Getting to Eddington and beyond in AGN and binaries!

Getting to Eddington and beyond in AGN and binaries!

Chris DoneUniversity of Durham

• L/LEdd determined by mass accretion rate onto the black hole

• LMXRB – roche lobe overflow of low mass companion

• HMXRB – wind accretion from high mass companion

• HMXRB – roche lobe overflow from high mass companion

• AGN

Accreting black holes





• AGN






• AGN






• AGN


LMXRB – Transient!• H ionisation instability in disc – eg Lasota 2001• If T(Rout)<H ionisation then disc globally unstable• T(Rout) depends on mass, mass accretion rate and binary size – all

correlated by roche lobe overflow condition!

DGK072 years

• Radius of star depends on type (ie mass)

• Star has to fill roche lobe for mass accretion

• Mass accretion rate from companion depends on type

• Outer edge of disc can’t be larger than ½ size of binary orbit.

• Find all have T(Rout)<H ionisation instability for all low mass companion stars

• See review by Lasota (2001)

Roche lobe overflow

Menou et al 1999

• Peak luminosity depends on how much mass can accumulate in quiescence and how fast it accretes onto black hole

• Lpeak R(out)3/R(out ) R(out)2 King & Ritter 1998

• Main sequence secondary is small so R(out) small so Lpeak<LEdd

Roche lobe overflow

• GRS1915+105 is huge! 33 day period. L/LEDD~1 for 20 years

• GS2023 – went to 2LEdd, then blew its disc apart…

• GX339 and XTEJ1550 are the next biggest but L<LEDD

J. Orosz

LMXRB BH in our galaxy

• Rare systems – most end as NS in population synthesis models

• High mass loss rate from stellar wind, but only capture small fraction so L<<LEdd

Winds fed HMXRB – BH

• Main sequence high mass star. Mass accretion rate from companion is high

• Keeps outer disc temperature in Cyg X-1 above H ionisation so persistent

• HUGE mass transfer as supergiant evolves – SS 433 in our galaxy

• Most ULX in other galaxies

Roche lobe overflow HMXRB-BH

• Main sequence high mass star. Mass accretion rate from companion is high

• Keeps outer disc temperature in Cyg X-1 above H ionisation so persistent

• HUGE mass transfer as supergiant evolves – SS 433 in our galaxy

• Most ULX in other galaxies

Roche lobe overflow HMXRB-BH

Rappaport et al 2005

Intermediate mass BH?

• Ultra - Luminous X-ray sources in spiral arms of nearby starforming galaxies – ULX

• L~1039-40 ergs s-1 so M~10-100 M

for L <LEdd

• Hard for stellar evolution to make BH > 50 M

Gao

et a

l 200

3

• Hard spectra plus soft excess looks like scaled up BHB in LHS?

• IMBH?

• But break above 7keV – NOT like LHS!!!

ULX state ?

Gladstone Roberts & Done 2008

ULX state ?


ULX state ?


ULX state ?


• L/LEdd determined by mass supply

• But ~0.5% of mass in star formation ends up in the black hole to make the M- relation

AGN

BH-Galaxy AND environment•Big black holes live in host galaxies with big bulges!

•Need 0.5% of bulge mass (ie starformation) to end up down the BH

Bla

ck h

ole

mas

s

Stellar system mass

103

109

1061012

Black hole mass

• SMBH grow by gas accretion and BH-BH mergers

• Mergers dominate only highest BH mass (> 109 M) . Spin of 0.7-0.8

• Accretion (thin disk) dominates for lower mass (<108 M)

• Accretion (hot flow) never really dominates

Fanidakis et al 2010

Black hole mass accretion rate

• Not many with L/LEdd>1 in local universe - and they are predominantly low mass BH

• What do they look like?

Fanidakis et al 2010

Disc spectra from 106 M L/LEdd ~1

• Much more soft X-ray flux than expected from either disc or power law

• Enormous soft X-ray excess !!

Jin et al 2011


• Standard SS disc temperature – assumes energy thermalises

• BHB discs - Colour temperature correction as scattering > absorption opacity. Tobs=1.8 Teff

• AGN discs even more scattering dominated as less dense !! Factor 2.4 !!

•

Done, Davis, Jin, Blaes Ward 2011


• Enourmous soft excess in REJ1034

• But actually a lot of it should be the bare disc!

• Plus a little bit of comptonisation !

• More like disc dominated black holes

Done, Davis, Jin, Blaes Ward 2011

Conclusions• Galactic BHB can’t get to Eddington very easily

• Exceptions are LMXRB in wide binaries with evolved companions – GRS1915+105

• And HMXRB evolving into supergiants – SS 433 and probably ULX.

• And some nearby low mass high mass accretion rate AGN like REJ1034 (QPO AGN).

• Disc in these AGN MUST extend into soft X-rays. Much of soft X-ray excess in these is the bare disc. Then need SMALL comptonisation to get shape of component

Unsolved problems• How does magnetic field stress dissipate and heat the accretion flow?

Can we get as high as ~0.1? Stress HEATS so eventually gives pressure so alpha prescription OK on average??

• What happens to the disc as we go to Eddington and beyond? – How does it stay optically thick up to ~0.5LEdd?

– How important are winds Fgrav=(1- abs/ es L/LEdd) GM/R2

– How important is advection of radiation – and what fraction of this escapes from the plunging region becoming optically thin: not radiatively inefficient?

• How does the B field manage to get the same (approx) vertical flux to launch the same power jet in lots of different BHB?

• How does thin cool disc truncated into hot flow? Simulations??• what are the HF QPOs ?

– method for measuring a*? But LF QPOs probably don’t !!!!

• Can we understand iron line profiles ? And get all methods for measuring a* giving the same answer?

Getting to Eddington and beyond in AGN and binaries!

Documents