Astro2020 Science White Paper

Tracing the feeding and feedback of active galaxies

Thematic Areas: � Planetary Systems � Star and Planet Formation� Formation and Evolution of Compact Objects � Cosmology and Fundamental Physics� Stars and Stellar Evolution � Resolved Stellar Populations and their EnvironmentsX Galaxy Evolution � Multi-Messenger Astronomy and Astrophysics

Principal Author:Name: Enrique Lopez-RodriguezInstitution: SOFIA Science Center / NASA Ames Researcher CenterEmail: enrique.lopez-rodriguez@nasa.govPhone: 6506045754

Co-authors:Robert Nikutta, National Optical Astronomy Observatory (NOAO)Nancy Levenson, Space Telescope Science Institute (STScI)Chris Packham, University of Texas at San Antonio (UTSA)Erin K. S. Hicks, University of Alaska Anchorage (UAA)Kohei Ichikawa, Tohoku University, Sendai, JapanSibasish Laha, University of California San Diego (UCSD)Matt Malkan, University of California Los Angeles (UCLA)Claudio Ricci, University Diego Portales, Chile; Kavli Institute for Astronomy and Astrophysics,Peking University, ChinaVivian U, University of California Irvine (UCI)Stephanie Juneau, National Optical Astronomical Observatory (NOAO)

Abstract:Supermassive black holes in active galaxies accrete mass from their surroundings, and inject

energy as well as mass to the galactic medium through outflows. Strong accretion signatures ontothe active nucleus are evident at sub-pc scales. Their host galaxies show the presence of outflowingand inflowing material from hundreds to tens of parsecs. The interface between the active nucleiand their host galaxies is a region of a few pc in size with a gas and dust flow cycle, a so-calledtorus. The torus and accretion disk fuel accretion and launch outflows, which fundamentally con-nect the black holes to their host galaxies. Under this scheme, the torus is dynamically connectedto galaxy evolution. However, many key questions remain: What is the origin and evolution of thetorus? What is the structure of the torus during its existence? What is the feeding and feedbackrole of the torus? The lack of observational capabilities that can resolve the torus hinders a fullunderstanding of the feeding material building up the mass of the central black holes in galaxies.New capabilities that resolve this interface will present empirical constraints that will be used tophysically guide the theories of galaxy evolution.

1

The physical interface between active nuclei and their host galaxiesThe relation between the masses of host galaxies and their central supermassive black holes (BHs)(e.g., Gebhardt et al. 2000) implies an evolutionary connection. However, the physical process bywhich this occurs is uncertain. The buildup of a central BH may occur over 107 − 109 yr (e.g.,Marconi et al. 2004), with active spurts of accretion over 106 − 108 yr (Netzer 2015). Hydrody-namical simulations suggest that early BH growth is limited by stellar feedback, which expels gasfrom the galactic nuclei (Angles-Alcazar et al. 2018). However, the immediate surroundings ofactive galactic nuclei (AGN) hide the direct signatures of this activity—which proceeds throughan accretion disk—along certain lines of sight. This obscuring material, the so-called torus, is alsoimportant as the physical interface between the central engine and host galaxy. The torus is theimmediate source of fuel for both accretion and outflow.

For a few decades, the best picture of the material between the active nucleus and the hostgalaxy has been a static structure that reprocesses the emission from the central engine. In its mostsimplistic geometrical form, this interface is an obscuring toroidal distribution of dusty and molec-ular matter. Some of the evidences for this structure are the lack of broad emission lines in somehidden AGN (Tran et al. 2003), the emergence of broad lines in polarized light in hidden AGN(Antonucci & Miller 1985), and the detection of discrete X-ray obscuration events (Markowitz etal. 2014). Infrared (IR) and X-ray observations suggest that the dust distribution is clumpy (RamosAlmeida et al. 2009, Markowitz et al. 2014), producing an emission morphology within the central10 pc that varies with wavelength (Ramos Almeida & Ricci 2017, Lopez-Rodriguez et al. 2018a).

