Early supermassive black holes
How did some black holes grow to billions of solar masses within the first billion years after the Big Bang?
Summary
Why it is open
Quasars, the bright centres of galaxies where gas falls onto a supermassive black hole, are seen at redshifts above 6, when the Universe was less than a billion years old [1]. In 2015 Xue-Bing Wu and colleagues reported the quasar SDSS J0100+2802 at redshift 6.30, with a black hole of about 1.2 × 10¹⁰ solar masses [1]. In 2021 Feige Wang and colleagues reported J0313−1806 at redshift 7.642, the most distant quasar then known, with (1.6 ± 0.4) × 10⁹ solar masses about 670 million years after the Big Bang [2]. Such masses are usually estimated from the widths of broad emission lines; Wu and colleagues put the systematic uncertainty of this method at up to a factor of three [1].
The difficulty is time. Radiation from accreting gas pushes back on the infall, defining the Eddington luminosity and a corresponding characteristic accretion rate; growing at that limit with a radiative efficiency of about 10%, a black hole's mass increases by a factor e roughly every 50 million years [3]. Kohei Inayoshi, Eli Visbal and Zoltán Haiman calculate that a seed of 100 solar masses would need about 0.8 billion years of uninterrupted Eddington-limited accretion to reach 10⁹ solar masses, comparable to the age of the Universe at redshift 6 [3].
The problem therefore has two parts that trade against each other: how massive the first seeds were, and how fast and how continuously they grew [3]. Inayoshi and colleagues expect that telling the scenarios apart "will be difficult", and favour combining gravitational-wave observations of mergers with multi-wavelength observations of accretion [3]. Since 2023 the James Webb Space Telescope (JWST) has found fainter, less massive black holes at earlier times; several key results rest on single objects, and their masses and nature are disputed [4, 5, 6].
What is known
Seeds in theory. Remnants of the first, metal-free stars (Population III) would give light seeds of about 10–100 solar masses, but their early growth "was typically stunted" by the shallow gravitational wells of their host galaxies [3]. Heavy seeds of 10⁵–10⁶ solar masses could form through a supermassive star where pristine gas in a larger halo stays warm, for example when strong ultraviolet light destroys the molecular hydrogen that would cool it; such models are often called direct collapse [3]. Runaway collisions in very dense star clusters could give seeds of 10³–10⁴ solar masses [3]. Accretion faster than the Eddington rate onto a light seed would act like a heavy seed, and for most seeding models a time-averaged rate only about twice the Eddington rate would suffice [3].
JWST black holes (2023–2025). In 2024 Jorryt Matthee and colleagues reported 20 sources at redshifts 4.2–5.5 with broad hydrogen lines, implying black holes of 10⁷–10⁸ solar masses at a number density an order of magnitude above the extrapolation from quasar luminosities [7]. Their paper calls these sources little red dots [7]; what such objects are is a separate open problem, Little red dots. In January 2024 Roberto Maiolino and colleagues interpreted the spectrum of GN-z11, a galaxy at redshift 10.6, as showing a black hole with log(M/M☉) = 6.2 ± 0.3, about 1.6 million solar masses, accreting at about five times the Eddington rate [4]. A 2025 mid-infrared spectrum by J. Álvarez-Márquez and colleagues found no dominant broad hydrogen line and was consistent with a compact, low-metallicity starburst, though a weak broad component could not be ruled out [8]. Fabio Pacucci and colleagues found in 2023 that black holes at redshifts 4–7 are 10–100 times more massive than local black holes in galaxies of the same stellar mass, which they state is not a selection effect [9]. Junyao Li and colleagues argued in 2025 that selection biases and measurement uncertainties "can largely explain" that offset [10]. Ignas Juodžbalis and colleagues reported in 2024 a dormant black hole of about 4 × 10⁸ solar masses at redshift 6.68, accreting at 0.02 times the Eddington rate, with a black hole to stellar mass ratio about 1,000 times the local value [11]. How black holes and galaxies reached the local relations is itself open: Origin of the M–sigma relation.
Recent claims (2023–2026). In November 2023 Ákos Bogdán and colleagues reported in Nature Astronomy Chandra X-ray emission from UHZ1, a lensed galaxy at redshift about 10, implying an obscured black hole of 10⁷–10⁸ solar masses if it accretes at the Eddington limit, comparable to its host's stellar mass, and consistent with a heavy seed [12]. On 14 October 2026 Fan Zou and colleagues, in The Astrophysical Journal, analysing the full 2.2 megaseconds of Chandra data, found the excess at only 2.0–2.9σ, wrote that the original 4.2–4.4σ detection "is not robustly reproducible", and described UHZ1 as a low-mass, star-forming galaxy [13]. In January 2026 V. Rusakov and colleagues argued in Nature (preprint March 2025) that in most broad-line sources with the best JWST spectra, little red dots included, the lines are widened by electron scattering, implying black hole masses of 10⁵–10⁷ solar masses, "two orders of magnitude lower than previous estimates" [5]. In 2025 Matilde Brazzini and colleagues found in MNRAS Letters that the line profiles of the brightest known little red dot, at redshift 2.26, are inconsistent with simple scattering, a "counterexample" to a universal overestimate [14]. In May 2026 Juodžbalis and colleagues, also in Nature, measured a rotation curve in one strongly lensed little red dot at redshift 7.04 and found a central mass of 50 million solar masses, consistent with the line-width estimate and with a black hole outweighing its host's stars by more than a factor of two [6].
