1. Dongyue Li 2018/4/27 Ref: Saxton et al. 2015 Esquej et al. 2008, 2012
Was the soft X-ray flare in NGC 3599 due to an AGN disk instability or delayed tidal disruption
Dongyue Li
2018/4/27
Ref: Saxton et al
Esquej et al. 2008, 2012

2. Background: NGC 3599 (XMMSL1 J111527.3+180638)
An early type galaxy in local universe: z=0.0028

3. X-ray observation
Detected in with an EP-pn keV count rate of 5.5 cts/s
RASS: ratio of 88 (upper limit)
ROSAT pointed: ratio of : 297 and 100 (upper limit)
Follow-up observations
Flux estimation: PIMMS
typical near maximum spectrum: KT=0.07 keV, modified with Galactic absorption
Unabsorbed luminosity in slew:

4. X-ray analysis
Best fit model：

5. X-ray light curve
(0.2-2 keV)

6. Optical observation (pre outburst)
The long-slit optical spectrum of NGC 3599 was taken in 2000−02−05 at the Whipple Observatory with the FAST spectrograph located in the 1.5 m Tillinghast telescope

7. Optical observation stellar population of 2.1 Gy Intensity ratios of:
[O III] 5007 Å/Hβ = 3.5(±1.0), [N II] 6584 Å/Hα = 1.1(±0.2)
The dotted curve shows the demarcation between starburst galaxies and AGN defined by Kewley et al. (2001). The dashed curve shows Kauffmann et al 2003 demarcation
Kauffmann et al.2003

8. X-ray light curve 2002-05-22: 0.2-2 keV: 19 photons in 4.0 s exposure
Flare phases are marked as: quiescent (Q), rise (R; ≤107 months), plateau (P; ≥18 months) and decay (D; ∼36 months).

9. X-ray spectral analysis
fittedwith a blackbody of kT∼45 eV plus a steep power law, with ∼2.7,
both absorbed by the Galactic column (1.42 × 1020 cm−2; Kalberla
et al. 2005). We fit this model, with free blackbody temperature
and power-law slope

10. TDE candidate Classical model of TDE: fast rise, days to months
-5/3 decay
if the measurement lies on the rise of the luminosity curve and the measurement on the decline
Rise time: 1.14 year

11. TDE candidate
Recently, numerical simulations and new analytical work have shown that the development of a TDE light curve is dependent on when the streams of tidal debris intersect each other (Guillochon & Ramirez-Ruiz). Early interactions appear to be rare and in the majority of cases circularisation occurs late and at a large distance from the BH, 5–10 times further away than predicted by the classical model (Piran et al. 2015). This leads to a longer viscous time-scale and a rise to peak that takes months, years or even decades.
For lower black hole masses (Mh 10  6), we find that flares are typically slowed down by about an order of magnitude, resulting in the majority of TDEs being sub-Eddington at peak. This also implies that current searches for TDEs are biased toward prompt flares, with slowed flares likely having been unidentified.

12. Evolved stars have weakly bound envelopes which may be stripped in black hole encounters without destroying the star (MacLeod, Guillochon & Ramirez-Ruiz 2012). These events, which result in light curves with a long rise to peak and a plateau phase lasting years

13. ULX The peak luminosity of NGC 3599,
LX = 5 E41 erg/s1, falls within the range attained by ULX
Stellar-mass accreting binaries and ULX have
significantly harder low-state spectra

14. A highly variable AGN
(i) an increase of factor 130 in X-ray flux within ≤9 yr;
(ii) an unusually soft peak spectrum ( index∼ 4/kT ∼ 90 eV);
(iii) A spectrum which remained soft while the flux dropped by a factor of
100 between 2003 and 2008
No obvious intrinsic absorption
Warm absorber: poorly fitted
Can not explain observation
An increase in intrinsic emission
A change in the disc structure
filling and emptying of the inner accretion disc
(2) Lightman-Eardley disc instability

16. (2) Lightman-Eardley (LE) disk instability
IC Grupe et al. 2015
The instability begins when the radiation pressure becomes comparable with the gas pressure. A heating wave is generated at the inner edge of the disc which propagates backwards through the disc at the sound speed (cs). The heating increases the local viscosity, scaleheight of the disc and the local accretion rate (Cannizzo 1996). At this point the disc is no longer a thin Shakura–Sunyaev disc but still emits as a thermalized plasma.
Rise time is longer than decay time

17. Summary
Based on new data we have shown that it is difficult to reconcile the
rise time of the soft X-ray flare seen in NGC 3599 with its previous
classification as a classical, fast-rising, short plateau TDE. It does,
however, fit into the emerging scheme of a disruption which has led
to a late, distant, circularization where the rise is flatter and the peak
lasts longer or alternatively to the tidal stripping of a red giant star.
If the factor of 100 soft X-ray flare is interpreted as a change
in the accretion from a persistent AGN, we find that the rise time
of ≤9 yr is too short to be explained purely by the bulk motion
of disc material about a 106M BH. The observed time-scales are
compatible with the LE instability which boosts the thermal soft
X-ray emission by heating the inner disk and raising its scaleheight.
This mechanism predicts that flares will repeat on the viscous timescale
of the truncation radius, i.e. every few decades. The behaviour
of NGC 3599 would then be analogous to flares seen in certain
Galactic binary systems, but is too rapid to be equivalent to the state
changes which are seen in those systems.