Formation of the first stars and quasars

Abstract

We examine various observable signatures of the first generation of stars and low-luminosity quasars, including the metal enrichment, radiation background, and dust opacity/emission that they produce. We calculate the formation history of collapsed baryonic halos, based on an extension of the Press-Schechter formalism, incorporating the effects of pressure and H 2-dissociation. We then use the observed C/H ratio at z=3 in the Lyman-α forest clouds to obtain an average the star formation efficiency in these halos. Similarly, we fit the efficiency of black-hole formation, and the shape of quasar light curves, to match the observed quasar luminosity function (LF) between z=2−4, and use this fit to extrapolate the quasar LF to faint magnitudes and high redshifts. To be consistent with the lack of faint point-sources in the Hubble Deep Field, we impose a lower limit of ∼ 75 km s −1 for the circular velocities of halos harboring central black holes. We find that in a ΛCDM model, stars reionize the IGM at z reion=9−13, and quasars at z=12. Observationally, z reion can be measured by the forthcoming MAP and Planck Surveyor satellites, via the damping of CMB anisotropies by ∼10% on small angular scales due to electron scattering. We show that if reionization occurs later, at 5 ≲ z reion ≲ 10, then it can be measured from the spectra of individual sources. We also find that the Next Generation Space Telescope will be able to directly image about 1–40 star clusters, and a few faint quasars, from z >10 per square arcminute. The amount of dust produced by the first supernovae has an optical depth of τ=0.1−1 towards high redshift sources, and the reprocessed UV flux of stars and quasars distorts the cosmic microwave background radiation (CMB) by a Compton y-parameter comparable to the COBE limit, y ∼ 1.5 × 10 −5.