🔭 astrophysics

Light, heavy, primordial: exploring the diversity of black hole seeding and growth mechanisms in the JWST era

This study utilizes DELPHI and PHANES models to demonstrate that while super-Eddington accretion onto heavy astrophysical seeds and primordial black hole (PBH) accretion can both explain the massive, metal-poor black holes observed by JWST, only the PBH scenario uniquely reproduces the observed low black hole-to-stellar mass ratios in specific halo mass ranges, offering a distinct clustering-based discriminant for future observations.

Original authors: Pratika Dayal

Published 2026-04-24

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Pratika Dayal

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 early Universe as a vast, dark construction site. For decades, astronomers have had a big mystery: How did some of the "construction bosses" (Supermassive Black Holes) get so huge, so fast, in the first billion years of the Universe?

Usually, black holes are thought to start small (like a baby) and grow slowly over billions of years by eating gas and merging with others. But the new James Webb Space Telescope (JWST) has found "obese" black holes that are already massive giants, sitting in tiny, metal-poor galaxies. It's like finding a fully grown elephant living in a mouse hole.

This paper, written by Pratika Dayal, tries to solve this puzzle by testing three different theories on how these black hole "babies" were born and how they grew up.

The Three Contenders: The "Seed" Theories

The authors compare three different ways these black holes might have started:

The "Light Seed" (The Humble Start):
- The Idea: Black holes start as tiny remnants of the very first stars (about 100 times the mass of our Sun).
- The Growth: They try to grow by eating gas.
- The Problem: If they eat at a normal, safe speed (the "Eddington limit," like a person eating one meal a day), they simply can't get big enough in time. They are too slow. The paper says this theory is ruled out because it can't explain the giant black holes we see.
The "Heavy Seed" (The Big Start):
- The Idea: Black holes start as massive "babies" (thousands of times the mass of the Sun), perhaps formed from collapsing clouds of gas or crowded star clusters.
- The Growth: They can grow normally, or even "super-fast" (super-Eddington), like a bodybuilder eating 10 meals a day.
- The Verdict: This works! If they start big and eat fast, they can reach the sizes we see. However, there's a catch: these models struggle to explain why the host galaxies are so "metal-poor" (lacking heavy elements like gold or iron) and why the black holes are sometimes heavier than the entire galaxy they live in.
The "Primordial Black Hole" (The Cosmic Ghost):
- The Idea: These black holes didn't come from stars at all. They were born instantly during the Big Bang itself, like ghosts appearing out of thin air.
- The Growth: They act as the foundation of the galaxy. Instead of the galaxy forming first and then the black hole, the black hole forms first, and the galaxy builds around it.
- The Verdict: This is the most surprising contender. Because the black hole is the "boss" from day one, it naturally ends up being huge compared to the tiny galaxy around it. It also explains why the galaxies are so metal-poor (they haven't had time to make heavy elements yet).

The Detective Work: How Do We Tell Them Apart?

The authors ran computer simulations to see which theory matches the "crime scene" evidence (the JWST observations). Here is what they found:

The "Obese" Ratio: In normal galaxies, the black hole is usually a tiny fraction of the galaxy's weight. But in the early Universe, JWST sees black holes that are 30% to 100% of the galaxy's weight.
- Analogy: Imagine a person weighing 300 lbs living in a house that only weighs 300 lbs total. That's an "obese" black hole.
- Result: Only the Heavy Seeds (eating super-fast) and the Primordial Black Holes can explain this. The "Light Seeds" are too small to pull this off.
The Metal Mystery: The galaxies hosting these black holes are incredibly "clean" (low metallicity).
- Analogy: Imagine a kitchen that has never been used to cook; there's no grease or food stains.
- Result: The Primordial Black Hole model is the best fit here. Because the black hole formed first and the galaxy is still very young, there hasn't been enough time to create heavy metals.
The "Halo" Test (The Most Important Clue):
- Every galaxy sits inside a giant, invisible bubble of dark matter called a "halo."
- The Twist: In normal (astrophysical) models, bigger black holes need bigger halos to grow.
- The Primordial Difference: In the Primordial Black Hole model, the black hole creates its own halo. So, you can have a massive black hole sitting in a surprisingly small dark matter bubble.
- The "Negative Trend": The paper found a unique fingerprint: For Primordial Black Holes, as the dark matter halo gets bigger, the black hole actually becomes smaller relative to the stars. In normal models, it's the opposite.

