🔭 astrophysics

The birth of the intracluster medium: the evolution of multiphase gas and Lyman- $\alpha$ haloes in a simulated $z\sim3$ protocluster

This study utilizes a high-resolution cosmological simulation of a $z\sim3$ protocluster to demonstrate how major mergers and AGN feedback drive an inside-out transformation from a filamentary, multiphase circumgalactic medium to a hot, X-ray emitting intracluster medium, while analyzing the resulting evolution of metal absorption lines and the transition from filamentary to spherical Lyman- $\alpha$ haloes.

Original authors: Jake S. Bennett, Aaron Smith, Fabrizio Arrigoni-Battaia, Debora Sijacki, Cassandra Lochhaas, Lars Hernquist

Published 2026-07-17

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Jake S. Bennett, Aaron Smith, Fabrizio Arrigoni-Battaia, Debora Sijacki, Cassandra Lochhaas, Lars Hernquist

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 or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the universe as a giant, cosmic construction site. In the very beginning, galaxies were like lonely houses being built in a vast, empty field. They were fed by cold, narrow rivers of gas—called filaments—that flowed straight from the surrounding space into the galaxy's heart. This surrounding gas, known as the Circumgalactic Medium (CGM), was a messy, multi-layered soup: some parts were freezing cold and dense, while others were hot and thin, all swirling together.

But as these galaxies grew up and gathered into massive groups called clusters, the neighborhood changed. The cold rivers got cut off, and the whole area heated up into a uniform, super-hot, X-ray glowing fog called the Intracluster Medium (ICM). Think of it like a house that used to have a cozy, drafty porch with a cold breeze, but then got sealed up and filled with a giant, hot air balloon that pushed all the cold air out. Scientists have seen the "before" (cold rivers) and the "after" (hot fog), but they've been missing the middle chapter: exactly how and when that cold, messy gas gets cooked into a hot, smooth fog. Understanding this "birth of the ICM" is crucial because it tells us how the universe organizes itself and why galaxies stop forming stars as they get older.

This paper takes a deep dive into that missing middle chapter by running a super-detailed computer simulation of a massive galaxy cluster being born. The researchers, led by Jake Bennett, zoomed in on a single, growing protocluster (a baby cluster) as it evolved from a young age (about 1.5 billion years after the Big Bang) to a slightly older age (about 2.7 billion years after the Big Bang). They didn't just watch the gas move; they used advanced physics to simulate how light would bounce off and through that gas, creating "mock" images of what we would see if we pointed our most powerful telescopes at it.

Here is what their simulation revealed about the transformation:

The Great Heat Wave
The story starts with a chaotic mix. At the beginning, the cluster is fed by cold, dense filaments of gas, like icy streams pouring into a warm bath. But then, two big things happen: a massive collision with another galaxy group (a merger) and a powerful burst of energy from a supermassive black hole in the center (AGN feedback). Together, these act like a giant cosmic blowtorch. They heat the gas from the inside out. The cold, dense filaments get disrupted and destroyed, and the gas is pushed outward. By the end of the simulation, the inner part of the cluster has turned into a hot, smooth, X-ray emitting fog, while the cold gas is mostly squashed into the galaxies themselves or pushed to the very edges.

The Disappearing Act
As the gas heats up, the "signatures" of the cold gas vanish. The researchers tracked specific chemical elements, like Magnesium and Carbon, which usually hang out in the cold gas. They found that as the cluster grew, these elements were rapidly ionized (stripped of their electrons) and scattered. The "middle" amounts of these elements disappeared, leaving behind only the very densest clumps of gas, which are usually stuck inside the satellite galaxies. It's as if the cold gas was a foggy mist that got blown away, leaving only the heavy rocks (the galaxies) behind.

The Light Show: Lyman-alpha vs. H-alpha
The team also simulated how this gas would glow. They looked at two types of light: Lyman-alpha (Lyα), which is a very bright, tricky light that bounces around a lot (like a pinball), and H-alpha (Hα), which is a dimmer, cleaner light that travels straight.

Lyα: Even without the black hole helping, the simulation showed a huge, glowing halo of Lyα light. At first, this halo looked like a long, messy filament (the cold river). As the cluster heated up and the gas got disrupted, the halo became rounder and more spherical, looking like a glowing bubble. However, the simulation predicted that the very center of this glow was dimmer than what astronomers actually see in real life. This suggests that our current models might be missing something: either the dust in the center isn't as thick as we think, or the black hole is helping to create more light than we calculated.
Hα: The Hα haloes were much smaller and dimmer than the Lyα ones. The authors suggest that with new telescopes like the James Webb Space Telescope (JWST), looking at Hα could be a great way to see the "clean" version of the gas without the confusing bouncing effects of Lyα.

The Verdict
The paper doesn't claim to have solved the mystery with a single observation, but it provides a very detailed roadmap of what should happen during this transition. It suggests that the shift from a cold, filamentary environment to a hot, smooth one happens quickly (in cosmic terms) and is driven by a combination of galaxy crashes and black hole energy. It also highlights a gap in our understanding: we still need to figure out exactly why the centers of these glowing haloes are so bright in real life compared to our simulations. The authors conclude that to fully understand the "birth of the ICM," we need even more powerful simulations and new data from telescopes that can see these glowing haloes in different colors of light.

