Starting in October 1987 the eight flight mirror shells of the ROSAT high-resolution X-ray mirror system have been assembled to a Wolter I telescope. A description of the assembly procedure and the alignment control measurement methods is given as well as a summary of the performance. I. Introduction II. Configuration and design III. Mirror manufacture IV. Assembly and adjustment V. Summary Acknowledgments

The European Space Agency ESA is currently planning its next cornerstone mission of the Horizon 2000 space science programme, which is the high throughput x-ray astronomy mission XMM. The heart of the scientific payload consists of a set of 3 highly nested Wolter type I telescopes. A review of the optical design and layout is given. The grazing incidence mirrors will be made by epoxy replication on CFRP from a highly polished glass mandrel. This technique has been used to produce flat mirror samples, of which x-ray measurements of reflectivity and scattering are presented.

LAXRIS (large area x-ray imaging spectrometer) is an experimental, balloon-borne, hard x-ray telescope that consists of a coaligned array of x-ray imaging spectrometer modules capable of obtaining high angular resolution (1-3 arcminutes) with moderate energy resolution in the 20- to 300-keV region. Each spectrometer module consists of a CsI(Na) crystal coupled to a position-sensitive phototube with a crossed-wire, resisitive readout. Imaging is provided by a coded aperture mask with a 4-m focal length. The high angular resolution is coupled with rather large area (-800 cm') to provide good sensitivity. Results are presented on performance and overall design. Sensitivity estimates are derived from a Monte-Carlo code developed to model the LAXRIS response to background encountered at balloon altitudes. We discuss a variety of observations made feasible by high angular resolution. For instance, spatially resolving the nonthermal x-ray emission from clusters of galaxies is suggested as an ideal program for LAXRIS.

The new AS&E Ultrahigh Resolution Soft X-Ray Solar Research Rocket Payload has been successfully flown twice on Black Brant IX Sounding Rockets from White Sands Missile Range. These flights, conducted on 15 August 1987 and 11 December 1987, provided the first test of the new payload which consists of 3.8X magnifying hyperboloid-hyperboloid grazing incidence relay optic used in conjunction with an existing Wolter-I primary mirror. An RCA SID 500 series CCD detector was utilized in a thinned, back-illuminated configuration for recording the images. The 5.4 m effective focal length of the compound optics system resulted in a plate scale of 1 arc second per pixel which is comparable to the inherent resolution of the primary mirror. These flights represent the first use in X-ray astronomy of either of these two new technologies. These observations are presented with comparison to laboratory measurements and theoretical expectations of the instrument performance.

The detection of point-like celestial sources of hard X-radiation in the presence of background could be improved by many orders of magnitude by using a telescope to concentrate the incoming X-ray flux. Unfortunately, conventional grazing incidence X-ray optics become impractical at energies greater than about 10 keV. We have developed a design concept for a new class of hard X-ray telescope using synthetic multilayer structures as X-ray reflecting elements. A proper choice of multilayer spacing as a function of position allows the telescope to be tuned to a specific X-ray wavelength. Our design is the multilayer analog of a popular, economical, grazing incidence X-ray telescope design. It has the advantage that the multilayers are deposited on flat surfaces of manageable size, unlike previous designs which required the deposition of multilayers on large concave surfaces. We have considered the design of the required multilayer structures, the design of the telescope mirror, and the design of a telescope tuned to 100 keV and sized for a shuttle launch.

We discuss the flight of a multilayer-coated mirror on a sounding rocket experiment on March 17, 1988, which was used to obtain photometric images of the Sun just prior to solar eclipse. This 1.5" superpolished mirror was coated with 17 layer-pairs of Mo/Si at the Lockheed Palo Alto Research Laboratory, and achieved a peak reflectivity of 33 percent at a wavelength of 171 A; the mirror passband was 13 A FWHM. The spherical mirror, having a focal length of 0.75 m, formed an image of the Sun at 2.3 degrees off-axis directly onto a 128 x 128 microchannel plate CODACON detector. This detector and the companion EUV spectrometer experiment were built at the University of Colorado. Eight full-disk images of the Sun were obtained by this XUV imager. Our aim was to obtain absolutely-calibrated photometry of solar structures, rather than the higher resolution photography we could have achieved with film. Images obtained were of excellent quality and are being analyzed together with coordinated ground-based, eclipse coronagraph observations and radio maps taken at the VLA and Greenbank.

