X-ray and radio observations of the magnetar Swift J1834.9-0846 and its dust-scattering halo

We present a long-term study of the 2011 outburst of the magnetar Swift J1834.9-0846 carried out using new Chandra observations, as well as all the available Swift, RXTE and XMM-Newton data. The last observation was performed on 2011 November 12, about 100 d after the onset of the bursting activity that had led to the discovery of the source on 2011 August 7. This long time-span enabled us to refine the rotational ephemeris and observe a downturn in the decay of the X-ray flux. Assuming a broken power law for the long-term light curve, the break was at \u2dc46 d after the outburst onset, when the decay index changed from \u3b1 \u2dc 0.4 to \u2dc4.5. The flux decreased by a factor of \u2dc2 in the first \u2dc50 d and then by a factor of \u2dc40 until 2011 November (overall, by a factor of \u2dc70 in \u2dc100 d). At the same time, the spectrum, which was well described by an absorbed blackbody all along the outburst, softened, the temperature dropping from \u2dc1 to \u2dc0.6 keV. Diffuse X-ray emission extending up to 20 arcsec from the source was clearly detected in all Chandra observations. Its spatial and spectral properties, as well as its time evolution, are consistent with a dust-scattering halo due to a single cloud located at a distance of 48200 pc from Swift J1834.9-0846, which should be in turn located at a distance of \u2dc5 kpc. Considering the time delay of the scattered photons, the same dust cloud might also be responsible for the more extended emission detected in XMM-Newton data taken in 2011 September. We searched for the radio signature of Swift J1834.9-0846 at radio frequencies using the Green Bank Radio Telescope and in archival data collected at Parkes from 1998 to 2003. No evidence for radio emission was found, down to a flux density of 0.05 mJy (at 2 GHz) during the outburst and \u2dc0.2-0.3 mJy (at 1.4 GHz) in the older data