Localized Deviations from the CO-PAH Relation in PHANGS-JWST Galaxies: Faint PAH Emission or Elevated CO Emissivity?

TL;DR

This work probes localized breakdowns of the CO–PAH relation in PHANGS–JWST galaxies by leveraging high-resolution JWST PAH maps, ALMA CO data, and MUSE optical data. It identifies 20 of 70 galaxies with regions where CO is over-luminous relative to 7.7μm PAH emission by more than an order of magnitude, predominantly along bar lanes and centers with little massive star formation. Through analyses across multiple PAH and dust bands and by examining CO kinematics and isotopologue data, the study disfavors a universal PAH suppression and instead finds strong evidence that elevated CO emissivity—driven by bar-induced shear and shocks and associated higher $Δv_{ m CO}$—is the primary cause of the deviations, with sub-kpc variations in $\\alpha_{ m CO}$ substantiated by velocity-dispersion–dependent relations and isotopologue indicators. The results have important implications for using PAH emission as a proxy for the cold ISM and for interpreting CO-based molecular gas masses in dynamically complex environments; they also motivate deeper CO isotopologue measurements and JWST integral-field spectroscopy to further pin down the physical conditions in these outlier regions.

Abstract

Polycyclic aromatic hydrocarbon (PAH) emission is widely used to trace the distribution of molecular gas in the interstellar medium (ISM), exhibiting a tight correlation with CO(2-1) emission across nearby galaxies. Using PHANGS-JWST and PHANGS-ALMA data, we identify localized regions where this correlation fails, with CO flux exceeding that predicted from 7.7$μ$m PAH emission by more than an order of magnitude. These outlier regions are found in 20 out of 70 galaxies and are mostly located in galaxy centers and bars, without signs of massive star formation. We explore two scenarios to explain the elevated CO-to-PAH ratios, which can either be due to suppressed PAH emission or enhanced CO emissivity. We examine PAH emission in other bands (3.3$μ$m and 11.3$μ$m) and the dust continuum dominated bands (10$μ$m and 21$μ$m), finding consistently high CO-to-PAH (or CO-to-dust continuum) emission ratios, suggesting that 7.7$μ$m PAH emission is not particularly suppressed. In some outlier regions, PAH size distributions and spectral energy distribution of the illuminating radiation differ slightly compared to nearby control regions with normal CO-to-PAH ratios, though without a consistent trend. We find that the outlier regions generally show higher CO velocity dispersions ($Δv_{\mathrm{CO}}$). This increase in $Δv_{\mathrm{CO}}$ may lower CO optical depth and raise its emissivity for a given gas mass. Our results favor a scenario where shear along the bar lanes and shocks at the tips of the bar elevate CO emissivity, leading to the breakdown of the CO-PAH correlation.

Figures (12)

