Hidden cooling flows in clusters of galaxies – III. Accretion on to the central black hole

ABSTRACT

Recently, we have uncovered hidden cooling flows (HCFs) in the X-ray spectra of the central brightest galaxies of 11 clusters, 1 group, and 2 elliptical galaxies. Here, we report such flows in a further 15 objects, consisting of 8 clusters, 3 groups, 3 ellipticals, and 1 Red Nugget. The mass cooling rates are about |$1\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$| in the ellipticals, 2 to |$20\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$| in the groups, and 20 to |$100\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$| in regular clusters. The Red Nugget, MRK 1216, has an HCF of |$10\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$|⁠. We review the fate of the cooled gas and investigate how some of it might accrete on to the central black hole. The gas is likely to be very cold and to have fragmented into low-mass stars and smaller objects before being swallowed whole, with little luminous output. If such a scenario is correct and operates at a few |$\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$| then such objects may host the fastest growing black holes in the low-redshift Universe. We briefly discuss the relevance of HCF to the growth of early galaxies and black holes.

1 INTRODUCTION

We have recently found hidden cooling flows (HCFs) in clusters and groups of galaxies, as well as a couple of nearby elliptical galaxies (HCFI and HCFII) (Fabian et al. 2022, 2023), using spectra from the XMM Reflection Grating Spectrometer (RGS). These soft X-ray-emitting flows are hidden within photoelectrically absorbing cold clouds and dust near the centres of the central brightest galaxies. They represent the cooler inner parts of larger, wider-scale cooling flows. Active galactic nucleus (AGN) feedback acts to reduce the main cooling flow in the larger body of these objects but the inner parts drop from direct view behind cold absorbing clouds. The total mass cooling rates can be 20–50 per cent or more of the unabsorbed rates inferred earlier from X-ray imaging studies.

The findings again raise the ‘cooling flow problem’ of what happens to the cooled gas? HCF mass cooling flow rates of tens of solar masses per year in regular clusters and |$1\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$| in early-type galaxies (ETGs) lasting |$\sim 8{\rm \, Gyr}$| (since redshift z = 1) means almost |$10^{11}$| and |$10^{10}\hbox{$\rm \, {\rm M}_{\odot }$}$|⁠, respectively, of accumulated cooled gas. Where does it go? The issue is not new¹ but has largely been ignored for the past two decades, even at the low rates allowed without absorption (see Liu et al. 2019 and Section 2).

We have proposed and discussed several possibilities, namely that (a) the gas cools to invisibility (i.e. so cold that it radiates little), (b) the cooled gas fragments into low-mass stars and substellar objects, and (c) cooled gas is dragged out from the centre by the bubbling action of AGN feedback. (a) and (b) mean that there is increasing unseen mass of gas and/or low-mass stars at the centres of these objects. (c) may be consistent with observed metal abundance profiles. These possibilities are not of course mutually exclusive.

Here, we investigate how much cooled gas can end up in the central black holes. Many of the most massive black holes at low redshift lie in brightest cluster galaxies (BCGs; McConnell & Ma 2013; Bogdán et al. 2018), and we include a couple here including Holm 15A, the central galaxy of A85, which has a black hole of mass 4 × 10¹⁰ (Mehrgan et al. 2019). There is some evidence that the black hole to galaxy stellar mass ratio of ETGs has increased significantly from z = 1 to the present day (Farrah et al. 2023). Since massive black holes can swallow stars whole, such accretion need not be luminous.

We now search for hidden cooling flows in eight cool core clusters, three X-ray luminous groups, and four relatively isolated elliptical galaxy including a Red Nugget. They are found in all objects and have the typical mass cooling rates found in HCFI and HCFII. One is the very X-ray luminous cluster, ZW3146, at medium redshift z ∼ 0.3, the results for which compare well with other high-luminosity clusters found at similar and higher redshift. A significant part of its high cooling rate of |$\sim 1000\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$| goes into observed normal star formation but it is unlikely to be a long-lived situation. As discussed in HCFII (Fabian et al. 2023), rapid accretion on to the central black hole has the potential to turn it into a luminous quasar, as seen in the Phoenix cluster, perhaps ending in a massive outburst such as has occurred in MS 0735+7421 (McNamara et al. 2005).

We then speculate whether hidden accretion is taking place into the central black holes of HCFs. Hidden in the sense of unobserved because the infalling matter consists of low mass stars, etc., which are swallowed whole without emitting radiation. Finally, we speculate on high-pressure star formation, which occurs in HCF, and discuss its relevance to early galaxy formation and in particular to the origin of ‘Red Nuggets’.

