Cosmic X-ray Background Spectrum Research Articles

The shape of the cosmic X-ray background: nuclear starburst discs and the redshift evolution of AGN obscuration

A significant number of active galactic nuclei (AGNs) are observed to be hidden behind dust and gas. The distribution of material around AGNs plays an important role in modeling the cosmic X-ray background (CXB), especially the fraction of Type-2 AGNs ($f_2$). One of the possible explanations for the obscuration in Seyfert galaxies at intermediate redshift is dusty starburst discs. We compute the 2D hydrostatic structure of 768 nuclear starburst discs (NSDs) under various physical conditions and also the distribution of column density along the line of sight ($N_{\text{H}}$) associated with these discs. Then, the $N_{\text{H}}$ distribution is evolved with redshift by using the redshift dependent distribution function of input parameters. The $f_2$ shows a strong positive evolution up to $z=2$, but only a weak level of enhancement at higher $z$. The Compton-thin and Compton-thick AGN fractions associated with these starburst regions increase as $\propto (1+z)^{\delta}$ where the $\delta$ is estimated to be 1.12 and 1.45, respectively. The reflection parameter $R_f$ associated with column density $N_{\text{H}} \geq 10^{23.5}$ cm$^{-2}$ extends from 0.13 at $z=0$ to 0.58 at $z=4$. A CXB model employing this evolving $N_{\text{H}}$ distribution indicates more compact ($R_{\text{out}}<120$ pc) NSDs provide a better fit to the CXB. In addition to "Seyfert-like" AGNs obscured by nuclear starbursts, we predict that 40 to 60 per cent of quasars must be Compton-thick to produce the peak of the CXB spectrum within observational uncertainty. The predicted total number counts of AGNs in 8-24 keV band are in fair agreement with observations from NuSTAR.

Mildly obscured active galaxies and the cosmic X-ray background

The diffuse cosmic X-ray background (CXB) is the sum of the emission of discrete sources, mostly massive black-holes accreting matter in active galactic nuclei (AGN). The CXB spectrum differs from the integration of the spectra of individual sources, calling for a large population, undetected so far, of strongly obscured Compton thick AGN. Such objects are predicted by unified models, which attribute most of the AGN diversity to their inclination on the line of sight, and play an important role for the understanding of the growth of black holes in the early Universe. The fraction of obscured AGN at low redshift can be derived from the observed CXB spectrum assuming AGN spectral templates and luminosity functions. We show that high signal-to-noise average hard X-ray spectra, derived from more than a billion seconds of effective exposure time with the Swift/BAT instrument, imply that mildly obscured Compton thin AGN feature a strong reflection and contribute massively to the CXB. A population of Compton thick AGN larger than that effectively detected is not required, as no more than 6\% of the CXB flux can be attributed to them. The stronger reflection observed in mildly obscured AGN suggests that the covering fraction of the gas and dust surrounding their central engines is a key factor in shaping their appearance. These mildly obscured AGN are easier to study at high redshift than Compton thick sources.

A JOINT MODEL OF X-RAY AND INFRARED BACKGROUNDS. II. COMPTON-THICK ACTIVE GALACTIC NUCLEUS ABUNDANCE

We estimate the abundance of Compton-thick (CT) active galactic nuclei (AGN) based on our joint model of X-ray and infrared backgrounds. At L_{rest 2-10 keV} > 10^42 erg/s, the CT AGN density predicted by our model is a few 10^-4 Mpc^-3 from z=0 up to z=3. CT AGN with higher luminosity cuts (> 10^43, 10^44 & 10^45 erg/s) peak at higher z and show a rapid increase in the number density from z=0 to z~2-3. The CT to all AGN ratio appears to be low (2-5%) at f_{2-10keV} > 10^-15 erg/s/cm^2 but rises rapidly toward fainter flux levels. The CT AGN account for ~ 38% of the total accreted SMBH mass and contribute ~ 25% of the cosmic X-ray background spectrum at 20 keV. Our model predicts that the majority (90%) of luminous and bright CT AGN (L_{rest 2-10 keV} > 10^44 erg/s or f_{2-10keV} > 10^-15 erg/s/cm^2) have detectable hot dust 5-10 um emission which we associate with a dusty torus. The fraction drops for fainter objects, to around 30% at L_{rest 2-10 keV} > 10^42 erg/s or f_{2-10keV} > 10^-17 erg/s/cm^2. Our model confirms that heavily-obscured AGN (N_HI > 10^23 cm^-2) can be separated from unobscured and mildly-obscured ones (N_HI < 10^23 cm^-2) in the plane of observed-frame X-ray hardness vs. mid-IR/X-ray ratio.

