Blowing star formation away in AGN Hosts (BAH) -- IV: Feeding and feedback in 3C 293 observed with JWST NIRSpec

TL;DR

3C 293 is analyzed as an AGN-host affected by a recent merger. The study uses JWST/NIRSpec IFU to map stellar kinematics and multiphase gas (hot molecular H2 and ionized Fe II Paα) within the inner ~2 kpc, resolving a rotating stellar disk and three gas kinematic components (disk, broad outflow, very broad outflow); the molecular gas shows inflows along dust lanes. Disk modeling yields a systemic velocity $V_{sys}=13412\pm15$ km s$^{-1}$, a line of nodes $ψ_0=46°\pm7°$, and an inclination $θ=53°\pm6°$, with the gas disks exhibiting higher maximum rotation speeds than the stars. The ionized outflow reaches up to $\dot{M} \sim 4.9$ M$_{\odot}$ yr$^{-1}$ with $\dot{K}\sim 9.6\times10^{41}$ erg s$^{-1}$, while the hot molecular outflow is $\dot{M} \sim 0.08$ M$_{\odot}$ yr$^{-1}$ and $\dot{K} \sim 1.5\times10^{40}$ erg s$^{-1}$. Jet-ISM coupling efficiencies are a few percent, indicating jet-driven feedback is energetically capable of quenching star formation, though a small hot molecular inflow (~$4\times10^{-4}$ M$_{\odot}$ yr$^{-1}$) suggests limited fueling from this phase; extinction is extreme ($A_V$ up to ~35), underscoring JWST's ability to probe dusty nuclear regions and fueling channels in AGN hosts.

Abstract

We use JWST/NIRSpec observations of the radio galaxy 3C 293 to map the emission, extinction, and kinematics of hot molecular and ionized gas, as well as stellar kinematics, within the inner ~ 2 kpc. The stellar velocity field is well described by a rotating disk model, with its kinematical center offset by ~ 0.5 arcsec from the continuum peak. The hot molecular gas is traced by the H$_2$2.12$μ$m emission line, and the ionized gas by [Fe II]1.64$μ$m and Pa$α$. The gas presents three main kinematic components: a rotating disk seen as a narrow component ($σ$ ~ 100 kms$^{-1}$); a blueshifted broad outflow ($σ$ ~ 250 kms$^{-1}$); and a fast ionized outflow as a very broad component ($σ$ ~ 640 kms$^{-1}$). Extinction maps reveal high A$_V$ values, up to ~ 35, spatially coincident with dust lanes seen in optical images. In addition to the disk and outflows components, inflows along the dust lanes are detected in H$_2$ gas, with a mass inflow rate of $\dot{M}_{in}$ ~ 4 x 10$^{-4}$ M$_{\odot}$ yr$^{-1}$, which is lower than the AGN accretion rate. For the outflows, we derive peak mass-outflow rates of 0.08 $\pm$ 0.02 M$_{\odot}$yr$^{-1}$ (molecular) and 6.5 $\pm$ 1.7 M$_{\odot}$yr$^{-1}$ (ionized). The outflow, driven by the radio jet, has a kinetic power of 5.7% of the jet power - enough to suppress star formation. Our results highlight 3C 293's turbulent post-merger history and JWST's unique capability to probe dust-obscured AGN.

Figures (11)

• Figure 1: Top left: Composite image of 3C 293 (UGC 8782) from Pan-STARSS data archive Chambers2019Flewelling2020 in the y (9633 Å), i (7545 Å), and g (4866 Å) bands. The black rectangle marks the JWST/NIRSpec FoV, while the gray rectangle indicates the FoV of the Hubble Space Telescope (HST) near-ultraviolet (NUV) Cs2Te Multianode Microchannel Array (MAMA) detector. Bottom left: STIS NUV MAMA image of 3C 293 Allen2002, with the JWST/NIRSpec FoV indicated by a black rectangle. The red circle marks the IR nuclear position, and the red square marks an extranuclear position. Top right: Nuclear spectrum of 3C 293, obtained at the position of the continuum peak, corresponding to the red circle in the images. Bottom right: Extranuclear spectrum of 3C 293, observed with JWST/NIRSpec, at the location of the red square in the images. The orange lines highlight the ionized emission lines, while the teal lines indicate the molecular ones. The position of the CO absorption bandheads is also identified.
• Figure 2: Examples of fits of the [Fe ii]$\,1.64\:\mu$m (top), Pa$\alpha$ (middle) and H$_2\,2.12\,\mu$m (bottom) emission-line profiles for the nuclear spaxel. The observed profiles are shown as thin gray lines and the best-fit models are shown in black. The narrow component is represented by dashed orange lines, the broad component by dashed dark red lines, and the very broad component by a dashed teal line.
• Figure 3: The top panel displays the stellar velocity field (in units of $\rm km~s^{-1}$) corrected for systemic velocity, while the bottom panel shows the stellar velocity dispersion (in units of $\rm km~s^{-1}$), corrected for instrumental broadening. The white partial ring indicates a region of lower velocity dispersion, attributed to an intermediate-age stellar population. All panels are oriented with north up and east to the left. The cross indicates the location of the NIR nucleus, which corresponds to the peak of the continuum emission. Regions with uncertainties larger than 20 km s$^{-1}$ are masked.
• Figure 4: The top row shows the flux distribution (in units of $\rm erg\, s^{-1} cm^{-2}$, logarithmic scale), velocity (in units of $\rm km ~s^{-1}$), and velocity dispersion (in units of $\rm km ~s^{-1}$) for the narrow component of the $\rm H_2\, 2.12 \,\mu m$ emission line. The second row displays the same maps for the broad component. The measurements are restricted to areas with at least $\rm 3\sigma$ detections above the continuum noise. The cross marks the nucleus, corresponding to the position of the continuum emission peak.
• Figure 5: Same as Fig. \ref{['fig:H2']}, but for the [Fe ii]$\rm 1.64 \mu m$ emission line. In addition, the third row shows the corresponding maps for the very broad component.