- Email: [email protected]
A sink for methane on Mars? The answer is blowing in the wind
Accepted Manuscript A sink for methane on Mars? The answer is blowing in the wind Svend J. Knak Jensen, Jørgen Skibsted, Hans J. Jakobsen, Inge L. ten Kate, Haraldur P. Gunnlaugsson, Jonathan P. Merrison, Kai Finster, Ebbe Bak, Jens J. Iversen, Jens C. Kondrup, Per Nørnberg PII: DOI: Reference:
S0019-1035(14)00162-6 http://dx.doi.org/10.1016/j.icarus.2014.03.036 YICAR 11018
To appear in:
Icarus
Received Date: Revised Date: Accepted Date:
11 September 2013 5 March 2014 15 March 2014
Please cite this article as: Knak Jensen, S.J., Skibsted, J., Jakobsen, H.J., ten Kate, I.L., Gunnlaugsson, H.P., Merrison, J.P., Finster, K., Bak, E., Iversen, J.J., Kondrup, J.C., Nørnberg, P., A sink for methane on Mars? The answer is blowing in the wind, Icarus (2014), doi: http://dx.doi.org/10.1016/j.icarus.2014.03.036
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
A sink for methane on Mars? The answer is blowing in the wind. Svend J. Knak Jensena,*, Jørgen Skibsteda, Hans J. Jakobsena, Inge L. ten Kateb, Haraldur P. Gunnlaugssonc, Jonathan P. Merrisonc, Kai Finsterd, Ebbe Bakd, Jens J. Iversenc, Jens C. Kondrupa, Per Nørnberge a
Instrument Center for Solid-State NMR Spectroscopy, Department of Chemistry, Interdisciplinary Nanoscience Center (iNANO), Langelandsgade 140, Aarhus University, DK-8000 Aarhus C, Denmark. b
Department of Earth Sciences, Utrecht University, Budapestlaan 4, 3584 CD Utrecht, The Netherlands.
c
Department of Physics and Astronomy, Ny Munkegade 120, Aarhus University, DK-8000 Aarhus C, Denmark. d
Department of Bioscience, Microbiology section, Ny Munkegade 114 – 116, Aarhus University, DK-8000 Aarhus C, Denmark.
e
Department of Geoscience, Høegh-Guldbergs Gade 2, Aarhus University, DK-8000 Aarhus C, Denmark.
Abstract Tumbling experiments that mimic the wind erosion of quartz grains in an atmosphere of enriched methane are reported. The eroded grains are analyzed by
13
C and
29
13
C-
Si solid-state NMR
techniques after several months of tumbling. The analysis shows that methane has reacted with the eroded surface to form covalent Si-CH3 bonds, which stay intact for temperatures up to at least 250 °C. The NMR findings offer an explanation for the fast disappearance of methane on Mars.
Keywords: Mars, Atmosphere; Aeolian processes; Atmospheres, Chemistry
*
Corresponding author: Svend J. Knak Jensen, Phone: +(45) 87155926, Fax: +(45) 8619 6199, Email: [email protected]
2
1. Introduction Recently methane (CH4) has been observed in the Martian atmosphere from a satellite orbiting the planet (Formisano et al., 2004) as well as from Earth based telescopes (Krasnopolsky et al., 2004). These observations are exciting as methane may be a fingerprint of past or present life on Mars. Currently, the Curiosity rover on Mars is exploring various issues related to methane. However, so far the source of methane has not been identified. Potential sources could be biological in origin, like methanogenesis in extreme environments (Reid et al., 2006) or geochemical processes, like serpentinisation (Oze and Sharma, 2005) or volcanism (Etiope et al., 2007) or methane release from clathrates (Chastain and Chevrier, 2007). Release of methane from such processes could occur as episoidic events. Likewise, the fate of methane is a mystery. Observations indicate that methane disappears within a few years after a plume eruption (Mumma et al., 2009 and Lefèvre and Forget, 2009), while the photochemical processes that operate on Earth predict a lifetime in the range of several hundred years (Mumma et al., 2009). Various mechanisms proposed for the disappearance of methane have difficulties reproducing its short residence time (ten Kate, 2010). Reported concentrations of methane are in the range of tens of ppbv. A significant feature of methane concentrations is that they show a substantial time and spatial variation (Geminale et al., 2008, Mumma et al., 2009, Lefèvre and Forget, 2009, Geminale et al., 2011, Krasnopolsky 2012). Here we show, using solidstate
13
C and
29
Si magic-angle spinning NMR spectroscopies, that wind driven erosion produces
highly reactive sites on mineral grain surfaces that sequester methane by forming covalent bonds with methyl groups and propose that this mechanism can be the hitherto undiscovered methane sink on Mars. The mechanism is supported by experiments where quartz (SiO2) grains are exposed to gentle mechanical agitation in a methane atmosphere, using a specially designed experimental setup, that mimics mineral grain transport created by a typical Martian wind. The predicted lifetime
3
of methane agrees well with the one experimentally observed. We anticipate that the designed agitation process can be useful in studies of surface erosion and surface chemistry. After this paper was submitted, in situ measurements by the Curiosity team found very little evidence, if any, for methane in the Gale crater on Mars (Webster et al., 2013). These findings are discussed in relation to our findings in Sec. 4. 2. Materials and methods. The wind driven erosion of surface material is simulated using the specially designed apparatus depicted in Fig. 1. Commercially available quartz (Merck, 1.07536) was chosen as an analogue for surface material because of its simple chemical composition. The quartz was placed in a borosilicate flask with
13
C-methane (Sigma-Aldrich, 490229, 99% enriched) to facilitate NMR
investigations. The basic purity of the 13C-methane gas was confirmed by 13C NMR experiments on the gas loaded in a glass tube. The flask with quartz and 13C-methane was sealed and mounted in a carousel where it was tumbled end-over-end at 30 RPM. The collision speed of the quartz grains as they tumble is about 1 m/s, which is a typical saltation collision speed close to threshold (HolsteinRathlou et al., 2010, Merrison et al., 2010 and Merrison, 2012). After several months of tumbling the quartz material was analyzed by solid-state NMR techniques (details of the NMR experiments are given in the Supplementary Information).
