The near-infrared background: Interplanetary dust or primordial stars?

The intensity of the diffuse similar to 1-4 mu m sky emission from which solar system and Galactic foregrounds have been subtracted is in excess of that expected from energy released by galaxies and stars that formed during the z less than or similar to 5 redshift interval. The spectral signature of this excess near-infrared background light (NIRBL) component is almost identical to that of reflected sunlight from the interplanetary dust cloud and could therefore be the result of the incomplete subtraction of this foreground emission component from the diffuse sky maps. Alternatively, this emission component could be extragalactic. Its spectral signature is consistent with that of redshifted continuum and recombination line emission from H II regions formed by the first generation of very massive stars. In this paper we analyze the implications of this spectral component for the formation rate of these Population III stars, the redshift interval during which they formed, the reionization of the universe, and evolution of collapsed halo masses. Assuming that these Population III stars are massive objects radiating at the Eddington luminosity and ending their lives by directly collapsing into black holes, we find that to reproduce the intensity and spectral shape of the NIRBL requires a peak star formation rate of similar to 2.5M(circle dot) yr(-1) Mpc(-3), with a (1 + z)(-2) dependence on redshift, until the epoch ends at redshifts z approximate to 7-9. It requires a comoving luminosity density of about 2.7 x 10(11) L-circle dot Mpc(-3), corresponding to a total energy input of 670-820 keV per baryon, and that about 10% of the total number of baryons in the universe be converted to Population III stars. All these numbers are higher by about a factor of 4-10 than those derived from models in which Population III stars form at a rate that is proportional to the collapse rate of halos in a cold dark matter dominated universe. Furthermore, an extragalactic origin for the NIRBL leads to physically unrealistic absorption-corrected spectra of distant TeV blazars. All these results suggest that Population III stars contribute only a fraction of the NIRBL intensity, with zodiacal light, star-forming galaxies, and/or nonnuclear sources giving rise to the remaining fraction. Further 0.1-10 mu m observations of the diffuse sky and the zodiacal cloud are therefore crucial for resolving the true spectrum and origin of the NIRBL.