Test campaign of mirror shells for eROSITA, to be launched in 2011, has been performed at the PANTER X-ray
test facility. The results of the campaign are adequately described in our companion paper. In this paper, we
focus on the detailed analysis of intra- and extrafocal images to investigate possible outcome of the out-of-focus
measurements in the framework of the eROSITA mirror test programme. The images taken at several out-offocus
positions are ring-shaped, that provide crucial keys to the understanding of the optics performance. With
those out-of-focus rings, we can easily examine where the deformation occurs on the shell. Conceivable error
causes of the image degradation are examined quantitatively. Properties of the in-focus image can be predicted
by an interpolation of the out-of-focus rings, a kind of "Hartmann test". Performing Hartmann test, we can
estimate how large portion of the image degradation can be attributed to the out-of-roundness error which is
mainly due to fluctuations in taper angles. The reconstructed image using Hartmann test appears similar to
the core of actual in-focus image. We also pursue the possibility to evaluate the contributions of slope errors
using the width of the radial profiles of the out-of-focus rings. These diagnostic techniques with the out-of-focus
measurements should be useful for future measurements, not only for the eROSITA mirror but also for other
missions (e.g. XEUS optics), where the images are only available at out-of-focus positions in PANTER facility
(i.e. impossible to take in-focus images) because of the significantly long focal lengths.
The PANTER X-ray Test Facility has been utilized successfully for developing and calibrating X-ray astronomical
instrumentation for observatories such as ROSAT, Chandra, XMM-Newton, Swift, etc.
Future missions like eROSITA, SIMBOL-X, or XEUS require improved spatial resolution and broader
energy band pass, both for optics and for cameras. Calibration campaigns at PANTER have made use of flight
spare instrumentation for space applications; here we report on a new dedicated CCD camera for on-ground
calibration, called TRoPIC. As the CCD is similar to ones used for eROSITA (pn-type, back-illuminated, 75 μm
pixel size, frame store mode, 450 μm micron wafer thickness, etc.) it can serve as prototype for eROSITA camera
New techniques enable and enhance the analysis of measurements of eROSITA shells or silicon pore optics.
Specifically, we show how sub-pixel resolution can be utilized to improve spatial resolution and subsequently the
characterization of of mirror shell quality and of point spread function parameters in particular, also relevant
for position reconstruction of astronomical sources in orbit.
We present X-ray characteristics of X-ray telescopes (XRTs) onboard the Astro-E2 satellite. It is scheduled to be launched in February 2005. We have been performed X-ray characterization measurements of XRTs at Institute of Space and Astronautical Science (ISAS) since January 2003. We adopted a raster scan method with a narrow X-ray pencil beam. Angular resolution of the Quadrants composed of the Astro-E2 XRT was evaluated to be 1'.6-2'.2 (HPD; Half Power Diameter), irrespective of the X-ray energy, while those of the Astro-E XRT was 2'.0-2'.2. The effective area of a telescope is approximately 450, 330, 250, and 170 [cm2] at energies of 1.49, 4.51, 8.04, and 9.44 keV, respectively. The field of view (FOV) of the XRTs which is defined as Full Width Half Maximum (FWHM) of the vignetting function is ≈18' at 4.51 keV. We summarize these characters of the XRTs.
We report a ground-based X-ray calibration of the Astro-E2 X-ray
telescope at the PANTER test facility. Astro-E2, to be launched in
February 2005, has five X-Ray Telescopes (XRTs). Four of them focus on
the X-Ray Imaging Spectrometers (XIS) while the other on the X-Ray
Spectrometer (XRS). They are designed with a conical approximation of
Wolter-I type optics, nested with thin foil mirrors to enhance their
throughput. A calibration test of the first Astro-E2 flight XRT for
XIS was carried out at the PANTER facility in August 2003. This
facility has an 130 meter long diverging beam from X-ray generator to
XRT. Owing to the small X-ray spot size of about 2 mm dia., we verified that the focal position of each quadrant unit converged within 10 arcsec. The energy band around Au-M edge structures was
scanned with a graphite crystal. The edge energy (Au M5) is consistent with that listed in Henke et al. 1997. Owing to the large area coverage of the PSPC detector which is a spare of the ROSAT satellite, off-axis images including stray lights at large off-axis angle (up to 6 degree) were obtained with a large field of view. We also compared the results with those measured with the parallel pencil beam at ISAS which is in detail reported in our companion paper by Itoh A. et al..
