This paper describes the mechanical configuration of the science instrument module (SIM) used on the Advanced X-ray Astrophysical Observatory (AXAF-I). The SIM houses the focal plane scientific instruments of AXAF and provides instrument selection and focusing capabilities. It is designed to meet a large number of requirements of both the science instruments and the observatory. An overview of the SIM is provided and a brief description of the components and their functions is presented.

The high resolution camera (HRC) is a microchannel plate based imaging detector for the Advanced X-Ray Astrophysics Facility (AXAF) that will be placed in a high earth orbit scheduled for launch in August, 1998. An end-to-end calibration of the HRC and the AXAF high resolution mirror assembly (HRMA) was carried out at the Marshall Space Flight Center's X-Ray Calibration Facility (XRCF). This activity was followed by several modifications to the HRC to improve its performance, and a series of flat field calibrations. In this paper, and the following companion papers, we discuss the calibration plans, sequences, and results of these tests. At the time of this conference, the HRC has been fully flight qualified and is being integrated into the science instrument module (SIM) in preparation for integration into the AXAF spacecraft.

The high resolution camera (HRC) will be one of the two focal plane instruments on the Advanced X-ray Astrophysics Facility, (AXAF). AXAF will perform high resolution spectrometry and imaging in the X-ray band of 0.1 to 10 keV. The HRC instrument consists of two detectors, the HRC-I for imaging and the HRC-S for spectroscopy. Each HRC detector consists of a thin aluminized polyimide window, a chevron pair of microchannel plates (MCPs) and a crossed grid charge readout. The HRC-I is a 100 by 100 mm detector optimized for high resolution imaging and timing, the HRC-S is an approximately 30 by 300 mm detector optimized to function as the readout for the low energy transmission grating spectrometer (LETGS). In this paper we present the absolute quantum efficiency, spatial resolution, point spread response function and count rate linearity of the HRC-I detector. Data taken at the HRC laboratory and at the Marshall Space Flight Center X-ray Calibration Facility are presented. The development of the HRC is a collaborative effort between The Smithsonian Astrophysical Observatory, University of Leicester UK and the Osservatorio Astronomico, G.S. Vaiana, Palermo Italy.

The high resolution camera (HRC) is one of two focal plane detector systems that will be flown on the Advanced X-ray Astrophysics Facility (AXAF). The HRC consists of two microchannel plate (MCP) detectors: one to provide large area, high position resolution imaging and timing (HRC-I), and a second (HRC-S) to provide a readout for the AXAF low energy transmission gratings. Each detector is composed of a chevron pair of CsI coated MCPs with a crossed grid charge detector and an Al/polyimide UV/ion shield. In this paper, we describe the operation, performance and calibration of the spectroscopic detector. In particular, we discuss the absolute quantum efficiency calibration, the point spread function of the instrument combined with the AXAF telescope, the count rate linearity, the spatial linearity, and the internal background of the instrument. Data taken in the laboratory and at the x-ray Calibration Facility at Marshall Space Flight Center are presented.

The high resolution camera (HRC) is one of AXAF's two focal plane instruments. It consists of two detectors: the HRC-I which is optimized for direct imaging of x-ray sources; and the HRC-S which is optimized as the spectroscopic read-out of the low energy transmission grating (LETG). Both detectors are comprised of a chevron pair of micro-channel plates (MCPs) with a crossed grid charge detector (CGCD) and a UV/ion shield (UVIS). The role of the UVIS is to minimize the detectors' sensitivity to low energy electrons, ions and UV light, while providing sufficient x-ray transmission in the 0.1 to 10 keV x-ray band. In this paper, we report on the results of the flight UVIS calibration measurements. Specifically, x-ray and UV transmission measurements obtained at the HRC X-ray Test Facility of the Smithsonian Astropysical Observatory, and x- ray transmission measurements of UVIS witness samples obtained at a synchrotron light source facility.

The Smithsonian Astrophysical Observatory is performing the calibration of the AXAF High Resolution Mirror Assembly (HRMA) at the Marshall Space Flight Center. In order to determine the absolute incoming photon flux at the calibration facility a high purity germanium (HPGe) detector and an identical spare detector have been chosen as flux monitor of the HRMA x-ray detection system (HXDS). The detectors must thus be calibrated in the photon energy range between 700 eV and 10 keV. The calibration and characterization of the detectors was performed at the radiometry laboratory of the Physikalisch- Technische Bundesanstalt (PTB) at the electron storage ring BESSY. Two methods have been applied for the determination of the detection efficiency. Firstly, the undispersed calculable synchrotron radiation emitted by a bending magnet of the primary source standard BESSY has been used. Secondly, the detection efficiency has been determined using monochromatized radiation and Si photodiodes as transfer detector standards. The photodiodes have been calibrated with a relative uncertainty of 0.5% in the photon energy range from 50 eV to 1500 eV against a cryogenic electrical substitution radiometer which is a primary detector standard. It is shown that the determination of the detection efficiency in the whole desired spectral range is possible with a relative uncertainty between 1% and 2%, even in the photon energy range where the detection efficiency is affected by the x-ray absorption finestructures of the Al light blocking filter of the detector.

The high throughput x-ray spectroscopy mission XMM is a 'Cornerstone' project in the ESA Horizon 2000 Programme for Space Science. This observatory has at its heart three highly nested Wolter 1 grazing incidence x-ray telescopes which will provide a large collecting area (1500 cm² each at 1.5 keV). This optical system has a spatial resolution of about 16 arcsec and when coupled with reflection grating spectrometers and x-ray CCD cameras, it will provide a major advance in astrophysics by the end of the century. In this paper, we first present the design of the telescope and then describe the manufacturing and the integration processes of the telescope, with the emphasis on the production of the x-ray mirrors. We then concentrate on the improvements made since the successful qualification of the XMM mirror module, last year. Last but not least, the results achieved with the first two flight models of the XMM telescope are also presented with some prospective on the next generation of ultra thin mirror shells.

Monolithic arrays of 12 CCDs, 3 by 1 cm² each, have been developed and produced for the focal plane instrumentation of the European photon imaging camera (EPIC) on XMM and the German ABRIXAS x-ray satellite mission. The design parameters have been optimized to match the properties of the x-ray imaging optics as well as the x-ray intensity, energy bandwidth and characteristic time constants of the objects to observe. The pixel size is 150 by 150 micrometer²; readout is performed in parallel; low noise, spectroscopic performance is realized by on-chip integrated JFET electronics; highohmic, ultrapure bulk material allows full depletion and enhances the efficiency for higher energy x-ray detection. The fabrication process, the layout topology and the operating conditions guarantee for a ten year operation in space without performance degradation.

