfect off-axis rejection of light by the instruments.

A brief review of recent models and measurements of the daytime limb radiance is given for the ultraviolet and visible spectral regions. Effects of molecular band systems and atomic emission lines as well as resonance fluorescence and Rayleigh scattering are included. The effects of sun position, atmospheric composition, solar activity and other variables are examined. Comparisons are made with recent experimental measurements. The transmission through the limb as a function of tangent height is calculated. A tentative clutter model, based on the variability of atmospheric species, is presented. Integration along the line of sight is performed to provide an improved limb clutter model.

Electrography has proven to be a useful tool for astronomical and aeronomical investigations. The addition of microchannel plate intensification provides significant enhancements in responsivity and resolution and offers the potential of providing detection of single photon events. In practice, long exposures and film fog have precluded the use of this kind of detector as a photon counting device. For low light levels or short exposures, it is possible with the correct algorithm to separate individual photon events from the film fog. This has several advantages including increased low light sensitivity, increased photometric accuracy, and increased resolution. A computer algorithm has been developed and applied to microchannel plate intensified electrographic emulsions to identify and centroid individual photon events. Modifications of this algorithm which treat the case of occasional overlap of photon events are discussed. This algorithm is very general and can be applied to other imaging detection techniques and can be used to analyze starfields in conventional astronomical photography.

An electron-bombarded Charge Coupled Device (EBCCD), when used with an external ultraviolet sensitive cathode, can be used as photon-counting UV detector with high quantum efficiency and wide dynamic range. We report on results obtained with the Texas Instruments VP1M 1024x1024 (18 micron square pixel) virtual phase CCD used with an electromagnetically focussed f/2 Schmidt camera. Excellent single photoevent discrimination and counting efficiency has been achieved with this device.

A number of different MAMA detector systems are now in use at both ultraviolet and visible wavelengths. MAMA detectors with formats of 256 x 1024 pixels and with pixel dimensions of 25 x 25 microns² are being used in the laboratory and at ground-based telescopes and an ultraviolet version has recently been flown on a Black Brant sounding rocket. Third-generation (224 x 960)-pixel MAMA detectors with 25 x 25 microns2 pixels are also being used in the laboratory and at ground-based telescopes and a (224 x 960)-pixel detector with 14 x 14 microns2 pixels is currently under test in the laboratory. Third-generation MAMA detectors with formats of 360 x 1024 pixels are under development for use at extreme ultraviolet (EUV) wavelengths on the European Space Agency/NASA Solar and Heliospheric Observatory (SOHO) mission and detectors with 1024 x 1024 and 2048 x 2048 pixels are under development for use at ultraviolet and far ultraviolet (FUV) wavelengths on the NASA Goddard Space Flight Center's Hubble Space Telescope Imaging Spectrograph (STIS). Both of these detectors have pixel dimensions of 25 x 25 microns2. This paper describes the configurations, modes-of-operation and some of the latest performance data for the different detector systems.

The performance of a high-gain curved-channel microchannel plate (C2MCP) with a spherical concave input face and a plane output face has been evaluated in the laboratory. This format allows the input face of the MCP to match a curved focal surface, such as in a Rowland circle spectrometer mounting, and, at the same time, permits the use of a high-resolution plane readout array in proximity focus with the output face. The MCP was evaluated in a discrete-anode Multi-Anode Microchannel Array (MAMA) detector system. The MCP tested had channel diameters of 12 gm, a rectangular active area of 9 x 27 mm2, and a front face radius-of-curvature of 250 mm. The length-to-diameter (LID) ratio of the channels varied from 136:1 at the edges of the active area to 106:1 at the center. The variation of the LID ratio across the active area of the MCP allowed the relationship between the saturated modal gain of the pulse-height distribution and the LID ratio to be examined from modal gain measurements. The saturated ,modal gain was found to be inversely proportional to the LID ratio and directly proportional to the applied MCP voltage. The measured performance characteristics are described and compared with gain models based upon the geometric parameters of the MCP.

