3 The catalogue

3.1 Starting samples

Our purpose is to have the most complete ensemble of spectral information (fluxes and spectral indices) in the X-ray band, from 0.1 to 10 keV, of all known blazars. We therefore considered all blazars detected in the X-ray band, for which also a measure of the X-ray spectral index is available. We collected the data obtained by five X-ray satellites: Einstein, EXOSAT, ROSAT, ASCA and BeppoSAX (see Table 2).

The first step was to recognize if a source belongs to the blazar class, and to which subclass (i.e. if a source is a FSRQ or an HBL or an LBL). We used several published lists of blazars and other publications describing single recognized sources. We considered the Slew Survey Sample (Elvis et al. 1992; Perlman et al. 1996), the 2 Jy sample of Wall & Peacock (1985), and the 1 Jy BL Lac sample (Stickel et al. 1991). In addition, we used the lists taken from the works of Bade et al. (1994); Bade et al. (1998); Brinkmann et al. (1994); Brinkmann et al. (1997); Cappi et al. (1997); Comastri et al. (1997); Ghisellini et al. (1993); Lamer et al. (1996); Laurent-Muehleisen et al. (1999); Sambruna et al. (1997); Wolter et al. (1998) and Worrall et al. (1990). We also checked the NASA Extragalactic Database (NED) for other objects classified as BL Lacs/blazars, or that could be classified as such.

The total number of considered blazars is 268. Of these, 227 have been observed by ROSAT and 88 have spectral information in the 2-10 keV band [of these latter sources, 77 have both soft (ROSAT) and hard X-ray data]. The details about the breakdown of source among HBL/LBL/FSRQ, and of the data among different X-ray telescopes is reported in Table 2. The data obtained with Einstein have large errors associated and for almost all sources better ROSAT data were available. For these reasons, the Einstein data are not used to derive any of the results (or figures) of this paper.

Some sources have been observed many times either by the same and/or by different satellites. For these sources, we chose the observation with the best $\chi_{\rm r}^2$ in the analysis.

For the most "famous" sources, like 3C 273, Mkn 421, Mkn 501, PKS 2155-304, we do not include the results of all the observations made by all satellites, but we have only listed few representative data (those with the best $\chi_{\rm r}^2$) for each of these sources (typically, one spectral datum for each observing satellite).

Of course, the resulting catalogue is not a complete sample. Nevertheless, it is the largest database of its kind, and we think it is representative of the entire blazar class. The large number of sources in each sub-category of blazars guarantees a meaningful comparison between their X-ray properties, and their relation with the fluxes in other bands.

 

 

Table 2: Number of observations obtained from various satellites and number of observed blazars (divided into different subclasses). For the total number of sources we have excluded multiple observations by different satellites of the same source.

	${\rm No.}\,$oss.	HBL	LBL	FSRQ
ASCA	52	14	9	24
EXOSAT	33	16	7	10
BeppoSAX	47	29	9	8
ROSAT	227	129	54	44
EINSTEIN	62	7	23	32
TOTAL	421	136	63	69

   
3.2 Format of the catalogue

Data are presented in Table 5 with the following format. For each source, Table 5 gives the IAU name, the redshift, the fluxes in the radio band (5 GHz), optical (V band) and X-ray (1keV) and the X-ray photon spectral index. In the last columns of Table 5 we also indicate to which subclass the blazar belongs to (1 for HBL, 2 for LBL and 3 for FSRQs) and the observing satellite (RO=ROSAT; AS=ASCA; EI=Einstein; EX=EXOSAT; SA=BeppoSAX).

For the radio fluxes we calculated the averaged value when there was more than one observation; the optical fluxes reported in the NED database are calculated using the indicated magnitude dereddened with the galactic extinction A_B as reported by the NED database. When in the literature we found only the 0.1-2.4 keV and/or the 2-10 keV integrated fluxes, we derived the monochromatic ones at 1 keV using the corresponding X-ray spectral index. All fluxes presented in Table 5 are not K-corrected.

To compute the luminosities, we used H₀= 50 km s^-1 Mpc^-1 and q₀=0.5, and for the K-correction we assumed a radio spectral index $\alpha=0$ for all sources; an optical spectral index $\alpha = 0.5$ for HBL and $\alpha =1$ for the rest of the sources; for the X-ray data we used the listed X-ray spectral index. Also the broad band spectral indices have been K-corrected.

The K-correction for sources with unknown redshift was computed using the average redshift appropriate for each sub-class (i.e. $\langle z \rangle_{\rm HBL}= 0.249$, $\langle z \rangle_{\rm LBL}= 0.457$ and $\langle z \rangle_{\rm FSRQ}= 1.265$).