Spectral monitoring of RX J1856.5-3754 with XMM-Newton. Analysis of EPIC-pn data

TL;DR

This study analyzed XMM-Newton data of the neutron star RX J1856.5-3754 over several years, finding minimal spectral and flux variability likely due to instrumental effects, and modeling its spectrum with two blackbodies.

Contribution

It provides a detailed spectral analysis of RX J1856.5-3754 over a long period, highlighting instrumental effects and establishing constraints on its variability and spectral features.

Findings

Minimal intrinsic spectral variability detected

Instrumental effects influence temperature measurements

Spectrum best fitted with two blackbody components

Abstract

Using a large set of XMM-Newton observations we searched for long term spectral and flux variability of the isolated neutron star RX J1856.5-3754 in the time interval from April 2002 to October 2011. This is the brightest and most extensively observed source of a small group of nearby, thermally emitting isolated neutron stars, of which at least one member (RX J0720.4-3125, Hohle et al., 2010) has shown long term variability. A detailed analysis of the data obtained with the EPIC-pn camera in the 0.15-1.2 keV energy range reveals small variations in the temperature derived with a single blackbody fit (of the order of 1% around kT^inf \sim 61 eV). Such variations are correlated with the position of the source on the detector and can be ascribed to an instrumental effect, most likely a spatial dependence of the channel to energy relation. For the sampled instrumental coordinates, we quantify this effect as variations of \sim 4% and \sim 15 eV in the gain slope and offset, respectively. Selecting only a homogeneous subset of observations, with the source imaged at the same detector position, we find no evidence for spectral or flux variations of RX J1856.5-3754 from March 2005 to present-day, with limits of Delta kT^inf < 0.5% and Delta f_X < 3% (0.15-1.2 keV), with 3sigma confidence. A slightly higher temperature (kT^inf \sim 61.5 eV, compared to kT^\inf \sim 61 eV) was instead measured in April 2002. If this difference is not of instrumental origin, it implies a rate of variation \sim -0.15 eV yr^-1 between April 2002 and March 2005. The high-statistics spectrum from the selected observations is best fitted with the sum of two blackbody models, with temperatures kT_h^inf = 62.4_{-0.4}^{+0.6} eV and kT_s^\inf = 38.9_{-2.9}^{+4.9} eV, which account for the flux seen in the optical band. No significant spectral features are detected, with upper limits of 6 eV on their equivalent width.