The High Resolution Camera (HRC) [12] is one of four instruments selected for detailed design definition as a focal plane detector for an upcoming NASA mission - the Advanced X-Ray Astrophysics Facility (AXAF) [2], which is scheduled for launch at the end of calendar year 1996. The camera is a microchannel plate (MCP) based detector utilizing a CsI photocathode to enhance X-ray quantum efficiency in the 0.1 to 10.0 keV energy band. An electronic readout provides better than 25pm FWHM spatial resolution over a 100mm x 100mm active area. The HRC is closely related to the High Resolution Imaging Detector (HRI) which was successfully flown on the Einstein (HEAO-2) Observatory [3] from 1978 to 1981. A similar detector has been built for the ROSAT Observatory [4] which is scheduled to be launched in February of 1990. The major differences between the HRC and its predecessors are in overall size (20 times greater active area), quantum efficiency (3 to 4 times higher), count rate limitation ( 2 to 5 times greater count rate), background (3 to 5 times lower), and energy resolution below 2 keV (some compared to none). While ,the basic x-ray detector is similar to previous instruments, the electronic readout and processing chain of the HRC incorporate major im-

Television sensors such as CCD's and vidicons can be coupled to convertors and (typically) image intensifiers to obtain active areas, high flux capabilities, quantum efficiency, high time resolution, or ease of construction and operation that may not be obtained with a directly illuminated sensor in the X-ray and XUV range. A general purpose system which makes use of these capabilities for a number of applications is described. Some of the capabilities and properties of this type of system are discussed, as are some of the considerations that should be kept in mind when configuring a system of this type.

Recent developments of scientific CCDs have produced sensors that achieve ultra low read noise performance (less than 2 electrons rms) and near perfect charge transfer efficiency (0.9999996) without the addition of a fat-zero. This progress has now made it possible to achieve Fano-noise-limited performance in the soft x-ray where the detector's energy resolution is primarily limited by the statistical variation in the charge generated by the interacting x-ray photon. In this paper, Fano-noise-limited test data is presented for two different CCD types and a CCD derived estimate of the Fano factor is determined. By evaluating ultra low-modulation images (less than 1 electron peak-to-peak) it is shown that the CCD's global CTE is now superior to its read noise floor. To capitalize on this capability CCD manufacturers are now focusing their attention on reducing the noise floor below the 1 electron level thereby matching the sensor's CTE performance. This improvement, if accomplished, will push Fano-noise-limited performance for the CCD into the extreme ultra-violet.

We have built a linear CCD with enhanced quantum efficiency to x-rays. The primary objective is to replace an existing scanning pinhole camera used to evaluate x-rays optics that we are developing. The scanning pinhole camera can take up to 3 hours to acquire a single scan of data. Several scans may be required in order to evaluate an optics part. By using a linear CCD, an entire scan of data can be taken at the same time, which would reduce the data collection time for an optics part to a few minutes.

A number of current developments in microchannel plate (MCP) manufacturing technology promise significant advances in detector performance both for X-ray astronomy and for other fields. These include the production of: (i) MCPs made from radioisotope-free ('low-noe') lead glasses (ii) low resistance MCPs for high count rate applications (iii) large area (-100 cm ) channel plates and (iv) MCPs with small (2-6 micron) channel diameters. In this paper, we present measurements made on prototype low noise, low resistance no MA.) MCPs fabricated by Galileo Electro-Optics. These MCPs exhibit a limiting dark noise count rate ten times lower than2 that of the same manufacturer's standard plates. Contributions to the residual count rate (-0.07 cm s ) from cosmic ray interactions, field emission, trace radioisotope contamination and pressure effects are discussed. Thermal modelling of the new Galileo MCPs is correlated with their observed noise and count rate characteristics. Results of an accelerated lifetest are described. Incidental noise images are presented from the large area detector developed at Leicester to evaluate MCPs for the AXAF High Resolution Camera (HRC). Calculations of induced radioactivity in satellite-borne MCP detectors and their associated photocathodes are described. A theoretical model is used to predict the gain and temporal response of MCPs with reduced pore sizes.

Microchannel plate (MCP) stacks have proven very useful for imaging detectors in X-ray and UV astronomy 1,2. However, the back to back stack configuration of flat MCP's which we often use is not optimal for some applications. Firstly, there are variations in the performance (gain and pulse height distribution) of these stacks due to warpage of the individual MCP's. Secondly, for grazing incidence optics with a curved focal plane, a curved detector is often desirable. We have studied two MCP developments, bonded MCP stacks, and stacks of MCP's with curved surfaces to determine if we can achieve the MCP requirements for future astrophysical detectors. Our experiments show that both configurations studied give superior MCP performance characteristics. However, there seem to be some problems with the fabrication of bonded MCP stacks resulting in poor flat field characteristics and increased background.

