Chandra and XMMNewton FollowUp Observations

Chandra and XMM-Newton allow us to measure the long-term evolution of the light curves and spectra of the flares. In particular, Chandra with its high spatial resolution enables us to pinpoint the counterpart of the flare precisely. In case of tidal disruption, the emission should come directly from the center of each galaxy.

RX J1242-1119 [13] was the first target of choice for follow-up X-ray observations because it flared most recently, so the probability of catching the source in the declining phase-before it had faded away completely-was highest.

Chandra detected a huge drop in X-ray flux by a factor ^200 [14], compared with the high-state ROSAT observation. The point-like late-phase flare emission

Fig. 21.1 Upper panel: Artist's sketch of the tidal disruption of a star. The star is ripped apart by the tidal forces of a massive black hole. Part of the stellar debris is then accreted. This causes a luminous flare of radiation that fades away as more and more of the matter disappears into the black hole. Lower panel: optical image of the galaxy (pair) RXJ1242-1119A, B (right) and "afterglow" of the X-ray flare from its center (left) [image credit: NASA/CXC/M.Weiss/ESO/MPE; Komossa et al. 2004]

Fig. 21.1 Upper panel: Artist's sketch of the tidal disruption of a star. The star is ripped apart by the tidal forces of a massive black hole. Part of the stellar debris is then accreted. This causes a luminous flare of radiation that fades away as more and more of the matter disappears into the black hole. Lower panel: optical image of the galaxy (pair) RXJ1242-1119A, B (right) and "afterglow" of the X-ray flare from its center (left) [image credit: NASA/CXC/M.Weiss/ESO/MPE; Komossa et al. 2004]

coincides with the center of the galaxy RX J1242-1119A (Fig. 21.1). With XMM-Newton, for the first time a good-quality X-ray spectrum of one of the few flaring galaxies was obtained [14]. The spectrum is well fit by a power law (photon index rx = -2.5), harder than during outburst. This spectral shape is typical for the emission spetrum of matter in the immediate vicinity of a black hole.

A further decline of the X-ray emission is expected as more and more of the stellar debris is accreted by the black hole. Indeed, reobservation of RXJ1242-1119 with Chandra in 2004 showed a further fading of the X-ray source by a factor of several (Komossa et al. 2007, in prep.) such that the total amplitude of variability exceeds a factor of 1 000.

Two more flares were followed up with Chandra. Only few, if any, photons from the galaxies' centers were detected, making their total amplitude of variability extremely large, a factor > 1000 (NGC 5905, [9]) and >6 000 (RX J1624+75, [9, 22]).

RXJ1420+53 o

RXJ 1624+75 A

1990 1992 1994 1996 1998 2000 2002 TIME [y]

Fig. 21.2 Collective X-ray light curves of the four flaring galaxies, all shifted to the same peak time. The dashed line follows a t-5/3 law matched to the high- and first low-state data point of RXJ1242-1119. The dotted line corresponds to the least fastest rise to high-state consistent with the upper limit of RX J1420+5334. The last data point of NGC 5905 corresponds to the central point source only, while previous X-ray measurements of NGC 5905 plotted in this figure include both extended and core emission not spatially resolved prior to Chandra [15]

Figure 21.2 shows the collective X-ray light curve of the flare events, shifted in time to the same high-state as NGC 5905. The light curves of all events are relatively similar, and consistent with a faster rise, and a slower decline on the timescale of months-years. Most of the data points are consistent with a t-5/3 decline law, which is expected for the "fall-back" phase of tidal disruption (e.g. [21,17]).1

In summary, the observed events match basic predictions from models of the tidal disruption of stars by supermassive black holes at the centers of the flaring galaxies (see [12,9,14] for a more detailed discussion). The observed X-ray flares have the highest amplitudes of variability-up to a factor >6 000 - ever recorded among galaxies, including active galaxies. The rate of tidal disruption events, estimated on the basis of ROSAT observations, is event per galaxy per 104 yrs [3], roughly consistent with theoretical predictions [24].

1 This law only holds if the bolometric correction is constant, i.e., if most of the emission remains in the X-ray band, as time evolves. It is approximate for other reasons as well.

0 0

Post a comment