Special Issue Reports

CHO-bearing organic compounds at the surface of 67P/Churyumov-Gerasimenko revealed by Ptolemy

See allHide authors and affiliations

Science  31 Jul 2015:
Vol. 349, Issue 6247, aab0673
DOI: 10.1126/science.aab0673


The surface and subsurface of comets preserve material from the formation of the solar system. The properties of cometary material thus provide insight into the physical and chemical conditions during their formation. We present mass spectra taken by the Ptolemy instrument 20 minutes after the initial touchdown of the Philae lander on the surface of comet 67P/Churyumov-Gerasimenko. Regular mass distributions indicate the presence of a sequence of compounds with additional -CH2- and -O- groups (mass/charge ratios 14 and 16, respectively). Similarities with the detected coma species of comet Halley suggest the presence of a radiation-induced polymer at the surface. Ptolemy measurements also indicate an apparent absence of aromatic compounds such as benzene, a lack of sulfur-bearing species, and very low concentrations of nitrogenous material.

The Philae lander contacted the surface of comet 67P/Churyumov-Gerasimenko (67P) at 15:34:04 UTC, but it failed to anchor itself to the surface as planned (1) and instead “bounced,” eventually coming to land at the site now known as Abydos. This bouncing may have occurred because of the presence of a surface layer of loosely consolidated “dust” overlying a relatively hard interior composed of sintered ice (2). Philae is equipped with a drilling system (3) capable of procuring samples from the surface and to depths of at least 20 cm. The Ptolemy instrument (4) is designed to accept these solid samples and measure stable isotope ratios of elements such as hydrogen, carbon, nitrogen, and oxygen. Additionally, the instrument can operate in a compositional analytical mode and can use its mass spectrometer to measure volatiles. As part of the First Science Sequence of Philae, the instrument was preprogrammed to perform analyses in “sniff mode” ~20 min after touchdown.

The Ptolemy instrument is located inside Philae and connected to space via a vent pipe, the exhaust for which is located on the top of the lander. Ptolemy was switched on 9 min after the touchdown signal, and (after executing sequences for initialization and confirmation that the oven tapping station was undocked) it began collecting mass spectra in sniff mode some 11 min later. The ion trap mass spectrometer (supplementary materials) collected a total of six mass spectra at 14-s intervals, each with an ionization time of 0.2 s. Two types of spectra, similar to those used for the Lutetia flyby (5) and for operations while in orbit, were acquired alternately. The first had a mass range of mass/charge ratio (m/z) 13 to 89; the second had a mass range of m/z 25 to 136. Based on a number of similar sniff-mode measurements taken during post-hibernation commissioning and at various distances from the comet, the mass spectra collected after the initial touchdown on 67P (Fig. 1) appear to be of cometary origin. Background observations, made in orbit at distances farther than 30 km from the comet while still attached to the spacecraft, typically had ion counts of 30, 2, and 4 for H2O, CO, and CO2, respectively. These ion counts are significantly lower than the measurements made at Agilkia and are similar to background measurements made during the Lutetia flyby (5). Instrumental biases can affect the relative sensitivities of different species and masses, but they do not affect the overall pattern of the mass spectra.

Fig. 1 Mass spectra taken by Ptolemy in sniff mode.

The first combined mass spectra taken 20 min after the initial touchdown event.

In our data, we expected to see the main components of the coma, namely, H2O, CO, and CO2 (m/z 18, 28, and 44, respectively). The most abundant peak arose from H2O, which occurred at m/z 19 because of protonation (i.e., H3O+; see the supplementary materials), with an accompanying cracking pattern at m/z 18, 17, and 16 from H2O+, OH+, and O+, respectively. The ions ascribed to H2O only constitute about 30% of the total number of ions in the mass spectra. The signal at m/z 45 represents a combination of protonated CO2 and organics. Based on sniff-mode measurements made after the lander came to rest and on data from ROSINA (the Rosetta Orbiter Spectrometer for Ion and Neutral Analysis) (6), the estimated H2O/CO2 ratio of about 5:1 indicates that 60% of the signal at m/z 45 is from CO2. Finally, the ion count at m/z 29 reflects a combination of signals due to the presence of CO and N2 organics and the cracking of CO2. The low m/z 14 signal indicates that N2 is not in high abundance, which is in agreement with the aerage N2/CO ratio of 5.7 × 10−3 obtained by ROSINA (7). Assuming that the CO2 contribution to m/z 29 is 10% of the CO2 peak at m/z 45 (8), the remaining spectra reflect the amount of CO and organics present. In this way, we put an upper limit of 2:1 on the CO2/CO ratio. Taken together, the ratio of H2O/CO2/CO is on the order of 10:2:>1.

