Talk given at the NASA Anuual UV-Vis SR&T Workshop, NASA Headquartes, 20-21 September 2011.

Published on: **Mar 3, 2016**

Published in:
Technology

- 1. Atomic Data and Spectral Models for Lowly Ionized Iron-peak Species Manuel Bautista, Vanessa Fivet (Western Michigan University) Pascal Quinet (Mons University, Belgium) Connor Ballance (Auburn University)
- 2. • Reliable modeling of neutral through doubly ionized Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu is of great importance in various areas, e.g. H II regions, SNe remnants, AGN, supernovae light curves as cosmological candles, atmospheres of the Sun and late type stars, afterglows of GRBs, etc.
- 3. Emission spectra of η Carinae
- 4. Absorption spectrum of QSO 0059-2735. The spectrum is dominated byabsorption features from the ground and excited levels of Cr II, Fe II, Fe III,Co II, Ni II, Mn II, and possibly Ti II
- 5. • For most of these ions there are yet no spectral models available because even the fundamental atomic parameters are unknown.• For those ions that have been studied in the past, such as Fe II and Fe III, there was mounting evidence on that the models were inaccurate.
- 6. • For instance, predicted line intensities for Fe II in the Orion nebula, the simplest and best known nebular environment to astronomy, disagree with observations by up to several factors.
- 7. Ratio of CLOUDY predicted [Fe II] line intensities toobserved values in the Orion nebula (Verner et al 2000).
- 8. [Fe III] and [Fe IV]• A discrepancy of about a factor ~3 remains in the Fe abundances derived from [Fe III] and [Fe IV]• Rodriguez & Rubin (2004) argue that the errors could be either in the collision strengths or the total Fe3++e -> Fe++ recombination• Current collision strengths (Zhang 1996), but McLaughlin et al. (2002) report LS collision strengths lower by a factor of 2
- 9. Fe II• Excitation mechanisms for Fe II include electron impact, photoexcitation by continuum radiation, and fluorescence by Lyα.• Current models include over 800 levels (>300.000 transitions), e.g. Bautista et al. (2004).• But data still incomplete and unchecked.
- 10. Collisional coupling of pseudo- metastable levels
- 11. Bautista et al. (2004)
- 12. Comparison between bound-free cross sections of Bautista (1997) and hydrogenic approximations
- 13. Comparison of theoretical and observed emergent fluxes of the solar atmosphere
- 14. Goals of the project• Computation of reliable and complete data sets (A- values for allowed and forbidden transitions, collision strengths, photoionization cross sections and recombination rate coefficients) for neutral, singly and doubly ionized iron-peak species• Construction of spectral and opacity models whose quality will be benchmarked by modeling spectra of AGN and Eta Carinae• Distribution of the data and models among the scientific community• Implementing the atomic models into the photoionization modeling codes XSTAR (Kallman & Bautista 2001) and CLOUDY (Ferland et al. 1998)
- 15. Atomic Physics Hi Ei i N pi2 N Ze2 e2H i 1 2me i 1 ri i j r rj i two electron 1 N ( N 1) one electron 2 ZN 1 1 For neutral atoms 4 2
- 16. Atomic Physics, cont.• The two electron terms yield electron- electron correlations (radial and angular)• Current methods deal with electron correlations by: 1) optimization of radial functions 2) configuration interaction (CI) (CI: correlated solutions are written as linear combinations of non-correlated configurations)
- 17. Why are low Fe-peak ions difficult?• Very large number of metastable levels that participate in the spectra.• Strong radial correlations• Strong angular correlations• CI: always large but difficult to reach convergence• Relativistic effects
- 18. Atomic structure calculations• We use a combination of methods and codes: - HFR (Cowan codes) - MCDF (GRASP/GRASP92) - TFD central potential (SUPERSTRUCTURE) - We derive non-spherical multipole corrections to the TFD potential (Bautista 2008) that account for polarization and electron-electron correlations of filled and half-filled shells.