The requirements for CCDs which may be used in future X-ray astronomical missions are discussed. Results from a programme designed to enhance commercially available CCDs for X-ray spectroscopic applications are presented. The use of deep depletion technology allows a combination of good X-ray detection efficiency (> 50% at 6 keV) and X-ray energy resolution (e, 100eV FWHM). The effects on the scientific performance of a radiation dose of as little as 14 krads are shown to be potentially serious. Further developments in the topics of back illumination and large area arrays are described.

The Advanced X-ray Astrophysics Facility CCD Imaging Spectrometer (ACIS) will use two arrays of CCD's to provide X-ray imaging and spectroscopy. Spectroscopy at medium resolution with the imaging array is accomplished by pulse height analysis of each X-ray interaction, while for high spectral resolution, an objective grating disperses the spectrum across a linear array of CCD's to provide a dispersed spectrum where the wavelength resolution is determined by the telescope imaging properties. Performance data for CCD's manufactured by Texas Instruments, MIT Lincoln Laboratory and Ford Aerospace Corp. will be presented. Plans for future CCD enhancements will be discussed.

Fully depletable pn CCD's on high resistivity 280 pm thick silicon (e e:-, 2.5 kIlcm) have been fabricated. Their operation is based on the semiconductor drift chamber principle proposed by Gatti and Rehak. They are designed as energy and position sensitive radiation detectors for X-rays and (minimum-) ionizing particles. Two-dimensional semiconductor device modeling demonstrates the basic charge transfer mechanisms. Prototypes of the detectors have been tested in quasistatic and dynamic conditions. A preliminary charge transfer inefficiency e, has been determined to less than 10-3 at shift frequencies varying between 500 kHz and 6 MHz. The charge loss during the transfer is discussed. As a consequence, an improved design had been developed for a new fabrication iteration which is currently being tested.

An imaging gas scintillation proportional counter (GSPC) having a 70 mm diameter thin polypropylene entrance window has been built. The detector is of the so called driftless type. In this concept the X-rays are absorbed directly in the scintillation region whereas in conventional GSPCs a separate drift region with a lower electric field is used. The advantages of a driftless GSPC are its improved low energy response and higher background rejection efficiency. Position sensitive read-out of the scintillation light is performed by means of a 5" Hamamatsu photomultiplier with mesh dynodes and crossed-wire anode. In this article we present energy and position resolution measurements obtained with this detector at several X-ray energies. For example values of 16.0 % and 1.9 mm full width at half maximum (FWHM) energy and position resolutions have been measured at an energy of 1487 eV. Some present limitations will be discussed and an outline of future developments will be given.

We report measurements supporting the feasibility of exploiting primary scintillation in xenon as fast trigger for deconvolving off penetration depth effects of energetic X-rays (E > 20 keV). The application of the technique to high pressure position sensitive gas scintillation proportional counters when used in conjunction with wide field (30° x 30°) coded masks is discussed.

Three different types of proportional gas detectors are under development at the Danish Space Research Institute for use in X-ray astronomy. A conventional multiwire proportional counter, a novel microstrip detector and a parallel gap counter with a segmented anode readout. Instead of wires, the micro strip counter employs very narrowly spaced conducting strips deposited on an isolating substrate. Excellent energy resolutions of 13% (5.9 keV) and 27% (1.5 keV) have been achieved. Several possible methods for position readout for this detector are reported. The parallel gap detec-tor is equipped with a wedge and strip electrode as the anode and has been operated with gas amplifications up to 105. The signals induced in the segmented anode allow for accurate two-dimensional position determination.

The complex behavior of compact galactic X-ray sources complicates their study. Since these objects are highly variable, and exhibit a multiplicity of "states", it is often impossible to anticipate when a desired observation can be performed. "Transient" objects appear without warning and may be detectable for only a short period of time. These difficulties can be reduced by an instrument which can monitor a large number of objects simultaneously, or nearly so. The function of an "All Sky Monitor" is to record a photometric history of known objects, and to detect and locate transient objects. The classic techniques of this field utilize the pinhole camera and the scanned slat collimators. For both of these approaches, there is an unfortunate tradeoff between angular resolution and "photon starvation": a high resolution instrument will capture very few photons. The multiple pinhole approach avoids this problem, but designs based on the Uniformly Redundant Array (URA) principle" have generally required sophisticated two dimen-sional imaging detectors, complicated auxiliary collimation, and a large telemetry bandwidth. The All Sky Monitor (ASM) for the X-ray Timing Explorer (XTE) avoids these problems. It uses the multiple pinhole approach to avoid photon starvation for good photometric capability together with good angular resolution and an excellent capability for locating transients. While the XTE ASM doesn't have the ideal imaging qualities of URA based designs, it is simpler to build and requires a modest telemetry bandwidth; since the ASM is intended for detection, location, and photometry of discrete point sources, imaging is of secondary importance.

