M giants with IGRINS II. Chemical evolution of fluorine at high metallicities

Context. The origin and evolution of fluorine in the Milky Way Galaxy is still under debate. In particular, the increase in the [F/Fe] in metal-rich stars found from near-IR HF lines is challenging to explain theoretically. Chemical evolution models with current knowledge of yields from different fluorine-producing stellar sources cannot reproduce these observations. Aims. The aim of this work is to observationally study the Galactic chemical evolution of fluorine, especially for metal-rich stars. We want to investigate whether the significant rise in fluorine production at high metallicities can be corroborated. Furthermore, we want to explore the possible reasons for this upturn in [F/Fe]. Methods. We determined the fluorine abundances from 50 M giants (3300 < T-eff < 3800 K) in the solar neighborhood spanning a broad range of metallicities (-0.9 < [Fe/H] < 0.25 dex). These stars are cool enough to have an array of lines from the HF molecule in the K band. We observed the stars with the Immersion GRating INfrared Spectrograph (IGRINS) spectrometer mounted on the Gemini South telescope and on the Harlan J. Smith Telescope at McDonald Observatory and investigate each of 10 HF molecular lines in detail. Results. Based on a detailed line-by-line analysis of ten HF lines, we find that the R19, R18, and R16 lines (22 699.49, 22 714.59, and 22 778.25 angstrom) should primarily be used for an abundance analysis. The R15, R14, and R13 lines at 22 826.86, 22 886.73, and 22 957.94 angstrom can also be used, but the trends based on these lines show increasing dependence on the stellar parameters. The strongest HF lines, namely R12, R11, R9, and R7 lying at 23 040.57, 23 134.76, 23 358.33, and 23 629.99 angstrom should be avoided. The abundances derived from these strongest lines show significant trends with the stellar parameters, as well as a high sensitivity to variations in the stellar microturbulence, especially for coolest and most metal-rich stars. This leads to a huge scatter and high fluorine abundances for super-solar metallicity stars, not seen in the trends from the weaker lines for the same stars. Conclusions. When estimating the final mean fluorine abundance trend as a function of metallicity, we neglect the fluorine abundances from the four strongest lines (R7, R9, R11, and R12) for all stars and use only those derived from R16, R18, and R19 for the coolest and most metal-rich stars. We confirm the flat trend of [F/Fe] found in other studies for stars in the metallicity range of -1.0 < [Fe/H] < 0.0 dex. We also find a slight enhancement at super-solar metallicities (0 < [Fe/H] < 0.15 dex) but we cannot confirm the upward trend seen at [Fe/H] > 0.25 dex. The HF line is intrinsically temperature sensitive, which calls for studies of stars with highly accurate and homogeneous stellar parameters. The spread in our trend is presumably caused by the temperature sensitivity. We need more observations of M giants at super-solar metallicities with a spectrometer that covers as many of the HF lines as possible, for instance the IGRINS spectrometer, to confirm whether the metal-rich fluorine abundance upturn is real or not.