The physical origin of stellar bars at redshifts $z > 2$: formation conditions and reproduction in cosmological simulations

The discovery of barred spiral galaxies in the first two billion years of cosmic time — ceers-2112 at $z\approx3$, COSMOS-74706 at $z_{\rm spec}=3.16$, and GN20 at $z=4.055$ — overturned the expectation that stellar bars form late, in gas-poor, dynamically settled disks, and exposed a failure of pre-JWST cosmological simulations, which produce almost no barred Milky-Way progenitors beyond $z=1.5$. The open problem is to establish the physical origin of these early bars: determine the causal roles of gas fraction, star-formation efficiency, turbulence, and baryon versus dark-matter dominance in enabling bar formation at $z>2$, and identify the simulation ingredients required to reproduce the observed population, particularly low-mass barred galaxies. Substantial 2024–2026 progress — bar-fraction surveys out to $z\sim4$, new $z>3$ barred galaxies including a gas-rich case that challenges the gas-consumption scenario, and candidate mechanisms such as dark-matter self-interactions and merger-triggered bar formation — has sharpened but not resolved the question: no consensus mechanism exists, and no cosmological simulation robustly reproduces low-mass barred galaxies at $z>1.5$.

Background

Roughly half of the disk galaxies in the present-day Universe, including the Milky Way, contain a stellar bar: an elongated, rigidly rotating ridge of stars in the central disk that develops when a rotationally supported, dynamically 'cold' (low stellar random velocities relative to rotation) disk becomes unstable to non-axisymmetric perturbations. Once formed, a bar exerts torques on the disk and funnels gas toward the center, driving 'secular' (slow, internal, non-merger-driven) evolution. Classical theory and pre-James Webb Space Telescope (JWST) observations with the Hubble Space Telescope supported a late-forming picture: gas-rich, turbulent disks at high redshift were expected to stay near-axisymmetric, the observed bar fraction was known to decline strongly by $z\approx1$ (redshift $z$ parametrizes cosmic time: $z=1$ corresponds to about 6 Gyr after the Big Bang, $z=3$ to about 2 Gyr), and cosmological hydrodynamical simulations such as TNG50 and NewHorizon predicted bars to be almost absent beyond $z=1.5$ in Milky-Way-like progenitors (Zana et al. 2022; Reddish et al. 2022) — a deficit sometimes called the 'missing bars' problem.

JWST near-infrared imaging upset this picture. Costantin et al. (2023) identified ceers-2112, a barred spiral at photometric redshift $z\approx3$ (redshift inferred from imaging colors rather than a spectrum) with stellar mass $M_\star=3.9\times10^9\,M_\odot$ — a low-mass Milky-Way progenitor whose bar, from its star-formation history, must have assembled in only about 400 Myr. The authors inferred that the disk became dynamically cold after a rapid episode of gas consumption and that baryons (stars plus gas) already dominated over dark matter in the inner galaxy at $z\approx3$; contemporaneous JWST surveys were reporting other barred galaxies at $z>2$. This left a coherent set of unresolved questions: how common early bars actually are; which physical conditions (gas fraction, star-formation efficiency, turbulence, baryon versus dark-matter dominance) allow a bar to grow so early; whether such early bars are transient or survive to become the barred disks of the present day; and what ingredients cosmological simulations lack that prevents them from producing low-mass ($M_\star<10^{10}\,M_\odot$) barred galaxies beyond $z=1.5$.

Problem Statement

Establish the physical origin of stellar bars in disk galaxies at $z\gtrsim2$, within roughly the first three billion years of cosmic time. The question, raised by the discovery of the low-mass barred spiral ceers-2112 at $z\approx3$ (Costantin et al. 2023), comprises two jointly linked parts serving one objective: (i) determine the causal conditions that permit bar formation so early — the relative roles of gas fraction, star-formation efficiency, turbulence, and baryon versus dark-matter dominance in the host disks — through direct measurements (e.g., molecular-gas masses, stellar and gas kinematics) and/or controlled theoretical or simulation-based analysis; and (ii) identify the physical ingredients (feedback implementation, gas treatment, dark-matter physics, interaction and merger history) that cosmological simulations must include to reproduce the observed early-bar population, particularly barred galaxies with $M_\star<10^{10}\,M_\odot$ at $z>1.5$.