The confluence of microelectronics developments, Space Shuttle and expendable launch vehicle maturation and emergence of miniature satellite technology (MST) is providing a new resource for astronomy research. A small, low cost satellite can provide significant on orbit data collection and transmission capability on a budget and time scale accessible to a diverse community of researchers. but like any other complex research tool careful application of MST is critical to obtaining good results. Small satellite capabilities in electric power, pointing, overall size and weight, orbit adjustment, downlink data rate and other parameters are limited compared with today's large scale systems (though in many cases they are comparable to or exceed large systems of a decade or two ago). However, the convenience and low cost of typical MST ground facilities provide enhanced resource accessibility compared to conventional approaches. Proliferation of ground stations and their operation at the research institution or observation site by research team members are features easily integrated within the MST approach. To introduce the potential of MST, missions are treated which most easily adapt to the small satellite environment. Design to operational requirements and close integration of payload and bus are fundamental to the MST approach. A summary of typical MST design capabilities and tradeoffs illustrates the current state of the art of MST application in astronomy. Construction of the small satellite is moot without attention to launch opportunities and costs. Here too properly conceived implementation of MST is important in securing the earliest, lowest cost, most dependable launch opportunity. the launch vehicle flexibility inherent to small satellites can be maximized by engineering compatibility to the largest number of potential launch vehicles. The capabilities, costs and availability of several launch options are assayed. With the current relative scarcity of launch opportunities MST payloads are increasingly appealing compared with conventional, larger systems which demand a larger share of the available orbit insertion resource.

Los Alamos and Sandia National Laboratories are, building an ultrasoft X-ray monitor experiment. This experiment, called ALEXIS (Array of Low-Energy X-Ray Imaging Sensors), consists of six compact normal-incidence telescopes. ALEXIS will operate in the 70 - 110 eV band. The ultrasoft X-ray/EUV band is nearly uncharted territory for astrophysics. ALEXIS, with its wide fields-of-view and well-defined wavelength bands, will complement the upcoming NASA Extreme Ultraviolet Explorer and ROSAT EUV Wide Field Camera, which are sensitive broadzband survey experiments. The program objectives of ALEXIS are to 1) demonstrate the feasibility of a wide field-of-view, normal incidence ultrasoft X-ray telescope system and 2) to determine ultrasoft X-ray backgrounds in the space environment. As a dividend, ALEXIS will pursue the following scientific objectives: 1) to map the diffuse background, with unprecedented angular resolution, in several emission line bands, 2) to perform a 3-color survey of point sources, 3) to search for transient phenomena in the ultrasoft X-ray band, and 4) to provide synoptic monitoring of variable ultrasoft X-ray sources such as cataclysmic variables and flare stars. The six ALEXIS telescopes are arranged in pairs to cover three 40° fields-of-view. During each spin of the satellite, ALEXIS will monitor more than half the sky. Each telescope consists of a layered synthetic microstructure (LSM) mirror, a curved microchannel plate detector, background-rejecting filters and magnets, and readout electronics. The mirrors will be tuned to 72 eV, 85 eV, and 95 or 107 eV bands, chosen to select and deselect interesting line features in the diffuse background. The geometric area of each ALEXIS telescope will be about 15 cm2. The telescopes employ spherical mirrors with the curved detector at prime focus and are limited by spherical aberration to a resolution of about 1°. Assuming nominal reflectivities, quantum efficiency, and filter transmission, the 5a survey sensitivity will be about 2 x 10-3 photons cm-2 s-1 for line emission at the center of the bandpass. ALEXIS is designed to be flown on a small autonomous payload carrier (a minisat) that could be launched from either a Shuttle Get-Away-Special Can or from an expendable launch vehicle. The experiment weighs 100 pounds, draws 40 watts, and produces 10 kbps of data. It can be flown in any low Earth orbit. Onboard data storage allows operation and tracking from a single ground station at Los Alamos.

