1. Hard X-ray Black Hole Surveys in Space and Time: EXIST Concept Josh Grindlay Harvard International Workshop High Energy Astrophysics in the Next Decade (Tokyo) June 21, 2006

2. Outline of talk Key HEA and EXIST science questions –Overview of EXIST as Black Hole Finder Probe (BHFP) Current surveys to answer them: –X-ray: Chandra, XMM, Swift-XRT, Suzaku –Hard X-ray: INTEGRAL, Swift-BAT, Suzaku/HXD & BAT-slew –TeV: HESS, VERITAS Upcoming surveys: –Agile, GLAST, LSST (approved) –MAXI, Lobster/eROSSITA, HXMT, Symbol-X (uncertain?) “Ultimate” spectral-temporal HX survey: EXIST –Mission study for BHFP in Beyond Einstein Program…

3. Key HEA Survey & EXIST Science (from the Galaxy to Pop III GRBs) Nature/number of BHs vs. NSs, WDs in Galaxy Blazar spectra @TeV vs. ~100keV: origin of EBL Dormant AGN (tidal disruption by SMBHs) –SMBHs in every galaxy? BH masses from power spec. Obscured AGN and origin of CXB GRBs from Pop III stars: birth of First BHs & probes of IGM

4. Overview of BHFP-EXIST Science and Design Hard X-ray (~3-600 keV) all-sky imaging each orbit to measure: Obscured AGN and accretion (BHs) vs. nuclear (stars) luminosity of universe GRBs out to z ~20 and first stellar Black Holes ( ~5-20X Swift sensitivity) Stellar Black Holes in Galaxy & IMBHs in Local Group & BHs as probes Mission Design parameters: Extend ROSAT sens. (~5 x 10 -13 cgs) to 3-150 keV with 0.9-5’ resolution & ~10” positions Two wide-field coded aperture telescopes: 10-600 keV (6m 2 CZT) & 3-30 keV (1m 2 Si) http://EXIST.gsfc.nasa.gov e.g., EXIST measures Cen-A every orbit: characteristic time variability (QPOs) constrain BH mass High Energy Telescope HET (10-600keV) (6 x 3 coded aperture telescopes; 131 o x 65 o FoV) Low Energy Telescope LET (3-30 keV) (4 arrays of 7 x 1 coded aperture telescopes; 116 o x 64 o FoV) IMDC design Dec. 2004

5. Expected EXIST Survey Sensitivity LET 3-30 keV; HET 10-600 keV 0.05mCrab = 5 x 10 -13 cgs, (over any band E low 2E low ) 5σ,  1yr, 20-40% duty cycle any source Continuum Narrow Line

6. Hard X-ray Sky Previous Hard X-ray Sky HEAO-1, BeppoSAX 2016(?) Hard X-ray Sky EXIST ≤2010 Hard X-ray Sky Swift & INTEGRAL Hard X-ray (10-600 keV) sky not yet surveyed to ROSAT sensitivity. EXIST would be ~20X more sensitive than Swift or INTEGRAL and cover full sky EXIST will detect ≥3 x 10 4 sources, ≤10  positions, 3-600 keV spectra EXIST would provide unique temporal survey: full sky imaging each orbit HXMT, eROSSITA?

7. Survey BHs, NSs and WDs in Galaxy Integral and Swift/BAT discovery of hard, variable giants Wind-fed accretion onto NS or BH: Provides variable, high NH hard sources, with Lx >10 36 erg/s as seen by Integral Complements ChaMPlane L x ~10 32 erg/s, high NH sources which are qLMXBs and CVs Complements Swift discovery of transients & hard pulsars (15-150 keV) L x >10 36 erg/s & HESS (TeV) discovery of Be-binary PSRs Simulations by J. Blondin

8. The Chandra Multiwavelength Plane Survey (ChaMPlane) Josh Grindlay, Ping Zhao, JaeSub Hong, Maureen van den Berg, Silas Laycock (Harvard, CfA) Distribution of ChaMPlane fields (●=ACIS-I, vs. o=ACIS-S) locations, exp. Times, & NH See Grindlay et al, Hong et al, Laycock et al, Zhao et al (2005, ApJ)

