Anorexic and Gluttonous Black Holes:OutlineGlossary of acronymsGlossary IISummary of Part 1Sag A*: Our RIAF lab2-Temperature plasmaADAFOther examples of RIAF3-part accretion rate classificationConvection Dominated FlowsADIOS: Inflow & OutflowConclusions to Part 1Summary of Part 2Primary EigenvectorX-ray slopeModel for NLS1Variability of X-raysAlternatives to mdotConclusionsReferencesAnorexic and Gluttonous Black Holes:Radiatively Inefficient Accretion Flows and Narrow Line Seyfert 1 AGNKenneth MoodyOctober 11, 2004Outline• Low mdot: Radiatively Inefficient Accretion Flows– Simple Model: spherical Accretion– Model with spin: ADAF– Other models• High mdot: Narrow Line Seyfert 1– Observed properties– Model for narrow lines– How to nail down mdotinterpretationGlossary of acronyms• RIAF: Radiatively Inefficient Accretion Flow• ADAF: Advection Dominated Accretion Flow• CDAF: Convection Dominated Accretion Flow• ADIOS: Advection Dominated Inflow-Outflow SolutionGlossary II• NLS1: Narrow Line Seyfert 1• BBB: Big Blue BumpPart 1: RIAFs, Dim Bulbs of the UniverseSummary of Part 1• Nuclei of some galaxies have L<10-7 to -9 LEdd• Low luminosity partially due to low accretion rate, partially small amount of energy turned into radiation• ADAF most popular model for low radiative efficiency disk• Convection/outflow model also worksSag A*: Our RIAF lab• Comparatively close at 8.5 kpc• Mass: 2.6x106Lo (Ghez et al. 2003)• Observed luminosity: K band L≤1035erg/s (Genzel et al. 1997), X-rays L=2.2x1033erg/s (Baganoff et al. 2001)• Normal AGN luminosity (L=0.1LEdd) for that mass is 1043erg/sNo Disk in the Middle?•Most energy of accretion not turned into radiation•Shakura-Sunyaev disk very efficient at turning grav. potential energy into radiation via viscous dissipation•Therefore, RIAFs like Sag A* don’t have regular disk•Sag A* SED from radio/sub-mm to X-ray•Disk models with various accretion rates (6x10-8to 6x10-10 Mo/yr)•Lack of fit is obviousOne possible solution to low luminosity: low accretion rateBondi (spherical) accretion (|L|=0)RcscBsccRMcGMR022,ρπ==&from Chandra: n=100 cm-3, kT=2keV close to Sag A*obsBoBLLyrMM4510,102.1 ≈×=−&synchrotronbremsstrahlungComptonization2-Temperature plasma• Heating of gas with collapse due to compression, magnetic reconnection• Electron radiation efficiency increases with temperature, so hot electrons too bright• Coupling of e- and ions only through Coulomb forces at low density• Electrons cool rapidly, ions retain high temp (1012K)ADAF• Geometrically thick disk, hot ions puff to nearly spherical shell• Energy from gravitational potential retained in gas as heat, swallowed by (advected into) BH• Rotation in disk further reduces accretion rate below Bondi• Coupling of viscous energy into electrons, δThe Illustrated ADAFions: T=1012Ke-: T=few 1010KRc•Excellent fit for range of δ•No other parameters in model•Flare observed in 2001 fits δ=1. Also could be increased coupling of ions with e-, non-thermal electronsOther examples of RIAF• Giant ellipticals like NGC 1399, NGC 4472 have dim nuclei • Accretion rate in ADAF set from density at Rc, much less than Bondi rate,• Low luminosity AGN/LINERs show hybrid disk/ADAF00011.01399=m&21.0 cMLBacc&<<3-part accretion rate classification•Full Shakura-Sunyaev disk, standard AGN•Transition object, LINER/LLAGN, M81, NGC4579•Rad. inefficient accretion, ADAF, Sag A*, NGC 1399EddaccLL 11.0−=EddaccLL541010−−−=EddaccLL5.810−=RcConvection Dominated Flows• Heating of gas during infall causes it to become unstable to convection• Gas in ADAF nearly unbound; addition of convective energy reduces accretion rate, luminosityADIOS: Inflow & Outflow• Radial density profile• p=0 for standard ADAF• Convection causes profile to be flatter, 0<p<1• CDAF acts like model with p=1• Effects of CDAF would be seen in mm/radio (reduction in synchrotron emission)pR+−∝2/3ρConclusions to Part 1• No disk in center of Sag A* due to low luminosity, other spectral features• Low accretion rate not enough to explain low luminosity• 2-temp plasma with nearly spherical geometry provides acceptably low efficiency (ADAF)• Rate further decreased by outflow due to convectionPart 2: NLS1s, The Other Side of the Dining RangeSummary of Part 2• Objects found with narrow Balmer Lines (FWHM Hβ<2000 km/sec), but other properties like type 1 AGN• Correlations of emission lines put these objects at an extremum of some process, mdot• Some observed X-ray properties confirm, others yet to be discoveredNGC 1275Broad (left) versus Narrow (right) line Seyfert 1 galaxiesAlso note presence/absence of FeIIDefinition: NLS1 has FWHM of Hb < 2000 km/secUnder represented in surveys made in hard X-raysPrimary Eigenvector• Correlation between FeII, [OIII], and Balmer line widths/shapes• NLS1 at one end of relation• Reference Boronson & Green 1992• More on this WednesdayX-ray slope• Optical emission powered by inner parts of disk• BBB extension into soft x-rays, origin believed to be thermal inner disk• Lots of emission from inner disk makes x-ray spectrum soft, could indicate large accretion rateModel for NLS1Broad Line Region clouds virialized, radius proportional to luminosityFor fixed luminosity, emission at close to LEddÆ small massLow mass means clouds move more slowlyVariability of X-rays• Flux varies by large amount and on short timescales, more so than other types of AGN of similar luminosity• Rapid fluctuation implies coordinated emission, possibly relativistic jets• Comparison to similar states in galactic black holes show the latter more stable on very short timescalesAlternatives to mdot• From Boronson & Green 92: If [O III] isotropic, cannot be nuclear orientation• Black hole spin:– Radio loud QSOs at high end of primary eigenvector– NLS1 have non-spinning BH– Analysis of Fe Ka lineConclusions• Low accretion rate produces a thin, 2-temperature plasma that forms an ADAF, very low L/LEdd• High accretion rate makes a hot disk around a low mass BH for a fixed L, soft X-ray continuum, and narrow lines from slow moving BLR cloudsReferences• Brandt, W. N.• Leighly, K. M. astro-ph/0402676• Narayan, R. astro-ph/0201260• Quataert, E. astro-ph/0304099• …and references