🔭 astrophysics

Dust and PAHs in late-stage galaxy evolution: Imprints of TP-AGB dust injection, grain growth and AGN feedback in high-z quiescent galaxies with JWST and ALMA

This paper presents a semi-analytic model demonstrating that delayed dust injection from TP-AGB stars and ISM grain growth sustain diverse dust and PAH reservoirs in high-redshift quiescent galaxies for up to 2.5 Gyr after quenching, creating detectable signatures with JWST and ALMA that serve as independent probes of late-stage interstellar medium evolution rather than mere remnants of prior star formation.

Original authors: D. Donevski, A. Nanni, K. E. Whitaker, A. W. S. Man, A. Faisst, A. Lapi, I. García-Bernette, M. Romano, T. Petrushevska, G. Gururajan, G. Lorenzon, D. Narayanan

Published 2026-08-31✓ Author reviewed ⓘ

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: D. Donevski, A. Nanni, K. E. Whitaker, A. W. S. Man, A. Faisst, A. Lapi, I. García-Bernette, M. Romano, T. Petrushevska, G. Gururajan, G. Lorenzon, D. Narayanan

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Galaxies are not static islands of stars; they are living systems that evolve over billions of years. For a long time, astronomers believed that once a galaxy stopped forming new stars, it would simply fade into a quiet, dusty graveyard. In this old view, the cold gas and dust that once fueled star birth would vanish quickly, leaving behind a system of aging stars and empty space. However, recent observations have complicated this picture. We now know that many galaxies that have stopped making stars still hold onto surprising amounts of dust and gas, sometimes long after they should have run dry. The big question is: how does this material survive, and what processes are reshaping it in the silence after the stars stop forming?

To answer this, a team of researchers has built a new computer model called UNDUST. This model acts like a time machine for the invisible ingredients of galaxies. Instead of just tracking stars, it follows the life cycle of dust grains—tiny solid particles of carbon and rock—and a special type of molecule called a polycyclic aromatic hydrocarbon, or PAH. These PAHs are complex carbon structures that glow brightly when heated by starlight. The researchers wanted to see how these materials behave in "quiescent" galaxies, which are systems that have shut down their star formation. They focused on the period after the shutdown, asking whether the remaining dust simply disappears, or if new dust is being made and old dust is being destroyed by different forces.

The model simulates a galaxy that has just stopped making stars and tracks its evolution for several billion years. The researchers found that the story of dust in these quiet galaxies is far more complex than a simple fade-out. They discovered that even after star formation ends, the galaxy does not immediately lose its dust. Instead, two new sources begin to play a major role. First, aging stars, specifically a type known as thermally pulsing asymptotic giant branch stars, continue to pump fresh dust into the galaxy. Second, the remaining gas in the galaxy can act as a factory, allowing new dust grains to grow on existing seeds. These two processes can sustain the galaxy's dust supply for up to two and a half billion years after the stars have stopped forming.

However, the fate of this dust depends heavily on the environment inside the galaxy. The researchers found that if a galaxy is left to its own devices, it can maintain a rich reservoir of dust for a long time, with amounts varying wildly from one galaxy to another. But if the galaxy hosts an active supermassive black hole at its center, the story changes. The black hole can heat up the gas and blow it away, effectively shutting down the factories that grow new dust. In these cases, the dust supply drops much faster, often falling to less than ten percent of its original amount within just a few hundred million years. The model suggests that the presence of this black hole activity is a key reason why some quiet galaxies appear completely dust-free while others still hold onto significant amounts.

One of the most intriguing findings concerns the size and type of the dust particles. The model shows that the mix of dust changes over time. In the early stages after star formation stops, the galaxy is dominated by large dust grains. But as time goes on, violent collisions between these large grains break them apart, creating a surge of tiny grains and PAHs. This means that even if a galaxy has lost most of its heavy, cold dust, it can still be rich in these tiny, glowing carbon molecules. The researchers calculated that for up to two billion years, a quiet galaxy could still contain a few percent of its dust mass in the form of these PAHs, even if the total amount of dust has dropped so low that it is invisible to current radio telescopes.