Figure 1: The morphology of the dust emission in the central 10 pc of NGC 1068 varies as a function ofwavelength and shows more extended structures connecting with the galaxy. The IR dust emission arisesfrom a compact pc-scale structure dominated by hot dust and a larger polar emission by thin dust (Left;Lopez-Gonzaga et al. 2014). The cold dust is mainly concentrated in the equatorial plane, which representsthe bulk of dust distribution in the torus (Middle; Garcıa-Burillo et al. 2016). The broadband cold reflectedX-ray emission of NGC 1068 can be explained by multiple reflectors with three different column densities(Right; Bauer et al. 2015). The highest NH component is the dominant contribution to the Compton hump.

The absorbed and reprocessed X-ray radiation is produced in a compact source within several grav-itational radii of the central engine (Zoghbi et al. 2012). X-ray surveys have shown that the coldcomponent with the highest column density (NH ≥ 1024 cm−2) covers ∼20–30% of the AGN (e.g.,Burlon et al. 2011, Ricci et al. 2015). Some of the most surprising results come from ALMA,which resolved the obscuring structure of NGC 1068, an AGN with a fully obscured core. Theseobservations indicate that the torus of NGC 1068 is a rotating, turbulent, compact and inhomoge-neous structure of ∼ 10 pc (∼ 0.16′′) in diameter (Garcia-Burillo et al. 2016, Imanishi et al. 2016,

2

2018) with possible outflows (Gallimore et al. 2016). However, in general, IR interferometricobservations show that most of the MIR emission of Seyfert galaxies arises from a polar elongatedstructure (Honig et al. 2012, 2013). Our current understanding has formed using sub-mm obser-vations with ALMA, IR single-dish observations with 8–10 m class telescopes, IR interferometricobservations with the Very Large Telescope Interferometer (VLTI), and X-rays (Figure 1). Thesestudies, with the exception of ALMA, are based on the integrated signature of the spatially unre-solved emission across the electromagnetic spectrum. Because of the small physical scale (≤10pc) of this structure, we cannot currently characterize the torus in detail.

Figure 2: The 1-13 µm dust emission of the torus of NGC 1068 can be resolved using 30-m class tele-scopes. The first row shows the pupil images of the JWST (6m), Keck (10m), GMT (25m), and TMT (30m)left to right. From the second row, the first column shows, at several wavelengths, the 2D brightness map ofa CLUMPY torus model of NGC 1068 with ’infinite’ resolution. The remaining panels show the syntheticobservations with each telescope (columns) at each wavelength (rows).

In the next decade, we will resolve the physical interface between the AGN and host galaxy.To obtain useful constraints on the putative torus, it is important to improve the spatial resolutionand sensitivity of observational studies. Then, we can measure torus sizes, covering factors, in-clinations, optical depths and masses at high spatial resolution. For example, using the radiativetransfer model CLUMPY (Nenkova et al. 2008a,b) and assuming the torus parameters inferred fromthe IR spectral energy distribution (SED) by Lopez-Rodriguez et al. (2018a), the torus emission inthe 1 − 13 µm wavelength range of NGC 1068 is resolvable using 30-m class telescopes (Figure2). Specifically, the torus is very compact, ≤ 2 pc, at near-IR (NIR) wavelengths while it is moreextended, ≤ 10 pc, in the mid-IR (MIR). At NIR, the hot dust (1000 − 1500 K) is located at theinner edge of the torus where it is directly irradiated by the central engine. In the MIR, the emittingdust is located further away and/or in the shadowed regions of the dusty clumps, which makes thetorus appear more extended. While the James Webb Space Telescope (JWST) will have sensitive

