Jets and accretion in -ray emitting Narrow Line Seyfert 1s J OSEFIN - - PowerPoint PPT Presentation

Jets and accretion in ɣ-ray emitting Narrow Line Seyfert 1s

1

JOSEFIN LARSSON

KTH & Oskar Klein Centre, Stockholm

Supermassive black holes and jets

2

Type 1 Type 2 Blazar Central supermassive black hole with mass ~ 106 - 1010 M⦿ ~ 15% of AGN have large-scale relativistic jets Jet formation is poorly understood Jets are almost exclusively found in large elliptical galaxies with massive black holes (M > 108 M⦿, e.g, Sikora+2007, Chiaberge+2011)

What is a NLSy1?

3

• FWHM (Hβ) < 2000 km s-1
• [O III]/Hβ < 3
• Strong Fe II bump

Defined from optical spectra:

It’s a Seyfert 1 where the broad lines are kind of narrow.

• Spiral host galaxies
• Lower fraction of radio loud objects

(~ 7%, Komossa+2006)

• X-rays: rapid large-amplitude variability,

steep spectra, strong soft excess. (Boller+1996, Gliozzi+2020) Other properties

1 10 0.5 2 5 1 1.5 2 2.5 3 ratio Energy (keV)

Energy (keV) Data/Model ratio

soft excess

What is a NLSy1?

4

Usually interpreted as “low-mass” black holes (~ 106 - 107 M⦿) with high accretion rates.

No fundamental difference compared to regular Seyfert 1s, but interesting as they occupy an extreme end of AGN parameter space

Detection of ɣ-ray emission from NLSy1s

5

In 2009, the Fermi gamma-ray space telescope detected gamma-ray emission from a NLSy1 for the first time, revealing blazar-like properties (Abdo+09).

Are we really seeing powerful jets from low-mass black holes hosted by spiral galaxies?

• 9 ɣ-ray NLSy1s (z ~ 0.06-0.8) in the 4th Fermi LAT source catalogue

(compared to ~2000 blazars, Abdollahi+2019)

• ~20 “candidate” ɣ-ray NLSy1s (e.g. Paliya+2018)

Evidence for relativistic jets

6

SEDs similar to Flat Spectrum Radio Quasars (FSRQs), including ɣ luminosities ~ 1044-1048 erg s-1 and steep ɣ-ray spectra (photon indices ~2.5).

(Abdo+09)

Evidence for relativistic jets

7

2.0 4.0 6.0 Total Flux (RC) [10-12erg/cm2/s] MJD=56281

• Tot. Flux

MJD=56283 10 20 30 40

• P. D.

[%] P.D. 90

• 90

17 18 19 20 21 UT [hour] 17

Rapid multiwavelength variability. High and variable polarisation.

Optical observations of PMN J0948+0022 (Itoh+13) Gamma-ray flares in SBS 0846+513

(D’Ammando+13)

Evidence for relativistic jets

8

Radio properties: flat or inverted spectra, strong variability, pc-scale core- jet structures and superluminal motion.

MOJAVE observations of the radio jet in 1H0323+342. quasi-stationary point at 7 mas (~100 pc). Recollimation shock

• possible site of ɣ-ray

emission. (Doi+18)

Black hole masses

9

Viral mass estimates give 106 - 107 M⦿, but may be underestimated due to effects of radiation pressure and/or a flattened BLR. Other methods, such as accretion disc fitting give higher masses (e.g.

Calderone+2013)

Summary of results for the ɣ-ray NLSy1s 1H0323+342

[M] 1H0323+342 single-epoch Hβ 107 [156] Pα 2 107 [89] reverberation mapping 6 106 [139], fBLR = 1 3 107 [139], fBLR = 6 (opt–UV) SED modelling 107 [2] host galaxy 2–4 108 [93] X-ray excess variance 107 [147] PSD break frequency 3–8 106 [117] PKS 1502+036 single-epoch Hβ 4–6 106 [150]

(Komossa+18)

Host galaxies

10

Limited information due to relatively high redshifts for many ɣ-ray NLSy1s. Some interesting examples:

1H0323+342 (z=0.063) Recent merger or one-armed spiral? (Leon

Tavares+14)

SDSS J2118-0732 (z=0.26) Disturbed morphology, possible merger (Yang+18).

NOT (R band) SDSS (r band)

11

PKS 1502+036 (z=0.41) Elliptical with some circumnuclear structure (D’Ammando+18). PKS 2004-447 (z=0.24) pseudobuldge+disc+bar (no evidence of merger) (Kotilainen+16). FBQS J1644+2619 (z=0.15) Barred lenticular (Olguin-Iglesias+17)

• r elliptical (D’Ammando+17).

