Hard X-ray spectral variability of the brightest Seyfert AGN in the Swift/BAT sample

TL;DR

This study analyzes the hard X-ray spectral variability of the five brightest Seyfert AGN using Swift/BAT data, revealing significant flux variations, correlations with black hole mass, and insights into the corona's temperature behavior.

Contribution

It provides the first detailed analysis of 20-100 keV variability in these AGN, highlighting the relationship between flux, spectral shape, and accretion states.

Findings

All objects show significant flux variability.

An anti-correlation exists between variability amplitude and black hole mass.

Spectral slope varies with flux in some objects, indicating changes in coronal temperature.

Abstract

Aims: We used data from the 58 month long, continuous Swift/Burst Alert Telescope (BAT) observations of the five brightest Seyfert galaxies at hard X-rays, to study their flux and spectral variability in the 20-100 keV energy band. The column density in these objects is less than 10^24 cm-2, which implies that the Swift/BAT data allow us to study the "true" variability of the central source. Results: All objects show significant variations, with an amplitude which is similar to the AGN variability amplitude at energies below 10 keV. We found evidence for an anti-correlation between variability amplitude and black hole mass. The light curves in both bands are well correlated, with no significant delays on time scales as short as 2 days. NGC 4151 and NGC 2110 do not show spectral variability, but we found a significant anti-correlation between hardness ratios and source flux in NGC 4388 (and NGC 4945, IC 4329, to a lesser extent). This "softer when brighter" behaviour is similar to what has been observed at energies below 10 keV, and cannot be explained if the continuum varies only in flux; the intrinsic shape should also steepen with increasing flux. Conclusions: The presence of significant flux variations indicate that the central source in these objects is intrinsically variable on time scales as short as 1-2 days. The intrinsic slope of the continuum varies with the flux (at least in NGC 4388). The positive "spectral slope-flux" correlation can be explained if the temperature of the hot corona decreases with increasing flux. The lack of spectral variations in two objects, could be due to the fact that they may operate in a different "state", as their accretion rate is less than 1% of the Eddington limit (significantly smaller than the rate of the other three objects in the sample).