2 SPECTRAL ANALYSIS

The objects and data used are listed in Table 1. SAS 20.0 was used for the data reduction. The spectra were extracted using rgsproc with a 95  per cent extraction in point spread function width (corresponding to 1.7 arcmin) and 95  per cent in pulse-height distribution. To create good time intervals (GTIs), light curves were created for each RGS instrument from CCD 9, with events with flag values of 8 or 16, extracted with a cross-dispersion angle of greater than 1.5 × 10⁻⁴, in time bins of 200 s. The GTIs were created when the rate was below 0.3 cts s⁻¹. Background spectra were created with rgsbkgmodel. The spectra and background spectra from RGS1 and RGS2 were combined using rgscombine before spectral fitting. The spectra were then analysed using xspec (Arnaud 1996) over the energy range of 8–22A, where background is minimized.

The spectral model used is tbabs(gsmooth*apec + gsmooth(partcov*mlayerz)mkcflow). The intrinsic absorption model mlayerz (see HCFII for details) represents a sequence of interleaved emission and absorption layers with a total column density N_H listed in Table 2. tbabs is the Galactic absorption in the direction of the target apec is a constant temperature thermal emission model which represents the outer cluster gas. Its temperature is also used as the hotter temperature in the cooling flow mkcflow model. partcov enables the measurement of the total mass cooling rate of both unabsorbed and absorbed components. A covering fraction of one means that all the cooling flow component is absorbed and if zero then none is absorbed. The model assumes no particular geometry for the absorbed and unabsorbed components. It does assume that all absorbed components are identical. The minimum temperature of the cooling flow model is set at 0.1 keV.

The spectra are shown in the appendix as Figs B1–B15, together with contour plots of absorbed mass cooling rate (⁠|$\dot{M}_{\rm a}$|⁠) versus intrinsic column density N_H and covering fraction.

Since the RGS is a slitless spectrometer (den Herder et al. 2001), there is some blurring of the energy scale associated with extended sources. This is included in the spectral model by smoothing the spectral components with separate Gaussian kernels for the outer apec component and inner HCF. When making the contour plots for the less bright objects, we often needed to freeze the smoothing parameters to their best-fitting values in order to have convergence. Detailed spectral results are given in Table 2 and are compared with data from other wavebands in Table 3.

3 THE SPECTRAL RESULTS

As noted in HCFII, χ²-space for the HCF model is often corrugated, which can lead to complex contour plots. We are using a very simplistic model and a real hidden cooling flow is expected to be far more complicated in both space and column density. RGS spectra provide no more than a rough average over the inner arcmin of the target source.

A source like ZW3146, where there is a large continuum fraction, can have a very uncertain abundance Z, with it anticorrelating with the mass cooling rate (see Fig. 1). In this case we fix it at Z = 0.4.

Of the 15 sources studied here, all but three require a best-fitting covering fraction of 0.95 or more. This emphasizes that they are indeed ‘hidden’. The intrinsic column densities range from 2 × 10²¹ to |$3\times 10^{22}\hbox{${\rm \, cm}^{-2}\, $}$|⁠.

We also refit the spectra with the covering fraction set to zero, in order to determine the mass cooling rate if there is no absorption, |$\dot{M}_{\rm u}$|⁠. This is listed in the last column of Table 2. As expected, it is generally very low, but quite large for 2A0335. The lowest χ² value for this no absorption case is however 14 above that for the best-fitting HCF mlayerz model, which is therefore the statistically preferred one. The value in the case of A2597 is about what is expected from the HCF model where the covering fraction is about 70 per cent.

When the temperature of the gas is above about 0.4 keV the fraction of the energy emerging in continuum is about 50 per cent and drops below 10 per cent below 0.2 keV (Fig. 2). Most of the flux below 0.4 keV is absorbed away in our HCF fits, meaning the continuum shape plays a significant role in our spectral fit results.

We reduced the energy band of the spectra of several lower temperature objects to 12–20A due to broad excess residuals around 10A. These are likely due to the apec component having a (small) spread in temperature.

The absorbed luminosities (L_a, Table 3) are all less than the far infrared luminosities, where available. This indicates that the energy lost in the cooling flow to absorption is energetically capable of emerging as radiation from dust in the absorbing gas.