This discovery has important consequences for how we observe the universe. The team realized that many quiet galaxies might be hiding in plain sight. They could be invisible to the powerful Atacama Large Millimeter/submillimeter Array (ALMA), which looks for cold dust, because their dust mass has fallen below the detection limit. Yet, these same galaxies might still glow brightly in the mid-infrared light that the James Webb Space Telescope (JWST) can see, thanks to the lingering PAHs. The model predicts that there is a specific window of time, lasting about one to two billion years after a galaxy stops forming stars, where it becomes "ALMA-faint" but remains "JWST-bright." This suggests that many of the dusty galaxies we see with JWST might be in this transitional phase, chemically enriched but no longer capable of making new stars.

The study also challenges the idea that we can understand a galaxy's history just by looking at its gas or its stars. The researchers showed that two galaxies with the same amount of gas and the same age could have very different amounts of dust, depending on whether they experienced black hole activity or how efficiently their internal dust factories worked. This means that dust and PAHs provide a unique, independent way to trace the history of a galaxy's evolution. They act as a record of the specific processes that happened after the stars died, revealing whether the galaxy was gently fading away or being violently stripped by a black hole.

Ultimately, this work paints a picture of galaxy evolution that is dynamic and diverse. It shows that the end of star formation is not a clean break but a long, complex process where different physical forces compete. Some galaxies manage to recycle their dust for billions of years, while others are stripped clean quickly. The researchers emphasize that their findings are based on simulations that isolate these specific processes, offering a clear framework to test against real observations. As telescopes like JWST and ALMA continue to peer deeper into the universe, they will be able to test these predictions, looking for the specific signatures of these fading dust reservoirs and the hidden PAHs that tell the story of a galaxy's final, quiet chapters.

Technical Summary: Dust and PAHs in Late-Stage Galaxy Evolution

Problem Statement
A central uncertainty in late-stage galaxy evolution concerns the regulation of the cold interstellar medium (ISM) following the quenching of star formation. While quiescent galaxies (QGs) are traditionally viewed as systems with depleted molecular gas and dust, recent observations by ALMA and JWST reveal a heterogeneous population extending to $z \sim 7$ . These observations show a wide scatter in dust-to-gas ratios ( $\delta_{\text{DGR}}$ ) and dust-to-stellar mass ratios ( $f_{\text{dust}}$ ), ranging from star-forming-like values to highly depleted states. Current theoretical frameworks struggle to isolate the specific physical mechanisms governing dust replenishment, destruction, and processing in the post-quenching phase, particularly given the simultaneous action of multiple channels such as stellar mass return, grain growth, and AGN feedback. Furthermore, there is a lack of physically motivated diagnostics to distinguish between rapid dust destruction and prolonged survival in these systems.

Methodology: The UNDUST Framework
To address these gaps, the authors present UNDUST (UNified model of DUST evolution across galaxy stages), the first semi-analytic framework specifically designed to couple the post-quenching evolution of cold gas, metals, dust, and polycyclic aromatic hydrocarbons (PAHs) in a one-zone approximation.

Key methodological features include:

Two-Size Grain Model: Dust is divided into small ( $a < 0.03$ $\mu$ m) and large ( $a > 0.03$ $\mu$ m) reservoirs, with PAHs treated as a sub-component of the small carbonaceous grains.
Coupled Evolution Channels: The model tracks four distinct processes:
1. Remnant Dust: The inherited reservoir from the star-forming progenitor, subject to destruction.
2. Delayed Injection: Dust production by thermally pulsing asymptotic giant branch (TP-AGB) stars, computed using yield tables convolved with the star-formation history (SFH).
3. ISM Grain Growth: Accretion of gas-phase metals onto existing grains, dependent on cold gas density and temperature.
4. Destruction & Processing: Thermal sputtering, supernova (SN) shocks, shattering (large to small grains), coagulation (small to large grains), and astration.
AGN Feedback: Implemented phenomenologically to test its impact on the cold gas phase. AGN activity is modeled as enhancing gas outflows (mass-loading factor $\eta_{\text{out}}$ ) and heating the ISM, which suppresses grain growth efficiency without directly destroying grains.
Initial Conditions: Models are initialized at the onset of quenching ( $t_q=0$ ) with empirically motivated parameters for massive QGs ( $M_\star \sim 8 \times 10^{10} M_\odot$ , $f_{\text{gas}} \sim 0.08$ , $\delta_{\text{DGR}} \sim 1/250$ ). The framework explores mild, moderate, and harsh destruction regimes, with and without AGN feedback.

