Organohalogens, a class of molecules that contain at least one halogen atom bonded to carbon, are abundant on the Earth where they are mainly produced through industrial and biological processes1. Consequently, they have been proposed as biomarkers in the search for life on exoplanets2. Simple halogen hydrides have been detected in interstellar sources and in comets, but the presence and possible incorporation of more complex halogen-containing molecules such as organohalogens into planet-forming regions is uncertain3,4. Here we report the interstellar detection of two isotopologues of the organohalogen CH3Cl and put some constraints on CH3F in the gas surrounding the low-mass protostar IRAS 16293–2422, using the Atacama Large Millimeter/submillimeter Array (ALMA). We also find CH3Cl in the coma of comet 67P/Churyumov–Gerasimenko (67P/C-G) by using the Rosetta Orbiter Spectrometer for Ion and Neutral Analysis (ROSINA) instrument. The detections reveal an efficient pre-planetary formation pathway of organohalogens. Cometary impacts may deliver these species to young planets and should thus be included as a potential abiotical production source when interpreting future organohalogen detections in atmospheres of rocky planets.
Organohalogens are well known for their use in industry and for their detrimental effect on the ozone layer1. Some organohalogens are also produced naturally5, through different geological and biological processes. Because of their relationship to biology and industry on Earth, organohalogens have been proposed as biomarkers on other planets2,6,7. Methyl chloride (CH3Cl), the most abundant organohalogen in the Earth’s atmosphere, has both natural and synthetic production pathways. Its total production rate approaches 3 megatonnes per year, with most originating from biological processes8. Recent observations of Cl-bearing organic molecules, including methyl chloride, on Mars by the rover Curiosity, has challenged a straightforward connection between organohalides and biology; one proposed source of Cl-bearing organic molecules on Mars is meteoritic impacts9,10. This naturally raises the question of whether circumstellar and interstellar environments can produce organohalogens abiotically, and, if so, in what amounts.
Our understanding of interstellar and protostellar halogen chemistry is limited. The only halogen-bearing species so far observed in interstellar environments are diatomic and triatomic molecules: HCl (ref. 3), HCl+ (ref. 4), H2Cl+ (ref. 11,12,), HF (ref. 12)and CF+ (ref. 13). Current astrochemical models can account for these small molecules, but do not include larger halogen-bearing species such as organohalogens. It is therefore not known how this simple halogen chemistry and organic chemistry couple under interstellar conditions. In this work, we have searched for the simplest organohalogens, CH3Cl and CH3F, in two sources with previous detections of HCl and a wealth of organic molecules: the envelope of a low-mass protostar IRAS 16293–2422 (see ref. 14 for HCl and refs. 15,16,17 amongst others for detections of organic species), and the coma of comet 67P/C-G (see ref. 18 for halogen detections and ref. 19 for an inventory of organic molecules).
IRAS 16293–2422 (hereafter IRAS 16293) is a low-mass protostellar binary system in the ρ Ophiuchi star-forming region at a distance of 120 pc (ref. 20). The masses of the individual protostars suggest that the two components of the binary, IRAS 16293A and IRAS 16293B, will eventually evolve into two solar-like T Tauri stars, each surrounded by a planet-forming disk21. IRAS 16293 is best known for its rich inventory of organic molecules, which are distributed on scales of tens to hundreds of astronomical units (au) around the two sources15,16,17. The detection of HCl in its envelope14, together with its rich organic chemistry, makes IRAS 16293 an ideal target to search for methyl chloride and related organohalogens. We use data from the Protostellar Interferometric Line Survey (PILS) programme22, an ALMA unbiased survey of IRAS 16293 covering a wide frequency window between 329 and 363 GHz with a spatial resolution of about 60 au, a spectral resolution high enough to distinguish individual rotational lines of CH3Cl, as well as its fluorinated homologue CH3F.