3. Results and discussion The reaction of
13
C-enriched methane with surface sites of highly active quartz particles is
unambiguously demonstrated by the 13C{1H} CP/MAS and 29Si{1H} CP/MAS NMR spectra shown in Figs. 2 and 3, respectively. In these spectra the cross-polarization (CP) NMR technique transfers 1
H magnetization to either the 13C or 29Si spins via heteronuclear dipolar couplings and thereby acts
as a filter for detecting only 13C and 29Si spin nuclei within a distance less than 3-5 Å to nearby 1H
4
nuclei. The
13
C{1H} CP/MAS NMR spectrum of the quartz grains agitated in an atmosphere of
methane, and stored under standard atmospheric and temperature conditions (Fig. 2A), exhibits a resonance at δ(13C) = –1.7 ppm, i.e., very similar to the
13
C chemical shift reported for Si–CH3
groups; for example, δ(13C) = 0.0 ppm for Si(CH3)4 (TMS). In addition to the 13C CP/MAS NMR spectrum, we note that several other
13
recovery T1(13C) measurements and
C MAS NMR experiments for the sample (e.g., inversion-
13
C{1H} cross-polarization – depolarization experiments)
reveal restricted rotation for the methyl group around the Si–CH3 bond. Samples from the methanequartz batch have been exposed to heat-treatment at 250 and 550 °C for one hour. The
13
C{1H}
CP/MAS NMR spectrum of the 250 °C sample (Fig. 2B) shows intact Si–CH3 groups while no 13C signal could be detected for the 550 °C sample. The line width of the resonance for the 250 °C sample (Fig. 2B) is slightly more narrow (FWHM = 5.5 ppm) compared to that shown in Fig. 2A for the original sample (FWHM = 7.7 ppm). The observed line-narrowing may be ascribed to annealing processes and/or to elimination of water dehydroxylated from the quartz surface. A second flask with quartz and
13
C-methane, prepared under identical conditions as used for the
sample described above, was not exposed to tumbling and kept stationary for the same period of time.
13
C{1H} CP/MAS NMR spectra of this sample show no
13
C NMR resonances at all, which
demonstrate that the reaction of methane with quartz requires activation of the quartz grains by the tumbling process. The standard one-pulse 29Si MAS NMR spectrum of the methane-quartz sample in Fig. 3B exhibits a narrow resonance (FWHM = 0.11 ppm) at δ(29Si) = –107.5 ppm, i.e., the well-known
29
Si
chemical shift for α-quartz (Lippmaa et al., 1980 and Smith and Blackwell, 1983), and thus is assigned to the bulk SiO2 structure of the sample. More importantly, the sample are selectively detected in the
29
29
Si surface sites of the
Si{1H} CP/MAS NMR spectrum (Fig. 3A), which reveals
two broadened resonances at –61 and –101 ppm. The high-intensity resonance at –101 ppm
5
originates from
29
Si sites associated with hydroxyl groups, following earlier
29
Si CP/MAS NMR
studies of silica gels. More importantly, only this resonance at –101 ppm is observed in a similar spectrum of pure quartz exposed to tumbling in ambient air under the same conditions as used for the SiO2/13CH4 sample. Most interestingly, from previous investigations of modified silica surfaces, HPLC materials, and heterogeneous catalysts (Albert and Bayer, 1991 and Kellberg et al., 1993) it is known that methyl groups directly bonded to a Si atom on a silica surface give 29Si resonances in distinct regions of the 29Si chemical shift scale according to the number of attached methyl groups. For example, for a (CH3)2Si(OSi)2 species: δ(29Si) = –14 to –20 ppm, while for CH3Si(OSi)3 species: δ(29Si) = –53 to –65 ppm. Thus, the observed resonance at –61 ppm can be assigned to a (SiO)3Si–CH3 site. This result and the absence of the resonance at –61 ppm for the tumbled sample of pure quartz present an unambiguous and direct proof that agitation of quartz in a methane atmosphere results in a methyl group being directly bonded to a Si atom on the surface of the quartz particles, according to the formal reactions:
(SiO)3Si–OH + CH4 → (SiO)3Si–CH3 + H2O
and/or the breakage of dehydroxylated surface (SiO)3Si–O–Si(OSi)3 linkages according to
(SiO)3Si–O–Si(OSi)3 + CH4 → (SiO)3Si–CH3 + HO–Si(OSi)3
We note that the tumbling of the quartz grains in the borosilicate glass flask leads to a minor degree of abrasion of the borosilicate glass. This is apparent from 11B MAS NMR spectra of the tumbled SiO2/13CH4 sample and a ground sample of the borosilicate glass. Both spectra show the exact same set of
11
B centerband resonances from trigonal (BO3) and tetrahedral (BO4) units from the glass
6
material. However, a quantitiative evaluation of the intensities in the two spectra reveal that the SiO2/13CH4 sample only includes 1 – 2 wt% of the borosilicate glass as an impurity. Moreover, a 29
Si MAS NMR spectrum of the pure borosilicate glass shows a broadened resonance from approx.
