The X-ray absorber of PKS2126-158

BeppoSAX observed the z= 3:27 quasar PKS 2126-158 on May 24-28 1999 when its 2 10 keV and 0.1 2.5 keV fluxes were 1:1 10 11 and 4:4 10 12 erg cm 2 s 1 respectively, a factor of 2 higher than in all previous ROSAT and ASCA observations and 40% higher than in two more recent Chandra and XMM-Newton observations. The shortest detected rest frame variability timescale is of a few months, comparable to the causal timescale associated with an emission region of10 Schwarzschild radii around a few10 10 M black hole. The source is detected with a signal to noise ratio S=N > 3 up to50 keV, 215 keV rest frame. The BeppoSAX observations confirm the presence of low energy absorption along the line of sight, independent of the continuum model adopted, at a high confidence level. Despite the limited spectral resolution of the BeppoSAX LECS and MECS it is possible to put constraints on dierent absorption and continuum models, but not to unambiguously determine the redshift of the absorber. If the absorber is not significantly ionized the BeppoSAX data do prefer an absorber at z< 2:7. Strong and complex metal line systems along the line of sight to PKS 2126-158 have been found at z= 0:6631 and at 2:64< z< 2:82. They could well be associated with the X-ray absorption. Conversely, an ionized ("warm") absorber at the quasar redshift provides a good fit only if the iron abundance is smaller than 0:3 solar, while that of the other elements is fixed to the solar value. Lower iron metallicity would imply a lower dust to gas ratio, since iron aggregates easily in dust. This can help in solving the apparent paradox of the lack of significant ultraviolet reddening in this source while strong absorption is detected in X-rays. Low iron abundance would be at odds with the supersolar abundances derived from the broad emission lines.