Supermassive black holes at high redshifts

TL;DR

A new MeV telescope survey will detect over 1000 blazars up to redshift 6, enabling insights into early supermassive black hole formation, jet emission regions, and accretion-spin interactions.

Contribution

The paper proposes that a next-generation MeV telescope can significantly advance understanding of supermassive black holes and jet physics at high redshifts.

Findings

Detection of >1000 MeV blazars up to z=6

Probing black hole formation and growth mechanisms

Locating emission regions in powerful blazars

Abstract

MeV blazars are the most luminous persistent sources in the Universe and emit most of their energy in the MeV band. These objects display very large jet powers and accretion luminosities and are known to host black holes with a mass often exceeding $10^9 M_{\odot}$. An MeV survey, performed by a new generation MeV telescope which will bridge the entire energy and sensitivity gap between the current generation of hard X-ray and gamma-ray instruments, will detect $>$1000 MeV blazars up to a redshift of $z=5-6$. Here we show that this would allow us: 1) to probe the formation and growth mechanisms of supermassive black holes at high redshifts, 2) to pinpoint the location of the emission region in powerful blazars, 3) to determine how accretion and black hole spin interplay to power the jet.

Full text

Astro2020 Science White Paper

Supermassive black holes at high redshifts

Thematic Areas:                    \text{\hbox to0.0pt{\checkmark\hss}}\square Formation and Evolution of Compact Objects            \text{\hbox to0.0pt{\checkmark\hss}}\square Cosmology and Fundamental Physics

Principal Authors:

Name: Vaidehi S. Paliya Institution: Clemson University & Deutsches Elektronen Synchrotron DESY, Germany Email: vaidehi.s.paliya@gmail.com Phone: +49 175 2112107

Name: Marco Ajello Institution: Clemson University Email: majello@g.clemson.edu Phone: +1 650 8049042

Name: Lea Marcotulli Institution: Clemson University Email: lmarcot@g.clemson.edu Phone: +1 864 6507068

Co-authors: J. Tomsick, Space Sciences Lab, University of California Berkeley; J. S. Perkins, NASA/GSFC; E. Prandini, University of Padova, Italy; F. D’Ammando, INAF-IRA Bologna; A. De Angelis, INFN and INAF Padova, Italy; D. Thompson, NASA/GSFC; H. Li, LANL; A. Domínguez, GAE and IPARCOS, Universidad Complutense de Madrid; V. Beckmann, CNRS / IN2P3; S. Guiriec, GWU/NASA GSFC; Z. Wadiasingh, NASA/GSFC; P. Coppi, Yale University; J. Patrick Harding, LANL; M. Petropoulou, Princeton University; J. W. Hewitt, University of North Florida; R. Ojha, UMBC/NASA GSFC; A. Marcowith, Laboratoire Univers et Particules de Montpellier; M. Doro, University and INFN Padova; D. Castro, Center for Astrophysics, Harvard & Smithsonian; M. Baring, Rice University; E. Hays, NASA GSFC; E. Orlando, Stanford University; S. Guiriec, GWU/NASA GSFC; V. Bozhilov, Sofia University; J. Tomsick UC Berkeley; I. Agudo, IAA-CSIC (Spain); T. Venters, NASA GSFC; J. McEnery, NASA/GSFC; L. The, Clemson University; D. Hartmann, Clemson University; S. Buson, Universität Würzburg, University of Maryland, Baltimore County, USA; Francesco Longo, University of Trieste and INFN, Trieste; Dario Gasparrini, SSDC and INFN, Roma Tor Vergata.

Abstract: MeV blazars are the most luminous persistent sources in the Universe and emit most of their energy in the MeV band. These objects display very large jet powers and accretion luminosities and are known to host black holes with a mass often exceeding 109$M_{\odot}$. An MeV survey, performed by a new generation MeV telescope which will bridge the entire energy and sensitivity gap between the current generation of hard X-ray and $\gamma$-ray instruments, will detect $>$1000 MeV blazars up to a redshift of $z=5-6$. Here we show that this would allow us: 1) to probe the formation and growth mechanisms of supermassive black holes at high redshifts, 2) to pinpoint the location of the emission region in powerful blazars, 3) to determine how accretion and black hole spin interplay to power the jet.