– 95 to –125 ppm, characteristic for glass materials, which thereby does not overlap with the resonance at –61 ppm from the (SiO)3Si–CH3 sites in the SiO2/13CH4 sample. In our investigation quartz is used as an analogue for general silicate minerals found on Mars in the airborne dust and on the surface rocks. Detailed chemical analyses of the Martian regolith (Gellert et al., 2004 and McSween, 2004) show the presence of olivines ((Mg,Fe)2SiO4), pyroxenes (diopside) (MgCaSi2O6), and plagioclases (NaAlSi3O8 – CaAl2Si2O8 solid-solution series), for which (SiO)3Si–O–Si(OSi)3 linkages are dominant chemical entities. However, the chemical analyses also show that the regolith has a typical FeO content of about 15-16 wt%. Martian regolith analogues with a similar composition can be found on Earth, but the high content of paramagnetic FeO would make NMR investigations of such samples impossible. Thus, the large content of different silicates mentioned above could act as alternative materials for a methane sink in the Martian winds. Evidence supporting this hypothesis comes from experiments where the pressure of the gas phase is monitored during the tumbling process. In addition to the SiO2/13CH4 experiments we also tumbled a sample of olivine (from the Spanish island of Tenerife) in ordinary CH4 (with an isotope distribution of about 12C = 99 % and 13C = 1 %). The rate of pressure change for olivine is lower than that for quartz, but not more than a factor of two. We note that 12CH4 may react faster than 13CH4 because the activation energy for breaking the 12C-H bond is lower than that for the 13CH bond, which in turn is due to the fact that the heavier isotope has a lower vibration frequency (Levine, 2002). However, the difference in reaction rates is expected to be small. The dominant gas in the Martian atmosphere is CO2 and one might wonder if this gas could also react with the active sites in the Martian dust and regolith and perhaps monopolize all active sites.
7
Preliminary tumbling experiments with SiO2/13CO2 and subsequent solid-state
13
C NMR analysis
show a room-temperature 13C resonance at δ(13C) = 125.6 ppm, which disappears after exposing the sample to 250 °C for one hour. This indicates a weaker interaction than in the case of methane. Moreover, the amount of dust and surface material is huge so there are enough active sites for methane to react with. Quantitative single-pulse well-defined second relaxed
13
13
C MAS NMR analysis of the original agitated quartz sample, using a
13
C intensity reference sample (hexamethylbenzene) and employing fully-
C MAS NMR spectra (without CP) for both samples, shows that one gram of quartz
tumbled for 115 days has established 1.6·10–5 moles of Si-CH3 bonds. Combining this number with a surface area of 22 m2/g, determined for a tumbled sample of pure quartz, indicates that roughly 3 – 4 % of the surface Si sites are present as (Si-O)3Si-CH3 sites. If these numbers are translated to Martian conditions, then a burst of 270 tons of methane (Krasnopolsky et al., 2004 and Krasnopolsky, 2006) will require only about 106 tons of activated surface material for methane to disappear from the atmosphere within about one third of a year. There is some debate regarding the actual amount of methane estimated by Krasnopolsky. Villanueva et al. estimate the amount to be at least 4500 tons (Villanueva et al., 2013). However, the amount of available surface material is still in big surplus. Also on Earth, eroded and atmosphere suspended silicates should be considered as a sink for atmospheric methane. In general, wind mediated chemistry could take place on all planets and moons with a solid surface and an atmosphere.
4. Recent Curiosity findings After the original version of this paper was submitted a highly interesting paper dealing with in situ concentration measurements of methane in the Gale crater on Mars (Webster et al., 2013) appeared.
