🔭 astrophysics

Population III star formation in an X-ray background: IV. On-the-fly calculation of radiation backgrounds and their impact on the intergalactic medium

Using cosmological zoom-in simulations with on-the-fly radiation background calculations, this study demonstrates that X-rays from Population III supernovae create a self-consistent feedback loop that enhances molecular hydrogen formation and increases Population III star density, particularly in underdense regions, while remaining consistent with Planck 2018 optical depth constraints.

Original authors: Jongwon Park, Massimo Ricotti

Published 2026-03-30

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Jongwon Park, Massimo Ricotti

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 Big Picture: The Cosmic "Chicken and Egg" Problem

Imagine the early Universe as a giant, dark, cold room filled with invisible gas. Scientists want to know how the very first stars (called Population III stars) turned on the lights.

Usually, we think of stars as the result of gas collapsing. But this paper asks a different question: What if the stars themselves change the room so that more stars can form?

The authors are studying a specific type of "light" called X-rays (high-energy radiation) and how it interacts with the gas before the first stars even fully ignite. They found that X-rays act like a cosmic "starter kit" that helps the first stars form more easily, but only if you calculate the process in real-time, not after the fact.

The Main Characters

Population III Stars (The First Stars): Think of these as the "giants" of the early universe. They are massive, hot, and made of pure hydrogen and helium. They are the ancestors of all stars today.
The IGM (The Intergalactic Medium): This is the "air" between the stars. In the early universe, it was a cold, neutral fog.
X-rays: These are high-energy photons. Think of them as cosmic microwaves. They don't just heat things up; they zap electrons off atoms, turning neutral gas into a slightly ionized (electrically charged) soup.
LW Radiation (Lyman-Werner): This is a different kind of light (ultraviolet) that acts like a cosmic weed-killer. It destroys the fuel (molecular hydrogen) that stars need to form.

The Problem: The "Weed-Killer" vs. The "Fertilizer"

In the early universe, there was a battle between two forces:

The Weed-Killer (LW Radiation): As soon as the first stars formed, they emitted UV light that tried to destroy the molecular hydrogen (H₂) needed to make new stars. This usually stops star formation.
The Fertilizer (X-rays): The first stars eventually exploded as supernovae. These explosions shot out X-rays. Unlike the UV light, X-rays can penetrate deep into the gas fog. They don't just heat the gas; they create a chemical reaction that actually helps make more molecular hydrogen.

The Analogy:
Imagine trying to start a campfire in a damp forest.

The UV light is like a strong wind blowing out your matches (it destroys the fuel).
The X-rays are like a special chemical spray that dries out the wood and makes it easier to catch fire (it helps create fuel).

The paper asks: Does the "spray" (X-rays) help us start more fires (stars) despite the "wind" (UV light)?

The Innovation: The "Live Stream" vs. The "Replay"

This is the most technical but crucial part of the paper.

The Old Way (Post-Processing):
Imagine you film a soccer game, wait until it's over, and then try to calculate how the players' energy levels changed based on the score. You are looking at the past.

In previous studies, scientists ran simulations of star formation without X-rays, recorded the results, and then added the X-ray effects later.
The Flaw: This misses the feedback loop. The stars create X-rays, which help make more stars, which create more X-rays. If you calculate it after the fact, you miss this chain reaction.

The New Way (On-the-Fly):
Imagine watching the soccer game live, where the players' energy levels update in real-time based on what's happening right now.

The authors built a new method where the computer calculates the X-ray background while the simulation is running.
The Result: Because the X-rays help make more stars, and those stars make more X-rays, the "live" calculation shows a much stronger effect (10 to 100 times stronger) than the "replay" method.

The Key Findings

X-rays are a "Positive Feedback Loop":
The X-rays from exploding first stars act like a booster rocket. They heat the gas slightly and help create the fuel (H₂) needed for new stars. This means more stars form than we previously thought, especially in empty, low-density regions of the universe.
It Depends on Where You Are:
- In dense regions (crowded neighborhoods), the stars form so early that the X-rays haven't built up enough to help yet.
- In underdense regions (empty suburbs), the stars form a bit later. By then, the X-ray background has built up enough to act as a powerful fertilizer, significantly boosting star formation.
The "Optical Depth" Mystery (The Foggy Window):
Scientists measure how much the early universe scattered light (like looking through a foggy window). This is called "optical depth."
- The X-rays slightly ionized the gas, making the "fog" a bit clearer earlier than expected.
- However, the total amount of fog at the end still matches what the Planck satellite observed. So, while X-rays changed the timing, they didn't break the rules of the universe.
This is a "Conservative" Estimate:
The authors only counted X-rays from supernovae (exploding stars). They didn't include X-rays from black holes or active galactic nuclei (which we now know existed thanks to the James Webb Space Telescope).
- The Takeaway: If we add black holes to the mix later, the X-ray effect will be even stronger. The numbers in this paper are actually the "minimum" possible effect.