X-Ray Spectrometer (XRS) is the microcalorimeter onboard the X-ray astronomy satellite Astro-E2 which is scheduled to be launched
early in 2005. For the XRS to achieve its best energy resolution
of 6 eV at 6 keV, X-ray intensity should be limited up to several c s-1 pixel-1. The filter wheel (FW) is the instrument to reduce incident X-ray intensity on the XRS using extinction filters. The FW consists of a stepping motor, extinction filters, and a filter disk which has six mounting positions for the extinction filters. Among the six mounting points, two are used for Neutral Density (ND) filters, another two are for Beryllium (Be) filters, and the other two are remained open. The biggest modification from Astro-E is that we attach radioisotopes of
55Fe and 41Ca on the filter disk, which illuminate the XRS pixels to monitor the gain in orbit. We present here the mechanical design of the FW especially on improvements from Astro-E, and the results of our calibration measurements on X-ray transmission of the extinction filters.
Astro-E2, to be launched in early 2005, will carry five X-ray Telescopes (XRT). The design of the XRT is the same as the previous original mission Astro-E, that is a conical approximation of Wolter Type-I optics, where about 170 thin-foil reflectors are nested confocally. Some modifications from Astro-E are adopted within the severe constraints due to the policy of "re-build" instruments. One of the major changes is the addition of pre-collimators for the stray light protection. Several modifications on the fabrication processes are also made. The replication glass mandrels are screened carefully, which is expected to reduce the figure error of replicated reflectors. We thus expect better performance than Astro-E especially in imaging capability. In order to qualify the performance of the Astro-E2 XRT, we have started ground calibration program of XRT at 30 meter X-ray beam facility of the Institute of Space and Astronautical Science (ISAS). We have found positive improvements on the telescope performance from the Astro-E, which probably arise from the applied modifications. The on-axis half-power diameter (HPD) has been evaluated to be 1.6-1.7 arcmin, which is improved from the Astro-E (2.0 ~ 2.1 arcmin HPD). The on-axis effective areas of quadrants are larger than the average of Astro-E by about 5%. The on-axis effective areas of the XRT for X-ray Imaging Spectrometers (XIS) are approximately 460, 340, 260, and 190 cm2 at energies of 1.49, 4.51, 8.04, and 9.44 keV, respectively. The present paper describes the recent results of
the performance of the first flight assembly of the Astro-E2 XRT.
Astro-E2 XRTs adopt Wolter Type-I optics and have nested thin foil structure to enhance their throughput. But this structure allows stray X-rays to come from the sky outside of the XRT field of view. Stray lights contaminate focal plane images, especially in the case of extended source observations. We intend to mount pre-collimators on top of the ASTRO-E2 XRTs to intercept stray lights. According to the success for the engineering model pre-collimator to protect the stray lights efficiently, we proceeded to product flight model pre-collimators. Some improements are made for the flight model (FM) pre-collimator: the introduction of heat forming to make slats accurate cylindrical shape, the change of the groove shape of alignment plates and the change of the housing design. We also established the method of pre-collimator mounting. In X-ray measurements, stray light images and the flux of each stray component at any off-axis angles are measured with/without FM pre-collimator. The secondary only reflection component is reduced down to 3% at a larger off-axis angle than 30', and the backside reflection component becomes more remarkable. On the other hand, X-ray measurement of the effective area at on-axis with/without FM pre-collimator verifies that pre-collimator does not interfere the telescope aperture. In addition, the decrease of XRT field of view is ≤8%, which is the same as the ray-tracing simulations.
ASTRO-E2 XRTs adopt Wolter Type-I optics and have nested thin
foil structure to enhance their throughput. But this structure allows
stray X-rays to come from the sky out of the XRT field of view. Stray light contaminates focal plane images, especially in the case of extended source observations. We intend to mount pre-collimators on top of the ASTRO-E2 XRTs to intercept stray light. On the other hand, reflection by the pre-collimator itself newly creates secondary stray light. To decrease these additional stray light as possible, the mil finish aluminium with its roller mark normal to the incident X-ray beam has been used for the slat material, whose reflectivity is reduced down to 1/20 of ideal specular reflection. Optical profilers tell us these samples have very rough surfaces, whose height varies with Σ; = 1-2 μm. According to the design parameters as are described in the related paper in this symposium (Paper I), an engineering model pre-collimator is fabricated with 46 slats out of 175. Before EM pre-collimator is mounted onto XRT, alignment plates are adjusted to align slats to the same position of XRT primary reflectors. In x-ray measurements, stray light images and the flux
of each stray light component at 30' off axis are measured
with/without EM pre-collimator. The secondary only reflection component is reduced down to 3.6%, and the backside reflection component becomes more remarkable. On the other hand, X-ray measurement of the effective area at on axis with/without EM pre-collimator verifies that pre-collimator does not interfere the telescope aperture. In addition, the decrease of XRT field of view is ~10%, which is the same as the ray-tracing simulations. As a whole, EM pre-collimator reduces stray light to 27% level with only ~10% decrease of the XRT F.O.V.
