🔭 astrophysics

Probing the Hot Gaseous Halos of Milky Way-like Galaxies in the TNG50 simulation

By comparing synthetic X-ray and oxygen absorption data from TNG50 simulations with observations, the study finds that while the model reproduces inner-halo properties, it fails to match extended X-ray profiles and O VIII absorption, suggesting that the simulation's feedback mechanism deposits energy too centrally and lacks sufficient hot-phase gas to sustain a realistic, extended corona.

Original authors: Zhijie Zhang, Xiaoxia Zhang, Taotao Fang, Hui Li, Greg L. Bryan, Federico Marinacci, Paul Torrey, Mark Vogelsberger, Junfeng Wang, Haiguang Xu, Qingzheng Yu, Feng Yuan

Published 2026-04-28

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Zhijie Zhang, Xiaoxia Zhang, Taotao Fang, Hui Li, Greg L. Bryan, Federico Marinacci, Paul Torrey, Mark Vogelsberger, Junfeng Wang, Haiguang Xu, Qingzheng Yu, Feng Yuan

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

The Tale of the "Too-Small, Too-Hot" Galaxy Halo

Imagine you are looking at a glowing, cosmic campfire. In the vast darkness of space, galaxies like our Milky Way aren't just lonely islands of stars; they are surrounded by a massive, invisible "atmosphere" of incredibly hot gas called a halo. This halo is like the warm air surrounding a campfire—it’s not the fire itself, but it’s a vital part of the environment that tells you how the fire is burning.

Scientists use massive supercomputers to create "digital universes" to see if they can recreate this cosmic campfire. In this paper, a team of researchers tested one of the most advanced digital universes ever made, called TNG50, to see if it could correctly simulate the Milky Way’s hot atmosphere.

Here is the breakdown of what they found, using a few simple analogies.

1. The Success: The "Right Amount of Smoke"

First, the good news. When the researchers looked at the "smoke" (the X-ray light) coming from the center of the digital galaxies, it looked great!

The Analogy: Imagine you are trying to recreate a specific brand of campfire. You check the brightness of the glow and the amount of smoke rising from the center. The TNG50 simulation got the brightness and the density of the smoke almost exactly right. It proved that the simulation knows how to "light the fire" and create the basic ingredients of a galaxy.

2. The First Problem: The "Shrinking Campfire"

However, when the scientists looked at the shape of the glow, things went wrong. In the real universe, the hot gas seems to spread out very far, like a wide, gentle warmth that reaches far into the dark woods. But in the TNG50 simulation, the gas is too "clumped up." It stays too close to the center.

The Analogy: It’s like you built a campfire that is incredibly bright and intense right at the logs, but the warmth disappears almost immediately. Instead of a wide, cozy circle of warmth where you can sit comfortably, the heat is trapped in a tiny, intense ball. The simulation's "atmosphere" is too compact; it lacks the long-reaching, gentle glow we see in real life.

3. The Second Problem: The "Missing Temperature"

The researchers also looked at specific "fingerprints" left by different temperatures of gas (using things called Oxygen ions). They found that while the simulation was good at making "warm" gas, it was missing a specific type of "very hot" gas.

The Analogy: Imagine you are making a soup. You get the warmth of the broth right, and you get the medium-heat spices right, but you completely forgot the boiling-hot chili oil that should be floating on top. The simulation has the "warm" part of the atmosphere, but it’s missing that specific "extra-hot" layer that real galaxies seem to have.

The "Why": The Feedback Problem

Why is the simulation failing? The scientists point to something called AGN Feedback.

In a galaxy, there is a supermassive black hole at the center. This black hole acts like a cosmic leaf blower. It shoots out massive jets of energy that are supposed to push the gas around, preventing it from all falling into the center and helping it spread out into a beautiful, wide halo.

The Metaphor: In the TNG50 simulation, the "leaf blower" (the black hole) is working, but it’s acting more like a pressure washer. Instead of gently blowing the leaves (the gas) into a wide, even pile around the yard, it’s blasting them so hard and so centrally that it just cooks them in place or keeps them trapped in a tight, high-pressure zone. It’s too violent and too focused. It’s "overcooking" the center and failing to "spread the warmth" to the edges.

