Adaptive X-Ray Optics with a Deformable Mirror Shunji Kitamoto*,a,b, Norimasa Yamamotob, Takayoshi Kohmurab.c, Kazuharu Sugaa.b , Hiroyuki Sekiguchia,b, ,Jun’ichi Satoa, Keisuke Sudoa, Takeshi Watanabea, Youhei Ohkuboa, Akiko Sekiguchia and Masahiro Tsujimotoa 　 Rikkyo University, 3-34-1, Nishi-Ikebukuro, Toshima-ku Tokyo, 171-8501, Japan

Abstract We started development of an ultra high precision X-ray Telescope using adaptive X-ray optics, named X-ray milli-arc-sec Project　(X-mas Project). I will report our current activity of this project.

OUTLINE Introduction Telescope Design Components Conclusion Primary Mirror Deformable Mirror Wave Front Sensor Back side CCD Optical Blocking Filter X-ray Source Conclusion

1. Introduction X-ray Astronomy Satellite “Chandara” was launched in July 1999 and it has ~0.5 arc-sec resolution. Chandra is providing us wonderful X-ray images and we are enjoying lots of important scientific results. However, the current achievement of the image quality is still far from the theoretical diffraction limit!

1.Introduction Telescopes plotted on the wave length-diameter plane If we have a diffraction limit X-ray telescope with 1 m diameter, the resolution will be less than 1 milli-arc-sec.

What is the problem? Requirement of Small-scale Roughness : several Å Requirement of Large-scale Figure Error: ～1nm Easy very difficult We are trying to overcome this difficulty by applying two ideas (1) optical monitoring of the optics (2) adaptive optics system with a deformable mirror

Some Technical Consideration A normal incident telescope is easier than the grazing incident telescope, in order to have a large effective area. Possible precision of the shape measurement is a few nm. 13.5nm band is currently best choice. Mo/Si Multi-Layers have more than 70% reflectivity for the normal incident mirror.

Test Configuration All components are in the vacuum chamber and on a stable table.

Primary Mirror Off Axis Paraboloid 80mm D=80mm f=2000mm The diffraction limit is ~41 milli-arc-sec 80mm Coating with Mo/Si Multi-layers 30-50% reflectivity Shinsedai Kakou System　＆ X-ray Company

Secondary Mirror Deformable Mirror (DFM) 31 elements-Bimorph piezo-electric plates two layer piezos with the opposite polarity. Each plate can make a curvature of concave or convex shape. 55mmφ effective diameter CILAS Bimorph piezo-electric plates

Flat plane made by DFM Using Zygo Interferometer 5.26nm rms flat plane had been achieved. Some examples of the deformation

Mo/Si Multi-Layer Coating on DFM Surface Roughness　 　　　rms 0.321nm Figure Error 5.9 nm rms (after remove tilt and sphere) reflectivity 65% for 13.5nm

Optical Image of Current Telescope Optical Image of a anode-cap shined by a filament Image of the Anode Cap 1mm 9mm ~0.1mm at 4350mm away from the primary mirror. ~5 arc-sec (Diffraction limit for 500nm ~1.5 arc sec) No adaptive Optics No tuning wrong optics

Wave Front Sensor HASO32 (Imagine Optic) Shack-Hartmann Sensor consist of a Micro-lens array and CCD cartoon of HASO32

Precision of the Wave Front Sensor Spherical wave is constructed by a pin hole with 1 m m diameter. Remove the tilt and spherical component form obtained wave shape. Calculate the residuals and rms variation We confirm the precision of Less than 3nm rms

Wave front Control with Optical Light 263nm 520nm closed loop control 0.060 mm (rms) 121nm 187nm Confirm the closed loop control all biases are 0 V 0.143 mm, (rms)

X-ray Detector Back-illumination CCD (HPK) 30% detection efficiency at 13.5nm 512x512 pixels 24 mm square Expected image size is 0.3 mm We have to study a sub-pixel read-out method and/or we need a X-ray detector with finer position resolution. Image of 55Fe X-rays

Optical Blocking Filter 2 x 150nm Zr Optical transmission ~ 10-9 13.5nm Transmission ~0.25 Transmission of 100nm Zr 0.01 1 0.0001 Wave Length(nm) 10 100 1000 135nm Otical UV blocked by SiO2 from Luxel

Optical Blocking Filter Measurement of the X-ray transmission at KEK-PF. ~45% transmission at 13.5nm Transmission of two filters is ~20%

X-ray Source Manson Ultrasoft X-ray Source (Model-2) Anode Cap; Al/Si alloy Si 16.4% 13.55nm Si L transition Home-made monochrromator Confirm 13.55nm X-rays by the measurement of the reflectivity of a know Mo/Si Multilayer (2d=26nm)

Conclusion All the components are almost prepared. Optical closed loop system has been demonstrated. X-ray source is now ready. However, current precision is far from the goal Next step fine tuning/alignment of the components the X-ray imaging with adaptive optics.