8
Detailed recent snapshots measurements from this investigation on-board the Curiosity rover have shown that the concentration of methane is very low, i.e., 0.18 ± 0.67 ppbv. This finding underscores the challenging question of what happened to the methane observed by several groups a few years earlier. It is obvious that methane on Mars is a big issue as already two spacecraft (NASA's Mars Atmosphere and Volatile Evolution Mission, MAVEN, and ISRO's Mars Orbiter Mission, MOM) are on their way to the planet to (among other tasks) perform detailed measurements of methane levels in the atmosphere over long periods of time. Also ESA is planning to launch the ExoMars Trace Gas Orbiter in 2016 with a similar objective. Current evidence shows that methane levels have a substantial spatial and temporal variation. However, in the words of Webster et al. (Webster et al., 2013): The very short methane lifetime of 0.4 to 4 years derived from the 2003–2006 observations requires powerful destruction mechanisms that have not been identified to date. Our laboratory studies show that a wind mediated erosion process of ordinary quartz crystals can produce activated quartz grains, which sequester methane by forming covalent Si-C bonds. If this process is operational on Mars, which our recent preliminary studies on olivine indicate could be the case then it can explain the observed fast destruction of methane. Of course, there is a possibility that the methylated material is a temporary sink which may undergo chemical reactions with surface regolith, perhaps triggered by the abundant UV radiation. Therefore, we propose that in future Mars explorations it will be of high interest to look for chemical evidence of methylated materials – or downstream reaction products- in the Martial regolith and dust.
9
Figure 1: Schematic drawing of the tumbling apparatus.
The flask (borosilicate,Simax 3.3) contains about 10 grams of SiO2 grains, which have been sieved to extract the fraction between 125 and 1000 µm. This fraction was ultrasonic treated and washed several times to remove all attached finer particles. The gas in the flask is
13
C-enriched CH4 at a
pressure of approximately 600 mbar. The tumbling rate is 30 RPM. The whole tumbling apparatus is enclosed in a box to avoid any potential interference from photochemistry.
10
Figure 2: 13C{1H} CP/MAS NMR spectra.
Spectra of the quartz sample, agitated in an atmosphere of 99 % 13C-enriched CH4 for 115 days at room temperature. The spectra are acquired at room temperature and at 7.05 T using a CP contact time τCP = 3.0 ms. (A) Spectrum of the original sample using a spinning frequency νR = 10.0 kHz and acquisition of 18944 scans. (B) Spectrum of the original sample, after heat-treatment at 250 oC for 1 hr, using a spinning frequency νR = 3.0 kHz and acquisition of 1024 scans.
11
Figure 3: 29Si MAS and CP/MAS NMR spectra.
Spectra of the quartz sample, agitated in an atmosphere of 99 % 13C-enriched CH4 for 115 days at room temperature. The spectra are acquired at room temperature and at 9.39 T. The different highlighted environments for the central Si atom refer to their 29Si chemical shifts discussed in the text. (A) 29Si{1H} CP/MAS NMR spectrum acquired using a spinning frequency νR = 4.0 kHz, a CP contact time τCP = 4.0 ms, a relaxation delay of 4 s , and 30720 scans. (B) Single-pulse 29Si MAS NMR spectrum acquired using a spinning frequency νR = 6.0 kHz, a relaxation delay of 120 s, and 644 scans.
12
References Albert and Bayer, 1991 K. Albert, E. Bayer Characterization of bonded phases by solid-state NMR spectroscopy J. Chromatogr. 544 (1991), pp. 345-370
Chastain and Chevrier, 2007 B.K. Chastain, V. Chevrier Methane clathrate hydrates as a potential source for Martian atmospheric methane Planet. Space Sci. 55 (2007), pp. 1246–1256 Etiope et al., 2007 G. Etiope, T. Fridriksson, F Italiano, W. Winiwarter, J. Theloke Natural emissions of methane from geothermal and volcanic sources in Europe J. Volcanol. Geoth. Res. 165 (2007), pp. 76-86 Formisano et al., 2004 V. Formisano, S. Atreya, T. Encrenaz, N. Ignatiev, M. Giuranna Detection of methane in the atmosphere of Mars Science 306 (2004), pp. 1758-1761 Gellert et al., 2004 R. Gellert, et al. Chemistry of rocks and soils in Gusev Crater from the Alpha Particle X-ray spectrometer Science 305 (2004), pp. 829-832 Geminale et al., 2008 A. Geminale, V. Formisano, M. Giuranna Methane in Martian atmosphere: Average spatial, diurnal, and seasonal behavior Planet. Space Sci. 56 (2008), pp. 1194-2003 Geminale et al., 2011 A. Geminale, V. Formisano, G. Sindoni Mapping methane in Martian atmosphere with PFS-MEX data Space Sci. 59 (2011), pp. 137-148 Holstein-Rathlou et al., 2010 C. Holstein-Rathlou, et al. Winds at the Phoenix landing site J. Geophys. Res., 115 (2010), p. E00E16 Kellberg et al., 1993 L. Kellberg, P. Zeuthen, H. J. Jakobsen Deactivation of HDT calatalysts by formation of silica gels from silicone oil. Characterization of spent catalysts from HDT of coker naphta using 29Si and 13C CP/MAS NMR J. Catal. 