Fig. 3 shows the HCF mass cooling rates (in red) compared (where available) with the ‘classical’ rates from X-ray imaging listed by (Hudson et al. 2010) (in black). The mean ratio of hidden to classical rates is 45 per cent, with a range from 4 to 180 per cent. (Most lie between 17 and 58 per cent.) The HCF rates are about |$1\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$| for elliptical galaxies, 2–|$20\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$| for brightest group galaxies (BGGs), and about 10–|$100\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$| for regular BCGs. There are then a group of more distant, exceptionally X-ray luminous, BCGs with 400 to |$\gt 1000\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$|⁠.

We suspect that the last group of rare objects may be highly time variable, with peak luminosity followed by a quasar eruption. The regular clusters and elliptical galaxies generally have low-luminosity nuclei, with radio emission from jets that blow bubbles in the intracluster medium. The bubbles and related activity generally lie outside the inner kpc studied here.

4 THE ACCUMULATION OF COOLED GAS

Over a billion years |$10^9\hbox{$\rm \, {\rm M}_{\odot }$}$| of gas will have cooled in a typical elliptical, up to |$10^{10}\hbox{$\rm \, {\rm M}_{\odot }$}$| in a BGG and up to |$10^{11}\hbox{$\rm \, {\rm M}_{\odot }$}$| in a BCG. These are large values, the higher end of which exceeds the cold molecular masses observed via CO emission in BCGs (Olivares et al. 2019; Russell et al. 2019). It is possible that the mass of molecular gas has been underestimated due to low abundance and an unseen diffuse component, but this is unlikely to make a very large difference.

In HCFI, we considered the following possibilities: (a) continued cooling to invisibility at 3 K, (b) fragmentation and collapse into substellar objects since the Jeans mass is less than |$0.1\hbox{$\rm \, {\rm M}_{\odot }$}$|⁠, (c) outward dragging of cooled clouds by the bubbling process, or (d) cold front formation. The gas and dark matter peaks may be offset by a kpc or more.

We also flagged the similarity in conditions (e.g. gas pressure) of cooled dusty molecular clouds of a BCG core to those in the Crab Nebula. More detailed observational comparisons are warranted.

It is likely that the cooled material is spread over the innermost few kpc of the core. Clear evidence of the dragging out of dust-enriched material from the centre is provided by the peaks in metal abundance seen |$\sim 10{\rm \, kpc}$| from the centre of low-redshift clusters (Panagoulia, Sanders & Fabian 2015; Lakhchaura, Mernier & Werner 2019; Liu et al. 2019).

Detailed measurements of the mass profile of each separate component in a cool core (black hole, dark matter, stars, gas, etc.) will be invaluable in sorting the possibilities out. We now consider whether some small fraction of the very cold clouds and substellar objects can be swallowed by the central black hole in the next section.

4.1 Accretion of fragmented cold matter by the central black hole

We showed in HCFI (Fabian et al. 2022) that, under the high-pressure conditions of an HCF (⁠|$nT\sim 10^{6.5}\!-\!10^{7.5}\hbox{${\rm \, cm}^{-3}{\rm \, K}\, $}$|⁠) and no heating, the gas cools rapidly (time-scale of tens of years) to ∼3 K. The Jeans mass is below about |$0.1\hbox{$\rm \, {\rm M}_{\odot }$}$| (Jura 1977; Ferland, Fabian & Johnstone 1994) and the gas expected to clump and fragment into low mass stars, brown dwarfs, etc., some of which will fall into the black hole emitting little radiation. Exactly how large a fraction will be swallowed depends on how angular momentum is transported outward. The turbulent viscosity of a luminous accretion disc is absent here and a possible path is that the innermost cooled gas forms a thick disc of low-mass stars and cold gas clouds around the black hole. Dynamical gravitational instabilities such as spiral waves² and bars within bars transport angular momentum outward in non-spherical systems so that some of the matter falls inward (Shlosman, Frank & Begelman 1989; Hopkins & Quataert 2011; Gualandris et al. 2017) to be swallowed directly by the central black hole without a standard accretion disc forming.³

A very crude estimate of the mass inflow rate may be obtained from an isothermal Bondi flow. This of course assumes the matter is a fluid and ignores rotation but does give some idea of the rate at which matter comes under the gravitational influence of the black hole. This simple rate is

where M, c_s, and ρ are the black hole mass |$M_9=10^9\hbox{$\rm \, {\rm M}_{\odot }$}$|⁠, the speed of sound (or random motions), and density of the surrounding gas. Taking |$c_\mathrm{ s}=300\hbox{${\rm \, km}{\rm \, s}^{-1}\, $}$| and ρ equal to the mass density if the medium has a mass in units of |$10^9\hbox{$\rm \, {\rm M}_{\odot }$}$| per sphere of radius |$1{\rm \, kpc}$|⁠, we obtain |$\dot{M}\approx 4 M_9^2\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$| and an accretion radius of ∼50 pc.

