In Table 4 (click here) we list the soft X-ray (0.1-2.4 keV) fluxes and luminosities
and the far-infrared (
) fluxes and luminosities
for the secure identifications.
To compute
the soft X-ray (0.1-2.4 keV) energy flux
from the PSPC count rate we assume a simple power-law spectrum
where 
is the galaxy's energy flux between photon energies
E and 
. We assume a fixed photon spectral index 
,
which is the typical value found for extragalactic objects with ROSAT
(cf. Hasinger et al. 1991; Walter & Fink 1993),
and
an absorbing column density of hydrogen fixed at the
Stark et al. (1992) Galactic value 
along the line of sight.
In other words, the normalization of the spectrum from which the energy flux
is derived is chosen such that a power law photon spectrum with
, absorbed by a column , produces the
observed count rate. These fluxes are referred to in the following as
.
For 42 objects with more than 100 detected source photons we were able
to improve the flux estimates by fitting power-law spectra with
free spectral index 
, and free absorption column density, 
; latter was however required to be larger or equal to the Galactic
value at the respective position, .
Under the assumption that the intrinsic spectrum is a power law,
we can thereby account for absorption of X-rays
inside the respective galaxy.
The flux derived this
way, 
, should usually be higher than that derived with fixed
and , since
it is derived from a larger absorbing column.
However, in some cases 
. This occurs when our best
fit 
, which results in a lower energy flux for a
given count rate, and when at the same time
is not much larger than the Galactic value.
It may also be
due to differences between the count rates
determined in the standard analysis and the count rates derived from
the Photon Event files using the EXSAS package (Zimmermann et al. 1994).
The integrated energy fluxes and other
spectral parameters for the 42 X-ray brightest galaxies we were able
to fit this way are listed in Table 4.
In Fig. 3 we compare the fluxes obtained from a free fit, , with
those of a fixed spectral shape fit, . As expected, most objects
have 
.
We are aware that in objects where detailed spectral modeling was
performed, it often though not always appears that the spectrum can be fit
with a hard power-law plus a soft excess component. Even if the spectrum
is well represented by a power-law, the spectral index may assume a wide
range of values (e.g. Boller et al. 1996) find a correlation
between the soft X-ray photon index and the FWHM of the 
line).
In these cases the power-law approximation may result to uncertain
fluxes and luminosities.
To illustrate the robustness of our flux determination we have searched the literature for detections by other X-ray satellites or other authors on RASS-detected IRAS galaxies from this paper. From the secure identification listed in our Table 4 the following sources have published X-ray fluxes obtained with the Einstein satellite (Fabbiano et al. 1992): IRAS F02321-0900, F03207-3723 (as well as ROSAT RASS I observations (Brinkmann et al. 1994)), F03372-1850, F04150-5554, F11034+7250, F11210-0823, F12125+3328, F13277+5840, F14157+ 2522; and within ROSAT pointed observations (Brinkmann et al. 1994) F04305+0514 and F12265+0219. Figure 4 (click here) shows a good agreement between different flux measurements. We are aware that source variability might also contribute to the scatter in Fig. 4 (click here), since the majority of ROSAT sources show variability (cf. Voges & Boller 1997 for a statistical analysis of the variability of ROSAT sources).
Figure 4: Comparison of the X-ray fluxes obtained in this paper
with X-ray fluxes obtained by other authors. We searched the literature
and found 11 sources out of the 197 secure identifications listed in
Table 4 with published soft X-ray fluxes.
The open symbols mark detections by the Einstein
satellite and the filled symbols were obtained within ROSAT pointed
observations.
The fluxes obtained from different measurements are in good agreement
and the differences
may be due to either
intrinsic variability
of the sources (cf. Voges & Boller 1997 for an statistical analysis
of the variability of ROSAT sources) or slight differences in the
spectral modeling used to convert count rates to fluxes
The total far-infrared (
m)
fluxes, 
, were computed following Helou (1985)
from the IRAS 60 
m and 100 m
band fluxes:
where f60 and f100 are given in Jansky.
The soft X-ray and far-infrared fluxes were converted to luminosities using
Eq. (7) of Schmidt & Green (1986):
where a power-law spectrum is assumed in the energy range (E1,E2),
so that the redshift-dependent
functions C(z) and A(z) are then given by:
For the photon index in the far-infrared we assumed
.
A Hubble constant 
and
cosmological deceleration parameter of 
were adopted.
Figure 5: Soft (0.1-2.4 keV)
X-ray to far-infrared (40-120 m) flux ratio versus far-infrared flux for
ROSAT All-Sky Survey detected IRAS galaxies. Only objects with
confident correlations, i.e., grade 1 and 2, are shown. Galaxies identified
as Seyfert in NED are marked by filled symbols.
The X-ray flux was obtained from a simple power-law model with a
photon index and an absorbing column density of hydrogen
fixed to the Galactic value along the line of sight (cf. Sect. 3)
In Fig. 5 (click here) we plot the ratio between
the far-infrared and soft X-ray flux, 
against the far-infrared flux for the
objects in Table 4 (click here).
Galaxies identified as Seyfert in NED are marked as filled symbols.
The ratio ranges over 4 orders of magnitude and
IRAS galaxies
identified as Seyfert in the NED tend to be located at the high
end of this distribution.