Abstract
We present the analysis of three intermediate-mass quiescent galaxies (QGs) with stellar masses of ${\sim} 10^{10}\, \mathrm{M}_{\odot }$ at redshifts $z\sim 3\!-\!4$ using NIRSpec low-resolution spectroscopy. Utilizing the spectral energy distribution fitting code BAGPIPES, we confirm these target galaxies are consistent with quiescent population, with their specific star formation rates falling below 2 dex the star-forming main sequence at the same redshifts. Additionally, we identify these QGs to be less massive than those discovered in previous works, particularly prior to the JWST era. Two of our target galaxies exhibit the potentially blended $\mathrm{ H} \, {\alpha }$ + [N II] emission line within their spectra with signal-to-noise ratio ${\gt} 5$. We discuss whether this feature comes from an active galactic nucleus (AGN) or star formation, although future high-resolution spectroscopy is required to reach a conclusion. One of the target galaxies is covered by JWST/NIRCam imaging of the Public Release IMaging for Extragalactic Research survey. Using the 2D profile fitting code GALFIT, we examine its morphology, revealing a disc-like profile with a Sérsic index of $n=1.1 \pm 0.1$. On the size-mass relation, we find a potential distinction between less massive ($\log _{10}{(M_*/\mathrm{M}_\odot)}\lt 10.3$) and massive ($\log _{10}{(M_*/\mathrm{M}_\odot)}\gt 10.3$) QGs in their evolutionary pathways. The derived quenching time-scales for our targets are less than $1 \, {\rm Gyr}$. This may result from these galaxies being quenched by AGN feedback, supporting the AGN scenario of the emission line features.