Glycolaldehyde and ethanol toward the L1157 outflow: resolved images and constraints on glycolaldehyde formation

TL;DR

This work uses high-resolution NOEMA maps to image glycolaldehyde and ethanol along the L1157 southern outflow, aiming to test whether GA forms predominantly in the gas phase via the ethanol-tree route or on grain mantles. A two-stage, pseudo time-dependent astrochemical model compares gas-phase formation from ethanol and grain-surface injection, with abundances constrained by three shocks of increasing age (B0, B1, B2). The results favor gas-phase formation from ethanol to explain GA in the younger regions but cannot simultaneously reproduce all three shocks with a single set of assumptions, effectively ruling out a grain-surface–dominated GA origin in this environment. The study underscores the need for time-dependent shock modeling and region-specific physical conditions to robustly constrain iCOM formation pathways in protostellar outflows. Overall, the findings contribute to discriminating iCOM formation routes in solar-like star-forming regions and motivate future multi-line, time-resolved observations and dynamic shock modeling.

Abstract

Two main formation routes have been proposed for interstellar complex organic molecules (iCOMs): on dust grain surfaces and in the gas phase. Observing such molecules in protostellar outflow shock regions - provided that their ages are well-constrained - can help distinguish between these pathways by probing chemical evolution over time. This study focuses on the potential daughter-mother relationship of glycolaldehyde (CH$_2$OHCHO) and ethanol (C$_2$H$_5$OH), previously proposed in the literature. We test whether gas-phase reactions converting ethanol into glycolaldehyde derived in these works can explain the observed abundance of the latter in star-forming regions. We target the southern outflow of L1157, which hosts three shock regions, B0, B1 and B2, of increasing ages: about 900, 1500 and 2300 yr. We obtained high-resolution IRAM NOEMA maps of three lines of glycolaldehyde and one line of ethanol. We derived their abundances in the three shocks and used a pseudo time-dependent astrochemical model to simulate gas-phase and grain-surface formation scenarios for glycolaldehyde. Ethanol is assumed to form on grains and be released in the gas by shocks, where it is gradually converted into glycolaldehyde via the ethanol-tree reaction network. We present the first spatially resolved maps of glycolaldehyde and ethanol in the L1157 southern outflow, and more generally toward solar-like star forming regions. The abundance ratio [CH$_2$OHCHO]/[C$_2$H$_5$OH] increases from B1 to B2, consistent with model predictions. However, the model cannot reproduce all three shocked regions simultaneously, suggesting that one of the assumptions of our model, such as the same excitation temperature and grain composition in B0, B1 and B2, or gas temperature evolution, is wrong. Nonetheless, our modeling rules out the possibility that all the observed gaseous glycolaldehyde is a grain-surface product.

Figures (8)

• Figure 1: The left panel shows a zoom of the GA and ethanol emission maps toward B0, B1 and B2 respectively, integrated over the velocity range $[-10,5]$ km s$^{-1}$. The velocity-integrated emission is shown by white and red contours for the GA and ethanol lines, respectively, overlapped to the emission of the CH$_3$OH line 5$_{1,5}$-4$_{0,4}$ in color scale. For the GA and ethanol lines, the first contours are at 3$\sigma$ and the following ones are plotted with a step of 1$\sigma$ (1$\sigma$ = 15 mJy/beam km/s for CH$_2$OHCHO and 12 mJy/beam km/s for C$_2$H$_5$OH in B0 and in B1; 18 mJy/beam km/s for CH$_2$OHCHO and 14 mJy/beam km/s for C$_2$H$_5$OH in B2). The ellipses in the corners of the maps show the synthesized beams for each GA and ethanol line. The white triangles and associated labels on all maps mark the different clumps, identified by Benedettini_2007. The right panel shows the velocity-integrated intensity map of the CH$_3$OH $5_{1,5}-4_{0,4}$ line, to show the morphology of the overall L1157 southern molecular outflow. The methanol map was used to define the polygons (shown by the magenta contours) where the intensity used for the column density computations is integrated (see text). The methanol map is centered at $\alpha(\rm J2000) = 20^{\rm h} 39^{\rm m}09.635^{\rm s}$ and $\delta(\rm J2000) = 68\textrm{°}01'19.80"$. The first contour is at 50$\sigma$ and the next contours are plotted with a step of 100$\sigma$ (1$\sigma$ = 11 mJy/beam km/s). The black ellipse in the bottom right corner shows the synthesized beam of the CH$_3$OH line imaged. The white squares centered on B0, B1 and B2 delimit the respective areas of the zoomed maps presented on the left panels. The white star at the top of map marks the position of the unresolved L1157 protobinary system Tobin_2013Tobin_2022.
• Figure 2: GA column density and abundance ratio, computed from the $10_{1,10}-9_{0, 9}$ line, the brightest one, assuming LTE population and optically thin line emission. The bands represent the uncertainty associated to the line intensity (see text). Upper panel: Column density of GA in B0 (light pink), B1 (pink) and B2 (dark pink), respectively, as a function of the excitation temperature $T_{\rm ex}$. Please note that the column density values partially overlap in B0 and B1. Lower panel: GA/ethanol abundance ratio versus $T_{\rm exc}$ in B0, B1 and B2 (same color coding as above), respectively.
• Figure 3: Results of the chemical modeling obtained for the Reference Model parameters (listed in Table \ref{['tab:chemical_modelling_parameters']}) against the observations. The curves shows the predicted abundances (with respect to H-nuclei) of GA (blue) and ethanol (red) (upper panel), and the GA over ethanol abundance ratios (lower panel) as a function of time. The symbols represent the values assuming $T_{\rm ex}$ equal to 15 (crosses), 30 (triangles) and 50 (squares) K, as measured at B0, B1 ans B2, respectively. The solid lines correspond to the Reference Model with no injection of GA from the dust grains, whereas the dashed lines correspond to the Reference Model with an injected GA abundance $X(\rm{CH}_2\rm{OHCHO})=1.5\times 10^{-8}$.
• Figure 4: Results obtained assuming that GA is exclusively formed in the gas-phase via the ethanol tree. Model predictions (curves) versus observations (symbols) toward B0, B1 and B2, assuming three different excitation temperatures $T_{\rm ex}$: 15 (crosses), 30 (triangles) and 50 (squares) K. In each figure, the upper panels show the abundances of GA (blue) and ethanol (red) as a function of time. The bottom panels show the evolution with time of the GA/Ethanol abundance ratio. Left panel: Predictions obtained assuming a temperature of 100 K and four values for the injected H$_2$O abundances: 1.0 (dotted), 2.5 (solid), 5.0 (dashed) and 10 (dotted-dashed) $\times10^{-5}$). Right panel: Predictions obtained assuming an injected H$_2$O abundance of $2.5\times10^{-5}$ and four values of temperature: 80 (dotted), 90 (dashed), 100 (solid) and 130 K (dotted-dashed).
• Figure 5: Results obtained assuming that GA is formed in the gas-phase by the ethanol tree and it is injected from the grain mantles into the gas-phase at the passage of the shocks. The injected ethanol is 3 (left panels) and 15 (right panels) $\times 10^{-8}$. Model predictions (curves) versus observations (symbols) toward B0, B1 and B2, assuming three different excitation temperatures $T_{\rm ex}$: 15 (crosses), 30 (triangles) and 50 (squares) K. In each figure, the upper panels show the abundances of GA (blue) and ethanol (red) as a function of time. The bottom panels show the evolution with time of the GA/Ethanol abundance ratio.