We review the use of transmission gratings with grazing-incidence telescopes in celestial X-ray astronomy. Gratings were used on the Einstein Observatory and on EXOSAT, and they are planned for AXAF and SPECTROSAT. We outline the basic properties of transmission grating spectrometers and the use of "phased" gratings to enhance the diffraction efficiency. Gratings are fabricated in a multistep process involving generation of a mask followed by replication into a final grating of the appropriate thickness and material. Special attention is given to the AXAF High Energy Transmission Grating (HETG) being fabricated at MIT. The HETG operates over the range 0.4-8 keV, gives resolving powers of 100-1000, effective areas of 10-300 cm2, and a minimum detectable line flux of 1-10 x 10-6 photons cm -2 s'1. The instrument consists of a single array with two types of grating facets: medium-energy gratings (0.6 pm-period, 0.5 pm-thick silver) mounted behind the outer three AXAF mirrors, and high-energy gratings (0.2 pm-period, 1.0 pm-thick gold) mounted behind the inner three mirrors. The materials and thicknesses are selected to maximize efficiency throughout the energy band. The facets are fabricated at MIT using a process involving X-ray lithography. AXAF will also carry a Low Energy Transmission Grating (LETG) supplied by the Laboratory for Space Research at Utrecht. It uses self-supported grating facets of 1.0 gm period and is optimized for operation down to 80 eV. Both the HETG and LETG disperse the spectrum of a source across either of the AXAF focal-plane/imaging detectors. They are most effective for the study of point sources, but they also give moderate resolution spectra of slightly extended sources and can be used to map the spatial distribution of the line emitting regions.

The use of microcalorimeters for high-resolution, high quantum efficiency, non-dispersive X-ray spectroscopy has been demonstrated over the past few years. During this time the energy resolution in the 1-10 keV band has improved from a value of-140 eV (FWHM), comparable to the best solid-state (Si/Li) detectors, to a value of-17 eV. We are presently working on improving the energy resolution by designing devices that have been optimized with respect to heat capacity, thermal conductance, X-ray absorbing material, thermistor geometry and thermistor implant concentration. In addition we have been working on the design of an integrated spectrometer consisting of a He4 cryostat containing an adiabatic demagnetization refrigerator, non-X-ray blocking filters and control and signal processing electronics. The essential components for such a system have been assembled at the University of Wisconsin and a similar system is being developed at Goddard. These systems will demonstrate the feasibility of this approach for space flight use, in particular for the NASA AXAF mission, and serve as research tools for further work in X-ray calorimetry. In this paper we give an overview of X-ray calorimetry, present the results of on-going X-ray tests, and discuss an approach for building an X-ray calorimeter spectrometer.

The high throughput X-ray astrophysics mission is the second cornerstone in ESA's long term space science programme. The long duration X-ray observatory consists of three heavily nested X-ray imaging telescopes coupled to X-ray CCD cameras and reflection gratings which provide a high throughput facility for cosmic X-ray spectroscopy. Details of the spacecraft, mirror modules and possible instrument designs are described. The mission is due for launch in 1998 with an anticipated lifetime of over ten years. This observatory will make possible major advances in X-ray astrophysics at the turn of the century.

Bragg crystal spectroscopy has been used to study solar X-ray emission since the early 1960's. Studies of non-solar celestial X-ray emitters, however, require special techniques in order to achieve the requisite sensitivity. We review some of the Bragg spectrometers (all of which use concentration to reduce the geometric area of the detector) which have been proposed in the last 15 years, with special emphasis on two classes of spectrometers which are used in combination with grazing incidence telescopes: objective crystal spectrometers and focal plane crystal spectrometers. A special case of the latter class are focal plane spectrometers which operate in the Johann mode (the crystal is curved so as to intercept the diverging X-ray beam at a nearly constant Bragg angle, and refocuses the X-ray beam onto an imaging detector). The Bragg Crystal Spectrometer (BCS) for the Advanced X-ray Astrophysics Facility is such a device. We review the techniques which have been developed at MIT and by our colleagues at the Goddard Space Flight Center (Hakim et al. this volume) for achieving high spectral resolving powers (E/AE) for the BCS diffractors. Two techniques which we have studied, and which we find to be effective for decreasing AE, are a) the use of an exponential spiral crystal shape and b) placement of the imaging detector in an out-of-focus mode. Both of these minimize the line-broadening caused by the Johann geometry. The resolving power of the BCS exceeds 2000 at some energies; DES 1 eV over a large range of energies. The BCS achieves sensitivity adequate for the study of X-ray spectra for z 1000 cosmic X-ray sources.

Results are presented which detail the performance of proportional counters when filled with a wide variety of argon- and xenon-based gas mixtures. Parameters measured include the energy resolution, gas gain, and factors affecting the useful lifetimes of the mixtures.

The use of transmission gratings as dispersive elements in X-ray telescopes with an angular resolution in the arcsec range is preferable to the use of reflection gratings, as the transmission grating can be more easily incorporated into an existing telescope design. SPEKTROSAT, the follow-up mission to ROSAT will be equipped with a transmission grating spectrometer having a spectral resolution of 0.2 A over the wavelength range between 6 A and 200 A. Third order aberrations are minimized, since the gratings are mounted according to the Rowland-torus geometry. The ROSAT mirror system will be slightly changed in order to meet the spectroscopic requirements. The investigation of grating distortion effects which would reduce the theoretically achievable resolution has been started. Extensive efficiency measurements on different grating materials have been conducted.