Normal incidence optics have been used with multilayers in EUV
region. The 2d of the multilayers has to be equal to the wavelength of
interest. At the same time, the reflectivity of the multilayers should
decrease with the increase of the interfacial roughness much faster
than grazing optics. In general, 2d of 10 nm is the shortest d-spacing
available for multilayers in normal incidence. As a challenge of shorter wavelength application, we made NiCr/C multiplayer mirror for the laboratory use at 4.47 nm(carbon K alpha line). The main dish of the Cassegrain optics is 20 cm diameter spherical mirror and the secondary mirror is a reflector in aspherical shape to correct astigmatisms. Its focal point is placed at the X-ray source to create a broad parallel beam of 20 cm in diameter. The flux of the parallel beam is slightly less than the expected value, and gradually decreases of 40% toward the outer region. The measured parallelism is about 25 arcsec, which is a little larger than the designed value. More pricise positioning of the focal point to the X-ray generator may reduce such divergence. The beam profile through a slit shows a core of about 20 arcsec and an extended tail which might be due to scattering tail by the roughness of 0.3 nm. An application of this system is demonstrated with the Astro-E X-ray telescope. The image core is sharper but the scattering tail is considerable. Even after the subtraction of the tail, still some wing is left. This system is bright and parallel enough to examine the optical alignment much faster than previous method, while careful measurements are necessary for quantitative calibration of X-ray telescopes.
X-ray telescopes (XRTs) of nested thin foil mirrors were developed for
Astro-E, the fifth Japanese x-ray astronomy satellite. Although the
launch was not successful, the re-flight of Astro-E mission is approved as Astro-E2 and will carry the same XRTs. Ground-based calibration of Astro-E XRT revealed that its image quality and effective area are somewhat worse than what are expected from the original design. Conceivable causes of these defects of the XRT performance (i.e., surface roughness, waviness, misalignment of
reflectors, and so on) are examined by X-rays and optical microscopic
measurements. In this paper, we distinguish quantitatively these
causes to limit the performance of the Astro-E XRT. Using the detail
measurements, we can attribute both degradation of the image quality
and a deficit of the effective area from the design values mainly to a
slope error with a mm scale in each reflector and shadowing effects of
neighboring reflectors due to various factors. There is still room
for improvement in the support system of reflectors (i.e., alignment
bars) in the XRT. One of the main aims of the mirror system calibration is to construct response function. Therefore, it is important that the development of a representative numerical model and its validation against extensive ground-based calibration. Taking account of the results of the pre-flight calibration and the microscopic measurements, we develop and tune a ray-tracing simulator which constructs the XRT response function for a point source at an arbitrary off-axis angle and spatial distributions of celestial X-ray sources.
Next Japanese ASTRO-E2 satellite carries five X-ray telescopes (XRTs),
and the pre-collimators for the stray light protection will be
installed on them. The pre-collimator is composed of the vertical
foil cylinders which line up at the top of the primary reflectors of
the XRT. This configuration is effective to reject the stray light of
the 'secondary only' reflection, which is the most of the stray light
from large off-axis angles up to 70' to the focal plane detector
within 18' x 18'. If the height of the pre-collimator is 30 mm (15%
of the total height of XRT), we can protect all the "secondary only"
component at > 30' off-axis. The field of view will become only about
10% less due to the collimation effect of the 30 mm high
pre-collimator. However, new stray light component is generated by the
reflection and scattering by the pre-collimator itself, especially at
small off-axis angles. As a result, we estimate the total flux of the
stray light to be about 10% at 30' off-axis and 5% at 60' off-axis,
compared with the case without pre-collimators.
We report a new calibration system for large size X-ray optics at
ISAS. We adapted a 'dynamical' pencil beam collimated from an X-ray
generator, the maximum voltage for which is 50 kV. By combining two
stage systems for the X-ray generator and a collimator, the pencil
beam dynamically sweeps across a circular region of a telescope with
the radius of 60 cm at maximum. In this case, the X-ray telescope and
the focal plane detector are both statically fixed. A 4.4~m long rail
for detector stage and two positions of the telescope stage provide
focal lengths from 4.5 to 12 m, while the previous system can
accommodate 4.5 or 4.75 m focal length. The preliminary performance of
this system is summarized in this paper. For the post-Astro-EII
satellite, a hard X-ray multi-layer supermirror with an unprecedented
sensitivity up to 80~keV is strongly expected. This beam facility is of importance because the hard X-ray mirrors always require a long focal length of 8-12 m due to the small reflection angle (about 0.3 degree). Focal length and diameter of future telescopes are always decided by the boundary conditions of the mission at the last moment of the design freeze. Our new X-ray beam facility is designed to match with any kind of X-ray telescope parameters.
