A comparison of abundance analyses of first generation stars in multiple populations in 47 Tuc and NGC 3201

TL;DR

This study compares abundance analyses of first-generation stars in 47 Tuc and NGC 3201, revealing that observed metallicity variations are likely due to analysis issues rather than real differences.

Contribution

It provides a critical comparison of spectroscopic analyses, highlighting potential biases and questioning previous claims of metallicity variations in globular cluster stars.

Findings

Metallicity trends are inconsistent between the two clusters.

Analysis methods may introduce spurious metallicity variations.

No significant intrinsic metallicity variation detected in first-generation stars.

Abstract

The distinction of the stellar content in globular clusters (GCs) in multiple stellar populations characterized by different amounts of proton-capture elements has been well assessed since a long time. On the other hand, the existence of noticeable variations in metallicity among GC stars is still debated. In particular, recent spectroscopic analyses claimed the presence of a small variation in metallicity, ~0.1 dex, for the first generation (FG) stars in NGC 3201 and NGC 104. However, in both cases the claim is not robust because of the internal error of 0.1 dex associated to the [Fe/H] values. To verify the reality of a metallicity variation we compared the two analyses, performed by the same authors with identical methodology. We found trends of metallicity as a function of the spectroscopically derived effective temperatures. However they are in opposite directions; in NGC 3201 cooler (and brighter) stars have higher [Fe/H] values, whereas in 47 Tuc they show lower metallicities. The trend is not statistically significant in the former case, but it is in the latter. The dependence of metallicity on the luminosity along the red giant branch seems to indicate problems in the abundance analysis for 47 Tuc. Finally, effective temperature do not show a significant variation as a function of the color spread along the HST pseudo-colour map, which we should instead observe were the trends temperature-metallicity a real effects of intrinsic scatter in iron. According to this comparison, we conclude that with these analyses, and the associated spurious trends, the issue of metallicity variations in FG stars is hardly settled.