Published January 20, 2026 | Version Published
Journal Article Open

Early Stages of Dusty Tori: The First Infrared Spectra from a Highly Multiscale Quasar Simulation

  • 1. ROR icon California Institute of Technology
  • 2. ROR icon Drexel University

Abstract

We present the first infrared (IR) spectral predictions from a self-consistent simulation of the formation of a quasar in a starburst galaxy, spanning the cosmological environment to scales well below the dust sublimation region. The IR emission is dominated by a torus-like dust structure composed of a highly magnetized, turbulence-supported outer accretion disk and of accreting gas tidally torn from the interstellar medium (ISM). At these early stages, the active galactic nucleus is buried and Compton thick. The near- to mid-IR escaping luminosity varies by almost an order of magnitude across sight lines, largely due to extinction from the inflowing stream of cold dust. Self-absorption within the torus suppresses silicate emission features, and further reprocessing by the ambient ISM leads to prominent silicate absorption and colder IR emission. The sublimation structure is stratified by composition and size, producing sight-line-dependent extinction curves that intrinsically vary in shape. However, after repeated scattering in the optically thick dusty medium, these curves emerge substantially grayed. We also demonstrate that bipolar outflows from the central black hole, which carves biconical cavities and reveals the central engine in later stages, can preserve IR anisotropy and silicate features. These results suggest that dusty starburst quasars can undergo a buried, IR-bright phase early in their evolution.

Copyright and License

© 2026. The Author(s). Published by the American Astronomical Society. Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.

Acknowledgement

The authors thank Christian Knigge, Matthew Temple, Samantha Creech, Sam Ponnada, Kyle Kremer, and Yashvardhan Tomar for insightful discussions and comments. The numerical calculations in this paper were run on the Texas Advanced Computing Center (TACC) allocation AST21010. J.B. acknowledges support from a Natural Sciences and Engineering Research Council of Canada (NSERC) doctoral scholarship. P.F.H. acknowledges support from a Simons Investigator Grant.

Software References

SKIRT3 : P. Camps & M. Baes (2020); yt4 : M. J. Turk et al. (2011); numpy5 : C. R. Harris et al. (2020) matplotlib6 : J. D. Hunter (2007) scipy7 : P. Virtanen et al. (2020) astropy8 : Astropy Collaboration et al. (2018) pandas9 : Pandas Development Team (2020) scikit-image10 : S. Van der Walt et al. (2014).

Files

Bardati_2026_ApJ_997_10.pdf

Files (4.6 MB)

Name Size
md5:6d77dc381ccebccfcd5c41447217761e
4.6 MB Preview Download

Additional details

Related works

Is new version of
Discussion Paper: arXiv:2509.09770 (arXiv)

Funding

Natural Sciences and Engineering Research Council
Simons Foundation

Dates

Submitted
2025-09-10
Accepted
2025-12-05
Available
2026-01-12
Published