Probing the dark ages with metal absorption lines

Abstract

Recent observations of high-redshift quasars at z ∼ 6 have finally revealed complete Gunn-Peterson absorption. However, this at best constrains the volume-weighted and mass-weighted neutral fractions to be xVHI ≥ 10-3 and xMHI ≥ 10-2 respectively; stronger constraints are not possible because of the high optical depth for hydrogen Lyman transitions. Here I suggest certain metal lines as tracers of the hydrogen neutral fraction. These lines should cause unsaturated absorption when the intergalactic medium is almost fully neutral, if it is polluted to metallicities Z ∼ 10-3.5 - 10-2.5 Z⊙. Such a minimal level of metal pollution is inevitable in the middle to late stages of reionization unless quasars rather than stars are the dominant source of ionizing photons. The O I line at 1302 Å is particularly promising: the O I and H ionization potentials are almost identical, and O I should be in very tight charge exchange equilibrium with H. The Si II 1260 Å transition might also be observable. At high redshift, overdense regions are the first to be polluted to high metallicity but the last to remain permanently ionized, as a result of the short recombination times. Such regions should produce a fluctuating O I and Si II forest, which, if observed, would indicate large quantities of neutral hydrogen. The O I forest may already be detectable in the Sloan Digital Sky Survey z = 6.28 quasar. If seen in future high-redshift quasars, the O I and Si II forests will probe the topology of reionization and metal pollution in the early Universe. If, in addition, the H I optical depth can be measured from the damping wing of a high-redshift γ-ray burst, they will yield a very robust measure of the metallicity of the high-redshift Universe.