Multiphase Gaseous Halo In Regulating Galaxy Formation
Multiphase gas regulates galaxy formation by supplying the fuel for star formation and enabling the growth of galactic disks. Dust and metals produced by stars are carried through these gas phases, influencing cooling, star formation efficiency, and chemical enrichment. In this way, multiphase gas links star formation, disk assembly, and feedback into a global cycle that drives galaxy evolution.
Disturbed Cold Gas at High Redshift
We studied the first complete sample of 66 C I absorbers at 1.5 < z < 3.0 selected from SDSS and re observed with VLT X Shooter and UVES. These systems are metal and dust rich (Zou+18), often contain molecular gas such as CO and H2 (Noterdaeme+18), and show disturbed kinematics. The detection of Na I and Ca II further suggests a strong influence from star formation feedback (Zou+18).​
​
Cool gas (T ~ 10^4 K) fuels stars, early disk and structure formation
The Mg II doublet traces cool gas (T ≈ 10^4 K) associated with both inflows and outflows around galaxies. Using 114 quasars in the COSMOS field observed during DESI commissioning, we identified 51 Mg II and 49 C IV absorbers at 1.5 < z < 3.1 and found that the Mg II covering fraction evolves strongly in main sequence star forming galaxies, indicating close coevolution between cool gas reservoirs and galaxy growth at 1.5 < z < 2.5 (Zou+21; Zou+24a). At higher redshift, we selected 32 strong Mg II systems (Wr > 0.8 Å) from 50 quasars at z > 5.7 in the Gemini NIR sample (Shen+12) and showed that the comoving line density of strong Mg II absorbers declines rapidly at z > 3, while their velocity widths are significantly broader than those of DLA associated Mg II systems at similar redshifts. Follow up observations with Magellan, JWST, and ALMA indicate that these absorbers are linked to intense star formation activity or interacting galaxy environments (Zou+24b; Zou+25).
​​
Possible First Star Signature from Pristine Gas and Primeval Galaxies
The pristine gas and primeval galaxies offers insights into the formation of the first generation stars by comparing their observed metal abundance. We have detected a Lyman limit system with a notably high carbon enhancement relative to iron ([C/Fe] ~ +2.2 at [Fe/H] ~ -1.6). This is reminiscent of a distinctive sub-class of the first generation of stars, known as the carbon-enhanced metal-poor (CEMP) stars. More interestingly, the strong C I absorption is associated with a neutral hydrogen column density of log N(H I) (/cm^2) = 18 — a value too small to shield the gas from any external UV flux. We also see such carbon and oxygen enhancement in the absorbing systems at the end of reionization epoch (z = 6.0-6.5). If the metals are from the nearby galaxies detected by JWST, the galaxies favor a top-heavy IMF and possibly, the PopIII IMF.
​
Dense gas and dust across different black hole growth stages
Dense circumnuclear gas and dust play a key role in AGN obscuration and black hole growth at high redshift. We report two z ∼ 3 broad line X-ray AGNs (cid 414 and cid 947) from JWST COSMOS 3D NIRCam F444W grism spectroscopy showing prominent He I λ10830 absorption and emission. Both sources are detected in the mid infrared with MIRI, indicating hot dust. Photoionization modeling places the dense absorbing gas on broad line region scales and likely coupled to torus obscuration. Together with similar He I absorption seen in compact little red dots (LRDs), these results indicate that dense circumnuclear gas is common at high redshift and regulates AGN obscuration and black hole host coevolution.​​​
​
Galaxy Cluster L-T relation
We present a study using Chandra observations of 23 galaxy groups and low-mass galaxies within the redshift range 0.03 < z <0.15. Our sample represents a statistically complete, flux-limited subset from the 400 deg2 cluster survey. In this research, I examined the scaling relation between X-ray luminosity (L) and temperature (T), taking the selection biases into consideration. The L-T relation slope aligns with the values typically observed for samples of more massive clusters, may suggest the feedback processes.
​
