The James Webb Space Telescope peered into an era when the universe was less than a billion years old and spotted strange objects—'little red dots.' These are black holes with masses of millions of suns, but in very compact, chemically pristine galaxies. Standard scenarios can't grow such heavyweights that early.
A new explanation resembles a shark being fed by a school of small fish. At the center of a dense gas cloud, a seed already sits — a black hole with a mass from a thousand to a hundred thousand suns. Within a few light-years, a swarm of tiny black holes buzzes around: each about thirty times heavier than our Sun. These little ones are primordial black holes, born in the first moments after the Big Bang. Gas slows their motion, and they spiral onto the central giant. Each capture releases about six percent of the victim's mass as gravitational waves, while the rest settles onto the central hole. The whole process takes from ten to fifty million years — by cosmic standards, an instant.
Mergers happen one after another, creating a distinctive gravitational-wave background — a hum that future detectors could hear. Its spectrum has a characteristic slope: loudness increases with frequency following a 'two-thirds' law. At frequencies around 13 millihertz for hundred-thousand-solar-mass holes and around 1 hertz for thousand-solar-mass ones, 'ringdowns' occur — damped oscillations of a newly born hole. Rare collisions of two large seeds would produce single bursts, distinguishable against the general noise.
Such a model doesn't require rare conditions. If primordial black holes exist and are strongly clustered in space, then 'red dots' are a natural outcome of their collective life.
as written by the authors arXiv:2607.09201
The James Webb Space Telescope (JWST) has revealed compact, red, overmassive accreting black holes - the so-called ``Little Red Dots'' (LRDs) - in chemically near-pristine hosts at $$z\simeq5 - 9$$, straining standard heavy-seed models. We show that a population of strongly clustered primordial black holes (PBH) with a broad mass function predicted by a QCD-epoch thermal history naturally realizes the configuration that assembles LRD-scale seeds: an intermediate-mass PBH \emph{nucleus} $$M_{\rm BH}\sim10^3$$-$$10^5\,M_\odot$$ surrounded, within a few parsecs, by a swarm of light ($$m\sim30\,M_\odot$$) PBHs embedded in dense baryonic gas. Gas dynamical friction keeps the loss cone full and lets the core contract, so the swarm sinks and is swallowed on $$t_{\rm seed}\sim10$$-50 Myr, well inside the cosmic time at $$z\sim10$$-$$15$$. Because a heavy nucleus is present \emph{ab initio}, the captures occur at extreme mass ratio $$q\sim10^{-4}$$-$$10^{-2}$$: the remnant is retained against gravitational recoil, and each capture reaches the innermost stable orbit under gravitational-wave (GW) emission, radiating $$\simeq0.06\,m c^2$$ so that the assembly efficiency is $$\zeta\simeq0.06$$ independent of $$M_{\rm BH}$$. The superposed swarm inspirals form a stochastic background $$\Omega_{\rm GW} h^2\sim10^{-13}$$-$$10^{-11}$$ with a $$\Omega_{\rm GW}\propto f^{2/3}$$ shape truncated below the gas-decoupling frequency and topped by a ringdown ``comb'' at $$f_{\rm ring}(M_{\rm BH})\simeq13$$ mHz for $$10^5\,M_\odot$$ and $$\simeq1.3$$ Hz for $$10^3\,M_\odot$$ nuclei at $$z_f\simeq12$$. The few comparable-mass nucleus-nucleus coalescences are instead individually resolvable LISA/deci-Hz sources. Detection, and discrimination of these signatures from a directly formed PBH seed of the same mass, would identify the LRDs as PBH-nucleus seeded black holes.
Gravitational waves from a thirty-solar-mass hole falling onto a thousand-solar-mass one carry away 6% of its mass — hundreds of times more efficient than nuclear fusion in stars.
In Fred Hoyle's novel 'The Black Cloud,' an intelligent gas cloud manipulates gravity to feed on stars — here, nature itself has arranged a similar feeding mechanism for black holes.