Fluorine Evolution in the Galactic Halo

TL;DR

The paper addresses the unclear origin of fluorine in the Galaxy at low metallicity by employing a stochastic chemical evolution model of the Galactic halo. It combines rotation-dependent yields for massive stars from Limongi2018 and Roberti2024 with additional sources (LIMS FRUITY and SNe Ia) to predict $[F/Fe]$ and $[F/O]$ trends across metallicities. The study finds that rotating massive stars can produce high $[F/Fe]$ values (up to $\sim 2$ dex) at $[Fe/H] \sim -4$, reproducing recent observations and producing a plateau in $[F/O]$ at low $[O/H]$, while non-rotating models fail to match these features. These results underscore the importance of rotation for fluorine production in the early Galaxy and imply significant fluorine production at high redshift, motivating future measurements at very low metallicity and high redshift to further constrain the models.

Abstract

The chemical evolution of fluorine is still a matter of debate in Galactic archaeology, especially at low metallicities, where it is particularly challenging to obtain the corresponding chemical abundances from observations. We present here the first detailed theoretical study of the chemical evolution of fluorine at low metallicity by means of a stochastic chemical evolution model for the Galactic halo, in light of the most recent data for fluorine, which further pushed observations to lower metallicities down to [Fe/H]$\sim$-4 dex, more than a factor of 10 lower in metallicity than previous detections. We employ a state-of-the-art stochastic chemical evolution model to follow the evolution in the Galactic halo, which has been shown to reproduce well the main observables in this Galactic component and the abundance patterns of CNO and neutron-capture elements, and we implement nucleosynthesis prescriptions for fluorine, focusing on the chemical evolution of this element. By comparing recent observations with model predictions, we confirm the importance of rotating massive stars at low metallicities to explain both the [F/Fe] vs [Fe/H] and [F/O] vs [O/H] diagrams. In particular, we showed that we can reach high [F/Fe]$\sim$2 dex at [Fe/H]$\sim$-4 dex, in agreement with recent observations at the lowest metallicity. With a stochastic chemical evolution model for the Galactic halo, we confirm the importance of rotating massive stars as fluorine producers, as hinted by previous studies using chemical evolution models for the Galactic disc. We also expect an important production of F at high redshift, in agreement with recent detections of supersolar N by JWST. Further data for fluorine at low metallicities and also at high redshift would be needed to put further constraints on the chemical evolution of fluorine and be compared to our theoretical predictions.

Figures (4)

• Figure 1: Observed and predicted [F/Fe] versus [Fe/H] for the Galactic halo for our reference chemical evolution model with variable rotational velocity as in Rizzuti2025 (left) and non-rotating massive stars (right). The data are taken from Guzman2025 (blue) and Lucatello2011 (green), both upper limits and measurements with corresponding error bars; smaller dots correspond to CEMP-s stars. We also show Carina data from Abia2015 (magenta).
• Figure 2: Same as Fig. \ref{['fluorine']}, but for [F/O] vs. [O/H].
• Figure 3: Fluorine total yields in the stellar models of Limongi2018 (left column) and only wind (right column), for three initial rotational velocities: 0 km/s (upper panels), 150 km/s (middle panels) and 300 km/s (lower panels).
• Figure 4: Fluorine yields in the stellar models of Roberti2024 for 15 (left) and 25 M$_{\odot}$ (right) stars, as a function of the star initial equatorial velocity in km/s. Different colours correspond to different metallicities: [Fe/H]=-4, -5 and $-\infty$ (zero metals).