The High-Resolution X-ray Microcalorimeter Spectrometer System for the SXS on ASTRO-H

Transcription

1 The High-Resolution X-ray Microcalorimeter Spectrometer System for the SXS on ASTRO-H Kazuhisa Mitsuda a, Richard L. Kelley i, Kevin R. Boyce i, Gregory V. Brown l, Elisa Costantini m, Michael J. DiPirro i, Yuichiro Ezoe c, Ryuichi Fujimoto d, Keith C. Gendreau i, Jan-Willem den Herder m, Akio Hoshino d, Yoshitaka Ishisaki c, Caroline A. Kilbourne i, Shunji Kitamoto g, Dan McCammon j, Masahide Murakami e, Hiroshi Murakami g, Mina Ogawa a, Takaya Ohashi c, Atsushi Okamoto b, Stéphane Paltani n, Martin Pohl n F. Scott Porter i, Yoichi Sato b, Keisuke Shinozaki b, Peter J. Shirron i, Gary A. Sneiderman i, Hiroyuki Sugita b, Andrew Szymkowiak k, Yoh Takei a, Toru Tamagawa h, Makoto Tashiro f, Yukikatsu Terada f, Masahiro Tsujimoto a, Cor de Vries m, Hiroya Yamaguchi h, Noriko Y. Yamasaki a a Institute of Space and Astronautical Science, JAXA, Sagamihara, Japan; b Aerospace Research and Development Directorate, JAXA, Tsukuba, Japan; c Tokyo Metropolitan University, Hachioji, Japan; d Kanazwa University, Kanazawa, Japan; e Tsukuba University, Tsukuba Japan; f Saitama University, Saitama, Japan; g Rikkyo University, Tokyo, Japan; h Riken, Wako, Japan; i NASA/Goddard, Greenbelt, MD, USA; j Univ. of Wisconsin, Madison, WI, USA; k Yale Univ. New Haven, CT, USA; l Lawrence Livermore National Laboratory, Livermore, CA, USA; m SRON Netherlands Institute for Space Research, Utrecht, Netherlands; n Geneva University, Geneva, Switzerland ABSTRACT We present the science and an overview of the Soft X-ray Spectrometer onboard the ASTRO-H mission with emphasis on the detector system. The SXS consists of X-ray focusing mirrors and a microcalorimeter array and is developed by international collaboration lead by JAXA and NASA with European participation. The detector is a 6 6 format microcalorimeter array operated at a cryogenic temperature of 50 mk and covers a 3 3 filed of view of the X-ray telescope of 5.6 m focal length. We expect an energy resolution better than 7 ev (FWHM, requirement) with a goal of 4 ev. The effective area of the instrument will be 225 cm 2 at 7 kev; by a factor of about two larger than that of the X-ray microcalorimeter on board Suzaku. One of the main scientific objectives of the SXS is to investigate turbulent and/or macroscopic motions of hot gas in clusters of galaxies. Keywords: X-ray astronomy, Soft X-ray, High resolution X-ray Spectroscopy 1. INTRODUCTION The ASTRO-H mission, 1 6th in series of the ISAS/JAXA s X-ray astronomy program, is a combination of wide band X-ray spectroscopy (3-80 kev) provided by multi-layer coating, focusing hard X-ray mirrors (HXT) and pixel detectors (HXI), and high energy-resolution soft X-ray spectroscopy ( kev) provided by the Soft X-ray Spectrometer (SXS) consisting of thin-foil X-ray optics (SXS-SXT, also called as SXT-S) and the X-ray Microcalorimeter Spectrometer system (SXS-XCS ). The mission will carry other two instruments, a soft X-ray telescope + an X-ray CCD camera (SXT-I + SXI) and a soft gamma-ray detector (SGD). The SXS is a recovery mission of the XRS 2 onboard Suzaku 3 launched in It was the first orbiting x-ray microcalorimeter spectrometer and demonstrated an energy resolution of 6.7 ev at 5.9 kev (Mn Kα) in orbit and a low background rate of cs 1 cm 2 in the kev band. However, it could not observe astronomical sources because of unexpected short lifetime of Liquid He. The spectroscopic capability of X-ray Further author information: Send correspondence to K.M: mitsuda@astro.isas.jaxa.jp Sometimes the SXS-XCS is called simply the SXS.

