Draft version June 6, 2024

Typeset using LATEX twocolumn style in AASTeX631

Fluorine Abundances in Local Stellar Populations

K. E. Brady,1 C. A. Pilachowski,1 V. Grisoni,2 Z. G. Maas,1 and K. A. Nault3
1 Indiana University Bloomington, Department of Astronomy, 727 East 3rd St., Bloomington, IN 47405, USA
2 INAF, Osservatorio Astronomico di Trieste, via G.B. Tiepolo 11, I-34131, Trieste, Italy
3 University of Iowa, Department of Physics and Astronomy, 203 Van Allen Hall, Iowa City, IA 52242, USA

ABSTRACT
arXiv:2406.02855v1 [astro-ph.GA] 5 Jun 2024

We present the first fluorine measurements in 12 normal giants belonging to the Galactic thin and
thick disks using spectra obtained with the Phoenix infrared spectrometer on the 2.1m telescope at
Kitt Peak. Abundances are determined from the (1−0) R9 2.3358 µm feature of the molecule HF.
Additionally, sodium abundances are derived in 25 giants in the thin disk, thick disk, and halo using the
Na I line at 2.3379 µm. We report fluorine abundances for thin and thick disk stars in the metallicity
range −0.7 < [Fe/H] < 0. We add two abundance measurements for stars with [Fe/H] < 0.5 dex
which are at a critical metallicity range to constrain models. We find a larger dispersion in fluorine
abundances than sodium abundances despite both species having similar overall uncertainties due to
atmospheric parameters, suggesting this dispersion is real and not observational. The dispersion is
slightly larger in the thick disk than the thin. The thin and thick disk average [F/Fe] for our sample
of stars combined with the literature differ by 0.03 dex. The observations are compared to available
chemical evolution models.

Keywords: Stars: Abundances, Stars: Late Type; Galaxy: abundances; Galaxy: disk

1. INTRODUCTION view of the proposed nucleosynthetic sources of fluorine,

The abundances of light elements provide insight into see Section 5.1 in Ryde et al. (2020).
the chemical evolution of the Milky Way and its stellar Determining the abundance of fluorine in stars re-
populations (see Matteucci 2021 for a review). The cos- mains difficult due to many observational challenges.
mic origin of light elements with even atomic numbers Due to the lack of strong, measurable atomic fluo-
(e.g., O, Mg, Si) is generally well understood, but the rine lines, the only indicator of fluorine abundance that
origin of many odd-Z elements (e.g., F, P, Cl) remains can be measured in cool atmospheres are spectral lines
under debate (Nomoto et al. 2013). Understanding the from the hydrogen fluoride (HF) molecule. Vibration-
formation of odd elements provides an opportunity to rotation HF lines exist in the K band, and pure rota-
constrain nucleosynthesis processes as, for example, 19 F tional HF lines in the N band. Telluric lines in the
(the lone stable isotope of fluorine) is integral in proton- near-infrared (near-IR) can complicate the derivation
capture reactions that include the CNO and Ne−Na cy- of abundances, and the HF molecule has a relatively
cles. low dissociation energy, limiting studies to stars below
The light, odd-Z element fluorine is thought to be pro- ∼4700 K.
duced in multiple stellar sites such as asymptotic giant While early work on the abundance of fluorine in stars
branch (AGB) stars (Jorissen et al. 1992), rapidly ro- focused primarily on AGB stars (e.g., Jorissen et al.
tating massive stars (Prantzos et al. 2018), Wolf−Rayet 1992; Cunha et al. 2000; Abia et al. 2009, 2010) and
stars (Meynet & Arnould 2000), the ν process in core- carbon-enhanced metal-poor stars (Schuler et al. 2007;
collapse supernovae (Woosley & Haxton 1988), and no- Lucatello et al. 2011), recent investigations have turned
vae (Olive & Vangioni 2019). In the context of Galac- to the determination of the fluorine abundance in main-
tic chemical evolution, several nucleosynthetic channels sequence and pre-AGB giant stars to identify the source
may contribute to the Galactic fluorine abundance, with of fluorine in the Milky Way. These studies, such as
different processes dominating certain stellar popula- Jönsson et al. (2014a, 2017), Pilachowski & Pace (2015),
tions (such as thin-disk, thick-disk, and bulge stars) Guerço et al. (2019), and Ryde et al. (2020), use high-
and/or at certain metallicities. For a comprehensive re- resolution spectrometers in the near-IR to measure HF
2 Brady et al.

at 2.3 µm. For the first time in abundance derivation 2.1. Target Selection
studies, Jönsson et al. (2014b) explored a pure rota- Candidate stars belonging to the Galactic disk were
tional HF line at 12.2 µm, finding that the derived flu- selected for observation from the surveys of Barbuy &
orine abundance agreed with the 2.3 µm vibrational- Erdelyo-Mendes (1989), McWilliam (1990), Alves-Brito
rotational HF line. Many of these studies have con- et al. (2010), Ramı́rez et al. (2013), Prugniel et al.
trasting conclusions regarding the chemical evolution of (2011), Wu et al. (2011), and Pakhomov (2013). UVW
fluorine in the Milky Way. Most recently, Ryde et al. space velocities were computed using the PyAstronomy
(2020) indicated a need for multiple cosmic sources of Galactic space velocity function2 with the peculiar so-
fluorine over time. lar motion correction from Ding et al. (2019). Coor-
The solar abundance of fluorine was originally deter- dinates, proper motions, distances, and radial veloci-
mined to be log ϵ(F) = 4.56 ± 0.30 by Hall & Noyes ties were taken from the Gaia Data Release 3 (DR3;
(1969) based on observations of sunspots. Since then, Gaia Collaboration et al. 2016, 2023). Membership in
new measurements of the oscillator strengths and ex- different kinematic groups was established by requiring
citation potentials of the IR (1−0) HF features have that the probability of belonging to one population be
become available in the HITRAN database (Rothman greater than 50%, following the methodology of John-
et al. 2013), and Maiorca et al. (2014) have reanalyzed son & Soderblom (1987) as described by Ramı́rez et al.
the solar abundance of fluorine. With the new, exper- (2013). The final sample includes nine thin-disk stars,
imental molecular data, Maiorca et al. (2014) found a sixteen thick-disk stars, and two halo stars, as shown in
solar fluorine abundance of log ϵ(F) = 4.40 ± 0.25 from Table 1.
a careful analysis of eight HF features in sunspot umbral
spectra from the Wallace et al. (2001) atlas. This value 2.2. Observations
is in good agreement with the meteoritic abundance at Spectra covering the relatively unblended 2.3359 µm
log ϵ(F) = 4.42 ± 0.06 (Lodders et al. 2009). (1−0) R9 feature of HF and the Na I line at 2.3379 µm
Just as new observational results became available, so were obtained using the Phoenix infrared spectrometer
too have new chemical-enrichment models that provide on the 2.1m telescope at the Kitt Peak National Obser-
a theoretical framework for interpreting the abundance vatory in 2012 November and December, and 2013 De-
of fluorine in the Galactic disk. Authors include differ- cember. The spectrometer was configured with a 0.”7
ent nucleosynthesis prescriptions for fluorine to predict slit and the 4308 filter to isolate grating order 32, pro-
the abundance of fluorine as a function of [Fe/H]1 and viding spectral coverage from 2.3285 to 2.3390 µm. Ob-
[O/H]. Recent studies include Kobayashi et al. (2011a, servations were obtained at two slit positions for sub-
2020), Spitoni et al. (2018), Prantzos et al. (2018), Olive traction of a simultaneous sky spectrum and dark cur-
& Vangioni (2019), Grisoni et al. (2020), and Womack rent. Hot stars were observed frequently on each night
et al. (2023). at similar air mass to allow effective telluric-line correc-
In this work, we report on new determinations of the tion, and stars were observed close to the meridian to
abundance of fluorine in giants belonging to the Galactic mimimize telluric absorption features.
thin and thick disks, and compare the dependence of
[F/Fe] on [Fe/H] and [F/O] on [O/H] both with the thin 2.3. Data Reduction
and thick disks and with theoretical models. In Section Spectra were reduced by subtracting background and
§2 we describe the target selection, observations, and night-sky emission using the ABBA observing technique,
data reduction. Section §3 details the spectral analysis. correcting for variations in pixel sensitivity by divid-
We compare our results to the literature in Section §4. ing by a continuous lamp spectrum, extracting one-
In Sections §5 and §6 we discuss fluorine enrichment in dimensional spectra, removing telluric-line contamina-
the thin and thick disks and the chemical evolution of tion by dividing by the spectrum of a hot star, and cali-
fluorine. Finally, in Section §7 we summarize our results. brating the wavelength scale using telluric methane lines
in the hot-star spectra. Finally, the spectra were nor-
malized in the continuum and shifted to match the lab-
oratory rest wavelength scale. The observations of each
2. OBSERVATIONS AND DATA REDUCTION star, including star identifications, spectral types and K
magnitudes from SIMBAD, the dates of observation, the

