A Pilot Survey for CIII] Emission in the Reionization Era: Gravitationally-Lensed z$\sim7-8$ Galaxies in the Frontier Fields Cluster Abell 2744

TL;DR

This study conducted a spectroscopic search for CIII] emission in gravitationally-lensed galaxies at redshifts 7-8, setting upper limits on emission strength and highlighting the need for more luminous targets before JWST observations.

Contribution

First attempt to statistically constrain CIII] emission in z~7-8 galaxies using deep Keck/MOSFIRE spectroscopy in lensing clusters, providing upper limits on line strength.

Findings

No convincing CIII] emission detected in the sample.

Estimated 95% confidence upper limit on CIII] equivalent width is <26±5 Å.

Results are consistent with earlier detections in brighter galaxies at similar redshifts.

Abstract

We report results of a search for CIII] $\lambda \lambda$1907,1909 {\AA} emission using Keck's MOSFIRE spectrograph in a sample of 7 $z_{phot}\sim7-8$ candidates ($H\sim27$) lensed by the Hubble Frontier Field cluster Abell 2744. Earlier work has suggested the promise of using the CIII] doublet for redshift confirmation of galaxies in the reionization era given $Ly\alpha$ ($\lambda$1216 {\AA}) is likely attenuated by the neutral intergalactic medium. The primary challenge of this approach is the feasibility of locating CIII] emission without advanced knowledge of the spectroscopic redshift. With an integration time of 5 hours in the H-band, we reach a $5\sigma$ median flux limit (in between the skylines) of $1.5\times10^{-18}$ ergs cm$^{-2}$ sec$^{-1}$ but no convincing CIII] emission was found. We also incorporate preliminary measurements from two other CLASH/HFF clusters in which, similarly, no line was detected, but these were observed to lesser depth. Using the known distribution of OH emission and the photometric redshift likelihood distribution of each lensed candidate, we present statistical upper limits on the mean total CIII] rest-frame equivalent width for our $z\simeq7-8$ sample. For a signal/noise ratio of 5, we estimate the typical CIII] doublet rest-frame equivalent width is, with 95\% confidence, $<26\pm5$ {\AA}. Although consistent with the strength of earlier detections in brighter objects at $z\simeq6-7$, our study illustrates the necessity of studying more luminous or strongly-lensed examples prior to the launch of the James Webb Space Telescope.