Two beam lines have been built at the Institute d'Astrophysique Spatiale (IAS) d'Orsay to perform absolute calibration of the EPIC (European photon imaging camera). EPIC consists of three x-ray charge coupled device (CCD) cameras having imaging and spectroscopic performances set at the Wolter telescope focal planes on board the x-ray multi mirror mission (XMM) planned to be launched by ESA in August 1999. To cover the desired 0.1 - 15 keV range a dedicated beam line has been built on each synchrotron sources of the Laboratoire pour l'Utilisation du Rayonnement Synchrotron (LURE): SACO (0.8 GeV) and DCI (1.5 GeV). Both beam lines are merging in a clean 23 m³ vacuum tank containing the camera to calibrate. (1) The SACO windowless beam line is equipped with a grating monochromator. Four plane VLS gratings are used to cover the low energy range (0.1 - 1.2 keV). A triple grazing incidence mirror system set in front of the entrance slit removes the overlapping orders. (2) The high energy beam line on DCI has a 50 micrometer beryllium window and a double flat crystals monochromator equipped with four different crystal pairs. A double grazing incidence mirror system set close to the source absorbs the high energy photon spectra. CCD calibrations will be performed during 1997 second semester and years 1998.

The pn-charge coupled device (pn-CCD) camera was developed as one of the focal plane instruments for the European photon imaging camera (EPIC) on board the x-ray multi mirror (XMM) mission. The homogeneously sensitive detector consists of four quadrants of three pn-CCDs each, which are integrated on a single silicon wafer. Each CCD has an area of 10 mm by 30 mm divided into 64 by 200 pixels with a depletion depth of 280 micrometers. Altogether the sensitive area is 60 mm by 60 mm. In the standard imaging mode (full frame mode) the CCDs are read out sequentially every 70 ms. In addition, different window modes allow imaging of brighter sources by restricting the detector area and reducing the integration time down to 6 ms. We have tested one quadrant of the EPIC pn-CCD camera at line energies from 0.52 keV to 17.4 keV at the long beam test facility PANTER in focus of the qualification mirror module for XMM as well as in a homogeneous x-ray beam. In this paper we describe the tests in the different imaging modes and report on the performance.

The pn-CCD camera is developed as one of the focal plane instruments for the European photon imaging camera (EPIC) on board the x-ray multi mirror (XMM) mission to be launched in 1999. The detector consists of four quadrants of three pn-CCDs each, which are integrated on one silicon wafer. Each CCD has 200 by 64 pixels (150 micrometer by 150 micrometers) with 280 micrometers depletion depth. One CCD of a quadrant is read out at a time, while the four quadrants can be processed independently of each other. In standard imaging mode the CCDs are read out sequentially every 70 ms. Observations of point sources brighter than 1 mCrab will be effected by photon pile- up. However, special operating modes can be used to observe bright sources up to 150 mCrab in timing mode with 30 microseconds time resolution and very bright sources up to several crab in burst mode with 7 microseconds time resolution. We have tested one quadrant of the EPIC pn-CCD camera at line energies from 0.52 keV to 17.4 keV at the long beam test facility Panter in the focus of the qualification mirror module for XMM. In order to test the time resolution of the system, a mechanical chopper was used to periodically modulate the beam intensity. Pulse periods down to 0.7 ms were generated. This paper describes the performance of the pn-CCD detector in timing and burst readout modes with special emphasis on energy and time resolution.

We have developed optical filters for ESA's x-ray astronomy project XMM (x-ray multi mirror mission). Specific CCDs will be used as detectors in the focal plane on board the observatory. Since these detectors are sensitive from the x- ray to the NIR (near infrared) spectral range, x-ray observations require optical filters, which combine a high transparency for photon energies in the soft x-ray region and a high opacity for UV (ultraviolet) and VIS (visible) radiation. With respect to the mission goal three types of flight model filters were designed having different spectral transmittance functions. We report on one of these types, a so-called 'thick' filter, which has been realized to be used as flight filter on board the observatory. This filter attenuates radiation below 10 eV by more than 7 orders of magnitude. It has an effective aperture of 73 mm without any support structure. A 0.35 micrometer thick polypropylene carrier foil is coated with metallic films of Al and Sn. We describe transmission measurements in the soft x-ray photon energy range to determine the thickness of the individual layers and present the optical performance data of the filter.

We describe the analysis of BeppoSAX gamma-ray burst monitor on-ground calibrations performed after the full integration of the spacecraft in order to explore in detail the dependence of the detectors efficiency on the direction and energy of impinging photons. Analytical techniques have been used to determine with reasonable accuracy this function by fitting the angular response at different calibration energies with simple models partially derived from the underlying physics and partially semiempirical. Satisfactory results have been obtained for the two detectors which have almost clean field of view and are co-aligned with the wide field cameras. Work is still in progress for the others. Preliminary results of ground calibration analysis have been already used to derive spectral information on gamma-ray bursts impinging parallel to the axis of the two best performing shields.

The Italian-Dutch satellite for x-ray astronomy BeppoSAX is successfully operating on a 600 km equatorial orbit since May 1996. We present here the in-flight performances of the gamma ray burst monitor experiment during its first year of operation. The GRBM is the secondary function of the four CsI(Na) slabs primarily operating as an active anticoincidence of the PDS hard x-ray experiment. It has a geometric area of about 400 cm² but, due to its location in the core of the satellite its effective area is dependent on the energy and direction of the impinging photons. A dedicated electronics allows to trigger on cosmic gamma-ray bursts. When the trigger condition is satisfied the light curve of the event is recorded from 8 s before to 98 s after the trigger time, with a maximum time resolution of 0.48 ms, in an energy band of 40 - 700 keV. As an instrument housekeeping the 1 s event ratemeter of the same detectors in the same energy band is stored regardless the trigger condition, allowing for an off- line detection of non-triggered events. Finally, the onboard software collects the event count rate that is used as anticoincidence, i.e. the events above a given energy threshold, typically kept at 100 keV. The flight-data screening is in progress, in order to extract real gamma ray bursts from the many sources of background. Already many results have been obtained, as those GRBs detected simultaneously with the wide field cameras oinboard BeppoSAX itself.

