REBELS-IFU: Dust Build-up in Massive Galaxies at Redshift 7

TL;DR

The paper investigates dust build-up in 12 REBELS-IFU galaxies at z ~ 6.5–7.7 by jointly analyzing JWST-based metallicities, ALMA [C II]-based gas masses, and dust masses. It quantifies dust-to-gas (DtG), dust-to-metal (DtM), and dust-to-stellar mass (DtS) ratios across metallicity and compares them to leading high-redshift dust evolution models. The results indicate a high DtG (~ -3.02 in log units), DtM around log ∼ -0.8, and DtS around log ∼ -2.15, implying a substantial fraction of metals locked in dust and a rapid early dust assembly. The interpretation favors a combination of rapid SN enrichment and efficient ISM dust growth, though substantial systematics in dust temperatures and the [C II]-to-H2 conversion remain. The study demonstrates that dust build-up in massive z ~ 7 galaxies is already highly efficient and that future multi-band ALMA and kinematic [C II] data are needed to refine the gas budget and metallicity measurements.

Abstract

In recent years, observations with the JWST have started to map out the rapid metal enrichment of the early Universe, while (sub)millimeter observations have simultaneously begun to reveal the ubiquity of dust beyond $z\gtrsim6$. However, the pathways that led to the assembly of early dust reservoirs remain poorly quantified, and require pushing our understanding of key scaling relations between dust, gas and metals into the early Universe. We investigate the dust build-up in twelve $6.5 \lesssim z \lesssim 7.7$ galaxies drawn from the REBELS survey that benefit from (i) JWST/NIRSpec strong-line metallicity measurements, (ii) ALMA [CII]-based redshifts and gas masses, and (iii) dust masses from single- or multi-band ALMA continuum observations. Combining these measurements, we investigate the dust-to-gas (DtG), dust-to-metal (DtM), and dust-to-stellar mass (DtS) ratios of our sample as a function of metallicity. While our analysis is limited by systematic uncertainties related to the [CII]-to-H$_2$ conversion factor and dust temperature, we explore a wide range of possible values, and carefully assess their impact on our results. Under a fiducial set of assumptions, we find an average $\log(\mathrm{DtG}) = -3.02 \pm 0.23$, only slightly below that of local metal-rich galaxies. On the other hand, at fixed metallicity our average $\log(\mathrm{DtS}) = -2.15 \pm 0.42$ is significantly larger than that of low-redshift galaxies. Finally, through a comparison to various theoretical models of high-redshift dust production, we find that assembling the dust reservoirs in massive galaxies at $z\approx7$ likely requires the combination of rapid supernova enrichment and efficient ISM dust growth.

Figures (6)

