The Cosmic Baryon Cycle in IllustrisTNG: flows of mass, energy, and metals

This study analyzes mass, energy, and metal flows in the IllustrisTNG100 simulations to demonstrate how the transition from supernova-dominated to kinetic AGN-dominated feedback regulates galaxy growth and quenching by balancing inflows and outflows across different halo masses and redshifts.

The Gist

Imagine the universe as a giant, bustling city where galaxies are the buildings. For a building to grow tall and impressive (like a massive galaxy), it needs a steady supply of materials: gas, energy, and metals (the "bricks" and "cement" of stars). But just like a construction site, there are also forces trying to blow the building apart or stop new materials from arriving.

This paper is a detailed investigation into the traffic flow of these materials in the "IllustrisTNG" universe—a super-computer simulation that acts like a time machine, showing us how galaxies have evolved from the dawn of time until today.

Here is the story of the Cosmic Baryon Cycle, explained simply:

1. The Two Main Construction Crews (Feedback)

In this cosmic city, there are two main crews that control the flow of materials:

• The Supernova Crew (Stars): When massive stars die, they explode like fireworks. These explosions (supernovae) push gas out of the galaxy. Think of this as a construction crew using leaf blowers to clear away dust and debris. In small galaxies, this is the main force.
• The Black Hole Crew (AGN): In the center of massive galaxies, there are supermassive black holes. When they eat gas, they sometimes get "full" and switch modes.
  • Thermal Mode: They heat up the gas like a giant space heater, making it too hot to cool down and form stars.
  • Kinetic Mode: They shoot out powerful jets of energy, acting like a high-pressure fire hose that blasts gas far away. This is the "heavy machinery" that stops massive galaxies from growing too big.

2. The Two Gates: The Neighborhood and the City Limits

The researchers looked at the flow of materials at two specific locations:

• The ISM Gate (The Neighborhood): This is right around the galaxy itself, where stars are born. It's like the construction site fence.
• The Halo Gate (The City Limits): This is far out, at the edge of the galaxy's gravitational influence. It's like the city border where materials enter or leave the entire metropolitan area.

3. The Traffic Patterns: What They Found

The Early Universe (High Redshift): The "Inflow Rush"

In the early days of the universe (when the cosmos was young and hot), the traffic was almost entirely inbound.

• The Analogy: Imagine a new city being built. Everyone is rushing in to bring bricks. Even though the Supernova Crew was blowing some gas out, the gravity of the dark matter halo was so strong that it sucked in fresh gas from the intergalactic medium faster than the explosions could push it out.
• Result: Galaxies were growing rapidly, feeding on a steady stream of fresh fuel.

The Transition: The "Great Balance"

As the universe aged, things changed.

• The Analogy: The city grew up. The Black Hole Crew (specifically the "Kinetic Mode" fire hoses) turned on in the massive galaxies.
• Result: The outflow of gas finally matched the inflow. The "fire hoses" blew away as much gas as the gravity pulled in. This is why massive galaxies stopped growing and went into "retirement" (a state astronomers call "quenching"). They ran out of fuel because the Black Hole Crew kicked it all out.

The Recycling Loop: "Galactic Fountains"

One of the coolest discoveries is what happens to the gas that gets blown out.

• The Analogy: In smaller galaxies, the gas blown out by supernovae doesn't always escape the city limits. It's like a fountain. The water shoots up, cools down in the air, and falls back down into the pool.
• Result: This "recycled gas" falls back onto the galaxy, providing fresh fuel for new stars. The paper found that in smaller galaxies, this recycling is very efficient.

4. The "Loading Factors": How Heavy is the Load?

The researchers measured "loading factors," which is a fancy way of asking: "For every star we make, how much gas do we blow away?"

• Small Galaxies: They blow away a lot of gas relative to their size (high loading). It's like a small kid trying to push a heavy cart; they have to work very hard to move even a little bit.
• Massive Galaxies: They blow away less gas relative to their size (low loading). They are like a giant truck; it's easier to move the load.
• Energy vs. Mass: Interestingly, while the amount of gas blown away changes a lot, the energy of that wind stays surprisingly constant. It's like the wind always blows with the same force, but sometimes it carries a heavy load of sand (gas) and sometimes just a light mist.

5. The "Metal" Factor (Pollution vs. Enrichment)

Stars create heavy elements (metals) which are essential for life.