The Bottom Line

The TNG50 simulation is a masterpiece of modern science, but this paper shows it isn't perfect. It tells us that our current "digital recipes" for how black holes interact with their galaxies are a little too aggressive. To truly simulate a galaxy like our Milky Way, we need to learn how to make the cosmic "leaf blower" act more like a gentle breeze, spreading the heat far and wide.

1. Problem Statement

The origin and thermodynamic structure of the hot ( $T \gtrsim 10^6$ K) gaseous corona (or Circumgalactic Medium, CGM) surrounding Milky Way (MW)-mass galaxies are fundamental to understanding galaxy evolution. This corona is shaped by the competition between cosmological gas accretion (heating via shocks) and internal feedback (energy injection from stars and Active Galactic Nuclei, AGN). While theoretical models predict these extended halos, observational data—ranging from local Milky Way absorption lines to large-scale X-ray stacking—remain difficult to reconcile with current hydrodynamical simulations. The central question is whether state-of-the-art cosmological simulations can accurately reproduce the spatial distribution and multi-phase temperature structure of this hot gas.

2. Methodology

The authors utilize the TNG50 simulation, the highest-resolution component of the IllustrisTNG project, which includes magneto-hydrodynamics and sophisticated stellar/AGN feedback models.

Sample Selection: They identified a sample of 32 Milky Way analogues based on specific criteria:
- Stellar mass: $10^{10.5} < M_*/M_\odot < 10^{10.9}$ .
- Halo mass: $M_{200,c} < 10^{13} M_\odot$ .
- Star Formation Rate (SFR): $1 < \text{SFR}/(M_\odot \text{yr}^{-1}) < 3$ (to match the MW's current evolutionary state).
Synthetic Observables: To ensure a "like-with-like" comparison with real telescopes, the authors generated synthetic data:
- X-ray Emission: They calculated soft X-ray (0.5–2.0 keV) emissivity assuming Collisional Ionization Equilibrium (CIE). They adopted two viewpoints: an internal (Solar) viewpoint (placing an observer at 8.2 kpc in the disk) and an external viewpoint (randomly oriented lines of sight for radial profiling).
- Absorption Lines: They simulated O VII and O VIII K $\alpha$ absorption profiles, accounting for thermal broadening, Hubble expansion, and peculiar velocities, mimicking the use of background AGN to probe the MW halo.

3. Key Results

The study reveals a nuanced picture where the simulation succeeds in global energetics but fails in spatial and thermal distribution.

Successes (Inner Halo & Global Scales):
- Luminosity: The simulated soft X-ray luminosity ( $L_X$ ) aligns well with the observed $L_X$ –SFR relation and the specific luminosity of the Milky Way.
- Density & Temperature: The Emission Measure (EM) and emission-weighted temperature ( $T_{ew}$ ) along inner-halo sightlines are consistent with observations, suggesting the simulation correctly predicts the total hot gas mass and characteristic temperature in the central regions.
- O VII Absorption: The strength of O VII absorption (tracing $T \sim 10^6$ K gas) is largely consistent with observational data.
Discrepancies (Outer Halo & Hotter Phases):
- Spatial Compactness: The X-ray surface brightness profiles decline too steeply at large radii. While consistent within $\sim 50$ kpc, the simulated emission falls up to 1 dex below eROSITA stacking observations at $R \gtrsim 100$ kpc. This indicates the simulated halos are too centrally concentrated.
- Thermal Deficit (O VIII): There is a significant underproduction of O VIII absorption. The median equivalent width (EW) is $\sim 65\%$ lower than observed, indicating a deficit of gas in the $T \approx (1.6\text{--}3.2) \times 10^6$ K range.

4. Significance and Implications

The findings provide a critical diagnostic for the implementation of AGN feedback in cosmological simulations.

Feedback Physics: The authors suggest that the "radio mode" (kinetic) AGN feedback in TNG50 may be too "violent" or centrally concentrated. Instead of gently lifting gas to create an extended, diffuse corona, the energy might be depositing too much heat in the center, potentially pushing gas into even higher temperature regimes (beyond the O VIII window) while failing to redistribute it to the outer halo.
Model Refinement: The study highlights that reproducing the mass of the hot halo is not sufficient; simulations must also reproduce its spatial extent and multi-phase temperature structure. This provides a roadmap for future iterations of hydrodynamical codes to refine how energy is coupled to the CGM to sustain a large-scale, multi-phase atmosphere.