3

NIR and MIR instruments that will resolve the gas out to hundreds of pc, the origin of the torusmay be traced to the smaller-scale torus interface. Recently, observations using IR interferometryand 10-m class telescopes of the central ∼ 40 pc of Circinus have been explained by the contri-bution of an equatorial disk and a hyperboloid/conical wind model (Stalevski et al. 2017, 2019).This unique study uses the current state-of-the-art facilities and can only be performed in a handfulof objects. The next generation of 30-m telescopes and the already on-going theoretical modelingwill explain the physical connection between the AGN and the host galaxy in a statistically sig-nificant sample of AGN. Assuming a 15 mas angular resolution at 2.2 µm, and 68 mas at 10 µmfor a 30-m telescope, and using the above typical torus sizes of 2 pc at 2.2 µm and 20 pc at 10µm, the potential AGN which have resolvable tori (torus size at least twice the angular resolutionat a given wavelength) need to be at a distance ≤ 50 Mpc. Several tens of AGN located within 50Mpc could be resolved with the next generation of 30-m class telescopes, which will allow for acomprehensive physical account of the torus.

Towards a dynamical AGN torusIn the last ten years, we learned that AGN inject mass to the galactic medium through outflowsobserved on scales from tens to a few hundred pc. Mapping the distribution and kinematics ofhigh excitation lines in both the NIR and MIR (e.g. 1.08 µm He I, 1.64 µm [Fe II], 1.96 µm [SiVI], 2.17 µm Brγ in the NIR, and 7.65 µm [Ne VI], 12.81µm [Ne II] in the MIR) has verified thatthe NIR high excitation lines trace an outflow. The typical full-width-at-half-maximum (FWHM)is estimated to be 25 pc, and further modeling places tight constraints on the outflow speeds andmass loss rates out to scales of a few hundred parsecs (Mueller Sanchez et al. 2011). Line emissionfrom molecular hydrogen (rovibrational lines in the NIR, e.g. 2.12 µm H2 1-0 S(1), and rotationalS(2) to S(13) H2 lines in the 3-13 µm range) has primarily revealed the morphology, mass, andkinematics of the molecular gas in the galaxy disk as well as traced the inflowing and outflowingmaterial down to scales near the torus (Figure 3).

Although the above results support the unification model (Antonucci 1993) of AGN, we nowhave a new paradigm. The torus is the central structure in a gas flow cycle (Krolik & Begelman1988) in which gas is brought in from the host galaxy disk (inflow) and then driven out radiativelyby an AGN wind (e.g., the Wada (2012) fountain model; Figure 4-a, Schartmann et al. 2011,Dorodnitsyn, Bisnovatyi-Kogan & Kallman 2011, Hopkins et al. 2012, Ricci et al. 2017) and/ormagnetic disk-wind (i.e. Blanford & Payne 1982, Emmering et al. 1992, Konigl & Kartje 1994,Elitzur & Shlosman 2006, Figure 4-b). These views are both supported by the presence of inflowsand outflows in nearby Seyferts (e.g., Garcia-Burillo et al. 2005; Barbosa et al. 2009; Muller-Sanchez et al. 2011; Davies et al. 2014), large reservoirs of molecular gas in their circumnucleardisks (typical sizes ∼ 100 pc, see Hicks et al. 2013; Izumi et al. 2016), and magnetic fields at scalesof few pc (i.e. Lopez-Rodriguez et al 2015, 2018a,c). However, since the torus is only detected asa point-like source, these studies are based on integrated emission of the whole structure. The lackof observational capabilities that can resolve the torus hinders a full understanding of the feedingmaterial building up the mass of the central black holes in galaxies. These resolve observationswill present empirical constraints that can then be used to physically guide the theories of galaxyevolution.