Summary in One Sentence

This paper uses a new "live-streaming" computer method to show that the X-rays from the first exploding stars acted as a cosmic fertilizer, helping to create more first stars than we thought, especially in the empty spaces of the early universe.

1. Problem Statement

The formation of the first stars (Population III or Pop III) and the subsequent reionization of the Universe are heavily influenced by radiation backgrounds, specifically Lyman-Werner (LW) UV radiation and X-rays.

The Feedback Loop: Pop III stars produce supernovae (SNe) that emit X-rays. These X-rays ionize and heat the intergalactic medium (IGM), enhancing the formation of molecular hydrogen ( $H_2$ ) via the $H^-$ channel. Enhanced $H_2$ cooling promotes further Pop III star formation, creating a positive feedback loop. Conversely, LW radiation photodissociates $H_2$ , suppressing star formation.
Methodological Limitations: Previous studies often estimated radiation backgrounds via post-processing (calculating the background after the simulation based on star formation history) or used small simulation boxes that could not capture the contribution of distant sources (mean free paths of X-rays $\gg$ box size). This approach fails to capture the real-time, self-consistent feedback loop between star formation and the radiation environment.
The Gap: There is a need for a method that simultaneously resolves high-resolution star formation in small volumes while accounting for the global radiation background from distant sources and the dynamic feedback loop, specifically to understand the impact on the IGM and Pop III number density.

2. Methodology

The authors employed a suite of cosmological zoom-in simulations using the RAMSES-RT code (AMR radiative hydrodynamics) with the following key methodological innovations:

A. Simulation Setup

Volumes: Two parent dark-matter-only simulations ( $L=8$ Mpc/ $h$ and $4$ Mpc/ $h$ ) were used to select 10 sub-volumes representing diverse cosmic environments (overdense, mean-density, and underdense).
Representative Volume: Volume 2 (mean density) was selected as the reference to compute the global radiation background on-the-fly, as its halo mass function matches the cosmic average.
Resolution: The innermost target region ($1$ Mpc/ $h$ or $0.5$ Mpc/ $h$ ) was refined to a physical resolution of $\sim 0.57$ pc at $z=9$ , sufficient to resolve Pop III star-forming cores.
Star Formation Physics:
- Pop III: Formed in metal-free ( $Z < 10^{-4} Z_\odot$ ) gas at densities $n_H > 10^4$ cm $^{-3}$ . Adopted a fixed mass of $100 M_\odot$ .
- Pop II: Formed in metal-enriched gas ( $Z > 10^{-4} Z_\odot$ ) using a Schmidt law.
- Feedback: Pop III stars emit UV during their main sequence. Upon death, they explode as energetic hypernovae ( $10^{53}$ erg), returning mass and metals, and emitting X-rays (modeled as thermal bremsstrahlung from the remnant).

B. On-the-Fly Radiation Background Calculation (Key Innovation)

Instead of post-processing, the authors implemented an analytic integration along the lightcone to compute the global background during the simulation:

Integration: The specific intensity $J_\nu$ is calculated by integrating the emissivity of sources along the line of sight from $z_{start}=40$ to the current redshift, accounting for redshift and absorption (optical depth).
Feedback Loop: The star formation history (SFH) derived from the simulation (specifically Volume 2) is used to update the emissivity at every time step. This captures the self-consistent reinforcement where more stars $\to$ more X-rays $\to$ more stars.
Hybrid Approach:
- Local: Calculated via ray-tracing (M1 closure) for sources within the zoom-in box.
- Global: Calculated analytically for sources outside the box, assuming the SFH of Volume 2 is representative of the cosmic mean.
- Early Universe Handling: For $z > z_{ana}$ (when few stars exist in the box), an analytic estimate based on the Press-Schechter halo mass function is used to avoid unphysical jumps in background intensity.