143 (1993), pp. 45-51 Krasnopolsky et al., 2004 V. Krasnopolsky, J. Maillard, T. Owen Detection of methane in the Martian atmosphere: evidence for life? Icarus 172 (2004), pp. 537-547 Krasnopolsky, 2006 V. Krasnopolsky Some problems related to the origin of methane on Mars Icarus 180 (2006), pp. 359-367 Krasnopolsky, 2012 V. Krasnopolsky Search for methane and upper limits to ethane and SO2 on Mars Icarus 217 (2012), pp. 144-152
13
Lefèvre and Forget, 2009 F. Lefèvre, F. Forget Observed variations of methane on Mars unexplained by known atmospheric chemistry and physics Nature 460 (2009), pp. 720-723 Levine, 2002 I. R. Levine Physical Chemistry McGraw-Hill (2002), pp. 903 Lippmaa et al., 1980 E. Lippmaa, M. Mägi, A. Samoson, G. Engelhardt and A. R. Grimmer Structural studies of silicates by solid-state high-resolution 29Si NMR J. Am. Chem. Soc., 102 (1980), pp. 4889-4893 McSween, 2004 H. Y. McSween Basaltic rocks analyzed by the Spirit Rover in Gusev Crater Science 305 (2004), pp. 842-845 Merrison et al., 2010 J. P. Merrison, H. P. Gunnlaugsson, S. J. Knak Jensen, P. Nørnberg Mineral alteration induced by sand transport; a source for the reddish color of Mars Icarus, 205(2) (2010), pp. 716-718 Merrison, 2012 J. P. Merrison Sand Transport, Erosion and Granular Electrification Aeolian Research 4 (2012), pp. 1–16 Mumma et al., 2009 M. J. Mumma, et al. Strong release of methane on Mars in Northern Summer Science 323 (2009), pp. 1041 -1045 Oze and Sharma, 2005 C. Oze, M. Sharma Have olivine, will gas: Serpentinization and the abiogenic production of methane on Mars Geophys. Res. Lett. 32 (2005) L10203 Reid et al., 2006 I.N. Reid et al., Terrestrial models for extraterrestrial life: methanogens and halophiles at Martian temperatures Int. J. Astrobio. 5 (2006) 89-97 Smith and Blackwell, 1983 J. V. Smith, C.S Blackwell Nuclear magnetic resonance of silica polymorphs Nature 303 (1983), pp. 223-225 ten Kate, 2010 Inge L. ten Kate Organics on Mars? Astrobiology 10 (2010), pp. 589-603 Villanueva et al., 2013 G. L. Villanueva, M. J. Mumma, R. E. Novak, Y. L. Radeva, H. U. Käufl, A. Smette, A. Tokunagaf, A. Khayat. T. Encrenaz, P. Hartogh A sensitive search for organics (CH4, CH3OH, H2CO, C2H6, C2H2, C2H4), hydroperoxyl (HO2), nitrogen compounds (N2O, NH3, HCN) and chlorine species (HCl, CH3Cl) on Mars using ground-based high-resolution infrared spectroscopy Icarus 223 (2013), pp. 11-27 Webster et al., 2013 C. R. Webster et al. Low Upper Limit to Methane Abundance on Mars Science 342 (2013), pp. 355-356
14
Acknowledgments: This research was supported by Villum Kann Rasmussen Foundation, the Danish National Research Council, the Danish Council for Independent Research, Natural Sciences and the Carlsberg Foundation. The authors wish to thank Dr. Morten Bo Madsen, University of Copenhagen, for useful discussions.
Supplementary Information. The solid-state
13
C and
29
Si MAS (magic-angle spinning) and CP/MAS NMR experiments were
performed on Varian INOVA 300 MHz (7.05 T) and 400 MHz (9.39 T) spectrometers, employing home-built CP/MAS NMR probes for 5 and 7 mm o.d. zirconia (PSZ) rotors. The
13
C{1H}
CP/MAS NMR spectra (7.05 T) were acquired with a 5-mm probe using spinning speeds of either νR = 3.0 kHz or 10.0 kHz, a 4-s relaxation delay, and rf field strengths of γB1/2π ≈ γB2/2π = 40 kHz for 13C and 1H during the CP contact and γB2/2π = 75 kHz for the initial 90o 1H pulse and 1H TPPM decoupling. The
13
C MAS NMR spectra acquired for the quantitative analysis used a
relaxation delay of 60 s. The 29Si MAS and 29Si{1H} CP/MAS NMR spectra (9.39 T) employed a 7mm probe, spinning speeds of νR = 6.0 kHz and 4.0 kHz, respectively, and γB1/2π ≈ γB2/2π = 36 kHz for the 29Si and 1H pulses in both types of experiments. 13C and 29Si chemical shifts are relative to neat Si(CH3)4 for both nuclei.