In the coming decade, we will demonstrate more directly the active role of the torus in galaxyevolution. To understand the dynamics of the torus, it is crucial to achieve spectral and angular

4

Figure 3: The H2 2.12 µm emission in Seyfert 1 galaxy NGC 5643 shows evidence of molecular hydrogenin disk rotation, as well as inflow and outflow (Davies et al. 2014). Upper row from left to right: H2 fluxdistribution, HST V-H dust structure, Brγ 2.16 µm emission and velocity. Lower row from left to right:ALMA CO (2-1) velocity (Alonso-Herrero et al. 2018), H2 2.12 µm velocity, non-circular velocity, andvelocity dispersion. The inflow seen in both H2 2.12 µm and CO (2-1) is coincident with the spiral dustlanes indicated by the reddened (dark) regions of the dust structure map and outlined by the orange lines.The outflow indicated by the redshifted and high velocity dispersion H2 2.12 µm emission in the northeastis coincident with the blue (white) region of the dust structure map and the Brγ emission and velocity. It isindicated by the pink lines at the boarder of the outflow region. The warm molecular gas traced by H2 2.12µm traces the cold gas measured by CO (2-1).

resolutions that allow characterizing the dynamics of the gas over a range of temperatures. 1− 13µm spectroscopic observations will facilitate connecting the gas flow cycle on scales of a few pcto the inflows and outflows on larger scales. For example, 2.12 µm H2 emission will be detectedthroughout the central few pc and therefore provide the ability to trace the inflow of gas observedon larger scales from the host galaxy down to the torus (Hicks et al. 2009). A key advantage of theMIR H2 lines is that they trace the molecular gas at the temperatures (∼ 100 K, Izumi et al. 2016)typical of the nuclear disks of AGN rather than the T ∼ 1500 K gas traced by the vibrationallyexcited NIR H2 lines (Rigopoulou et al. 2002; U et al. 2019). By measuring the inflowingand outflowing gas down to the torus scales and across a range of gas and dust temperatures,we will show the effect on host galaxy evolution. Thus, dependencies of these key processes onfundamental AGN and host galaxy properties can be established reliably, advancing current work(e.g. Davies et al. 2014, 2015, 2017, U et al. 2019). Moreover, facilities such as the ngVLA willbe able to characterize the content and the dynamics of the gas at comparable resolution in thecentral few pc, which will allow us to quantify the effect of winds on BH growth (Angles-Alcazaret al. 2018) and explore jet-ISM feedback from the multi-phase perspective (Nyland et al. 2018).

In addition to previous models, a magneto-hydrodynamical (MHD) (i.e. Blanford & Payne1982, Emmering et al. 1992, Konigl & Kartje 1994, Elitzur & Shlosman 2006, Figure 4-b) outflowwind can lift the plasma from the mid-plane of the accretion disk to form a geometrically thick dis-tribution of dusty clouds surrounding the central engine. The potential influence of the magnetic

5

Figure 4: a) Cartoon representing an edge-on view of the nuclear regions of the radiation-driven fountainmodel (Izumi et al. 2018). The traditional torus is the 300 − 1500 K dust component, the extended dustemission and outflows along the polar direction are part of the fountain flow, and the nuclear star formation(SF) is traced by the molecular gas in the circumnuclear disk (CND) as an extension of the dusty torus. b)Hydromagnetic outflow wind model (Emmering et al. 1992) of AGN. Close to the black hole, the accretiondisk is ionized and becomes molecular at larger distances. Dense molecular clouds are lifted and acceleratedaway from the accretion disk moving along the magnetic field. The torus is that region of an outflow windwhere the clouds are optically thick and dusty.

field on the outflowing material and accreting activity can be investigated using its IR polarizationsignature. Dust grains can be aligned by the presence of magnetic fields, described by theories ofradiative torques (RATs, Lazarian & Hoang 2007) and also by intense radiation fields or outflow-ing media. As radiation propagates through these aligned dust grains, preferential extinction ofradiation along one plane leads to a measurable polarization in the transmission/emission of thisradiation, a term called dichroic absorption (at NIR)/emission (at MIR). The short axis of the dustgrains aligns parallel to the local magnetic field lines, and the observed position angle (PA) of thepolarization traces the direction of the magnetic field. Recent studies using IR polarimetry founda coherent magnetic field in the inner wall (∼ 1 pc) of the torus of radio-loud (i.e. IC5063 andCygnus A; Lopez-Rodriguez et al. 2013, 2018c), and radio-quiet (NGC 1068, Lopez-Rodriguezet al. 2015) AGN. These results suggest that the magnetic fields can play an important role in theorigin, kinematics, and evolution of the torus. As higher angular resolution and better sensitiv-ity become available, polarimetric techniques are more efficient. To characterize the origin anddynamics of the torus, it is crucial to provide IR polarimetric capabilities.

