Abstract

We present a detailed analysis of high-resolution Chandra observations of the merger system NGC 3256, the most infrared-luminous galaxy in the nearby universe. The X-ray data show that several discrete sources embedded in complex diffuse emission contribute ? 20 per cent of the total emission (LtotX ? 8 × 1041 erg s-1 in the 0.5-10 keV energy range). The compact sources are hard and extremely bright and their emission is probably dominated by accretion-driven processes. Both galaxy nuclei are detected with LX ? 3-10 × 1040 erg s-1. No evidence is found for the presence of an active nucleus in the southern nucleus, contrary to previous speculation. Once the discrete sources are removed, the diffuse component has a soft spectrum that can be modelled by the superposition of three thermal plasma components with temperatures kT = 0.6, 0.9 and 3.9 keV. Alternatively, the latter component can be described as a power law with index ? ? 3. Some evidence is found for a radial gradient of the amount of absorption and temperature of the diffuse component. We compare the X-ray emission with optical, H? and NICMOS images of NGC 3256 and find a good correlation between the inferred optical/near-infrared and X-ray extinctions. Although inverse Compton scattering could be important in explaining the hard X-rays seen in the compact sources associated with the nuclei, the observed diffuse emission is probably of thermal origin. The observed X-ray characteristics support a scenario in which the powerful X-ray emission is driven solely by the current episode of star formation.