The gamma ray burst monitor onboard the BeppoSAX satellite is a secondary function of the anticoincidence shields of the phoswich detection system hard x-ray experiment. For this reason the four CsI slabs operating as gamma ray bursts detectors have a not uniformly clear field of view. Actually the other SAX experiments partially obstruct the GRBM FOV in a way that strongly depends both on direction and energy. This peculiarity makes very hard to build-up a real response matrix of the experiment by simply interpolating the on-ground calibration. Therefore a complex activity of Monte Carlo simulation has been started using the MCNP code, in which the entire SAX satellite is described in a 3D geometrical reconstruction. This code is being used for the simulation of the on-ground calibration set-up, and once a good level of confidence is reached on that, it will be used to reconstruct direction, intensity and spectrum of the cosmic gamma ray bursts. In this paper we present the Monte Carlo set-up, discussing the approach to the work and the approximations that need to be done. Then the first results of the simulations are shown and compared, for some monochromatic energies and for several incoming directions, to the results obtained during the on-ground calibrations.

The phoswich detection system (PDS) is one of the four narrow field experiments on board the x-ray astronomy satellite BeppoSAX. PDS is devoted to deep temporal and spectral studies of celestial x-ray sources in the 15 - 300 keV energy band. It also includes a gamma-ray burst monitor. In this paper we compare the expected and observed in-flight performances. Our estimate of systematic errors in the background subtraction and in the spectral reconstruction are also presented and discussed.

We report here the result of the structure measurement of a charge-coupled device (CCD) pixel with sub pixel resolution by using a new technique. The new technique makes use of a parallel x-ray beam and a metal mesh placed just in front of the CCD. The CCD camera we used in the first experiment, is a conventional system using the TC213 [Texas Instrument Japan (TIJ)] whose pixel size is 12 micrometers by 12 micrometers with one million pixels. The mesh has 4 micrometer diameter holes spaced at 12 micrometer intervals. We produced an efficiency map within a typical pixel showing the gate structure in detail. In the reconstruction process, we have to determine the mutual alignment between the CCD and the mesh in detail. The method we used can easily determine it with enough precision. By selecting single pixel events, we determined a pixel boundary. The distribution of two pixel split event can give us more information in the behavior of the primary charge cloud.

Measuring the polarization of x-rays emitted from cosmological objects yields explanations of the structure which characterize these sources. Polarization detection efficiencies of up to 18% have been measured for two, small pixel, charge coupled devices (CCDs) using an 80% polarized monochromatic synchrotron beam between energies of 7.5 keV and 35 keV. The device efficiencies at less than 15 keV are of particular interest for astronomical purposes where imaging, spectroscopy and polarization measurements can be carried out simultaneously. Polarization measurements using a CCD rely on the preferential direction of the ejected photoelectron along the E-field of the incident x-ray beam. The resultant charge cloud is sampled by the pixellated array of the CCD. It will be shown that the CCD polarization detection efficiency (modulation factor) is a function of the pixel size and the energy of the incident photons. The effect of depletion depth and impact of a field-free layer in the detector are reviewed. The two devices used were a commercial optical CCD, Kodak KAF1400, with 6.8 by 6.8 micrometer squared pixels and a specialized CCD, designed by EEV Ltd., deeply depleted with 4 by 9 micrometer squared pixels.

In the recent past both the CCD camera controller hardware and software have witnessed a dynamic change to keep pace with the astronomer's imaging requirements. Conventional data acquisition software is based on menu driven programs developed using structured high level languages in non-window environment. An application under windows offers several advantages to the users, over the non-window approach, like multitasking, accessing large memory and inter-application communication. Windows also provides many programming facilities to the developers such as device-independent graphics, support to wide range of input/output devices, menus, icons, bitmaps. However, programming for windows environment under structured programming demands an in-depth knowledge of events, formats, handles and inner workings. Object-oriented approach simplifies the task of programming for windows by using object windows which manage the message- processing behavior and insulate the developer from the details of inner workings of windows. As a result, a window application can be developed in much less time and effort compared to conventional approaches. We have designed and developed an easy-to-use CCD data acquisition and processing software under Microsoft Windows 3.1 operating environment using object-Pascal for windows. The acquisition software exploits the advantages of the objects to provide custom specific tool boxes to implement different functions of CCD data accusation and image processing. In this paper the hierarchy of the software structure and various application functions are presented. The flexibility of the software to handle different CCDs and also mosaic arrangement is illustrated.

The performance of a prototype photon counting imaging detector, being developed for the international UV space mission spectrum-UV, is presented. The detector is based on a 4-cm diameter, Z stack, high gain microchannel plate (MCP) intensifier endowed with a RbTe photocathode. The electron cascade generated by the MCP intensifier is transduced, via a phosphor screen and a 1:3.6 fiber optics reducer, into a 5 by 5 pixel², quasi-Gaussian charge distribution on a 15 micrometer, 512 by 512 pixel² format CCD matrix read out in the frame-transfer mode at 20 MHz, corresponding to 60 frame sec^-1 in the full frame mode and to 220 frame sec^-1 in the window (128 by 512 pixel²) mode. The data flow is acquired serially as to generate a 5 by 5 pixel² event sash that sweeps dynamically the CCD matrix at the 50 ns place of the readout clock. Each and every event sash is searched for the presence of events whose charge distribution lie within set thresholds and satisfy given morphological rules, i.e. a peaked charge profile. The centroid coordinates of identified events are subsequently determined with sub-pixel accuracy and stored in an external, high resolution memory. The data acquisition and processing system, based on field programmable gate array technology, is well able to resolve the front MCP pore geometry (10 micrometer diameter pores at 12 micrometer pitch).

The principles and a detailed study of the basic operation of a relatively new type of electron multipliers -- microsphere plates (MSPs) has been reported recently. In this paper we extend these studied by presenting measurements of bare MSP quantum efficiency at incoming radiation wavelength range of 250 - 1450 angstrom. MSP efficiency appeared to be by an order of magnitude lower than that of bare microchannel plates (MCPs), having maximum of about 1% at 350 - 900 angstroms. We also extend the previous investigation of angular dependence of MSP gain and detection efficiency to an angular range of 90 plus or minus 40 degrees, when no gain depression was observed, while detection efficiency varied only by approximately 7%. The spatial charge cloud distribution of microsphere plates was measured with the help of a phosphor screen, showing that the dependence is quasi-symmetrical although featuring granular formations caused by the intrinsic structure of the plate. We also present a detailed study of combined MCP/MSP stack operation, suggested earlier by L. B. C. Worth et al. The gain of the stack was measured to be relatively high (10⁸) with pulse height distribution FWHM values as low as approximately 62% and dark noise count rates less than 0.1 counts cm^-2s^-1, limited by the front MCP. The spatial resolution reached the best value of about 80 micrometers with a 250 micrometer gap between the plates and an accelerating bias in the gap of 50 V. The counting rate capabilities of this hybrid stack are much better (no gain drop was observed at count rates of 3.3 (DOT) 10⁵ counts cm^-2s^-1) than those of purely MSP detector (10³ counts cm^-2s^-1).

