Published on August 29, 2007
The Origin of X-Ray Emission in the Nuclei of Radio Galaxies : The Origin of X-Ray Emission in the Nuclei of Radio Galaxies Dan Evans (Harvard) -with- Diana Worrall (U. Bristol), Martin Hardcastle (U. Herts), Ralph Kraft (SAO), Mark Birkinshaw (U. Bristol), Judith Croston (U. Herts) Contents: Contents Brief review of AGN physics andamp; phenomenology Nuclear X-ray emission: accretion flow or jet? The 3CRR sample of radio galaxies Is an obscuring torus ubiquitous? A nuclear Fanaroff-Riley dichotomy? Implications for accretion-flow structure Accretion Processes In AGN: Accretion Processes In AGN Animation of accretion onto a supermassive black hole Accretion Processes In AGN: Accretion Processes In AGN Accretion flow surrounded by dusty torus BB radiation from disk ‘big blue bump’ B-field loops optically thin corona Isotropic X-rays from Comptonization of disk photons in hot corona Power law X-ray spectrum Fe Ka Production: Fe Ka Production Fe Ka lines are the most commonly used accretion diagnostic Width and centroid of Fe Ka line give location of fluorescing material w.r.t. black hole Fe Kα George andamp; Fabian (1991) Radiatively Inefficient vs. Efficient Accretion Flows: Viscous energy heats ions Radiative losses mainly from electrons Energy advected onto BH X,Edd (=LX/LEdd) small (~10-5) Radiatively Inefficient vs. Efficient Accretion Flows m=M/M Inefficient flow (e.g. ADAF) Efficient flow High accretion rates Geometrically thin, optically thick accretion disk (e.g. Shakura-Sunyaev 1973) forms X,Edd higher (up to ~10%) Astrophysical Jets in Radio-Loud AGN: Astrophysical Jets in Radio-Loud AGN Centaurus A: Chandra / VLA Hardcastle et al. (2003) Slide8: Hotspot Core Jet Hotspot Lobe High-power (FRII) Low-power (FRI) Slide9: The Fanaroff-Riley Dichotomy Is the dichotomy Environmental? Interaction of the jet with ambient medium either causes the jet to decelerate (FRI) or propagate supersonically to large distances (FRII) Intrinsic? Properties of the central engine govern large-scale morphology (FRI/FRII) Astrophysical Jets in Radio-Loud AGN: Astrophysical Jets in Radio-Loud AGN Chandra commonly resolves kpc-scale X-ray jet emission in nearby RL AGN: FRIs kpc X-ray emission synchrotron in nature (e.g., Worrall et al. 2001) FRIIs X-ray emission tends to be inverse-Compton (e.g., Sambruna et al. 2004) What about (unresolved) parsec-scale X-ray jets? Parsec-scale X-ray Jets: Parsec-scale X-ray Jets Anisotropic emission, power law X-ray spectrum Relativistic Doppler beaming, dependent on bulk speed (Γ), angle to line of sight Radio-Galaxy Nuclei – Two Competing Models: Radio-Galaxy Nuclei – Two Competing Models Is the nuclear X-ray emission dominated by: The parsec-scale jet? -or- The accretion flow? Evidence for jet-dominated nuclear X-ray emission: Evidence for jet-dominated nuclear X-ray emission Correlations between the ROSAT soft X-ray and VLA radio core fluxes and luminosities in the B2 (Canosa et al. 1999) and 3CRR (Hardcastle andamp; Worrall 1999) samples Parsec-scale radio emission is jet-generated and strongly affected by beaming Tight correlations suggest X-ray emission affected by beaming in same manner as radio Soft X-ray emission originates in a jet Double-peaked SED (modeled with syn+SSC) 10 pc NGC 6251 - 5 GHz VLBI Jones et al. (1986) Evidence for accretion-dominated nuclear X-ray emission: Evidence for accretion-dominated nuclear X-ray emission Often accompanied by high intrinsic absorption (NLRGs) Short (~ks) timescale variability in broad-line FRII 3C 390.3 (Gliozzi et al. 2005) Broadened Fe K line emission in some radio galaxies (e.g., Gliozzi et al. 2004) Implies Fe K origin in inner regions of accretion flow