1 We use the standard spectroscopic notation where [A/B] ≡

2 https://pyastronomy.readthedocs.io/en/latest/pyaslDoc/aslDoc
log(NA /NB )star − log(NA /NB )⊙ and logϵ(A) ≡ log(NA /NH ) +
12.0 for elements A and B. /gal uvw.html
 Fluorine Abundances in Local Stellar Populations 3

Table 1. Stars Observed with Phoenix and the Kitt Peak 2.1 m Telescope

Spec. UT Exp. S/N

Star HIP Alt. Type K Mag. Date Time (s) Ratio

Thin Disk
HD 6555 5213 BD +22 174 K0III 5.410 28 Dec. 2013 4 x 600 160
HD 10057 7632 BD +1 293 K2III 3.059 30 Dec. 2013 4 x 300 300
HD 29065 21297 HR 1452 K3III 1.820 2 Dec. 2012 4 x 60 120
HD 49520 32844 58 Aur K3III 2.024 28 Dec. 2013 4 x 90 200
HD 106760 59856 HR 4668 K0.5III 2.110 29 Dec. 2013 4 x 90 300
HD 211683 110141 BD +09 5019 K2 4.612 31 Dec. 2013 4 x 400 205
HD 212074 110382 BD +14 4782 K1IV 5.013 1 Jan. 2014 4 x 600 200
HD 216174 112731 HR 8688 K1III 2.625 30 Dec. 2013 4 x 150 195
HD 233517 ··· BD +53 1239 K2 6.637 30 Nov. 2012 4 x 900 160
Thick Disk
HD 249 607 BD +25 5073 K1IV 4.879 28 Dec. 2013 4 x 600 190
HD 3546 3031 ϵ And G7III 2.070 28 Dec. 2013 4 x 120 260
HD 9138 7007 µ Psc K3III 1.660 30 Nov. 2012 4 x 30 210
HD 15596 11698 27 Ari G8III 3.861 31 Dec. 2013 4 x 400 300
HD 37171 26386 HR 1908 K5III 2.120 28 Dec. 2013 4 x 90 145
HD 44030 30142 BD +25 1223 K4III 4.139 28 Dec. 2013 4 x 300 300
HD 50778 33160 tet CMa K4III 0.660 30 Nov. 2012 4 x 20 130
HD 71597 41538 BD +0 2305 K1IV 4.503 1 Dec. 2012 4 x 500 190
HD 77729 44592 BD +26 1895 K4III 4.170 30 Dec. 2013 4 x 400 160
HD 96436 54336 65 Leo K0III 3.329 29 Dec. 2013 4 x 300 240
HD 103813 58306 BD +27 2073 K0III 5.057 28 Dec. 2013 4 x 600 125
HD 105740 59334 BD +17 2445 G9III 5.847 29 Dec. 2013 4 x 600 165
HD 107328 60172 c Vir K0.5III 2.200 29 Dec. 2013 4 x 90 180
HD 121146 67589 HR 5227 K2IV 3.655 17 Feb. 2014 4 x 300 200
HD 221345 116076 14 And G8III 2.331 28 Dec. 2013 4 x 120 190
BD +27 2057 ··· ··· G7III 7.220 30 Dec. 2013 4 x 900 100
Halo
HD 8724 6710 BD +16 149 G5 5.636 31 Dec. 2013 4 x 500 195
HD 63791 38621 BD +62 959 G0 5.427 1 Jan. 2014 4 x 900 300

Note—Spectral types and K magnitudes from SIMBAD. Sixteen stars have probabilities greater
than 90% of belonging to their predicted population, while seven stars are in the range 0.75 ≤ P
< 0.9 (HD 29065, thin; HD 37171, thick; HD 49520, thin; BD +27 2057, thick; HD 6555, thin; HD
106760, thin; and HD 211683, thin), and four stars are in the range 0.5 < P < 0.75 (HD 10057,
thin, at P =0.52; HD 50778, thick, at P =0.56; HD 3546, thick, at P =0.64; and HD 212074, thin,
at P =0.72). While HD 10057 is designated as a thin-disk member with P = 0.52, it was identified
as a thick-disk giant with a 97% probability by Pakhomov (2013) based on criteria from Mishenina
et al. (2004).

