paper

Iron Gradients in Cooling Flow Galaxies and Groups

arXiv:astro-ph/0001329 · doi:10.1086/309224

Abstract

Previous studies of the Fe abundances in the hot gas of galaxies and groups have reported conflicting results with most studies finding very sub-solar Fe abundances that disagree with standard theory. To investigate the possible role of Fe abundance gradients on these measurements we present deprojection analysis of the ROSAT PSPC data of 10 of the brightest cooling flow galaxies and groups. The PSPC allows for spatially resolved spectral analysis on a half-arcminute scale, and interesting constraints on both the temperatures and Fe abundances are possible because the ~1 keV temperatures of these systems are well matched to the bandpass of the PSPC. In 9 out of 10 systems we find clear evidence that the Fe abundance decreases with increasing radius: Z_Fe ~1 solar - (several) solar within the central radial bin (r < ~10 kpc) which decreases to Z_Fe ~0.5 solar at the largest radii examined (r ~50-100 kpc). The Fe abundances (and temperatures) are consistent with the average values for these systems that we obtained in our previous analyses of the ASCA data using multitemperature models which confirms that previous inferences of very sub-solar Fe abundances from ASCA arise from the incorrect assumption of isothermal gas and not the presence of Fe abundance gradients. We discuss why this ``Fe Bias'' affects much more seriously the measurements of Z_Fe from ASCA data than from ROSAT data. We show that the Fe abundance profiles for these galaxies and groups are consistent with a gas-dynamical model where the gas is enriched by stellar ejecta and supernovae in the ``solar supernova proportion'', the stars formed with a Galactic IMF, and the gas is diluted by mixing with primordial gas at large radii.

16 pages (7 figures), Accepted for publication in ApJ. Added a new section which discusses the relative importance of the Fe Bias and Fe gradients. Also included are results from a new gas-dynamical metal enrichment model

References in corpus (3)

Cited by in corpus (93)