VLT/ISAAC (J band) GTC/CIRCE (J band) VLT/ISAAC (J band)

X-ray emission from non-jetted AGN

12

0.1 1 10 100 1000 104 2000 5000 E FE Energy (keV)

Power-law from corona

Energy (keV)

E FE

0.1 1 10 100 1000 104 E FE Energy (keV)

Reflection spectrum Reflection spectrum

Energy (keV)

E FE

Compton hump Fe line

(from Athena White paper)

+ reflection from

distant material

• absorption

+ soft excess (?)

Models for the soft excess

13

0.1 1 10 100 1000 104 E FE Energy (keV)

Reflection spectrum Reflection spectrum

Energy (keV)

E FE

Compton hump Fe line Soft excess

• 1. It is simply part of the

reflection spectrum.

Soft excess seems too strong for this explanation in many cases. Solved with higher density disk (Jiang+19)?

• 2. It is due to Comptonisation

in an topically thick (𝝊~10-20) warm (kT~0.5 keV) corona

How is this produced? Would it be stable?

(Petrucci+20)

X-ray emission from blazars

14

Relativistically beamed emission from the jet completely dominates the X-ray

• spectrum. Power-law spectrum.

Synchrotron or Inverse Compton depending on the type of blazar. What are the properties of the inner accretion flow and coronae in AGN with jets?

X-ray spectra ɣ-ray NLSy1s

15

2-10 keV spectra well fitted by hard power law (Γ ~ 1.2-1.8) More similar to FSRQs than RQ NLSy1s! No broad Fe lines. Narrow Fe lines from distant material in

• ne case (1H0323+342).

Inverse Compton emission from the jet dominate the spectra above 2 keV.

Energy (keV) (Kynoch+18)

Soft X-ray excess

16

1 10 0.5 2 5 1 1.5 2 2.5 3 ratio Energy (keV)

Energy (keV)

Data/Model

The majority of sources with good- quality X-ray spectra show excess emission when the hard power law is extrapolated to low energies. This is vary rare in blazars.

(nergy (NeV)

1orm CtV V−1 NeV−1

DDtD/0odel (nergy (NeV)

1orm CtV V−1 NeV−1

DDtD/0odel (nergy (NeV)

1orm CtV V−1 NeV−1

DDtD/0odel

PMN J0948+0022 1H0323+342 FBQS J1644+2619 J1222+0413

Soft X-ray excess

17

The full 0.5-10 keV spectrum is usually well described by a broken power law with a break at ~ 2 keV. No evidence for intrinsic absorption.

1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8

Γ1

1.0 1.2 1.4 1.6 1.8 2.0 2.2

Γ1

FB46 M1644+2619 301 -0948+0022 1+ 0323+342 3.6 2004-447 3.6 1502+036

• 1222+0413
• 2118-0732

Broken power-law fits to the X-ray spectra of ɣ-ray NLSy1s (only highly significant ɣ-ray sources with XMM observations)

Physical models for the X-ray emission: FBQS J1644+2619 as an example

18

20000 40000 60000 80000

Time (R)

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6

iste (couttR R−1 )

Grey intervals affected by high background, excluded from spectral analysis

XMM EPIC pn light curve

107 109 1011 1013 1015 1017 1019 1021 1023 1025 1027

Frequency (Hz)

10-16 10-15 10-14 10-13 10-12 10-11 10-10

νFν (erg cm−1 s−1 )

FB46 -1644+2619 this work FB46 -1644+2619 arFhival

Highest ɣ-ray flux recorded in 2012 (D’Ammando+15)

80 ks XMM-Newton observation in 2017 together with MW campaign (Medicina, REM, Swift, Fermi). (Larsson+18)

Simple models

The X-ray spectrum is well described by a broken power law The jet dominates the spectrum above the break: The hard photon index is compatible with FSRQs, but much harder than in radio-quiet NLSy1s. No detection

• f an Fe line at 6.4 keV. No intrinsic absorption.

The low-energy part of the spectrum likely has a contribution from the underlying Seyfert emission.

19

Power law Γ = 1.82 ± 0.01 Χ2/d.o.f. = 478/355 Broken power law Γ1 = 1.90 ± 0.02 Γ2 = 1.66 ± 0.03 Ebreak = 1.9 ± 0.3 keV Χ2/d.o.f. = 349/353

etergk (keVD

10-3 10-2 10-1 100

Norm Cts s−d keV−d

0.3 1.0 10.0 0.8 1.0 1.2 1.4 1.6

Dsts/Model

I: Jet emission + ‘standard’ corona

20

Energy (keV)

0.3 1.0 10.0 −4 −4 −2 2 4

χ

2 pow

4 relxilllp

Power law + power law

Γ1 2.01+0.14

0.07

Norm 4 4 0 4 104

1 0 ⇥

Γ2 1.0+0.3

0.4 10 5 4 ⇥

χ2/d.o.f. 348/353

The photon index for the 2nd power law is extremely hard, even for a jet. The predicted hard X-ray flux is inconsistent with the non-detection by Swift/BAT.

