Data Centre for Astrophysics
Astronomy Department of the University of Geneva

INTEGRAL and RXTE observations of XTE J1946+274 in outburst

The Be/X-ray binary XTE J1946+274 was caught undergoing a new X-ray outburst by Swift (BAT) and Fermi (GBM) on 2010 June 4 (ATel #2663, Atel #2677).

This is the second bright event recorded from the source since its discovery in 1998 (Smith & Takeshima 1998, Wilson et al. 1998). In that occasion, XTE J1946+274 entered in outburst and remained active in X-rays for almost 3 years.

The currently on-going outburst of XTE J1946+274 was observed also by INTEGRAL on 2010 July 20 and 21, when the satellite was pointing in the direction of Cyg X-1 (PI J. Wilms).

The source was in the FOV of both the IBIS/ISGRI and JEM-X2 telescopes starting from MJD 55367.5, and was clearly detected up to 50 keV. The total exposure time for the two instruments was ~16.4 ks and ~5.7 ks, respectively. The combined spectrum ISGRI+JEM-X2 could be well described (χ2red=1.01/36 dof) by using a cutoff-pl model; this gave Γ=1.3±0.2, Ecutoff=22±2, and Efold=10±1 (90% c.l.). The estimated flux was 1.1 × 10-9 erg/cm2/s (20-40 keV). We could not find any evidence for the cyclotron line at 35 keV detected during the previous outburst (Heindl et al. 2001, ApJ 563, L35).

A preliminary timing analysis of the ISGRI data revealed pulsations with a spin frequency of 63.426[1] mHz (not corrected for orbital motion). No spin-up/down is detected. This pulse frequency is slightly higher than the value obtained with Fermi GBM (63.4155(2) mHz, ATel #2677). The ISGRI pulse profile presents a main narrow peak, a broad shoulder and a sharp minimum, consistent with the peak pulse profile observed in 1998 (Wilson et al. 2003).

We also performed a preliminary analysis of the two RXTE observations carried out in the direction of the source on 2010 June 20. We used the PCA background model SkyVLE (derived from the predicted SAA file) and modeled the HEXTE-A background by using the HEXTE-B off-source spectrum. In both observations, the 3-50 keV spectrum could be well described by using an absorbed powerlaw with a high energy cutoff. Some some residuals were evident in the PCA data around 35 keV, but we verified that the inclusion of an absorption line did not provide a statistically significant improvement in the fit.

From the spectrum of the first observation, carried out on MJD 55367.12 (PCA exposure time ~0.82 ks), we obtained NH=5.0+0.7-0.8 × 1022 cm-2, Γ=1.25±0.04, Ecutoff=15.8±0.5 keV, and Efold=13.7±0.7 keV (90% c.l., χ2red=1.42/102 dof). The estimated 20-40 keV flux was 1.15×10-9 erg/cm2/s.

From the spectrum of the second observation (MJD 55367.18, PCA exp. ~0.53 ks), we found NH=5.4±0.5 × 1022 cm-2, Γ=1.24±0.03, Ecutoff=16.1+0.4-0.3, Efold=11.8±0.6 (90% c.l.), and measured an X-ray flux of 1.10×10-9 erg/cm2/s (χ2red=1.27/102 dof).