9. Stellar mass BHs vs. CVs in Galactic Bulge? ChaMPlane survey and optical-IR followups: a mixture of magnetic CVs, NSs and BHs in SgrA* and Bulge… Quantile diagram (see Hong et al 2004) for ~900 sources within 7’ of SgrA*: spectral classification & IDs (Hong et al 2006) IR variable (red star in 0.5” Chandra circle) for a “Muno source” in H1 class: Possible CH-Cam BH-LMXB ? (Laycock et al 2006) Optical IDs with OGLE variables In “Stanek Window” : probable Wind-fed CVs, symbiotics ( IPs?) (van den Berg et al 2006)

10. Survey Stellar BHs & IMBHs in Local Group with EXIST EXIST detects all bright stellar BHs in transients (Lx(>10 keV) ~10 37-38 erg/s) throughout Galaxy, LMC/SMC and M31. Reveal population of obscured HX sources found with INTEGRAL/Swift: discrete sources at >20 keV Isolated stellar BHs in Galaxy and IMBHs in Local Group accreting via Bondi-Hoyle (with ~10 -4 efficiency) from GMCs nearly Compton thick Faint BH transients in Central Galactic Bulge ?: BHs in nuclear cusp (cf. Alexander & Livio 04) detected if Lx(>10 keV) ~10 35 erg/s SgrA* CI Cam type outbursts (~1-2d?) of Bulge BH vs. WD binaries around SgrA* Chandra view of central Bulge (~ 2 o x 1 o )

11. New Swift/BAT Slew Survey (BATSS) (being developed at CfA for Swift/BAT science & EXIST testing) BAT slews ~60X/day between pointings; photon data could be sent down (1min/slew) for imaging all-sky survey (15-150 keV), ~1h/day fast timing gives: GRBs & Long (High z) GRBs (30% more coverage) New transients & pulsar monitoring (~70mCrab/orbit) Stacked sky images for fainter transients (~20mCrab/d~90%sky) New persistent sources Test scanning for EXIST 70sec BAT slew image across CygX-1 (S/N~19) to a GRB (which was the z =6.3 event!)

12. BATSS enables ~30% more GRBs & ~90% sky/day

13. BATSS coverage & sensitivity vs. time (e.g. 10days in May, 2006…) ~30% increased detection probability for GRBs, transients, or steady sources is relatively constant w.r.t. GRB/transient duration. BUT only BATSS would provide photon timing; normal BAT data binned in 5min integ. Enables first all-sky PSR monitoring Survey since BATSE

14. TeV Blazar surveys: constrain EBL & IRB (but only if constrain HX spectrum and variability) Ongoing HESS and (soon) VERITAS surveys discover new Blazars for comparison with X-ray & HX spectra New blazars detected by HESS at large z ! H2356-309 (z=0.165), 1ES1101-232 (z=0.186) Reconstructing EBL (HESS collab., submitted to Nature)

15. X-  Blazar surveys Agile & GLAST will discover flaring Blazars (~30MeV-30 GeV) Swift/BAT, INTEGRAL give partial HX coverage Swift/XRT, Suzaku, Chandra, XMM give partial X-ray coverage Need ~5-300 keV imaging & monitoring to constrain SSC GLASTMkn 421 flares– 0.5h – 2d Blazar SSC models Synch. IC

16. EXIST VERITAS GLAST SSC model for Mkn 501 (Coppi & Aharonian 1999) continuous HX spectral-monitoring nuclear vs. accretion luminosity of the universe EXIST will provide the continuous HX spectral-monitoring to study Blazars and non-thermal AGN to constrain diffuse IR (~10-100μ) background from obscured AGN and thus nuclear vs. accretion luminosity of the universe Complements GRB science: star formation vs. redshift from LGRBs vs. z Blazar Spectral variability: Extragalactic Background Light (EBL) & Stellar vs. Accretion Luminosity of Universe Blazars EBL: Hard x-ray (synchrotron) spectral breaks (~5-200keV) for Blazars at known redshift allow SSC  -ray (~10 GeV - 10 TeV) spectral breaks measured by GLAST and HESS/VERITAS to constrain origin of diffuse IR background Time-variability: spectral breaks required from simultaneous HX measurements. Wide-field HX imaging needed to match GLAST