Main approaches
Light seeds with bursts of fast growth. Remnants of the first stars grow in short episodes above the Eddington rate [3, 15]. Strongest evidence: in 2026 cosmological simulations with resolutions down to 0.1 parsec, Daxal Mehta, John Regan and Lewis Prole found light seeds growing to about 10⁴ solar masses in bursts of up to 1,000 times the Eddington rate [15]; their discoverers read GN-z11 and the dormant black hole at redshift 6.68 as consistent with episodic super-Eddington growth, though GN-z11 also fits a heavy seed [4, 11]. Main difficulty: in those simulations only 0.05–1.5% of black holes doubled their seed mass, and supernova feedback, black hole feedback or lack of gas eventually stopped all growth [15].
Heavy seeds. The first massive black holes began at 10⁴–10⁶ solar masses [3, 12]. Strongest evidence: Wang and colleagues find that J0313−1806, growing at the Eddington limit from redshift 15–30, needs a seed of 10⁴–10⁵ solar masses, inconsistent with a Population III remnant or with dense star clusters [2]; and the "naked" black hole at redshift 7.04, which its discoverers call "a massive black-hole seed caught in its earliest accretion phase" [6]. Main difficulty: the conditions are "fully satisfied only in rare special environments" [3], and the original detection significance for UHZ1 was not reproduced in the full Chandra data [13].
Smaller masses than inferred. Part of the tension may lie in the measurements: selection effects make detected black holes look overmassive [10], and electron scattering may widen lines in little red dots [5]. Strongest evidence: forward modelling of survey limits [10] and fits to the best JWST spectra [5]. Main difficulty: the multi-line counterexample [14] and the dynamical mass at redshift 7.04, which agrees with the line-width estimate [6].
Open sub-questions
- UHZ1 in X-rays. Does a deeper or independent hard X-ray observation recover a persistent source at the 4σ level first reported, or confirm the 2.0–2.9σ result from the full Chandra data [12, 13]?
- GN-z11's broad line. Does deeper rest-frame optical spectroscopy detect a broad hydrogen component in GN-z11, and does its width match a black hole of about 10⁶ solar masses [4, 8]?
- Electron scattering. Among little red dots with Hα, Hβ and Paβ all measured, what fraction have profiles fitted by a single electron-scattering model, and does it depend on luminosity or redshift [5, 14]? See Little red dots.
- More dynamical masses. Can rotation curves of other strongly lensed broad-line sources above redshift 6 test line-width masses as at redshift 7.04, and do they agree within a factor of three [1, 6]?
- Overmassive or selected. Does a sample chosen without requiring broad lines, including dormant black holes, still lie above the local black hole to stellar mass relation after forward modelling like that of Li and colleagues [9, 10, 11]?
- From 10⁴ to 10⁹. In simulations like those of Mehta and colleagues, what fraction of seeds reaching 10⁴ solar masses keep growing to 10⁶ by redshift 10, and does that supply enough progenitors for quasars like J0313−1806 [2, 15]?
Work on this in Papex
If you have a testable answer to one of the sub-questions above, you can draft it as a private hypothesis in Papex, or prepare the draft from Claude or ChatGPT with the Papex connector. You review the exact version, sign it and decide whether to publish it; a public record that names this problem is listed on this page. A Papex record shows who signed which version and when. It is not peer review and does not mean the hypothesis is correct.
