Lithium abundances from the 6104 Å line in cool Pleiades stars

¹ Department of Physics, Keele University, Keele, Staffordshire, ST5 5BG, UK
² Department of Physics and Astronomy, The Open University, Walton Hall, Milton Keynes, MK7 6AA, UK e-mail: alison.ford@open.ac.uk; rdj@astro.keele.ac.uk or

Corresponding author: A. Ford, alison.ford@open.ac.uk

Received: 7 January 2002
Accepted: 17 May 2002

Abstract

Lithium abundances determined by spectral synthesis from both the 6708 Å resonance line and the 6104 Å subordinate line are reported for 11 Pleiades late-G and early-K stars observed at the William Herschel Telescope. Firm detections of the weak subordinate line are found for four objects, marginal detections for four, and upper limits for the remaining three stars. Some of these spectra were previously analysed by Russell (1996), where he reported that abundances derived from the 6104 Å line were systematically higher than those obtained from the 6708 Å line by 0.2–0.7 dex. He also reported a reduced spread in the 6104 Å line abundances compared with those determined from the 6708 Å feature. Using spectral synthesis we have re-analysed Russell's data, along with our own. Our results do not entirely support Russell's conclusions. We report a ~0.7 dex scatter in the abundances from 6708 Å and a scatter at least as large from the 6104 Å line. We find that this is partly explained by our inclusion of a nearby line and careful modelling of damping wings in the strong metal lines close to the 6104 Å feature; neglect of these leads to overestimates of the Li abundance which are most severe in those objects with the weakest 6104 Å lines, thus reducing the abundance scatter. We find a reasonable correlation between the 6104 Å and 6708 Å Li abundances, although four stars have 6104 Å-determined abundances which are significantly larger than the 6708 Å-determined values by up to 0.5 dex, suggesting problems with the homogeneous, 1-dimensional atmospheres being used. We show that these discrepancies can be explained, although probably not uniquely, by the presence of star spots with plausible coverage fractions. The addition of spots does not significantly reduce the apparent scatter in Li abundances, leaving open the possibility that at least some of the spread is caused by real star-to-star differences in pre-main sequence Li depletion.