EDP Sciences
Free Access
Volume 366, Number 3, February II 2001
Page(s) 981 - 1002
Section Stellar atmospheres
DOI https://doi.org/10.1051/0004-6361:20000287

A&A 366, 981-1002 (2001)
DOI: 10.1051/0004-6361:20000287

Kinetic equilibrium of iron in the atmospheres of cool dwarf stars

I. The solar strong line spectrum
T. Gehren1, K. Butler1, L. Mashonkina1, 2, 3, J. Reetz1 and J. Shi1, 4

1  Institut für Astronomie und Astrophysik der Universität München, Universitäts-Sternwarte München, Scheinerstr. 1, 81679 München, Germany
2  Max-Planck Institut für Astrophysik, Karl-Schwarzschild-Str. 1, 85741 Garching, Germany
3  Department of Astronomy, Kazan State University, Kremlevskaya 18, Kazan 8 420008, Russia
4  Beijing Astronomical Observatory, Chinese Academy of Sciences, 100012 Beijing, China

(Received 16 August 2000 / Accepted 17 November 2000 )

Line formation calculations of Fe I and Fe II in the solar atmosphere are presented for atomic models of iron including all observed terms and line transitions with available f-values. Recent improved calculations of Fe I photoionization cross-sections are taken into account, and the influence of collision processes is investigated by comparing synthesized and observed solar line flux profiles. The background is represented by the opacity of all important non-iron elements with iron lines added. Using a representative sample of sufficiently unblended strong Fe I and Fe II line profiles, it is evident that line formation is affected by (a) velocity fields and (b) deviations from local thermodynamic equilibrium (NLTE). The calculations are extended to a systematic analysis demonstrating that the ionization equilibrium of iron is recovered for solar parameters ( $T_{\rm eff}= 5780$ K, $\log g = 4.44$) either using the empirical atmospheric model of Holweger & Müller (1974) and assuming LTE for both Fe I and Fe II or a line-blanketed theoretical atmospheric model with NLTE iron line formation. In the latter case the kinetic equilibrium of Fe I shows a substantial underpopulation of Fe I terms which depends sensitively on both the improved photoionization calculations and the choice of hydrogen collision rates while the Fe II ion is well approximated by LTE. Although the source functions of most of the Fe I lines are nearly thermal, their formation is shifted deeper into the photosphere. NLTE wings of strong Fe I lines are therefore shallower than under the LTE assumption, whereas the cores of the strongest lines display the usual chromospheric contributions. Based on both calculated and laboratory f-values the abundances of 37 Fe II lines range between $\log\varepsilon_{FeII,\odot} =
7.50$ and 7.56, depending on atomic and atmospheric models, and those of 117 Fe I lines between $\log\varepsilon_{FeII,\odot} = 7.47$ and 7.56, both with a relatively large scatter of 0.08 ... 0.12. The collisional coupling of Fe I levels is investigated. Electron collisions seem to play only a minor role. Hydrogen collisions are very important between terms of low excitation, and they efficiently thermalize the line source functions but not necessarily the populations of the lower levels that determine the line optical depth. Thermalization of those low-excitation terms that are responsible for most of the lines analyzed is achieved only if the collisional coupling among highly excited Fe I terms and their Fe II parent terms is increased by large factors compared with standard collision rates. Solar flux profiles are reproduced under the assumption of both LTE or NLTE, with nearly all types of atomic and atmospheric models, because the Fe ionization equilibrium depends on the corresponding sets of f-values.

Key words: line: formation -- line: profiles -- Sun: photosphere -- Sun: abundances

Offprint request: T. Gehren, gehren@usm.uni-muenchen.de

© ESO 2001

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.