Long-term X-ray changes in the emission from the anomalous X-ray pulsar 4U 0142+61

We present results obtained from X-ray observations of the anomalous X-ray pulsar (AXP) 4U 0142+61 taken between 2000-2007 using XMM-Newton, Chandra and Swift. In observations taken before 2006, the pulse profile is observed to become more sinusoidal and the pulsed fraction increased with time. These results confirm those derived using the Rossi X-ray Timing Explorer and expand the observed evolution to energies below 2 keV. The XMM-Newton total flux in the 0.5-10 keV band is observed to be nearly constant in observations taken before 2006, while an increase of ~10% is seen afterwards and coincides with the burst activity detected from the source in 2006-2007. After these bursts, the evolution towards more sinusoidal pulse profiles ceased while the pulsed fraction showed a further increase. No evidence for large-scale, long-term changes in the emission as a result of the bursts is seen. The data also suggest a correlation between the flux and hardness of the spectrum, with brighter observations on average having a harder spectrum. As pointed out by other authors, we find that the standard blackbody plus power-law model does not provide the best spectral fit to the emission from 4U 0142+61. We also report on observations taken with the Gemini telescope after two bursts. These observations show source magnitudes consistent with previous measurements. Our results demonstrate the wide range of X-ray variability characteristics seen in AXPs and we discuss them in light of current emission models for these sources.

Years of RXTE Monitoring of Anomalous X-ray Pulsar 4U 0142+61: Long-Term Variability

We report on 10 years of monitoring of the 8.7-s Anomalous X-ray Pulsar 4U 0142+61 using the Rossi X-Ray Timing Explorer (RXTE). This pulsar exhibited stable rotation from 2000 March until 2006 February: the RMS phase residual for a spin-down model which includes nu, nudot, and nuddot is 2.3%. We report a possible phase-coherent timing solution valid over a 10-yr span extending back to March 1996. A glitch may have occured between 1998 and 2000, but is not required by the existing timing data. The pulse profile has been evolving since 2000. In particular, the dip of emission between its two peaks got shallower between 2002 and 2006, as if the profile were evolving back to its pre-2000 morphology, following an earlier event, which possibly also included the glitch suggested by the timing data. These profile variations are seen in the 2-4 keV band but not in 6-8 keV. We also detect a slow increase in the pulsed flux between 2002 May and 2004 December, such that it has risen by 36+/-3% over 2.6 years in the 2-10 keV band. The pulsed flux variability and the narrow-band pulse profile changes present interesting challenges to aspects of the magnetar model.