2 Table 1. ASTRO-H SXS key requirments Requirement Goal Energy range kev Effective area at 1 kev 160 cm 2 Effective area at 6 kev 210 cm 2 Energy resolution 7 ev 4 ev Array format 6 6 Field of view Angula resolution 1.7 (HPD) 1.3 (HPD) Lifetime 3 years 5 years Time assignment resolution 80 µs Maximum counting rate 150 c s 1 pixel 1 Energy-scale calibration accuracy 2 ev 1 ev Line-spread-function calibration accuracy 2 ev 1 ev Table 2. ASTRO-H SXS key design parameters Parameter Value SXS XRT (SXT-S, Thin-foil mirrors) Focal length 5.6 m Diameter of most outer mirror 45 cm Reflecting surface Gold Thermal shield Al (300 nm) + PET (0.22 µm) with SUS mesh of a 94 % open fraction SXS XCS (6 6 microcalorimeter array) Operating temperature 50 mk Pixel size 814 µm 814µm Pixel pitch 832µm Field of view X-ray absorber HgTe, 8µm thickness Optical Blocking filters 5 filters, polyimide (460 nm) + Al (400nm) total, Si mesh on two filters See Okajima et al. (2008) 4 for more details of the mirror design. microcalorimeters is very unique in X-ray astronomy, since no other spectrometers can achieve high energy resolution, high quantum efficiency, and spectroscopy for spatially extended sources at the same time. The demand from the science community for spectroscopy with microcalorimeters is even increasing. The following three scientific objectives are the core science of the ASTRO-H SXS: (1) uncover entire pictures of galaxy-cluster thermal energy, kinetic energy of intracluster medium, and directly trace the dynamic evolutions of galaxy clusters, (2) measure the motion of matter governed by gravitational distortion at extreme proximity of black holes, and reveal the structure of relativistic space-time, and (3) map the distribution of dark matter in galaxy clusters and the total mass of galaxy clusters at different distances, and thereby study the roles of dark matter and dark energy in the evolution of galaxy clusters. The SXS system is being developed by international collaboration lead by ISAS/JAXA and NASA with European contribution. In this paper, we describe the science and the instruments of the ASTRO-H SXS with emphasis on the detector system. 2. SXS REQUIREMENTS AND KEY DESIGN PARAMETERS In Table 1, we summarize the key requirements of the SXS with goals for some of the parameters. In Table 2, key design parameters to fulfill the requirements/goals are shown. The photon collecting effective area is determined

3 1000 SXS Effective Area (cm 2 ) 100 RGS MEG HEG 10 LETG Energy (kev) Figure 1. Effective areas of high-resolution X-ray spectroscopy missions as functions of X-ray energy. The curve for the SXS is the present best estimate for a point source, where we assumed to sum all photons detected on the whole array, 1.3 HPD of the X-ray mirrors, and no contamination of the optical blocking filters. The two crosses show the requirements. The RGS effective areas is a sum of first order of the two instruments (RGS-1 and RGS-2), and was derived from the RGS response matrix in SAS v9.0. The effective areas of LETG, MEG and HEG onboard Chandra are, respectively, derived from the response files for the cycle 12 proposal, and are a sum of first order dispersions in ± directions. (Color on-line) by the various design parameters of X-ray mirrors, the X-ray absorber thickness of the detector, and the X-ray transmissions of thermal/optical blocking filters on the X-ray mirrors and inside the Dewar. It is also dependent on the point spread function of the mirror and the array size of the detector. In Figure 1, the effective area for a point source is plotted as a function of X-ray energy. The effective are of the SXS is larger than those of high resolution spectrometers on board Chandra and XMM-Newton at most energies plotted here. Above 2 kev, the difference is as large as a factor of five, and a factor of ten at 6 kev. In Figure 2, we show the resolving power (E/ E) as a function of energy for two cases, E = 7 ev (requirement) and 4 ev (goal). In the figure the resolving powers of not only high resolution spectrometers but also a typical X-ray CCD camera are shown. The SXS has highest resolving power above 2 kev. The broken lines show a resolving power required to resolve some spectral features of interest. The energy separation of resonant and inter-combination lines of K emission of He-like ions is shown with a broken line labeled with Res. I.C(Helike) ). The energy of the emission of each atom is indicated with the labels at the bottom of the panel. It is found that the SXS can resolve these fine structures of emission for most of ions from O to Fe. If non X-ray background and continuum emission is negligible compared to the line emission, the line-centroid energy can be determined to a statisitical accuracy of E/N, where N is the number of photons in the line. Accuracy of line-centroid determination is also limited by an uncertainty of energy scale calibration, which will be 1 ev (Table 1). In Figure 2, we show E/ E for doppler shift of 100 km s 1, the shift divided by a factor of 10, and E/ E for the calibration accuracy (dot-dash line denoted with 1eV). From these, we find that the SXS is able to detect doppler shift of emission energy down to a velocity of 100 km s 1 if we accumulate