INTEGRALhard X-ray spectra of the cosmic X-ray background and Galactic ridge emission

We derive the spectra of the cosmic X-ray background (CXB) and of the Galactic ridge X-ray emission (GRXE) in the ~20-200 keV range from the data of the IBIS instrument aboard the INTEGRAL satellite obtained during the four dedicated Earth-occultation observations of early 2006. We analyse the modulation of the IBIS/ISGRI detector counts induced by the passage of the Earth through the field of view of the instrument. Unlike previous studies, we do not fix the spectral shape of the various contributions, but model instead their spatial distribution and derive for each of them the expected modulation of the detector counts. The spectra of the diffuse emission components are obtained by fitting the normalizations of the model lightcurves to the observed modulation in different energy bins. The obtained CXB spectrum is consistent with the historic HEAO-1 results and falls slightly below the spectrum derived with Swift/BAT. A 10% higher normalization of the CXB cannot be completely excluded, but it would imply an unrealistically high albedo of the Earth. The derived spectrum of the GRXE confirms the presence of a minimum around 80 keV with improved statistics and yields an estimate of ~0.6 M_Sun for the average mass of white dwarfs in the Galaxy. The analysis also provides updated normalizations for the spectra of the Earth's albedo and the cosmic-ray induced atmospheric emission.

Cosmic X‐Ray Background and Earth Albedo Spectra withSwiftBAT

We use Swift/BAT Earth occultation data at different geomagnetic latitudes to derive a sensitive measurement of the Cosmic X-ray background (CXB) and of the Earth albedo emission in the 15–200keV band. We compare our CXB spectrum with recent (INTEGRAL, BeppoSAX) and past results (HEAO-1) and find good agreement. Using an independent measurement of the CXB spectrum we are able to confirm our results. This study shows that the BAT CXB spectrum has a normalization � 8 ± 3% larger than the HEAO-1 measurement. The BAT accurate Earth albedo spectrum can be used to predict the level of photon background for satellites in low Earth and mid inclination orbits.

A new measurement of the cosmic X-ray background

Aims. We present a new measurement of the cosmic X-ray background (CXRB) in the 1.5-7 keV energy band, performed by exploiting the Swift X-ray telescope (XRT) data archive. We also present a CXRB spectral model in a wider energy band (1.5-200 keV), obtained by combining these data with the recently published Swift-BAT measurement.Methods. From the XRT archive we collect a complete sample of 126 high Galactic latitude gamma-ray burst (GRB) follow-up observations. This provides a total exposure of 7.5 Ms and a sky-coverage of ~7 square degrees which represents a serendipitous survey, well suited for a direct measurement of the CXRB in the 1.5-10 keV interval. Our work is based on a complete characterization of the instrumental background and an accurate measurement of the stray-light contamination and vignetting calibration.Results. We find that the CXRB spectrum in the 1.5-7 keV energy band can be equally well fitted by a single power-law with photon index or a single power-law with photon index and an exponential roll-off at 41 keV. The measured flux in the 2-10 keV energy band is erg cm-2 s-1 deg-2 in the 2-10 keV band. Combining Swift-XRT with Swift-BAT (15-200 keV) we find that, in the 1.5-200 keV band, the CXRB spectrum can be well described by two smoothly-joined power laws with the energy break at keV corresponding to a peak located at keV.Conclusions. Taking advantage of both the Swift high energy instruments (XRT and BAT), we produce an analytical description of the CXRB spectrum over a wide (1.5-200 keV) energy band. This model is marginally consistent with the HEAO1 measurement (~10% higher) at energies higher than 20 keV, while it is significantly (30%) higher at low energies (2-10 keV).