Plan Propose the X-mas satellite mission in futur Fukue et al. Challenge Direct imaging of the Black Holes. Plan Propose the X-mas satellite mission in futur Try to challenge the shorter wave length and larger diameter

Thank you for attention.

X-ray Optical Separation Filter Zr filter has a good transmission Zr 150nm On the donuts-shape frame with a few A surface roughness and 5nm rms figure error. 20mm X-ray Company ＆ Luxel

Testing the precision of the wave front sensor Installed in a clean booth covered by a Black Curtain Laser source with pin hole Wave Front Sensor Imagine Optic

X-ray landing position and its event pattern Ｘ線入射位置とそのイベントパターンはどう対応しているのだろうか？ 1 pixel シングルイベントを作るＸ線入射位置の分布。 シングルイベント 全電荷が一画素に集められたイベント。 3×3 pixels

X-ray Detection on a CCD Ｘ線 CCD平面模式図 電極 Single event Corner event 電子雲 Horizontally split event Vertically Split event 空乏層 CCD断面の模式図 まず、Ｘ線をＣＣＤがどうやって検出するか説明します。Ｘ線光子一個がＣＣＤに入射した場合、ＣＣＤ内部で光電吸収され、ここで、入射エネルギーに比例した数の電子を作ります。こちらに、ＣＣＤの断面、ＣＣＤを上からみた模式図を示します。生成された電子の数は数百から数千個にもなり、この電子の集団を電荷雲と呼んでいます。この電荷雲は、空乏層内の強い電場に引かれ、電荷転送路と呼ばれるポテンシャルミニマムまでドリフトします。ここで画素ごとに集められた電荷量を信号として検出しているわけですが、電荷雲がドリフト中にある程度広がってしまうので、電荷雲が一画素には治まらず、隣の画素にあふれて検出されることがあります。たとえば、Ｘ線が画素の中心に入射した場合、生成された電荷雲は一画素にすべて納まり、シングルイベントとなりますが、このように端に入射した場合、電荷雲が一画素には納まらず、隣の画素にあふれ、スプリットイベントになります。画素のコーナー付近に入射した場合は、３，４画素イベントとなり、Ｘ線光子一個からできた信号にはさまざまなパターンができます。また、エネルギーの高いＸ線がＣＣＤに入射すると、より深いところで光電吸収されるので、電荷雲もより大きくなると考えられます。 さて、ＣＣＤ表面はこのように、電極、チャンネルストップなどが配置されており、一画素といってもの表面構造は非常に複雑です。 この表面部分は、Ｘ線の吸収体として働くので、検出効率は一画素内で決して一様ではなく、詳細なレスポンスを得るためには、画素内のdead layerの厚みの違いを知る必要があり、そのためにはＸ線入射位置を画素よりも高い精度で制御できる実験技術が必要です。細いペンシルビームを作って、画素内を精密にスキャンニングする方法がアイディアとしては明快ですが、ビームをミクロン単位で制御し、画素内での入射位置を正確に知るのは不可能で、現実的ではありません。そこで、われわれはメッシュを用いて、画素ない入射位置を制御する方法を考案しました。 　Charge cloud 一個のＸ線光子により生成される信号を計測するため、入射フラックスは弱くする。 一個のＸ線光子がつくる電子雲はある程度広がる。 Ｘ線光子一個を検出したイベントにはさまざまなパターンがある。 Event pixel Split pixel Hiraga 2018/11/19 phD thesis 2002

2.Design for Laboratory Experiment The primary mirror has 80mm diameter and 2000mm focal length. The optical source, deformable mirror, wave front sensor, make it possible to monitor the shape of the telescope and the adaptive feedback system.

3. 望遠鏡の組み上げ真空化

X-ray Source Measure the reflectivity of known Mo/Si multilayers (2d~26nm) Clear peak at the incident angle of 33 deg.