We will present results showing the single pixel counting linearity of the MAMA (MultiAnode Microchannel Array) detector system for three different tubes. The relationship of MCP (microchannel plate) gain characteristics, amplifier discrimination level and ion feedback to single pixel counting linearity will be discussed. Comparisons of single pixel linearity between tubes using curved MCF's and a chevron pair will also be discussed.

We present a discussion of the critical issues in the design of a solid state photo-multiplier based on avalanche multiplication of carriers out of confined quantum states. Standard avalanche photodiodes utilize interband impact ionization to provide gain and as such produce electron-hole pairs. As is well known, low noise, high speed performance in these devices is typically frustrated by secondary hole ionization events. In order to overcome these limitations it is desirable to develop an avalanching unipolar solid-state device. Such a scheme is possible based on impact ionization of carriers out of confined quantum states. We present calculations of the impact ionization rate, gain, and dark current in representative GaAs/AlGaAs structures. Our model calculations are based on the average ionization rate calculated from the quantum mechanical transition rate assuming a drifted Maxwellian initial carrier distribution. It is found that the highest gain is attainable in an asymmetric quantum well structure in which the second barrier height is half as large as the initial barrier height. The effect of the electric field, and quantum well doping concentration on the device performance is assessed. Parallels are then drawn between the GaAs/A1GaAs device and a GaN/A1GaN device. It is expected that the GaN/A1GaN materials system is better suited than the GaAs/A1GaAs system for quantum confined state photomultipliers.

Two 0.125 meter low resolution spectrographs have been built covering the wavelength ranges of 125 - 231 nm and 180 - 339 nm. Each spectrograph was designed to have high sensitivity, large dynamic range, and modest spectral resolution. There were hard design constraints on collecting area and pointing knowledge. These spectrographs measured the spectrum using a linear array. The wavelength conversion was done via an image intensifier, converting a UV spectrum to green light. The spectrographs had a 0.5° nearly circular field of view with resolutions of <2 nm. After electronics processing, the spectrograph output consisted of a 128 bin spectrum every 0.2 seconds (one scan). The minimum sensitivity was 10 photons/cm2- sec. in a single output scan. With the use of a neutral density filter and an image intensifier automatic range system, the spectrographs had a dynamic range of >109.

The design and operational features of a spectroradiometer for the vacuum ultraviolet region are reviewed. A three-channel detector attachment has been developed for installation on a grating spectrograph. Each channel covers approximately 40 nm, and the concatenated range of the three channels extends from 156 to approximately 270 nm. The instrument is designed to perform spectral radiance measurements of extended sources. The design provides relatively fast spectral scanning and convenient calibration. All data are acquired and processed by microcomputer. An overview of the applications software is presented, along with representative calibration and application results.

A multianode MCP photomultiplier with a recovery time two orders of magnitude shorter than in conventional MCP devices has been developed for plasma diagnostics by laser scattering. The detector has an S20 photocathode, a low resistivity Z--plate electron multiplier and a 10 x 10 square array anode configuration. The Z-plate has a diameter of 25 mm, a resistance of 8.8 MO (at 20 °C) and is made of MCPs rated for operation at a DC strip current density in excess of 100 pA/cm². Once inside the vacuum tube the Z-plate must be operated with pulsed high voltage, to avoid overheating. For this reason this PMT is suitable for the detection of fast bursts of light pulses, occurring at a repetition rate sufficiently low to permit the heat dissipation between successive bursts. The thermal behaviour of the detector has been investigated. First the thermal coefficient of the resistance was measured giving a value of -6.00 ± 0.04 x 104 0/°C (around 20 °C); then the temperature increase of the Z-plate due to a single high voltage square pulse was measured as a function of the pulse energy. The heat capacity and the thermal time constant were found to be 0.93 J/°C and 175 s respectively. The PMT electrical behaviour has also been studied in pulse current mode. At 900 V/plate and with 25 ns FWHM input light pulses, the maximum output charge density was found to be 8.2 x 10^-10 C/cm² in linear regime and 4.9 x 10-9 C/cm2 in saturated gain mode. The Z-plate recovers from the maximum linear pulse in about 50 las.

