Atomic data needs for X-ray analysis and AtomDB v2.0


Published on

Talk presented by Randall Smith (Smithsonian Astrophysical Observatory) at the meeting The X-ray Universe 2011, Berlin, 27-30 June 2011.

Published in: Education, Technology
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Atomic data needs for X-ray analysis and AtomDB v2.0

  1. 1. Atomic Data Needs for X-ray Analysis and AtomDB v2.0 Randall Smith Smithsonian Astrophysical Observatory
  2. 2. Collaborators and Thanks• AtomDB Collaborators: – Adam Foster, Nancy Brickhouse, John Raymond, Li Ji, Tim Kallman, Mike Witthoeft• Thanks to: – ADAS team, CHIANTI team, ITAMP, Kate Kirby, Verne Jacobs, Ulyana Safronova, Guo- Xin Chen, Brendan McLaughlin, Brad Wargelin, and many others…
  3. 3. A Few Words about X-ray NeedsMajor needs are H-like, He-like, and Fe(and to a lesser extent Ni) L-shell linedata. Sutton’s Law: “First consider the obvious.”’s_la
  4. 4. A Few Words about X-ray Needs Major needs are H-like, He-like, and Fe (and to a lesser extent Ni) L-shell line data. Sutton’s Law: “First consider the obvious.”(Willie Sutton is the bank robber who supposedly answered the question “Why do yourob banks?” with “Because that’s where the money is.”)
  5. 5. A Few Words about X-ray NeedsMajor needs are H-like, He-like, and Fe(and to a lesser extent Ni) L-shell linedata. TW Hya, Chandra HETG Brickhouse et al. 2010
  6. 6. Current Problems: He-like“In the case of Si XIII,we find some evidencesuggesting that the low-density limit of the ratio1s2s3S→g/1s2p3P→gpredicted by both APEDand Porquet et al. (2001)are too low by 20%–40%or so based on ourobserved values.” Measured R ratios for Si XIII vs. the stellar X‐ray surface flux for all the analyzed stars (Testa et al. 2004).
  7. 7. Current Problems: He-likeComparison between the temperatures derived from the G ratio ofHe‐like triplet lines and the temperatures derived from the ratio of theLyα line of the H‐like ion and the resonance line of the He‐like ion, forMg. The dashed lines mark the locus of equal values. (based on Testaet al 2004)
  8. 8. Current Problems: He-likeData updated for He-like Ne IX using an R-Matrixcalculation for the levels up to n=5, and thenextending that to n=10 using a distorted wavecalculation. Smith et al. 2009
  9. 9. Current Problems: H-like Bandpass for Mg XIGoing to higher n-shell is needed – and still an outstandingproblem for all but H-like, He-like ions.
  10. 10. Current Problems: Ionization BalanceChanging the ionization balance can dramatically change the emission
  11. 11. Current Problems: Ionization Balance• 30% errors on the ionization & recombination rates leads to ~30% errors in the ion population.• However, at the limits of an ion’s population, these errors are increased up to 60%.
  12. 12. Current Problems: Fe XVIIFe XVII 3C/3D from the point of view of observational data and variouscalculations
  13. 13. Current Problems: Fe XVII Gillaspy et al. 2010Fe XVII 3C/3D from the point of view of laboratory data and various calculations
  14. 14. Coming Next: Future ProblemsINTERMISSION
  15. 15. Low Abundance Elements Suzaku 100ks Tycho observation shows Mn and Cr fluorescence lines from partially- ionized material. Relevent atomic parameters had to be estimated. (Tamagawa+08)Badenes+(2008) considered modelSNIa explosions with differentneutron excesses and variousclasses of explosionsFor the progenitor of Tychos SN,the data requires a near-solar orsupersolar metallicity Badenes et al 2008
  16. 16. Charge ExchangeNew measurements of astrophysically-relevant CX.
  17. 17. Charge Exchange norma l ???? ? Leutenegger et al 20101. Never get involved in a land war in Asia2. Never go against a Sicilian when death is on the line3. As President Reagan said, “Trust, but verify.”
  18. 18. Wavelengths Fit using ‘raw’ HULLAC wavelengths Fit using lab (Brown et al. 1998) wavelengths Fit using newly calculated wavelengths Kotochigova et al 2010• By combining laboratory measurements and theoreticalstructure calculations, can get highly accurate (few mÅ)wavelengths.• Detectors with R>1000 require this kind of accuracy!
  19. 19. Wavelengths Lee et al 09Absorption edges and related features still need a lot of
  20. 20. Fluorescent Line Data Bandler et al. 2010•The fluorescent lines fromradioactive sources are not simpleLorentzians, but complexcombinations of many lines.•Line widths measured by current Hoelzer et alcalorimeters are dominated by the 1997natural line widths – which are notknown for many useful elements!
  21. 21. Conclusions• Send lawyers, guns, and money
  22. 22. Conclusions• Send lawyers, guns, and money• We need a balanced program of focused – Laboratory Measurements – Theoretical Calculations – Data Curation of Both• Astro-H will usher in a new world of high- resolution spectra, especially in the 2-10 keV band. We must be ready for it!
  23. 23. BACKUP
  24. 24. Charge Exchange Leutenegger et al 2010
  25. 25. Photoionized Plasmas NGC 3783 – X-ray bright Seyfert galaxy;This talk has focused on 900 ksec HETG, 280 ksec RGS observationsthermal plasmas;photoionized plasmas areeven more complex as moreions come into play, andboth collisional and photoncross sections are relevant.• PL =1.53 + kTBB~0.1keV• 6 param 100 features• Two phase absorber w/ singleoutflow + turbulent velocity.• Fe M shell UTA vital• Continuous range of ionizationparameters not favored.Krongold et al. 2003; also Netzer et al. 2003 and Behar et al. 2003
  26. 26. Current Problems: 50-150ÅChandra has a wide bandpass, but the long-wavelength end suffers frompoorly understood excitation rates and many unidentified lines, often fromL-shell ions of Si, S, and other elements.
  27. 27. A Few Words on X-ray Missions Why is the atomic data we need almost always in the 1-50Å (0.25-10 keV) range? – Top end set by elemental abundances; Ni the last abundant element: all lines have  > 1Å. – Low end set by absorption • “Minimal” absorption is ~1020 cm-2. • At 50Å, this absorption implies  ~ 0.5• Resolution required (for thermal plasmas) set by thermal broadening:
  28. 28. Current Problems: Fe L-shell Desai et al (2005)
  29. 29.
  30. 30. A3112: A cluster with excess soft X-ray emission?The excess soft X-rays could be thermal or non-thermal;CCD and proportional counter data could not distinguish
  31. 31. RGS spectra shows no significant Fe XVII or Fe XVIII lines
  32. 32. Results…constraining any thermal soft X-ray excess to be 30-100x less than claimed