total exposure time, and the signal-to-noise (S/N) ratio sphere parameters (effective temperature, surface grav-
of the resulting spectra are provided in Table 1. S/N ity, [Fe/H], and microturbulance) were adopted from lit-
ratios of the sample star spectra range from 100 to 300, erature sources identified in Table 2. Model atmospheres
while the S/N ratios of telluric-line stars were in excess were interpolated from the MARCS grid (Gustafsson et
of 300. Sample spectra of the sample stars are shown in al. 2008) and spherical models were used for our sample
Figure 1. of giants. Sample best fit syntheses for the HF and Na
I line are shown in Figure 2.
3. ANALYSIS
The abundances of fluorine and sodium in the sample Following Ryde et al. (2020), we adopt the meteoritic
stars were determined using spectrum synthesis with the solar fluorine abundance of A(F )⊙ = 4.42 from Lodders
local thermodynamic equilibrium (LTE) spectral analy- et al. (2009) due to the large uncertainties of the solar
sis code MOOG (V. 2019, Sneden 1973). Synthetic spec- fluorine values derived from modeling umbral spectra.
tra covering the 2.3359 µm (1-0) R9 feature of HF and The molecular line data used for analysis of the
the Na I line at 2.3379 µm were convolved with Gaussian 23358.329 Å (1−0) R9 feature of HF has historically
profiles to fit the final reduced spectra. Model atmo-
4 Brady et al.

Table 2. Stellar Parameters and Abundances

Star Tef f Log g ξ [Fe/H] Ref log ϵ(F ) [F/Fe] log ϵ(N a) [Na/Fe]

(K) (km s−1 )

Thin Disk
HD 6555 4720 3.00 1.15 –0.09 b ··· ··· 6.40 0.25
HD 10057 4130 1.70 1.35 –0.30 b 4.15 0.03 5.89 -0.05
HD 29065 3990 1.77 2.30 –0.35 c 4.00 -0.07 5.89 0.00
HD 49520 4300 2.35 2.20 0.00 c 4.20 -0.22 6.24 0.00
HD 106760 4510 2.70 2.10 –0.29 c 4.16 0.03 5.95 0.00
HD 211683 4450 1.80 1.35 –0.13 b 4.62 0.33 6.11 0.00
HD 212074 4700 2.55 1.30 +0.05 b ··· ··· 6.39 0.10
HD 216174 4440 2.53 2.40 –0.53 c ··· ··· 5.71 0.00
HD 233517 4475 2.25 1.85 –0.37 p ··· ··· 5.87 0.00
Thick Disk
Arcturus 4286 1.66 1.74 –0.52 q 3.73 -0.17 5.91 0.19
HD 249 4850 2.96 1.17 –0.15 b ··· ··· 6.29 0.20
HD 3546 5109 2.83 1.73 –0.50 i ··· ··· 5.79 0.05
HD 9138 4395 2.07 1.98 –0.34 a 4.56 0.48 6.25 0.35
HD 15596 4936 3.06 1.19 –0.53 i ··· ··· 5.76 0.05
HD 37171 4000 1.25 1.35 –0.55 b 3.90 0.03 5.74 0.05
HD 44030 4300 1.00 3.00 –0.70 d 4.23 0.51 5.79 0.25
HD 50778 4084 1.70 1.50 –0.22 e 4.13 -0.07 ··· ···
HD 71597 4271 1.90 1.80 –0.40 f <3.65 <-0.37 5.89 0.05
HD 77729 4270 2.00 3.00 –0.40 d 4.20 0.18 ··· ···
HD 96436 4750 2.75 1.50 –0.50 k ··· ··· 5.94 0.20
HD 103813 4816 3.11 1.01 –0.36 l ··· ··· 6.08 0.20
HD 105740 4700 2.50 1.00 –0.51 m ··· ··· 5.93 0.20
HD 107328 4442 1.95 1.51 –0.39 n ··· ··· 5.85 0.00
HD 121146 4400 2.25 1.90 –0.06 o 4.64 0.28 6.43 0.25
HD 221345 4740 2.61 1.44 –0.27 i ··· ··· 5.97 0.00
BD +27 2057 4810 2.25 1.25 –0.51 g ··· ··· 6.03 0.30
Halo
HD 8724 4757 1.77 1.65 –1.53 h ··· ··· 4.41 -0.30
HD 63791 4725 1.70 1.60 –1.68 j ··· ··· 4.46 -0.10

Note—We use log ϵ(F e)⊙ = 7.50 from Asplund et al. (2009), log ϵ(F )⊙ = 4.42 from Lodders et al.
(2009), and log ϵ(N a)⊙ = 6.24 from Asplund et al. (2009).

References— (a) Luck & Heiter (2007); (b) Pakhomov (2013); (c) McWilliam (1990); (d) Gratton &
Ortolani (1984); (e) da Silva et al. (2006); (f) Barbuy & Erdelyo-Mendes (1989); (g) Afşar et al. (2012);
(h) Hayes et al. (2018); (i) da Silva et al. (2015); (j) Burris et al. (2000); (k) Cottrell & Sneden (1986);
(l) Deka-Szymankiewicz et al. (2018); (m) Shetrone (1996); (n) Smiljanic et al. (2016); (o) Luck &
Challener (1995); (p) Balachandran et al. (2000); (q) Ramı́rez & Allende Prieto (2011).

varied. We adopted the excitation potential χ = 0.227 et al. (2014a) and would result in an abundance differ-
from Decin (2000), the HITRAN database (Rothman et ence of only ∼0.01. Some uncertainty remains in the
al. 2013), and Jönsson et al. (2014a), log gf = -3.962 HF molecule’s dissociation energy, which could result in
from Jönsson et al. (2014a), and dissociation energy of a lower fluorine abundance of ∼0.04 dex (see Section
5.869 from Sauval & Tatum (1984). In early abundance 4.1.1 in Guerço et al. 2019).
determinations the excitation potential of 0.49 eV was Atomic line data for the Na I line at 2.3379 µm were
used, which results in a higher fluorine abundance of adopted from Pilachowski & Pace (2015), who used
typically 0.36 (see Section 3.2 in Nault & Pilachowski the atomic and molecular line list generator linemake
2013). Jönsson et al. (2014b) confirmed that the parti- (Placco et al. 2021), then confirmed by fitting the Na
tion functions used with MOOG and the Decin (2000) I line in the Infrared Atlas of the Arcturus Spectrum
line data are compatible with their results. Several other (Hinkle et al. 1995).
log gf values have been determined, such as -3.971 (Lu- Multiple CO (2−0) and (3−1) vibration-rotation lines
catello et al. 2011) and -3.956 (Maiorca et al. 2014), are found near the HF and Na features. The neighbor-
which are similar to the adopted value from Jönsson ing C12 O16 lines are included in the synthesis calculation
 Fluorine Abundances in Local Stellar Populations 5