Hopkins & Quataert (2011) give an analytical estimate of the accretion rate from gravitational torques which agrees with their numerical simulations and find |$1\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$| in the middle of the range. The predicted rate has a weak black hole mass dependence of M^1/6. Using observations and analytical work, Genzel et al. (2023) show that such torques operating in disc galaxies at z ∼ 2 lead to large-scale inflow on about 10 dynamical times.

In the case of the elliptical and brightest group galaxies studied here, the accretion rate could exceed the HCF mass cooling rate, which would then become the determining rate. We conclude that rates of a few |$\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$| may be possible. We are of course assuming a high efficiency with which the cold matter is swallowed by the black hole.

The possibility thus emerges that the mass of black holes in low redshift elliptical galaxies is increasing due to inflow from HCF at a rate of several |$\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$|⁠. Angular momentum transfer is due to gravitational torques. The black hole mass can thus increase by up to |$\sim 10^{10}\hbox{$\rm \, {\rm M}_{\odot }$}$| since z = 1 and possibly even more for the most massive objects in the BCGs of the most massive clusters. Such objects need not have a luminous AGN, although an ADAF due to a weak gaseous inflow may persist and power jets thus a radio source in these objects. They would be the highest accretion rate black holes in the low-redshift Universe. If the accretion rate continues for several Gyr then this would lead to the most massive black holes appearing now. Examples in our sample include NGC1600 and the central galaxy of A85, Holm 15A, with |$1.7\times 10^{10}\hbox{$\rm \, {\rm M}_{\odot }$}$| and |$4\times 10^{10}\hbox{$\rm \, {\rm M}_{\odot }$}$| black holes, respectively. Other BCGs have very high mass black hole including that of the BCG of A1201 for which a gravitationally lensed arc reveals a central mass of |$3\times 10^{10}\hbox{$\rm \, {\rm M}_{\odot }$}$| (Nightingale et al. 2023).

4.2 Relevance to galaxy formation

The standard model of galaxy formation involves gas falling into dark matter haloes and heated by shocks and compression, gas that can cool quickly (on a dynamical time or less) then leads to star formation enriching the gas with metals and dust and the core of a new galaxy (White & Rees 1978). Further accretion, mergers and feedback later build the outer galaxy. The gas which has a longer cooling time than the local dynamical time, but shorter that the age of the Universe at the time, can form a cooling flow (Nulsen & Fabian 1995). If the conditions such as metal and dust enrichment and especially the pressure are high (⁠|$nT\gt 10^6\hbox{${\rm \, cm}^{-3}{\rm \, K}\, $}$|⁠) then they may resemble the nearby hidden cooling flows discussed here. If at high redshift, the higher temperature of the cosmic microwave background will in turn require a higher Jeans mass. If cloud collapse does lead to large populations of low-mass stars and brown dwarfs then early supermassive black holes can grow by swallowing such fragments whole, independent of the Eddington limit.

4.3 Red Nuggets and MRK 1216

A population of compact ETGs have been identified at redshifts of 2 and above which may be examples of galaxies that did not progress beyond the early core formation galaxy stage (Daddi et al. 2005). These are known as ‘Red Nuggets’ (Damjanov et al. 2009) and have stellar masses of |$1\!-\!2\times 10^{11}\hbox{$\rm \, {\rm M}_{\odot }$}$| and effective radii of only 1–2 kpc. Later some examples have been identified at low redshifts, e.g. NGC1277 (Trujillo et al. 2014) a galaxy unable to grow larger by mergers, or by accretion of cold gas, since it lies in the core of the rich Perseus Cluster. More recently, further examples have been found (Ferré-Mateu et al. 2015) including the isolated rotating ETG MRK 1216 (Ferré-Mateu et al. 2017) that lies at a distance of 94 Mpc and hosts a black hole of mass |$4.5\times 10^9\hbox{$\rm \, {\rm M}_{\odot }$}$| (Walsh et al. 2017).

