The Case for Super-Eddington Accretion: Connecting Weak X-ray and UV Line Emission in JWST Broad-Line AGN During the First Gyr of Cosmic Time

TL;DR

JWST reveals an overabundance of early-universe SMBHs whose multi-wavelength signatures challenge standard sub-Eddington accretion models. The authors combine deep X-ray upper limits, JWST NIRSpec line measurements, and Cloudy simulations to test whether high Eddington ratios can explain the observed X-ray weakness and lack of high-ionization UV lines in z~5 broad-line AGN. They find pronounced X-ray weakness and non-detections of C IV/He II/Ne V, and show that slim-disk, super-Eddington SEDs naturally suppress UV/X-ray emission and raise Balmer decrements, without invoking heavy dust attenuation. The results imply that super-Eddington accretion may be common in the early Universe, reducing the need for massive seed black holes and informing bolometric corrections and growth environments for the first SMBHs.

Abstract

A multitude of JWST studies reveal a surprising over-abundance of over-massive accreting super-massive blackholes (SMBHs) -- leading to a deepening tension between theory and observation in the first billion years of cosmic time. Across X-ray to infrared wavelengths, models built off of pre-JWST predictions fail to easily reproduce observed AGN signatures (or lack thereof), driving uncertainty around the true nature of these sources. Using a sample of JWST AGN identified via their broadened Halpha emission and covered by the deepest X-ray surveys, we find neither any measurable X-ray emission nor any detection of high-ionization emission lines frequently associated with accreting SMBHs. We propose that these sources are accreting at or beyond the Eddington limit, which reduces the need for efficient production of heavy SMBH seeds at cosmic dawn. Using a theoretical model of super-Eddington accretion, we can produce the observed relative dearth of both X-ray and ultraviolet emission, as well as the high Balmer decrements, without the need for significant dust attenuation. This work indicates that super-Eddington accretion is easily achieved through-out the early Universe, and further study is required to determine what environments are required to trigger this mode of black hole growth.

Figures (7)

• Figure 1: Significant X-ray Weakness: (Left Panel) We show the upper limits of $\alpha_{\mathrm{OX}}$ for the JWST z$\sim5$ BL AGN sample (purple triangles). The blue contours are the spectroscopically-selected BL AGN sample derived from SDSS and confirmed by lusso17. The blue solid line shows the $\alpha_{\mathrm{OX}}$ relationship parameterized by lusso17. The dash-dot, dashed, and dotted lines show the 1$\sigma$, 3$\sigma$, and 5$\sigma$ scatter, respectively. The blue points are the high-z Type 1 BL AGN sample (z $> 3$) from sacchi22. The orange points are candidate super-Eddington sources from the WISSH QSO sample zappacosta20. (Right Panel) We show the offset from the lusso17 relationship ($\Delta$$\alpha_{\mathrm{OX}}$) where the colors are consistent with the previous panel. The red dashed line is the canonical X-ray weakness threshold ($\Delta$$\alpha_{\mathrm{OX}}$$< 0.3$, as is shown in laurenti22).
• Figure 2: Lower than predicted M$_{\mathrm{BH}}$: The dark purple points are the values from maiolino23. The light orange line is the relation between black hole and stellar mass for a sample of bulge dominated AGN and total sample of local AGN from reines15. The green line is the black hole to stellar mass relationship derived from greene20. The light purple triangles are the upper-limits of the black hole mass of our sample that overlaps with maiolino23, derived from the upper-limits of the X-ray power slope, $\Gamma$. These black hole mass upper limits are, on average, an order of magnitude below maiolino23.
• Figure 3: JWST/NIRSpec medium resolution (R$\sim1000$) spectroscopy for the sources with G140M coverage from JADES deugenio24. For each galaxy, the three subpanels show zoomed in regions centered on the expected locations of the C iv, $\mathrm{H}{\beta}$ + [O iii], and $\mathrm{H}{\alpha}$ + [N ii] emission lines, respectively. The high-ionization C iv emission line is undetected in every source that has wavelength coverage. Details can be found in the Methods section.
• Figure 4: Predicted line luminosities for C iv, He ii, and [Ne v], respectively. Dark blue is sub-Eddington prescription (agnsed), light purple is slim disk prescription (agnslim). Red triangles are median 3 sigma upper limits from the JWST/NIRSpec G140M/G235M spectra.
• Figure 5: Fits to the $\mathrm{H}{\alpha}$ + [N ii] emission lines for the 8 sources in the sample with NIRSpec/G395M observations from JADES deugenio24 in order of increasing redshift. Each is fit with a combination of four Gaussians: narrow components to $\mathrm{H}{\alpha}$ (light green) and both [N ii] lines (blue and red), plus a broad component to the $\mathrm{H}{\alpha}$ line (dark green). Best fit parameters for each individual Gaussian are shown on the top right of each plot, and the combined fit (purple) values are shown on the left of each plot. Details are described in the Methods section and values can be found in Table \ref{['tab:emlines']}.