Cumulative hard X-ray spectrum of local AGN: a link to the cosmic X-ray background

We determine the cumulative spectral energy distribution (SED) of local AGN in the 3-300 keV band and compare it with the spectrum of the cosmic X-ray background (CXB) in order to test the widely accepted paradigm that the CXB is a superposition of AGN and to place constraints on AGN evolution. We performed a stacking analysis of the hard X-ray spectra of AGN detected in two recent all-sky surveys, performed by the IBIS/ISGRI instrument aboard INTEGRAL and by the PCA instrument aboard RXTE, taking into account the space densities of AGN with different luminosities and absorption column densities. We derived the collective SED of local AGN in the 3-300 keV energy band. Those AGN with luminosities below 10^43.5 erg/s (17-60 keV) provide the main contribution to the local volume hard X-ray emissivity, at least 5 times more than more luminous objects. The cumulative spectrum exhibits (although with marginal significance) a cutoff at energies above 100-200 keV and is consistent with the CXB spectrum if AGN evolve over cosmic time in such a way that the SED of their collective high-energy emission has a constant shape and the relative fraction of obscured AGN remains nearly constant, while the AGN luminosity density undergoes strong evolution between z~1 and z=0, a scenario broadly consistent with results from recent deep X-ray surveys. The first direct comparison between the collective hard X-ray SED of local AGN and the CXB spectrum demonstrates that the popular concept of the CXB being a superposition of AGN is generally correct. By repeating this test using improved AGN statistics from current and future hard X-ray surveys, it should be possible to tighten the constraints on the cosmic history of black hole growth.

The 2–8 keV cosmic X-ray background spectrum as observed with XMM-Newton

We have measured the spectrum of the Cosmic X-ray Background (CXB) in the 2-8 keV range with the high throughput EPIC/MOS instrument onboard XMM-Newton. A large sample of high galactic latitude observations was used, covering a total solid angle of 5.5 square degrees. Our study is based on a very careful characterization and subtraction of the instrumental background, which is crucial for a robust measurement of the faintest diffuse source of the X-ray sky. The CXB spectrum is consistent with a power law having a photon index Gamma=1.41+/-0.06 and a normalization of 2.46+/-0.09 photons/(cm^2 s sr keV) at 3 keV (~11.6 photons/(cm^2 s sr keV) at 1 keV), corresponding to a 2-10 keV flux of (2.24+/-0.16)x10^(-11) erg/(cm^2 s deg) (90% confidence level, including the absolute flux calibration uncertainty). Our results are in excellent agreement with two of the most recent CXB measurements, performed with BeppoSAX LECS/MECS data (Vecchi et al. 1999) and with an independent analysis of XMM-Newton EPIC/MOS data (Lumb et al. 2002), providing a very strong constrain to the absolute sky surface brightness in this energy range, so far affected by a ~40% uncertainty. Our measurement immediately implies that the fraction of CXB resolved by the recent deep X-ray observations in the 2-10 keV band is of 80+/-7% (1sigma), suggesting the existence of a new population of faint sources, largely undetected within the current sensitivity limits of the deepest X-ray surveys.

The spectrum of the cosmic X-ray background observed by RTXE/PCA

We have analyzed a large set of Rossi X-ray Timing Explorer/Proportional Counter Array (RXTE/PCA) scanning and slewing observations performed between April 1996 and March 1999. We obtained the 3-20 keV spectrum of the cosmic X-ray background (CXB) by subtracting Earth-occulted observations from observations of the X-ray sky at high galactic latitude and far away from sources. The sky coverage is approximately 22.6 x 10(exp 3) square degrees. The PCA spectrum of CXB in 3-20 keV energy band is adequately approximated by a single power law with photon index GAMMA approximately 1.4 and normalization at 1 keV approximately 9.5 phot/s/square centimeter/keV/sr. Instrumental background uncertainty precludes accurate RXTE/PCA measurements of the spectrum of cosmic X-ray background at energies above 15 keV and therefore we cannot detect the high energy cutoff observed by the High Energy Astronomical Observatory (HEAO)-1 A2 experiment. Deep observations of the 6 high latitude points used to model the PCA background provide a coarse measure of the spatial variation of the CXB. The CXB variations are consistent with a fixed spectral shape and variable normalization characterized by a fractional rms amplitude of approximately 7% on angular scales of approximately 1 square deg.

A Composite Seyfert 2 X-Ray Spectrum: Implications for the Origin of the Cosmic X-Ray Background

We present a composite 1-10 keV Seyfert 2 X-ray spectrum, derived from ASCA observations of a distance-limited sample of nearby galaxies. All 29 observed objects were detected. Above ~3 keV, the composite spectrum is inverted, confirming that Seyfert 2 galaxies as a class have the spectral properties necessary to explain the flat shape of the cosmic X-ray background spectrum. Integrating the composite spectrum over redshift, we find that the total emission from Seyfert 2 galaxies, combined with the expected contribution from unabsorbed type 1 objects, provides an excellent match to the spectrum and intensity of the hard X-ray background. The principal uncertainty in this procedure is the cosmic evolution of the Seyfert 2 X-ray luminosity function. Separate composite spectra for objects in our sample with and without polarized broad optical emission lines are also presented.