Projects/Programs Emphasized:

All-sky Medium Energy Gamma-ray Observatory (AMEGO)

https://asd.gsfc.nasa.gov/amego/ 2. 2.

Lunar Occultation Explorer (LOX)

https://www.hou.usra.edu/meetings/deepspace2018/pdf/3094.pdf 3. 3.

Compton Spectrometer and Imager (COSI)

http://cosi.ssl.berkeley.edu 4. 4.

e-ASTROGAM

http://eastrogam.iaps.inaf.it 5. 5.

adEPT

http://ui.adsabs.harvard.edu/#abs/2014APh....59...18H

\justify

1 Introduction

Black holes as massive at $10^{9-10}$ $M_{\odot}$ have been discovered all the way up to redshift $z=6-7$ (Wu et al. 2015, Mortlock et al. 2011, Bañados et al. 2018). Their existence poses crucial questions on our understanding of black hole formation and growth. Indeed, already at $z>4$, the Universe’s age is barely compatible with the time it takes to grow a 109-10 $M_{\odot}$ black hole just by accretion (at the Eddington limit) starting from a stellar mass black hole. Thus, the detection of a large number of massive black holes early on implies that: 1) they could have grown from a massive (105-6$M_{\odot}$) seed (formed through, e.g., the collapse of a massive molecular torus following the merging of the giant protogalaxies; Mayer et al. 2010), 2) experienced sustained super-Eddington accretion (Volonteri & Rees 2005), or 3) have experienced a large number of merging events.

Quasars are the most luminous active galactic nuclei (AGN) known, and, as such, they enable unparalleled studies of the Universe at the earliest cosmic epochs. The space density of massive black holes ($M>10^{9}$ $M_{\odot}$) hosted in radio-quiet quasars111Formally, a source is called radio-quiet if $R<10$, where radio-loudness parameter $R$ is the ratio of the rest-frame 5 GHz to optical B-band flux density (Kellermann et al. 1989). A radio-loud object has $R>10$. peaks at $z\approx 2$ (Hopkins et al. 2007, see also blue line in the right panel of Fig. 1). However, massive black holes in radio-loud quasars (objects with strong jetted radio emission) are present at much earlier epochs, close to $z=4-5$ (orange line in the right panel of Fig. 1, Sbarrato et al. 2015). This points to the fact that relativistic jets may play a crucial role in rapid black hole growth at these early stages of the Universe. The observation of high-redshift jetted quasars will thus allow us to probe the formation and growth mechanisms of massive black holes at the dawn of the Universe.

Blazars, radio-loud AGN having relativistic jets oriented close to the line of sight, have been detected to $z>5$ (Romani et al. 2004). Already at moderately high redshifts ($z>3$), blazars host massive ($>$109 $M_{\odot}$) black holes and display extremely luminous ($>$1046 erg s*-1*) accretion disks (e.g., Ghisellini et al. 2010). Most remarkably, these powerful objects emit most of their energy output in the MeV band and have thus been dubbed ‘MeV’ blazars. These sources, among all other blazars, tend to host the most massive black holes and are also found at the highest redshifts, as seen in Fig. 1 (right panel). Therefore a survey of ‘MeV’ blazars, as explained below, will allows us to constrain the space density of massive black holes in the early Universe.