4 kt=1 kev) Figure 2. Resolving power of the ASTRO-H SXS as a function of X-ray energy for the two cases, 4 ev resolution (goal) and 7 ev (requirment). The resolving power of high resolution instruments on board Chandra and XMM-newton and typical resolving power of X-ray CCD cameras are also shown for comparison. The typical energy separations between K emission of H-like and He-like ions ( H Helike ) and between resonant and inter combination lines of He-like ions ( Res. I.C(Helike) ) are shown with broken lines, while the emission energies are shown at the bottom of the panel. The broken line denoted with Ion thermal motion is the line broadening due to thermal motion of ions in a kt = 1 kev plasma. The broken lines indicated with 100 km/s and 100 km/s (100 photons) are, respectively, the doppler shift by a bulk motion of the velocity and a typical detection limit with 100 photons in photon-statistics limit, i.e.continuum emission and non X-ray background are negligible. The dot-dash line denoted with 1 ev shows the line shift/broadening detection limit determined by 1 ev energy-scale or line-spread-function calibration uncertainty. (Color on-line) more than 100 photons per line for He-like ions of Ar or heavier atoms. A similar argument is valid for line broadening where line-spread-function uncertainty limits the detectability instead of energy-scale uncertainty. From Figure 2 we find that thermal broadening of Fe-K line would not be detected for a 1 kev plasma (43 km s 1 ). However, if kt = 10 kev (130 km s 1 ), it should be possible to detect this with sufficiently high statistics. In Figure 3, we show a simulation spectrum for the central region of Centaurus cluster of galaxies. The SXS cannot resolve spatially the central structure of the intra cluster medium (ICM) of the cluster spatially (left top panel of the fitgure). However, it will resolve the fine structure of emission lines and thus macroscopic motions of the ICM down to a speed of a few 100 km s 1 by the doppler shift of the line center and/or the line broadening. 3. SXS SYSTEM In Figure 4, we show the block diagram of the ASTRO-H SXS X-ray calorimeter spectometer (XCS) system. The main responsibilities of international partners are indicated with national flags. On the left of the diagram is the cryogenic Dewar whose outer shell is connected to radiators with heat pipes in order to remove the heat generated by the mechanical coolers. The microcalorimeter array is mounted inside the detector assembly (DA) and cooled to 50 mk by the adiabatic demagnetization refrigerator (ADR). The ADR are pre-cooled by superfluid