Interpretation of spectral paradox of cosmic X-ray background

The integrated spectrum of discrete X-ray sources (mainly the active galactic nuclei, AGN) is inconsistent with the observed spectrum of cosmic X-ray background (CXB), and it is so called CXB spectral paradox. The medium X-ray spectra of 68 AGNs are adopted, the evolution function of X-ray spectral indices is analyzed statistically, the fraction of CXB is calculated due to AGNs X-ray emission, which shows that almost 100% CXB comes from AGNs X-ray emission. Especially, the integrated spectrum in 2-10 keV is consistent with the observed spectrum of CXB. The spectral paradox of CXB can be interpreted by this result.

Observations and theory of X-rays and gamma-rays from radio-quiet seyferts

Recent OSSE observations of soft γ-ray emission from radio-quiet Seyfert galaxies as well as their theoretical implications are reviewed. For some Seyferts, the OSSE data can be combined with X-ray data from Ginga and ROSAT. Seyfert 1s have intrinsic power-law spectra with α ∼ 0.9 extending without a break to at least ∼ 200 keV. There is some evidence for a break at higher energies. On top of this intrinsic spectrum there is a Compton reflection component, which corresponds to cold matter covering a ∼ 2π solid angle as seen by the power-law source. Compton reflection is responsible for a hardening of the spectrum in the ∼ 5–30 keV range and a softening at ∼ 30–200 keV. This spectrum is then absorbed by a partially ionized external medium. The spectra of Seyfert 2s are marginally consistent with those of Seyfert 1s modified by strong absorption. However, there are indications that their intrinsic X-ray spectra are harder than those of Seyfert 1s. This is in fact the case for NGC 4151, a Seyfert 1.5. The Seyfert spectra integrated over redshift with cosmological evolution can explain the cosmic X-ray background spectrum from 1 to 100 keV. The simplest model for the intrinsic Seyfert spectra is thermal Comptonization of UV photons. The scattering medium in Seyfert 1s is optically thin and relativistic whereas in NGC 4151 and possibly in Seyfert 2s it is optically thick and nonrelativistic. Nonthermal models with the power injected in electrons with low Lorentz factors are also possible.

Contributions of Starburst Galaxies and Reflection-dominated Active Galactic Nuclei to the Cosmic X-Ray Background

Starburst galaxies and low-luminosity reflection-dominated active galactic nuclei (AGNs) have been proposed as models of the cosmic X-ray background (CXB). We calculate the CXB spectrum to be expected from an evolving population of both massive and low-mass X-ray binaries in starburst galaxies, and find that star-bursts cannot produce much of the background at low energies (&lt;~20 keV) without significant evolution. We find their contribution at higher energies to be negligible, even with strong evolution. We also investigate the consistency of joint starburst and AGN models with the spectrum of the CXB between 3 and 500 keV, by means of detailed χ^2^ fits. If X-ray binaries in starburst galaxies do produce much of the low-energy CXB, some other type of source must be responsible for the high-energy background. If reflecting AGN are this other type of source, up to 89% of their emission must be reprocessed. Even without a starburst contribution, the AGN spectrum must still be about 86% reprocessed. However, for the simple cold disk reflection AGN models currently considered, the χ^2^ fit statistical significance is too low, indicating the need for either alternative models or new elements in the standard cold disk reflection model.

The diffuse cosmic X-ray background spectrum in the 2–10 keV energy range

We report a measurement of the background spectrum based on 10000 counts observed in the energy range 2–10 keV. The rocketborne detector system was optimised for cosmic ray noise rejection. A best fit power law spectrum $$\frac{{dN}}{{dE}} = 16E^{ - 1.8} photons{\text{ }}cm^{ - 2} s^{ - 1} sr^{ - 1} keV^{ - 1} .$$ resulted from the analysis. At 10 keV this result is consistent with recently assessed higher energy data. We show therefore that the change in spectral index between 5 and 50 ke V is approximately −0.2.