Gliozzi et al. (2004) Summary of Introduction: Summary of Introduction X-ray continuum emission in the nuclei of RL AGN consists of: 'Radio-quiet' accretion-related component 'Radio-loud' jet-related component ROSAT Chandra/XMM Which dominates the X-ray emission? The 3CRR Sample: The 3CRR Sample Criteria: 178-MHz flux density andgt; 10.9 Jy Declination andgt; 10o |b| andgt; 10o Advantages: No orientation bias Spectroscopic identification High-resolution radio observations Select sources with zandlt;0.1 Unambiguously spatially separate unresolved nuclear emission from contaminating emission Rich variety (FRI/FRII, broad/narrow lines, large luminosity range) 19/35 X-ray observations of low-z 3CRRs, 16 of them with Chandra Complete X-ray spectral analysis of each 5 GHz VLA (Leahy, Bridle, andamp; Strom) The 3CRR Sample: Aims: The 3CRR Sample: Aims Dominant X-ray emission mechanism: Accretion Flow: Fe K, variability -or- Jet: Radio-X-ray luminosity correlations, SED Nature of accretion flow: thin disk? RIAF? Is the torus ubiquitous? FRI-FRII dichotomy? Unified AGN scheme: Distribution of intrinsic absorption: Distribution of intrinsic absorption Measure intrinsic absorption associated with dominant component of X-ray emission Note that some sources have two continuum components (one absorbed; one unabsorbed) Bimodal distribution FRIs have low (or no) intrinsic absorption FRIIs have NH andgt; 1023 cm-2 NB BLRG FRII 3C 390.3 Origin of intrinsic absorption: Origin of intrinsic absorption FRI sources with the highest intrinsic absorption are associated with host galaxies with circumnuclear disks at high inclinations Heavily absorbed emission in FRIIs likely in gas associated with dusty torus (c.f. Seyfert 2s) Slide20: Consider LX and LR Considerable scatter Luminosity-Luminosity Correlations Slide21: Luminosity-Luminosity Correlations Consider LX and LR Considerable scatter Strong correlation between components with NH ≤ 5 x 1022 (significant at 99.99%) Apparent in flux-flux too Suggests X-ray emission affected by beaming in same manner as radio Origin of X-ray emission in pc-scale jet (outside any torus) Jet Slide22: Luminosity-Luminosity Correlations FRIs are dominated by these soft components only FRIIs, although heavily absorbed, also contain these components Cannot distinguish between X-ray jet components in FRIs and FRIIs Jet Slide23: Components with NH ~ 1023 lie above trendline As does 3C 390.3, unobscured BLRG All have Fe K lines Accretion-dominated and surrounded by a torus FRIIs are dominated by these components Luminosity-Luminosity Correlations NH ≤ 5 x 1022 Jet Where is the torus in FRIs?: Where is the torus in FRIs? X-ray emission is jet dominated and exists on scales larger than any torus Cannot determine presence or absence of torus directly Assume there exists a ‘hidden’ accretion component obscured by a torus of intrinsic absorption 1023 atoms cm-2 Find upper limits to luminosity of accretion-related emission Data don’t exclude luminosities of LX,acc ~ 1039-1041 ergs s-1 (X,Edd ~ 10-7-10-5) Substantially lower than FRIIs, LX,acc ~ 1043-1044 ergs s-1 (X,Edd ~ 10-3-10-2) e.g. 3C 274 (M87) ‘Hidden’ Accretion Intermediate Summary: Intermediate Summary X-ray emission of FRI radio-galaxy nuclei is dominated by a parsec-scale jet, with little or no intrinsic absorption X-ray emission of FRII radio-galaxy nuclei is dominated by an accretion flow and is heavily absorbed (except BLRG 3C 390.3) Each FRII also has an unabsorbed component of X-ray emission jet origin Cannot rule out a torus in FRIs ‘Hidden’ accretion flows in FRIs are substantially sub-Eddington (c.f. FRIIs) e.g. 3C 264 (FRI) e.g. 3C 403 (FRII) Nuclear F-R Dichotomy?: Nuclear F-R Dichotomy? How can a dichotomy in the subparsec-scale accretion-flow mode influence the kpc-scale deceleration of jets into FRI and FRII structures (e.g., Bicknell 1995)? 