• The Outflow: When gas leaves the galaxy, it carries these metals with it.
• The Inflow: When fresh gas enters, it's usually "pristine" (clean, with few metals).
• The Surprise: The researchers found that the gas falling back into the galaxy (the fountain) is often more metal-rich than the fresh gas coming from deep space. This means the galaxy is constantly reusing its own "pollution" to build new stars.

The Big Picture Takeaway

This paper tells us that galaxy growth isn't just about gravity pulling things in. It's a delicate tug-of-war:

1. Gravity tries to pull gas in to make stars.
2. Supernovae (in small galaxies) and Black Holes (in big galaxies) try to push gas out to stop the party.

In the early universe, gravity won, and galaxies grew fast. In the modern universe, the Black Holes in massive galaxies have learned to turn on the "fire hoses" just enough to balance the scales, stopping the galaxy from growing any larger. This balance is what creates the diverse population of galaxies we see today: some small and starry, others massive and quiet.

The authors hope that by understanding these traffic patterns, we can build better "blueprints" (models) for how the universe works, helping us predict how galaxies will behave in the future.

Technical Summary

Here is a detailed technical summary of the paper "The Cosmic Baryon Cycle in IllustrisTNG: Flows of Mass, Energy, and Metals."

1. Problem Statement

Understanding galaxy evolution within the $\Lambda$CDM framework requires explaining the "baryon conversion efficiency"—the low fraction of baryons in halos that end up in stars. Observations show this efficiency peaks at $\sim20\%$ and drops significantly in low- and high-mass halos. This discrepancy implies that feedback mechanisms (Supernovae [SNe] and Active Galactic Nuclei [AGN]) must either prevent gas from cooling/accreting (preventative feedback) or eject it from the interstellar medium (ISM) and circumgalactic medium (CGM) (ejective feedback).

While numerical simulations like IllustrisTNG reproduce global galaxy properties (e.g., stellar mass functions), the specific physical pathways of the baryon cycle—how mass, energy, and metals flow between reservoirs (IGM, CGM, ISM) across different scales and redshifts—remain poorly quantified. Furthermore, it is unclear how these flows differ based on the dominant feedback mechanism (SN vs. AGN) and how they relate to the subgrid physics implemented in the simulation.

2. Methodology

The authors analyzed the TNG100-1 simulation (the highest resolution run of the 110.7 Mpc box) from the IllustrisTNG suite.

• Sample: 9,522 halos across 10 redshifts ($z = 0, 0.1, 0.2, 0.5, 1, 2, 4, 6, 8, 10$) and a wide range of virial masses ($10^{10} - 10^{14+} M_\odot$).
• Spatial Scales:
  • ISM Scale: A shell between $0.05R_{vir}$and$0.15R_{vir}$ (approximating the interface between the ISM and CGM).
  • Halo Scale: A shell between $0.95R_{vir}$and$1.05R_{vir}$ (the virial radius).
• Flow Quantification:
  • Particles were classified as inflows ($v_r < 0$) or outflows ($v_r > 0$).
  • Mass, Energy, and Metal Rates: Calculated by summing contributions of gas particles crossing these shells. Energy flows included both thermal ($E_{th} \propto c_s^2$) and kinetic ($E_{kin} \propto v^2$) components.
• Dominant Feedback Identification:
  • The authors integrated the energy injection rates from SNe, thermal AGN, and kinetic AGN over three characteristic travel times (based on 10th, 50th, and 90th percentile outflow velocities) to determine which mechanism dominated the energy budget for each halo.
• Loading Factors: Defined relative to the Star Formation Rate (SFR) for mass ($\eta_M$), SN energy ($\eta_E$), and metal yield ($\eta_Z$).

3. Key Contributions

• Comprehensive Flow Mapping: First detailed measurement of mass, energy, and metal inflow/outflow rates for IllustrisTNG across both ISM and halo scales, separated by dominant feedback mechanism.
• Energy Flow Analysis: Unlike previous studies focusing primarily on mass, this work explicitly quantifies energy flows, revealing that energy and mass flows often behave differently (e.g., net mass inflow with net energy outflow).
• Feedback Regime Separation: Explicitly distinguishes between SN-dominated, thermal AGN-dominated, and kinetic AGN-dominated halos, clarifying the specific role of each feedback mode in regulating the baryon cycle.
• Recycling and Entrainment: Quantifies the "galactic fountain" effect (rapid recycling of ejected gas) and the entrainment of CGM gas into outflows, showing how these processes evolve with redshift.

