Upper limit on HF(1-0) absorption in a dusty star-forming galaxy at $z = 6$: Constraints on early fluorine enrichment

TL;DR

This study tests early fluorine production by massive stars by measuring HF(1-0) absorption in a highly star-forming, lensed galaxy at z=6.024 with known metallicity, using ALMA Band 5. It derives a 5σ upper limit of $N_ ext{HF}/N_{ ext{H}_2} < 2.2\times10^{-9}$ (log10 $< -8.7$), indicative of inefficient fluorine enrichment about 0.9 Gyr after the Big Bang and consistent with chemical-evolution models that exclude WR yields at this epoch. The result suggests WR contributions to fluorine were not yet significant at z≈6 and emphasizes the need for a larger sample of HF measurements in high-redshift galaxies with robust metallicity constraints to map the onset of fluorine production over cosmic time.

Abstract

Wolf-Rayet (WR) stars have recently attracted attention as possible drivers of early chemical enrichment, including the production of fluorine, whose nucleosynthetic origin remains debated. To test the contribution of massive stars to fluorine production in the early Universe, we conducted Atacama Large Millimeter/submillimeter Array Band 5 spectroscopy of the HF(1-0) absorption line toward a dusty star-forming galaxy at $z=6.024$. This galaxy has a known gas-phase metallicity and is too young for low-mass AGB stars to have contributed significantly, providing a clean environment to isolate massive-star yields. We do not detect significant HF absorption ($\sim2σ$) and derive a conservative 5$σ$ upper limit of $N_\mathrm{HF}/N_\mathrm{H_2} < 2.2\times10^{-9}$. This limit is about an order of magnitude below typical local measurements, indicating inefficient fluorine enrichment $\sim0.9$ Gyr after the Big Bang. Comparison with chemical evolution models shows that our constraint is consistent with scenarios without WR yields at this epoch. Expanding the sample of HF absorption measurements in high-redshift galaxies with well-characterized metallicities will be crucial for tracing the onset of WR enrichment and fluorine production across cosmic time.

Figures (4)

• Figure 1: ALMA continuum and velocity-integrated (moment-0) maps of G09.83808. From left to right: dust continuum at $\sim170$ GHz, HF(1--0) absorption, CO(10--9), and H$_2$O($2_{20}$--$2_{11}$) lines. White contours indicate the clean mask. Red crosses mark the two continuum peaks used for line spectrum extraction and photometry. The synthesized beam is shown as a white ellipse in the lower left of each panel. White scale bars indicate 1, which corresponds to 5.7 kpc. The colorbar units are mJy beam$^{-1}$ for the continuum and mJy beam$^{-1}$ km s$^{-1}$ for the lines.
• Figure 2: Continuum-subtracted spectra centered on the systemic velocity. The HF(1--0) absorption line (red) is shown with $1\sigma$ error bars and binned to 50 km s$^{-1}$. For comparison, the CO(10--9) (blue) and H$_2$O($2_{20}$--$2_{11}$) (orange) lines from the same dataset are overlaid. We note that the positive feature near $+800$ km s$^{-1}$ in the HF spectrum is the adjacent H$_2$O($2_{20}$--$2_{11}$) line.
• Figure 3: HF column density versus H$_2$ column density. Literature measurements of HF in absorption and emission (circles; Neufeld2010AA...518L.108NPhillips2010AA...518L.109PSonnentrucker2010AA...521L..12SMonje2011ApJ...734L..23MEmprechtinger2012ApJ...756..136EKamenetzky2012ApJ...753...70KIndriolo2013ApJ...764..188IPereira-Santaella2013ApJ...768...55PKawaguchi2016ApJ...822..115KLu2017ApJS..230....1LKavak2019AA...631A.117K) are color-coded by redshift, and the new measurement from this work is shown as a star. Arrows indicate upper and lower limits. The red dashed line denotes the fluorine abundance in the solar neighborhood ($\mathrm{F/H}=2.5\times 10^{-8}$; Maiorca2014ApJ...788..149M), and the black dotted line shows the typical $N_{\rm HF}/N_{\rm H_2}$ value found in Galactic diffuse molecular clouds ($(1-2)\times 10^{-8}$; e.g., Sonnentrucker2010AA...521L..12S). Three high-redshift sources are highlighted for comparison: the DSFG NGP-190387 at $z=4.4$Franco2021NatAs...5.1240F, the QSO BR1202$-$0725 at $z=4.7$ where HF is detected in emission Lehnert2020AA...641A.124L, and the QSO Cloverleaf at $z=2.56$Monje2011ApJ...742L..21M.
• Figure 4: Predicted evolution of the oxygen abundance A(O) (left) and fluorine abundance A(F) (right) for chemical–evolution models constrained to reproduce the observed metallicity of G09.83808 ($Z \approx 0.5$–$0.7\,Z_\odot$; Tadaki2022), shown as the gray band in the A(O) panel and its gas fraction ($f_{\rm gas} = 0.2$; Zavala2022). (Top) Fiducial models, all of which assume a Kroupa IMF with slope $x=1.3$. WR–included cases adopt a stellar rotation velocity of $v_{\rm rot}=300~{\rm km\,s^{-1}}$ except for the thin line. Blue and orange curves denote models with and without WR yields, respectively. The solid curves show the model with $\tau_{\rm SF}=0.37$ Gyr and $\tau_{\rm in}=1$ Gyr, while the dashed curves adopt $\tau_{\rm SF}=0.3$ Gyr. A slower–rotation WR model ($v_{\rm rot}=150~{\rm km\,s^{-1}}$, blue thin curve) is shown for comparison and differs only slightly from the $300~{\rm km\,s^{-1}}$ WR model. The vertical dotted line marks $z=6.024$. In the A(F) panel, the gray horizontal line shows the solar abundance Maiorca2014ApJ...788..149M, and the red star marks our $5\sigma$ upper limit. This limit is consistent with models without WR enrichment and with longer star-formation timescales. (Middle) Models with different star-formation and infall timescales. The bold curves show an alternative SFH ($\tau_{\rm SF}=0.18$ Gyr, $\tau_{\rm in}=0.5$ Gyr) that also satisfies the observed metallicity and gas–fraction constraints, while the faint curves are the fiducial models. To satisfy the constraints, the models yield nearly identical A(F) regardless of the adopted timescales. (Bottom) Models exploring variations in the IMF slope. The dotted lines adopt IMF slopes of $x=1.1$ (red) and $x=1.7$ (blue) while keeping all other parameters at the fiducial values; these models fail to reproduce the observed metallicity. The solid red and blue curves retune the star–formation timescale to $\tau_{\rm SF}=0.45$ and $0.32$ Gyr, respectively, so that the models match the observed metallicity, in which case the predicted A(F) remains similar to the fiducial case.