orine abundance is high, as the HF feature disappears

in warmer stars due to molecular dissociation.
Our analytic methods can be confirmed by measuring
the 2.3359 µm (1−0) R9 HF feature in the cool giant
standard Arcturus from the atlas of Hinkle et al. (1995).
We derived a fluorine abundance of A(F) = 3.73 using
the atmospheric parameters from Ramı́rez & Allende
Prieto (2011) (Tef f = 4286 K, log g = 1.66, [Fe/H] =
-0.52, and ξ = 1.74 km s−1 ). This value is comparable
with recent Arcturus fluorine measurements by Guerço
et al. (2019) (A(F) = 3.63), Abia et al. (2015) (A(F) =
3.78), Jönsson et al. (2014a) (A(F) = 3.75), Jönsson et
al. (2014b) (A(F) = 3.65 from the 2.3 µm line and A(F)
= 3.73 from the 12.2 µm line), and Nault & Pilachowski
Figure 1. Sample spectra showing the features of HF at
23359 Å and Na I at 23379 Å used in this analysis. The top (2013) (A(F) = 3.75).
spectrum is HD 71597, the middle spectrum is HD 44030, We detect rotation in HD 233517, which is a Li-rich,
and the bottom spectrum is HD 29065. The spectra have single K giant known to have a high rotation velocity
been shifted vertically for display purposes. Spectra were (v sin i). By fitting the rotational broadening of the
selected to show the range of HF equivalent width and signal- 2.3379 µm Na I line, we find a v sin i of 17.5 ± 4 km
to-noise ratio used in the analysis. s−1 . The rotational line broadening of HD 233517 was
previously measured by Balachandran et al. (2000) and
using adopted wavelengths, log(gf ) values, and excita- Strassmeier et al. (2015), who report values of 17.6 ±
tion potentials from Goorvitch (1994). Due to the lack 1.0 km s−1 and 17.8 ± 0.1 km s−1 , respectively.
of carbon and oxygen abundances for our program stars No known non-LTE (NLTE) corrections exist for the
in the literature, we are unable to measure individual 2.3359 µm (1-0) R9 HF feature. LTE should generally be
abundances of these elements from only the observed CO expected for molecular vibrational-rotational transitions
lines. The carbon abundance was adjusted to match the in the ground electronic state (Hinkle & Lambert 1975).
observed CO strength, then the abundances of fluorine Li et al. (2013) explored the deviations from radiative
and sodium were measured in a similar manner. equilibrium for the 2.3359 µm (1-0) R9 HF line in the
In one of our program stars (HD 121146), the (2−0) low-metallicity range near [M/H] = -2.0. They deter-
R25 line of C12 O17 is present and partially blended with mined that the effects of three-dimensional (3D) model
the HF line. This C12 O17 line at 23357.177 Å was in- atmospheres on their derived fluorine abundances were
cluded in the synthesis using line data from linemake. insignificant at low gravity and temperature, but larger
linemake also provided line data for the C12 O17 line at corrections (> 0.1 dex) were expected at higher grav-
23358.237 Å (Mura-Guzmán et al. 2020), which was not ity and temperature. Our sample consists of stars with
detected in HD 121146. temperatures less than 4500 K, with the exception of HD
Final fluorine and sodium abundance measurements 106760, which has an effective temperature of 4510 K,
are shown in Table 2. We report fluorine abundances surface gravity of 2.70, and a derived fluorine abundance
for five thin-disk and six thick-disk stars, and one thick- of [F/Fe] = 0.03. In Figure 3 we detect no dependence
disk star upper limit. Stars must be sufficiently cool of fluorine abundance on effective temperature, provid-
(less than ∼4700 K) to permit the formation of HF in ing some assurance that NLTE and 3D effects are not
their atmospheres. In our sample stars that are at tem- substantial for stars less than ∼4500 K. With the selec-
peratures above this limit, the HF line was not detected tion effect of the HF line, where the line is detectable at
and only measurements of the Na I line were obtained. higher effective temperatures if the abundance is high,
The [F/Fe] abundance ratios are plotted versus effective it is difficult to identify possible NLTE and 3D effects.
temperature in Figure 3, with the metallicity shown in However, we can examine the derived abundances at the
the color bar following Ryde et al. (2020). As observed cool end of our sample, which show no obvious increase
and discussed in Pilachowski & Pace (2015) and Ryde et in abundance that is indicative of such effects.
al. (2020), we see a rise in fluorine abundance with tem- No NLTE corrections currently exist for the Na I line
perature that is likely a systematic effect. In the warmer at 2.3379 µm. Cunha et al. (2015) discuss NLTE cal-
stars, the fluorine feature will only be detected if the flu- culations for the studied Na I lines in the APOGEE
spectral region in the near-IR H band, finding depar-
6 Brady et al.

Figure 2. The observed and synthetic spectra for the (1−0) R9 2.3358 µm feature of the HF molecule in HD 50778, and the
Na I line at 2.3379 µm in HD 10057. Three syntheses are plotted in each panel, with the best fits being A(F) = 4.14 and A(Na)
= 5.89.
tures from LTE to be minimal (≤ 0.04 dex) for their & Pace (2015), Guerço et al. (2019), Ryde et al. (2020),
sample of giants in NGC 6791. Our sample of giants and the data from Jönsson et al. (2017) that Ryde et
has a similar parameter range as Cunha et al. (2015)’s al. (2020) reanalyzed are included. The measurements
sample, though our metallicities are lower. We observe from these studies have been scaled to align with the
no temperature trend with derived sodium abundance, molecular data and solar fluorine abundance used in
as seen in Figure 3. this study. The derived sodium abundances are also pre-
Uncertainties in our abundance results arise from sented in Figure 4, with sodium abundances of red-giant
observational uncertainties and uncertainties in the stars in the thin/thick disk and halo from Alves-Brito
adopted atmospheric parameters for each star. Obser- et al. (2010) for comparison.
vational uncertainties were estimated by altering the Our sample covers a metallicity range from −0.70
smoothing of the synthetic spectrum and the fits them- < [Fe/H] < 0.00, with measurements at low- and
selves until the difference in abundance measurement approximately-solar metallicity. Recent fluorine abun-
were distinguishable. To estimate the uncertainties in dance studies with measurements in this regime include
the adopted stellar atmospheric parameters, the sensi- Jönsson et al. (2014b) (−0.62 < [Fe/H] < −0.14), Pila-
tivity of the fluorine abundance to each individual pa- chowski & Pace (2015) (−0.6 < [Fe/H] < +0.3), Guerço
rameter was measured. These uncertainties are sum- et al. (2019) (−1.3 < [Fe/H] < 0.00), and Ryde et al.
marized in Table 3. Uncertainties may arise from the (2020) (−1.11 < [Fe/H] < +0.37). For our sample, the
telluric correction of the near-IR spectra. Fortunately, average fluorine abundance is [F/Fe] = +0.14 ± 0.07
H2 O only has a few weak lines in the region of inter- (standard error of the mean, SEM) with a standard de-
est. The total uncertainty is determined by adding the viation of 0.23. The Pilachowski & Pace (2015) sam-
individual sources in quadrature. Depending on the ef- ple of ∼40 normal giants has an average abundance of
fective temperature of the star, the total uncertainty for [F/Fe] = +0.07 ± 0.04 (SEM). Average fluorine abun-
the derived fluorine abundances is less than 0.23 dex dances in Guerço et al. (2019) and Ryde et al. (2020)
across the temperature range from 4000 to 4500 K. are [F/Fe] = −0.10 ± 0.06 (SEM) and [F/Fe] = +0.03
± 0.02 (SEM), respectively. By taking an average of the
fluorine abundances from the 2.3µm and 12.2µm where
measurements exist from each line, the average fluorine
4. COMPARISON TO PREVIOUS WORK
abundance from Jönsson et al. (2014b) is [F/Fe] = −0.09
To our knowledge, we present the first fluorine mea- ± 0.03 (SEM).
surements for our sample stars. Figure 4 shows the de- At approximately-solar metallicity, the abundance
rived abundances versus metallicity, compared to nor- measurements from all studies have similar [F/Fe] val-
mal giants from the literature. Fluorine abundances ues, though there is an observed scatter (see Figure 4).
from Li et al. (2013), Jönsson et al. (2014b), Pilachowski
 Fluorine Abundances in Local Stellar Populations 7