Werner et al. (2018) noted that MRK 1216 might lie in a halo of mass up to |$10^{13}\hbox{$\rm \, {\rm M}_{\odot }$}$| and so have an X-ray halo. They indeed found extended thermal emission with an X-ray luminosity |$L_{\rm X}=7\times 10^{41}\hbox{${\rm \, erg}{\rm \, s}^{-1}\, $}.$| Buote & Barth (2019) found that its dark matter halo has a high concentration, implying early formation. We have included MRK1216 in our sample and find a significant HCF of |$9.7\pm 2.7 \hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$|⁠, larger than the rate of typical ETGs. Ferré-Mateu et al. (2017) show it to have a very bottom-heavy initial mass function (IMF) that is consistent with a significant accumulation of low mass stars and brown dwarfs.

MRK1216 could provide the nearest link between low- and high-redshift HCF and clearly merits deeper study.

4.4 Observational possibilities

Further observations at the time of writing are limited. Hopefully, XRISM will be launched soon and provide new high-resolution, non-dispersive, X-ray spectra of the inner regions of clusters, groups, and ETGs. Its field of view is larger than that of the RGS so can see how any HCF region matches into the rest of the cluster. High spatial resolution X-ray studies await next-generation telescopes such as AXIS. As well as resolving the expected irregular appearance of HCF due to the absorption, it will be particularly helpful for examining the immediate surroundings of the central black hole. The X-IFU of Athena will spectroscopically map HCFs in great detail, as will the Light Element Emission Mapper Probe. JWST may open up the inner regions in the near-IR. Since most of the flow of cooled gas takes place at very low temperatures below |$10{\rm \, K}$|⁠, the bulk of the flow will be inaccessible, except to absorption measurements.

ALMA has opened up molecular absorption studies of cool BCGs using the central radio source as a backlight (David et al. 2014; Tremblay et al. 2016a; Rose et al. 2019, 2023). Four objects in the last study show molecular gas moving towards the central source at |$200\!-\!300\hbox{${\rm \, km}{\rm \, s}^{-1}\, $}$|⁠, each plausibly forming part of an inward cold accretion flow.

5 CONCLUSION

We find that significant cooling flows, closely linked with cold absorbing gas, are common in the brightest galaxies of cool core clusters and groups as well as large elliptical galaxies. The mass cooling rates range from 1 to over |$1000\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$|⁠. In most cases they are reduced by AGN feedback to a factor 2–3 times lower than the simple cooling rates derived from X-ray imaging. The gas in the central hidden/absorbed part can cool to below 10 K, collapsing and fragmenting into low mass stars, brown dwarfs, etc., most of which are dragged outward by the bubbling and cold front processes. We speculate that some matter within the inner tens pc may fall into the black hole, with a rate of a few |$\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$| being plausible. Such accretion emits little radiation, although it is likely that some thin plasma is present, possibly in the form of a low-luminosity ADAF, to power the jets usually seen in radio images. If cooled collapsed matter does fall in, then the mass accretion rate can be among the highest in the low redshift Universe.

ACKNOWLEDGEMENTS

BRM acknowledges the Natural Sciences and Engineering Research Council for their support. We thank the referee for a prompt report.

DATA AVAILABILITY

All data used here are available from ESA’s XMM–Newton Science Archive.

Footnotes

References

APPENDIX A: NOTES ON OBJECTS

A1 2A0335-096

The Chandra X-ray data (Sanders, Fabian & Taylor 2009) of 2A0335 show that the central arcmin consists of six or so blobs of emission. Some are softer than the others. Cavities are seen and in the optical a bright H α nebula is found. ALMA CO data (Vantyghem et al. 2016) of the core reveal |$10^9\hbox{$\rm \, {\rm M}_{\odot }$}$| of cold molecular gas. The paper also discusses the patchy dust extinction reported earlier by Donahue et al. (2007). We use the column density of Sanders et al. (2009) in the spectral fitting of this object.

A2 A85

Holm 15 A, the BCG of A85, hosts one of the most massive black holes known, at |$4\times 10^{10}\hbox{$\rm \, {\rm M}_{\odot }$}$| (Mehrgan et al. 2019). The cluster is merging with a subcluster seen to the South. Cold fronts and a cavity are seen in Chandra X-ray data presented by (Ichinohe et al. 2015).

A3 A496

The RGS data have been studied by Tamura et al. (2001) in which no evidence for X-ray emitting gas cooler than |$1{\rm \, keV}$| was reported. (No additional absorption was considered.) Chandra X-ray images showing cold fronts have been presented by Dupke & White (2003). Optical emission lines spectra have been studied by Crawford & Fabian (1992).

