Supporting Material for GCN 17548 and ATel #7181

Swift J0045.2+4151: analysis of XMM-Newton archival data

A. Belfiore (INAF/IASF Milano), A. Tiengo (IUSS Pavia), G. L. Israel (INAF/OA Roma), A. M. Read (U. Leicester), R. Salvaterra, D. Salvetti (INAF/IASF Milano), G. Novara (IUSS Pavia), S. Mereghetti, M. Marelli (INAF/IASF Milano), G. Rodriguez (INAF/OA Roma), G. Lisini (IUSS Pavia), S. R. Rosen (U. Leicester), A. De Luca (INAF/IASF Milano), on behalf of the EXTraS collaboration.


Based on the preliminary results of the EXTraS project (DeLuca et al. 2015, arXiv:1503.01497), we report on the analysis of XMM-Newton archival data of 3XMM J004514.7+415035 (aka [PFH2005] 622, Pietsch et al. 2005, ApJ 434, 483), the candidate X-ray counterpart of GRB 150301C (aka Swift J0045.2+4151, GCN Circ 17512, 17516, 17536, 17544). The source is detected by the EPIC instrument at a 0.2-12 keV flux level of ~10-14 erg cm-2 s-1 in two observations (2002-01-26 and 2007-01-05), while only upper limits of ~5x10-14 erg cm-2 s-1 can be set from 5 shorter observations performed in July 2006. Shallow upper limits of ~10-12 erg cm-2 s-1 can be derived from three non-detections during slew observations on 2005-07-03, 2007-01-06, and 2010-01-31.

The EPIC light curves of the two observations show no evidence for variability. The search for coherent pulsations in the PN data, between 0.15 and 104 s, gives negative results. The blind search for transients does not report any significant candidates within the BAT error circle of GRB 150301C in any of the seven XMM-Newton observations.

We extracted the EPIC spectra of 3XMM J004514.7+415035 and, after verifying that there are no significant spectral differences between the two observations, we fit them simultaneously with an absorbed power-law model. The best-fit parameters (errors at the 90% confidence level) are NH<3x1021 cm-2 and photon index 1.9 (-0.6, +0.8). Fixing the NH to the Galactic value of 1021 cm-2, as appropriate for sources in the M31 disk, the photon index can be constrained in the 1.6-2.2 range (C-stat=142 for 143 degrees of freedom) and the unabsorbed flux in the 0.1-10 keV range is (1.3 +/- 0.3) 10-14 erg cm-2 s-1, corresponding to ~7x1035 erg s-1 at the M31 distance.

We note that the properties of 3XMM J004514.7+41503 are consistent with those of an active magnetar, like, e.g., SGR1806-20, at the distance of M31. Considering that the duration, fluence and spectrum of the burst detected by BAT are compatible with an intermediate flare from a magnetar in M31, 3XMM J004514.7+41503 might therefore be considered as a magnetar candidate in the Andromeda galaxy.

Fig. 1Long-term light curve for the MOS1 camera

Fig. 2Long-term light curve for the MOS2 camera

Fig. 3Long-term light curve for the PN camera. Upper limits of ~10-12 erg cm-2 s-1 from three non-detections during slew observations on 2005-07-03, 2007-01-06, and 2010-01-31 are not shown in the plot.

Fig. 4Short-term light curve combining data from the 3 EPIC cameras for the observation 0109270301

Fig. 5Short-term light curve combining data from the 3 EPIC cameras for the observation 0402561501

Fig. 6Power density spectrum for the PN camera with 3-sigma upper limits over the full period range 0.15-104 sec

Fig. 7Power density spectrum for the PN camera with 3-sigma upper limits over the restricted period range 1-20 sec which includes the spin period of all the known magnetars. The lack of coherent signals is not conclusive given that the greatest part of the known magnetars have spin period pulsed fractions smaller than the upper limit of 60% inferred from the XMM data.