We examine the performance of high spatial frequency "phased" X-ray transmission gratings developed for the High Energy Transmission Grating Spectrometer on the Advanced X-ray Astrophysics Facility (AXAF). The gratings tested here nominally consist of 1 thick gold lines of 0.2 μm period covering approximately 5 cm 2 of a polyimide membrane. A table-top setup at MIT employs the gratings in reflection to diffract UV (325 nm) laser light. It is used to measure grating periods and indicates that period variations within and between gratings are a few parts in 104. Tests performed at the Marshall Space Flight Center 304 m X-ray Facility using 1.5 keV (Al Ka) X-rays in transmission corroborate the UV measurements and demonstrate geometrically-limited resolving powers of E/ΔE-750. Finally, X-ray transmission tests performed in the MIT 25 m X-ray facility provide measurements of period, line thickness, space-to-period ratio, tilt of grating lines, and efficiency. We find that the agreement between the design parameters and the measured parameters is good.

A series of prototype blazed reflection gratings, designed for incorporation in a satellite-borne, high efficiency, moderate resolution, astronomical x-ray spectrometer, have been tested for x-ray reflectivity in the relevant spectral orders. Both mechanically-ruled and ion-etched holographic master and replica grat-ings produced by various manufacturers, have been measured at 8.34 A, 13.34 A, and 44.7 A. We find near theoretical performance from a particular ion-etched sample and from one of the mechanically ruled samples. The other mechanically ruled samples exhibit lower efficiency, which can in part be ascribed to imperfections in the groove profile. A comparision between scalar diffraction theory and the rigorous electromagnetic calculations of grating efficiency for these samples is also presented.

The purpose of this paper is to survey some of the recent advances in the state of the art of soft X-ray spectrometers, particularly as they might be applied to Solar Observations. The discussions will center on the windowless region from roughly 1 to 100 A, and will cover both grating and crystal instruments. During the interval that has elapsed since the launches of the Solar Maximum Mission and P 78-1 satellites in the early 1980's, there have been a number of significant technological developments. Among these, are multi-layer mirrors, large format CCD detectors sensitive to X-rays, position sensitive photon counting detectors, new kinds of X-ray films, and the discovery of a number of ingenious new optical systems based on gratings with non-uniform ruling spacings. There have also been a number of improvements in the extent and accuracy of the atomic physics data sets on which the analysis of spectroscopic observations depend. Finally, the analysis of the data from the two missions mentioned above, from the earlier Skylab mission, and from recent sounding rocket flights, have raised a number of interesting questions that will greatly benefit from a new generation of instruments and the observations that they will produce. I will begin with a short discussion of the solar soft X-ray spectrum and its interpretation, followed by a few general comments on problems peculiar to soft X-ray instruments. The main portion of the paper will be devoted to a review of recent developments in spectrometer optical design, which has been a lively field during the last dozen years. This is particularly true in the case of grating spectrometers, because of the development of diffraction gratings with variable groove spacings or with curved rulings. The availability of multi-layer mirror coatings means that designs are not necessarily restricted to the glancing incidence region, giving additional design flexibility. The paper will conclude with a short section on telescope considerations, and some hopeful remarks on future flight opportunities.

We discuss the possibilities opened up by the newly-available soft x-ray and XUV multilayer coated optics for observations of the Solar outer atmosphere and corona. Presently available material combinations and the achievable quality of coatings provide basic limitations to the possible spectral bands which can be observed. The spectral regions within which Solar emission lines are formed are compared with the available multilayer passbands in a discussion of the types of instruments which we can consider building. In addition, we discuss briefly the possible spectral lines which are of interest for studying the various temperature regimes and variability time constants which are known to be present in the Solar atmosphere.

We report on the first high resolution images of the sun in the soft x-ray/extreme ultraviolet (XUV) regime obtained with normal incidence multilayer optics. The images were obtained during a sounding rocket flight on October 23, 1987 from White Sands Missile Range, New Mexico. Multilayer coated optics were used in three different configurations: as optical elements in two Cassegrain Telescopes, as off-axis primary mirrors, and as tertiary mirrors operating with a Wolter Schwarzschild grazing incidence mirror. The inherent energy selective property of multilayer coated optics allowed distinct groups of emission lines to be isolated in the solar corona and the transition region. Images were recorded at 173 A and 256 A with the Cassegrain Telescopes, at 256 A with a Herschelian Telescope, and at 44 A, 173 A and 256 A with the Wolter Schwarzschild hybrid telescope. In addition, soft x-ray images in the 8 - 18 A bandpass were obtained at the prime focus of the Wolter Schwarzschild optics. The images show many features of the solar corona and transition region, including magnetically confined loops of hot solar plasma, coronal plumes, polar coronal holes, supergranulation, and features associated with overlying cool prominences.