15
Mars wind erosion is simulated by tumbling quartz (SiO2) and olivine along with CH4 Tumbling of SiO2 and olivine along with CH4 leads to consumption of CH4 Solid-state 13C/29Si NMR of tumbled SiO2 material shows formation of Si-CH3 bonds The results offer an explanation for the fast disappearance of CH4 on Mars
S0019-1035(14)00162-6 http://dx.doi.org/10.1016/j.icarus.2014.03.036 YICAR 11018
To appear in:
Icarus
Received Date: Revised Date: Accepted Date:
11 September 2013 5 March 2014 15 March 2014
Please cite this article as: Knak Jensen, S.J., Skibsted, J., Jakobsen, H.J., ten Kate, I.L., Gunnlaugsson, H.P., Merrison, J.P., Finster, K., Bak, E., Iversen, J.J., Kondrup, J.C., Nørnberg, P., A sink for methane on Mars? The answer is blowing in the wind, Icarus (2014), doi: http://dx.doi.org/10.1016/j.icarus.2014.03.036
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
A sink for methane on Mars? The answer is blowing in the wind. Svend J. Knak Jensena,*, Jørgen Skibsteda, Hans J. Jakobsena, Inge L. ten Kateb, Haraldur P. Gunnlaugssonc, Jonathan P. Merrisonc, Kai Finsterd, Ebbe Bakd, Jens J. Iversenc, Jens C. Kondrupa, Per Nørnberge a
Instrument Center for Solid-State NMR Spectroscopy, Department of Chemistry, Interdisciplinary Nanoscience Center (iNANO), Langelandsgade 140, Aarhus University, DK-8000 Aarhus C, Denmark. b
Department of Earth Sciences, Utrecht University, Budapestlaan 4, 3584 CD Utrecht, The Netherlands.
c
Department of Physics and Astronomy, Ny Munkegade 120, Aarhus University, DK-8000 Aarhus C, Denmark. d
Department of Bioscience, Microbiology section, Ny Munkegade 114 – 116, Aarhus University, DK-8000 Aarhus C, Denmark.
e
Department of Geoscience, Høegh-Guldbergs Gade 2, Aarhus University, DK-8000 Aarhus C, Denmark.
Abstract Tumbling experiments that mimic the wind erosion of quartz grains in an atmosphere of enriched methane are reported. The eroded grains are analyzed by
13
C and
29
13
C-
Si solid-state NMR
techniques after several months of tumbling. The analysis shows that methane has reacted with the eroded surface to form covalent Si-CH3 bonds, which stay intact for temperatures up to at least 250 °C. The NMR findings offer an explanation for the fast disappearance of methane on Mars.
Keywords: Mars, Atmosphere; Aeolian processes; Atmospheres, Chemistry
*
Corresponding author: Svend J. Knak Jensen, Phone: +(45) 87155926, Fax: +(45) 8619 6199, Email: [email protected]
2
1. Introduction Recently methane (CH4) has been observed in the Martian atmosphere from a satellite orbiting the planet (Formisano et al., 2004) as well as from Earth based telescopes (Krasnopolsky et al., 2004). These observations are exciting as methane may be a fingerprint of past or present life on Mars. Currently, the Curiosity rover on Mars is exploring various issues related to methane. However, so far the source of methane has not been identified. Potential sources could be biological in origin, like methanogenesis in extreme environments (Reid et al., 2006) or geochemical processes, like serpentinisation (Oze and Sharma, 2005) or volcanism (Etiope et al., 2007) or methane release from clathrates (Chastain and Chevrier, 2007). Release of methane from such processes could occur as episoidic events. Likewise, the fate of methane is a mystery. Observations indicate that methane disappears within a few years after a plume eruption (Mumma et al., 2009 and Lefèvre and Forget, 2009), while the photochemical processes that operate on Earth predict a lifetime in the range of several hundred years (Mumma et al., 2009). Various mechanisms proposed for the disappearance of methane have difficulties reproducing its short residence time (ten Kate, 2010). Reported concentrations of methane are in the range of tens of ppbv. A significant feature of methane concentrations is that they show a substantial time and spatial variation (Geminale et al., 2008, Mumma et al., 2009, Lefèvre and Forget, 2009, Geminale et al., 2011, Krasnopolsky 2012). Here we show, using solidstate
13
C and
29
Si magic-angle spinning NMR spectroscopies, that wind driven erosion produces
highly reactive sites on mineral grain surfaces that sequester methane by forming covalent bonds with methyl groups and propose that this mechanism can be the hitherto undiscovered methane sink on Mars. The mechanism is supported by experiments where quartz (SiO2) grains are exposed to gentle mechanical agitation in a methane atmosphere, using a specially designed experimental setup, that mimics mineral grain transport created by a typical Martian wind. The predicted lifetime
3
of methane agrees well with the one experimentally observed. We anticipate that the designed agitation process can be useful in studies of surface erosion and surface chemistry. After this paper was submitted, in situ measurements by the Curiosity team found very little evidence, if any, for methane in the Gale crater on Mars (Webster et al., 2013). These findings are discussed in relation to our findings in Sec. 4. 2. Materials and methods. The wind driven erosion of surface material is simulated using the specially designed apparatus depicted in Fig. 1. Commercially available quartz (Merck, 1.07536) was chosen as an analogue for surface material because of its simple chemical composition. The quartz was placed in a borosilicate flask with
13
C-methane (Sigma-Aldrich, 490229, 99% enriched) to facilitate NMR
investigations. The basic purity of the 13C-methane gas was confirmed by 13C NMR experiments on the gas loaded in a glass tube. The flask with quartz and 13C-methane was sealed and mounted in a carousel where it was tumbled end-over-end at 30 RPM. The collision speed of the quartz grains as they tumble is about 1 m/s, which is a typical saltation collision speed close to threshold (HolsteinRathlou et al., 2010, Merrison et al., 2010 and Merrison, 2012). After several months of tumbling the quartz material was analyzed by solid-state NMR techniques (details of the NMR experiments are given in the Supplementary Information).