The microchannel plate, delay line, detectors developed for the far ultraviolet spectroscopic explorer mission to be launched in 1998 are described. The two FUSE detectors have a large format (approximately equals 184 mm by 10 mm split into two 88.5 by 10 mm segments), with high spatial resolution (less than 20 micrometers by 50 micrometers FWHM, greater than 9000 by 200 resolution elements) and good linearity (plus or minus 25 micrometers), high image stability, and counting rates in excess of 4 by 10⁴ events sec^-1. KBr opaque photocathodes have been employed to provide quantum detection efficiencies of 30 - 40% in the 900 - 1200 angstrom range. Microchannel plates with 10 micrometer pores and an 80:1 pore length to diameter ratio, with a 95 mm by 20 mm format have been used in a Z stack configuration to provide the photon amplification (gain approximately equals 2 by 10⁷). These show narrow pulse height distributions (less than 35% FWHM) even with uniform flood illumination, and good background levels (less than 0.3 event cm^-2sec^-1). Flat field images are demanded by the microchannel plate multifiber boundary fixed pattern noise and are stable.

Results are presented on the development of superconducting tunnel junctions suitable as imaging spectrometers for use at extreme ultraviolet (EUV) wavelengths (10 - 100 nm). Progress in improving the spectral resolution of both niobium and tantalum based devices appears such that tunnel limited resolutions (d(lambda) approximately 0.1 nm at 10 nm) should be achievable. In addition these detectors offer not only a high degree of linearity (less than 1%) but a significant efficiency together with a capability to handle rather high count rates. Progress in the development of an imaging instrument through the fabrication of close packed arrays is such that it has become clear that such an imaging spectrometer would represent a powerful and logical next step in the development of instrumentation for EUV astronomy.

Arrays of superconducting tunnel junctions (STJs) provide the possibility to perform high-resolution imaging spectrophotometry at x-ray wavelengths. We describe the applications of STJ arrays to x-ray astronomy, and present measurements on a 3 by 3 test array of Nb/Al-based STJs, operated at a temperature of 1.2 K, which illustrate the current photometric and spectroscopic capabilities of such devices. These results demonstrate the basic experimental feasibility of STJ arrays and indicate that there are no fundamental problems to be expected in the development of large-format x-ray detector arrays based on STJs.

Using a CdZnTe sample detector, a variety of diagnostic tools are applied, so as to assess the crystal characteristics and to compare these to the x-ray response measured with synchrotron radiation. Correlations are found, such that x-ray response degrading processes can be identified. In this respect the performance of the detector is found to be limited by both large scale defects such as some grain boundaries and also pipes and by crystal imperfections, together with impurities and other crystal defects, both at the surface and in the bulk crystal. The relatively soft CdZnTe crystals are very sensitive to improper handling, producing particularly surface damage, which in turn deteriorates the detector performance, as is rather clearly established by photoluminescence measurements.

Linear arrays of single photon avalanche detectors (SPADs) designed for use in low light level imaging applications were fabricated using a novel planar process that is compatible with standard CMOS technology. The device characteristics for these arrays are presented here to investigate their suitability for high efficiency low light level imaging. A new scheme is proposed to eliminate the problem of optical crosstalk between pixels in the array by introducing a trench isolation process coupled with silicon-on-insulator (SOI) technology.

The burst and all sky imaging survey (BASIS) is a proposed mission to provide plus or minus 3 arc-second locations of an estimated 90 gamma-ray bursts (GRBs) per year. The BASIS coded aperture imaging system requires a segmented detector plane able to detect the position of photon absorption to less than 100 microns. To develop prototype detector arrays with such fine position resolution we have fabricated many 15 mm by 15 mm by 2 mm 100 micron pitch CdZnTe strip detectors. A 2 by 2 prototype 100 micron CdZnTe strip detector array has been fabricated and has been used to test the capabilities of the BASIS imaging system. Preliminary shadowgrams of a 1 mm wide gap between two tungsten straight edges indicate that our position resolution is on the order of 69 micrometers. Both the array and imaging tests are described. A 6 by 6 element CdZnTe detector array is also being fabricated at GSFC. The assembly of this flight prototype array is discussed as well as applications for BASIS.

Silicon photodiodes offer a number of advantages over conventional photocathode type soft x-ray detectors in pulsed power experiments. These include a nominally flat response, insensitivity to surface contamination, low voltage biasing requirements, sensitivity to low energy photons, excellent detector to detector response reproducibility, and ability to operate in poor vacuum or gas backfilled experiments. Silicon photodiodes available from International Radiation Detectors (IRD), Torrance, California have been characterized for absolute photon response from 1 eV to 10 keV photon energy, time response, and signal saturation levels. We have assembled individually filtered photodiodes into an array designated the XUV-7. The XUV-7 provides seven photodiodes in a vacuum leak tight, electrically isolated, low noise, high bandwidth, x-ray filtered assembly in a compact package with a 3.7 cm outside diameter. In addition we have assembled the diodes in other custom configurations as detectors for spectrometers. Our calibration measurements show factor of ten deviations from the silicon photodiode theoretical flat response due to diode sensitivity outside the center 'sensitive area.' Detector response reproducibility between diodes appears to be better than 5%. Time response measurements show a 10 - 90% rise time of about 0.1 nanoseconds and a fall time of about 0.5 nanoseconds. Silicon photodiodes have proven to be a versatile and useful complement to our standard photocathode detectors for soft x-ray measurement and are very competitive with diamond for a number of applications.

The four kinds of crystals; RAP(001), Si(111), LiF(220) and the Co/C multilayer on the super polished Si(111) crystals, together make up the objective crystal spectrometer OXS. They cover a wide energy range extending from 0.16 eV to 8 keV. A study of crystal reflectivity and energy resolution including measurements on RAP, LiF and Co/C and a calculation of Si crystals in the respective wavelength bands has been performed and the results are presented.

The stellar x-ray polarimeter (SXRP) will be more than an order of magnitude more sensitive than any previous x-ray polarimeter in the 2 - 15 keV energy band. The SXRP is a focal plane detector for a Danish-Russian SODART telescope, which will be launched on the Russian spectrum-x-gamma (SXG) mission. The SXRP exploits the polarization dependence of Bragg reflection from a graphite crystal, and of Thomson scattering from a target of metallic lithium. The SXRP flight model (FM) was calibrated at a facility at Lawrence Livermore National Laboratory (LLNL) equipped with polarized and unpolarized x-ray sources producing x-rays in the band pass for the graphite and lithium scatterers. By adjusting the orientation of the SXRP with respect to the incident x-ray beam, it was possible to simulate the converging beam from a SODART telescope and to measure the SXRP response to telescope pointing errors. In this paper, we describe the SXRP-FM calibration and present results for the graphite polarimeter.

