Connecting the Dots: UV-Bright Companions of Little Red Dots as Lyman-Werner Sources Enabling Direct Collapse Black Hole Formation

TL;DR

The paper tests whether UV-bright companions near Little Red Dots (LRDs) supply the Lyman-Werner radiation needed to suppress molecular cooling and enable direct-collapse-like formation of massive black hole seeds in the early universe. Using JWST imaging of 83 spectroscopically confirmed LRDs, the authors perform component-resolved analysis to identify UV-bright companions and measure their separations, then compute local LW fields $J_{LW}$ from the companion SEDs, finding values in the range $J_{LW} \sim 10^{2.5}-10^{5}$, often exceeding the threshold $J_{crit} \sim 10^{3}$. A two-stage component identification, fixed-$n$ Sérsic fitting, and forced photometry yield a clear red component plus UV-bright companions, with the blue companions typically having $M_{\star} \sim 10^{8}$–$10^{9} M_\odot$ and the red component showing a steep, BH*-like continuum. The results provide observational support for a LW-regulated direct-collapse channel linking the observed morphologies and spectra of LRDs to environments conducive to rapid, fragmentation-free collapse, while highlighting biases and the need for resolved spectroscopy to confirm metallicity, shielding, and dynamics.

Abstract

We compile a sample of 83 Little Red Dots (LRDs) with JWST imaging and find that a substantial fraction ($\sim$43%, rising to $\gtrsim$85% for the most luminous LRDs) host one or more spatially offset, UV-bright companions at projected separations of $0.5\rm \, kpc \lesssim d\lesssim 5 \rm \,kpc$, with median of $\langle d \rangle = 1.0\,\mathrm{kpc}$. This fraction is even higher when smaller spatial scales are probed at high S/N ratio: we show that the two most strongly lensed LRDs known to date, A383-LRD and the newly discovered A68-LRD, both have UV-bright companions at separations of only $d\sim0.3$ kpc, below the resolution limit of most unlensed JWST samples. We explore whether these ubiquitous red/blue configurations may be physically linked to the formation of LRDs, in analogy with the "synchronized pair" scenario originally proposed for direct-collapse black hole formation. In this picture, ultraviolet radiation from the companions, which typically have modest stellar masses ($M_\ast \sim 10^{8-9}M_\odot$), suppresses molecular hydrogen cooling in nearby gas, allowing nearly isothermal collapse and the formation of extremely compact objects, such as massive black holes or quasi-stars. Using component-resolved photometry and SED modeling, we infer Lyman-Werner radiation fields of $J_{21,LW} \sim 10^{2.5}$-$10^{5}$ at the locations of the red components, comparable to those required in direct-collapse models, suggesting that the necessary photodissociation conditions are realized in many LRD systems. This framework provides a simple and self-consistent explanation for the extreme compactness and distinctive spectral properties of LRDs, and links long-standing theoretical models for early compact object formation directly to a population now observed with JWST in the early universe.

Figures (6)

• Figure 1: RGB composites of the full LRD sample. All images are $1.5\arcsec\times1.5\arcsec$, constructed from the available JWST/NIRCam filters for each object, selected as a function of redshift to approximately sample rest-frame UV emission (blue), wavelengths near the Balmer/4000 Å break (green), and redder rest-frame optical emission (red), depending on filter availability. For the three strongly lensed LRDs, the lensing magnifications are indicated in the lower left corner. The images are shown for illustrative purposes; all quantitative measurements are derived from the structural modeling described in Section \ref{['sec:structuralmodeling']}.
• Figure 2: Same as Fig. 1, but showing residual images from the GALFIT modeling after subtracting the red component. The fits are performed in F200W and then applied to the other bands with fixed structural parameters and free magnitudes. To highlight faint UV emission, the images are displayed with a more aggressive scalings and contrast than in Fig. 1. Symbols indicate the companion classification: a check mark ($\checkmark$) denotes a robust companion, while a question mark (?) indicates a tentative candidate, identified either through a dropout consistency or the presence of faint residual emission not robustly isolated as an independent source (see Section \ref{['sec:discussion:companionfraction']} for details). The generally low residuals demonstrate that the F200W-based model provides a good representation of the source morphology across bands.
• Figure 3: Top: JWST/NIRCam cutouts in three filters (F115W, F200W, F444W), the corresponding segmentation mask, best-fit GALFIT model, and residual image for A68-LRD1. Ellipses indicate the fitted component parameters from F200W, overlaid on all bands. The gray circle shows the 0.5$\arcsec$ photometric aperture. Bottom left: SED decomposition for A68-LRD1. Gray points show the observed total (observed) aperture photometry, while the red and blue points show the forced photometry obtained from the GALFIT fits shown in the top panel, adopting the same color scheme. Solid curves show the best-fit EAZY templates. These SEDs are shown in the observed frame, meaning that they include lensing magnification and IGM absorption. Overplotted is the Cliff spectrum from Graaff2025_Cliff, only corrected for the redshift difference (without any renormalization), showing remarkable resemblance to the SED shape and intrinsic luminosity. Bottom right: Stacked intrinsic rest-frame SEDs for all LRDs with companions. For each system, the total SED (gray) is normalized to the total flux density at a rest-frame wavelength of 0.3645$\mu$m (Balmer break). The red and blue component SEDs are shown relative to this normalization. Shaded regions indicate object-to-object scatter (16th–84th percentile). For systems with multiple companions, the blue curve represents the sum of all companions.
• Figure 4: The total LW magnitude of the companion(s), $M_{\rm LW,tot}$, is shown as a function of the effective projected separation from the compact red component, $d_{\rm eff}$ (defined such that $L_{\rm LW,tot}/d_{\rm eff}^2 = \sum_i L_{{\rm LW},i}/d_i^2$ when an LRD has multiple companions). Diagonal lines indicate constant LW radiation intensity, $J_{21,\rm LW}$, incident on the red component. Systems toward the upper left correspond to brighter companions at smaller separations and therefore higher LW intensities. The red dashed line marks a commonly adopted critical threshold for molecular hydrogen dissociation ($J_{\rm crit}=10^3$), above which fragmentation is expected to be strongly suppressed. Shaded regions illustrate increasing likelihood for collapse in an LW-regulated regime.
• Figure 5: Left panel: Effective projected separation between the compact red component and its associated companion(s) as a function of the inferred luminosity $L_{5100}$. The luminosity is measured using the NIRCam filter closest to rest-frame $5100\,\AA$ and converted to a luminosity using the source redshift. The top panel shows the companion fraction as a function of luminosity; we find a striking trend where the companion fraction increases with $L_{5100}$, reaching nearly 100% completeness for the most luminous LRDs in our sample. Right panel: the companion(s) UV magnitude (rest-frame 1500$\rm \AA$) as a function of $L_{5100}$, indicating that companions associated with these high-luminosity LRDs are themselves significantly brighter in the UV.