Spatially resolved X-ray spectroscopy of cooling clusters of galaxies

¹ SRON National Institute for Space Research Sorbonnelaan 2, 3584 CA Utrecht, The Netherlands
² Department of Physics, Columbia University, 550 West 120th Street, New York, NY 10027, USA
³ Paul Scherrer Institute, Laboratory for Astrophysics, 5232 Villigen, Switzerland
⁴ Institute of Theoretical Physics, University of Zürich, Winterthurerstrasse, 190, 8057 Zürich, Switzerland
⁵ Mullard Space Science Laboratory, University College London, Holmbury St. Mary, Dorking, Surrey, RH5 6NT, UK
⁶ Max-Planck-Institut für extraterrestrische Physik, Giessenbachstrasse, 85748 Garching, Germany

Corresponding author: J.S. Kaastra, J.Kaastra@sron.nl

Received: 8 January 2003
Accepted: 22 September 2003

Abstract

We present spatially resolved X-ray spectra taken with the EPIC cameras of XMM-Newton of a sample of 17 cooling clusters and three non-cooling clusters for comparison. The deprojected spectra are analyzed with a multi-temperature model, independent of any a priori assumptions about the physics behind the cooling and heating mechanisms. All cooling clusters show a central decrement of the average temperature, most of them of a factor of . Three clusters (Sérsic 159-3, MKW 3s and Hydra A) only show a weak temperature decrement, while two others (A 399 and A 2052) have a very strong temperature decrement. All cooling clusters show a weak pressure gradient in the core. More important, at each radius within the cooling region the gas is not isothermal. The differential emission measure distribution shows a strong peak near the maximum (ambient) temperature, with a steep decline towards lower temperatures, approximately proportional to T³, or alternatively a cut-off at about a quarter to half of the maximum temperature. In general, we find a poor correlation between radio flux of the central galaxy and the temperature decrement of the cooling flow. This is interpreted as evidence that except for a few cases (like the Hydra A cluster) heating by a central AGN is not the most common cause of weak cooling flows. We investigate the role of heat conduction by electrons and find that the theoretically predicted conductivity rates are not high enough to balance radiation losses. The differential emission measure distribution has remarkable similarities with the predictions from coronal magnetic loop models. Also the physical processes involved (radiative cooling, thermal conduction along the loops, gravity) are similar for clusters loops and coronal loops. If coronal loop models apply to clusters, we find that a few hundred loops per scale height should be present. The typical loop sizes deduced from the observed emission measure distribution are consistent with the characteristic magnetic field sizes deduced from Faraday rotation measurements.