5 Centaurus cluster Chandra image HPD 3 He-like Fe K Counts s -1 kev -1 6x6 SXS pixels O Fe-L Ne Mg Si Counts s -1 kev -1 S Ar CCD Response Ca Z = (3420 km/sec) Ni Figure 3. SXS pixel format and half power circle of the SXS X-ray telescope point spread function overlaid on the X-ray image of Centaurus cluster central region observed with Chandra 5 (top left) and simulated SXS energy spectra of the cluster. The spectrum plotted with a broken line is for a typical X-ray CCD resolution and is multiplied by a factor of 3 for display purpose. (Color on-line) liquid Helium. The 4 He Joule-Thomson (JT) cooler which is pre-cooled by two sets of double stage Stirling-Cycle coolers (2ST-PCa and 2ST-PCb), and two shield cooler (2ST-SCa and 2ST-SCb) provide thermal shields for the liquid He. The ADR and the cryo-cooelrs are controlled by four cooler driver electronics. Analog signal from the DA is amplified and digitized by the X-ray processing box (Xbox). The digitized data stream is sent to the Pulse Shape Processor (PSP), where X-ray events are detected and their pulse heights are determined by applying the optimum digital filtering algorithm. The PSP will also format the science and house keeping data into telemetry packets and send them to the spacecraft data-handling system. The PSP consists of the Mission Input/output (MIO) boards and Space Card boards, both of which are commonly used in other science instruments of ASTRO-H. Hardware logic constructed in a FPGA (Field Programmable Gate Array) on the MIO board will detect X-ray events from the digitized data stream, and the software running on a CPU on the Space Card board will perform optimum digital filtering and telemetry formatting. The Power Supply Unit (PSU) provides extremely low-noise regulated power to the Xbox. It will also provide a sync signal to the ADR control (ADRC) so that the switching pulses of all the DC-DC converters in the PSU and in the ADRC are synchronized with one another, which will reduce any beat frequency noises. The filter wheel Mechanism (FWM) is mounted at a distance of 90 cm from the detector on the lower panel of the fixed optical bench of the spacecraft. The FWM has six filter positions including open. The choice of filters are not determined yet, but at least we will have a Be filter to cut low-energy X-rays, a neutral density filter to reduce X-ray flux without modifying energy spectrum, and an optical blocking filter to observe optically bright objects. The Be filter will also be employed to protect the detector and optical blocking filters inside the Dewar

6 X-rays MD Valve drv DWR FWM MXS DA motor drv, cal source drv Aperture analog sig. heater X-ray processing Box (Xbox) LVDS Pulse Shape Processor (PSP1,2) FWE SpW SpW S/C BUS Radiators MSE HP s He vent status NEA valves (x4) NEA drv Loop HPs NEAC heater ADR LHe JT 2ST-PCa 2ST-PCb 2ST-SCa 2ST-SCb Heater Control Electronics (HCE) mag. cur. AC drv AC drv AC drv AC drv AC drv Gate valve, He ll, MS vent status S/C Bus ADR Control (ADRC) JT Driver (JTD) Pre Cooler Driver (PCD) Shield Cooler Driver (SCD) Sync Power Supply Unit (PSU) DIST Space Wire Router (SWR) A/B SpW SpW S/C BUS MD only on SWR-A MD S/C BUS Figure 4. Electrical block diagram of the ASTRO-H SXS. from micro-meteoroids while the SXS is not observing a source. On the FWM, we will mount commandable, modulated X-ray sources (MXS), which produce nearly monochromatic fluorescent X-rays as desired. The entire detector array will be illuminated by the MXS for time intervals as short as 10 ms. We can thereby calibrate the energy scale of the full array and at the same time screen the data in which the celestial X-ray spectrum is also exposed to the calibration sources, which produce electron- and photo- loss events. Both the FWM and MXS are controlled by the electronics, FWE. For more details of the filter wheel system and the MXS, please see de Vries et al. (2010). 6 The SpaceWire protocol is used for communication between the SXS electronics boxes and for communication with the spacecraft bus. The power distribution unit (DIST), one of the SpaceWire Router (SWR), and the Dewar are equipped with special GSE interfaces called MD (Manually-Disconnect) connectors. Using these interface we will be able to operate the ADR and mechanical coolers on ground without space craft power. We will use this function for the He top-off operation and to keep the Liquid He superfluid during the hold time before lift off. At that time, the spacecraft will be inside the nose fairing of the launch vehicle. We will access the MD connectors through an access hole of the nose fairing. The total mass of the SXS-XCS is estimated to be about 360 kg, and the total power consumed from the space craft primary power line is estimated to be about 580 W. A large amount of mass and power consumption are in the cooling system.