The optical filters on board the JET-X telescope comprise thin foils of aluminum coated Lexan. During ground calibration of the filters, narrow spectral regions of high UV leakage, with peak levels of up to a few percent, were observed in broad band optical measurements in the 1000 to 10,000 angstrom range. Furthermore, transmission values were typically up to two orders of magnitude higher than calculated for the aluminum thickness. Investigation showed that these effects were attributed to a combination of aluminum oxidation, which reduces the opacity, and the use of a double sided aluminum layer in the filter design which behaves as a Fabry-Perot interference filter. These effects were verified by a multi- layer model of the filter UV response. Recent redesign of the filters for the flight program eliminated the UV leakage by adopting a single aluminum layer configuration, thus eliminating interference effects, and increasing the thickness by 30% to compensate for oxidation levels. The integrated x- ray transmission below 1 keV was found to be only reduced by 3%. In parallel with the production of the new Lexan flight filters, a set of qualification model filters was produced by the Luxel Corporation in the USA. These filters use polyimide as a substrate material which has the advantage that it is optically opaque to wavelengths below 3000 angstroms, unlike Lexan which is transparent. These new filters were found to have superior mechanical strength, being able to survive extended qualification vibration without any visible degradation in performance, and had a higher cosmetic quality and attenuation levels. As a result, these filters have now been included in the JET-X flight program. We report on the optical tests results from both Lexan and polyimide filters along with high resolution x-ray transmission results carried out at the BESSY synchrotron facility in Germany. Results of the mapping of the filter edge structures, global transmission values and uniformity are presented.

Construction of the flight model joint European X-ray telescope (JET-X) for the Russian spectrum-X mission has been completed and performance tests and calibration of the instrument have been carried out. Separate measurements of the responses of the x-ray mirrors, the CCD detectors and the optical filters already indicate that JET-X will achieve spatial resolutions of around 20 arcsec, an on-axis collecting area of 310 cm² at 1.5 keV and an energy resolution of 130 eV at 6 keV. As a final step in the calibration of the telescope assembly, end-to-end x-ray tests on the complete instrument have been performed in the x-ray beam line facility at MPE Garching. Results from this calibration program are reported and the overall response of the two x-ray telescopes are compared with the previously measured responses of the mirror, the CCD detectors and the optical filters. In-orbit sensitivity responses are derived from these calibration data sets, for the normal operating modes of JET-X.

The x-ray monitor JEM-X will be flown aboard the ESA mission INTEGRAL for gamma-ray astronomy. It has been conceived to furnish images of celestial sources in fields of the sky 9.1 deg wide in the 3 - 100 keV extended energy band. Careful on- ground calibrations are needed to verify the spectral capabilities of the JEM-X detector. These operations in the 20 - 100 keV energy band are planned to be performed using a hard x-ray (15 - 140 keV) facility operated at University of Ferrara (Italy). We present the apparatus that will be used to carry out these calibrations. In particular we describe a double-reflection fixed-exit system of monochromatization based on silicon crystals that are implemented in the facility.

NASDA (National Space Development Agency of Japan) has selected MAXI as an early payload of the JEM (Japanese experiment module) Exposed Facility (EF) on the space station. MAXI is designed for all sky x-ray monitoring, and is the first astrophysical payload of four sets of equipment selected for JEM. MAXI will monitor the activities of about 1000 - 2000 x-ray sources. In the present design, MAXI is a slit scanning camera system which consists of two kinds of x-ray detectors; one with one-dimensional position sensitive proportional counters and the other with an x-ray CCD array employed for one-dimensional imaging. MAXI will be able to detect one milli-Crab x-ray sources in a few-day observations. The whole sky will be covered completely in every orbit of the space station. MAXI will be capable of monitoring variability of galactic and extragalactic sources on timescales of days with a sensitivity improvement of a factor of 5 or more over previous missions. NASDA and RIKEN have jointly begun the design and construction of MAXI. The payload will be ready for launch in 2003. In this paper we present the scientific objectives of MAXI, a basic design and some simulation results, after introducing the present status of JEM.

The burst and all-sky imaging survey (BASIS) project is a proposed small explorer (SMEX) mission to image the gamma-ray sky in the 10 - 150 keV energy range with high angular and energy resolution. It will be able to determine the locations of gamma-ray bursts (GRBs) to within a few arcseconds, sending accurate positions to ground-based telescopes for simultaneous and follow-up observations within seconds of the beginning of the GRB. It will also produce all-sky maps with 30 arcsecond resolution and 2 milliCrab sensitivity. The instrument uses a two-scale coded aperture mask to modulate gamma-rays falling on a cadmium zinc telluride (CZT) detector plane consisting of both 100 micrometer pitch strip detectors and 4 mm square spectroscopy detectors. The spatial pattern of gamma-rays will be deconvolved with the mask pattern to produce an image. This paper presents results from a prototype of this system, using a mask and strip detectors to produce an image of a radioactive source. The prototype functions as expected, producing images which, when scaled to the dimensions of the proposed instrument, achieve the desired resolution.

A wide field-of-view, arcsecond imaging, high energy resolution x-ray and low energy gamma ray detector is proposed for a future space mission. It is specifically designed to monitor and study gamma ray bursts (GRBs) with high energy and angular resolution and also find counterparts at other wavelengths. Detection of GRBs requires wide field-of-view ((pi) to 2 (pi) field-of-view) and high sensitivity. This is achieved by using high quantum efficiency CdZnTe pixel detectors with a low energy threshold (few keV) to observe the larger flux levels at lower energies, and large effective area (625 to 1,000 cm²) per coded aperture imaging module. Counterpart searches can only be done with ultra high angular resolution detectors (10 to 30 arcsecond FWHM) which gives 1 to 5 arcsecond position determination especially for strong GRBs. A few arcsecond size error box is expected to contain at most one object observed at another wavelength. This will be achieved by using ultra high spatial resolution pixel detectors (100 by 100 microns) and a similar resolution coded aperture to achieve the required angular resolution. AXGAM also has two other important advantages over similar detectors: (1) excellent low energy response (greater than 1 keV) and (2) high energy resolution (less than 6% at 5.9 keV, less than 3% at 14 keV, less than 4% at 122 keV). The low energy range may provide important new information on GRBs and the high energy resolution is expected to help in the observation and identification of emission and absorption lines in the GRB spectrum. The effective energy range is planned to be 2 to 200 keV which is exceptionally wide for such a detector. AXGAM will be built in the form of a 'bucky ball' using a coded aperture mask in a semi-geodesic dome arrangement placed over a two-dimensional, high resolution CdZnTe pixel detector array using newly developed p-i-n detector technology. The p-i-n structure decreases the electron and hole trapping effect and increases energy resolution significantly. The major scientific goals of the proposed mission in addition to continuously monitoring gamma- ray bursts, is to observe AGNs, transient phenomena, isolated and binary pulsars, and solar flares. A space deployed AXGAM detector is expected to observe several hundred gamma ray bursts per year.

