COMPLETE IONIZATION OF THE NEUTRAL GAS: WHY THERE ARE SO FEW DETECTIONS OF 21 cm HYDROGEN IN HIGH-REDSHIFT RADIO GALAXIES AND QUASARS

From the first published z greater than or similar to 3 survey of 21 cm absorption within the hosts of radio galaxies and quasars, Curran et al. found an apparent dearth of cool neutral gas at high redshift. From a detailed analysis of the photometry, each object is found to have a lambda = 1216 angstrom continuum luminosity in excess of L-1216 similar to 10(23) W Hz(-1), a critical value above which 21 cm has never been detected at any redshift. At these wavelengths, and below, hydrogen is excited above the ground state so that it cannot absorb in 21 cm. In order to apply the equation of photoionization equilibrium, we demonstrate that this critical value also applies to the ionizing (lambda <= 912 angstrom) radiation. We use this to show, for a variety of gas density distributions, that upon placing a quasar within a galaxy of gas, there is always an ultraviolet luminosity above which all of the large-scale atomic gas is ionized. While in this state, the hydrogen cannot be detected or engage in star formation. Applying the mean ionizing photon rate of all of the sources searched, we find, using canonical values for the gas density and recombination rate coefficient, that the observed critical luminosity gives a scale length (3 kpc) similar that of the neutral hydrogen (H (1)) in the Milky Way, a large spiral galaxy. Thus, this simple yet physically motivated model can explain the critical luminosity (L-912 similar to L-1216 similar to 10(23) W Hz(-1)), above which neutral gas is not detected. This indicates that the non-detection of 21 cm absorption is not due to the sensitivity limits of current radio telescopes, but rather that the lines of sight to the quasars, and probably the bulk of the host galaxies, are devoid of neutral gas.