INTEGRAL observations of the cosmic X-ray background in the 5–100 keV range via occultation by the Earth

Abstract

Aims. We study the spectrum of the cosmic X-ray background (CXB) in energy range ∼5−100 keV. Methods. Early in 2006 the INTEGRAL observatory performed a series of four 30 ks observations with the Earth disk crossing the field of view of the instruments. The modulation of the aperture flux due to occultation of extragalactic objects by the Earth disk was used to obtain the spectrum of the Cosmic X-ray Background (CXB). Various sources of contamination were evaluated, including compact sources, Galactic Ridge emission, CXB reflection by the Earth atmosphere, cosmic ray induced emission by the Earth atmosphere and the Earth auroral emission. Results. The spectrum of the cosmic X-ray background in the energy band 5−100 keV is obtained. The shape of the spectrum is consistent with that obtained previously by the HEAO-1 observatory, while the normalization is ∼10% higher. This difference in normalization can (at least partly) be traced to the different assumptions on the absolute flux from the Crab Nebulae. The increase relative to the earlier adopted value of the absolute flux of the CXB near the energy of maximum luminosity (20−50 keV) has direct implications for the energy release of supermassive black holes in the Universe and their growth at the epoch of the CXB origin.

Supermassive black holes in deep multiwavelength surveys

“…Figure 8b shows the allowed regions in terms of the Compton-thick AGN factor versus reflection parameter. That is, the shaded regions show the values of these two parameters that produce an integrated X-ray background intensity in the 20-40 keV range, the region most affected by both the number of Compton-thick AGN and R, consistent with: (green region) the Gruber et al (1999) data increased by 40% (the maximum suggested renormalization to match higher estimates with imaging instruments at lower energies; Treister & Urry 2005); (red region) the recent INTEGRAL X-ray background intensity measured using Earth occultation by Churazov et al (2006); (blue region) the original Gruber et al (1999) measurements, not renormalized. In each case uncertainties in the X-ray background intensity were assumed to be 5%.…”

“…However, such a high value of the Compton-thick AGN factor is clearly inconsistent with the new observational constraints on the density of Compton-thick AGN. So, either the intensity of the X-ray background is lower, as suggested by recent INTEGRAL measurements (Churazov et al 2006), or the average value of the reflection parameter is high, R ∼2, or some combination of the two. Observations of individual sources seem to indicate that such a high value for the reflection component is unlikely.…”

Cosmic X-ray and gamma-ray background from dark matter annihilation

“…They all give approximately the correct relic density. Observations from soft X-rays to gamma-rays are marked with red to violet, following approximately a rainbow color pattern: red symbols [46], red arrows (Chandra, [33]), orange symbols (INTEGRAL, [47]), yellow symbols (SWIFT BAT, [32]), yellow arrows [48], green area (SMM, [49]), light blue (COMPTEL, [50]), blue (EGRET, [51]), violet (Fermi-LAT, [52]), violet arrows [34,35]. The points with error bars are absolute measurements with 2σ or 1σ errors.…”

“…We take the absolute measurements obtained using the latter two satellites according to the analysis of [46] (red symbols in Fig. 2), [47] (orange symbols) and [32] (yellow symbols). At intermediate energies, 300 keV ≤ E 0 ≤ 30 MeV, the measurements come from the Solar Maximum Mission (SMM) [49] and COMPTEL [50].…”

X-ray observations of highly obscuredτ9.7μm > 1 sources: an efficient method for selecting Compton-thick AGN?

“…The peak of the X-ray background at 20−30 keV (e.g. Frontera et al 2007;Churazov et al 2007;Moretti et al 2009) can be reproduced only by invoking a significant number of Compton-thick sources at moderate redshifts. However, the exact density of Comptonthick sources required by X-ray background synthesis models still remains an open issue (Gilli et al 2007;Sazonov et al 2008;Treister et al 2009a).…”