Key references
- Wu, X.-B., Wang, F., Fan, X., Yi, W., Zuo, W., Bian, F. et al. (2015). An ultraluminous quasar with a twelve-billion-solar-mass black hole at redshift 6.30. Nature 518(7540), 512–515. 10.1038/nature14241 · Checked 15 Oct 2026
- Wang, F., Yang, J., Fan, X., Hennawi, J. F., Barth, A. J., Bañados, E. et al. (2021). A luminous quasar at redshift 7.642. The Astrophysical Journal Letters 907(1), L1. 10.3847/2041-8213/abd8c6 · Checked 15 Oct 2026
- Inayoshi, K., Visbal, E. & Haiman, Z. (2020). The assembly of the first massive black holes. Annual Review of Astronomy and Astrophysics 58, 27–97. 10.1146/annurev-astro-120419-014455 · Checked 15 Oct 2026
- Maiolino, R., Scholtz, J., Witstok, J., Carniani, S., D'Eugenio, F., de Graaff, A. et al. (2024). A small and vigorous black hole in the early Universe. Nature 627(8002), 59–63. 10.1038/s41586-024-07052-5 · Checked 15 Oct 2026
- Rusakov, V., Watson, D., Nikopoulos, G. P., Brammer, G., Gottumukkala, R., Harvey, T. et al. (2026). Little red dots as young supermassive black holes in dense ionized cocoons. Nature 649(8097), 574–579 (first posted as arXiv:2503.16595 in March 2025). 10.1038/s41586-025-09900-4 · Checked 15 Oct 2026
- Juodžbalis, I., Marconcini, C., D'Eugenio, F., Maiolino, R., Marconi, A., Übler, H. et al. (2026). A direct black-hole mass measurement in a little red dot at high redshift. Nature 653(8116), 1017–1021. 10.1038/s41586-026-10579-4 · Checked 15 Oct 2026
- Matthee, J., Naidu, R. P., Brammer, G., Chisholm, J., Eilers, A.-C., Goulding, A. et al. (2024). Little red dots: an abundant population of faint active galactic nuclei at z ∼ 5 revealed by the EIGER and FRESCO JWST surveys. The Astrophysical Journal 963(2), 129. 10.3847/1538-4357/ad2345 · Checked 15 Oct 2026
- Álvarez-Márquez, J., Crespo Gómez, A., Colina, L., Langeroodi, D., Marques-Chaves, R., Prieto-Jiménez, C. et al. (2025). Insight into the starburst nature of galaxy GN-z11 with JWST MIRI spectroscopy. Astronomy & Astrophysics 695, A250. 10.1051/0004-6361/202451731 · Checked 15 Oct 2026
- Pacucci, F., Nguyen, B., Carniani, S., Maiolino, R. & Fan, X. (2023). JWST CEERS and JADES active galaxies at z = 4–7 violate the local M•–M⋆ relation at >3σ: implications for low-mass black holes and seeding models. The Astrophysical Journal Letters 957(1), L3. 10.3847/2041-8213/ad0158 · Checked 15 Oct 2026
- Li, J., Silverman, J. D., Shen, Y., Volonteri, M., Jahnke, K., Zhuang, M.-Y. et al. (2025). Tip of the iceberg: overmassive black holes at 4 < z < 7 found by JWST are not inconsistent with the local M_BH–M⋆ relation. The Astrophysical Journal 981(1), 19. 10.3847/1538-4357/ada603 · Checked 15 Oct 2026
- Juodžbalis, I., Maiolino, R., Baker, W. M., Tacchella, S., Scholtz, J., D'Eugenio, F. et al. (2024). A dormant overmassive black hole in the early Universe. Nature 636(8043), 594–597. 10.1038/s41586-024-08210-5 · Checked 15 Oct 2026
- Bogdán, Á., Goulding, A. D., Natarajan, P., Kovács, O. E., Tremblay, G. R., Chadayammuri, U. et al. (2024). Evidence for heavy-seed origin of early supermassive black holes from a z ≈ 10 X-ray quasar. Nature Astronomy 8(1), 126–133 (published online 6 November 2023). 10.1038/s41550-023-02111-9 · Checked 15 Oct 2026
- Zou, F., Gallo, E., Zuo, Z., Hodges-Kluck, E., Nguyen, D. D., Roberts-Borsani, G. et al. (2026). Revisiting the claim for a direct-collapse black hole in UHZ1 at z = 10.05. The Astrophysical Journal 1010(2), 164 (published online 14 October 2026; preprint arXiv:2603.24893). 10.3847/1538-4357/aea0cc · Checked 15 Oct 2026
- Brazzini, M., D'Eugenio, F., Maiolino, R., Juodžbalis, I., Ji, X., Scholtz, J. et al. (2025). Ruling out dominant electron scattering in little red dots' Rosetta Stone using multiple hydrogen lines. Monthly Notices of the Royal Astronomical Society: Letters 544(1), L167–L173. 10.1093/mnrasl/slaf116 · Checked 15 Oct 2026
- Mehta, D. H., Regan, J. A. & Prole, L. (2026). The growth of light seed black holes in the early Universe. Nature Astronomy 10(4), 583–592 (published online 21 January 2026). 10.1038/s41550-025-02767-5 · Checked 15 Oct 2026
Source
Open, as listed by Wikipedia · Unsolved problems in astronomy. Checked 08 Oct 2026.
The status is the source’s. Papex does not decide whether a problem is solved.
Records on this problem
No public record addresses this problem yet.