12.07.2015 Views

Seminar

Seminar

Seminar

SHOW MORE
SHOW LESS
  • No tags were found...

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Observed disc spectra• Pick only ones that look like disc• L/L Edd ∝T 4 max (Ebisawa et al 1993; Kubotaet al 1999; 2001)• Constant size scale – last stableorbit!!• Proportionality constant gives ameasure R lso i.e. spin• Consistent with low to moderatespin not extreme spin norextreme versions of higherdimensional gravity -braneworlds (Gregory, Whisker, Beckwith& Done 2004)


But rest are not simple…• Bewildering variety ofspectra from single object• Underlying pattern• High L/L Edd : soft spectrum,peaks at kT max often disclike,plus tail• Lower L/L Edd :hardspectrum, peaks at highenergies, not like a discGierlinski & Done 2003


And the radio jet…• No special μQSO class –they ALL produce jets,consistent with sameradio/X ray evolution• Jet links to spectral state• Steady jet in low/hardstate, power depends onaccretion rate! i.e. L/L Edd(Merloni et al 2003; Falke et al 2004)• Bright radio flares inrapid low/hard tohigh/soft transition (Fenderet al 2004)• No jet in disc dominatedstateGallo et al 2003


No jetDecrease fraction of power innonthermal reconnectionabove discHSHSDisc to minimum stable orbitVHSDecrease inner disc radius,and maybe radial extent ofcorona giving increasing LFQPO frequencyJet gets faster, catches upwith slower outflow, getflares of radio emission frominternal shocks Fender (2004)VHSLS(bright)LS(dim)Done Gierlinski & Kubota 2007


• Disc temperature M -1/4 forsame L/L Edd• 10 eV from 1 keV for factor10 8 change in BH mass• UV/EUV hidden by ISM socan’t easily see peak of discScale up to AGN


AGN spectral states• Stretched out by lower disctemperature so states not soobvious as in BHB but still….• …AGN NOT all the sameunderneath absorption as insimplest unified models!! TheirSED should depend intrinsicallyon L/L Edd in same way as BHB• So need to know AGN mass inorder to calculate L Edd


Mass of AGN??• Magorrian-Gebhardt relation gives BH mass!! Big black holes livein host galaxies with big bulges! Either measured by bulgeluminosity or bulge mass (stellar velocity dispersion)• Plus scaling relations for BLR in AGN (Kaspi et al 2000)10 310 9 10 6 10 12Black hole massStellar system mass


AGN/QSO Zoo!!! Optical


Seyfert1 -QuasarsSimilar spectra and line ratios,strong UV flux to excite lines,probably similar L/L Edd ~ 0.1-0.3Increasing LIncreasing M


Seyfert 1 – Seyfert 2• Intrinsically same except forobscuration• But differences in HIGH energyspectra (20-200 keV). S1’ssofter than S2’s – but also havehigher than S2 sampleso same correlation as in BHB -softer when higher L/L Edd inLHS (Middleton, Done & Schurch 2008;Winter et al 2009)


LSHard (low L/L Edd )Soft (high L/L Edd )LINERS?VHSNLS1HSQSOUSQSODone & Gierliński 2005


Implications for high L/L Edd• High L/L Edd objects easyto find. Typically most PGQSO’s have L>0.05 L Edd• For these, soft excessesshould be very rare inXMM bandpass. Whenseen they should be verysteep, and low temperature• Power law at high energiesshould be steep, Γ=2-2.5• PG1211- what not to see!Strong soft excess to~1keV, flat power law athigh energies


Soft excess? NOT from the disc!• NOT THE DISC -doesn’t get close torise in data at 1keV• unless extreme spinand/or modified byadvection – butdisc tail very steepwhile SX gradual• Compton scatteringof disc by low T e ,high τ material?Magdziarz et al 1998,Czerny et al 2003PG1211: disk for M=10 8 M L/L Edd =1Gierliński & Done 2004


NOT from Comptonisation• ALL need soft excess• Fit with comptonisation...• ALL have same kT e for softexcess!! Yet big range inexpected disc kT (mainly M)Walter & Fink 1993, Czerny et al 2003,Gierlinski & Done 2004, Crummy et al 2006• Expect electron temperature tochange if seed photons fromdisc change – differentefficiency of Compton cooling• NOT COMPTON SCATTERINGGierlinski & Done 2004


Partially ionised, relativistic material• Opacity jump at OVII/VIII at 0.7 keV: fixed energy• Atomic features not seen so relativistic smearing sometimes extreme• Reflection: needs intrinsic power law suppressed for large SX• Absorption: larger SX by larger column, curvature gives red wingGierliński & Done 2004, Chevallier et al 2006 Fabian et al 2002; 2004 Miniutti & Fabian 2004


Partially ionised, partial covering• Opacity jump at OVII/VIII at 0.7 keV: fixed energy• Atomic features not seen due to dilution by unabsorbed flux• Absorption: curvature gives red wingInoue 2002; Miller et al 2007; 2008; Turner et al 2007 Fabian et al 2002; 2004 Miniutti & Fabian 2004


Alternative geometries for soft excessfrom partially ionised materialReflection Absorption


More soft excesses in AGN


More soft excesses in AGNMiller et al 2007, Miller et al 2008


More soft excesses in AGNMiller et al 2007, Miller et al 2008


More soft excesses in AGNMiller et al 2007, Miller et al 2008


But this was a dipper!!!Diaz-Trigo et al 2006• Neutron star binary at high inclination• Phase dependent complex (photo and collisionalionsation) absorption needed for RGS data! van Peet et al 2009• Complexity (multiple pcf) doesn’t rule it out!!!


