skip to content
ASCA Guest Observer Facility

ASCA science highlights

Diffuse X-ray background

Since its discovery in 1962 (Giacconi et al. 1962 Phys Rev Lett 9, 439) the origin of the diffuse X-ray background (DXRB) has been one of the main unsolved problems in X-ray astronomy. Its very high relative isotropy and uniformity combined with the perfect fit to a 40 keV bremmstrahlung spectrum over the 2-40 keV band (Marshall et al. 1980 ApJ 235, 4) gave impetus to models of its origin in a hot intergalactic medium. The sharp rise in the spectrum at E < 1 keV (McCammon et al. 1983 ApJ 269, 107) was attributed to a warm galactic component. However, there was no one experiment that observed this spectral transition and there were indications (Wu et al. 1991 ApJ 379, 564) of another spectral component. The COBE upper limits on the Compton y parameter for the microwave background (Mather et al. 1990 ApJ 354, L37) ruled out a hot IGM origin for the E > 2 keV background. Thus point source models of its origin are now considered to be most likely. These models require a large number (> 200 /sq degree) of cosmologically distant ( ~ 1), faint (F(x) < 10^-13 ergs/cm^2/s) sources and indeed the ROSAT deep survey results (Hasinger et al. 1993 A&Ap 275, 1) have found such a population. However, the X-ray spectra of the faint ROSAT sources are in general too steep to fit the spectrum of the E > 1 keV background. In fact, the absence of any observed population of sources with the observed spectrum of the background has been a source of continuing confusion. The solution, reached at the same time by several authors is that the flat observed x-ray background spectrum is due to the superposition of the spectra of sources with a wide range in column density, ranging up to log N(H) ~ 24 over a wide range in redshifts with power law spectral slopes consistent with the HEAO-1/EXOSAT/Ginga values for active galaxies (Zdziarski et al. 1993 ApJ 414, L81; Setti & Woltjer 1989 A& Ap 224, L21; Matt & Fabian 1994 MNRAS 267, 187; Madau et al. 1993 ApJ 410, L7).

Another of the controversies about the X-ray background, is the apparent rather different normalizations of the soft (0.2-2 keV) and hard (2-10 keV) source counts (the logN-logS laws). It has been known ever since the HEAO-1 and Einstein epochs (Piccinotti et al. 1982 ApJ 253, 485; Gioia et al. 1984 ApJ 283, 495) that the normalization of the hard X-ray logN-logS law is 3 times that of the soft band. However, before ASCA the hard X-ray logN-logS relation was only derived from source counts at F(x) > 10^{-11} ergs/cm^2/s and the direct comparison to the Einstein and ROSAT data relied on a fluctuation analysis from HEAO-1 and Ginga. If the fluctuation analysis were correct then this discrepancy could be explained by the presence of absorbed active galaxies, more of which would be visible in the hard band than in the soft band.

However, without reliable measures of the source counts at low (F(x) < 10^-11 ergs/cm^2/s) fluxes and without a detailed spectral form of the background over the 0.4-10 keV band these explanations were just ``best guesses''. ASCA data has enabled these tests to be made and resulted in a strong indication that the long standing mystery of the origin of the diffuse X-ray background is close at hand.

ASCA spectra of the DXRB (Gendreau et al. 1995 PASJ 47, 5; Gendreau 1995 PhD thesis; Ishisaki 1996 Waseda meeting) have shown that the 0.4-10 keV spectrum can be well modeled as the sum of two components; a hard power law (G = 1.42 +/- 0.02) or hot bremmstrahlung component that dominates at E > 1 keV and a softer component that can be well modeled as a kT ~ 0.16 keV subsolar abundance thermal plasma. The interpretation is that the soft component is the sum of a galactic thermal contribution and a steep spectrum line-less extragalactic component, while the hard emission is very similar to that reported previously by non-imaging proportional counters. The high quality E > 1 keV ASCA spectrum is devoid of spectral features of amplitude > 5% and thus places very strong constraints on any model of the X-ray background (Gendreau 1995 PhD thesis).

Deep surveys by ASCA (Inoue 1996 Waseda meeting) indicate that the hard X-ray logN-logS law continues as a S^-1.5 power law whose normalization is roughly 3 times that of the soft (0.2-2 keV) log N-logS down to flux levels of 10^-13 ergs/cm^2/s . At this level more than 25% of the hard X-ray background has been resolved out into sources. Combined ROSAT and ASCA spectra of a deep field (Chen, Fabian and Gendreau 1996 MNRAS submitted) have shown that the spectra of faint quasars are too steep at E > 1 keV to account for the observed background flux indicating that another population makes up the bulk of the E > 1 keV background. Analysis of the faintest resolved sources, F(X) < 2.5 x 10^-13 ergs/cm^2/s (Ueda 1996, PhD Thesis) indicates that they have a very flat spectrum with a photon index of 1.5 +/- 0.2, consistent with the spectrum of the diffuse background itself.

While optical identification work on the faint hard X-ray sources is still in progress several objects have been found (Makishima et al. 1994 PASJ 46, L77; Ohta et al. 1996 ApJ 458, L57) whose spectra are consistent with recent unified models (Madau et al. 1994 MNRAS 270, L17, Zdziarski et al. 1995 ApJ 438, L63) of the origin of the X-ray background.

ASCA science highlights

Last modified: Tuesday, 26-Jun-2001 14:22:33 EDT

If you have any questions concerning ASCA, visit our Feedback form.

This file was last modified on Tuesday, 26-Jun-2001 14:22:33 EDT

NASA Astrophysics

  • FAQ/Comments/Feedback
  • Education Resources
  • Download Adobe Acrobat
  • A service of the Astrophysics Science Division (ASD) at NASA/ GSFC

    ASCA Project Scientist: Dr. Nicholas E. White

    Responsible NASA Official: Phil Newman

    Privacy Policy and Important Notices.