Modelling the non-equilibrium chemistry of the Milky Way's cold nuclear wind

TL;DR

This study addresses how cold atomic and molecular gas can persist in the Milky Way’s hot nuclear wind by testing time-dependent, non-equilibrium chemistry. Using the DESPOTIC framework with a zoned, 32-zone cloud model and five environmental irradiations, the authors compare equilibrium and non-equilibrium scenarios for two wind clouds (C1, C2). They find that chemical equilibrium cannot reproduce the observed combination of HI column density and CO luminosity without unrealistically small radii and excessive external pressures, while non-equilibrium, stripping-driven evolution can—predicting elevated CO-to-H2 conversion factors ($X_ ext{CO}$) and substantial molecular masses in the wind. The results imply that cold outflows can originate from disc molecular clouds that survive acceleration but lose their diffuse envelopes, leading to higher wind mass loading than previously estimated and underscoring the importance of non-equilibrium chemistry for interpreting multiphase galactic winds.

Abstract

Cold atomic and molecular gas are commonly observed in the winds of both external galaxies and the Milky Way, yet the survival and origin of these cool phases within hot galactic winds is poorly understood. To help gain insight into these problems, we carry out time-dependent chemical modelling of cool clouds in the Milky Way's nuclear wind, which possess unusual molecularto-atomic hydrogen ratios that are inconsistent with both disc values and predictions from chemical equilibrium models. We confirm that CO and Hi emission comparable to that in the observed nuclear wind clouds cannot be produced by gas in chemical equilibrium, but that such conditions can be produced in a molecule-dominated cloud that has had its atomic envelope rapidly removed and has not yet reached a new chemical equilibrium. Clouds in this state harbour large reservoirs of molecular gas and consequently have anomalously large CO-to-H2 conversion factors, suggesting that the masses of the observed clouds may be significantly larger than suggested by earlier analyses assuming disc-like conversions. These findings provide a new framework for interpreting cold gas in galactic winds, providing strong evidence that cold outflows can originate from the galactic disc molecular clouds that survive acceleration into the wind but lose their diffuse atomic envelopes in the process, and suggesting that the Milky Way's nuclear outflow may be more heavily mass-loaded than previously thought.

Figures (13)

• Figure 1: H i column density ($N_\mathrm{H~\textsc{i}}$; top panel) and integrated CO $2\to 1$ luminosity ($L_\mathrm{CO}$; bottom panel) as a function of mean hydrogen column density $\langle N_\mathrm{H} \rangle$ and volume density $\langle n_\mathrm{H} \rangle$ for clouds in chemical and thermal equilibrium. In both panels, the colour scale has been set so that white corresponds to the observed values of $N_{\mathrm{H\,\textsc{i}}}$ and $L_\mathrm{CO}$ for C1, with the red to blue range corresponding to a factor of two variation about this value. The dashed and solid black lines in the top and bottom panels, respectively, correspond to the loci where the model-predicted and observed values of $N_{\mathrm{H\,\textsc{i}}}$ and $L_\mathrm{CO}$ match. The loci of ($\langle N_\mathrm{H} \rangle$, $\langle n_\mathrm{H} \rangle$) for which the cloud radius lies within a factor of two of the observed radius of C1 is shown in the grey hatched region.
• Figure 2: The same as \ref{['fig:PS1']}, but now showing just the solid and dashed contour lines indicating the loci where $N_{\mathrm{H\,\textsc{i}}}$ (dashed) and $L_{\rm{CO}}$ (solid) match the observed values for C1. The red marker indicates where the contours intersect and thus a cloud can reproduce both the observed $N_{\mathrm{H\,\textsc{i}}}$ and $L_{\rm{CO}}$ values in equilibrium. The grey hatched region again indicates the locus where the cloud radius lies within a factor of two of the observed value.
• Figure 3: Abundances of select hydrogen-bearing (top row) and carbon-bearing (bottom row) chemical species as a function of cloud radius (left column) and total H column density measured from the cloud surface (right column). The case shown corresponds to $\log (\langle N_\mathrm{H} \rangle{}/\mathrm{cm}^{-2}) \approx 21.57$ and $\log (\langle n_\mathrm{H} \rangle{}/\mathrm{cm}^{-3}) \approx 2.85$, the red point shown in \ref{['fig:intersect']}. The abundances shown for the hydrogen-bearing species are normalised to the total abundance of H nuclei, while those for the carbon-bearing species are normalised to the total abundance of carbon nuclei.
• Figure 4: Same as in \ref{['fig:intersect']}, but with markers showing a selection of $\langle n_\mathrm{H} \rangle$$_i$ (circle markers) and $\langle n_\mathrm{H} \rangle$$_f$ (diamond markers) values. The $\langle n_\mathrm{H} \rangle$$_i$ values connected by the horizontal line correspond to the $\langle n_\mathrm{H} \rangle$$_f$ point at the same $\langle N_\mathrm{H} \rangle$. The exact coordinates of the points shown are given in \ref{['tab:models']}. The filled markers represent the models that successfully reproduce the observed $L_{\rm{CO}}$, $N_{\mathrm{H\,\textsc{i}}}$, and radii of (a) C1 and (b) C2, while empty markers show models that do not reproduce the observations. The grey hatched regions show loci where the cloud radii are within a factor of two of the observed radii, and the solid and dashed lines show the loci for which the equilibrium values of $L_{\rm{CO}}$ and $N_{\mathrm{H\,\textsc{i}}}$ match that of the observed clouds; these lines in (a) are identical to those shown for C1 in \ref{['fig:intersect']}.
• Figure 5: H i (top panel) and CO (bottom panel) normalised abundances as a function of cloud radius (left column) and total H column density to the cloud surface (right column) for the C1 model $\log(\langle n_\mathrm{H} \rangle_i/\mathrm{cm}^{-3}) = 0.85$, $\log(\langle n_\mathrm{H} \rangle_f/\mathrm{cm}^{-3}) = 1.85$ and $\log(\langle N_\mathrm{H} \rangle/\mathrm{cm}^{-2}) = 21.32$ (the most stripped model for row two in \ref{['tab:models']}); the abundances $x_\mathrm{H^0}$ and $x_\mathrm{CO}$ are the number of H$^0$ and CO per H nucleon, while $x_\mathrm{C}$ is the number of C nuclei per H nucleon. The dashed blue line shows the initial, equilibrium abundances of the cloud before stripping, and the solid purple line shows the non-equilibrium abundances immediately after the cloud has been stripped.