17. Dormant SMBHs revealed by Tidal disruption (and BH masses from AGN power spectra) Artists conception of tidal disruption of star in RXJ1242-1119 detected with ROSAT (1991) and confirmed with Chandra (Komossa et al 2004). Tidal disruption of stars spiraling into Dormant SMBHs with mass ~10 7 Mo: if 1% of L acc in HX band, ~10 -5 events/year/Mpc 3 allow EXIST to see ~ 10-30 flares/yr out to ~200Mpc! (Grindlay 2004). HX spectral comp. “confirmed” with PL spectral decay of RX1242 Sub-giants with WD cores are gravitational wave LISA triggers. Possible soft (~5keV) prompt (~1d) burst detectable out to ~30 Mpc directly with EXIST (LET) and with MAXI or LOBSTER; LISA trigger Measure 10 6-7 M  SMBH content/evolution of nearby galaxies (understand BH-Bulge mass relation & BH-galaxy evolution!)

18. Obscured AGN and origin of the CXB NGC1068NGC6240 (Vignati et al 99) EXIST will find >1-10 obscured AGN/square degree and obtain first all-sky all-sky measure of Seyfert 2 QSO 2 luminosity function and constrain obscuration vs. z for supermassive BHs. Provide required all-sky survey for rare (Type 2 QSOs) and Lx dependence of NH EXIST survey HEAO Swift SAX/PDS Chandra & XMM surveys find >40% unresolved CXB from obscured AGN

19. Birth of Stellar BHs at z~5-20 “Long”-GRBs are from SNIb,c & likely due to stellar BH formation Likely that first stars were ~100M o and collapse to BHs GRBs “Short”-GRBs from merging NSs in globulars (Grindlay et al 2006) suggests Short GRBs enhanced at z of globular cluster formation? EXIST detects GRBs to z~20 from PopIII BHs at re-ioniz. epoch suggested by WMAP. Photometric z from Lum-E peak need response to E>300keV, And from Lum-Variability (Paczynski Relation) need large area det. X-ray flashes and high z GRBs need response to E~5 keV ~5sr instantaneous GRB coverage And increased sensivity: Rare (high z?) events; 3-5GRBs/d! Flux vs. detected E peak for GRBs from z=1 (top +) to 10 (bottom +) for E peak =30, 100, 300, 1000keV if emitted at z=1 vs. sensitivities (Band 2004).

20. Highest z stellar universe uniquely from GRBs Swift GRB at z = 6.3 shows high z universe is accessible Broader energy band, higher sensitivity needed for z~20 PopIII IR from space needed for z! GRBs provide “back-light” for IR spectroscopy of IGM, gas, galactic structure back to re-ionization: EXIST + JWST? Record-setting z vs. time: GRBs are gaining fast!

21. Zenith (Yaw) Orbit Normal (Pitch) Velocity (Roll) Solar Panels (fold-down) Thermal Radiator EXIST: Current Baseline Mission Design Concept Survey the hard X-ray sky 50x deeper than previous, with 10  source positions (5  ) Cover the 3 to 600 keV band with two telescope systems: HET: 10 - 600 keV (5.6 m 2 CZT ) and LET: 3 - 30 keV (1.1 m 2 Si) (GSFC ISAL & IMDC runs Oct. & Dec. '04) Low Energy Telescopes (LET: 3-30 keV) High Energy Telescopes (HET: 10 – 600 keV)

22. EXIST Mission Design Parameters Orbit Normal Zenith Nod direction, ┴ scan (+/-20 o ea. ~10min) Orbital scan direction (orbital velocity vector) Free-Flyer (500km, i  5-7º, low bkgnd) Zenith pointer - scanning & nodding for  full sky coverage each orbit (95min) 18 coded aperture HE telescopes (6m 2 total area CZT pixel det.) 28 coded aperture LE telescopes (1m 2 total area Si drift det.) Mass, power, telemetry: 9500kg, 3KW, 3Mbs Delta IV launch (to i ~5 o orbit) Mission lifetime: 5 years