The Berlin electron storage ring BESSY can be used as a primary source standard to es-tablish a spectral emission scale. The uncertainty is about 0.23 % in the near infrared and increases to about 2 % in the soft X-ray region at 5 keV. At the radiometric laboratory of PTB at BESSY special beamlines have been built to calibrate transfer source standards utilizing BESSY as a primary standard. Presently laser produced plasmas and high current hollow cathode sources are investigated in respect of their use as transfer source standards. To establish a spectral responsivity scale in the soft X-ray region Si(Li)-detectors are calibrated using undispersed synchrotron radiation. Solid state diodes are calibrated as transfer detector standards by comparison with Si(Li)-detectors. The characterization of optical components can be performed using a reflectometer operating in the 5 nm to 34 nm range.

We describe the vacuum calibration facilities located at the Space Sciences Laboratory, University of California at Berkeley, which are designed for the calibration and testing of extreme and far ultraviolet space-borne instrumentation in the spectral range 44 to 2500 A. The facility includes one large cylindrical vacuum chamber (3 m x 5 m) containing two EUV collimators, and it is equipped with a 4-axis manipulator of angular control resolution of 1 arcsec for payloads weighing up to 500 kg. In addition, two smaller cylindrical chambers, each 0.9 m x 1.2 m, are available for vacuum and thermal testing of UV detectors, filters, and space electronics hardware. All three chambers open into Class 10,000 clean rooms, and all calibrations are referred to NBS secondary standards. Users of the facility to date include the EUVE, FAUST, and UVX Shuttle payload projects.

The Optical Research Section at the Goddard Space Flight Center is establishing the diffraction grating evaluation facility (DGEF) to evaluate the performance of new technology diffraction gratings for future space flight instrumentation. Performance is most accurately evaluated with the grating in the optical configuration for which it is designed. For this reason the DGEF has been devised to evaluate gratings in an optical breadboard scheme that emulates design concepts or design options. The working volume of the DGEF was sized to accommodate proposed spectrographic designs for future missions, e.g. Lyman or the Far Ultraviolet Spectroscopic Explorer (FUSE)1, and second generation Space Telescope instruments such as the Space Telescope Imaging Spectrograph (STIS)2. Space was also allocated to set up supporting equipment, including stimulus sources and diagnostic systems. The spectral range which we propose to eventually cover extends from the extreme ultraviolet into the infrared. The facility, designed and built by Baker Manufacturing Co. in Evansville, Wisconsin, consists of a large, vibration isolated granite slab, 3m wide, 6.7m long, and 0.6m thick, with an aluminum surface plate 7.6cm thick for the optical set-up surface and vacuum sealing3. Vacuum capability is provided by a stainless steel vacuum cover and collar that will provide a working height of roughly 1.2m. The vacuum chamber will be pumped with four cryopumps to a pressure of one microtorr or less. Electrical, cooling water, and cryogenic interfaces to devices inside the chamber will be provided through the vacuum collar. Internal motions for mirrors, detectors, etc. will be implemented by means of computer controlled precision optical stages with stepper motors and encoded read-outs. The granite slab, aluminum surface plate, and vibration isolation system, have been delivered and installed on a temporary basis until construction of the permanent laboratory is completed in late 1988. Test results for several STIS new technology gratings, including an ion etched cross disperser are presented.

Spectrometer calibrations have been performed at the National Bureau of Standards, recently renamed the National Institute of Standards and Technology, for over 10 years using the calculable synchrotron radiation from the SURF-II electron storage ring. SURF is now a high performance storage ring which can operate at 300 MeV with over 250 mA of stored current. One beam line is dedicated for use as a calibration facility for outside users.

The Extreme Ultraviolet Explorer (EUVE) has a spectrometer which utilizes variable line-spaced, plane diffraction gratings in the converging beam of a Wolter-Schwarzschild type II mirror. The gratings, microchannel plate detector, and thin film filters have been calibrated with continuum radiation provided by the NBS SURF II facility. These were calibrated in a continuum beam to find edges or other sharp spectral features in the transmission of the filters, quantum efficiency of the microchannel plate detector, and efficiency of the gratings. The details of the calibration procedure and the results of the calibration are presented.