4. Key Results

A. Dominant Feedback Mechanisms

• Transition: A transition from SN-dominated to AGN-dominated feedback occurs at $M_{vir} \approx 10^{12} M_\odot$ at $z \sim 2$.
• Thermal vs. Kinetic: Thermal AGN feedback (high accretion) occurs in halos $4\times10^{10} < M_{vir} < 2\times10^{13} M_\odot$but does not drive significant outflows. Kinetic AGN feedback (low accretion/jet mode) becomes dominant in massive halos ($M_{vir} \gtrsim 10^{12} M_\odot$) at$z \lesssim 2$ and is the primary driver of quenching and strong outflows.

B. Mass Flows

• High Redshift ($z \gtrsim 2$): Strong net inflows on halo scales for all masses. On ISM scales, inflows are net positive but weaker.
• Low Redshift ($z \lesssim 0.5$):
  • Low Mass: Net inflows on halo scales; strong "galactic fountains" on ISM scales (ISM inflows > halo inflows, indicating rapid recycling).
  • Intermediate Mass ($M_{vir} \sim 10^{12}-10^{13} M_\odot$): The onset of kinetic AGN feedback leads to a balance where inflows and outflows nearly cancel, resulting in net outflows and galaxy quenching.
  • High Mass ($M_{vir} \gtrsim 10^{14} M_\odot$): Strong net inflows on halo scales again, as the deep gravitational potential well overcomes AGN feedback, allowing gas to accumulate in the CGM.
• Loading Factors: Mass loading factors ($\eta_M$) decrease with increasing halo mass and redshift. At $z=0$, $\eta_M$ ranges from $\sim 20$ (low mass) to $\sim 5$ (intermediate mass).

C. Energy Flows

• Decoupling of Mass and Energy: While mass flows are predominantly inward (accretion), energy flows are predominantly outward for most masses and redshifts.
• Mechanism: Galaxies gain mass but lose energy. This is driven by the injection of thermal/kinetic energy by feedback, which heats the CGM and prevents cooling (preventative feedback) or drives gas out (ejective feedback).
• AGN Impact: Kinetic AGN feedback drives significant net energy outflows from the halo scale, particularly in the $10^{12}-10^{13} M_\odot$ range.

D. Metal Flows

• Enrichment: ISM outflows are metal-rich ($Z \sim 0.4-1.0 Z_\odot$). Halo-scale outflows have metallicities similar to the CGM, suggesting mixing.
• Inflows: Inflows into the ISM are enriched relative to the average CGM metallicity, consistent with radial metallicity gradients in the CGM (higher metallicity closer to the galaxy).
• Loading: Metal loading factors ($\eta_Z$) are roughly constant with halo mass but increase with decreasing redshift.

E. Preventative vs. Ejective Feedback

• Low Mass: Feedback is primarily ejective (driving winds) but with significant recycling (fountains). Preventative feedback is weak.
• Intermediate Mass: Strong preventative feedback is observed at the halo scale (gas enters the halo but is heated/prevented from reaching the ISM) and ISM scale.
• High Mass: Preventative feedback weakens as the halo potential deepens, leading to gas accumulation in the CGM despite AGN activity.

5. Significance

• Guiding Empirical Models: The detailed measurements of flow rates and loading factors provide critical constraints for Semianalytic Models (SAMs) and Gas Regulator Models. The authors note that traditional SAMs often fail to reproduce these specific flow dynamics without tuning, particularly regarding energy flows and halo-scale recycling.
• Subgrid Physics Validation: The results highlight discrepancies between the "input" subgrid parameters (e.g., wind launch velocities) and the "emergent" properties (e.g., actual mass loading at the halo scale). This suggests that high-density CGM at high redshift stalls winds, while low-density CGM at low redshift allows for entrainment.
• Understanding Quenching: The study clarifies that the transition to kinetic AGN feedback is the primary driver of the "quenching" of star formation in massive galaxies, acting through a combination of ejective outflows and preventative heating of the CGM.
• Future Simulations: The findings suggest that future subgrid recipes (like the Arkenstone project) must account for multiphase outflows and the complex interaction between winds and the CGM density to accurately reproduce the baryon cycle.

In summary, this paper provides a rigorous, multi-scale accounting of the baryon cycle in IllustrisTNG, demonstrating that galaxy growth and quenching are regulated by a complex interplay of mass accretion, energy injection, and metal enrichment, heavily dependent on halo mass and the specific mode of feedback active.