Download Martin`s presentation

Martin Elvis, Astrophysics 2020, 14 November 2007
Active X-ray Optics
For The Next High Resolution X-ray Observatory
Martin Elvis
Harvard-Smithsonian Center for Astrophysics
Cambridge, Massachusetts, USA
1
Martin Elvis, Astrophysics 2020, 14 November 2007
45 Years of X-ray Astronomy:
1 billion times more sensitive
1962
1/10,000
Sco X-1
Good for 1 (one) Nobel Prize
good enough for my thesis
Detector Area,
Exposure time
1962
1/100,000
1982 HEAO-1
2002 Chandra
angular resolution
2022
1978Gen-X
2007
2
Martin Elvis, Astrophysics 2020, 14 November 2007
The Chandra Revolution:
Quantitative : 70 to 1400 Sources
ROSAT: ~5”
Chandra: ~0.5” [2.4 mrad]
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
Star Formation Region in Orion
ROSAT:The
~10”
3
Martin Elvis, Astrophysics 2020, 14 November 2007
The Chandra Revolution:
Qualitatively new Features
Hubble: ~0.1”
ROSAT: ~5”
Chandra: ~0.5”
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
The Antennae Colliding Galaxies
System and a
QuickTime™
TIFF (Uncompressed) decompressor
are needed to see this picture.
4
Martin Elvis, Astrophysics 2020, 14 November 2007
A High Resolution X-ray Successor to Chandra
 Obviously
desirable
 Obviously infeasible?
 Have to define what is needed
 Derive Technology
requirements
 Pursue promising technologies
5
Martin Elvis, Astrophysics 2020, 14 November 2007
The First Galaxies and Black Holes:

X-rays at z~10-20
Faint: 1st BH fluxes: ~10-3 of Deepest
Chandra surveys
 Area, Aeff ~ 100 m2
 Angular
resolution
HEW ~ 0.1”, 0.5mrad
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
 Reduce background
 Discriminate from foreground z=3 galaxies
 Energy
range
0.1-10 keV
 spectra kT~10keV / (1+z) ~1 keV

Z=10, Age = 480 Myr, 3.5%)
Defines next generation high resolution
large X-ray Observatory:
Generation-X
6
Martin Elvis, Astrophysics 2020, 14 November 2007
A High Resolution X-ray Successor to Chandra: Issues



No current plans for a Chandra-class sub-arcsec - mission - world-wide
No space agency developing high
resolution X-ray mirrors
Reason: Cannot make a bigger Chandra
 mirrors are heavy

QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
1.5 cm thick glass cylinders
 Can’t use integral shells
7
Martin Elvis, Astrophysics 2020, 14 November 2007
A High Resolution Successor to Chandra:
New Mirror Technology
HEW (arcsec)
100
Con-X
10
Chandra
1
0.0
0.5
1.0
Mass/unit area (kg cm-2)
Citterio et al.199x [Brera]
8
Martin Elvis, Astrophysics 2020, 14 November 2007
Generation-X Vision Mission Study





Gen-X selected as NASA Vision
Mission study in 2003
0.1” FWHM
100 m2 @ 1 keV
Nominal Launch date ~ 2020
Mission concept studies
Generation-X Vision Mission Study Report
March 9, 2006
Prepared for
National Aeronautics and Space Administration
(NASA) Headquarters
 JPL ‘Team-X’ : formation flying
 GSFC ‘IMDC’: single spacecraft