We combine our search for organohalogens in IRAS 16293 with an analogous search in the coma of Jupiter-family comet 67P/C-G. Comets are believed to preserve the chemical composition of the solar nebula, and potentially the Sun’s birth cloud as well. By comparing abundances of organohalogens in comet 67P/C-G with the nascent solar nebula analogue IRAS 16293, we can assess whether such an inheritance is likely in the case of organohalogens. We focus on 67P/C-G because its volatile inventory is known with unprecedented detail and sensitivity for a comet, thanks to the Rosetta mission18,19. Here we make use of results from ROSINA and its Double Focus Mass Spectrometer (DFMS), whose very high mass resolution and sensitivity23 enables detections of minor constituents, including organohalogens if present.
We unambiguously detect both 35Cl and 37Cl isotopologues of CH3Cl towards IRAS 16293B. The spectra in the top and middle panels of Fig. 1 show the clear identification of the J′ K –J″ K = 13 K –12 K rotational progression of CH335Cl and CH337Cl, with all the transitions attributed up to K = 4 in a spectrum extracted one beam (0.5″) away from source B (position ii in Fig. 2). This off-source position has been identified as an optimal location for most molecular identification in IRAS 16293, because it contains strong and narrow lines that are relatively unaffected by continuum absorption. J is the rotational quantum number and K the component of J along the principal axis of symmetry for symmetric rotors. The observed transitions have an upper energy level between 116 and 386 K for CH335Cl and between 114 and 384 K for CH337Cl. The hyperfine splitting induced by the nuclear spin of 3/2 for both chlorine isotopes is partially resolved for K ≥ 4. Based on the fitting of synthetic spectra, which assumes that the gas is in local thermal equilibrium (LTE; see Methods), the CH335Cl excitation temperature is 102 ± 3 K, and the column density is (4.6 ± 0.7) × 1014 cm−2 in a 0.5″ beam, where the main source of uncertainty is the absolute flux calibration (Table 1). The reported error does not take into account possible error introduced by adopting a LTE framework. The excitation temperature is at the lower end of the range of temperatures obtained for some other organic molecules through the PILS dataset24. The CH337Cl column density is constrained to be (2.2 ± 0.4) × 1014 cm−2 using the excitation temperature and full-width at half-maximum (FWHM) derived from the stronger CH335Cl lines. This yields a 35Cl/37Cl ratio of 2.1 ± 0.4, which is consistent with the previously reported 35Cl/37Cl ratio of 2.1 derived for HCl in the envelope of IRAS 16293 encompassing both sources A and B14. The species CH3F is not securely detected (bottom panel of Fig. 1) through its J′–J″ = 7–6 rotational progression. Because of the high number of overlapping rotational lines from other species, only an upper limit on the column density of 4.6 × 1013 cm−2 can be derived, assuming the same excitation temperature and linewidth as those derived for CH335Cl.
In addition to the spectra displayed in Fig. 1, spectra of the three molecules are extracted at two more source positions: half a beam and two beams away from IRAS 16293B (positions i and iii in Fig. 2). The excitation temperature, column density and upper limits for the CH3Cl isotopologues and CH3F are reported in Table 1 for all three locations. The column density ratios of the organohalogens to methanol, the most widespread organic molecule found in protostars and key parent to many interstellar organic species, are also reported in Table 1. The ratios are about 7 × 10−5 for CH3Cl (combining the two isotopologues) and below <2 × 10−6 for CH3F, using methanol column densities reported for the corresponding locations (ref. 22, and J. K. Jørgensen et al., manuscript in preparation) and assuming that the emission is uniform over the beam (that is, a beam filling factor of unity). We also identified CH335Cl at a position 0.3″ away from IRAS 16293A, and the inferred column density ratio with respect to methanol is consistent with what is extracted from the spectra toward source B where the lines are better resolved.
