1. The Sharpest Spatial View of a Black Hole Accretion Flow from the Chandra X-ray Visionary Project Observation of the NGC 3115 Bondi Region Jimmy Irwin 1, Ka-Wah Wong 1, Roman Shcherbakov 2 Mihoko Yukita 1, William Mathews 3, Chris Reynolds 2 1 University of Alabama 2 University of Maryland 3 UC-Santa Cruz See Roman Shcherbakov’s talk next for a theoretical view.

2. Bondi Flows Around Black Holes The “sphere of influence” around a black hole is defined by its Bondi radius, R B = 2GM BH /c s 2, where M BH is the mass of the black hole, and c s is the sound speed of gas in the vicinity of the black hole. For supermassive black holes in early-type galaxies, c s assumed to be representative of the ambient hot ISM ∝ √kT of gas at R B. Edgar (2009)

3. How Can We Constrain Accretion Flow Models Observationally? Models converging on agreement that: - T(R) ∝ R -1 - ρ (R) ∝ R -(½+p) where p = 0 – 1 (most show p = 0 – ½) Can we spatially resolve the hot gas within the Bondi radius of a SMBH to derive T(R) and ρ (R) profiles? Since R Bondi ∝ M BH /kT gas, need systems that have: – large black hole mass – cool ISM temperature – small distance

4. Best Resolved Bondi Radius Candidates Garcia et al. (2010) ×  NGC 3115   Only reasonable candidates for determining T(R) and ρ (R) are: M87: extremely bright jet knot makes analysis frustratingly difficult M31: low X-ray gas counts, luminous supersoft source within R B NGC 3115: low X-ray gas counts If only Chandra had 0.1” spatial resolution!

5. The S0 Galaxy NGC 3115 D = 9.7 Mpc (Tonry et al. 2001) M BH = 1-2 x 10 9 M  (Kormendy et al. 1996; Emsellem et al. 1999) kT gas ~ 0.35 keV (Wong et al. 2011) R B = 2.5”-5” No bright X-ray AGN/jet at its center Very low radio flux (Nyland & Wrobel 2012) ESO/VLT + Chandra (blue) – Wong et al. (2011)

6. 1 Megasecond Chandra Image of NGC 3115 30” x 30” Green D25 contour Green 5” Bondi radius

7. Multiple X-ray Components Within R B Spectrum: hot gas + unresolved XRBs + CV/ABs APEC (1 or 2): free kT, norm Power law: free norm All components fixed, norm by L K All components absorbed by Galactic column density (N H = 4 x 10 20 cm -2 ) Each region fit with XSPEC to obtain best-fit parameters, flux for each component + uncertainties.

8. Temperature Profile of Hot Gas Within NGC 3115 1” = 47 pc @9.7 Mpc 1-temperature thermal model2-temperature thermal model Effects of cooling? Effects of projection or mixed phases (e.g., Gaspari et al. 2013)?

9. De-projected Density Profile of Hot Gas Within NGC 3115 ρ at each radius determined from emission measure: EM = ∫ n e n h dV /4πD 2  Determine n e profile, and de-project From 5”-40” ρ (R) ∝ R -s s = 1.34 +0.09 Within 5” Bondi radius ρ (R) ∝ R -s s = 0.95 +0.24 -0.24 -0.15

10. Summary Our Megasecond Chandra observation of the Bondi region of the 2 x 10 9 M  SMBH of NGC3115 has so far revealed: 1) First temperature/density profile of hot gas within the Bondi radius of a supermassive black hole. 2) Evidence that temperature increases inside Bondi radius, as expected, but also a cooler component inside 3”. Projection or mixed phases? New physics? 3) Density profile within 5”: ρ (R) ∝ R -s, s =0.95 +0.24, in line with many simulations (but not a pure ADAF nor CDAF). More work to do! - 0.24

11. Bonus Science! 453 total point sources detected in all observations ~150 sources detected within one D25 contour (green ellipse) Limiting luminosity of ~10 36 ergs s -1

12. Surface Brightness Profiles of Components 1” = 47 pc @9.7 Mpc

13. Comparison to Other Galaxies 1” = 47 pc @9.7 Mpc Fukazawa et al. (2006)