Extragalactic Hard X-ray Surveys: From INTEGRAL to Simbol-X

We present some results of the deepest extragalactic survey performed by the INTEGRAL satellite. The fraction of very absorbed AGN is quite large. The sharp decrease in the absorption fraction with X-ray luminosity observed at lower-energy X-rays is not observed. The current lack of truly Compton-thick objects, with an upper limit of 14% to the size of this population, is just compatible with recent modeling of the cosmic X-ray background. We also study the prospects for a future hard X-ray serendipitous survey with Simbol-X. We show that Simbol-X will easily detect a large number of serendipitous AGN, allowing us to study the evolution of AGN up to redshifts about 2, opening the door to the cosmological study of hard X-ray selected AGN, which is barely possible with existing satellites like Swift and INTEGRAL.

💡 Research Summary

The paper presents two complementary studies: (1) the deepest extragalactic hard‑X‑ray survey performed to date with the INTEGRAL satellite, and (2) a forward‑looking assessment of the scientific reach of the upcoming Simbol‑X mission.

For the INTEGRAL work, the authors used the IBIS/ISGRI detector to observe a high‑latitude sky area of roughly 300 deg² with a total exposure of about 12 Ms, covering the 20–60 keV band. This effort yielded 34 hard‑X‑ray detections, 30 of which correspond to previously known active galactic nuclei (AGN) identified at optical or softer X‑ray energies. By comparing the 2–10 keV and 20–60 keV fluxes and fitting simple absorbed power‑law models, the hydrogen column densities (N_H) of each source were estimated. The key result is that roughly 60 % of the sample shows significant absorption (N_H > 10²² cm⁻²), a fraction that is markedly higher than the ≈30 % typically reported in soft‑X‑ray surveys.

A central focus of the analysis is the relationship between intrinsic X‑ray luminosity (L_X) and the absorption fraction. In the soft‑X‑ray regime a well‑known anti‑correlation exists: more luminous AGN tend to be less absorbed. In the INTEGRAL hard‑X‑ray sample this trend largely disappears. While low‑luminosity AGN (L_X < 10⁴³ erg s⁻¹) still show a high absorption rate (~70 %), even high‑luminosity objects (L_X > 10⁴⁴ erg s⁻¹) retain an absorption fraction of ≳40 %. This demonstrates that hard X‑rays can penetrate column densities that would render sources invisible at lower energies, thereby providing a less biased census of the absorbed AGN population.

The authors also address the elusive Compton‑thick (CT) AGN, defined by N_H > 1.5 × 10²⁴ cm⁻². No secure CT detections emerged from the INTEGRAL sample, allowing the authors to place an upper limit of 14 % on the CT fraction in the local hard‑X‑ray AGN population. This limit is compatible with recent cosmic X‑ray background (CXB) synthesis models, which require a CT contribution of roughly 10–20 % to reproduce the observed background spectrum. The lack of detected CT sources, however, underscores the current sensitivity limits of INTEGRAL and the need for deeper observations.

The second part of the paper projects the capabilities of Simbol‑X, a proposed focusing hard‑X‑ray telescope covering 0.5–80 keV with an angular resolution better than 30 arcsec and a flux limit near 10⁻¹⁴ erg cm⁻² s⁻¹ for a 1 Ms exposure. Simulations indicate that a single deep field observed with Simbol‑X would uncover roughly 1,200 AGN, of which about 30 % would lie at redshifts z > 1 and a non‑negligible number (20–30) would be bona‑fide CT objects. This represents an order‑of‑magnitude increase over the CT samples currently available from Swift/BAT or INTEGRAL.

With such a dataset, Simbol‑X would enable, for the first time, a statistically robust study of the evolution of hard‑X‑ray‑selected AGN out to z ≈ 2. Researchers could directly measure how the absorbed fraction evolves with cosmic time, test whether the luminosity‑dependent obscuration seen at soft energies persists at higher energies, and refine the contribution of CT AGN to the CXB. Moreover, the combination of Simbol‑X data with multi‑wavelength surveys (optical, infrared, radio) would allow detailed modeling of the torus geometry, cloud distribution, and feedback processes that regulate black‑hole growth.

In summary, the INTEGRAL survey confirms that hard X‑ray observations are essential for an unbiased view of the absorbed AGN population, revealing a high overall absorption fraction and a lack of a strong luminosity‑absorption anti‑correlation. Simbol‑X promises to extend this work dramatically, delivering large, deep, and high‑resolution samples of both Compton‑thin and Compton‑thick AGN across a broad redshift range, thereby opening a new era of cosmological hard‑X‑ray AGN studies that are currently out of reach for existing missions.