Table 3. Sensitivity to Stellar Parameters

Parameter log ϵ(F) log ϵ(Na)

Tef f = 4000 K Tef f = 4500 K Tef f = 4000 K Tef f = 4500 K

∆Tef f = ±100 K 0.18 0.22 0.14 0.12

∆Log g = ±0.3 0.02 0.00 0.00 0.00
∆ξ = ±0.5 km s−1 0.02 0.00 0.12 0.10
∆[Fe/H] = ±0.1 0.05 0.03 0.06 0.07
Observational Uncertainty 0.05 0.05 0.05 0.05
Total Uncertainty 0.20 0.23 0.20 0.18

The scatter is greater at solar metallicity than at su-

persolar metallicities. As seen in Table 3, the dominant
source of uncertainty on [F/Fe] ratios is from errors in
the effective temperature. The methods of determin-
ing temperature among different studies may partially
explain the observed scatter and differences in average
fluorine abundances across different samples. However,
fluorine and sodium have similar overall uncertainties
due to atmospheric parameters, but the observed scat-
ter in the fluorine abundances is greater than the scatter
in the sodium abundances, as seen in Figure 4.
Previously, it has been suggested that fluorine in
higher-metallicity stars is dominated by a secondary
source (see Section 5.1 in Ryde et al. 2020 for an
overview on primary and secondary fluorine production Figure 3. Derived abundance vs. effective temperature,
sites and processes). Ryde et al. (2020) found a sec- with metallicity ([Fe/H]) indicated by color.
ondary behavior in their sample stars and concluded
that AGB stars are the main channel of formation in the least ∼0.3 dex greater than the values derived in low-
metallicity range of −0.6 < [Fe/H] < 0. This behavior metallicity stars in Guerço et al. (2019) and the upper
is demonstrated in the A(F) versus A(Fe I) plot in Fig- limits from Ryde et al. (2020). Our abundance deter-
ure 4 of Guerço et al. (2022), where the slope indicative mination for HD 44030, which has a [Fe/H] = −0.70, is
of secondary yields is shown to fit the observed trend comparable to the abundance determinations in the low-
of fluorine at higher metallicities. Our derived fluorine metallicity regime for the two metal-poor red giants from
abundances in the metallicity range of −0.5 to 0.0 are Guerço et al. (2019), determined to be probable mem-
consistent with the values from the literature, following bers of the Monoceros overdensity. The other abundance
the observed secondary behavior of disk stars. measurement at low metallicity for the star HD 37171
At lower metallicities, a primary source of fluorine at [Fe/H] = −0.55 is comparable to measurements from
dominates (Guerço et al. 2019; Ryde et al. 2020; Guerço Pilachowski & Pace (2015) and the data from Jönsson et
et al. 2022). Guerço et al. (2019) proposes primary be- al. (2017) that Ryde et al. (2020) reanalyzed at a similar
havior at [Fe/H] < −0.5, and secondary behavior at metallicity.
[Fe/H] > −0.5. Ryde et al. (2020) suggests at lower
metallicities the abundance ratios may decrease more 5. FLUORINE ENRICHMENT IN THE THIN AND
slowly than at higher metallicities, and Ryde et al. THICK DISKS
(2020) and Guerço et al. (2019) report a near-flat trend
Using our classification procedure for kinematic pop-
of abundance with [Fe/H] at low metallicity. Guerço et
ulations detailed in Section 2.1, we assigned the stars
al. (2019) reports a gradient of [F/Fe] for their probable
from the literature to be members of the thin disk, thick
thick-disk/halo stars to have a shallow slope of 0.02 ±
disk, or halo. Gaia DR3 data were utilized when avail-
0.03 dex. We report two fluorine measurements in low-
able. In instances where Gaia DR3 data were unavail-
metallicity ([Fe/H] < −0.5 dex) stars. Our two mea-
able, proper motions, parallaxes, and radial velocities
surements have derived fluorine abundances that are at
came from Gaia DR2 (Gaia Collaboration et al. 2018),
8 Brady et al.

