Multiple X-ray Observations of the Transient Neutron Star Binary XTE J1946+274

ATel #3077; S. Müller (Remeis Observatory & ECAP), M. Kühnel (Remeis Observatory & ECAP), K. Pottschmidt (GSFC/UMBC), I. Caballero (CEA Saclay), F. Fürst (Remeis), L. Barragán (Remeis), M. Finger (NSSTC/USRA), A. Santangelo (IAA Tuebingen), C. Ferrigno (ISDC), I. Kreykenbohm (Remeis), and J. Wilms (Remeis)
on 11 Dec 2010; 18:00 UT
Credential Certification: Joern Wilms (j.wilms@sternwarte.uni-erlangen.de)

Tweet

We report on INTEGRAL, Swift, and RXTE observations of the 2010 November/December outburst of the Be/X-ray binary XTE J1946+274. This source was first detected in the year 1998 (ASM/RXTE: Smith & Takeshima 1998 and BATSE/CGRO: Wilson et al. 1998) during a giant outburst. In 2001 the source went into quiescence. A new sequence of outbursts started in 2010 June (ATEL #2663, #2677, #2692, #2847, #3048). The third of these outbursts is ongoing (ATEL #3048). Based on the RXTE/ASM count rate, the first outburst in this series was about two times brighter and the second comparable in brightness than the current outburst. A particularity of the system is that it goes into outburst two times every orbit (about 169 d; Wilson et al., 2003), which is still not understood. XTE J1946+274 has been detected in the INTEGRAL ISGRI field of view during Cyg X-1 observations. Swift and RXTE TOO observations were then triggered to follow the outburst. The INTEGRAL observations took place at the onset and around the peak of the outburst (MJD 55500.7, 55518.0, 55518.7, and 55530.6). The flux during the observations was around 10, 47, 56, and 40 mCrab (20-40 keV) respectively. The four Swift/XRT observations were performed during the maximum of the outburst (MJD 55526, 55528, 55530, 55532) with exposure times of 2.5, 3.4, 3.4, and 1.5 ksec, respectively. The source reached values in flux up to ~65 mCrab in the XRT detector. RXTE observed the outburst on MJD 55523, 55524, 55525, 55526, 55528 and 55530.

The source shows a double peak pulse profile in the soft 2-9 keV band. Pulse arrival time analysis without binary orbit correction reveals a pulse period of 15.75397(7) s, slightly shorter than that found in previous work (ATEL #2692), but consistent with measurements from Fermi/GBM (Finger et al.).

The 2-9 keV spectrum is well described by an absorbed power law with reduced chi^2 values between 0.9 and 1.6 for all observations. The fluxes for the four observations in the 2-9 keV energy band are $(4.7\pm 0.6)$, $(4.2\pm 0.6)$, $(3.8\pm 0.6)$, and $(4.8\pm 0.8)$E-10 erg s^{-1} cm^{-2}, respectively. No spectral evolution is seen, with the photon index remaining constant at $1.39\pm 0.07$. Furthermore there is no change in the absorbing column during the four observations, with all $N_H$ values consistent with $4.4\pm 0.4\times 10^{22}\,\text{cm}^{-2}$. There is no evidence for a 6.4keV iron line.

To constrain the overall continuum better, we fit near-simultaneous Swift/XRT (Nov 26), RXTE/PCA (Nov 25), and INTEGRAL/ISGRI (Nov 30) data together. As no spectral evolution in the Swift data was evident, this approach is justified. We find that a best fit with a reduced chi^2 of 1.6 is achieved when using an absorbed Fermi-Dirac cutoff powerlaw, with a photon index of $0.75^{+0.05}_{-0.07}$, a cutoff energy of $18^{+2}_{-4}$\,keV, and folding energy of $8.5^{+1.2}_{-0.8}$\,keV. The absorbing column is fitted to a low $(1.7\pm0.4)\times10^{22}\,\text{cm}^{-2}$. The cyclotron line detected in the 1998 outburst at about 35 keV (Heindl et al. 2001) is not required in the model. Contrary to earlier outbursts a "highecut" model does not describe the data adequately in terms of chi^2 (ATel #2692).