The Final Verdict

The paper concludes that we can't just pick one winner yet, but we can rule out the "slow starter" (Light Seeds eating normally).

The Best Candidates: The universe likely contains a mix of Heavy Seeds (big babies eating fast) and Primordial Black Holes (ghosts from the Big Bang).
How to Solve the Mystery: To know for sure, we need to look at the "neighborhood" (clustering) of these black holes.
- If we find a massive black hole in a tiny, low-mass dark matter halo, it's almost certainly a Primordial Black Hole.
- If we find them in huge, massive halos, they are likely Heavy Seeds.

In a nutshell: The early Universe was a wild place. The black holes we see today might be the result of giant babies eating too much, or they might be ancient ghosts that built their own homes from scratch. The James Webb Space Telescope is giving us the clues, and soon, we might finally know which story is true.

1. Problem Statement

The James Webb Space Telescope (JWST) has revealed a population of massive black holes (BHs) in the first billion years of the Universe ( $z \sim 5-10$ ) that pose significant challenges to standard theoretical models. Key observational puzzles include:

Early Massiveness: Existence of BHs with masses up to $10^8 M_\odot$ at $z \sim 8-10$ .
"Obese" Black Holes: BHs that are over-massive relative to their host stellar masses, with ratios ( $M_{BH}/M_*$ ) as high as 30–100%.
Low Metallicity: Host galaxies inferred to be extremely metal-poor ( $Z \lesssim 0.01-0.1 Z_\odot$ ).
X-ray Weakness: Some early BHs appear X-ray weak despite high optical/IR luminosities.

The central question is whether these objects formed from astrophysical seeds (remnants of Population III stars or direct collapse) or cosmological Primordial Black Holes (PBHs) that acted as seeds for structure formation itself.

2. Methodology

The authors compare two distinct theoretical frameworks using semi-analytic and analytic models within a $\Lambda$ CDM cosmology ( $\Omega_\Lambda=0.673, \Omega_m=0.315, \Omega_b=0.049, H_0=67.3$ km s $^{-1}$ Mpc $^{-1}$ ).

A. Astrophysical Seeding Models (DELPHI)

Code: delphi semi-analytic model.
Seeding:
- Light Seeds (ls): $\sim 100 M_\odot$ (Pop III remnants).
- Heavy Seeds (hs): $10^3 - 10^5 M_\odot$ (formed via dynamical interactions or runaway mergers).
Growth Mechanisms:
- Accretion (Eddington-limited and Super-Eddington).
- Mergers (galaxy and BH mergers).
Variants: Four models tested by varying seed mass and Eddington fraction ( $f_{Edd}$ $f_{E dd}$ ):
- hs-edd1: Heavy seeds, Eddington-limited ( $f_{Edd}=1$ ).
- hs-edd5: Heavy seeds, Super-Eddington ( $f_{Edd}=5$ ).
- ls-edd1: Light seeds, Eddington-limited.
- ls-edd5: Light seeds, Super-Eddington.
Physics: Includes ISM physics, star formation efficiency, and SNII feedback calibrated against UV luminosity functions and stellar mass functions at $z \sim 5-14$ .

B. Cosmological Seeding Models (PHANES)

Code: phanes analytic model.
Seeding: PBHs formed at the Big Bang/inflation with a power-law mass spectrum ( $10^{0.5} - 10^6 M_\odot$ ).
Growth Mechanism:
- PBHs act as seeds for structure, assembling halos around themselves.
- Growth occurs only via accretion (no mergers of seeds).
- Accretion is capped at sub-Eddington rates ( $f_{Edd} \le 0.25$ ).
Calibration: Normalized to match the number density and mass of high- $z$ candidates UHZ1 and GHZ9 ( $z \sim 10-10.4$ ).

3. Key Contributions

Systematic Comparison: Provides a direct, quantitative comparison between astrophysical and PBH seeding scenarios against the latest JWST constraints ( $z \sim 5-10$ ).
Discriminators: Identifies specific observables (BH mass function, luminosity function, $M_{BH}-M_*$ relation, metallicity, and halo mass dependence) that can distinguish between models.
New Clustering Diagnostic: Proposes a novel observational test based on the negative correlation between the BH-to-stellar mass ratio and halo mass for PBHs, contrasting with the positive correlation in astrophysical models.