In these mass spectra, some peaks (compounds) are notable by their absence (i.e., the compounds were below detection limits). For instance, there is no evidence for any sulfur-bearing components (neither H2S nor SO2 signals were present). There was no significant signal at m/z 78 (benzene, C6H6+), suggesting that aromatic compounds are not present in large concentrations. The very small peak at m/z 14 constrains the presence of nitrogen-bearing compounds to very low levels. We believe that the peak at m/z 15 is from CH3+ rather than from NH+. Unambiguous detection of NH3 was not possible because of the presence of H2O and other potential compounds, such as CH4. Clear detection of NH3 would require a concentration of >10% relative to that of H2O.

Almost all of the species present had m/z ratios <105 (ions of higher mass were detected, but the signals were very low, making it difficult to be confident of their exact nature) (Fig. 1). Therefore, it appears that relatively low-mass compounds and fragments dominated the sample. Ptolemy only operates over a range of m/z 13 to 136, meaning that heavier compounds may have been present. However, were such compounds present, it would have been reasonable to expect some peaks from cracking in the range of m/z 105 to 136. An alternative explanation for the absence of peaks in this range is that high-molecular-weight components were present but did not reach the mass spectrometer ionization source. Greater insight into the nature of higher–molecular-weight compounds may come from the Cometary Sampling and Composition experiment (COSAC) if it is able to analyze a drilled sample (9). For the COSAC measurements that were made at Agilkia (10), the conditions of analysis were similar to those of Ptolemy’s sniff mode. There are similarities in the mass spectra of the two instruments but also some intriguing differences (10).

In a process guided by the pattern of peaks observed by the PICCA (Positive Ion Cluster Composition Analyzer) instrument (11), which identified polyoxymethylene in mass spectral measurements of coma materials from comet Halley during the Giotto mission (12), we superimposed mass increments of 14 and 16 (representing additions and losses of -CH2- and -O2-, respectively) on the mass spectra, with peaks ascribed to H2O and CO2 removed (Fig. 2). The data presented here do not reflect a single idealized compound polymer [e.g., (CH2O)n] with its ends terminated by H atoms. Rather, they indicate a number of different terminations, with the chain running as either -O-CH2- or -CH2-O- (i.e., repeating units of 16:14 or 14:16 m/z) (Fig. 3). In principle, such terminations result from H-, HCO-, or CH3CO-, which can be thought of as radicals arising from hydrogen, formaldehyde, and acetaldehyde, respectively (although it depends on exactly where in the chain one considers the termination to occur). The mass spectra are consistent with the presence of formaldehyde (peaks at m/z 29 to 31) and acetaldehyde (peaks at m/z 29, 43, and 44), although Ptolemy has insufficient mass resolution to identify these individual compounds.

Fig. 2 Proposed polyoxymethylene fit to the Ptolemy spectra.

(A) Schematic for proposed mass fragments of polyoxymethylene with different terminations. (B) Peaks that are considered to be from polyoxymethylene. (C) Ptolemy spectra with peaks from H2O and CO2 removed.

Fig. 3 Idealized polyoxymethylene chains.

(A) Idealized polyoxymethylene chain with repeating units of 16:14 m/z (-O-, -CH2-). For the mass spectra taken by Ptolemy, we considered three different types of termination: H-, HCO-, and CH3CO-. The H- termination would produce peaks at 1, 17, 31, 47, 61, 77, 91, 107, and 121 m/z. For HCO-, peaks would occur at 29, 45, 59, 75, 89, 105, and 119. For CH3CO-, peaks would occur at 43, 59, 73, 89, 103, and 119. Here, we consider only those peaks up to a mass of about 120. (B) The equivalent chain, but with repeating units in the reverse order [14:16 m/z (-CH2-, -O-)]. In this case, the H- termination would produce peaks at 1, 15, 31, 45, 61, 75, 91, 105, and 121 m/z. For HCO-, peaks would occur at 29, 43, 59, 73, 89, 103, and 119. For CH3CO-, peaks would occur at 43, 57, 73, 87, 103, and 117.

