The 28th February 1997 has been a date that changed the
thirty-years old history of the GRB astronomy.
On 02:58:00.8 Universal Time (UT)
a moderately intense, multi-peak GRB triggered
the GRBM on-board BeppoSAX .
The same GRB was also detected in one of
the two WFCs and therefore localized at about 10 arcmin precision,
later reduced to 3 arcmin. The peak intensity of the GRB in
the GRBM band (40-700 keV) was 
,while it was 
in the WFC band
(2-26 keV).
Figure 5.3 shows
the light curve of this GRB as seen from the GRBM and WFC instruments.
The NFI were pointed to the GRB location in the sky
in a time as short as 8 hours after the trigger detection.
The previously unknown X-ray source
1SAXJ0501.7+1146
was detected in the field of view of the LECS and MECS instruments
at the celestial coordinates:
Right Ascension (R.A.)=05h01m44s and
Declination (Decl.)=+11
46'.4 (equinox 2000.0).
This first observation lasted about 14.000 s of net observing time,
and the source mean flux in the 2-10 keV energy range
was 
(that is about 10.000 times weaker
than the reference celestial source for X-ray astronomy, the Crab Nebula,
in the same band).
What was particularly intriguing in this detection is that the new source appeared to be fading away during the observation. So, we found in a small region of the sky that few hours earlier hosted a GRB a previously undetected source with a time behavior indicating it was disappearing. We therefore decided to point again the NFI to the source, and this happened on March 3, 17:37 UT, for an exposure time of 16.000 s. Figure 5.4 shows the comparison between the results of the two observations and the discovery of the first 'afterglow'.
The evaluation of the temporal decay law of the
X-ray flux detected from the counterpart
of GRB970228 was initially performed by using the two NFI observations only.
The mean flux of the source appeared to decrease following a power
law dependency on time (
) with index
. A further X-ray observation with the Japanese
X-ray satellite ASCA detected again the source about one week later
with a flux consistent with the same law ([Yoshida et al. 1997]).
This kind of temporal
behavior agrees with the general predictions of the fireball models
for GRBs ([Wijers et al. 1997]).
If one observes the GRBM and WFC light curves of the GRB970228 in figure 5.3, it appears rather clearly that the second train of pulses of the GRB are much more soft than the first pulse, that is, the ratio between X-rays and gamma-rays is higher. We therefore decided to compare this second part of the GRB flux with its afterglow. In figure 5.5 the result is shown: the mean X-ray flux of the second part of the GRB (indicated as WFC (35-70 s) in the figure) is perfectly consistent with the backward extrapolation of the power law decay derived from a fit to the NFI data only. In the same figure is also shown the average X-ray flux of the entire GRB (indicated as WFC (0-100 s) in the figure) and it appears less clearly correlated to the above law. Our guess from this analysis is therefore that the afterglow to GRB970228 actually started soon after the GRB. A further confirmation of this comes from the spectral analysis that shows a spectral continuity of the GRBM/WFC spectra with the NFI only for the second set of pulses.
Another important result of the spectral analysis of the GRB970228 afterglow is that it seems to exclude a thermal origin of the emission, therefore suggesting that a model in which the radiation comes from the cooling of the surface of a neutron star cannot work in this case.