The Multi Anode Microchannel Array (MAMA) detector is a photon counting imaging detector capable of operation over a wide spectral range. There is a fundamental difference between the MAMA detectors and "analog" detectors. The MAMA detector requires arrival time coincidence of a charge packet at multiple anodes to determine location while "analog" detectors derive location information from relative amplitudes of analog signals.

Neutral density filters have been designed that are rendered non-reflective. This has been achieved by adding anti-reflective layers and also subdividing the metal layer with dielectric spacers. Filters have been fabricated that obey Beer's law: Tt=Ta*Tb,

The Ultra Violet Plume Instrument (UVPI) is designed to image and make radiometric measurements of rocket plumes; image and conduct earth background measurements; and gather earth background clutter data in the 200 to 450 nm region. The instrument will be one of several experiments in an earth orbiting satellite to be launched by the Naval Research Laboratory. The optical system uses a ten cm aperture telescope with the image divided into two overlapping spectral regions, 200 to 360 nm and 250 to 450 nm. Four spectral filters isolate selected portions of the lower region for the plume camera and the upper region (i.e., 250 to 450 nm) is used primarily for tracking. Two intensified CCD cameras are used as sensors. Pointing and tracking capability for the UVPI system is provided by a two axis gimballed mirror with supporting electronics and software. The instrument can be configured for a specific encounter by ground command, or can be completely autonomous through use of its own onboard computer and data recorder.

The Ultra Violet Plume Instrument (UVPI) will measure launch vehicle plume intensities, clutter and earth background in the near and mid UV. Two intensified CCD (ICCD) cameras are used to collect UV irradiance and guide the UVPI via its own pointing and tracking capability. Radiometric calibration provided correlation between video output and irradiance for the 16 intensifier gain levels while tracking optimization emphasized acquiring and tracking low level radiance sources. The science, or plume, camera obtains radiometric imagery in a 0.13° FOV using four selectable filters to cover a 220 to 340 nm bandpass. The tracker camera output drives gimbal servos and keeps the UV source within target FOV. The tracker has a 280 to 440 nm bandpass and 2.2° FOV. Reliable tracking was achieved with a target irradiance of 1 x 1044 watts/cm2.

STARLAB is a seven-day Spacelab mission, presently scheduled for 1991, whose principal objective is to demonstrate acquisition, tracking, and pointing technology relevant to SDI requirements. This will be accomplished using various onboard sensors and lasers, and ground-launched missiles with target boards and other diagnostics. The STARLAB mission also involves a number of other experiments, including measuring and cor-recting optical aberrations and atmospheric effects on laser propagation, assessing the feasibility of submarine laser communication from space, and making high-resolution, multi-wavelength observations of missile plumes and hardbodies, and space, earth, and limb backgrounds. The STARLAB UV camera assembly (UVCA), developed by Lockheed's Palo Alto Research Laboratories, operates in the 0.20 to 0.32 urn waveband, and will support the passive plume imagery and background measurements. The UVCA can be operated automatically by the experiment computer or manually, with the shuttle payload specialist selecting filters, gain settings, and data-taking modes. The UVCA system consists of an f/25.5, 20-cm aperture Cassegrain telescope with an 8-filter wheel, and a two-stage, UV/visible image intensifier coupled to a CCD array. The Backgrounds Measurements experiment will examine the radiant intensities, as well as the spatial, spectral, and temporal variabilities, of the earth and limb. Pre-launch and on-orbit camera calibrations and characterizations will allow for absolute radiometry and camera boresighting. For all measurements, the onboard flight computer will store target coordinate locatidns and scene sun angle if appropriate. Though the STARLAB control system will normally point the line-of-sight and acquire data, the payload specialist will be able to manually point the LOS to view targets of opportunity such as hurricanes, abnormal weather patterns, aurora, nocti-lucent clouds or other mesospheric structures.