• Figure 1: The physical properties of the 12 targets with combined JWST + ALMA observations studied in this work. Left: dust mass vs. redshift. We show the two galaxies for which we have multi-band dust mass estimates -- REBELS-25 and REBELS-38 algera2024algera2024b -- twice; once using the measured dust mass (large symbols with thick outline), and once assuming the same fiducial $T_\mathrm{dust} = 45 \pm 15\,\mathrm{K}$ as for the rest of the sample (fainter symbols; offset in redshift by $\Delta z = 0.01$ for clarity). The two REBELS-IFU targets without a continuum detection are shown as empty markers. For these, only upper limits on $M_\mathrm{dust}$ are available. Right: [C ii]-based molecular gas mass vs. metallicity. The gas mass is inferred using the calibration from zanella2018, who determined a mass-to-light ratio of $\alpha_\text{[C\,{\sc ii}]{}} = 31 M_\odot\,L_\odot^{-1}$. The solar metallicity [$12 + \log(\mathrm{O/H})_\odot = 8.69$] is indicated via the vertical dashed line. Our $6.5 \lesssim z \lesssim 7.7$ sample spans $\sim1\,\mathrm{dex}$ in dust mass, gas mass and metallicity.
• Figure 2: The effect of the adopted $\alpha_\text{[C\,{\sc ii}]{}}$ and $T_\mathrm{dust}$ on the inferred dust-to-gas ratio of a hypothetical galaxy at the median redshift ($z=6.8$) and [C ii] luminosity [$\log(L_\text{[C\,{\sc ii}]{}}/L_\odot) = 8.9$] of our sample. The thickest contour corresponds to $\log(\mathrm{DtG}/\mathrm{DtG}_0) = 0$, where the fiducial $\mathrm{DtG}_0$ is represented by the white star and assumes $T_\mathrm{dust} = 45\,$K and the [C ii]-to-H$_2$ conversion factor from zanella2018. The white, shaded region shows the uncertainties adopted on the dust temperature ($\pm15\,\mathrm{K}$) and $\alpha_\text{[C\,{\sc ii}]{}}$ (approximately $\pm0.3\,\mathrm{dex}$). Increasingly thinner contours show offsets of $\pm0.5, 1.0, 1.5\,\mathrm{dex}$ with respect to the fiducial DtG ratio. Various $\alpha_\text{[C\,{\sc ii}]{}}$ determinations from the literature are furthermore overplotted madden2020rizzo2021vizgan2022casavecchia2024kaasinen2024ramambason2024.
• Figure 3: Gas fraction as a function of metallicity for the REBELS-IFU sample (orange), six $z>6$ literature galaxies with [C ii] observations and metallicity measurements (light blue; Appendix \ref{['app:literature']}) and local galaxies from devis2019. We overlay the model tracks from palla2024, both using their fiducial model, and a closed-box model. The latter, lacking in- or outflows, provides the theoretical maximum gas fraction for a given metallicity. The width of the model tracks represents the variation in $f_\mathrm{gas}$ based on the adopted star-formation history. Moreover, we overlay tracks from the analytical model by erb2008, whereby $f_i$ ($f_0$) represents the inflow (outflow) rate in units of the SFR. The effect of in- and outflows on the gas fraction and metallicity is shown schematically by the arrows in the lower left corner, under the simplifying assumption that outflows deplete molecular gas and metals equally. At low metallicities [$12+\log(\mathrm{O/H})\sim 7.8 \sim 0.1\,Z_\odot$], the REBELS and literature samples show lower gas fractions than the model predictions, while at high metallicities [$12+\log(\mathrm{O/H})\gtrsim 8.5 \sim 0.6\,Z_\odot$] their gas fractions match or even exceed the closed box model.
• Figure 4: The dust-to-gas (DtG; top panel), dust-to-metal (DtM; middle) and dust-to-stellar mass (DtS; bottom) ratios as a function of gas-phase metallicity, drawn from a variety of low-, intermediate- and high-redshift observational studies. The REBELS-IFU sources, as well as additional high-redshift galaxies with constraints on their dust and metal contents (Appendix \ref{['app:literature']}), are shown in orange and light blue, respectively. Galaxies with filled markers are dust-detected in either a single band (narrow black outline) or multiple bands (thick black outline and larger marker). Open markers represent galaxies without a dust continuum detection. At low and intermediate redshifts, we show either individual galaxies devis2019peroux2020, or average trends remyruyer2014galliano2021. The REBELS-IFU galaxies, at relatively high metallicities of $Z\approx0.1-1.1\,Z_\odot$, show similar or slightly lower DtG and DtM ratios compared to local galaxies. Conversely, their DtS ratios are larger than those for galaxies at $z\approx0$.
• Figure 5: The dust-to-gas (upper), dust-to-metal (middle) and dust-to-stellar mass (bottom) ratios as a function of gas-phase metallicity. Similar to Fig. \ref{['fig:DustBuildupObservations']}, we overplot the twelve REBELS-IFU sources and other $z\gtrsim6$ galaxies in the literature. Moreover, we compare to a variety of studies modeling the production and evolution of dust at $z\approx7$popping2017vijayan2019triani2020dayal2022palla2024. Overall, our measurements are best reproduced by models predicting rapid dust build-up, resulting in high DtG, DtM and DtS ratios across the metallicity range spanned by the REBELS-IFU sample.