Karataş et al. (2004), Famaey et al. (2005), Gontcharov as a result of its quicker evolution. At solar metallic-
(2006), van Leeuwen (2007), Massarotti et al. (2008), ity, the majority of models show declining [F/Fe] with
Famaey et al. (2009), Soubiran et al. (2018), and Jönsson [Fe/H]. The “two-infall” model from Spitoni et al. (2018)
et al. (2020). The final thin-/thick-disk and halo sam- matches our derived abundances at low metallicity and
ples from the literature are plotted with our results in solar metallicity quite well. Nonetheless, the data show
Figures 4, 5, and 6. Combining our derived abundances an increase in [F/Fe] at ∼+0.25 which is unexplained by
with the values from the literature, we find a final thin- the models. The Grisoni et al. (2020) models have the
disk fluorine abundance of [F/Fe] = +0.04 ± 0.02 (SEM) closest slope to the observed fluorine abundances from
with a standard deviation of 0.24, and a thick disk abun- the literature at high metallicity.
dance of [F/Fe] = +0.07 ± 0.08 (SEM) with a standard Thin- and thick-disk stars were distinguished when
deviation of 0.29. For thin-disk stars in the metallicity comparing [F/O] ratios to [O/H] ratios in Grisoni et
range –0.6 < [Fe/H] < 0.3, Pilachowski & Pace (2015) al. (2020), possibly due to more uncertain yields of Fe.
find an average [F/Fe] ratio of +0.07 ± 0.04 (SEM), We plot [F/O] vs. [O/H] in Figure 6 using oxygen abun-
which is comparable to our results. Within the range dances from Smith & Lambert (1990), Luck & Challener
from approximately −0.7 < [Fe/H] < 0, the thin- and (1995), Meléndez et al. (2008), Hinkel et al. (2014) (Hy-
thick-disk average abundances do not differ by more patia Catalog), and Jönsson et al. (2020) (APOGEE).
than 0.03 dex. We do not see differences between populations, though
Both our sample and our sample combined with the this may be due to higher uncertainties. At lower metal-
literature values show a slightly larger dispersion in the licities, the thick-disk stars have higher [F/O] measure-
thick disk than the thin disk, with standard deviations ments than the models.
larger in thick-disk samples. This can be seen in Figures While sophisticated chemical evolution models of flu-
4 and 5, and is even more pronounced when considering orine have been created, the predictions have yet to re-
our upper-limit thick-disk measurement for HD 71597. produce all features of the observed abundances. This
Scatter in [F/Fe] is higher at lower metallicity, which may in part be due to uncertainties remaining in Galac-
suggests there could be a variation from star to star in tic chemical evolution models, in addition to the un-
the thick disk. The dispersion could be either real or certainty in deriving fluorine abundances. Grisoni et al.
observational as scatter increases as the HF line gets (2020) point out that the largest uncertainty in Galactic
weaker. The [Na/Fe] in Figure 4 shows a relatively flat chemical evolution modeling is nucleosynthesis prescrip-
trend for both the thin and thick disks with the popula- tions, leading to proposed corrections to help explain the
tions overlapping each other, similar to the distribution observed abundance trends. Many prescriptions may
and pattern observed by Li et al. (2018). be underestimated: Grisoni et al. (2020) multiplied the
yields of AGB stars by 5, Prantzos et al. (2018) pro-
6. CHEMICAL EVOLUTION OF FLUORINE posed a factor of 2 increase in AGB star yields, and
Spitoni et al. (2018) increased the yields of AGB stars,
Finally, we compare the observed fluorine abundances
novae, and Wolf−Rayet stars. Moreover, there are dif-
to chemical evolution models to constrain possible nu-
ferent nucleosynthesis prescriptions for AGB stars that
cleosynthetic sources of fluorine. Figure 5 shows our
provide distinctive results when implemented in chemi-
derived fluorine abundances and the kinematically as-
cal evolution models. Figure 1 of Grisoni et al. (2020)
signed literature sample with recent chemical evolution
illustrates how different sets of low- and intermediate-
models fully described in Kobayashi et al. (2011a, 2020),
mass star yields result in vastly different fluorine predic-
Prantzos et al. (2018), Spitoni et al. (2018), and Grisoni
tions on a plot of [F/Fe] versus [Fe/H]. These nucleosyn-
et al. (2020) overlain. As in Figure 4, the literature
thesis prescriptions affect the model behavior at high
abundance measurements and theoretical models have
metallicity. Instances where the slope of the literature
been scaled to align with the molecular data and solar
abundances at higher metallicities is not reproduced by
fluorine abundance used in this study.
a model may be due to the nucleosynthesis prescriptions
Some models show a peak in [F/Fe] at ∼−0.1 which
used for low- and intermediate-mass stars in that model.
is explained by different star-formation histories. These
Further data and investigation into the prescriptions of
histories can lead to different abundance patterns, which
different nucleosynthetic sources of fluorine are needed
can be understood in terms of the time-delay model
to draw definitive conclusions on the origin of fluorine
(Matteucci 2021). The Grisoni et al. (2020) models
in different stellar populations.
use a parallel approach (see Grisoni et al. 2017, 2018),
where the disks are formed in parallel, but the thick-disk
track is shifted to higher metallicities than the thin disk 7. SUMMARY
 Fluorine Abundances in Local Stellar Populations 9

Figure 4. [F/Fe] (upper panel) and [Na/Fe] (lower panel) vs. [Fe/H] for thick disk (red circles), thin disk (blue circles), and
halo (green circles) stars. Literature measurements for the thin-disk, thick-disk, and halo populations are shown in light blue,
orange, and light green, respectively. Abundances for giants are taken from Li et al. (2013) (pentagons), Jönsson et al. (2014b)
(right-pointing arrows), Pilachowski & Pace (2015) (squares), Guerço et al. (2019) (triangles), Ryde et al. (2020) (left-pointing
arrows), and the Jönsson et al. (2017) data reanalyzed by Ryde et al. (2020) (diamonds) for [F/Fe] and from Alves-Brito et al.
(2010) (squares) for [Na/Fe].

Fluorine and sodium abundances have been deter- real and not observational. The dispersion is
mined for a sample of giants in the thin disk, thick disk, slightly larger in the thick disk than the thin.
and halo. We summarize our findings below.
• Finally, we compare our observations with Galac-
• Our sample covers a metallicity range from −0.70 tic chemical evolution models including different
< [Fe/H] < 0.00. The average fluorine abundance nucleosynthesis prescriptions for fluorine. Still, it
for our sample is [F/Fe] = +0.14 ± 0.07 (SEM) remains difficult to draw firm conclusions and fur-
with a standard deviation of 0.23. The derived flu- ther data might be needed in this context.
orine abundances are consistent with other studies 8. ACKNOWLEDGEMENTS
of fluorine in these populations.
We thank the referee for helpful comments that im-
• We report two robust fluorine measurements for proved this paper. We are grateful to the Kitt Peak
stars with [Fe/H] < 0.5 dex: [F/Fe] = +0.51 (HD National Observatory and particularly to Karen Butler,
44030) and [F/Fe] = +0.03 (HD 37171). Ken Hinkle, Dick Joyce, and Christian Soto for their
assistance during the observing runs. We also thank
• We identify two stars with high [F/Fe]: HD 9138 Rashad Givens for his contributions in the early phases
with [F/Fe] = +0.48 and HD 44030 with [F/Fe] = of the analysis of the data. C.A.P. acknowledges the gen-
+0.51. erosity of the Kirkwood Research Fund at Indiana Uni-
• The thin- and thick-disk average [F/Fe] for our versity. V.G. acknowledges financial support from INAF
sample of stars combined with the literature differ under the program “Giovani Astrofisiche ed Astrofisici
by 0.03 dex. di Eccellenza - IAF: Astrophysics Fellowships in Italy”
(Project: GalacticA, “Galactic Archaeology: recon-
• There is a larger dispersion in fluorine abundances structing the history of the Galaxy”). Z.G.M is partially
than sodium abundances despite both species hav- supported by a NASA ROSES-2020 Exoplanet Research
ing similar overall uncertainties due to atmo- Program Grant (20-XRP20 2-0125). This research has
spheric parameters, suggesting this dispersion is made use of the NASA Astrophysics Data System Bib-
10 Brady et al.