References

1. Costantin, L., et al. (2023). A Milky Way-like barred spiral galaxy at a redshift of 3. Nature 623, 499–501, doi:10.1038/s41586-023-06636-x; arXiv:2311.04283. https://arxiv.org/abs/2311.04283
2. Le Conte, Z. A., et al. (2024). A JWST investigation into the bar fraction at redshifts 1 ≤ z ≤ 3. Monthly Notices of the Royal Astronomical Society 530, 1984–2000, doi:10.1093/mnras/stae921; arXiv:2309.10038. https://arxiv.org/abs/2309.10038
3. Guo, Y., et al. (2025). The Abundance and Properties of Barred Galaxies out to z ~ 4 Using JWST CEERS Data. The Astrophysical Journal 985, 181, doi:10.3847/1538-4357/adc8a7; arXiv:2409.06100. https://arxiv.org/abs/2409.06100
4. Géron, T., et al. (2025). Galaxy Zoo CEERS: Bar fractions up to z ~ 4.0. The Astrophysical Journal 987, 74, doi:10.3847/1538-4357/add7d0; arXiv:2505.01421. https://arxiv.org/abs/2505.01421
5. Ivanov, D., et al. (2026). Discovery of a Barred-spiral Galaxy at z_spec = 3.16. I. Bar Identification and Properties. The Astrophysical Journal 1007, 24, doi:10.3847/1538-4357/ae8023; arXiv:2606.23792. https://arxiv.org/abs/2606.23792
6. Cheng, Y., et al. (2026). Discovery of a Barred-spiral Galaxy at z_spec = 3.16. II. The Star Formation History. The Astrophysical Journal 1004, 248, doi:10.3847/1538-4357/ae75fe; arXiv:2606.23793. https://arxiv.org/abs/2606.23793
7. Boogaard, L. A., et al. (2026). A stellar bar hidden in an extreme gas-rich disk galaxy at z = 4.055. The Astrophysical Journal Letters 1006, L29, doi:10.3847/2041-8213/ae84c7; arXiv:2605.15273. https://arxiv.org/abs/2605.15273
8. Dattathri, S., et al. (2026). Self-interacting dark matter promotes bar formation in disk galaxies. arXiv:2606.27480 (submitted to MNRAS). https://arxiv.org/abs/2606.27480
9. Lokas, E. L. (2025). High-redshift Merger-induced Bar-like Galaxies in IllustrisTNG. The Astrophysical Journal Letters 991, L52, doi:10.3847/2041-8213/ae07c2; arXiv:2509.22083. https://arxiv.org/abs/2509.22083
10. Kalita, B. S., et al. (2026). Galactic bars are already mature at Cosmic Noon: bar strength and flatness at z ~ 1.5. The Astrophysical Journal 997, 247, doi:10.3847/1538-4357/ae2755; arXiv:2512.04163. https://arxiv.org/abs/2512.04163
11. Huang, S., et al. (2025). Large gas inflow driven by a matured galactic bar in the early Universe. Nature 641, 861–865, doi:10.1038/s41586-025-08914-2; arXiv:2511.16078. https://arxiv.org/abs/2511.16078
12. Umehata, H., et al. (2025). ADF22-WEB: A giant barred spiral starburst galaxy in the z = 3.1 SSA22 protocluster core. Publications of the Astronomical Society of Japan 77, 432–445, doi:10.1093/pasj/psaf010; arXiv:2410.22155. https://arxiv.org/abs/2410.22155
13. Reddish, J., et al. (2022). The NewHorizon simulation – to bar or not to bar. Monthly Notices of the Royal Astronomical Society 512, 160–185, doi:10.1093/mnras/stac494; arXiv:2106.02622. https://arxiv.org/abs/2106.02622
14. Zana, T., et al. (2022). Morphological decomposition of TNG50 galaxies: methodology and catalogue. Monthly Notices of the Royal Astronomical Society 515, 1524–1543, doi:10.1093/mnras/stac1708; arXiv:2206.04693. https://arxiv.org/abs/2206.04693
15. Worrakitpoonpon, T. (2025). Bar instability and formation timescale across Toomre's Q parameter and central mass concentration: slow bar formation or true stability. The Astrophysical Journal 979, 166, doi:10.3847/1538-4357/ada35f; arXiv:2412.18098. https://arxiv.org/abs/2412.18098
16. Kuhn, V., et al. (2024). JWST Reveals a Surprisingly High Fraction of Galaxies Being Spiral-like at 0.5 ≤ z ≤ 4. The Astrophysical Journal Letters 968, L15, doi:10.3847/2041-8213/ad43eb; arXiv:2312.12389. https://arxiv.org/abs/2312.12389