No Show Stoppers
Mirrors are the greatest challenge
 On-orbit adjustment of figure?
9
Martin Elvis, Astrophysics 2020, 14 November 2007
Active X-ray Optics Solution:
Piezoelectric Bi-morph (PBM)
•
Working at Synchrotrons
•
news to astronomers
•
10 year program by Signorato et al.
•
Operational
•
•
16-, 32- element
•
~1 m long optics
•
2 cm sized actuators
Kirkpatrick-Baez configuration
10
Martin Elvis, Astrophysics 2020, 14 November 2007
Active X-ray Optics:
figure improvement
•
Need factor ~100 correction:
•
•
~400 nm errors to ~4 nm
Finite element analysis shows
feasibility of control… in principle!
•Begin with Con-X optic goal,
•2 cm axial actuators give figure correction n < 0.025 mm- 1
I.e. Fourier low pass filter
•Correct to:
•6.5 nm rms 0.001<n<0.01 mm-1
~ 2 times Con-X goal
•1.6 nm rms 0.01<n<0.1 mm-1
~ 10 times Con-X goal
11
Martin Elvis, Astrophysics 2020, 14 November 2007
PBM Development needed for X-ray Astronomy
•
•
•
•
•
•
•
Thin replica substrates - bonding PBM
2-D Wolter geometry
• axial + azimuthal curvature
Radiation hard piezo materials
Cold operation piezos
Getting the wires out
Mass production: 100 m2 Aeff
104 m2 polished area
• Cost
• Speed - ~3 year production
~2x105 (2 cm actuators)/m2 Aeff :
•
•
Calibration
Calculation problem • closed loop essential in orbit
THERMAL
RADIATOR
THERMAL
COLLECTOR
INFLATABLESOLAR ARRAY
SUN SHIELD
GRATING ASSEMBLY
INFLATIABLE
INSTRUMENT
SHIELD
1.5M ANTENNA
SOLAR ARRAY
12
Martin Elvis, Astrophysics 2020, 14 November 2007
Active X-ray Optics:
A More Immediate Goal?
•
~50 times Con-X area is a big leap
•
Chandra/Con-X ‘Hubble/Keck’
complementarity be lost after Chandra.
•
Great Observatories IR/Opt/X
diminished
•
Budget future is always murky.
•
Should we consider a small active Xray optics mission before Gen-X?
•
Mirror cost/speed advantage may not be so
Blind men and the elephant. Manga VIII Hokusai, Katsushika (1760-1849)
great for smaller diameter mirrors;
•
Won’t reach 1st black holes;
•
… but still more powerful than Chandra
13
Martin Elvis, Astrophysics 2020, 14 November 2007
Active X-ray Optics
For The Next High Resolution X-ray Observatory
• Only way to go?
• Feasible: Piezoelectric Bimorph
Mirrors address biggest technical
challenge
• Development needed for
telescope use
• Rapid development program
could further all imaging X-ray
astronomy missions
The Crab Nebula
A Cosmic Synchrotron
14
Martin Elvis, Astrophysics 2020, 14 November 2007
Active X-ray Optics
opens a new long-term vision for X-ray Astronomy
Angular resolution
Gen-X
Hubble
0.1”
1”
Quic kTime™ and a
TIFF (Unc ompres sed) dec ompres sor
are needed to see this pic ture.
Galileo
1610
10”
Chandra
100”
Dawn of History
1600
1700
1800
1900
2000
From 0.5” to 0.1”, and on to the Diffraction limit <20masr
15
Martin Elvis, Astrophysics 2020, 14 November 2007
The Chandra Revolution:
qualitatively new structures
ROSAT: ~5”
Chandra: ~0.5”
The Supernova Remnant Cassiopeia A
16
Martin Elvis, Astrophysics 2020, 14 November 2007
Chandra only gives this Detail on the nearest of
each Class of Celestial Object
ROSAT: ~5”
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see thi s picture.
Chandra: ~0.5”
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
The Giant Galaxy M87 in the Virgo Cluster
17
Martin Elvis, Astrophysics 2020, 14 November 2007
The Chandra X-ray Observatory
Sub-arcsecond imaging by Chandra
has revolutionized astronomy
18
Martin Elvis, Astrophysics 2020, 14 November 2007
High Resolution X-ray Optics for Astronomy:
Challenging Requirements
•
High angular resolution, large area  thin shells
•
Axial figure errors comparable to Chandra
•
Azimuthal figure errors substantially better
 On-orbit adjustment of figure?
•
•
•
•
Advantages
Reduced ground calibration
Reduced launch stability
requirements
Can operate away from room
temperature
Slow adjustments ~10-5 Hz high orbit
C.f. 10 Hz on ground-based
telescopes
Challenges
•
Optical path clearance
•
Sensing misalignments
•
Calculating adjustments
•
Applying corrections
•
Stable actuators
19
Martin Elvis, Astrophysics 2020, 14 November 2007
Gen-X Study: Feasibility
2 options:
1. 125m f.l. Separate mirror detector s/c
2. 6 25m f.l., 8m dia mirrors
THERMAL
RADIATOR
THERMAL
COLLECTOR
INFLATABLE
SOLAR ARRAY
SUN SHIELD
 No show stoppers
GRATING ASSEMBLY
 Launch capability to Sun-Earth L2 OK
INFLATIABLE
INSTRUMENT
SHIELD
 Power budget OK
 Main Challenge: Mirror technology
1.5M ANTENNA
SOLAR ARRAY
 1/100 Chandra mass/area
 10 x better angular resolution
20
Martin Elvis, Astrophysics 2020, 14 November 2007
Piezoelectric Bi-morph Mirrors (PBM):
•
Good Properties for Astronomy I
Thin: no optical path blockage
•
Natural match to thin reflectors
•
0.2 mm
•
Low power, weight
•
Existing synchrotron K-B mirrors
comparable size to telescope
segments
•
Pairs of oppositely directed piezos
remove T dependence
•
Stable over days, months
•
No anticlastic effect (‘saddling’)
Suzaku Mirror segment
21
Martin Elvis, Astrophysics 2020, 14 November 2007
Piezoelectric Bi-morph Mirrors (PBM):
Good Properties for Astronomy II
•
Piezos parallel to mirror surface
•
Reduce amplitude of errors by factor 15
•
•
From 150 nm to 10 nm
•
Factor 100 more improvement possible
Citterio et al.
C.f. mechanical actuators:
No •
•
•
•
Optical path blockage
lubricants
hysterisis
backlash
22
Martin Elvis, Astrophysics 2020, 14 November 2007
Active X-ray Optics Alignment:
Signal & Compute Challenges
•
105 actuators! How to sense adjustments?
•
Form defocussed image
 Separate images of each shell, and azimuthal
sector of parabola-hyperbola pair
•
Factorizes calculation
•
Each shell segment, P-H pair is independent
•
Separate P, H via finite focus source?
•
Computation not overly demanding
• Example: 20m dia mirror, 10cm actuators
• Annular images 400 mm thick: 20
resolved elements with 20 mm pixels
23