AGN Jets: A Review for Comparison with Microquasars & GRBs

Alan MarscherBoston University

Research Web Page: www.bu.edu/blazars

Emission Regions in a Radio-Loud AGN

Differences with BH X-ray binaries:

Inner accretion disk not hot enough to emit X-rays (but can have X-ray emitting ADAF if accretion rate is low) harder X-ray spectrum

Core is always present in nearly all radio-loud AGN jets

Unbeamed

Beamed

Jets of Low-Luminosity AGNs

Jets often seem to be interacting with clouds

Apparent motion usually < c

Liners

Seyfert

III Zw 2 (Brunthaler et al. 2005, A&A, 435, 497): 0.6c at 15 GHz, 1.2c at 43 GHz

Blazar Jets: 3C 279

Superluminal motion between ~5c & 20c, bulk Lorentz factor up to 25, Doppler factor up to 50

Changes in apparent speed may be due solely to change in direction of jet by about ±2o

Blazar Jets: PKS 1510-089

Apparent speeds up to 45c (fastest known blazar containing well-defined superluminal knots) bulk Lorentz factor of at least 45 in jet

Radio-Loud AGN: The General Population

Relativistic beaming causes strong selection effect in flux-limited radio surveys Bias toward high- jets pointing almost directly along line-of-sight

Population simulation (Lister & Marscher 1997): observed apparent-motion & redshift distribution reproduced if:

1. Radio-galaxy luminosity function measured at low z is valid at higher z

2. Lorentz factor distribution is a power law, N() -a, a = 1.5-1.75, with a high- cutoff of 45 (highest observed app)

12-17% of jets in population have = 10-45 5-7% have = 20-45, 2-3% have = 30-45, 0.5-0.9% have = 40-45

Spine-sheath models for compact AGN jets requiring a very high- spine in a typical jet are untenable unless radiation from the spine is suppressed

- But such ultra-fast spines should be prodigious emitters of inverse Compton X-rays off ambient photon field (e.g., CMB)

Intrinsic Half Opening Angles of Jets (Jorstad et al. 2005, AJ, 130, 1418)

Intrinsic Half Opening Angles of Jets (Jorstad et al. 2005, AJ, 130, 1418)

Blazars:

1/

Agrees with models in which jet is

focused as it is accelerated over an

extended region.(HD: Marscher 1980;

MHD: Vlahakis & Königl 2004)

Explains why apparent opening angle is

uncorrelated with apparent speed

Side-on radio galaxies:

Opening angles typically 1-4o

Knots in Jets

stationary

8c

Polarization:BL Lac objects usually have B ~ transverse to local jet axis well downstream of core

Knots in Jets

Polarization:Quasars generally have oblique direction of B after aberration taken into account

Shock Model for Knots in Jets

Best-liked model: Shocks propagating down turbulent jetMagnetic field compressed at shock frontElectrons accelerated at shock front

Polarization indicates that in general such shocks must be oblique, especially after correcting for aberration

Need supersonic relative motion to get shock waves strong shocks are difficult for high- flows with relativistic equation of state (but don’t need very strong shocks for substantial enhancement of radiation)

Bends in Jets

Bending:Apparent bends amplified greatly by projection effectsIntrinsic bends by only a few degrees

3C 4460528+134

Changes in Direction

Change in apparent speed can be due solely to change in directionNonthermal luminosity seems to be related to direction of jetChanges amplified greatly by projection effectsVelocity seems ballistic in some jets but seems to follow twisting jet in many others

Changes in direction appear to be abrupt, unlike precession (more like an unstable firehose)

The Core of Blazar Jets

Frequencies below ~ 40 GHz: ~ 1 surfaceAt higher frequencies:a. Conical standing shock? (Daly & Marscher 1988)- See poster by Cawthorne et al. (e.g., 1803+784 shown below)-In favor: reproduces polarization pattern if randomly oriented B field is compressed by conical shock-b. End of zone of accelerating flow- Where Doppler factor reaches asymptotic value

1803+784

Jet Acceleration over Extended RegionJet Acceleration over Extended Region

HD: Pressure gradient p r-a Lorentz factor increases with cross-sectional radius R:

Γ R p-1/4 ra/4

If a < 4/(3+1) and viewing angle is small, brightest emission is where Γ reaches its asymptotic value