II: Jet emission + warm corona

Γ 1.64+0.05

−0.08

PL Norm 4 18 0 03 10−4

−0.05 ×

kT0 (eV) 26f kTe (keV) 0.30+0.21

−0.11

τ 16+11

−4

Norm 1.4 0.3 10−2

−0 5 ×

χ2/d.o.f. 346/352

21

Energy (keV)

0.3 1.0 10.0 −4 −2 2 4

χ

pow +comptt

Optical depth and temperature of the ‘warm’ corona similar to typical parameters found for RQ sources. Power law + CompTT

Energy (keV)

0.3 1.0 10.0 10-6 10-5 10-4 10-3

T

• tal

PL CompTT

keV2 (Ph cm−2 s−1 keV−1 )

10 10 10 10 10 10

III: Jet emission + reflection

22

Energy (keV)

0.3 1.0 10.0 −4 −4 −4 −2 2 4

χ

relxilllp

Large height for the base of the jet. No significant emission from the innermost accretion disc. Rexilllp

h (rg) a incl () 5 Rin(rg)

Γ

45⇤

34

0.998f ) 5f 1.4+40.8

⇤

1.78 ± 0.01

log ξ ( AFe (k ± 1.6+0.3

0.2

1f

R

χ2/d.o.f. R 0.88⇤

0.15 3 8

• ⇥

355/351

10-6 10-5 10-4 10-3

T

• tal

PL relxilllp

Energy (keV)

0.3 1.0 10.0

keV2 (Ph cm−2 s−1 keV−1 )

10 10 10 10 10 10

23

Scenarios I-III all provide good fits, but scenario I unlikely due to the very hard jet emission. Combinations of the models also possible, but poor constraints on parameters. The spectrum is consistent with having a sub-dominant contribution from the underlying Seyfert emission.

24

Similar results for other ɣ-ray NLSy1s with long XMM-Newton observations

1H0323+342

(Kynoch+18)

Total model Jet Corona Disc reflection

PMN J0948+0022

(D’Ammando, Larsson et al., 2014)

X-ray variability

25

RMS spectra of PMN J0948+0022 and 1H023+342 show different variability properties below/above ~ 1.5 keV.

1 10 0.2 0.5 2 5 0.05 0.1 0.15 RMS Energy (keV)

PMN J0948+0022

D’Ammando, Larsson et al., 2014

1 10 0.2 0.5 2 5 0.05 0.1 0.15 0.2 RMS Energy (keV)

1H0323+342

Larsson et al., in prep.

Jet + Seyfert emission?

26

RQ NLSy1s typically show very strong soft-excess emission. It is not surprising that this is detected above the jet emission also in the ɣ-ray NLSy1s.

Only jet emisison?

X-ray observations in different flux states is needed in order to further test different scenarios for the soft excess in these sources. Tail of the synchrotron emission from the jet extending to soft X-rays? No evidence of this from SED modelling of ɣ-ray NLSy1s.

Testing the origin of the soft excess in ɣ-ray NLSy1s

27

If the soft excess is due to the underlying Seyfert emission it should

• be more prominent when the jet emission is fainter
• disappear (or weaken) when the jet emission is brighter

Testing the latter with XMM-Newton anticipated TOO programme (PI

Larsson). Trigger when ɣ-ray flux of a NLSy1 with a previously observed soft

excess is > 5 times the average value.

X-ray spectra of NLSy1s with radio detections

28

• ●
• ●
• 5e+42

5e+43 5e+44 5e+45 5e+46 0.5 1.0 1.5 2.0 2.5 3.0 X−ray Luminosity (2 − 10 keV) [ergs/s] Photon Index

• Current XMM NLS1 sample

Gamma−ray NLS1s

• 1e+22

1e+24 1e+26 1e+28 0.5 1.0 1.5 2.0 2.5 3.0 Radio Luminosity [W/Hz] Photon Index

• Current XMM NLS1 sample

Gamma−ray NLS1s

NLSy1 sample observed with SDSS+FIRST+XMM- Newton

Many NLSy1s with strong radio emission are probably similar to the ɣ-ray population (but not detected in ɣ-rays yet).

Kaur & Larsson, in prep.

Summary

29

NLSy1s are thought to harbour rapidly accreting low-mass supermassive black holes. ~ 10-20 NLSy1s have been detected in gamma rays. They have powerful relativistic jets close to the line of sight. The X-ray properties of ɣ-ray NLSy1s are intermediate between blazars and non-jetted NLSy1s. Most of them show a “soft excess”. These sources can offer new insights about the soft excess, the disc- jet connection and jet formation in general.