On the nature of unabsorbed Seyfert 2 galaxies

TL;DR

This study investigates six Seyfert 2 galaxies that appear unabsorbed in X-ray observations, revealing that many are heavily obscured or lack broad-line regions, challenging traditional AGN unification models.

Contribution

The paper provides detailed multi-wavelength analysis showing that some unabsorbed Seyfert 2s are actually obscured or intrinsically lack broad-line regions, offering new insights into AGN classification.

Findings

Many galaxies are under-luminous in 2-10 keV X-ray band.

Some Seyfert 2s show signs of heavy obscuration despite unabsorbed X-ray spectra.

Certain sources may intrinsically lack a broad-line region.

Abstract

We present an analysis of six 12 um selected Seyfert 2 galaxies that have been reported to be unabsorbed in the X-ray. By comparing the luminosities of these galaxies in the mid-IR (12um), optical ([O III]) and hard X-ray (2-10 keV), we show that they are all under-luminous in the 2-10 keV X-ray band. Four of the objects exhibit X-ray spectra indicative of a hard excess, consistent with a heavily obscured X-ray component and hence a hidden nucleus. In these objects the softer X-rays may be dominated by a strong soft scattered continuum or contamination from the host galaxy, which is responsible for the unabsorbed X-ray spectra observed, and accounts for the anomalously low 2-10 keV X-ray luminosity. We confirm this assertion in NGC4501 with a Chandra observation, which shows hard X-ray emission coincident with the nucleus, consistent with heavy absorption, and a number of contaminating softer sources which account for the bulk of the softer emission. We point out that such 'Compton thick' sources need not necessarily present iron Ka emission of high EW. An example in our sample is IRASF01475-0740, which we know must host an obscured AGN as it hosts a HBLR seen in scattered light (Tran 2003). The X-ray spectrum is nonetheless relatively unobscured and the iron Ka line only moderate in strength. These observations can be reconciled if the hidden nuclear emission is dominated by transmitted, rather than reflected X-rays, which can then be weak compared to the soft scattered light or galactic emission. Despite these considerations, we conclude that NGC3147 & 3660 may intrinsically lack a BLR, confirming the recent results of Bianchi et al. (2008) for NGC3147. Neither X-ray spectrum shows signs of hidden hard emission and both sources exhibit X-ray variability leading us to believe we are viewing the nucleus directly.