Figure 2 illustrates the spatial extent of the CH3Cl emission around IRAS 16293B. The brightest line, CH335Cl 130–120, is clearly spatially resolved (line integrated intensity shown in the left column, middle panel), and is detected almost out to the edge of the dust continuum emission, 1″ or 120 au from the source centre (left column, upper panel). Similar to other lines of other species, CH335Cl 130–120 is under-luminous toward the continuum peak of the B source owing to continuum absorption22,25. Compared with CH3OCHO, an archetypical complex organic molecule found around protostars, CH3Cl is somewhat more extended, but the difference is too small to be conclusive. The CH3OCHO spatial extent for lines with similar upper-level energies is shown in the bottom left panel of Fig. 2 through the imaging of one of its line integrated intensities, and the relative spectral line strengths for both species extracted at the three positions i, ii and iii away from source B is displayed in the right panel of Fig. 2.
The origin of the detected CH3Cl is not known and is difficult to constrain without considerable theoretical and experimental efforts, as astrochemical networks do not include its formation, or that of any other organohalogens beyond CH2Cl+. Amongst the potential formation routes, we suggest that there are at least two that are consistent with the observed CH3Cl excitation temperature of about 100 K and its extended spatial emission: (1) ion–molecule gas-phase chemistry with CH4Cl+ as an intermediate, and (2) formation on grains through successive hydrogenation and halogenation of carbon followed by sublimation. Details on these pathways are available in the Supplementary Information.
The species CH3Cl is also detected in the coma of the 67P/C-G comet as shown in Fig. 3 by the mass spectrometer response centred on mass/charge 49.99 u/e (where u is unified mass unit and e the number of electronic charge), corresponding to the presence CH335Cl. The species CH337Cl and CH3F are below the detection limit. Although there are possible abundance variations in the coma along the comet orbit, a relative abundance of CH3Cl/HCl = (4 ± 2) × 10−3 can be obtained by fitting six spectra similar to the one presented in Fig. 3 centred at mass/charge values of 49.99 u/e (CH335Cl) and 35.98 u/e (H35Cl), obtained in May 2015 when the HCl signal in the coma was at its highest. Adopting the HCl/H2O ratio of 2 × 10−6−3 × 10−4 (ref. 18) and a CH3OH/H2O ratio of 3.1 × 10−3−5.5 × 10−3 (ref. 19), we estimate the bulk ratio of CH3Cl to CH3OH to be 7 × 10−7−6 × 10−4. This is consistent with the value derived for the protostar IRAS 16293, and therefore with a scenario in which comets in our Solar System and in extra-solar systems generally contain interstellar organohalogens.
The non-detection of CH337Cl in the coma of comet 67P/C-G is consistent with a 35Cl/37Cl ratio of 3.1, the average value in the Solar System. The 35Cl/37Cl ratio of 2.1 ± 0.2 obtained in the vicinity of IRAS 16293B by using CH3Cl is lower but agrees well with that measured14 using HCl in the gas around IRAS 16293. The origin of this lower isotopic ratio in IRAS 16293 compared with the Solar System has so far been explained by explosive nucleosynthesis, in which massive star nucleosynthesis is disrupted by a nearby explosive event such as a supernova explosion14. Although a chemical origin for fractionation cannot be excluded, the nucleosynthesis process alone can explain the difference in 35Cl/37Cl between IRAS 16293 and the Solar System.
Based on the cometary methyl chloride abundance, we can estimate how much methyl chloride could have been delivered to the young Earth, and may currently be delivered to nascent exoplanets, assuming that 67P/C-G is representative for comets and exocomets alike. For this calculation, we assume the maximum ratio of 1.2 × 10−6 of CH3Cl with respect to H2O derived for 67P/C-G, and use recent estimates of the maximum peak influx of impactors onto the young Earth during the late heavy bombardment of 10−6 Earth masses Myr−1 during an 80-Myr period26. We further assume that 20% of the impactors have a cometary origin, and that 40% of the comet nucleus is water ice. Under these conditions, we estimate that up to 600 tonnes per year of CH3Cl could have been delivered to the young Earth, with an accumulated maximum of 50 gigatonnes.