7 (a) (b) Figure 5. Layout of the SXS detector array on a Si frame (a), and Mn Kα emission spectrum obtained with a prototype SXS detector (b).7, 8 (Color on-line) 4. DETECTOR The calorimeter array for the SXS is being designed and developed with significant heritage from Suzaku XRS2. However, at the same time, significant improvements are also implemented. Although the array format is the same 6 6, pixels are larger and maintain a 3 x 3 FOV with the longer focal length of the telescope. Furthermore, by improvement in the X-ray absorber material and also by operating at a lower temperature, a better energy resolution is obtained. In Figure 5 (b), we show an example of Mn Kα emission spectrum obtained with a prototype SXS detector. The FWHM resolution is better than 4 ev. One of the key issue in obtaining a good energy resolution in orbit is heating of the Si detector frame by charged particle.2 In the SXS detector, a layer of gold (1.5 µm) will be deposited partially on the Si frame (yellow area in Figure 5 (a)) to obtain a better heat sinking. More details of the detector and the DA are described in Porter et al. (2010).7 5. CRYOGENICS SYSTEM A conceptual block diagram of the ASTRO-H SXS cooling chain is shown in Figure 6. A temperature of 50 mk is obtained by the double-stage adiabatic de-magnetization refrigerator (dadr) which consists of a high temperature stage of 150g of Gadolinium Lithium Floride (GLF) + 3 T (Tesla) superconducting magnet, a low temperature stage of 300 g of Chromium Potassium Alum (CPA) + 2 T superconducting magnet, and two sets of gas-gap active heat switches (GGHS). The dadr dumps the heat to the superfluid liquid He (LHe) at 1.2 K. The third ADR between the 4 He JT cooler and LHe tank is nominally off and the two heat switches (GGHS) on both sides of the third ADR are also off. Thereby, the 4 He JT cooler provides a 4.5 K thermal shield to the He tank. The full volume of the He tank is 40". After the He top-off operation, ground hold before lift off, and initial cool down in orbit, the amount of initial LHe is expected to be > 30". We thus assume 30" for our base of calculation. Then the LHe lifetime is expected to be 4.5 years in the nominal case (Table 3). The third ADR can be operated to remove heat from the He tank. If we fully operate it in orbit, the LHe lifetime will be extended to 6.9 years. The third ADR will be also operated during ground hold before launch. This will also extend the lifetime of LHe by increasing the initial amount of LHe in orbit. This case has not been studied in detail yet. In nominal case, we can cool the detector to 50 mk and continue observations even after all LHe has exhausted. In this cryogen-free operation, the observation efficiency become lower than with LHe, because the hold time of the ADR until next regeneration would be shorter. In Table 3, we also show the LHe lifetime for several failure cases of a single mechanical coolers or of unexpected loss of LHe. In the nominal case, all cyrocoolers are operated at about half maximum power. In

8 HP s LHP s Radiators 2ST 2ST 2ST 4 He JT 2ST LHe Cryostat ADR (GLF) 150g, 3T ADR (GLF) 150g, 3T ADR (CPA) 300g, 2T Figure 6. Block diagram of the ASTRO-H SXS cooling chain. contingency cases, we will increase the cyrocooler drive power to compensate for the loss of cooling power and the heat load through the non-operational cooler. However, if one of the precoolers fails, the remaining precooler cannot pre-cool the He gas in the JT cooler to a low enough temperature to obtain the JT effect. Thus, in this case as well as in JT contingency case, cryogen-free operation is not possible. However, the temperature of the thermal shield, which the JT cooler is supposed to cool, can be kept low enough to preserve the LHe for 2 years. In Figure 7, the mechanical layouts of the SXS Dewar are shown. The mass of the Dewar is about 270 kg, while the total mass of the four cooler driver boxes are estimated to be 40 kg. In nominal case, the four cooler drivers consumes 480 W of power from the space craft primary power. This includes not only the cooler drive power but also power conversion efficiency and power used for house keeping and communication. For more details of the cooling system, see Fujimoto et al. (2010) 9 and Shirron et al. (2010). 10 ACKNOWLEDGMENTS We are grateful to the engineers in Sumitomo Heavy Industries (SHI), in NASA/GSFC, in Mitsubishi Heavy Industrories (MHI), in SRON, and in Geneva University, and scientists and engineers in JAXA and univrsiteis for the design of the SXS microcalorimeter system, and graduate students who contributed laboratory experiments and design; especially K. Kanao, S. Yoshida, M. Miyaoka, K. Narasaki, S. Tsunematsu, T. Hiroishi, A. Okabayashi, K. Ootsuka (SHI), J. Lander, J. Crow, A. Mattern, A. Kashani, R. Thorpe, D. McGuinness, N.Galassi, P. Chen, P. Arsenovic, J. Miko, T. Bialas, K. Mil, M. Chiao, K. Kawaja, I. Linares, C. Masters, T. Sullivan (NASA/GSFC), K. Masukawa, K. Matsuda, Y. Kuroda (MHI), H. Aarts, M. Frericks, P. Laubert, P. Lowes (SRON), D. Haas, and R. Dubosson (Geneva), M. Nomachi (Osaka University), H. Ogawa, K. Minesugi,