Magnetic field focusing of electrons affords higher spatial resolution as compared to the proximity focusing employed in most microchannel plate image intensifier tubes currently being manufactured.1,2,3 Magnetic focusing also affords a means of rotating the optical image between the photocathode and anode. A special 40 mm diameter MCP tube fabricated by ITT with 98 mm spacing between the photocathode and MCP has been evaluated to compare its spatial frequency response with that of a conventional proximity focused MCP image tube and to study effecting controlled electronic image rotation as a function of the ratio of magnetic field strength at the photocathode and MCP.

The Digicon detector, (in which accelerated photoelectrons are detected by solid state devices), has been used in various UV and visible photon counting applications, including the spectrographs in the Hubble Space Telescope'. Since the input photons are converted to electrons prior to detection, electronic image control such as gating, deflection and magnification is possible. A novel Digicon in which all three modes of image control are used has been designed. In this tube, magnetic deflection is used to randomly select a small area of the photocathode to be magnified for detection by a two dimensional diode array. Features of this design are presented along with preliminary data from demountable design-verification experiments. The data include magnified images in addition to the time dependence of the deflection magnetic field.

In support of programs in surveillance, star tracking and the use of intensified sensors, a series of CCD activities were begun in 1984 by Science Applications International Corporation. They now include the 1989 company-funded development of 1024 x 1024, 18 micron pixel CCD's, benefiting from technology transfer from JPL, and fabricated for SAIC by Ford Aerospace. The result has been a most successful series of CCD's and continued SAIC investment in 1) augmentations of the original Multi-Pinned Phase (MPP) version, 2) the first wafer configuration and funding commitment for Open-Pinned-Phase (OPP) (invented by J. Janesick), and now the configuration of a variety of larger and smaller versions of 18 micron pixel MPP CCD's, to be operated with either front- or back-illumination. This paper surveys activities which are now leading toward reduced noise and toward improved visible region QE and toward the acquisition of UV/Xray responsivity coupled with the features of the MPP CCD's.

We have developed an intensified charge-coupled device (CCD) bialkali photocathode image tube, based on the incorporation of a thinned backside-illuminated CCD as the target anode of a Hubble Space Telescope design Digicon tube. The delicate construction of the CCD target required development of a special low temperature (<230°C) processing schedule. The technique required to desorb all gasses, process a visual photocathode with adequate quantum efficiency and still remain within the time/temperature constraints dictated by the CCD is somewhat different than that employed for the typical multi-alkali photocathode tube fabrication. These techniques may be applicable to other types of complex design tubes where processing at >350°C is not possible. We discuss the specific approach we followed in developing this tube.

One of the military's newest carrier-based attack aircraft is powered by a modern turbojet engine. This turbojet engine provides additional thrust for combat maneuvers and carrier takeoffs from an afterburner (the burning of additional fuel "after the burning" in the main or core engine). To improve reliability of "light-off" characteristics and to provide safeguards against undesired afterburner "light-off" conditions, an ultraviolet flame sensor is employed. The flame sensor contains a UV detector which is a gaseous-discharge-type tube manufactured by Armtec Industries, Inc. In order to better determine the relationship between flame and sensor, measurements of the UV spectral response of the sensors were made at ambient and elevated (400°F) temperatures. The measurements, which were made by Optronic Laboratories at Armtec's Manchester, New Hampshire facility, utilized a double monochromator-based system optimized for measuring the spectral response of detectors over the 200 to 300 nm wavelength region. The paper describes the UV-flame sensor, the instrumental set-up, the measurement procedure, the NIST-traceable standards and the results and uncertainties associated with the measurements.