The high-resolution EUV spectroheliometer (HiRES) is our first step towards the 0.1 arc sec angular resolution goal at EUV wavelengths. The HiRES instrument consists of a Gregory telescope with a 45-cm-diameter primary mirror, and an imaging EUV spectrometer, employing a single-reflection toric diffraction grating in a Rowland circle mounting and an imaging pulse-counting multi-anode microchannel array (MAMA) detector system. The MAMA detector covers the spectral range from 560 to 631 angstrom with a spectral resolution of 70 mAngstrom. A Pt/Ne lamp impresses a wavelength calibration spectrum on the upper part of the detector while the solar spectrum is recorded simultaneously on the lower part. The times-of-arrival and addresses of the detected photons are transmitted directly to the ground to allow correction of the SPARCS pointing jitter. The HiRES spectrometer slit is pointed to a specific location using real-time control of the SPARCS, and SPARCS can be commanded to execute a linear (push broom) scan across the sun. Two EUV photodiodes are used to measure the absolute sola irradiances and the atmospheric extinction profiles in wavelength bands centered at 304 angstrom and 584 angstrom respectively.

We present a conceptual design for a new x-ray all sky monitor (ASM). Compared with previous ASMs, its salient features are: (1) it has a focusing capability that increases the signal to background ratio by a factor of 3; (2) it has a broad-band width: 200 eV to 15 keV; (3) it has a large x-ray collection area: approximately 10² cm²; (4) it has a duty cycle of nearly 100%, and (5) it can measure the position of a new source with an accuracy of a few minutes of arc. These features combined open up an opportunity for discovering new phenomena as well as monitoring existing phenomena with unprecedented coverage and sensitivity.

We constructed an optical imaging x-ray detector with a capillary gas proportional counter (CGPC), a focusing mirror system, and an image intensified CCD camera. The CGPC consists of an absorption region for x-rays, which also operates as a drift region of electron clouds, and a gas proportional scintillation region in a capillary plate. With a gas mixture of Xe plus 2% CH₄ at 1 atm and a gas gain 8000, the light output from the CGPC is several tens thousand times larger than that of a typical NaI(Tl) scintillator and it is enough to image electron clouds due to a few tens keV x rays. To investigate the imaging ability, images for tracks due to (alpha) particles (²⁴¹Am) of 5.5 MeV were taken in the gas mixture of Xe plus CH₄, for the several reduced fields in the drift region. We confirmed that the images were most clearly observed at around the reduced fields of 40 V/cm(DOT)atm.

The Argonne/Toulouse collaboration is developing a crystal lens diffraction telescope for use as an astrophysical detector in the energy range of 200 keV to 1.3 MeV. The lens consists of 8 rings of diffraction crystals that all focus a narrow band of energies on a common HPGe detector. The inclination angle of these crystals controls the energy band being focused and will need to be adjusted over a range of 0.5 to 1.5 degrees with arcsecond precision to cover this energy band. At Argonne National Laboratory, a new lens frame was constructed and the inner ring was equipped with 16 Ge crystals of 1 cm³ size. The orientation of each crystal could be adjusted using a piezo-based picomotor in combination with a noncontact eddy-current sensor. The sensors have 0.1 - 0.2 arcsecond resolution; the motors have a step size of 0.05 - 0.2 arcseconds. By changing the crystal inclination and the distance of the detector from the lens, we were able to focus the 662 keV radiation from a ¹³⁷Cs source at 24.75 m as well as line energies at 276, 303, 356, and 383 keV from a ¹³³Ba source at 24.45 m. The sensor and system stability were demonstrated by alternately focusing line energies. We were able to simulate scans in energy of a spaceborne instrument as well as the enlargening of the energy repone by a slight detuning of the lens crystals. At the Advanced Photon Source (APS) Facility, an experiment to directly measure the diffraction efficiency of lens crystals from 200 - 500 keV using a beam with 3 arcsecond divergence was carried out. A double-crystal monochromator using two 3-mm-thick Ge crystal in Laue geometry was realized. The experimental results imply diffraction efficiencies for an astrophysical point source of 38% to 41% over the energy range for the crystals used.

The results of research of proton and neutron space radiation fluxes influence on spectrometric characteristics of high pressure xenon gamma-spectrometer (HPXGS) which has been working on board the orbital station 'MIR' since 1990 are submitted. It is shown that energy resolution and test gamma- line position in registered spectra have not practically changed for six years of gamma-spectrometer's work. Other type of HPXGS (the cylindrical ionization chamber) was subjected to neutrons irradiation of Pu-Be source in laboratory conditions. The analysis of these measurement results are given.

We have measured the permeation rates of helium and water through thin-foil UV-blocking filters used in the ASTRO-E/x- ray spectrometer (XRS) instrument. In the XRS program, there is a concern that outgassed contaminants such as water could permeate through the outermost filter which will be at room temperature and freeze on the inner filters which will be at cryogenic temperatures. The filters tested consisted of approximately 1000 angstroms Al on approximately 1000 angstroms of either Lexan or polyimide. Measurements were made using a vacuum apparatus consisting essentially of two small chambers separated by the filter under test. A helium leak detector was used to measure helium permeation rates, and a residual gas analyzer (RGA) was used to detect water. Results discussed include permeation rate as a function of pressure difference across a filter, the ratio of helium permeation rate over water permeation rate, and the effect of the aluminum layer thickness on permeation.

We report the scientific motivation for and performance measurements of a prototype detector system for SONTRAC, a solar neutron tracking experiment designed to study high- energy solar flare processes. The full SONTRAC instrument will measure the energy and direction of 20 to 200 MeV neutrons by imaging the ionization tracks of the recoil protons in a densely packed bundle of scintillating plastic fibers. The prototype detector consists of a 12.7 mm square bundle of 250 micrometer scintillating plastic fibers, 10 cm long. A photomultiplier detects scintillation light from one end of the fiber bundle and provides a detection trigger to an image intensifier/CCD camera system at the opposite end. The image of the scintillation light is recorded. By tracking the recoil protons from individual neutrons the kinematics of the scattering are determined, providing a high signal to noise measurement. The predicted energy resolution is 10% at 20 MeV, improving with energy. This energy resolution translates into an uncertainty in the production time of the neutron at the Sun of 30 s for a 20 MeV neutron, also improving with energy. A SONTRAC instrument will also be capable of detecting and measuring high-energy gamma rays greater than 20 MeV as a 'solid-state spark chamber.' The self-triggering and track imaging features of the prototype are demonstrated with cosmic ray muons and 14 MeV neutrons. Design considerations for a space flight instrument are presented.

