Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 342, L41-L44 (1999)

Previous Section Next Section Title Page Table of Contents

4. Discussion and conclusions

BeppoSAX high energy instrument, PDS, has succeeded in measuring for the first time the X-ray spectrum of the Seyfert 2 galaxy NGC 2110 at energies above 15 keV. The spectrum, having a photon index [FORMULA] is fully compatible with what expected from a Seyfert 1 nucleus (Nandra and Pounds 1994). The interpretation of the flat ([FORMULA]) spectrum observed between 2 and 10 keV in terms of reflection is neither required by the data ([FORMULA]) nor is compatible with them, since the observed Fe EW would imply a reflection a factor of 5 greater than the observed upper limit. The flat 2-10 keV spectrum is reconciled with unified models and with the [FORMULA] keV power-law slope assuming the presence of a complex absorber. In the framework of unified models the two ([FORMULA] and [FORMULA]) column densities can be associated with the torus or with a combination of the BLR plus an additional absorber.

In the first scenario the dual absorber can be explained in terms of a complex configuration of the torus: an inner, denser and tidally disrupted region plus an outer more homogeneous one which account for the partial and total absorbing column respectively. Alternatively, the complex absorber can be associated with the presence of a torus "atmosphere" (Feldmeier et al. 1998). In the second case the partial absorber is physically associated with the clouds of the BLR, assuming that they are small if compared with the nuclear X-ray source size or the source appears much larger due to nearby X-ray scattering induced by highly ionized gas. The upper limit on the BLR clouds Doppler velocity ([FORMULA] few 103 km/s) deduced from the FeK width is also consistent with this hypothesis. The second (total) absorber can again be the torus itself (Hayashi 1996) or the Intermediate Line Region introduced by Cassidy & Raine (1997) for the Seyfert 1.5 NGC 4151. In NGC 2110 no obvious broad line components have been detected (Veilleux et al. 1997). Moreover the observed nuclear reddening provides only a lower limit ([FORMULA] mag) on the amount of obscuration to the nucleus (Mulchaey et al. 1994). Therefore the present data do not allow to firmly assess where the absorption regions are to be located and therefore to discriminate among the proposed scenarios.

The interpretation of the flat 2-10 keV spectrum of Seyfert 2 galaxies as the result of an artifact caused by the presence of a complex absorber is not new, since it has been suggested to explain the X-ray spectrum of NGC 5252 (Cappi et al 1996), IRAS 04575-7537 (Vignali et al. 1998), and NGC 7172 (Guainazzi et al. 1998).

The detection of an "excess" FeK edge in the spectrum of ESO 103-G35, NGC 7314 (Turner et al. 1997), and possibly NGC 7582 (Xue et al. 1998), suggests that an additional unmodelled absorber is present in these sources. While this could indicate that the dual absorber scenario can be applicable to Seyfert 2 galaxies and NELGs in general, other observational results show that this model is not the universal solution for reconciling flat and "canonical" (i.e. [FORMULA]) sources. Namely, the BBXRT spectrum of NGC 4151 results in a flat ([FORMULA]) power law also after the addition of a complex ionized absorber (Weaver et al. 1994). Moreover, Turner et al. (1997) show that the ASCA spectrum of NGC 2110 remains flat ([FORMULA]) also after the addition of ionized material partially covering the source. Finally, the recent results on the ASCA spectrum on the Seyfert 1.5 LB 1727 shows an unattenuated flat ([FORMULA]) power law (Turner et al. 1999).

Clearly, the easiest interpretation of the flat spectrum of NGC 2110 is that the source is actually intrinsically flat. In this case the observed spectral steepening at [FORMULA] keV could be the exponential cut-off of the primary power law. We have tested this model with our data. The best fit ([FORMULA]) gives a photon index [FORMULA] with a cutoff energy value [FORMULA]. We cannot, therefore, discriminate among this model and the dual absorber one on a statistical basis. As far as the physical interpretation in concerned, however, we prefer the interpretation of the 0.5-150 keV spectrum of NGC 2110 in terms of a dual absorber for several reasons. The exponentially cut-off power law fails in explaining the observed FeK line EW and absorption edge optical depth, unless a large ([FORMULA]) iron overabundance is introduced. Moreover, the high energy ([FORMULA] keV) spectrum in the dual absorber model is fully consistent with the slope measured using the PDS data only. Finally, the results listed above regarding flat X-ray spectrum Seyfert galaxies have all been obtained with instruments operating only up to 10 keV, while we have shown in this work that the steep, intrinsic spectrum of NGC 2110 is measurable only above 13-20 keV.

In summary, the new result on the steepening of the NGC 2110 spectrum above 13 keV is well explained assuming the presence of a complex absorber. On the other hand, this model fails in explaining the flat 2-10 keV spectrum observed in several other Seyfert galaxies, and cannot, therefore be accepted as a universal explanation of "flat-spectrum" Seyfert galaxies. It is important to point out that this issue can be fully addressed only with future broad-band observations of an extended sample of flat-spectrum Seyfert galaxies covering the entire 2-100 keV region.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1999

Online publication: February 23, 1999