- 19. Angular electron correlationCalculated vs. measured energies in O I (2p4) Experiment (Ry) Theory (Ry)3P 0 01D 0.144 0.1601S 0.308 0.374
- 20. The O I problem• Ground configuration 1s22s22p4 3P 0 Ry J 1D 0.144 Ry 1S 0.307 Ry• Two important lines are the trans-auroral line at 2972Å (1S0-3P1) and the green line at 5577Å (1S0-1D2)
- 21. • The A-values recommended by NIST areA(2972 Å) = 7.54e-2A(5577 Å) = 1.26 and A(5577Å)/A(2972Å) = 16.7From Froese Fischer (1983) and Baluja & Zeippen (1988)Accuracy rating: B+
- 22. Theoretical Determination of theOI 557.7/297.2 nm Intensity Ratio• Condon, 1934 11.1• Pasternack, 1940 24.4• Garstang, 1951 16.4• Yamanouchie and Horie, 1952 30.4• Garstang, 1956 17.6• Froese Fischer and Saha, 1983 13.6• Baluja and Zeippen, 1988 13.0• Galavis, et al., 1997 14.2• Froese Fischer and Tachiev, 2004 16.1• NIST 16.7
- 23. Observational Determination of the OI 557.7/297.2 nm Ratio• Sharp and Siskind, 1989 ~9• Slanger et al., 2006 9.8±1.0• Gattinger et al., 2009 9.3±0.5• Gattinger et al., 2010 9.5±0.5
- 24. A(1S0-1D2 ) A(1S0-3P1 ) Ration=3 single prom. 1.50 0.28 5.3n=3 double prom. 1.44 0.29 4.9n=4 single prom. 6.36 0.38 18.8n=4 double prom. 1.45 0.29 4.9n=4 trip prom. 1.45 0.24 6.2n=5 double prom. 1.50 0.069 21.8n=5 triple prom. 2.26 0.073 30.9
- 25. The Fe III problemRatio of observed [Fe III] lines in the Orion nebula lines topredictions by previous models.
- 26. Approaches for scattering calculations• LS R-matrix + ICFT: allows for very large CI/CC expansions and ICFT includes relativistic effects in the outer-box region• Breit Pauli R-matrix: includes relativistic effects, but limited CI=CC expansion• DARC: fully relativistic calculation but for small CC expansion.
- 27. Maxwellian-averaged CollisionStrengths at 10,000 K for Fe III Upper lRM+ICFT DARC Zhang 5D 4.57E+0 2.54E+0 2.92E+0 3 5D 1.94E+0 1.11E+0 1.24E+0 2 5D 8.79E-0 5.33E-1 5.95E-1 1 5D 2.51E-1 1.60E-1 1.80E-1 0 3P2 7.14E-1 7.14E-1 5.80E-1 2 3P2 1.84E-1 1.96E-1 1.65E-1 1 3P2 3.83E-2 3.25E-2 2.13E-2 0 3H 2.66E+0 1.21E+0 1.34E+0 6 3H 1.10E+0 9.84E-1 4.89E-1 5 3H 2.41E-1 5.33E-1 9.26E-2 4 3F2 1.47E+0 4.54E-1 1.07E+0 4 3F2 6.42E-1 1.91E-1 4.35E-1 3 3F2 2.11E-1 1.73E-1 1.57E-1 2 3G 1.11E+0 1.36E+0 1.10E+0 5 3G 1.24E+0 1.11E+0 4.28E-1 4 3G 4.52E-1 4.21E-1 1.09E-1 3
- 28. (Itheo-Iobs)/Iobs New Fe III model
- 29. • Collision strengths for forbidden transitions are dominated by resonances.• All previous calculations use LS-coupling R-matrix, which does not include relativistic effects in resonance positions• Fully relativistic R-matrix methods are needed
- 30. Photoionization of Fe+ 3p 6 3d 6 3p 3d 4s h 3p 6 3d 5 4s 6 6 e 3p 6 3d 5 nl 3p 6 3d 6 3p 5 3d 8 5 7 3p 6 3d 5 4s e 3p 3d 4s 3p 6 3d 5 nl
- 31. Top: LS cross section of Nahar & Pradhan (2002). Middle: present DARC calculations. Lower: experiment (Kjeldsen et al. (2002)
- 32. The Fe II problem• We are carrying out fully relativistic R-matrix calculations for Fe II• We compare here with 75 lines measured in the optical spectrum of Orion by Mesa-Delgado et al. (2009).• Density and temperature are known from other species (ne=1.4x104 cm-3, T=9000K)
- 33. Spectral models for iron-peak ions Sc Ti V Cr Mn Fe Co Ni I II B07 B06 M06 B10a B05 B04 III B10b B01 IV Z97 M05
- 34. Photoionization cross sections for Iron-peak ions
- 35. Conclusions• Atomic data underpins most astronomical studies, from modeling microphysics processes, to diagnostics of plasma conditions, to full analysis of spectra.• Atomic data for neutral and singly ionized species are important in Op/UV astronomy.• Though, these computations test the limits of atomic methods.
- 36. Conclusions, cont.• For effective collision strengths @ 104 K one must a good representation of low energy resonances, mostly formed in the inner-box region => relativistic calculations must be performed (DARC)• When doing LS-calculations the larger the expansion the worse the results
- 37. Conclusions, cont.• New theoretical methods and computational tools are needed to treat electron-electron correlations.• We have created a new open forum (blog) to discuss atomic data issues in astronomyhttp://astroatom.wordpress.com/