4. Key Results

A. Black Hole Mass Function (BHMF)

Evolution: All models show increasing BHMF amplitude and mass range with decreasing redshift.
PBHs: Predict a significant overabundance (2–3 orders of magnitude) of intermediate-mass BHs ( $10^3 - 10^4 M_\odot$ ) at $z \sim 5-10$ compared to astrophysical models. They match the massive end ( $>10^8 M_\odot$ ) well.
Light Seeds: The ls-edd1 model (light seeds, Eddington-limited) fails to produce BHs massive enough to match observations at $z \sim 7$ and is ruled out.
Convergence: By $z \sim 5$ , most astrophysical models (except ls-edd1) converge in their BHMF predictions.

B. Bolometric Luminosity Function (BLF)

High- $z$ ( $z \sim 10$ ): Only the hs-edd5 (super-Eddington heavy seeds) and PBH models are consistent with the lower limits of BLF derived from UHZ1 and GHZ9.
Low- $z$ ( $z \sim 5-7$ ): All models except ls-edd1 reproduce observed BLFs within uncertainties.
PBH Luminosity: PBHs under-predict the bright end of the BLF at $z \sim 7$ because they are limited to sub-Eddington accretion, whereas observations often assume Eddington-limited accretion for mass estimates.

C. Black Hole Mass vs. Stellar Mass ( $M_{BH} - M_*$ )

Ratios: JWST observations show high ratios ( $M_{BH}/M_* \sim 0.3 - 1$ $M_{B H} / M_{*} \sim 0.3 - 1$ ).
- ls-edd1: Cannot reproduce these high ratios.
- hs-edd1: Reproduces the bulk of objects but fails to explain the most extreme "obese" BHs ( $M_{BH} \sim 10^7 M_\odot$ with $M_* < 10^7 M_\odot$ ).
- hs-edd5 & PBH: Both successfully reproduce the high $M_{BH}/M_*$ ratios across all redshifts.
Halo Mass Dependence (Crucial Finding):
- Astrophysical Models: $M_{BH}/M_*$ increases with increasing halo mass (deeper potentials allow more gas accretion).
- PBH Models: $M_{BH}/M_*$ decreases with increasing halo mass. Light PBHs in low-mass halos have inefficient star formation (high ratio), while as they grow larger halos, star formation catches up, lowering the ratio.

D. Metallicity Evolution

Trend: Metallicity ( $Z$ ) generally scales with $M_*$ and $M_{BH}$ .
PBHs: Due to their unique assembly history (BH precedes the halo), PBH hosts are more susceptible to feedback, resulting in the lowest metallicities for a given BH mass.
Comparison: The PBH model is the only one that simultaneously reproduces the observed combination of high $M_{BH}$ , low $M_*$ , and extremely low metallicity ( $Z \le 0.01 Z_\odot$ ) inferred for specific systems (e.g., Tripodi et al. 2025).

5. Significance and Conclusions

Ruled Out: The model where all BHs are seeded by light seeds growing at Eddington-limited rates (ls-edd1) is conclusively ruled out by current data.
Leading Contenders:
- Super-Eddington Heavy Seeds (hs-edd5): Consistent with BLF and high $M_{BH}/M_*$ ratios but struggles with the specific low-metallicity/low-stellar-mass combinations.
- Primordial Black Holes (PBHs): Consistent with high $M_{BH}/M_*$ ratios, low metallicities, and the BHMF at the massive end. They uniquely predict a population of intermediate-mass BHs and a negative $M_{BH}/M_*$ vs. halo mass trend.
Proposed Observational Test: The authors propose a clustering-based discriminant to be tested with future JWST, Roman, and Euclid data:
- Select systems at $z \sim 7$ with $M_{BH}/M_* > 0.1$ and $L_{Bol} \sim 10^{44-46}$ erg s $^{-1}$ .
- If these systems reside in low-mass halos ( $10^9 - 10^{11} M_\odot$ ) and exhibit a negative correlation between $M_{BH}/M_*$ and halo mass, it would strongly favor the PBH seeding scenario.
Caveats: The study highlights uncertainties in observational inferences (SED fitting, bolometric corrections) and theoretical assumptions (PBH mass spectrum, spin, and feedback coupling). Future deep X-ray observations (e.g., NewAthena) and statistical samples of SEDs are required to definitively distinguish between these scenarios.