1H0707-495: iron• Huge drop at iron K(plus huge SX)• If absorption thenshould come out asline and if its notrelativistic then it isnarrow and easy tosee! Reynolds et al 2009But if it’s a windthen its easy! PCygni line profile isbroad Sim 2005; Sim et al2008; Done et al 2008• Big feature at iron Lin the data though !!!Miller et al 2007, Done st al 2008


Winds• Most opacity in resonance lines• Accelerating wind shifts energy of absorption so make notch…Log νfνLog E


Winds• Most opacity in resonance lines• Accelerating wind shifts energy of absorption so make notch…Log νfνLog E


P Cygni line profilessymmetric emission + blueshifted = P Cygni profileabsorption


UV line driven Winds ?• UV bright disc launches vertical wind. Rises up but thenilluminated by central X-ray source which overionises it so thatno UV transitions! Only X-ray lines and these don’t absorb asmuch momentum L x


Similar drop in deepest dips!!Hyodo et al 2009


1H0707-495: iron• Huge drop at iron K(plus huge SX)• If absorption thenshould come out asline and if its notrelativistic then it isnarrow and easy tosee! Reynolds et al 2009But if it’s a windthen its easy! PCygni line profile isbroad Sim 2005; Sim et al2008; Done et al 2008• Big feature at iron Lin the data though !!!Miller et al 2007, Done st al 2008


RE1034+396• Huge SX (similarsize to softest1H0707 spectra) andno clear drop at Fe…Miller et al 2007 Middleton et al 2008


RE J1034+396: spot that period!!100ks of XMM-Newton data, co-adding MOS1, 2 and PN dataGierlinski, Middleton, Ward & Done 2008


Smoothed lightcurvePeriod much clearer in last ~60ks – almost periodic Q = υ/Δυ > 16Gierlinski, Middleton, Done & Ward 2008


Power spectrumEven sampling so analyticPower law P PL ∝ f -αα=1.35 ± 0.18f QPO =2.7x10 -4 Hz (=3700 s)Much bigger than 99.99%significance (chanceprobability is 10 -7 )Gierlinski, Middleton, Ward & Done 2008Derived from same methodsused to reduce significanceof previous claims Vaughan etal. 2006


REJ1034: rms variability• Not as expected if havesame modulation at allenergies!• Obvious energydependance of QPO• Much higher amplitudeat higher energiesMiddleton et al 2008


REJ1034: rms variability• Not as expected if havesame modulation at allenergies!• Obvious energydependance of QPO• Much higher amplitudeat higher energies• Sharp change inproperties at ~2keV• Similar to all rapidvariability• But not to longtimescale!Middleton et al 2008


cf other NLS1’s …• Nothing like!!• Often (not always) peaks between 0.7-3 keV – smoking gun foratomic processes (but also warm absorber in these systems)Gierliński & Done 2006


REJ1034: separate soft component• Vary power lawnorm, keepcomptoniseddisc constant!!• Like in BHB,QPO is in tail,not disc!• Can make longertimescale byslow variabilityof comptoniseddisc BUTlightcurvecomplex!!Middleton et al 2008


REJ1034: separate soft componentdisc slim disc comptonMiddleton et al 2008


Range of local high L/L Edd AGN• Soft excess:factor ~2 at 0.5keV but can bemuch larger• Lsx/Lbol


What else is special about REJ1034?Casebeer et al 2006Extreme SED comparedto normal even for NLS1Highest temperaturedisc/largest soft excessSmall mass and highL/L Edd ?L bol ~ 5x10 44 ergs s -1Extreme accretion rate?ML Edd


GRS1915+105• See a lot low kTe spectraspectra in the superEddingtonBHB GRS1915+105• Can’t assume electrons alwayshigh temperature kT e = 50 keV– gives completely differentspectral decomposition (andblack hole spin!) and doesn’t fitso well• Obvious wind features even atRXTE resolution!kT e free kT e =50 keVMiddleton, Davis, Done & Gierlinski 2005


REJ1034+396: Comptonised discMiddleton et al 2008RE J1034+396GRS1915+105Looks very similar tosome spectra seen inGRS1915+105 if shiftenergy scale by ~20=> mass of ~2x10 6 M


Conclusions• BHB give template for sub-Eddington accretion flow• Simple disc L-T 4 in high/soft state• NOT always simple disc – low/hard and very high states aredominated by Comptonisation• Scale up to AGN – LINERs are dim low/hard, some Seyferts (espS2) are bright low/hard, some (esp S1) are high/soft as are radioquiet QSO’s• NLS1 as very high state? But SX. And sometimes hard 2-10 keVspectra. Many (most?) of these are a distortion from atomicprocesses (reflection/absorption)• But some SX REAL! Optically thick, low temperature, opticallythick Comptonised disc!! see also in uniquely luminous BHBGRS1915+105 and in AGN with X-ray QPO (and ULX!)

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!