Collimated Fast Wind in the Pre-Planetary Nebula CRL 618

Collimated fast winds (CFWs) have been proposed to operate during the post-AGB evolutionary phase (and even earlier during the late AGB phase), responsible for the shaping of pre-planetary nebulae (PPNs) and young planetary nebulae (PNs). This paper is a follow-up to our previous study of CFW models for the well-studied PPN CRL 618. Previously, we compared our CFW models with optical observations of CRL 618 in atomic and ionic lines and found that a CFW with a small opening angle can readily reproduce the highly collimated shape of the northwestern (W1) lobe of CRL 618 and the bow-like structure seen at its tip. In this paper, we compare our CFW models with recent observations of CRL 618 in CO J=2-1, J=6-5, and H2 1-0 S(1). In our models, limb-brightened shell structures are seen in CO and H2 at low velocity arising from the shocked AGB wind in the shell, and can be identified as the low-velocity (LV) components in the observations. However, the shell structure in CO J=2-1 is significantly less extended than that seen in the observations. None of our models can properly reproduce the observed high-velocity (HV) molecular emission near the source along the body of the lobe. In order to reproduce the HV molecular emission in CRL 618, the CFW is required to have a different structure. One possible CFW structure is the cylindrical jet, with the fast wind material confined to a small cross section and collimated to the same direction along the outflow axis.

💡 Research Summary

This paper investigates the role of collimated fast winds (CFWs) in shaping the pre‑planetary nebula (PPN) CR 618, focusing on the highly collimated north‑western (W1) lobe. Building on earlier work that compared CFW models with optical atomic and ionic line images, the authors now confront the same models with recent millimeter and near‑infrared observations: CO J = 2‑1, CO J = 6‑5, and H₂ 1‑0 S(1). The hydrodynamic simulations adopt a two‑dimensional axisymmetric framework in which a fast wind (≈ 200 km s⁻¹, density ≈ 10⁴ cm⁻³, opening angle ≈ 10°) collides with a slow AGB wind (≈ 15 km s⁻¹, density ≈ 10⁶ cm⁻³). The interaction produces a thin, limb‑brightened shell that emits low‑velocity (LV) molecular lines. This shell reproduces the observed LV components in CO and H₂ quite well, confirming that the shocked AGB material dominates the low‑velocity emission.

However, the models fail in two critical respects. First, the simulated CO J = 2‑1 shell is considerably less extended than the observed CO structure, indicating that the model shell dissipates too quickly. Second, and more importantly, the high‑velocity (HV) molecular emission (50–150 km s⁻¹) detected near the central source and along the body of the W1 lobe is absent from the simulations. In the standard CFW configuration, the fast wind spreads over a relatively wide cone; after impact it decelerates sharply, leaving little high‑speed molecular material.

To resolve these discrepancies, the authors explore alternative wind geometries. The most promising configuration is a cylindrical jet: a fast, dense flow confined to a narrow cross‑section (radius ≈ 10 AU) that remains collimated along the outflow axis. In this scenario, the jet material suffers minimal lateral expansion and retains a high velocity over large distances. Simulations incorporating a cylindrical jet generate HV molecular emission that aligns with the observed CO and H₂ line profiles along the entire lobe, while still preserving the LV shell emission produced by the shocked AGB wind. Parameter studies reveal that the strength and spatial distribution of the HV component are highly sensitive to the jet radius and initial temperature; smaller radii and higher temperatures both enhance the HV CO signal.

The paper concludes that reproducing the full suite of CR 618 observations requires a wind structure more collimated than the previously assumed wide‑angle CFW. A cylindrical jet, possibly driven by binary interaction or magneto‑centrifugal processes, provides a physically plausible mechanism that simultaneously accounts for both LV shell emission and HV molecular outflows. The authors recommend future work employing three‑dimensional magnetohydrodynamic simulations coupled with high‑resolution ALMA and JWST observations to probe jet–AGB wind mixing, shock chemistry, and the detailed kinematics of the transition from PPN to planetary nebula. Such studies will deepen our understanding of how fast, collimated outflows sculpt the diverse morphologies observed in evolved stellar environments.