As the interface between active nuclei and their host galaxies, AGN tori are dynamically con-nected to galaxy evolution. Completing a quantitative picture of the dust and gas flow cycle istherefore critical in establishing the mechanisms that drive the co-evolution of the central super-massive black holes and their host galaxies.

6

References

Alonso-Herrero, A., Pereira-Santaella, M., Garcıa-Burillo, S., Davies, R. I., Combes, F., Asmus, D., Bunker, A., Dıaz-Santos, T., Gandhi, P., Gonzalez-Martın, O., Hernan-Caballero, A., Hicks, E., Honig, S., Labiano, A., Levenson, N.A., Packham, C., Ramos Almeida, C., Ricci, C., Rigopoulou, D., Rosario, D. Sani, E., Ward, M. J. 2018, ApJ, 859,144

Angles-Alcazar, D., Dave, R., Faucher-Giguere, C.-A., Ozel, F., Hopkins, P., F. 2017, MNRAS, 464, 2840

Antonucci, R. R. J., Miller, J. S. 1985, ApJ, 297, 621

Antonucci, R. 1993, AR&A, 31, 473

Barbosa, F. K. B., Storchi-Bergmann, T., Cid Fernandes, R., Winge, C., Schmitt, H. 2009, MNRAS, 396, 2

Bauer, F. E., Arevalo, P., Walton, D. J., Koss, M. J., Puccetti, S., Gandhi, P., Stern, D., Alexander, D. M., Balokovic,M., Boggs, S. E., Brandt, W. N., Brightman, M., Christensen, F. E., Comastri, A., Craig, W. W., Del Moro, A., Hailey,C. J., Harrison, F. A., Hickox, R., Luo, B., Markwardt, C. B., Marinucci, A., Matt, G., Rigby, J. R., Rivers, E., Saez,C., Treister, E., Urry, C. M., Zhang, W. W. 2015, ApJ, 812, 116

Blandford, R. D., Payne, D. G., 1982, MNRAS, 199, 883

Burlon, D., Ajello, M., Greiner, J., Comastri, A., Merloni, A., Gehrels, N., 2011 ApJ 728, 58

Davies, R. I., Maciejewski, W., Hicks, E. K. S., Emsellem, E., Erwin, P., Burtscher, L., Dumas, G., Lin, M., Malkan,M. A., Muller-Sanchez, F., Orban de Xivry, G., Rosario, D. J., Schnorr-Muller, A., Tran, A. 2014, ApJ, 792, 101

Davies, R. I., Burtscher, L., Rosario, D., Storchi-Bergmann, T., Contursi, A., Genzel, R., Gracia-Carpio, J., Hicks, E.K. S., Janssen, A., Koss, M., Lin, M. -Y., Lutz, D., Maciejewski, W., Muller-Sanchez, F., Orban de Xivry, G., Ricci,C., Riffel, R., Riffel, R. A., Schartmann, M., Schnorr-Muller, A., Sternberg, A., Sturm, E., Tacconi, L., Veilleux, S.,2015, ApJ, 806, 127

Davies, R. I., Hicks, E. K. S., Erwin, P., Burtscher, L., Contursi, A., Genzel, R., Janssen, A., Koss, M., Lin, M. -Y.,Lutz, D., Maciejewski, W., M-Sanchez, F., Orban de Xivry, G., Ricci, C., Riffel, R., Riffel, R. A., Rosario, D., Schart-mann, M., Schnorr-Muller, A., Shimizu, T. Sternberg, A., Sturm, E., Storchi-Bergmann, T., Tacconi, L., Veilleux, S.2017, MNRAS, 466, 4917