liographic Services, the HITRAN database operated by cosmos.esa.int/gaia), processed by the Gaia Data Pro-
the Center for Astrophysics, and the WEBDA database, cessing and Analysis Consortium (DPAC, https://www.
operated at the Department of Theoretical Physics and cosmos.esa.int/web/gaia/dpac/consortium). Funding
Astrophysics of the Masaryk University. Additionally, for the DPAC has been provided by national institu-
this research has made use of the SIMBAD database, tions, in particular the institutions participating in the
operated at CDS, Strasbourg, France. The research Gaia Multilateral Agreement.
shown here acknowledges use of the Hypatia Catalog
Database, an online compilation of stellar abundance
data as described in Hinkel et al. (2014), which was sup- Facilities: KPNO:2.1m (Phoenix)
ported by NASA’s Nexus for Exoplanet System Science Software: IRAF (Tody 1986, 1993), MOOG (Sneden
(NExSS) research coordination network and the Vander- 1973, V. 2019), matplotlib (Hunter 2007), numpy (van
bilt Initiative in Data-Intensive Astrophysics (VIDA). der Walt et al. 2011)
Finally, this work has made use of data from the Euro-
pean Space Agency (ESA) mission Gaia (https://www.

REFERENCES
Abia, C. Recio-Blanco, A., de Laverny, P., Cristallo, S., Famaey, B., Jorissen, A., Luri, X., et al. 2005, A&A, 430,
Domı́nguez, I., & Straniero, O. 2009, ApJ, 694, 971 165. doi:10.1051/0004-6361:20041272
Abia, C., Cunha, K., Cristallo, S., de Laverny, P, Famaey, B., Pourbaix, D., Frankowski, A., et al. 2009,
Domı́nguez, I., Eriksson, K., Gialanella, L., Hinkle, K., A&A, 498, 627. doi:10.1051/0004-6361/200810698
Imbriani, G., Recio-Blanco, A., Smith, V. V., Straniero, Gaia Collaboration, Prusti, T., de Bruijne, J. H. J., et al.
O., & Wahlin, R. 2010, ApJ, 715, L94 2016, A&A, 595, A1. doi:10.1051/0004-6361/
Abia, C., Cunha, K., Cristallo, S., et al. 2015, A&A, 581, Gaia Collaboration, Brown, A. G. A., Vallenari, A., et al.
A88. doi:10.1051/0004-6361/201526586 2018, A&A, 616, A1. doi:10.1051/0004-6361/201833051
Afşar, M., Sneden, C., & For, B.-Q. 2012, AJ, 144, 20 Gaia Collaboration, Vallenari, A., Brown, A. G. A., et al.
Alves-Brito, A., Meléndez, J., Asplund, M., et al. 2010, 2023, A&A, 674, A1. doi:10.1051/0004-6361/202243940
A&A, 513, A35. doi:10.1051/0004-6361/200913444 Gontcharov, G. A. 2006, Astronomy Letters, 32, 759.
Asplund, M., Grevesse, N., Sauval, A. J., & Scott, P. 2009, doi:10.1134/S1063773706110065
Goorvitch, D. 1994, ApJS, 95, 535.
ARA&A, 47, 481
Gratton, R. G. & Ortolani, S. 1984, A&A, 137, 6
Balachandran, S. C., Fekel, F. C., Henry, G. W., &
Grisoni, V., Spitoni, E., Matteucci, F., et al. 2017,
Uitenbroek, H. 2000, ApJ, 542, 978
MNRAS, 472, 3637. doi:10.1093/mnras/stx2201
Barbuy, B. & Erdelyi-Mendes 1989, A&A, 214, 239
Grisoni, V., Spitoni, E., & Matteucci, F. 2018, MNRAS,
Burris, D. L., Pilachowski, C. A., Armandroff, T. E., et al.
481, 2570. doi:10.1093/mnras/sty2444
2000, ApJ, 544, 302
Grisoni, V., Romano, D., Spitoni, E., et al. 2020, MNRAS,
Cottrell, P. L. & Sneden, C. 1986, A&A, 161, 314
498, 1252. doi:10.1093/mnras/staa2316
Cunha, K.; Smith, V. V.; Lambert, et al. 2003, AJ, 126,
Guerço, R., Cunha, K., Smith, V. V., et al. 2019, ApJ, 885,
1305
139. doi:10.3847/1538-4357/ab45f1
Cunha, K., Smith, V. V., Johnson, J. A., et al. 2015, ApJL, Guerço, R., Ramı́rez, S., Cunha, K., et al. 2022, ApJ, 929,
798, L41. doi:10.1088/2041-8205/798/2/L41 24. doi:10.3847/1538-4357/ac5c55
da Silva, L., Girardi, L., Pasquini, L., et al. 2006, A&A, Gustafsson B., Edvardsson B., Eriksson K., et al. 2008,
458, 609 A&A486, 951.
da Silva, R., Milone, A. de C., & Rocha-Pinto, H. J. 2015, Hall, D. N. B. & Noyes, R. W. 1969, ApJL, 4, 13
A&A, 580, A24 Hayes, C. R., Majewski, S. R., Shetrone, M., et al. 2018,
Decin, L. 2000, Catholique University of Leuven, Ph.D. ApJ, 852, 49
Thesis Hinkle, K. H. & Lambert, D. L. 1975, MNRAS, 170, 447.
Deka-Szymankiewicz, B., Niedzielski, A., Adamczyk, M., et doi:10.1093/mnras/170.3.447
al. 2018, A&A, 615, A31 Hinkle, K., Wallace, L, & Livingston, W. E. (ed.) 1995,
Ding, P.-J., Zhu, Z., & Liu, J.-C. 2019, Research in Infrared Atlas of the Arcturus Spectrum, 0.9-5.3 microns
Astronomy and Astrophysics, 19, 068 (San Francisco, CA: ASP)
 Fluorine Abundances in Local Stellar Populations 11