A4 A2597

A2597 has consistently shown evidence for residual cooling, starting with O vi ultraviolet emission detected with FUSE (Oegerle et al. 2001), together with C iii]. The O vi line originates from thermal gas at around |$7\times 10^5{\rm \, K}$| and could be due to a cooling flow of |$20\pm 15\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$|⁠, or |$75\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$| when corrected for dust extinction.

Morris & Fabian (2005) analysed XMM data, including the RGS, and found a possible cooling flow rate of |$\sim 45\hbox{$\hbox{$\rm \, {\rm M}_{\odot }$}{\rm \, yr}^{-1}\, $}$| (corrected to |$H_0=70\hbox{${\rm \, km}\, {\rm \, s}^{-1}{\rm \, Mpc}^{-1}\, $}$|⁠). McNamara et al. (2001) found ghost bubbles in the Chandra X-ray images and Tremblay et al. (2012a, b, 2016a, b) found evidence for some residual cooling as well as CO absorption from ALMA data. They argue for a multiphase outflow or fountain in the cluster core.

A5 A2199

Nulsen et al. (2013) presented the Chandra data for A2199. It shows signs of sloshing and has a complex radio source 3C338.

A6 M87

M87 is the second most luminous galaxy but is situated at the centre of the nearest cluster to us, the Virgo cluster. Its central black hole has a mass of |$6.5\times 10^9\hbox{$\rm \, {\rm M}_{\odot }$}$| as measured by the Event Horizon Telescope (Event Horizon Telescope Collaboration 2019). Werner et al. (2013) studied the Chandra emission in the core of M87, showing that it is highly multiphase and fitting an intrinsic absorption model. Temi, Brighenti & Mathews (2007) report the Spitzer far-infrared (FIR) flux. A power-law component with photon index of 2 was included in the fit to represent the nucleus and jet. Oldham & Auger (2018) found that M87 has a bottom-heavy IMF.

A7 NGC1399

This is the BCG of the Fornax cluster that is the next most distant cluster after Virgo. Su et al. (2019) present analysis of the Chandra X-ray data, revealing a pair of X-ray cavities coincident with radio lobes. Optical spectroscopy studied by Vaughan et al. (2018) shows that IMF in the central region is bottom heavy. They also comment on a filament of ionized gas.

A8 NGC720

NGC720 is an isolated nearby elliptical galaxy. Its Chandra X-ray data have been studied by Buote et al. (2002).

A9 NGC1550

NGC1550 is the brightest galaxy in an X-ray bright group. It has been studied recently with Chandra by Kolokythas et al. (2020). It shows signs of activity near its core due to interactions with its radio source.

A10 NGC1600

NGC1600 is the brightest galaxy in a small group. It hosts a very massive black hole of |$1.6\times 10^{10}\hbox{$\rm \, {\rm M}_{\odot }$}$|⁠, as determined by stellar velocity measurements by Thomas et al. (2016). Runge & Walker (2021) present the Chandra data on NGC1600 at the Bondi radius. They find evidence for a multiphase (2-temperature) X-ray emitting gas. H α emission is reported by Trinchieri & di Serego Alighieri (1991)

A11 NGC3091

NGC3091 is a bright elliptical galaxy in the relatively isolated compact group, Hickson 42. It has been selected for not being otherwise special. Our results show that it hosts a modest HCF.

A12 NGC5813

NGC5813 is the brightest galaxy in a subgroup of the larger NGC5846 group. They constitute the third nearest massive cluster/group assembly after the Virgo and Fornax clusters. The Chandra X-ray data of NGC5813 reveal a series of bubbles extending from the BGG (Randall et al. 2015). ALMA CO and other data are reported by Temi et al. (2018).

A13 NGC5846

Chandra X-ray data on NGC5846 have been studied by Machacek et al. (2011). Further X-ray analysis and multiband data on NGC5846 and many other elliptical galaxies are given by Lakhchaura et al. (2018). Radio images of this and some of the above galaxies can be found in Grossová et al. (2022).

A14 MRK1216

The Red Nugget MRK1216 is discussed in Section 4.3.

A15 ZW3146

ZW3146 is a massive X-ray luminous cluster at redshift 0.29. X-ray and ALMA CO and other data are discussed by Vantyghem et al. (2021). The centroid of the |$5\times 10^{10}\hbox{$\rm \, {\rm M}_{\odot }$}$| of molecular gas is offset by 2.6 kpc from the central AGN. The FIR detection is presented by Egami et al. (2006). MUSTANG-2 SZ results are reported by Romero et al. (2020).

APPENDIX B: SPECTRA OF OBJECTS