2 Importance of the MeV Band

The most powerful and distant blazars emit the majority of their bolometric output between 0.2 MeV and 10 MeV, making the MeV band the optimal one to search for them. So far, the detection of such elusive monsters has been achieved using $\gamma$-ray (Fermi-Large Area Telescope, LAT, $\rm[20\,MeV,2\,TeV]$, Atwood et al. 2009) or hard X-ray (Swift-Burst Area Telescope, BAT, $\rm[14,195\,keV]$, Gehrels et al. 2004; Nuclear Spectroscopic Telescope Array, NuSTAR, $\rm[3,80\,keV]$, Harrison et al. 2013) observatories, or a combination of the two. Peaking at MeV energies, these sources become harder to detect in the $\gamma$-ray band, due to both the softening of the spectrum ($\Gamma_{\gamma}>2.7$) as well as instrument sensitivity threshold. Only 8 blazars have been detected by the LAT beyond redshift 3 (Marcotulli et al. 2017, Ackermann et al., 2017), the farthest being at $z=4.31$. On the other hand, the shallow sensitivity of the BAT has allowed us to find only a handful of the brightest sources (Oh et al., 2018). Even if the exceptional sensitivity of NuSTAR allows for the detection of fainter sources, its pointing strategy does not permit a population study. Moreover, hard X-ray observations alone of these sources (without $\gamma$-ray ones) are not able to pinpoint the peak of the emission. This, in turn, is of fundamental importance to measure jet parameters, such as the total jet power and the bulk Lorentz factor ($\Gamma$). Moreover, since their jetted radiation is Doppler boosted and collimated within an angle $\sim 1/\Gamma$, for each detected blazar there are 2$\Gamma^{2}$ ($\sim 200-450$, for typical $\Gamma\sim 10-15$) radio-loud quasars with similar intrinsic properties and jets pointed elsewhere. Thus, the detection of even a few blazars and an accurate measurement of $\Gamma$ can constrain in a sensitive way the space density of massive black holes at high redshifts (see Fig. 1 and Sbarrato et al. 2015, Ackermann et al. 2017). An all-sky MeV instrument is therefore the necessary bridge in both sensitivity and energy coverage between the existing state-of-the-art instruments in the other high-energy bands. Directly sampling blazars’ peak emission, it is the best solution to uncover the population of extreme high-redshift blazars, to precisely constrain their jet properties and to uncover the most massive black holes in the Universe and their connection to jets.

3 Science with MeV Observations of Powerful blazars

An all-sky MeV mission like AMEGO, LOX, e-ASTROGAM or adEPT, with a continuum sensitivity in the 0.1-10 MeV energy band of 10*-11* erg cm*-2* s*-1*, will detect $\approx$2000 blazars. Half of these objects will be detected at redshift $>2$ and, importantly, $\sim$50 will be found between redshift of 5 and 6 (Ajello et al. 2009). Moreover, LAT-known blazars will be uncovered by such instruments. As can be seen from Fig. 2, if these sources display steeply falling spectral energy distribution (SED) in the GeV band, they are much brighter in the MeV band and will be easily captured by these MeV missions.

Evolution and Massive Black Hole Growth: This sample will be used to study the evolution of the most powerful blazars, which are known to experience extremely strong positive evolution (i.e. space density growing quickly with increasing redshift) and to trace the evolution of the most massive elliptical galaxies (Ajello et al. 2009).

As shown in the right panel of Fig. 1, we expect most of the blazars at $z\geq$3 to host supermassive black holes with a mass in excess of 109$M_{\odot}$. Therefore, this set of blazars can be used to measure the space density of billion solar mass black holes at high redshifts. Already at $z=4$ this function for radio-quiet and radio-loud systems is the same, implying that the radio-loud phase (likely triggered by a major merger) may be essential for quick black hole growth. This survey of powerful blazars will allow us to extend this measurement beyond redshift of $z=4$, where it becomes virtually impossible to grow a black hole purely by accretion at the Eddington limit.

Location and Size of the Emission Region: The high-energy emission of FSRQs is typically interpreted as owing to inverse Compton scattering on either the photons of the broad line region (BLR) or those of the torus (Begelman & Sikora 1987, Paliya et al. 2017). Since the characteristic energy of BLR photons ($\sim$10 eV) is larger compared to the torus ones ($\sim$0.1 eV), inverse Compton scattered emission will peak at higher frequencies if upscattering BLR photons rather than photons from the torus. The peak frequency is particularly sensitive to (and scales linearly with) the energy of the external photon fields and can thus be used to discriminate between different external field components and whether the emission region resides within the BLR or beyond (see Fig. 2, right). Virtually all the high-energy peaks of known FSRQs fall in the MeV regime and, as of now, sensitive observations at these wavelengths are crucially lacking. Moreover, in a conical jet scenario, an emission region located within the BLR is $\sim$5$-$10 times smaller than that one located outside of it and within the torus. Thus, the typical timescale for the variability of FSRQs in the MeV band will strongly be affected by the location, hence size, of the emission region (with way faster variability in the BLR scenario with respect to the torus one). An MeV mission will thus yield two critical observables to determine the location and size of the emission region in FSRQs.