We are developing depth-graded, multilayer-coated mirrors for astrophysical hard x-ray focusing telescopes. In this paper, we discuss the primary technical challenges associated with the multilayer coatings, and report on progress to date. We have sputtered constant d-spacing and depth-graded W/Si multilayers onto 0.3 - 0.5 mm thick DURAN glass (AF45 and D263) and 0.4 mm thick epoxy replicated aluminum foils (ERAFs), both of which are potential mirror substrates. We have characterized the interfacial roughness, uniformity, and stress of the coatings. The average interfacial roughness of each multilayer was measured from specular reflectivity scans ((theta) _i equals (theta) _r) using Cu K_alpha x-rays. The thin film stress was calculated from the change in curvature induced by the coating on flat glass substrates. Thickness and roughness uniformity were measured by taking specular reflectivity scans of a multilayer deposited on the inside surface of a quarter cylinder section. We found that interfacial roughness ((sigma) ) in the multilayers was typically between 3.5 and 4.0 angstrom on DESAG glass, and between 4.5 and 5.0 angstrom on the ERAFs. Also, we found that coatings deposited on glass that has been thermally formed into a cylindrical shape performed as well as flat glass. The film stress, calculated from Stoney's equation, for a 200 layer graded multilayer was approximately 200 MPa. Our uniformity measurements show that with no baffles to alter the deposition profile on a curved optic, the layer thickness differs by approximately 20% between the center and the edge of the optic. Interfacial roughness, however, remained constant, around 3.6 angstrom, throughout the curved piece, even as the layer spacing dropped off.

The high energy focusing telescope (HEFT) is a balloon-borne system for obtaining arcminute imagery in the 20 - 100 keV energy band. The hard x-ray optics are baselined to use thin epoxy-replicated aluminum foil substrates coated with graded-d multilayers, and we show some results on x-ray performance of prototype foil substrates. We also propose an extremely promising alternative substrate -- thermally formed glass. The advantages of thermally formed glass substrates, their fabrication and preliminary metrology on sample pieces are discussed. If ultimately feasible, the thermally formed glass is a better substrate due to its superior hard x-ray reflectivity and scattering properties in comparison to similarly coated epoxy-replicated aluminum foil. We also discuss some preliminary work on the HEFT mirror mounting concept and the associated angular resolution error budget.

We report the fabrication and evaluation of silicon diffraction gratings for use in the far- and extreme- ultraviolet. An interference technique was used to expose a layer of photoresist on a 10 cm silicon wafer in a series of parallel strips. V-shaped grooves were then etched into the wafer anisotropically. Diffraction efficiencies for a first pair of gratings at groove periods of 1.0 and 2.5 micrometers were measured. The process leaves much room for refinement, but shows promise in that the gratings produce clear diffraction orders with reasonable efficiency and have groove facets free of pitting or sharp ridges.

With the current generation of x-ray astronomy observatories either in orbit (ASCA, ROSAT) or well on the road to completion (Spectrum-X(gamma) , XMM, AXAF), plans are being developed for what is being classed as the next generation of x-ray observatories for launch by the space agencies of Europe, USA and Japan. In all of these mission concepts, together with many smaller mission concepts, the charge coupled device (CCD) is seen as playing a key detector role, building upon its successful use in current missions where it is used for both dispersive, and imaging, x-ray spectroscopy applications. This paper reviews the goals of the concept observatories and discusses near-term CCD developments which are needed to meet the future instrument requirements. In addition, since future observatories will have larger collecting areas, a concept design is proposed for a high throughput x-ray imaging/spectroscopy camera as a progression beyond that of the EPIC MOS camera on XMM.

The extreme UV imaging telescope (EIT) on-board SOHO is performing a global survey of the extreme ultraviolet (EUV) solar corona. Operating since January 96, EIT has been producing tens thousands of images of the Sun in four narrow channels (171, 195, 284 and 304 angstrom). orbiting around the L1 Lagrangian point and oriented permanently towards the Sun, the EIT mission is a unique opportunity to study an instrument continuously exposed to solar EUV radiations. The backside thinned CCD detector is showing significant changes in its overall signal and in local 'burn in' regions. Periodic bakeouts allowed to restore a good efficiency. However, a specific observation program has been set up to diagnose the origin of the signal decay. In this framework, photon transfer analyses are performed on solar EUV images, providing good indications on the local charge collection efficiency status. Calibration lamp images are also used to eluate the signal recovery in the visible range. The signal degradation seems to be the result of two competing effects: periodic deposition of a contamination layer, and charge mobility change in the CCD Si layer as a function of the accumulated EUV dose. In this paper, the CCD quantum properties evolution is discussed, as well as the contamination issue. Preliminary diagnostics on the CCD aging under EUV radiations are exposed.

The x-ray multi-mirror mission is the second of the four cornerstone projects of the ESA long term Program for Space Science. The payload comprises three co-aligned high throughput imaging telescopes called mirror modules. The 'Centre Spatial de Liege' (CSL) is in charge of optical and environmental qualification test of each of these MMs. To perform optical tests, a vertical test facility (FOCAL X) has been developed by CSL. An EUV channel providing an 800 mm diameter collimated beam is used. A trade-off leading to the selection of an electron cyclotron resonance EUV source is presented. Impact of the coating on the microroughness of the EUV optics is assessed and its homogeneity across the optical surfaces is measured. Another feature of the facility is an x- ray channel providing a 50 (DOT) 8 mm² collimated beam. It has been characterized for x-ray effective area measurement. It is the first time that an off-axis parabolic mirror is used for this purpose.

The x-ray multi-mirror mission is one of the four 'cornerstone' projects in the ESA Long-Term Programme for Space Science. The image quality of two complete flight telescopes has been evaluated in a facility whose vertical optical axis minimizes deformation induced by gravity. The specimens illuminated by an EUV (58.4 nm) collimated beam allows the measurement of the point spread function, and the effective area across the field of view. Additionally, an x- ray pencil beam (1.5 and 8 keV) was used to measure the reflectivity on selected shells. The optical performance of the two first XMM flight telescopes was assessed during an environmental test campaign. The impact of the the thermal and vibration tests is presented and the performance of the two telescopes are compared.