Technical Summary: The Birth of the Intracluster Medium: The Evolution of Multiphase Gas and Lyman-α Haloes in a Simulated z ∼3 Protocluster

Problem Statement
While the circumgalactic medium (CGM) of high-redshift galaxies is characterized by complex, multiphase structures fed by cold, filamentary inflows, mature low-redshift galaxy clusters are dominated by a hot, enriched, X-ray emitting intracluster medium (ICM) where cold gas is largely confined to member galaxies. The transition between these regimes—the "birth of the ICM"—remains poorly constrained observationally and theoretically. Existing cosmological simulations often lack the resolution to capture the detailed structure of the CGM at intermediate redshifts ( $z \sim 3$ ), a critical epoch where massive haloes grow rapidly and feedback mechanisms are expected to drive the phase transition. Furthermore, linking hydrodynamical evolution to observables requires robust modeling of radiative transfer, dust attenuation, and scattering, which are computationally expensive and often omitted.

Methodology
The authors present a cosmological zoom-in simulation of a massive protocluster progenitor, evolved from $z \approx 8$ to $z = 2.7$ , destined to become a $10^{15} M_\odot$ cluster at $z=0$ . The simulation utilizes the moving-mesh code Arepo with a specific focus on enhancing CGM resolution.

Resolution: The authors employ a shock-refinement scheme that increases resolution by a factor of 512 in baryonic cell mass compared to the fiducial Fable model, achieving an average cell radius of 200–300 pc in the CGM at $z \sim 3$ . This allows for the explicit tracking of cold gas structures that are typically under-resolved.
Physics: The simulation includes stellar feedback (hydrodynamically decoupled winds) and AGN feedback (quasar and radio modes). The UV background is active from $z \sim 10.6$ .
Post-Processing:
- Absorption: Ionization states for Mg II, C II, O VI, and O VII are calculated in post-processing using trident, assuming photoionization equilibrium with the metagalactic UV background and self-shielding.
- Emission: Radiative transfer for Lyman-α (Lyα) and Hα emission is performed using the colt (Cosmic Lyman-α Transfer) Monte Carlo code. This accounts for photoionization, dust attenuation (testing various dust-to-metal ratios), and resonant scattering. The models consider both stellar ionizing sources and a central AGN (scaled to Eddington luminosity).

Key Contributions and Results

Inside-Out Transformation of the CGM:
Between $z \sim 4.4$ and $2.7$, a major merger (mass ratio $\sim 1:2.5$ ) and powerful AGN feedback drive an "inside-out" transformation of the gaseous halo. The inner halo ( $\sim 50-250$ kpc) is heated from $\sim 10^7$ K to $> 3 \times 10^7$ K, entering the X-ray regime. This process flattens the density, temperature, and metallicity profiles, enriching the forming ICM to a near-uniform $\sim 0.3 Z_\odot$ .
Evolution of Gas Phases and Ion Tracers:
- Cold Gas Depletion: Intermediate column density cold gas (traced by Mg II and C II) is rapidly destroyed in the inner halo due to heating and disruption. The covering fraction of Mg II ( $N > 10^{11}$ cm $^{-2}$ ) drops from $\sim 70\%$ at $z > 6$ to $\sim 15\%$ at $z \sim 3$ in the 50–250 kpc range.
- Survival of Dense Clumps: The densest gas associated with satellite galaxies survives the transition, leading to a clumpier distribution of cold gas at lower redshifts.
- Hot Gas Ionization: As the halo heats, oxygen is ionized beyond O VI to O VII and higher states. The inner halo becomes dominated by highly ionized gas, while the O VII column density distribution shifts to lower values as the core is heated.
Lyman-α and Hα Halo Evolution:
- Morphology: The Lyα halo evolves from a filamentary structure tracing cold inflows to a more spherical, extended nebula as the inflows are disrupted. Extended Lyα emission remains detectable even without AGN photoionization.
- Resolution Dependence: Higher numerical resolution significantly boosts the inferred Lyα surface brightness (by a factor of $\sim 2.8$ ) compared to lower-resolution runs, due to the formation of more multiphase, dense structures.
- Central Emission Discrepancy: The fiducial model underpredicts the central Lyα surface brightness compared to observations. The authors attribute this to either an overestimation of dust attenuation in the unresolved ISM or an underestimation of Lyα production/escape efficiency in the nuclear region.
- AGN Impact: Including AGN ionization produces only a modest change in the extended Lyα profile, suggesting the halo is already highly ionized by stellar radiation. However, AGN feedback is likely necessary to create channels for ionizing photons to escape the central galaxy.
- Hα Comparison: Hα haloes are predicted to be significantly smaller and dimmer than Lyα haloes but offer a cleaner probe of the CGM as they are not affected by resonant scattering. The Lyα/Hα ratio is consistent with observations at low radii ( $\sim 3-5$ ) and approaches the Case B value ( $\sim 8.7$ ) at larger radii.

Significance and Claims
The paper claims to address a critical gap in understanding the formation of galaxy clusters by providing a highly resolved view of the CGM-to-ICM transition at $z \sim 3$ . The study demonstrates that this transition is a coupled process involving:

The disappearance of intermediate-column cold absorbers.
The growth and subsequent over-ionization of hot-phase tracers in the inner halo.
A morphological shift in Lyα emission from filamentary to clump-dominated.

The authors emphasize that their results highlight the sensitivity of Lyα observables to CGM resolution and the importance of sub-grid modeling for the ISM and dust. They conclude that while their case study suggests a specific evolutionary path for a massive protocluster, a larger suite of high-resolution simulations is required to determine if this "inside-out" heating is ubiquitous or contingent on specific merger and feedback histories. The work positions Hα observations with JWST as a complementary and potentially less ambiguous probe of the evolving CGM at this critical epoch.

More like this