C. Experimental Design

Comparisons: For each sub-volume, two simulations were run:
1. LW-only: Includes LW feedback but no X-rays.
2. LW + X-ray: Includes both LW and the on-the-fly calculated X-ray background.
Post-processing Test: A control test applied a post-processed X-ray background (derived from the LW-only SFH) to an underdense volume to quantify the error introduced by neglecting the feedback loop.

3. Key Contributions

Novel On-the-Fly Method: Developed a computationally efficient method to self-consistently include global radiation backgrounds in high-resolution zoom-in simulations, bridging the gap between small-box resolution and large-volume cosmological context.
Quantification of the Feedback Loop: Demonstrated that neglecting the real-time feedback loop (via post-processing) significantly underestimates the X-ray background intensity and its impact on star formation.
Environmental Dependence: Mapped how X-ray feedback varies across different cosmic densities, showing that the positive feedback is most effective in mean-density and underdense regions.

4. Key Results

A. Radiation Background Evolution

X-ray Background: Generated primarily by Pop III supernovae, reaching a peak intensity of $J_X \sim 10^{-5}$ (in units of $J_{21}$ ) at $z \sim 15$ . It declines gradually as Pop III formation is suppressed by rising LW radiation.
LW Background: Remains moderate ( $J_{LW} \sim 10^{-1}$ ) until $z \sim 12$ , then rises sharply to $10^1 - 10^2$ with the onset of Pop II star formation.
Comparison: The on-the-fly X-ray background is 1–2 orders of magnitude stronger than post-processed estimates because it captures the cumulative effect of the feedback loop.

B. Impact on the Intergalactic Medium (IGM)

Heating and Ionization: The X-ray background mildly heats the neutral IGM to $T \sim 100$ K and increases the electron fraction to $x_e \sim 5 \times 10^{-3}$ by $z=9$ .
H2 Fraction: While the global $H_2$ fraction in the diffuse IGM remains low (due to efficient LW photodissociation), X-rays significantly enhance $H_2$ formation inside dark matter minihaloes and dense filaments. This is because $H_2$ formation scales with density squared, allowing it to outpace dissociation in dense regions.

C. Impact on Pop III Star Formation

Positive Feedback: The enhanced $H_2$ cooling in minihaloes lowers the critical halo mass required for collapse, leading to a net increase in Pop III star counts.
Magnitude:
- In Volume 2 (mean density), Pop III counts increased from 92 (LW-only) to 165 (LW+X-ray).
- In underdense regions (Volumes 3–10), the enhancement was most pronounced (e.g., Volume 3: 1 to 7 stars).
- In overdense regions (Volume 1), the effect was minimal because haloes formed early ( $z > 16$ ) when the X-ray background was still weak.
Post-processing Failure: Applying a post-processed background to Volume 3 yielded only 3 stars, compared to 7 with the on-the-fly method, proving that the feedback loop is critical for accurate predictions.

D. Reionization and Optical Depth

Thomson Optical Depth ( $\tau_e$ ): The X-ray background raises the optical depth at high redshifts ( $z \gtrsim 10$ $z ≳ 10$ ).
- The contribution from Pop III X-rays alone increases $\tau_e$ by $\sim 0.004$ (a 150% increase relative to the LW-only case in the pre-reionization era).
- However, the total optical depth ( $\tau_e \approx 0.058$ ) remains consistent with Planck 2018 constraints ( $0.054 \pm 0.007$ ), as the majority of ionization occurs later during full reionization.

5. Significance and Conclusion

Conservative Lower Limit: The model assumes X-rays originate solely from Pop III supernovae. Given JWST observations of early black holes and AGN, the actual early X-ray background is likely stronger, implying the positive feedback on Pop III formation could be even more significant.
Methodological Shift: The study establishes that on-the-fly calculation is essential for modeling early Universe physics. Post-processing methods fail to capture the self-reinforcing nature of X-ray feedback, leading to substantial underestimates of Pop III populations.
Cosmological Implications: The results suggest that X-ray feedback plays a crucial role in sustaining Pop III star formation in underdense regions, potentially influencing the rate of Pair-Instability Supernovae (PISNe), early metal enrichment, and the timing of the transition to Pop II star formation.

This work provides a robust framework for future studies incorporating additional X-ray sources (e.g., X-ray binaries, accreting black holes) to fully constrain the role of radiation backgrounds in the Cosmic Dawn.