9 Detector Assembly Shield cooler (2ST) dadr Aperture Assembly He tank Dewar HK connectors He ll/ vent lines Detector and ADR connectors Shield cooler (2ST) Support struts Pre coolers (2STx2) JT cooler Cooling fans for ground operation JT compressors Figure 7. External view (left) and cutaway drawing (right) of the ASTRO-H SXS cryogenic Dewar. The gate valve which will be mounted on the top of the Dewar outer shell is not shown. Two shield coolers are mounted on the side of the outer shell, while the JT cooler and its pre coolers are mounted on the bottom. The detector assembly (DA) and the ADR are installed as a unit on the He tank from the upper side. The third ADR is on the side of the DA but is not shown here because it is in the unseen part of the Dewar in the cutaway drawing. The outer diameter of the Dewar outer shell excluding mechanical coolers and other sticking-out structures is 950 mm. The height of the Dewar is 1292 mm including the support struts but excluding the gate valve. The total mass of the Dewar is about 270 kg.

10 Table 3. Expected LHe lifetime and observation lilfetime Case Cooler powers a LHe lifetime (y) Observation (y) Nominal JT=90W PC=50W 2 SC=50W b > 4.5 c > 6.9 bc 1 SC contingency JT=90W PC=50W 2 SC=90W 1 JT contingency JT=0W PC=50W 2 SC=90W 2 1 PC contingency JT=0W PC=90W 1 SC=90W > 2.6 c Unexpected He loss JT=90W PC=50W 2 SC=90W 2 0 > 0 c a Power supplied to the cooler. The conversion efficiency of the cooler driver electronics is not include. If we operate the 3rd ADR in nominal case. Observation can be continued without LHe using the 3rd ADR as long as the remaining mechanical coolers are working. K. Ishimura, N. Iwata, T. Irikado, I. Funak, N. Yamanishi (ISAS/JAXA), K. Ishikawa, Y. Abe (Tokyo Metropolitan University), H. Seta, Y. Shimoda, T. Yasuda, M. Asahina (Saitama University), T. Hagihara, I. Mitsuishi, H. Yositake (ISAS/JAXA), and T. Yuasa (University of Tokyo). We would also like to thank members of Akari and SPICA teams, in particular, T. Nakagawa, and the SMILES group at ISAS/JAXA for the development of mechanical coolers. REFERENCES [1] Takahashi, T. and et al., The ASTRO-H mission, Proc. SPIE 7732, (2010). [2] Kelley, R. and et al., The Suzaku High Resolution X-Ray Spectometer, Publ. Astron. Soc. Japan 51, S77 (2007). [3] Mitsuda, K. and et al., X-ray Observatory Suzaku, Publ. Astron. Soc. Japan 51, S1 (2007). [4] Okajima, T. and et al., Soft x-ray mirrors onboard the NeXT satellite, Proc. SPIE 7011, 70112X 70112X 10 (2008). [5] Fabian, A. C. and et al., A deep Chandra observation of the Centaurus cluster: bubbles, filaments and edges, Mon. Not. of Roy, Ast. Soc. 360, L20 (arxiv:astro ph/ ) (2005). [6] de Vries, C. and et al., Filters and calibration sources for the soft x-ray spectrometer (SXS) instrument on ASTRO-H, Proc. SPIE 7732, (2010). [7] Porter, F. S. and et al., The detector subsystem for the SXS instrument on the ASTRO-H Observatory, Proc. SPIE 7732, (2010). [8] Porter, F. S. and et al., The Astro-H Soft X-ray Spectrometer (SXS), AIP conference series CP1185, 91 (2009). [9] Fujimoto, R. and et al., Cooling system for the soft x-ray spectrometer (SXS) onboard ASTRO-H, Proc. SPIE 7732, (2010). [10] Shirron, P. J. and et al., Design of a 3-stage ADR for the soft x-ray spectrometer instrument on the ASTRO-H Mission, Proc. SPIE 7732, (2010).