'HYMORS' sources (Gopal-Krishna andamp; Wiita 2000) Nuclear F-R Dichotomy?: Nuclear F-R Dichotomy? Low-excitation – High-excitation Dichotomy? Low-Excitation Radio Galaxies: Low-Excitation Radio Galaxies ‘Low-excitation’ vs ‘High-excitation’ (e.g., Hine andamp; Longair 1979). Based on strength of high-excitation lines like [OIII] FRIs are almost entirely low-excitation Significant population of low-excitation FRIIs at 0.1andlt;zandlt;0.5 Extend sample to these redshifts X-ray Spectra: X-ray Spectra NLRG FRIIs FRIs / LERG FRIIs Sample results: Sample results All of the NLRG FRIIs show evidence for an absorbed nuclear component with NH ~ 1023 cm-2 accretion flow dominates Almost all the FRIs and LERG FRIIs show no evidence for an absorbed component (exceptions are debatably classed as LERG) jet dominates Again, can place upper limits on the accretion-flow luminosity in the presence of a torus Luminosities: Luminosities White = LERG Red = NLRG Green = BLRG; Blue = quasar Circle =andgt; FRI Line goes through FRII NLRG Limits assume NH = 1023 cm-2 For a given 178-MHz radio power, FRIs and LERG FRIIs produce significantly less radiative accretion luminosity Pointing to a dichotomy in the excitation properties of AGN? Understanding the Dichotomy: Understanding the Dichotomy Do LERGs represent a class of radio galaxies that don’t participate in unified AGN schemes? What implications does the observed dichotomy have on the structure of the central engine in AGN? Several possible interpretations… Model 1: FRI/LERG FRII tori are Compton thick: Model 1: FRI/LERG FRII tori are Compton thick Luminous accretion flow surrounded by thicker torus than NLRG FRIIs Would give rise to reduced observed flux from accretion flow Assume NH = 1024 cm-2 X,Edd ~ 10-6-10-4, still lower than NLRG FRIIs For FRI/LERG FRII and NLRG FRII accretion-flow efficiencies to match, need extreme NH (andgt; 1025 cm-2) Ruled out from infrared data (Muller et al. 2004, Haas et al. 2004, Birkinshaw et al., in prep.) Absorbed X-ray lum. against mid-IR (Ogle et al 2006 data, except for Cen A). Line is the line of equality, not a fit. Model 2: An intrinsic nuclear dichotomy: Model 2: An intrinsic nuclear dichotomy Fundamentally different accretion mode in FRI/LERG FRIIs and NLRG FRIIs Accretion-flow luminosities and radiative efficiencies of LERGs systematically lower than NLRGs Widely discussed model: hybrid ADAF/thin disk + jet Critical mass-accretion rate mcrit (=Mcrit/M) Esin et al. (1997) Model 2: An intrinsic nuclear dichotomy: Hi-exc Low-exc Model 2: An intrinsic nuclear dichotomy Analogous to XRBs Step change in accretion luminosity at mcrit At low accretion rates, jet still strong and comes to dominate Esin et al. (1997) Körding, Falcke, andamp; Markoff (2002) Accretion flow Jet mEdd mcrit m LX Model 2: An intrinsic nuclear dichotomy: Model 2: An intrinsic nuclear dichotomy In this model, both FRI and FRII kpc-scale structures can be produced by nuclei with low accretion luminosity FRI/FRII dichotomy entirely due to environment and jet power Excitation dichotomy controlled by accretion mode 3C 388 (LERG FRII) Summary: Summary The excitation of an AGN is a vital parameter in unification schemes X-ray emission of FRI and LERG FRII radio-galaxy nuclei is unabsorbed and dominated by a parsec-scale jet X-ray emission of NLRG FRII radio-galaxy nuclei is heavily absorbed and accretion-related Data do not exclude the presence of a heavily obscured, accretion-related emission in LERG-type sources An intrinsic nuclear dichotomy (analogous to XRBs)?