The multi-anode microchannel array (MAMA) is a photon counting detector which decodes the position of an event through coincidence discrimination. The decoding algorithm which associates a given event with the appropriate pixel is determined by the geometry of the array. In a standard MAMA detector, the pixel size is determined by the spacing of the anode array; however, the actual limiting factor to the pixel resolution is the spacing of the channels in the microchannel plate. The analog amplitude of charge amplifier pulses can be converted to digital quantities and employed for increasing the resolution of the detector by calculating the centroid position of each event in real time. This centroiding procedure can be implemented as an independent module of the decoder, operating in parallel with the standard digital decoding circuitry and providing the least significant bit(s) of the pixel address. Decoding hardware and techniques for enhanced pixel resolution are discussed in light of speed and complexity issues. A space-based realization of the MAMA detector requires that the decoding circuit be a single-chip monolithic or hybrid integrated circuit because of power, size and weight constraints. A single-chip CMOS gate array version of the decoder is discussed and contrasted with existing multi-chip decoders in terms of size, speed and power.

An 18-in. Cassegrain telescope and UV imaging system is used to observe and record starlight in the wavelength range 3000-3700 A . Such observations provide a real-time sampling of the atmosphere, allowing an accurate study of remote UV sources of unknown intensity. The spatial dependence for the trans-mission of radiation through the atmosphere has been calculated using the LOWTRAN 6 computer code, and the results are compared to data from various UV-emitting stars. Variations in the atmosphere are seen to strongly depend on aerosol concentration which affects meteorological range. Ground-level photometer measurements of extinction coefficients are extrapolated to yield atmospheric visibility profiles for our observation point.

Digital photometry of point-like images has been developed for the calibration of image systems sensitive to the UV and visible wavelengths and which use RS170 video. The photometry algorithm integrates the total net brightness signal over the area of the image. A threshold is determined based on the noise statistics in the image in order to separate signal from a noisy background. The response of this algorithm is linear with respect to brightness over a dynamic range of at least 3000 to 1.

The camera chosen for this study uses a narrow-band uv bandpass filter with a center wavelength at 254_nm and a refractive-optic uv lens. The uv image detector is a an ITT Type #F4146 photon-counting imaging detector with a uv sensitive photocathode used in conjunction with a Surface Science Laboratories position computer to generate the output images. This combination of filter and photocathode make the camera essentially solar-blind. On a clear day and in full sunlight there are essentially no background photon counts from outside scenes. Examples of a few open-flame and other uv sources and images produced by this camera are presented.

Imaging quality ultraviolet filters have been developed for research programs of the Mr Force Geophysics Laboratory (AFGL), for use in acquisition of high-resolution spectral images of rocket plumes. Narrowband and broadband filters have been produced with both high peak transmittance (10 to 25%) and superior rejection, typically 8 to 10 decades to the far infrared. Inherent image resolution quality of such filters has previously been estimated at 40 line-pairs per mm.

The design and preparation of polarizers used in high power laser systems at operating wavelength of 1060 nm are presented. Ta205 TiO2 composite films and Si02 are used in the manufacture of the polarizers, which are deposited on substrates with index of 1. 514. A Kaufman type ion source has beenapplied with ion energy ranging from 200 eV to 400 eV and current density up to 60 /4A/cm2 dur-ing deposition. The laser damage threshold and optical stability of the polarizers have been improved. The transmittance of the polarizers for p-polarized radiation is greater than 97% and for s-polarized radiation 0. 1 O. 5% hence extinction ratio is about 200-1000.

The Solar Blind Ultraviolet (SBUV) spectral region covers the interval between 230 nm and 290 nm. The lower limit is set by the edge of Shumann-Runge band and the upper limit is determined by solar radiation penetrating the stratospheric ozone shield. The SBUV spectral region is interesting from experimental point of view, since the lack of solar background in the troposphere is favorable in such applications as lidar, atmospheric communications and remote sensing. Our recent measurements indicated that previous model LOWTRAN-6 does not include 02 absorption at Herzberg I band. Newly released LOWTRAN-7 model includes only Herzberg continuum. It is suggested that LOWTRAN-7 use should be limited to stratospheric calculations. New experimental results are reported confirming the necessity of inclusion of Herzberg I absorption band in tropospheric calculations. Future plans for high spectral resolution measurements are discussed.