The survival of these organohalogens upon impact depends on the comet size, impact speed and angle, as well as the thermal decomposition parameters of CH3Cl. We expect that, similar to most organic molecules, the survival rate of the original CH3Cl brought by comets will be negligible for mid-size, high-speed vertical impactors27,28, which can induce temperatures up to 15,000 K. By contrast, slower impacts at grazing angles and impacts from larger comets that can provide shielding of the comet nucleus should increase the likelihood of survival. Reformation of CH3Cl from decomposed parent species through shock chemistry may also be important. Further studies of impact chemistry and physics are needed to assess the net delivery. The subsequent accumulation of CH3Cl in the planet atmosphere depends on the atmosphere composition and the ultraviolet irradiance (CH3Cl absorbs at wavelengths shorter than 203 nm), as well as geochemistry and presence of oceans. Dedicated studies are required to evaluate the balance between CH3Cl sources (cometary delivery, formation upon impact or outgassing from the planet interior) versus sinks such as atmospheric reaction with OH radicals, adsorption on rocks and solvation in oceans. In this context, the current protostellar and cometary detections of organohalogens represent a new reservoir of halogens and a new potential source of this family of molecules for exoplanets, especially young ones that are heavily impacted. This has consequences for the proposed use of CH3Cl detection in the atmospheres of exoplanets as a biosignature2, as a substantial percentage of this species found in rocky planet atmospheres could have been inherited from abiotic formation pathways prior to or during planet formation.
ALMA observations of the binary system IRAS 16293–2422 were taken during Cycle 2 as part of the PILS programme and are explained in detail elsewhere22. Briefly, the dataset consists of spectral cubes centred at αJ2000 = 16 h 32 min 22.72 s; δJ2000 = −24° 28′ 34.3″ with a 0.5″ beam (~60 au), covering 329–363 GHz at a spectral resolution of about 0.244 MHz. The whole frequency range is covered by 72 spectral settings of 468.75 MHz each divided into 1,920 spectral channels. Each setting was observed by the 12-m array for 13 min and by the Atacama Compact Array (ACA) for 26 min on source, resulting in a spectral noise level of about 7–10 mJy beam−1 channel−1. Titan and Ceres were used as flux calibrators, the quasar J1517–2422 was used as bandpass calibrator, and the quasar J1625–2527 was used as phase calibrator. Standard calibration through the ALMA software CASA, including flagging of bad datasets and phase-only self calibration, was performed on the obtained data cubes. The 12-m array and ACA data sets were further combined, and a continuum-subtracted dataset with a circular restoring beam of 0.5″ over the whole spectral window was produced22.
Organohalogen rotational lines were searched for in spectra extracted from the data cubes half a beam, one beam, and two beams away from the peak continuum position of IRAS 16293B. The corresponding coordinates used for the extraction from a 0.5″ beam are: αJ2000 = 16 h 32 min 22.60 s, δJ2000 = −24° 28′ 32.7″ for position i; αJ2000 = 16 h 32 min 22.58 s, δJ2000 = −24° 28′ 32.8″ for position ii; and αJ2000 = 16 h 32 min 22.54 s, δJ2000 = −24° 28′ 33.0″ for position iii. The Markov-chain Monte Carlo ensemble sampler emcee was used to fit the synthetic spectra of CH335Cl at the three extracted positions with the column density, Gaussian line FWHM, and excitation temperature taken as free parameters. A conservative absolute flux uncertainty of 15% was adapted, as well as an effective root mean square of 14 mJy beam−1 channel−1 that takes into account the potential presence of low-intensity unresolved lines for positions i and ii around source B. A total of 200 walkers and 1,000 steps were used with a burn-in period of 100 and 500 steps, adjusted after experimentation for the three extracted spectra. The obtained column densities were corrected to account for the high continuum optical depth. The FWHM values obtained are 1.1, 0.9 and 0.5 ± 0.1 km s−1 for positions i, ii and iii around IRAS 16293B, respectively. The derived temperature and linewidth were further used as inputs to derive the column density of CH337Cl and the CH3F upper limit at the different positions away from source B, using the same method described for CH335Cl.