An x-ray calibration facility for use in the 0.2-25 keV region is described. The facility employs several types of specially modified sources and detectors to produce and detect both line and continuum radiation in this energy range. We describe an inexpensive commercial x-ray source which has been modified for efficient high intensity operation as well as production of x-rays up to 25 keV. We also describe a system that utilizes multilayer mirrors alone or in a double Bragg geometry to select an energy bandpass. This system is controlled by a microcomputer which translates and rotates the multilayers to provide an easily selectable monochromatic beam with good resolu-tion over a broad energy range. A long focal length Kirkpatrick-Baez mirror pair has been coupled to a pivoting beam line in order to accurately characterize gratings for use in soft x-ray astronomy. The beam line is scanned through the various grating orders. All aspects of the facility incorporate a high degree of flexibility so that a wide variety of calibrations can be easily performed.

Toroidal gratings can be used for imaging spectrometers. These can provide stigmatic images and are very useful for space applications. Two cases are described: an Ultraviolet Coronagraph for the solar SOHO mission and a high resolution spectrometer for the stellar Lyman mission. The toroidal gratings have been produced by replicating an elastically deformed spherical one. By properly applying the distorting forces also coma-corrected surfaces can be achieved.

The Advanced X-ray Astrophysics Facility (AXAF) will be a 15-year lifetime observatory, intended to investigate the nature of celestial objects and physical processes, and to study the history and evolution of the universe. We are investigating how to translate the requirements for scientific measurements (e.g., fluxes, spectra and sizes of sources) into operational requirements for the calibration of the AXAF X-ray mirrors and scientific instruments. We carry out a complete simulation of the process of acquiring data from an X-ray source, and analyze the data using various calibration errors. We generate an ideal spectrum of an X-ray source and convolve it with hypothetical "true" mirror and instrument responses to simulate an expected spectrum. This spectrum may be randomized, independently, a large number of times to simulate the effects of photon statistics. These data are analyzed as if they were actual flight data. We introduce controlled perturbations in the mirror and instrument responses used in the analysis. Such perturbations may include: the neglect of extended X-ray absorption fine structure (EXAFS), errors in the jump of absorption coefficients across atomic shell energies, changes in the optical constants of mirror coatings, effects of dust contamination, or arbitrary calibration measurement errors. We perform x2 fits, and assess the changes in the fitted results from those resulting when the data are analyzed with the "true" mirror and instrument responses. In this paper we discuss an initial study of exponential energy spectra with galactic absorption. We assume perfect detector response and consider only the effects of errors in the on-axis effective area of the mirror.

Two Dall-Kirkham, normal incidence collimators have been designed to calibrate the imaging properties of the Extreme Ultraviolet Explorer over the wavelength region from 114 to 2000 A. The mirrors of the short-wavelength, 25-cm diameter collimator are superpolished Zerodur which have been multilayer coated for optimal reflectivity at 114 A. The mirrors of the long-wavelength, 41.25-cm diameter collimator are gold coated Zerodur for high reflectance above 300 A. The design, performance, and future use of these collimators in the extreme ultra-violet is discussed.

A three-crystal laboratory X-ray spectrometer is used to measure the Bragg reflection from concave cylindrically curved crystals to be used in the High Resolution X-ray Spectrometer in NASA's Advanced X-ray Astrophysics Facility (AXAF). The first two crystals, in the dispersive (1,1) arrangement, select a narrow collimated monochromatic beam in the Cu Ka 1 line at 1.54A (8.1 keV), which illuminates the test crystal. The angular centroids of rocking curves measured along the surface provide a measure of the conformity of the crystal to the desired radius of curvature. Individual and combined rocking curve widths and areas provide a measure of the resolution and efficiency at 1.54A. The crystals analyzed included LW (200), PET, and acid phthalates such as TAP.

The prime focal instrument of the ROSAT X-ray telescope, the position sensitive proportio-nal counter (PSPC), has been calibrated extensively with X-rays in the energy range from 0.28 to 1.7 keV. We developed a method to measure the differential spatial non-linearities caused by the digitizing effects of the anode and cathode grids, and the bulging of the thin entrance window, and also the position dependent gain of the counter. From these measurements we derived tables, which are used to correct each single recorded event in position as well as in pulse height. We will present the calibration method and the performance achieved for the PSPC.

We present X-ray measurements at FeKa (6.4 keV) which range over 3-4 orders of magnitude of reflected intensity from four metal coated samples (Au, Ir, Pt and Ni). For one Au sample, the total reflectivity was also measured at MoKa (17.5 keV). The results indicate that a model consisting of a bulk region with a top layer having a variation from bulk density to zero is a good description of the coating. They also indicate that the coating process rather than the material is crucial in achieving high reflectivity.