3. Results and discussion The reaction of
13
C-enriched methane with surface sites of highly active quartz particles is
unambiguously demonstrated by the 13C{1H} CP/MAS and 29Si{1H} CP/MAS NMR spectra shown in Figs. 2 and 3, respectively. In these spectra the cross-polarization (CP) NMR technique transfers 1
H magnetization to either the 13C or 29Si spins via heteronuclear dipolar couplings and thereby acts
as a filter for detecting only 13C and 29Si spin nuclei within a distance less than 3-5 Å to nearby 1H
4
nuclei. The
13
C{1H} CP/MAS NMR spectrum of the quartz grains agitated in an atmosphere of
methane, and stored under standard atmospheric and temperature conditions (Fig. 2A), exhibits a resonance at δ(13C) = –1.7 ppm, i.e., very similar to the
13
C chemical shift reported for Si–CH3
groups; for example, δ(13C) = 0.0 ppm for Si(CH3)4 (TMS). In addition to the 13C CP/MAS NMR spectrum, we note that several other
13
recovery T1(13C) measurements and
C MAS NMR experiments for the sample (e.g., inversion-
13
C{1H} cross-polarization – depolarization experiments)
reveal restricted rotation for the methyl group around the Si–CH3 bond. Samples from the methanequartz batch have been exposed to heat-treatment at 250 and 550 °C for one hour. The
13
C{1H}
CP/MAS NMR spectrum of the 250 °C sample (Fig. 2B) shows intact Si–CH3 groups while no 13C signal could be detected for the 550 °C sample. The line width of the resonance for the 250 °C sample (Fig. 2B) is slightly more narrow (FWHM = 5.5 ppm) compared to that shown in Fig. 2A for the original sample (FWHM = 7.7 ppm). The observed line-narrowing may be ascribed to annealing processes and/or to elimination of water dehydroxylated from the quartz surface. A second flask with quartz and
13
C-methane, prepared under identical conditions as used for the
sample described above, was not exposed to tumbling and kept stationary for the same period of time.
13
C{1H} CP/MAS NMR spectra of this sample show no
13
C NMR resonances at all, which
demonstrate that the reaction of methane with quartz requires activation of the quartz grains by the tumbling process. The standard one-pulse 29Si MAS NMR spectrum of the methane-quartz sample in Fig. 3B exhibits a narrow resonance (FWHM = 0.11 ppm) at δ(29Si) = –107.5 ppm, i.e., the well-known
29
Si
chemical shift for α-quartz (Lippmaa et al., 1980 and Smith and Blackwell, 1983), and thus is assigned to the bulk SiO2 structure of the sample. More importantly, the sample are selectively detected in the
29
29
Si surface sites of the
Si{1H} CP/MAS NMR spectrum (Fig. 3A), which reveals
two broadened resonances at –61 and –101 ppm. The high-intensity resonance at –101 ppm
5
originates from
29
Si sites associated with hydroxyl groups, following earlier
29
Si CP/MAS NMR
studies of silica gels. More importantly, only this resonance at –101 ppm is observed in a similar spectrum of pure quartz exposed to tumbling in ambient air under the same conditions as used for the SiO2/13CH4 sample. Most interestingly, from previous investigations of modified silica surfaces, HPLC materials, and heterogeneous catalysts (Albert and Bayer, 1991 and Kellberg et al., 1993) it is known that methyl groups directly bonded to a Si atom on a silica surface give 29Si resonances in distinct regions of the 29Si chemical shift scale according to the number of attached methyl groups. For example, for a (CH3)2Si(OSi)2 species: δ(29Si) = –14 to –20 ppm, while for CH3Si(OSi)3 species: δ(29Si) = –53 to –65 ppm. Thus, the observed resonance at –61 ppm can be assigned to a (SiO)3Si–CH3 site. This result and the absence of the resonance at –61 ppm for the tumbled sample of pure quartz present an unambiguous and direct proof that agitation of quartz in a methane atmosphere results in a methyl group being directly bonded to a Si atom on the surface of the quartz particles, according to the formal reactions:
(SiO)3Si–OH + CH4 → (SiO)3Si–CH3 + H2O
and/or the breakage of dehydroxylated surface (SiO)3Si–O–Si(OSi)3 linkages according to
(SiO)3Si–O–Si(OSi)3 + CH4 → (SiO)3Si–CH3 + HO–Si(OSi)3
We note that the tumbling of the quartz grains in the borosilicate glass flask leads to a minor degree of abrasion of the borosilicate glass. This is apparent from 11B MAS NMR spectra of the tumbled SiO2/13CH4 sample and a ground sample of the borosilicate glass. Both spectra show the exact same set of
11
B centerband resonances from trigonal (BO3) and tetrahedral (BO4) units from the glass
6
material. However, a quantitiative evaluation of the intensities in the two spectra reveal that the SiO2/13CH4 sample only includes 1 – 2 wt% of the borosilicate glass as an impurity. Moreover, a 29
Si MAS NMR spectrum of the pure borosilicate glass shows a broadened resonance from approx.
