Detection of Al II in the Ultraviolet Spectra of Metal-Poor Stars: An Empirical LTE Test of NLTE Aluminum Abundance Calculations

TL;DR

This study detects Al II lines in metal-poor stars and uses these to empirically validate NLTE corrections for Al I lines, confirming their accuracy in abundance calculations.

Contribution

It provides the first empirical test of NLTE aluminum abundance corrections using Al II lines in metal-poor stars.

Findings

Al II line detection in 11 metal-poor stars

LTE-based Al abundance matches high-excitation Al I lines

NLTE corrections for Al I resonance lines are validated

Abstract

We report the detection of an Al II line at 2669.155 Angstroms in 11 metal-poor stars, using ultraviolet spectra obtained with the Space Telescope Imaging Spectrograph on board the Hubble Space Telescope. We derive Al abundances from this line using a standard abundance analysis, assuming local thermodynamic equilibrium (LTE). The mean [Al/Fe] ratio is -0.06 +/- 0.04 (sigma = 0.22) for these 11 stars spanning -3.9 < [Fe/H] < -1.3, or [Al/Fe] = -0.10 +/- 0.04 (sigma = 0.18) for 9 stars spanning -3.0 < [Fe/H] < -1.3 if two carbon-enhanced stars are excluded. We use these abundances to perform an empirical test of non-LTE (NLTE) abundance corrections predicted for resonance lines of Al I, including the commonly-used optical Al I line at 3961 Angstroms. The Al II line is formed in LTE, and the abundance derived from this line matches that derived from high-excitation Al I lines predicted to have minimal NLTE corrections. The differences between the abundance derived from the Al II line and the LTE abundance derived from Al I resonance lines are +0.4 to +0.9 dex, which match the predicted NLTE corrections for the Al I resonance lines. We conclude that the NLTE abundance calculations are approximately correct and should be applied to LTE abundances derived from Al I lines.