Dorodnitsyn, A., Bisnovatyi-Kogan, G. S., Kallman, T. 2011, ApJ, 741, 29

Elitzur, M., Shlosman, I. 2006, ApJ, 648, 101L

Emmering, R. T., Blandford, R. D., Sclosman, I. 1992, ApJ, 385, 460

Gallimore, J. F., Elitzur, M., Maiolino, R., Marconi, A., O’Dea, C. P., Lutz, D., Baum, S. A., Nikutta, R., Impellizzeri,C. M. V., Davies, R., Kimball, A. E., Sani, E. 2016, ApJ, 829, 7L

Garcia-Burillo, S., combes, F., Schinnerer, E., Boone, F., Hunt, L. K. 2005, A&A, 441, 1011

Garcia-Burillo, S., combes, F., Ramos Almeida, C., Usero, A., Krips, M., Alonso-Herrero, A., Aalto, S., Casasola, V.,Hunt, L. K., Martin, S., Viti, S., colina, L., Costagliola, F., Eckart, A., fuente, A., Henkel. C., Marquez, I., Neri, R.,Schinnerer, E., Tacconi, L. J., van der Werf, P. P. 2016, ApJL, 823, 12L

Gebhardt, K., Bender, R., Bower, G., Dressler, A., Faber, S. M., Filippenko, A. V.; Green, R., Grillmair, C., Ho, L. C.;Kormendy, J., Lauer, T. R., Magorrian, J., Pinkney, J., Richstone, D., Tremaine, S. 2000, ApJ, 539, 13

Hicks, E. K. S., Davies, R. I., Malkan, M. A., Genzel, R., Tacconi, L. J., Muller Sanchez, F., Sternberg, A. 2009, ApJ,696, 448

Hicks, E. K. S., Davies, R. I., Maciejewski, W., Emsellem, E., Malkan, M. A., Dumas, G., Muller-Sanchez, F., Rivers,A. 2013, ApJ, 768, 107

7

Honig, S. F., Kishimoto, M., Antonucci, R., Marconi, A., Prieto, M. A., Tristram, K., Weigelt, G. 2012, ApJ, 755, 149

Honig, S. F., Kishimoto, M., Tristram, K. R. W., Prieto, M. A., Gandhi, P., Asmus, D., Antonucci, R., Burtsher, L.,Duschl, W. J., Weigelt, G. 2013, ApJ, 771, 87

Hopkins, P. F., Hayward, C. C., Narayanan, D., Hernquist, L. 2012, MNRAS, 420, 320

Imanishi, M., Nakanishi, K., Izumi, N. 2016, ApJ, 822, 10L

Imanishi, M., Nakanishi, K., Izumi, T., Wada, K. 2018, ApJ, 853, 25L

Izumi, T., Kawakatu, N., Kohno, K. 2016, ApJ, 827, 81

Izumi, T., Wada, K., Fukushige, R. Hamamura, S., Kohno, K. 2018, ApJ, 867, 48

Krolik, J. H., Begelman, M. C., 1988, ApJ, 329, 702

Lopez-Gonzaga, N., Jaffe, W., Burtscher, L., Tristram, K. R. W., Meisenheimer, K. 2014, A&A, 565, A71.

Lopez-Rodriguez, E., Packham, C., Young, S., Elitzur, M., Levenson, N. A., Mason, R. E., Ramos Almeida, C.,Alonso-Herrero, A., Jones, T. J., Perlman, E. 2013, MNRAS, 431, 2723

Lopez-Rodriguez, E., Packham, C., Jones, T. J., Nikutta, R., McMaster, L., Mason, R. E., Elvis, M., Shenoy, D.,Alonso-Herrero, A., Ramirez, E., Gonzalez Martin, O., Honig, S. F., Levenson, N. A., Ramos Almeida, C., Perlman,E. 2015, MNRAS, 452, 1902