Key Results

Diversity in Dust States: At $z \sim 1$ , QGs with similar stellar mass and age span 2–3 dex in $M_{\text{dust}}/M_\star$ and $\delta_{\text{DGR}}$ . This diversity arises not solely from the initial conditions but from the interplay between delayed TP-AGB injection and ISM grain growth, which can sustain dust enrichment for up to $\sim 2.5$ Gyr after quenching. Without these channels, pre-quenching dust levels would drop below 10% of their initial value within $\lesssim 0.5–1$ Gyr.
Decoupling of Dust and Gas: The model demonstrates that dust evolution can decouple from cold gas depletion. Even when the cold gas fraction declines significantly, dust replenishment via grain growth and TP-AGB injection can maintain dust reservoirs. Consequently, $f_{\text{dust}}$ and $\delta_{\text{DGR}}$ are not uniquely determined by the remaining gas mass or the time since quenching.
Role of AGN Feedback: AGN feedback accelerates the decline of dust content primarily by suppressing the conditions necessary for grain re-accretion (via gas removal and heating) rather than through direct grain destruction. In AGN-active scenarios, $f_{\text{dust}}$ drops below $10^{-4}$ within $\lesssim 1$ Gyr, roughly twice as fast as in AGN-off cases.
Grain Size and PAH Evolution:
- Carbonaceous vs. Silicate: The small-grain mass fraction ( $f_{\text{small}}$ ) exhibits non-monotonic evolution. A temporary rise in small carbonaceous grains occurs at $t_q \sim 1.4–2.3$ Gyr due to the shattering of large TP-AGB carbon grains, whereas silicate grains remain dominated by large-grain injection.
- PAH Survival: PAH fractions ( $q_{\text{PAH}}$ ) can reach $\sim 2–3\%$ in the first $\sim 1$ Gyr post-quenching, matching values seen in metal-rich star-forming galaxies, even as the total cold dust budget declines. This is driven by the shattering of large carbon grains.
Observational Implications (ALMA vs. JWST):
- ALMA Faintness: Galaxies can fall below typical ALMA continuum detection limits ( $M_{\text{dust}}/M_\star \lesssim 10^{-4}$ ) while still retaining a chemically enriched dust phase.
- JWST/MIRI Detectability: Despite being ALMA-faint, these systems may remain detectable in deep JWST/MIRI imaging (e.g., F1500W) due to PAH emission and residual dust continuum. This creates a "JWST MIRI" window where galaxies are observable in the mid-infrared but faint in the sub-millimeter, lasting $\sim 1–2$ Gyr post-quenching in mild/moderate regimes.
- Diagnostic Power: Joint constraints on $f_{\text{dust}}$ , $\delta_{\text{DGR}}$ , $q_{\text{PAH}}$ , and $t_q$ are required to break degeneracies and identify the dominant ISM processing mechanisms.

Significance and Claims
The paper claims that dust and PAHs serve as independent probes of distinct stages of ISM evolution in quiescent galaxies, rather than simply tracing the residual cold ISM of the preceding star-forming phase. The authors emphasize that:

Non-Unique Evolution: Similar observational locations in the dust-gas plane can arise from different evolutionary routes (e.g., efficient destruction without AGN vs. moderate destruction with AGN), necessitating multi-wavelength, multi-phase observations.
Reformulation of QG Evolution: The presence of dust and PAHs in QGs cannot be attributed solely to passive aging or offsets from scaling relations; active post-quenching processes (TP-AGB injection and grain growth) substantially modify the dust budget.
Observational Strategy: ALMA non-detections should not be interpreted as evidence for chemically depleted, dust-free galaxies. Instead, they may indicate a phase where cold dust is faint but PAH-bearing reservoirs are accessible to JWST/MIRI.
Laboratory for Dust Physics: QGs provide a unique environment to study fundamental dust physics (shattering, coagulation, sputtering) with reduced degeneracies from ongoing star formation.

The study concludes that future joint observations with ALMA, JWST, and upcoming FIR facilities are essential to constrain the duration of chemically enriched dust processing after star formation ceases and to distinguish among the proposed dust-evolution scenarios.

More like this