23. Detector packaging: DCU, DCA, DM, Sub-Tel & HET Building a very large area CZT detector/telescope Detector Crystal Array (DCA) 2x2 DCUs (4x4 cm 2 ) + FPGA Board Detector Module (DM) 7x7 DCAs (28x28 cm 2 ) + MicroProcessor Board Sub-Tel Module 2x2 DMs (56x56 cm 2 ); each of 18! Active & Passive shields Detector Crystal Unit (DCU) Crystal (2x2 cm 2 ) (with Interposer Board?) + 2 x 128 channel ASIC (with micro-via tech?) HET 3x6 Sub-Tels (5.6 m 2 )

24. Baseline Mission parameters: EXIST/HET & LET Parameters EXIST/HETEXIST/LET Energy 10 – 600 keV3 - 30 keV Modules 3x64x7 Mask (W) 5mm thick, 2.5 mm pix0.05mm thick, 0.2mm pix. Detector (CZT) Area (Mod./Tot.) 5 mm thick, 1.25mm pix 56x56 cm 2 /5.6 m 2 1mm thick, 0.16 mm strip. 20x20cm 2 / 1.1 m 2 F.C. FoV (Mod./Tot.) 21º x21º / 65º x131º16º x16º /64º x112º Ang. Res./5σ Loc. 5.7/ 1.20.95/ 11  Mask-Det. Sep. 1.5 m0.72 m Temporal Res. < 0.1 ms Shields (GRB spectra to 10MeV ) CsI/passive side, CsI rear shields Passive Sensitivity (5σ) 0.05mCrab (<150 keV, ~1yr) 0.5mCrab (>150 keV, ~1yr) 0.15mCrab

25. Advancing the Mission Design The contiguous layout of HET sub-tel modules allows very efficient shield sharing and makes the overall instrument design simple and straightforward. The current mission design already meets the mission parameters required for science goals. However, the current layout introduces some unnecessary constraints on spacecraft design due to asymmetric packaging for launch. Goal: Redistribute sub-tel modules symmetric for launch and increase total area in LET (and HET).

26. New (General Dynamics) Design Concept Symmetric, smaller solar panels & increased LET Area & FoV Stowed Delta IV (5m Fairing) On Orbit Zenith Velocity Orbit Normal Solar Panels LET HET Spacecraft Bus

27. Side and Top Views of New Design

28. Instrument Layout & FoV Comparison Baseline Design HET: 5.6m 2 (18 mods) LET: 1.1m 2 (28 mods) HET 65  131  64  112  154  64  160  65  New Configuration HET: 6.0m 2 (19 mods) LET: 1.3m 2 (32 mods)  HET LET

29. Mission parameters for the new configuration Parameters EXIST HETEXIST LET Energy 10 – 600 keV4 - 30 keV Modules8+5+8=196+10+10+6=32 Mask (W) 5mm thick, 2.5 mm pix0.05mm thick, 0.2mm pix. Detector (CZT) 5 mm thick, 1.25mm pix1mm thick, 0.16 mm strip. Det. Area (Mod./Tot.) 56x56 cm 2 / 5.96 m 2 20x20cm 2 / 1.28 m 2 F.C. FoV (Mod./Tot.) 21º x21º / 65º x 154º 16º x16º / (32º  64º) x 160º Ang. Res./5σ Loc. 5.7/ 1.20.95/ 11  Mask-Det. Sep. 1.5 m0.72 m Temporal Res. < 1 ms Shields CsI/passive side, CsI rear shields Passive Sensitivity (5σ) 0.05mCrab (<150 keV, ~1yr) 0.5mCrab (>150 keV, ~1yr) 0.05mCrab

