CO Emission in Infrared-selected Active Galactic Nuclei

Abstract

In order to better understand how active galactic nuclei (AGN) effect the interstellar media of their host galaxies, we perform a meta-analysis of the CO emission for a sample of z = 0.01−4 galaxies from the literature with existing CO detections and well-constrained AGN contributions to the infrared (67 galaxies). Using either Spitzer/IRS mid-IR spectroscopy or Spitzer+Herschel colors we determine the fraction of the infrared luminosity in each galaxy that can be attributed to heating by the AGN or stars. We calculate new average CO spectral line ratios (primarily from Carilli & Walter 2013) to uniformly scale the higher-J CO detections to the ground state and accurately determine our sample's molecular gas masses. We do not find significant differences in the gas depletion timescales/star formation efficiencies (SFEs) as a function of the mid-infrared AGN strength (f (AGN) MIR or L IR (AGN)), which indicates that the presence of an IR-bright AGN is not a sufficient sign-post of galaxy quenching. We also find that the dust-to-gas ratio is consistent for all sources, regardless of AGN emission, redshift, or L IR , indicating that dust is likely a reliable tracer of gas mass for massive dusty galaxies (albeit with a large degree of scatter). Lastly, if we classify galaxies as either AGN or star formation dominated, we do not find a robust statistically significant difference between their CO excitation.

The ALMA Spectroscopic Survey in the Hubble Ultra Deep Field: CO Excitation and Atomic Carbon in Star-forming Galaxies at z = 1–3

“…Based on the low number of sources, we are unable to draw strong conclusions here. However, the results are consistent with recent work that did not find statistically different excitation, up to CO(7-6), between galaxies with and without an active nucleus (e.g., Sharon et al 2016;Kirkpatrick et al 2019).…”

“…These studies find that the CO ladders of SMGs can continue to rise up to J∼7, testifying to a warm and dense (  n 10 5.5 cm −3 ) ISM. The differences between the (low-J) CO excitation in SMGs and (mid-IR selected) AGNs have not been found to be statistically significant (Sharon et al 2016;Kirkpatrick et al 2019).…”

More than star formation: High-J CO SLEDs of high-z galaxies

“…IR-bright AGN are known to produce CO SLEDs that peak near J upper ∼ 7 (e.g., Weiß et al 2007), while starburst-dominated galaxies like SMGs have CO SLEDs that peak near J upper ∼ 5 (e.g., Carilli & Walter 2013). While initial systematic differences in low-J CO line ratios between SMGs and AGN (e.g., Swinbank et al 2010;Harris et al 2010;Ivison et al 2011;Danielson et al 2011;Riechers et al 2011;Thomson et al 2012) were not statistically significant in larger samples (Sharon et al 2016), a more quantitative classification of AGN strength based on mid-IR spectral features suggests systematically higher excitation CO line ratios across the entire SLED for AGN (albeit with limited statistical power; Kirkpatrick et al 2019). Looking for evidence of AGN feedback via CO excitation not only probes the fraction of the molecular gas affected by the AGN, but can also constrain how the AGN interacts with the gas (for example, via X-ray or shock heating; e.g., van der Werf et al 2010; Meijerink et al 2013).…”

The impact of AGN on the life of their host galaxies at z ∼ 2

“…When dealing with SFRs of AGN hosts, an additional complication is the difficulty to properly account for the AGN contribution to the far-infrared luminosity. Different methods to estimate the AGN contribution can lead to different results and place the AGN population on the same star-formation law of normal galaxies (Kirkpatrick et al 2019). Finally, AGN samples at high redshift are usually small and likely biased toward brighter objects (e.g.…”

“…Brusa et al 2018) or are heterogeneous when assembled from literature data (e.g. Fiore et al 2017;Perna et al 2018;Kirkpatrick et al 2019).…”