ForumSpringerAstron. Astrophys.
ForumWhats NewSearchOrders

Astron. Astrophys. 317, 25-35 (1997)

Previous Section Next Section Title Page Table of Contents

5. Summary and conclusions

Close to the well defined optical position of the QSO Ton S180 ROSAT discovered the bright and soft X-ray source RX J0057.3-2222 during the all-sky survey in December, 1990. There were pointed follow-up observations of the X-ray source at four occasions in 1992 and 1993. The optical image of the QSO is well contained in the 90% error circle of the centroid of the X-ray point source image, so that an identification of RX J0057.3-2222 with the optical QSO is highly probable.

A study of the temporal behaviour of the X-ray emission of RX J0057.3-2222 in ROSAT's (0.1 - 2.4)keV energy band revealed intensity as well as spectral variability. Since the observations took place in three different years, each time for several but not always successive orbits, a temporal analysis of the observed X-ray emission cannot be undertaken on a continous grid of time scales. The limited time structure of the observations allows only to study the intra-orbit variability with a shortest time scale of 400s, the orbit-to-orbit variability where observations extend over adjacent orbits, and the long-term year-to-year variations. Nevertheless, it can be stated that Ton S180 shows a great variety of variability in its soft X-ray emission. The long-term variability on a time scale of years, which we observed with no indications of spectral changes, shows amplitudes up to a factor of two. On the other hand, the amplitudes usually found for the orbit-to-orbit variability [FORMULA] do not exceed 30 per cent, a level which is also measured for the intra-orbit intensity variations [FORMULA]. Generally, the latter variations are smooth. No "noisy" orbits were found as in the case of BL Lac objects.

Now and then Ton S180 shows outbursts like that measured during the pointed observations in June 1992. The X-ray emission rose from the actual persistent level of about 3.3 cts/s by more than a factor of two within less than a day. Although the ROSAT measurements show rather extended gaps, it looks as though the top level of the outburst might decay on a much longer time scale than that of the outburst's rise. Again, no indications of a spectral change accompanying the intensity jump could be found.

The investigation of the quasar's variable X-ray emission revealed that most of the time the spectrum is unchanged even in the case of strong intensity variations. In order to investigate the mean characteristics of the quasar's soft X-ray spectrum during those time intervals we merged the data sets of the pointed observations which were performed in June 1992 and one year later. These data do show intensity variations but no clear spectral variability. The mean spectrum is well represented by a two-component model consisting of a power law and a steep low energy component dominant below 0.3 keV. The influence of the low energy absorption (Galactic value [FORMULA], Stark et al., 1992) on this component of the intrinsic continuum prevents the exact determination of its spectral shape. We represented it by a black body spectrum, but it should be noted that another choice of model spectra like a second steeper power law or thermal bremsstrahlung does not change the goodness of the fit. The mean photon index of the power law component yielded by fitting the two-component model to the merged count rate spectrum is rather steep, [FORMULA], whereas the temperature of the black body is unusually low, [FORMULA], thus demonstrating that this component is dominant only at very soft photon energies. Estimating the flux of the X-ray source in the (0.1 - 2.4)keV band on the basis of the two-component model for the continuum, we obtained an unabsorbed mean flux of [FORMULA] which yield a soft X-ray luminosity of [FORMULA] for a redshift of z = 0.06198 ( [FORMULA] ).

Despite superb photon statistics of the mean spectrum, the strength of the steep low energy component remains more or less undefined. It is, therefore, difficult to compare the ROSAT (0.1 -2.4)keV flux with the (0.07 - 0.21)keV EUVE flux which has to be derived from the measured count rate of 0.053 cts/s by adopting a model spectrum for the continuum. The EUVE/Lexan energy range covers almost exactly that part of the spectrum, which is most affected by the cold matter absorption in our Galaxy, and where a possible steep component begins to dominate the power law continuum. Assuming the continuum to be represented by a power law only, Vennes et al. (1995) estimated a monochromatic flux density of [FORMULA] at [FORMULA] (or [FORMULA] ) adopting a photon index of [FORMULA] and an [FORMULA]. With our fit parameters we derive from the mean X-ray spectrum the ROSAT flux density of [FORMULA] at 0.14 keV. This value is more than a factor of five larger than the reported monochromatic EUVE flux, but it should be kept in mind that both observers adopted different slopes and absorbing column densities to reduce their flux densities. These different choices do not allow us to compare the reported EUVE flux directly with that one measured with ROSAT. Due to the exponential effect of the absorbing column density on the measured flux below 0.2 keV we speculate that the quasar's emission at very soft X-rays was clearly weaker at the time of the EUVE measurement than during the pointed observations with ROSAT.

The detailed study of the parameters of individual orbit spectra confirms that there are spectral variations neither of the power law continuum nor of the black body component for the data measured in mid 1992 and one year later. On the other hand, a clear change of the quasar's spectrum can be stated for the time interval between the observation on Dec 18, 1992, and the measurement performed 23 days later in Jan, 1993. In Dec 1992 the power law component of the spectrum is steep ( [FORMULA] ) and the low energy component is definitely existent. Twenty three days later the power law component has become flatter ( [FORMULA] ) without a noticeable change of its flux contribution. This component looks as though it is rotated relative to the steeper spectrum measured in Dec 1992 around a crossover point near 0.6 keV. Simultaneously with the flattened power law component we find the black body component weakened by at least a factor of four in Jan 1993. In fact, the strength as well as the temperature of this component is consistent with zero for this observation. The relative change in the total flux within (0.1 - 2.4)keV accompanying the spectral variation is only about 10 per cent. Half a year later all spectral parameters are back again to the values measured a year before and during the all-sky survey.

By inspecting the residuals of all orbit spectra it can be excluded that the ROSAT spectra show any sign of a warm absorber on the line of sight.

As mentioned above, Ton S180 is a narrow line QSO ( [FORMULA], Wisotzki et al. (1995)) with an optical Seyfert 1 spectrum. Also its soft X-ray properties go with those of the class of narrow line Seyfert 1 galaxies (NLSY1): with its characteristic parameters [ [FORMULA], [FORMULA] )] = [3.10, 900] the QSO belongs well to the domain of NLSY1s in Fig. 8 of Boller, Brandt, and Fink (1995) which shows that NLSY1s usually have steeper soft X-ray spectra than broad-line Seyfert 1 galaxies. A second feature of NLSY1s is their X-ray variability, the doubling time of which ranges from 800 s to several hours. The formal doubling time of the intra-orbit variations of Ton S180 was measured to be about 2 hours, whereas for the doubling time of the outburst's rise only an upper limit of one day could be given due to an observational gap of 15 orbits. Therefore, as far as the time scales of the observed variability are concerned, the QSO Ton S180 behaves like a NLSY1. To date, the medium energy spectrum of Ton S180 above 2 keV has not been measured. Only for one of the brightest NLSY1 galaxies, RE 1034+39, discovered with the ROSAT Wide Field Camera (Pounds, 1994), a measurement of the medium energy spectrum with the detectors aboard the ASCA satellite was reported (Pounds, Done, and Osborne, 1995). Surprisingly, the source shows a steep (2 - 10)keV power law spectrum with a photon index of [FORMULA]. It will be of high interest to see whether such steep hard X-ray spectra are a class property of NLSY1s, and whether narrow line QSOs like Ton S180 behave analogously.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997