Accretion-Jet Connection: It has been long postulated that jets are powered by the accretion disk (Rawlings & Saunders, 1991). In recent years, it has come to attention that, for jetted quasars at $z<3$, an extra source of power (e.g. black hole spin) is also needed to energize these jets (Ghisellini et al., 2014). Whether this holds true beyond $z=3$ is still an open issue. However, these theories strongly rely on accurate measurements of both the accretion disk power as well as the jet one. Though the former can be constrained from optical spectroscopy (e.g., Shaw et al., 2012), it is tedious to precisely measure the latter due to lack of sensitive hard X-ray-to-$\gamma$-ray (i.e., MeV band) observations. To this goal, the MeV regime is crucial. Firstly because these powerful blazars radiate a major fraction of their bolometric luminosity in this energy range. Secondly, measurements in this band are needed in order to accurately parametrize the underlying particle population. The synchrotron peak in MeV blazars is typically located at radio to sub-mm wavelengths. Because of synchrotron self-absorption, it is impossible to characterize the spectral shape of the underlying electron energy distribution (EED) from the synchrotron emission alone. In such MeV blazars, the EED can be constrained using hard X-ray and $\gamma$-ray data, which (in EC scenario222The large Compton dominance in powerful MeV blazars indicates a relatively low magnetic field and thus a negligible contribution from synchrotron self Compton at X-rays.) constrain the low-energy and high-energy electron population, respectively. In particular, the spectral index before the break energy of the EED333The EED is typically modeled as broken power law. (as measured from MeV observations, Fig. 2, right panel) and the minimum energy of the EED control the total amount of radiating electrons, thus also the number of cold protons, assuming one cold proton per electron (Celotti & Ghisellini, 2008) which dominate the total power of the jet. Therefore, the knowledge of the hard X-ray-to-MeV energy band is crucial to accurately determine the total jet power. A comparison of the accretion disk luminosity with the jet power will enable us to determine whether jets are mainly powered by accretion or there are also other reservoir of energy, such as black hole spin. Most importantly, we will be able to study this connection for a whole population of high-redshift sources, possibly unveiling new jet physics.

References:

Atwood, et al. 2009, ApJ, 697, 1071

Ackermann et al., 2017, ApJL, 837, 5

Ajello et al., 2009, ApJ, 699, 603

Ajello et al., 2012, ApJ, 751, 108

Bañados et al., 2018, Nature, 553, 473

Begelman & Sikora, 1987, ApJ, 322, 650

Blandford & Rees, 1978, Phys. Scr, 17, 265

Bloemen et al., 1995, A&A, 293, 1

Celotti & Ghisellini, 2008, MNRAS, 385, 283

Dermer, 1995, ApJ, 446, 63

Dermer & Schlickeiser, 1993, ApJ, 416, 458

Gehrels, et al. 2004, ApJ, 611, 1005

Ghisellini et al., 1998, MNRAS, 301, 451

Ghisellini et al., 2010, MNRAS, 405, 387

Ghisellini et al., 2014, Nature, 515, 376

Harrison et al. 2013, ApJ, 770, 103

Hopkins et al., 2007, ApJ, 654, 731

Kellermann et al., 1989, AJ, 98, 1195

Mayer et al., 2010, Nature, 466, 1082

Marcotulli et al. 2017, ApJ, 839, 96

Mortlock et al., 2011, Nature, 474, 616

Paliya et al., 2017, ApJ, 851, 33

Rawlings & Saunders, 1991, Nature, 349, 138

Romani et al., 2004, ApJL, 610, 9

Sbarrato T., et al. 2015, MNRAS, 446, 2483

Shaw et al., 2012, ApJ, 748, 49

Volonteri & Rees, 2005, ApJ, 633, 624

Volonteri M., et al., 2011, MNRAS, 416, 216

Wu et al., 2015, Nature, 518, 512