If viewing angle is large, brightest emission is at lowest r where high-E electrons are accelerated

(Marscher 1980 ApJ)

MHD: Models still being developed

Vlahakis & Königl (2004, ApJ) solution appears similar to HD solution, except that Γ decreases away from jet axis & there is no distinct boundary

In either case, energy density at base of jet must exceed ~ 2Γρc2

Might require a magnetosphere (pulsar or ergosphere of spinning BH)

Theory: A jet with > ~10 cannot propagate out of nuclear region (Phinney 1987)

Predicts toroidal field, but perhaps only close to central engine, where opacity is too high to image

Cygnus A (Bach et al. 2004, 2005)FR II radio galaxy, jet at large angle to l.o.s.

Gap between core & counterjet < 0.7 masApparent speed increases with distance from core

CoreCounter-core

Evidence for Collimation of Jets Well Outside Central Engine

• VLBA observations of M87: jet appears broad near core→ Flow appears to be collimated on scales ~1000 Rs

Junor et al. 2000 Nature

The FR I Radio Galaxy 3C 120 (z=0.033)

HST image (Harris & Cheung)

Scale: 1 mas =

0.64 pc = 2.1 lt-yr

(Ho=70)

• Superluminal apparent motion, ~5c (1.8-2.8 milliarcsec/yr)

• X-ray spectrum similar to Seyferts• Mass of central black hole ~ 3x107

solar masses (Marshall, Miller, & Marscher 2004; Wandel et al. 1999)

Sequence of VLBA images (Marscher et al. 2002)

X-Ray Dips in 3C 120

Superluminal ejections follow X-ray dips Similar to microquasar GRS 1915+105

Radio core must lie at least 0.4 pc from black hole to produce the observed X-ray dip/superluminal ejection delay of ~ 60 days

Comparison of GRS1915+105 with 3C 120 Light Curves

BH mass of 3C 120 ~2x106 times that of GRS 1915+105, so timescales of hours to months in the former are similar to the scaled-up quasi-periods (0.15 to 10 s) & duration of X-ray dips in the latter.Typical fractional amplitude of dips is also similar Long, deep dips not yet seen in 3C 120

blow-up

← GRS 1915+105 over 3000 s on 9/9/97Light curve (top) & PSD (bottom)(Taken from Markwardt et al. 1999 ApJL)

Perhaps low-hard X-ray state corresponds to 3C 120

150 s of blow-up should scale up to ~10 yr in 3C 120 if timescales Mbh

Below: X-ray light curve of 3C 120 over 2.2 yr

FR II Radio Galaxy 3C 111 (z=0.0485) Seems to Do the Same

Superluminal ejection follows minimum of deep X-ray by 0.3 yr

Radio core must lie at least 0.4 pc from black hole to produce the observed X-ray dip/superluminal ejection delay

May 2004

August2004

New knot

1 mm flare

1 milliarcsec

Accretion States of AGNs

Power spectral density of Seyferts similar to high-soft state of Cygnus X-1 (McHardy et al. 2004)-Weak jets of Seyferts consistent with weak/no jet in high-soft state of GRS1915+105 (Fender & Belloni 2004)-Inner accretion disk not hot enough to emit X-rays spectrum not so soft (mean spectral index of 0.9)

X-ray spectra of radio galaxies 3C 120 (FR 1) & 3C 111 (FR 2) flatter than this- Suggestive of low-hard state with ~ steady, optically thick jet seen in GRS1915+105

Liners and low-luminosity Seyferts may have ADAFs near black hole

Seyfert PSDs from McHardy et al. (2004)

High break timescale scales approximately linearly with mass

Sketch of Physical Structure of Jet, AGN

CORE

Relation of AGN jets to XRBs & GRBs

• Bulk Lorentz factors of jet flows can exceed 40c - not too dissimilar to GRBs - but only rarely

→ Ultra-fast ( > 10) spines cannot be general feature in AGN → Blandford-Payne type jet launching might be sufficient in high fraction of AGN

• X-ray variability of high-luminosity Seyferts has similar PSD to XRBs in high-soft state, with weak jets

• X-ray spectrum of radio galaxies with strong jets flatter than in Seyferts, similar to low-hard state

• Evidence for acceleration & focusing of jet over an extended region is mounting

→ Conforms with HD & some MHD models for jet launching