Hinkel, N. R., Timmes, F. X., Young, P. A., et al. 2014, AJ, Olive, K. A. & Vangioni, E. 2019, MNRAS, 490, 4307.
148, 54. doi:10.1088/0004-6256/148/3/54 doi:10.1093/mnras/stz2893
Hunter, J. D. 2007, Computing in Science and Engineering, Pakhomov, Yu. V. 2013, Astronomy Letters, 39, 54
9, 90. doi:10.1109/MCSE.2007.55 Pilachowski, C. A. & Pace, C. 2015, AJ, 150, 66.
Johnson, D. R. H., & Soderblom, D. R. 1987, AJ, 93, 864 doi:10.1088/0004-6256/150/3/66
Jönsson, H., Ryde, N., Harper, G. M., et al. 2014, A&A, Placco, V. M., Sneden, C., Roederer, I. U., et al. 2021,
564, A122. doi:10.1051/0004-6361/201423597 Research Notes of the American Astronomical Society, 5,
Jönsson, H., Ryde, N., Harper, G. M., et al. 2014, ApJL, 92. doi:10.3847/2515-5172/abf651
789, L41. doi:10.1088/2041-8205/789/2/L41 Prantzos, N., Abia, C., Limongi, M., et al. 2018, MNRAS,
Jönsson, H., Ryde, N., Spitoni, E., et al. 2017, ApJ, 835, 476, 3432. doi:10.1093/mnras/sty316
50. doi:10.3847/1538-4357/835/1/50 Prugniel, Ph., Vauglin, I’, & Koleva, M. 2011, A&A, 531,
Jönsson, H., Holtzman, J. A., Allende Prieto, C., et al. A165
2020, AJ, 160, 120. doi:10.3847/1538-3881/aba592 Ramı́rez, I. & Allende Prieto, C. 2011, ApJ, 743, 135.
Jorissen, A.; Smith, V. V.; & Lambert, D. L. 1992, A&A, doi:10.1088/0004-637X/743/2/135
261, 164 Ramı́rez, I., Allende Prieto, C., & Lambert, D. L. 2013,
Karataş, Y., Bilir, S., Eker, Z., et al. 2004, MNRAS, 349, ApJ, 764, 78. doi:10.1088/0004-637X/764/1/78
1069. doi:10.1111/j.1365-2966.2004.07588.x Rothman, L. S., Gorden I. E., Babikov, Y., et al. 2012,
Kobayashi, C., Izutani, N., Karakas, A. I., et al. 2011, JQSRT, 130, 4
ApJL, 739, L57 Ryde, N., Jönsson, H., Mace, G., et al. 2020, ApJ, 893, 37.
Kobayashi, C., Karakas, A. I., & Lugaro, M. 2020, ApJ, doi:10.3847/1538-4357/ab7eb1
900, 179. doi:10.3847/1538-4357/abae65 Sauval, A. J. & Tatum, J. B. 1984, ApJS, 56, 193.
Li, H. N., Ludwig, H.-G., Caffau, E., et al. 2013, ApJ, 765, doi:10.1086/190980
51 Schuler, S. C., Cuhna, K., Smith, V. V., et al. 2007, ApJ,
Li, C., Zhao, G., Zhai, M., et al. 2018, ApJ, 860, 53. 667, 81
doi:10.3847/1538-4357/aac50f Shetrone, M. D. 1996, AJ, 112, 1517
Lodders, K., Palme, H., & Gail, H.-P. 2009, LanB, 4B, 712. Smiljanic, R., Romano, D., Bragaglia, A., et al. 2016, A&A,
doi:10.1007/978-3-540-88055-4 34 589, A115
Lucatello, S.; Masseron, T.; Johnson, J. A.; Pignatari, M.; Smith, V. V. & Lambert, D. L. 1990, ApJS, 72, 387.
Herwig, F. 2011, ApJL, 729, 40 doi:10.1086/191421
Luck, R. E. & Challener, S. L. 1995, AJ, 110, 2968. Sneden, C. 1973, ApJ, 184, 839
doi:10.1086/117741 Soubiran, C., Jasniewicz, G., Chemin, L., et al. 2018, A&A,
Luck, R. E. & Heiter, U. 2007, AJ, 133, 2464 616, A7. doi:10.1051/0004-6361/201832795
Maiorca, E., Uitenbroek, H., Uttenthaler, S., et al. 2014, Spitoni, E., Matteucci, F., Jönsson, H., et al. 2018, A&A,
ApJ, 788, 149. doi:10.1088/0004-637X/788/2/149 612, A16. doi:10.1051/0004-6361/201732092
Massarotti, A., Latham, D. W., Stefanik, R. P., et al. 2008, Strassmeier, K. G., Carroll, T. A., Weber, M., et al. 2015,
AJ, 135, 209. doi:10.1088/0004-6256/135/1/209 A&A, 574, A31. doi:10.1051/0004-6361/201424130
Matteucci, F. 2021, A&A Rv, 29, 5. Tody, D. 1986, Proc. SPIE, 627, 733. doi:10.1117/12.968154
doi:10.1007/s00159-021-00133-8 Tody, D. 1993, Astronomical Data Analysis Software and
McWilliam, A. 1990, ApJS, 74, 1075 Systems II, 52, 173
Meléndez, J., Asplund, M., Alves-Brito, A., et al. 2008, van der Walt, S., Colbert, S. C., & Varoquaux, G. 2011,
A&A, 484, L21. doi:10.1051/0004-6361:200809398 Computing in Science and Engineering, 13, 22.
Meynet, G. & Arnould, M. 2000, A&A, 355, 176 doi:10.1109/MCSE.2011.37
Mishenina, T. V.; Soubiran, C.; Kovtyukh, V. V. 2004, van Leeuwen, F. 2007, A&A, 474, 653.
A&A, 418, 551 doi:10.1051/0004-6361:20078357
Mura-Guzmán, A., Yong, D., Abate, C., et al. 2020, Wallace, L., Hinkle, K., & Livingston, W. C. 2001, NSO
MNRAS, 498, 3549. doi:10.1093/mnras/staa2610 Technical Report #01-001; Tucson: National Solar
Nault, K. A. & Pilachowski, C. A. 2012, AJ, 146, 153 Observatory
Nomoto, K.; Kobayashi, C., & Tominaga, N. 2013, Womack, K. A., Vincenzo, F., Gibson, B. K., et al. 2023,
ARA&A, 51, 457 MNRAS, 518, 1543. doi:10.1093/mnras/stac3180
12 Brady et al.

Woosley, S. E. & Haxton, W. C. 1988, Nature, 334, 45. Wu, Y., Singh, H. P., Prugniel, P., Gupta, R., & Kolea, M.
doi:10.1038/334045a0 2011, A&A, 525, 71
 Fluorine Abundances in Local Stellar Populations 13

Figure 5. [F/Fe] vs. [Fe/H] for our sample of thin- and thick-disk stars with measurements from the literature. Chemical
evolution models from the literature are shown for different combinations of production channels. Literature measurements from
Li et al. (2013), Pilachowski & Pace (2015), Guerço et al. (2019), and Ryde et al. (2020) are shown in gray. Thin-disk stars are
indicated with open squares, thick-disk stars with filled triangles, and halo stars with crosses.
14 Brady et al.

Figure 6. [F/O] vs. [O/H] for our sample of thin- and thick-disk stars with measurements from the literature. Symbols are as
defined in Figure 5.