High deuteration of methanol in L1544

TL;DR

The paper addresses how methanol becomes heavily deuterated in a prototypical pre-stellar core, linking ice-surface chemistry to inheritance in star formation. It combines IRAM 30 m observations of CH$_2$DOH and CHD$_2$OH toward L1544 with state-of-the-art gas-grain chemical models, showing deuteration levels around $N$(CH$_2$DOH)/$N$(CH$_3$OH) ≈ $0.21$–$0.22$ and $N$(CHD$_2$OH)/$N$(CH$_3$OH) ≈ $0.03$–$0.06$ at the peaks. Among four models, the D5 version that includes H-abstraction reactions best reproduces the observed $R_D$ and $R_{D_2}$ and remains consistent across L1544, L694-2, and HMM-1 within a factor of two, highlighting the importance of surface abstraction chemistry; non-LTE excitation effects are crucial for reliable column densities. The results emphasize the role of core structure and physical conditions in setting methanol deuteration, and they motivate future work to obtain collisional rates for deuterated methanol and to extend similar analyses to related COMs. Overall, the study advances our understanding of molecular complexity buildup and inheritance in the early stages of star formation.

Abstract

Isotopic fractionation is a very powerful tool to follow the evolution of material from one stage to the next in the star-formation process. Pre-stellar cores exhibit some of the highest levels of deuteration because their physical conditions greatly favor deuteration processes. Deuteration maps are a measure of the effectiveness of the deuteration across the core, and they are useful to study both the deuteration as well as the formation mechanism of the main species. Methanol is the simplest O-bearing complex organic molecule (COM) detected in the interstellar medium (ISM). It represents the beginning of molecular complexity in star-forming regions, thus a complete understanding of its formation and deuteration is a necessary step to understand the development of further chemical complexity. In this paper, we use single-dish observations with the IRAM 30 m telescope and state-of-the-art chemical models to investigate the deuteration of methanol towards the prototypical pre-stellar core L1544. We also compare the results of the chemical models with previous observations of deuterated methanol towards the pre-stellar cores HMM1 and L694-2. The spectra extracted from the CHD$_2$OH map show that the emission is concentrated in the center and towards the north-west of the core. Using deep observations towards the dust and the methanol peaks of the core, we derive a very large deuterium fraction for methanol ($\sim20\%$) towards both peaks. The comparison of our observational results with chemical models has highlighted the importance of H-abstraction processes in the formation and deuteration of methanol. Deep observations combined with state-of-the-art chemical models are of fundamental importance in understanding the development of molecular complexity in the ISM. Our analysis also shows the importance of non-LTE effects when measuring the D/H ratios in methanol.

Figures (9)

• Figure 1: Left panel: the five different areas where the CHD$_2$OH spectra have been extracted from the OTF map observed with the IRAM 30 m telescope are shown as dotted squares on the H$_2$ column density map of L1544 computed from $Herschel$/SPIRE data at 250, 350 and 500 $\mu$mspezzano16. The solid white contours are the 30$\%$, 60$\%$ and 90$\%$ of the peak intensity of the N(H$_2$) map. The $Herschel$/SPIRE beam is shown in the bottom left of the map. The black saltire shows the position of the methanol peak and the full black circle shows the dust emission peak. Right panel: $J$$_{K_a,K_c}$ = 2$_{0,2}$-1$_{0,1}$$e_{0}$ CHD$_2$OH spectra extracted from the IRAM 30m OTF map. The vertical dashed lines shows the v$_{LSR}$ of the source (7.2 km/s), and the horizontal dotted lines show the 3$\sigma$ noise level. The number in each spectra refers to the area where the spectra was extracted from, shown in the left panel of the figure.
• Figure 2: Spectra of CHD$_2$OH observed with single-pointing observations toward the dust peak (upper panel) and the methanol peak (lower panel) of L1544. The vertical dashed line shows the v$_\mathrm{{LSR}}$ of the core, 7.2 km/s. The horizontal dotted lines show the 3$\sigma$ noise level.
• Figure 3: Column density ratios for the deuteration of methanol in L1544 computed with four of the models presented in Riedel et al. (2025)riedel25. The horizontal dashed lines show the result from the observations towards the dust peak of L1544 and the shaded region indicates the error bars of the observed ratios. Models D2 and D3 apply the single collision model proposed by Hasegawa et al. (1992)hasegawa92 with either tunnel diffusion (D2) or fast diffusion (D3). Models D4 and D5 apply the reaction-diffusion competition model proposed by Chang et al. (2007)chang07. Additionally, D5 allows for H abstraction reactions. For model D4, a zoom-in for low values of column density ratios has been added within the plot.
• Figure 4: The column density ratios of [CH$_2$DOH]/[CH$_3$OH] (upper panel) and [CHD$_2$OH]/[CH$_2$DOH] (lower panel) as a function of source types, rearranged from Lin et al. (2023a)lin23a with L1544 values in the dust (blue) and methanol (orange) peaks (from this work). Filled markers indicate single-dish observations and open markers indicate interferometric observations. The plotted values for L1544 take into consideration T$_\mathrm{{ex}}$ variations from 5 to 8 K, and use the CH$_3$OH column density derived in Lin et al. (2023b)lin23b with non-LTE models. The references for S68N and B1c are from van Gelder et al. (2022)vangelder22; for IRAS4A and IRAS2 are Taquet et al. (2019) and Parise et al. (2006)taquet19parise06; for IRAS16293A and IRAS16293B are Manigand et al. (2019)manigand19, Jørgensen et al. 2016jorgensen16, Drozdovskaya et al. (2022)dro22; for comet 67P/C-G is Drozdovskaya et al. (2021)dro21.
• Figure S1: Results from the best model (D5) from Riedel et al. (2025)riedel25 for the pre-stellar cores L694-2 and HMM-1. The horizontal dashed lines show the result from the observations and the shaded region indicates the error bars of the observed ratio lin23a.