2 Data analysis and results

This burst was detected by only one lateral shield of the GRBM on-board BeppoSAX. The event shows a single peak structure lasting about 7 s (Coletta et al. 1997) and the peak intensity was $(2.6\pm0.3)$ photons cm^-2 s^-1 in the energy range 40-700 keV. The 256-channel GRBM data permits a sensitive spectral analysis with a 128 s accumulation interval. Fitting GRBM data with a single power law we found a photon index of $\Gamma=-2.58\pm0.33$. The fluence is $(6.6\,\pm\,0.7)\, 10^{-7}$ erg cm^-2 in the (40-700) keV band and $(9.5\,\pm\,0.9)\, 10^{-7}$ erg cm^-2 in the (20-300) keV. Assuming as a spectral model a broken power law with energy cut-off above 300 keV (BATSE results) we found $\Gamma_1=-2.30 \pm 0.43$, $E_{\rm cut-off}=155 \pm 1$ keV and $\Gamma_{2}\leq-9$ ($\chi_{r}^{2}$=1.3). The GRBM results are consistent with results from BATSE and show evidence of a quite soft $\gamma$-ray emission.

This GRB was also detected by the WFC unit 2, with a similar time profile structure and about the same duration. The 2-10 keV peak flux was $3.6\ 10^{-8}$ erg cm^-2 s^-1 (Coletta et al. 1997) a 2-10 keV fluence of 1.0 10^-7 erg cm^-2 and a mean flux, in the same energy band, of $1.4\ 10^{-8}$ erg cm^-2 s^-1. The position given by the quick-look analysis was R.A. = 12$^{\rm h}$ 57$^{\rm m}$ 29$^{\rm s}$, Dec. = +59$^\circ$16$.\mkern-4mu^\prime$3 (equinox 2000.0). The event was detected in a part of the orbit when the satellite was not in an optimal pointing configuration, resulting in an estimated error radius of 10 arcmin.

The 10' error box of GRB 971227 was imaged with the narrow field X-ray instruments (NFI) on-board BeppoSAX as a Target of Opportunity observation. The GRB follow-up observation started 51960 s after the trigger time, from 27 December 22:49 U.T. to 28 December 18:06 U.T., for a total net exposure time of 37000 s with the MECS and 16316 s with the LECS (the latter being operated only during satellite night-time).

Two faint X-ray sources were detected in the WFC error box by the BeppoSAX NFI (Piro et al. 1997). It is worth noting that other faint sources may be present in the error circle and they could be hidden by the detector strongback. The source closer to the center of the field, 1SAX J1257.5+5915, has a measured count rate of $(1.6\,\pm\,0.4)\ 10^{-3}$ cts s^-1 corresponding to a flux of $1.1\ 10^{-13}$ erg cm^-2 s^-1. All fluxes quoted in this paper are calculated assuming a spectrum described by an absorbed power law with the galactic value for the absorption and $\Gamma=2.1$. This source does not show a fading behaviour and may be a field source. We expect, from the BeppoSAX logN-logS (Giommi et al. 1998), to find in the 2-10 keV band about 20 sources per squaredegree at the measured flux: so, in the WFC error box, we expect to find 1.7 sources. The other source, 1SAX J1257.3+5924, showed, during the NFI observation, a fading behaviour making it the best candidate for the X-ray afterglow. We divided the total observation in two parts and the source counts varied from $(3.6\pm1.0)\ 10^{-3}$ cts s^-1 measured in the first 11555 s to $(2.93\ 10^{-3}$ cts s^-1 ($3\sigma$ U.L.) in the last 23784 s. The corresponding fluxes varied from 2.6 10^-13 erg cm^-2 s^-1 to $2.1\ 10^{-13}$ erg cm^-2 s^-1 ($3\sigma$ upper limit). These values together with the WFC value show a decaying law $F(t)\propto t^{-\alpha}$ with $\alpha=1.12_{-0.05}^{+0.08}$ (see figure). The refined position of this source is R.A. = 12$^{\rm h}$ 57$^{\rm m}$ 17.2$^{\rm s}$ and Dec. = $+59^\circ 24'02\hbox{$^\prime$}$ (equinox 2000.0) with an error circle of 1.5 arcmin due to the poor statistic.