The development of hard X-ray focusing optics is widely recognized as
one of key technologies for future X-ray observatory missions such as
NeXT(Japan), Constellation-X(US) and possibly XEUS(Europe). We have developed hard X-ray telescope employing depth-graded multilayers, so-called supermirrors. Its benefit is to reflect hard X-rays by Bragg reflection at incidence angles larger than the critical angle of total external reflection. We are now continuously fabricating platinum-carbon(Pt/C) supermirror reflectors for hard X-ray observations. In this paper we focus on our development of the
hard X-ray telescope for the first balloon flight observation
(InFOCuS) and its results. InFOCuS is an international balloon-borne hard X-ray observation experiment initiated by NASA/GSFC. InFOCuS hard X-ray telescope have been jointly developed by Nagoya University and GSFC. The telescope is conical approximation of Wolter-I optics with 8m focal length and 40cm diameter. It consists of 255 nested ultra-thin reflector pairs with incidence angles of 0.10 to 0.36deg. Reflectors are coated with Pt/C supermirrors with periodic length of 2.9 to 10nm and bi-layer number of 25 to 60, depending on incidence angles. The effective area and imaging quality are expected as 100 cm2 at 30 keV and 2 arcmin in half power diameter, respectively. The InFOCuS experiment was launched on July 5, 2001, from National Scientific Balloon Facility in Texas, USA. We successfully observed Cyg X-1, chosen for a calibration target, in 20-40keV energy band. We are planning to carry out next flight for scientific observations as soon as additional telescopes, detectors, and upgraded gondola system are implemented.
X-ray characterization measurements of capillary (X-ray Optical Systems Inc) were carried out at ISAS (Japan) X-ray beam facility. Since capillary system has a capability of collecting x-rays effectively by the small angle reflections in the narrow tubes, it is expected to apply it for the observation of astronomical x-ray objects. The depth of focus on the on-axis is 6 mm, and the image size at the focal plane is 0.4 mm (FWHM) with Cu-K. By using continuum x-ray (2 to 15 keV) beam, the on-axis efficiency was evaluated to be 20 %. The efficiency decreases gradually toward low energy range, this could be explained by the absorption effect of Si which is one of the constituent element of capillary tubes. The field of view is defined as the off-axis angle at which the efficiency becomes half of the on-axis value. The diameter of the field of view was 22', 19' at 4.51 keV and 8.04 keV, respectively. Capillary has no imaging capability, in other words, the light-concentrating direction is independent from the incident angle of x-ray beam. When the x-ray incident angle varies from -15' to +15', the focused images are distributed within 3'. From our measurements, we could confirm the performance of capillary and its potential for astronomical applications.
We have been developing the high throughput hard X-ray telescope, using reflectors coated with the depth graded multilayer known as supermirror, which is considered to be a key technology for future satellite hard X-ray imaging missions. InFOC(mu) $S, the International Focusing Optics Collaboration for (mu) -Crab Sensitivity is the project of the balloon observation of a cosmic hard X-ray source with this type of hard X-ray telescope and CdZnTe pixel detector as a focal plane imager. For the fist InFOC(mu) S balloon experiment, we developed the hard X-ray telescope with outermost diameter of 40cm, focal length of 8m and energy band pass of 20-40 keV, for which Pt/C multilayer was used. From the pre-flight X-ray calibration, we confirmed its energy band and imaging capability of 2 arcmin HPD and 10 arcmin FOV of FWHM, and a effective area of 50 cm2 for 20-40 keV X-ray. We report the current status of our balloon borne experiment and performance of our hard X-ray telescope.
Mass production of replicated thin aluminum x-ray reflecting foils for the InFOC(mu) S (International Focusing Optics Collaboration for Micro-Crab Sensitivity) balloon payload is complete, and the full mirror has been assembled. InFOC(mu) S is an 8-meter focal length hard x-ray telescope scheduled for first launch in July 2001 and will be the first instrument to focus and image x-rays at high energies (20-40 keV) using multilayer-based reflectors. The individual reflecting elements are replicated thin aluminum foils, in a conical approximation Wolter-I system similar to those built for ASCA and ASTRO-E. These previous imaging systems achieved half-power-diameters of 3.5 and 1.7-2.1 arcminutes respectively. The InFOC(mu) S mirror is expected to have angular resolution similar to the ASTRO-E mirror. The reflecting foils for InFOC(mu) S, however, utilize a vertically graded Pt/C multilayer to provide broad-band high-energy focusing. We present the results of our pre-flight characterization of the full mirror, including imaging and sensitivity evaluations. If possible, we will include imaging results from the first flight of a multilayer-based high-energy focusing telescope.