[Paper Review] Oxygen Line Formation in Late-F through Early-K Disk/Halo Stars: Infrared O I Triplet and [O I] Lines

Yoichi Takeda|May 14, 2001

Stellar, planetary, and galactic studies34 references53 citations

TL;DR

This study investigates non-LTE effects in oxygen lines—specifically the infrared O I triplet (7771–5) and forbidden [O I] lines (6300/6363)—in late-F to early-K disk and halo stars. Using extensive non-LTE calculations, it finds that the O I triplet lines are well described by a two-level-atom scattering model with non-LTE corrections dependent only on line strength and atmospheric parameters, not metallicity, while LTE is valid for [O I] lines. A systematic 0.3 dex abundance discrepancy between O I and [O I] lines is confirmed in metal-poor halo stars, persisting even after non-LTE corrections, suggesting a fundamental issue in oxygen abundance determinations for these stars.

ABSTRACT

In order to investigate the formation of O {\sc i} 7771--5 and [O {\sc i}] 6300/6363 lines, extensive non-LTE calculations for neutral atomic oxygen were carried out for wide ranges of model atmosphere parameters, which are applicable to early-K through late-F halo/disk stars of various evolutionary stages. The formation of the triplet O {\sc i} lines was found to be well described by the classical two-level-atom scattering model, and the non-LTE correction is practically determined by the parameters of the line-transition itself without any significant relevance to the details of the oxygen atomic model. This simplifies the problem in the sense that the non-LTE abundance correction is essentially determined only by the line-strength ($W_λ$), if the atmospheric parameters of $T_{ m eff}$, $\log g$, and $ξ$ are given, without any explicit dependence of the metallicity; thus allowing a useful analytical formula with tabulated numerical coefficients. On the other hand, our calculations lead to the robust conclusion that LTE is totally valid for the forbidden [O {\sc i}] lines. An extensive reanalysis of published equivalent-width data of O {\sc i} 7771--5 and [O {\sc i}] 6300/6363 taken from various literature resulted in the conclusion that, while a reasonable consistency of O {\sc i} and [O {\sc i}] abundances was observed for disk stars ($-1 \la [{ m Fe}/{ m H}] \la 0$), the existence of a systematic abundance discrepancy was confirmed between O {\sc i} and [O {\sc i}] lines in conspicuously metal-poor halo stars ($-3 \la [{ m Fe}/{ m H}] \la -1$) without being removed by our non-LTE corrections, i.e., the former being larger by $\sim 0.3$ dex at $-3 \la [{ m Fe}/{ m H}] \la -2$.

Motivation & Objective

• To investigate the formation mechanisms of infrared O I triplet and [O I] forbidden lines in late-F through early-K disk and halo stars.
• To assess the impact of non-LTE effects on oxygen abundance determinations from these lines.
• To resolve the long-standing discrepancy between oxygen abundances derived from O I and [O I] lines in metal-poor stars.
• To determine whether the discrepancy is due to non-LTE effects or intrinsic astrophysical issues.
• To provide a robust, parameter-dependent analytical correction for O I triplet lines applicable to metal-poor stars.

Proposed method

• Conducted extensive non-LTE radiative transfer calculations for neutral oxygen across a wide range of model atmosphere parameters (Teff, log g, ξ) relevant to disk and halo stars.
• Applied a two-level-atom scattering model to describe O I triplet line formation, with source function solutions derived using Rybicki’s numerical scheme.
• Assessed the dependence of non-LTE corrections on line strength (Wλ), effective temperature, surface gravity, and atomic model parameters.
• Compared LTE and non-LTE emergent fluxes to compute equivalent widths and derive non-LTE abundance corrections (Δ).
• Reanalyzed published equivalent width data from multiple literature sources to compare O I and [O I] abundances across different metallicity regimes.
• Evaluated the influence of 3D hydrodynamical model atmospheres on abundance corrections, concluding their effect is minor for O I and [O I] lines.

Experimental results

Research questions

• RQ1 Does non-LTE significantly affect the formation of the O I 7771–5 triplet lines in late-F to early-K stars?
• RQ2 Are the observed discrepancies between O I and [O I] oxygen abundances in metal-poor stars due to non-LTE effects or intrinsic astrophysical factors?
• RQ3 Can a universal non-LTE correction formula be derived for the O I triplet lines based on line strength and atmospheric parameters alone?
• RQ4 How does the non-LTE correction for O I triplet lines depend on metallicity, and is this dependence negligible as suggested by the model?
• RQ5 Why is the O I–[O I] abundance discrepancy more prominent in metal-poor halo stars than in disk stars?

Key findings

• The non-LTE correction for the O I 7771–5 triplet lines is primarily determined by the line strength (Wλ) and atmospheric parameters (Teff, log g, ξ), with negligible dependence on metallicity.
• The O I triplet line formation is well described by a two-level-atom scattering model, simplifying non-LTE corrections to depend only on line strength and local atmospheric conditions.
• LTE is robustly valid for the forbidden [O I] 6300/6363 lines, with non-LTE corrections being negligible (|Δ| < 0.1 dex).
• A systematic 0.3 dex discrepancy is confirmed between O I and [O I] abundances in metal-poor halo stars (−3 ≤ [Fe/H] ≤ −1), persisting even after non-LTE corrections.
• The discrepancy is less pronounced in higher Teff/log g stars, indicating that dwarf or subgiant stars are more suitable for reliable oxygen abundance studies in metal-poor populations.
• The 3D hydrodynamical model atmosphere corrections do not significantly alter the conclusions, as their effect on O I and [O I] lines is minor or negligible.