Little Red Dots Are Nurseries of Massive Black Holes

Fabio Pacucci, Lars Hernquist, Michiko Fujii

First listed 2025-09-02 | Last updated 2025-10-21

Abstract

The James Webb Space Telescope (JWST) has revealed a previously unknown population of compact, red galaxies at $z \sim 5$, known as "Little Red Dots" (LRDs). With effective radii of $\sim 100$ pc and stellar masses of $10^9-10^{11} \, M_\odot$, a purely stellar interpretation implies extreme central densities, $ρ_\star\sim10^4-10^5 \, M_\odot \, \mathrm{pc}^{-3}$ and in some cases up to $\sim 10^9 \, M_\odot \, \mathrm{pc}^{-3}$, far exceeding those of globular clusters. At such densities, the dynamical friction time for $10 \, M_\odot$ stars in the central $0.1$ pc is $< 0.1$ Myr, driving rapid mass segregation. We investigate the dynamical consequences of such an environment using: (i) a Fokker-Planck analysis of long-term core evolution, (ii) an analytical model for the collisional growth of a very massive star (VMS), and (iii) direct $N$-body simulations. All approaches show that runaway collisions produce a VMS with mass $9\times10^3 < M_{\rm VMS} \, [M_\odot] < 5\times10^4$ within $<1$ Myr. Once the supply of massive stars is depleted, the VMS contracts on a $\sim 8000$ yr Kelvin-Helmholtz timescale and undergoes a general relativistic collapse, leaving a massive black hole of mass $M_\bullet \sim 10^4 \, M_\odot$. We conclude that LRDs are natural nurseries for the formation of heavy black hole seeds via stellar-dynamical processes. This pathway produces seed number densities that far exceed those expected from direct collapse models, and, owing to the dense residual stellar core, can sustain high rates of tidal disruption events.

Short digest

Assuming the star-only interpretation of compact, red z∼5 Little Red Dots, the authors model ultradense cores where dynamical friction for 10 M⊙ stars in the inner 0.1 pc is <0.1 Myr, triggering rapid mass segregation. A Fokker–Planck core-evolution calculation, an analytic collisional-growth model, and direct N-body simulations all converge on runaway mergers that build a very massive star of 9×10^3–5×10^4 M⊙ within <1 Myr. The VMS then contracts on a ∼8000 yr Kelvin–Helmholtz timescale and collapses to an ∼10^4 M⊙ black hole. This route yields seed number densities above direct-collapse expectations and, with a dense residual core, can power elevated tidal disruption event rates.

Key figures to inspect

• Figure 1: Track the steepening of the central density and the four-order-of-magnitude mass gain inside the central parsec to see explicit core collapse while the outer profile remains static.
• Figure 2: Compare pre/post-contraction dynamical friction times; the fall to <0.1 Myr in the inner 0.1 pc explains rapid mass segregation, while the rise in σ1D quantifies the collisional environment.
• Figure 3: Read the VMS mass growth curve against the accretion and loss rates to see when net growth becomes runaway and why the final M_VMS is reached on a timescale comparable to the central t_df.
• Figure 4: Use the N-body track—initial discrete mergers followed by a runaway phase completing within <1 Myr—to confirm the analytic prediction and visualize the stepwise assembly of the VMS.