Multilayer coatings for narrowband imaging in the extreme ultraviolet (EUV) have been designed and prepared. Multilayers were designed to optimize reflectivity at O⁺ 83.4 nm spectral line, and simultaneously to reject radiation at 121.6 nm Lyman (alpha) hydrogen line. Al/MgF₂/Mo multilayer coatings were prepared and high reflection/suppression ratios at the above wavelengths were measured. The coatings also exhibited a dip in reflectivity at 102.6 nm Lyman (beta) . The coatings showed some slow degradation over time, but maintained a high reflection/suppression ratio after a few months. Sample cleaning was found effective in restoring a very low reflectivity at 121.6 nm for samples aged of over 100 days.

The performances of a CCD have been evaluated in a very wide spectral region, which comprises near IR, visible, near and far UV, EUV and soft x-ray spectral regions. The CCD detector is a back illuminated one, 512 by 512 format, 24 by 24 micrometer squared pixel. The performed measurements consist mainly on the determination of the quantum efficiency in the 0.3 - 1100 nm (4.5 keV - 1 eV) spectral region. Three different experimental setup have been used: a Czerny-Turner monochromator for the 1100 - 250 nm region, a Johnson-Onaka monochromator for the 250 - 30 nm region, and a grazing incidence Rowland monochromator for the 25 - 0.3 nm region. The tested CCD exhibits high values of quantum efficiency in the analyzed spectral range, representing a very useful detector of radiation in the extended optical domain.

This paper describes the original concept and the optical design of the IMAGE mission FUV spectrographic imager (SI). The instrument goal is to spectrally separate and image the electron and proton auroras. A 30 angstrom (3 nm) spectral resolution is required to isolate the electron auroras (1356 angstrom). The proton aurora imaging requires to efficiently mask the geocoronal Lyman-alpha line (1216 angstrom), in order to image the Doppler shifted Lyman-alpha light (1217 - 1223 angstrom). A classical SI combines a telescope with a spectrometer. Our SI is consisting of a reverse combination: (1) a multi-slits Wadsworth monochromator designed to spectrally isolate the two bandwidths (electrons and protons auroras), (2) a two mirror imager with a crossed delay line detector producing the final imaging on each spectral channel.

The high pressure gas scintillation proportional counter, HPGSPC, was launched on April 30, 1996, as part of the narrow field instruments package of the satellite for x-ray astronomy BeppoSAX. Sensitive in the 4 - 100 keV band it fills the gap between the LECS, MECS and PDS instruments, making BeppoSAX a real wide band satellite. In the following paper we present performance and calibration results on in-flight energy response, spectral and timing capabilities, in-orbit background.

Reduction, subtraction and modeling of the in-flight background has been one of the main goal of the BeppoSAX high pressure gas scintillation proportional counter commissioning and science verification phase. In this paper we report studies of the observed in-orbit background.

The x-ray background spectroscopic survey (XBSS) is a SMEX mission proposed to perform a high spectral resolution all-sky survey of diffuse x-ray emission in the 50 - 2000 eV range. This spectral exploration of the x-ray background with high energy resolution will resolve important questions about the role of hot gas in the structure and evolution of the interstellar medium, the Galactic halo, and nearby intergalactic space that cannot be answered in any other way. The temperature distribution, emission measure, ionization distribution, and metallicity of the gases responsible for this emission are unknown. The survey is performed with a 6 by 6 array of cryogenic microcalorimeters that have spectral resolution of approximately 4 eV FWHM. The satellite is spin- stabilized with the spin axis directed toward the sun. The detectors look 90 degrees to the spin axis with a mechanically collimated field of view that is 5 degrees in radius. The instrument scans the entire sky twice in twelve months. During the second survey, deep exposures are performed along selected ecliptic meridians with the span axis fixed for up to 20 days at a time.

The transmission of thin indium filters was measured in the EUV wavelength range between 500 and 1200 angstroms. Our results for the shape of the transmission peak are consistent with previous measurements, indicating a FWHM bandpass between 750 and 900 angstrom with the maximum at 770 angstrom. The absolute transmission values however differ significantly from former measurements. The results are compared to currently used predictions. The filter transmission was remeasured after 20 months storage in N₂. On average, the transmission was observed to be reduced to about 60% of the original value.

We have developed absorptive Mg coatings for the vacuum UV (VUV) wavelength range. The total hemispherical reflectivity at normal incidence was measured at 121.6 nm. The reflectivity of the Mg coating produced by magnetron sputtering (presumably from the vapor phase) is close to 1%, comparable to the best known coatings. In contrast to the latter they are mechanically robust. After storage in air for 2 years the reflectivity increased by a factor 2. The reflectivity of Mg and Ag coatings produced by thermal evaporation was observed to be about 2 - 3%. The reflectivity of the Ag coatings proved to be stable over a period of 58 months in air.

We present the results of the measurement of transparency of five round beryllium windows for the LEIPC (low energy imaging proportional counter) of the stellar x-ray polarimeter (SXRP) experiment which will be flown on board the spectrum x-gamma Russian satellite. Each window was tested across its entire surface by using an x-ray fluorescence beam produced by a Cm²⁴⁴ alpha source. We mapped the physical properties of the whole set in order either to verify the performance of the manufacturing method and to select the window having the highest counting rate and the most homogeneous transparency. This is crucial in order to both enhance the scientific capability of the experiment and to reduce the impact of possible systematic effects due to pointing instability which could occur during the observation of celestial sources.

We propose a new astrophysics space mission for a low energy gamma-ray-burst observatory (LEGO) that will fit the envelope of a small-explorer (SMEX) type mission. The LEGO instrument combines silicon pixel detectors with ultra-high energy resolution and a novel cost effective fine-pitch coded mask, to image the sky with sub-arcminute accuracy in the 0.3 - 30 keV range with a wide field-of-view. LEGO is well adapted to study hundreds of short transients such as gamma-ray bursts and soft gamma repeaters in the unexplored energy range below 5 keV. LEGO takes one of the next logical steps in GRB studies in the post-BeppoSAX era by attacking the astrophysics questions raised by recent discoveries of variable radio, optical, and x-ray counterparts to burst sources. In addition to monitoring the sky for gamma-ray bursts, LEGO would provide a first all-sky monitor in the 0.3 - 30 keV range. LEGO will be sensitive to all mCrab sources in the sky in a day and to 0.1 mCrab sources in a year, and thus, may provide daily light curves and sensitive spectral measurements on about 10³ objects and yearly data on an order of magnitude more sources.