The synthetic spectra for the CH3Cl isotopologues used above were obtained assuming a Boltzmann distribution with the gas at LTE and using the line frequencies, upper-level energies, upper-state degeneracies and line intensities available in the JPL catalogue, based on laboratory line identification of the J′–J″ = 13–12 transition29. For CH3F, a line catalogue was generated using the molecular constants from the literature (ref. 30 and references therein), obtained through line identification in the laboratory from the millimetre (30–300 GHz) to the submillimetre (THz) regime. The program SPCAT31 was used to translate spectroscopic molecular parameters into Hamiltonian terms and solve the Schrödinger equation to predict line positions and strengths for the CH3F symmetric rotor. We adopted a spin isomer ratio of unity, which is expected in the explored environments here and fits the data.
Using non-LTE escape probability calculations for CH3OH, it can be shown that LTE is a reasonable assumption for the immediate environment of IRAS16293 owing to the high density (≳3 × 1010 cm−3)22. Similar estimates cannot be made for CH3Cl directly because no collisional rate coefficients with H2 are available, but for the main transitions the high density should ensure that LTE is a good approximation. The CH3Cl excitation temperature derived from a single J-ladder should be a good proxy for the gas kinetic temperature; however to obtain a complete view of the rotational population distribution would require additional lines from different J level.
The ROSINA-DFMS spectrum was obtained in May 2015 when the HCl abundance in the coma was maximal. The comet was between 1.7 and 1.5 au from the Sun and the spacecraft at 130–200 km from the comet centre. The high mass resolution of the DFMS allows for the distinction of isotopologues with different elemental composition (m/Δm of ~9,000 at FWHM at mass 28 u/e). Thus the signal centred at 49.99 u/e is univocally a fragment containing a combination of one carbon, one chlorine and three hydrogens. As CH3Cl is a saturated species, direct fragmentation from heavier molecules only yields CH3Cl upon secondary reaction after the ionizing stage in the DFMS. Because these reactions are extremely inefficient in low pressure environments, we can securely attribute the signal at 49.99 u/e signal to CH335Cl. An abundance ratio can be derived by performing a least-squares fit of the spectra using a linear combination of pre-flight measures of the fragmentation responses for the various species present in the coma19,23.
The data that support the plots within this paper and other findings of this study are available from the corresponding author upon reasonable request. This paper makes use of the ALMA dataset ADS/JAO.ALMA#2013.1.00278.S.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This work is based on observations from ALMA, a partnership of ESO (representing its member states), NSF (USA) and NINS (Japan), together with NRC (Canada), NSC and ASIAA (Taiwan), and KASI (Republic of Korea), in co-operation with the Republic of Chile. The Joint ALMA Observatory is operated by ESO, AUI/NRAO and NAOJ. Data from ROSINA, an instrument part of Rosetta mission, were used in this work. Rosetta is a European Space Agency (ESA) mission with contributions from its member states and NASA, and we acknowledge herewith the work of the whole ESA Rosetta team. E.C.F. and K.I.O. acknowledge financial support from the Simons Foundation (SCOL award 321183, KO) and to Northrop Grumman Corporation. The group of J.K.J. acknowledges support from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement no. 646908) through ERC Consolidator Grant S4F. Research at the Centre for Star and Planet Formation is funded by the Danish National Research Foundation. Work at the University of Bern was funded by the State of Bern, the Swiss National Science Foundation, and the ESA PRODEX programme (Programme de Développement d’Expériences scientifiques). E.F.v.D. acknowledges A-ERC grant CHEMPLAN 291141. M.N.D. acknowledges the financial support of the Center for Space and Habitability (CSH) Fellowship and the IAU Gruber Foundation Fellowship. S.F.W. acknowledges financial support from a CSH fellowship.
Electronic supplementary material
Supplementary Text and Supplementary References.