2025ApJ...978...90A

The measured ages of massive, quiescent galaxies at z ∼ 3-4 imply that massive galaxies quench as early as z ∼ 6. While the number of spectroscopic confirmations of quiescent galaxies at z < 3 has increased over the years, there are only a handful at z > 3.5. We report spectroscopic redshifts of one secure (z = 3.757) and two tentative (z = 3.336 and z = 4.673) massive ($\mathrm{log}({M}_{* }/{M}_{\odot })\gt 10.3$) quiescent galaxies with 11 hr of Keck/MOSFIRE K-band observations. Our candidates were selected from the FLAMINGOS-2 Extragalactic Near-Infrared K-band Split (FENIKS) survey, which uses deep Gemini/Flamingos-2 K_bK_r imaging optimized for increased sensitivity to the characteristic red colors of galaxies at z > 3 with a strong Balmer/4000 Å break. The rest-frame UVJ and (ugi)_s colors of three out of four quiescent candidates are consistent with 1-2 Gyr old stellar populations. This places these galaxies as the oldest objects at these redshifts, and challenges the notion that quiescent galaxies at z > 3 are all recently quenched, post-starburst galaxies. Our spectroscopy shows that the other quiescent-galaxy candidate is a broad-line active galactic nucleus (z = 3.594) with strong, redshifted Hβ + [O III] emission with a velocity offset > 1000 km s^–1, indicative of a powerful outflow. The star formation history of our highest redshift candidate suggests that its progenitor was already in place by z ∼ 7-11, reaching ∼10¹¹M_⊙ by z ≃ 8. These observations reveal the limit of what is possible with deep near-infrared photometry and targeted spectroscopy from the ground and demonstrate that secure spectroscopic confirmation of quiescent galaxies at z > 4 is feasible only with JWST.