supermassive black holes in active galactic george/ay21/eaa/eaa-agn-smbh.pdf · pdf...

Click here to load reader

Post on 25-Aug-2018




0 download

Embed Size (px)


  • eaa.iop.orgDOI: 10.1888/0333750888/2365

    Supermassive Black Holes in Active Galactic NucleiLuis C Ho, John Kormendy


    Encyclopedia of Astronomy & AstrophysicsP. Murdin

    IOP Publishing Ltd 2006

    ISBN: 0333750888

    Downloaded on Thu Mar 02 23:39:17 GMT 2006 []

    Institute of Physics PublishingBristol and Philadelphia

    Terms and Conditions

  • Supermassive Black Holes in Active Galactic Nuclei ENCYCLOPEDIA OF ASTRONOMY AND ASTROPHYSICS

    QUASARS are among the most energetic objects in the uni-verse. We now know that they live at the centers of galax-ies and that they are the most dramatic manifestation ofthe more general phenomenon of ACTIVE GALACTIC NUCLEI(AGNs). These include a wide variety of exotica such asSYEFERT GALAXIES, RADIO GALAXIES and BL LACERTAE OBJECTS.Since the discovery of quasars in 1963, much effort hasgone into understanding their energy source. The suite ofproposed ideas has ranged from the relatively prosaic,such as bursts of star formation that make multiplesupernova explosions, to the decidedly more colorful,such as supermassive stars, giant pulsars or spinars,and supermassive black holes (SMBHs). Over time,SMBHs have gained the widest acceptance. The keyobservations that led to this consensus are as follows.

    Quasars have prodigious luminosities. Not uncom-monly, L~1046 erg s1; this is 10 times the luminosity ofthe brightest galaxies. Yet they are tiny, because they varyon timescales of hours. From the beginning, the need foran extremely compact and efficient engine could hardlyhave been more apparent. Gravity was implicated,because collapse to a BLACK HOLE is the most efficientenergy source known. The most cogent argument is dueto Donald Lynden-Bell (1969). He showed that anyattempt to power quasars by nuclear reactions alone isimplausible. First, a lower limit to the total energy outputof a quasar is the energy, ~1061 erg, that is stored in itsradio-emitting plasma halo. This energy weighs 1040 g or107Mo. . However, nuclear reactions produce energy withan efficiency of only =0.7%. Then the waste mass leftbehind in powering quasars would be at least M~109Mo. .Lynden-Bell argued further that quasar engines are2R~~1015 cm in diameter because large variations inquasar luminosities are observed on timescales as shortas 10 h. However, the gravitational potential energy of109Mo. compressed into a volume as small as 10 light-hours is ~GM2~1062 erg. As Lynden-Bell noted,Evidently although our aim was to produce a modelbased on nuclear fuel, we have ended up with a modelwhich has produced more than enough energy by gravi-tational contraction. The nuclear fuel has ended as anirrelevance. We now know that the total energy outputis larger than the energy that is stored in a quasars radiosource; this strengthens the argument. Meanwhile, acaveat has appeared: the objects that vary most rapidlyare now thought to contain relativistic jets that arebeamed at us. This boosts the power of a possibly smallpart of the quasar engine and weakens the argument thatthe object cannot vary on timescales less than the lighttravel time across it. However, this phenomenon wouldnot occur at all if relativistic motions were not involved,

    so SMBH-like potential wells are still implicated. Theseconsiderations suggest that quasar power derives fromgravity.

    The presence of deep gravitational potentials haslong been inferred from the large velocity widths of theemission lines seen in optical and ultraviolet spectra ofAGNs. These are typically 200010000 km s1. If the largeDoppler shifts arise from gravitationally bound gas, thenthe binding objects are both massive and compact. Theobstacle to secure interpretation has always been therealization that gas is easy to push around: explosionsand ejection of gas are common astrophysical phen-omena. The observation that unambiguously points torelativistically deep gravitational potential wells is thedetection of radio jets with plasma knots that are seen tomove faster than the speed of light, c. Apparent expan-sion rates of (110)c are easily achieved if the true expan-sion rate approaches c and the jet is pointed almost at us.

    The final pillar on which the SMBH paradigm isbased is the observation that many AGN jets are well collimated and straight. Evidently AGN engines canremember ejection directions with precision for up to 107 yr. The natural explanation is a single rotating bodythat acts as a stable gyroscope. Alternative AGN enginesthat are made of many bodiessuch as stars and super-novaedo not easily make straight jets.

    A variety of other evidence also is consistent withthe SMBH picture, but the above arguments were theones that persuaded a majority of the astronomical com-munity to take the extreme step of adopting SMBHs asthe probable engine for AGN activity. In the meantime,SMBH alternatives such as single supermassive stars andspinars were shown to be dynamically unstable andhence short lived. Even if such objects can form, they arebelieved to collapse to SMBHs.

