Quasar accretion disks are gaseous structures around supermassive black holes where matter heats up to hundreds of thousands of degrees and shines from X-rays to infrared. They cannot be seen directly: even for the nearest quasars, the disk subtends an angle smaller than one millionth of an arcsecond. So astronomers use indirect methods: reverberation mapping (the response of emission to central variability) and gravitational microlensing (brightening by stars in the lensing galaxy). Both methods give sizes several times larger than predicted by the classical thin-disk model of Shakura and Sunyaev. However, previous microlensing estimates relied on heterogeneous ultraviolet data and were limited to inner disk regions with radii up to five light-days. The work by Motta and Mediavilla extends the measurements to 14 light-days using a homogeneous set of near-infrared spectra.
The researchers took Hubble Space Telescope data for seven gravitationally lensed quasars, a total of 21 image pairs. In the spectra, they isolated narrow emission lines, such as [OIII] at a wavelength of 5007 angstroms, which arise in an extended region and are not affected by microlensing. By comparing the brightness of the continuum and these lines in different images, the authors calculated the microlensing amplitude. They then modeled magnification maps for stellar masses in the lensing galaxy and convolved them with Gaussian disk profiles of different sizes. A Bayesian analysis yielded the most likely disk radius at the reference wavelength of 5007 Å, and combining it with a previous measurement at 1026 Å gave the slope of the size-wavelength relation.
The average disk radius at 5007 Å is 13.3+2.9−2.7 light-days. This is three times larger than the thin-disk model predicts. The power-law index of the size-wavelength relation is 0.68±0.23. The theoretical value of 4/3 is rejected at the 98% confidence level (a 2.33-sigma deviation). The result is consistent with reverberation mapping estimates, which give p around 0.8–0.9, though their uncertainties are larger. The disk size curve from X-ray to infrared lies systematically above the thin-disk prediction.
The work shows that the discrepancy with thin-disk theory is not limited to the inner regions but persists at radii twice as large. This complicates explanations involving contamination from the broad-line region or winds: any additional component must contribute to the emission from X-rays to optical without upsetting the energy balance. Among neighboring microlensing studies of quasars, this investigation stands out for using narrow emission lines in the near-infrared, which provide a more reliable zero point for microlensing than broad-line cores. Other groups have used reverberation mapping or ultraviolet microlensing, but they did not have such a wide wavelength baseline.
The next step could be to expand the sample of lensed quasars with high-resolution infrared spectra. New observations with the James Webb Space Telescope will provide even more precise temperature profiles and test whether the deviation persists at radii beyond 14 light-days. Theorists will need to revise the thin-disk model, possibly incorporating magnetic fields, turbulence, or nonthermal processes.
The results will affect models of active galactic nuclei and estimates of supermassive black hole masses, which often rely on accretion disk sizes. They are also important for interpreting quasar variability in future sky surveys.
The plan is to apply the same method to a larger number of lensed quasars and add mid-infrared data. It is also necessary to refine the stellar mass distribution in lensing galaxies to reduce the systematic uncertainty in size.
For the astrophysics of accretion disks, the question of angular momentum transport and heating mechanisms remains open. The deviation of the temperature profile from the three-quarter law indicates that the standard alpha viscosity model is incomplete. The result also relates to the origin of broad emission lines and the role of winds in quasars.
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
The authors used microlensing of seven quasars with near-infrared Hubble data, anchoring on narrow lines to measure the amplitude. They obtained a disk radius of 13.3 light days at 5007 Å, three times larger than theory, and a size–wavelength slope of 0.68±0.23, which rejects 4/3 at 98% confidence.
Where its strength is
Using narrow emission lines in the near-infrared as a non-microlensed reference point provides a more reliable calibration, and extending to 14 light days shows that the discrepancy with the thin disk persists at larger radii.
Where it could go next
Worth trying
Why this may matter
If the discrepancy with thin-disk theory holds at larger radii and across different wavelength regimes, it would require revising basic assumptions about quasar accretion disk structure, which is important for understanding supermassive black hole growth and their feedback with galaxies.
Nearby works in our archive
The archive is sparse around this work: the closest review is only 0.54, against a usual 0.64. It looks like few have drilled nearby.
A disk size of 13 light-days means that light from its far side takes 13 days longer to reach us than from the near side. If such a disk were placed in the Solar System, its edge would lie far beyond the orbit of Neptune.