An apparent regularity in the mass spectra does not necessarily confirm the presence of a polymer; rather, the spectra could represent a collection of individual compounds. In principle, a polymer might be expected to produce peaks at higher masses, but we cannot be certain that some aspect of its volatility did not prevent higher-mass fragments from entering the instrument. If the data do reflect the presence of a polymer, that does not necessarily imply the presence of polyoxymethylene (13, 14). Other polymers may be possible, including an “ice tholin” (15). Ice tholins can form through the irradiation of mixtures of simple components such as water, methanol, carbon dioxide, and ethane, a combination of molecules similar to those present on 67P. Related experiments (16) indicate that acetaldehyde is not produced in ice tholins, yet it appears likely that it is one of the main termination compounds reflected by the Ptolemy data. Glycerol (glycerine, C3H8O3) is produced in ice tholins and has a peak at m/z 92, which is prominent in our results (and ostensibly is not accounted for by a polyoxymethylene fit). In principle, there is no reason why polyoxymethylene could not be a component of an ice tholin, but it has been suggested (16) that the structure of the latter is based on a substituted polyalcohol (i.e., a material that results from direct polymerization of methanol radicals and water, as opposed to formaldehyde, which produces polyoxymethylene).

Three notable peaks in the mass spectra do not fit with an interpretation of the presence of polyoxymethylene. These occur at m/z 33, 41, and 37. For the peak at 33, we considered the possibility of hydroxylamine (H2NOH). Although it has not been detected either in the interstellar medium or in cometary comae, it may be a precursor to amino acids (17). A cracking pattern for this molecule could fit within the mass spectra obtained by Ptolemy (the molecule has, for instance, a relatively minor peak at m/z 14). However, considering the apparent lack of other N-bearing species, this molecule seems implausible. We propose instead that the peak at m/z 33 probably indicates an oxonium ion, namely CH3.OH2+ (analogous to the peak at m/z 19, which is effectively H.OH2+). The possibility of protonated methanol (CH3OH.H+) is discounted because of a lack of fit for the cracking pattern.

It is tempting to ascribe the peak at m/z 41 to acetonitrile or methyl isocyanide (CH3CN), both of which have been detected in cometary comae (18). However, we discount this in favor of C3H5+; that is, a fragment of a hydrocarbon (assuming that hydrocarbons are present). The peak at m/z 37 also has several possible explanations. If it is organic in origin, then the only possibility is C3H+, and therefore it could be related to the peak at m/z 41. However, it seems unlikely that there would be such a prominent fragment without the larger peaks expected to accompany it. On the other hand, argon readily protonates in the ion trap mass spectrometer, making 36Ar.H+ a possibility. This would require that argon be present at an extremely high level of ~2%; we consider this unlikely because a high argon value has not been reported in ROSINA measurements. An alternative possibility is that an ion molecule reaction between neutral H2O and H.OH2+ in the ion trap formed an ion cluster of H2O.H.OH2+, although this identification remains speculative.

The mass spectra obtained during the first contact with the comet 67P indicate that the surface contains a complicated mixture of organics, as well as H2O and CO2. Many of the features of the mass spectra can be explained by the presence of polyoxymethylene, but undoubtedly many other compounds are present at low concentrations. Before the Giotto encounter with Halley, researchers had already conjectured that interstellar grains may contain polyoxymethylene (19). The possible presence of similar materials on a comet, as postulated by a consideration of the PICCA results, raised interest in the subject of prebiotic polymers. The most immediate scientific impact of the possible presence of polyoxymethylene was that it provided an explanation for the presence of observable formaldehyde in cometary comae—that is, at distances beyond which the molecular species, if released directly from the nucleus, would be expected to survive (20). However, after somewhat straightforward initial interpretations (21), more detailed enquiries have uncovered issues that remain unresolved (2225).

Supplementary Materials


  1. ACKNOWLEDGMENTS: The paper is dedicated to the memory of C. Pillinger, who was the original proposer of the Ptolemy instrument. We acknowledge the interest and inputs from our colleagues involved with the Ptolemy science team. Ptolemy was developed in a collaborative venture between The Open University and RAL Space, with funding provided by the Science and Technology Facilities Council and the UK Space Agency. We acknowledge the efforts and patience of our colleagues at the Lander Control Centre and the Science Operation and Navigation Centre, as well as the lander principal scientists J.-P. Bibring and H. Böhnhardt. All Ptolemy data are available from the Planetary Science Archive of the European Space Agency and the Planetary Data System archive of NASA.

Stay Connected to Science

Navigate This Article