Lopez-Rodriguez, E., Alonso-Herrero, A., Diaz-Santos, T., Gonzalez-Martin, O., Ichikawa, K., Levenson, N. A.,Martinez-Paredes, M., Nikutta, R., Packham, C., Perlman, E., Ramos Almeida, C., Rodriguez-Espinosa, J. M., Tele-sco, C. M. 2018a, MNRAS, 478, 2350

Lopez-Rodriguez, E., Fuller, L., Alonso-Herrero, A., Efstathiou, A., Ichikawa, K., Levenson, N. A.; Packham, C.,Radomski, J., Ramos Almeida, C., Benford, D. J., Berthoud, M., Hamilton, R., Harper, D.,; Kovavcs, A., Santos, F. P.,Staguhn, J., Herter, T. 2018b, ApJ, 859, 99

Lopez-Rodriguez, E., Antonucci, R., Chary, R. R, Kishimoto, M. 2018c, ApJ, 861, 23

Marconi, A., Risaliti, G., Gilli, R., et al. 2004, MNRAS, 351, 169

Markowitz, A. G., Krumpe, M., Nikutta, R. 2014, MNRAS, 439, 1403.

Muller-Sanchez, F., Prieto, M. A., Hicks, E. K. S., Vives-Arias, H., Davies, R. I., Malkan, M., Tacconi, L. J., Genzel,R. 2011, ApJ, 739, 69

Nenkova, M., Sirocky, M. M.; Ivezic, Z., Elitzur, M. 2008a, ApJ, 685, 147

Nenkova, M., Sirocky, M. M.; Nikutta, R., Ivezic, Z., Elitzur, M. 2008b, ApJ, 685, 160

Netzer, H. 2015, ARA&A, 53, 365

Nyland, K.; Mukherjee, D.; Lacy, M.; Prandoni, I.; Harwood, J.; Alatalo, K.; Bicknell, G.; Emonts, B. 2018, ASPC,517, 457

Ramos Almeida, C., Levenson, N. A., Rodrıguez Espinosa, J. M., Alonso-Herrero, A., Asensio Ramos, A., Radomski,J. T., Packham, C., Fisher, R. S., Telesco, C. M. 2009, ApJ, 702, 1127

Ramos Almeida, C., Ricci, C., 2017, Nature, 1, 679.

Ricci, C., Ueda, Y., Koss, M. J., Trakhtenbrot, B., Bauer, F. E., Gandhi, P., 2015 ApJ 815L, 13

Ricci, C., Trakhtenbrot, B., Koss, M. J., Ueda, Y., Schawinski, K., Oh, K., Lamperti, I., Mushotzky, R., Treister, E.,Ho, L. C., Weigel, A., Bauer, F. E., Paltani, S., Fabian, A. C., Xie, Y., Gehrels, N., 2017, Nature 549, 488

Rigopoulou, D., Kunze, D., Lutz, D., Genzel, R., Moorwood, A. F. M. 2002, A&A, 389, 374

Schartmann, M., Krause, M., Burkert, A. 2011, MNRAS, 415, 741

8

Stalevski, M., Asmus, D., Tristram, K. R. W. 2017, MNRAS, 472, 3854

Stalevski, M., Tristram, K. R. W., Asmus, D. 2019, MNRAS, 484, 3334

Tran, H. D. 2003, ApJ, 632

U, V., Medling, A. M., Inami, H., Armus, L., Dıaz-Santos, T., Charmandaris, V., Howell, J., Stierwalt, S., Privon,G. C., Linden, S. T.; Sanders, D. B.; Max, C. E.; Evans, A. S.; Barcos-Munoz, L., Chiang, C. W. K., Appleton, P.,Canalizo, G., Fazio, G., Iwasawa, K., Larson, K., Mazzarella, J., Murphy, E., Rich, J., Surace, J. 2019, ApJ., 871, 166.

Wada, K. 2012, ApJ, 758, 66

Zoghbi, A., Fabian, A. C., Reynolds, C. S., Cackett, E. M. 2012, MNRAS, 422, 129

9