The abundances of about fifteen elements in seven red giants of NGC 2360 and NGC 2447 are derived with a global uncertainty of about dex (essentially due to photon noise, normalization of the spectra and uncertainties related to atomic line parameters). From those abundances, the following conclusions are drawn.
The metallicity [M/H] of the two open clusters is directly derived from their iron-group elements. It amounts to 0.10 for NGC 2360 and 0.05 for NGC 2447.
The sodium abundance reveals a positive correlation with stellar mass in the range, as expected from model predictions. Quantitatively, the surface abundances are in good agreement with those predicted after 1DUP for stars (i.e. for stars in NGC 2360), but not at (NGC 2447). The disagreement in NGC 2447 reaches 0.08 dex, which we consider marginally significant given our estimated uncertainties on the Na abundances. The excess of Na observed in two of our stars of NGC 2447, if confirmed, cannot be explained by current stellar models. Other physical mechanisms, such as stellar rotation, should be investigated in the future. More observations in open clusters of various ages (and thus turn-off masses) are also required to increase the small number statistic of this study.
The oxygen abundance is very deficient (by 0.2 to 0.5 dex depending on the true gf value of the Ni i line) relative to iron in all our four red giants for which the oxygen abundance could be measured. This is not compatible with stellar model calculations which predict an oxygen deficiency by at most 0.03 dex after the first dredge-up. The high deficiencies can be partly explained if we assume that the Sun is slightly oxygen-rich relative to most stars in its neighborhood, as suggested by several authors. If not, then a new mechanism in stellar models should be thought of to account for this discrepancy.
© European Southern Observatory (ESO) 2000
Online publication: August 17, 2000