The effect of intergalactic helium on hydrogen reionisation: implications for the sources of ionising photons at z > 6

TL;DR

This study uses hydrodynamical simulations to explore how helium influences hydrogen reionisation and the thermal history of the IGM, suggesting stellar sources with soft EUV spectra drove reionisation at z>6.

Contribution

It provides new insights into helium's role in delaying reionisation and its impact on IGM temperature, constraining the nature of ionising sources at high redshift.

Findings

Helium slightly delays hydrogen reionisation.

Helium photo-heating increases IGM temperature by about 20%.

Pop-II stars with soft EUV spectra likely drove reionisation.

Abstract

We investigate the effect of helium on hydrogen reionisation using a hydrodynamical simulation combined with the cosmological radiative transfer code CRASH. The simulations are run in a 35.12/h comoving Mpc box using a variety of assumptions for the amplitude and power-law extreme-UV (EUV) spectral index, alpha, of the ionising emissivity. We use an empirically motivated prescription for ionising sources which ensures all of the models are consistent with constraints on the Thomson scattering optical depth and the hydrogen photo-ionisation rate at z=6. The inclusion of helium slightly delays reionisation due to the small number of ionising photons which reionise neutral helium instead of hydrogen. However, helium has a significant impact on the thermal state of the IGM. Models with alpha=3 produce IGM temperatures at the mean density at z=6 which are about 20 % higher compared to models without helium photo-heating. Harder EUV indices produce even larger IGM temperature boosts. A comparison to recent observational estimates of the IGM temperature at z=5 - 6 suggests that hydrogen reionisation was primarily driven by pop-II stellar sources with a soft EUV index, alpha<3. We also find that faint, as yet undetected galaxies, characterised by a luminosity function with a steepening faint-end slope and an increasing Lyman continuum escape fraction (fesc=0.5), are required to reproduce the ionising emissivity used in our simulations at z>6. Finally, we note there is some tension between recent observational constraints which indicate the IGM is > 10% neutral by volume z=7, and estimates of the ionising emissivity at z=6 which indicate only between 1 and 3 ionising photons are emitted per hydrogen atom over a Hubble time. This tension may be alleviated by either a lower neutral fraction at z=7 or an IGM which still remains a few % neutral by volume at z=6.