    The above picture became the paradigm long beforethere was direct evidence for SMBHs. Dynamical evi-dence is the subject of this article and the following one(SUPERMASSIVE BLACK HOLES IN ACTIVE GALAXIES).Meanwhile, there are new kinds of observations thatpoint directly to SMBH engines. In particular, recentobservations by the Advanced Satellite for Cosmologyand Astrophysics (ASCA) have provided strong evidencefor relativistic motions in AGNs. The x-ray spectra ofmany Seyfert galaxy nuclei contain iron K emissionlines (rest energies of 6.46.9 keV; see figure 1). Theselines show enormous Doppler broadeningin somecases approaching 100 000 km s1 or 0.3cas well asasymmetric line profiles that are consistent with rela-tivistic boosting and dimming in the approaching andreceding parts, respectively, of SMBH accretion disks assmall as a few Schwarzschild radii.

    The foregoing discussion applies to the most pow-erful members of the AGN family, namely quasars andhigh-luminosity Seyfert and radio galaxies. It is less

    Supermassive Black Holes in ActiveGalactic Nuclei

    Copyright Nature Publishing Group 2002Brunel Road, Houndmills, Basingstoke, Hampshire, RG21 6XS, UK Registered No. 785998and Institute of Physics Publishing 2002Dirac House, Temple Back, Bristol, BS21 6BE, UK 1

  • Supermassive Black Holes in Active Galactic Nuclei ENCYCLOPEDIA OF ASTRONOMY AND ASTROPHYSICS

    compelling for the more abundant low-luminosityobjects, where energy requirements are less demandingand where long jets or superluminal motions are seenless frequently or less clearly. Therefore a small butvocal competing school of thought continues to arguethat stellar processes alone, particularly those that occurduring bursts of STAR FORMATION, can reproduce manyAGN characteristics. Nonetheless, dynamical evidencesuggests that SMBHs do lurk in some mildly activenuclei, and, as discussed in the next article, even in themajority of inactive galaxies.

    Measuring AGN masses: direct methodsVery general arguments suggest that quasar engineshave masses M~106109Mo. . Gravitational collapse isbelieved to liberate energy with an efficiency of ~0.1;Lynden-Bells arguments then imply that typical rem-nant masses are M~108Mo. . Better estimates can bederived by asking what we need in order to powerquasar luminosities, which range from 1044 to 1047 erg s1

    or 1011Lo. to 1014Lo. . For =0.1, the engine must consume0.0220Mo. yr1. How much waste mass accumulatesdepends on how long quasars live. This is poorly known.If they live long enough to make radio jets that are colli-mated over several Mpc, and if their lifetimes are conser-vatively estimated as the light travel time along the jets, then quasars last >~107 yr and reach massesM>~105108Mo. . However, the most rigorous lower limiton M follows from the condition that the outward forceof radiation pressure on accreting matter not overwhelmthe inward gravitational attraction of the engine, a con-dition which, admittedly, strictly holds only if the accret-ing material and the radiation have spherical symmetry.This so-called Eddington limit requires that L

  • Supermassive Black Holes in Active Galactic Nuclei ENCYCLOPEDIA OF ASTRONOMY AND ASTROPHYSICS

    equivalently that M> 8105(L/1044 erg s1)Mo.. Here G isthe gravitational constant, mp is the mass of the protonand T is the Thompson cross section for electron scat-tering. We conclude that we are looking for SMBHs withmasses M~106109Mo. . Finding them has become one ofthe Holy Grails of astronomy because of the importanceof confirming or disproving the AGN paradigm.

    AGNs provide the impetus to look for SMBHs, butACTIVE GALAXIES are the most challenging huntingground. Stellar dynamical searches first found centraldark objects in inactive galaxies (see SUPERMASSIVE BLACKHOLES IN INACTIVE GALAXIES), but they cannot be appliedin very active galaxies, because the nonthermal nucleusoutshines the star light. We can estimate masses usingthe kinematics of gas, but only if it is unperturbed bynongravitational forces. Fortunately, this complicationcan be ruled out a posteriori if we observe that the gas isin Keplerian rotation around the center, i.e. if its rotationvelocity as a function of radius is V(r):r1/2. We can alsostack the cards in our favor by targeting galaxies that areonly weakly active and that appear to show gas disks inimages taken through narrow bandpasses centered onprominent emission lines.

    Kinematics of optical emission linesHigh-resolution optical images taken with ground-basedtelescopes and especially with the Hubble SpaceTelescope (HST) show that many giant ELLIPTICAL GALAX-IES contain nuclear disks of dust and ionized gas. Themost famous case is M87 (figure 2(a)). The disk m

View more