• Figure 1: Three examples of galaxies (NGC 1566, NGC 4303, and NGC 1097) hosting regions with abnormally high CO-to-PAH ratios. The first column shows the ratio of CO flux ($I_{\rm CO}$) to that expected from 7.7$\mu$m PAH emission ($I_{\mathrm{CO}}^{\mathrm{7.7\mu m}}$), using the relations of chown24. We apply the same intensity cut as chown24, including only pixels likely dominated by molecular gas with $I_{7.7\,\mu m}>0.5\mathrm{MJy\,sr^{-1}}$, and excluded regions are shown in gray. Outlier regions, defined as having a flux ratio ($I_{\mathrm{CO}}/I_{\mathrm{CO}}^{\mathrm{7.7\mu m}}$) higher than 10 (see text), are highlighted in red solid line. Subsequent columns show the 7.7$\mu$m PAH, CO and H$\alpha$ observations. For NGC 1566 and NGC 4303, the H$\alpha$ maps are from PHANGS–MUSE data corrected for extinction belfiore22, while for NGC 1097 the H$\alpha$ map is from narrowband imaging that has not been corrected for extinction (Razza et al. in prep.).
• Figure 2: The upper panels show the ratio of CO flux to that expected from other PAH- and dust continuum-dominated bands ($I_{\mathrm{CO}}/I_{\mathrm{CO}}^{\mathrm{JWST}}$), based on correlation coefficients reported in chown24 and measured in this work (for 10$\mu$m and 21$\mu$m). Red contours highlight outlier regions identified using the 7.7$\mu$m band. The lower panel shows the relation between $I_{\mathrm{CO}}/I_{\mathrm{CO}}^{7.7\mu\mathrm{m}}$ and the corresponding ratios derived from other JWST bands. Data points with abnormal $I_{\mathrm{CO}}/I_{\mathrm{CO}}^{7.7\mu\mathrm{m}}$ values are shown in red, whereas pixels with normal ratios are shown in gray.
• Figure 3: Comparison of PAH properties between abnormal and null-test regions in JWST Cycle 1 galaxies that host outlier regions. The first panel shows the median and 1$\sigma$ distribution of $R_{\mathrm{PAH}}$ measured in outlier and null-test regions. The second and third panels show $I_{\mathrm{11.3\mu m}}/I_{\mathrm{3.3\mu m}}$, which is sensitivity to the PAH size distribution and the hardness of the radiation field, and $I_{\mathrm{11.3\mu m}}/I_{\mathrm{7.7\mu m}}$, which trace the radiation field hardness and PAH charge states. The gray solid line indicates the one-to-one relation. As abnormal regions exist both in center and disk in NGC 1566, two environments are shown separately.
• Figure 4: Top left: CO luminosity-weighted velocity dispersion ($\langle\Delta v_{\rm CO}\rangle$) is higher in abnormal regions compared to their null-test regions. $\langle\Delta v_{\rm CO}\rangle$ versus the median CO-to-PAH ratio, $\langle I_{\rm CO}/I_{\rm CO}^{7.7\,\mu{\rm m}}\rangle$, for each abnormal (red) and null-test (gray) region across the 20 galaxies hosting CO/PAH outliers. The outermost contour encloses 68% of the points (solid line; 1$\sigma$). The dotted black line shows the $\langle\Delta v_{\rm CO}\rangle$ required to explain the outliers solely via enhanced CO emissivity at fixed H$_2$/PAH. Top right: Distributions of $I_{\rm CO}/I_{\rm CO}^{7.7,\mu{\rm m}}$ for abnormal (red) and null-test (gray) regions. Middle left: As in the upper panel, $\alpha_{\rm CO}$ (from the teng24 prescription) is shown as a function of $\langle I_{\rm CO}/I_{\rm CO}^{7.7,\mu{\rm m}}\rangle$. Middle right: Distributions of $\Sigma_{\rm H_2}/I_{\rm CO}^{7.7,\mu{\rm m}}$, where $\Sigma_{\rm H_2}=I_{\rm CO}\,\alpha_{\rm CO}$. Bottom:$\Delta v_{\mathrm{CO}}$ as a function of molecular gas surface density ($\Sigma_{\mathrm{H_{2}}}$) for NGC 1566, NGC 2566, and NGC 6300. Data points are color-coded by the observed-to-expected CO flux ratio ($I_{\mathrm{CO}}/I_{\mathrm{CO}}^{7.7,\mu{\rm m}}$), where points associated with outliers are outlined in red. We show the median $\Delta v_{\mathrm{CO}}$ in logarithmically spaced bins of $\log \Sigma_{\mathrm{H_{2}}}$ together with 1$\sigma$ scatter for each abnormal and null test region (red and gray squares, respectively). While $\Sigma_{\mathrm{H_{2}}}$ is estimated using the $\alpha_{\mathrm{CO}}$–$\Delta v_{\mathrm{CO}}$ relation from teng24, we also show the median of $\Delta v_{\mathrm{CO}}$ as a function of $\Sigma_{\mathrm{H_{2}}}$ when a constant galactic $\alpha_{\mathrm{CO}}=4.35 \mathrm{M_{\odot}pc^{-2}/(K~km~s^{-1})}$ is adopted instead (triangles).
• Figure 5: CO line ratio maps of NGC 3351 center from teng22, overlaid with contour levels of $I_{\mathrm{CO}}/I_{\mathrm{PAH}}^{7.7\,\mu\mathrm{m}}$ at 5 and 10. Left:$^{13}$CO($J{=}3{-}2$)/($J{=}2{-}1$), primarily sensitive to excitation temperature. Right:$^{12}$CO/$^{13}$CO($J{=}2{-}1$), primarily sensitive to isotopologue abundance and/or optical depth. Regions with enhanced $I_{\mathrm{CO}}/I_{\mathrm{PAH}}^{7.7\,\mu\mathrm{m}}$ coincide with lower $^{13}$CO($3{-}2$)/($2{-}1$) and higher $^{12}$CO/$^{13}$CO($2{-}1$), indicating that the elevated CO-to-PAH ratio is not driven by higher excitation temperature. Instead, it is consistent with reduced optical depth, which allows more $^{12}$CO emission to escape and increases $^{12}$CO/$^{13}$CO($2{-}1$).