30. A wide dynamic range (>~10 3-4 ) to achieve high sensitivity impose challenges for coded-aperture imaging. The main factor for limiting the performance in coded- aperture telescopes is noise caused by unknown systematics in the system. These systematics include non-uniformity in the detectors such as detector gaps, dead pixels, efficiency variations, background variations, etc. EXIST’s scanning/nodding motion designed for surveying the sky is also a key to reduce the unknown systematics. We demonstrate imaging performance of EXIST by a series of progressively realistic simulations. Swift/BAT slew imaging tests are in progress; and demonstration of EXIST detectors and imaging planned with ProtoEXIST balloon experiments. Ensuring EXIST Imaging Performance

31. Artifacts due to gaps start appearing at ~0.5 mm gap in the sky image generated by a simple correlation without any correction Example: Imaging noise caused by detector gaps (click for movies) Scanning motion automatically averages out the coding noise without need for additional correction

32. Assuming a fraction of unknown dead pixels in the detector, 1 day exposure of one sub-tel in EXIST HETs Simulated Scanning vs. Pointing with dead-pixels 5  detection limit with no other source or with a 1.5 Crab source Dynamic Range with a 1.5 Crab source

33. Assuming a partially coded source in the FoV & unknown 1 mm crystal gap, apply a CLEAN procedure for the partially coded source Scanning vs. Pointing with a partially-coded source 5  detection limit with no other source or with a 1.5 Crab source Dynamic Range with a 1.5 Crab source

34. BAT slew imaging provides on-orbit test And new Swift/BAT survey science… ~70sec scan detects CygX-1 @19  vs. 14  for pointing with BAT Noise reduced due to averaging effects of scan

35. Prototyping EXIST Balloon program to develop ProtoEXIST (Harvard, GSFC, Caltech, Stanford collaboration) Phased development ( ProtoEXIST1 ProtoEXIST2 ) (2.5mm 1.2mm pixel pitch & direct-bonded ASIC) Demonstrate detectors & packaging, scanning imaging and (if ULDB?!) early EXIST survey science

36. The current test board (poptart) allows testing 4 Detector Crystal Units (DCUs = crystal + IPB+ASIC) 1 DCU: 2cm x 2cm X 0.5 cm CZT Custom ASIC 64ch, 3 x 9mm pkg., ~150microW/channel

37. Prototype Detector Performance Parasitic Capacitance of Input traces on Interposer board (IPB) (from simulations) FWHM measurements of a bare IPB using pulser Internal to ASIC

38. Radiation Tests on prototype DCUs for ProtoEXIST1 ~4.7 keV FWHM from 120 keV ( 57 Co) ~4.5 keV FWHM from 60 keV ( 241 Am) We expect ~3 keV FWHM from the next batch (Aug ’06) of DCUs with improved (lower capac.) IPBs. (vs. ~1.5 keV for direct-bonded ASICs in ProtoEXIST2 and EXIST)

39. Detector Crystal Arrays (DCAs) for ProtoEXIST1 Detector Crystal Unit (DCU) Top Socket Board Bottom FPGA Board FPGA (controls/reads out 8 ASICs on DCA) ASIC

40. Assembled ProtoEXIST1 DCA (design complete; under fab.) Enable tiling of 4cm x 8cm CZT arrays with 0.4mm gaps

41. Detector Module for ProtoEXIST1 (256cm 2 close-tiled array of 8 DCAs) Detector Crystal Arrays (DCAs) HV bias

42. Detector Module for ProtoEXIST1 Near gapless Packaging of 16cm x16cm CZT arrays EXIST sub-tels will have 56 cm x 56 cm CZT arrays.

43. Summary imagingINTEGRAL and Swift imaging have revealed the rich HX sky (obscured binaries, AGN, high z GRBs…!) Broad band (≥3 – 300 keV), large area and wide FoV are key. Unique all-sky imaging each orbit HX surveys have enormous synergy with GeV-TeV (GLAST- VERITAS) studies of Blazars (and EBL) wide- fieldBoth obscured and dormant SMBHs best studied with wide- field very sensitive HX imager measuredHighest z universe uniquely measured via GRBs couldEXIST under study for BHFP – could launch in ~2016 –See EXIST website ( http://EXIST.gsfc.nasa.gov) for Study & team ProtoEXIST1 balloon in 2008, ProtoEXIST2 in 2010… ProtoEXIST1 balloon in 2008, ProtoEXIST2 in 2010…