– 95 to –125 ppm, characteristic for glass materials, which thereby does not overlap with the resonance at –61 ppm from the (SiO)3Si–CH3 sites in the SiO2/13CH4 sample. In our investigation quartz is used as an analogue for general silicate minerals found on Mars in the airborne dust and on the surface rocks. Detailed chemical analyses of the Martian regolith (Gellert et al., 2004 and McSween, 2004) show the presence of olivines ((Mg,Fe)2SiO4), pyroxenes (diopside) (MgCaSi2O6), and plagioclases (NaAlSi3O8 – CaAl2Si2O8 solid-solution series), for which (SiO)3Si–O–Si(OSi)3 linkages are dominant chemical entities. However, the chemical analyses also show that the regolith has a typical FeO content of about 15-16 wt%. Martian regolith analogues with a similar composition can be found on Earth, but the high content of paramagnetic FeO would make NMR investigations of such samples impossible. Thus, the large content of different silicates mentioned above could act as alternative materials for a methane sink in the Martian winds. Evidence supporting this hypothesis comes from experiments where the pressure of the gas phase is monitored during the tumbling process. In addition to the SiO2/13CH4 experiments we also tumbled a sample of olivine (from the Spanish island of Tenerife) in ordinary CH4 (with an isotope distribution of about 12C = 99 % and 13C = 1 %). The rate of pressure change for olivine is lower than that for quartz, but not more than a factor of two. We note that 12CH4 may react faster than 13CH4 because the activation energy for breaking the 12C-H bond is lower than that for the 13CH bond, which in turn is due to the fact that the heavier isotope has a lower vibration frequency (Levine, 2002). However, the difference in reaction rates is expected to be small. The dominant gas in the Martian atmosphere is CO2 and one might wonder if this gas could also react with the active sites in the Martian dust and regolith and perhaps monopolize all active sites.
7
Preliminary tumbling experiments with SiO2/13CO2 and subsequent solid-state
13
C NMR analysis
show a room-temperature 13C resonance at δ(13C) = 125.6 ppm, which disappears after exposing the sample to 250 °C for one hour. This indicates a weaker interaction than in the case of methane. Moreover, the amount of dust and surface material is huge so there are enough active sites for methane to react with. Quantitative single-pulse well-defined second relaxed
13
13
C MAS NMR analysis of the original agitated quartz sample, using a
13
C intensity reference sample (hexamethylbenzene) and employing fully-
C MAS NMR spectra (without CP) for both samples, shows that one gram of quartz
tumbled for 115 days has established 1.6·10–5 moles of Si-CH3 bonds. Combining this number with a surface area of 22 m2/g, determined for a tumbled sample of pure quartz, indicates that roughly 3 – 4 % of the surface Si sites are present as (Si-O)3Si-CH3 sites. If these numbers are translated to Martian conditions, then a burst of 270 tons of methane (Krasnopolsky et al., 2004 and Krasnopolsky, 2006) will require only about 106 tons of activated surface material for methane to disappear from the atmosphere within about one third of a year. There is some debate regarding the actual amount of methane estimated by Krasnopolsky. Villanueva et al. estimate the amount to be at least 4500 tons (Villanueva et al., 2013). However, the amount of available surface material is still in big surplus. Also on Earth, eroded and atmosphere suspended silicates should be considered as a sink for atmospheric methane. In general, wind mediated chemistry could take place on all planets and moons with a solid surface and an atmosphere.
4. Recent Curiosity findings After the original version of this paper was submitted a highly interesting paper dealing with in situ concentration measurements of methane in the Gale crater on Mars (Webster et al., 2013) appeared.
8
Detailed recent snapshots measurements from this investigation on-board the Curiosity rover have shown that the concentration of methane is very low, i.e., 0.18 ± 0.67 ppbv. This finding underscores the challenging question of what happened to the methane observed by several groups a few years earlier. It is obvious that methane on Mars is a big issue as already two spacecraft (NASA's Mars Atmosphere and Volatile Evolution Mission, MAVEN, and ISRO's Mars Orbiter Mission, MOM) are on their way to the planet to (among other tasks) perform detailed measurements of methane levels in the atmosphere over long periods of time. Also ESA is planning to launch the ExoMars Trace Gas Orbiter in 2016 with a similar objective. Current evidence shows that methane levels have a substantial spatial and temporal variation. However, in the words of Webster et al. (Webster et al., 2013): The very short methane lifetime of 0.4 to 4 years derived from the 2003–2006 observations requires powerful destruction mechanisms that have not been identified to date. Our laboratory studies show that a wind mediated erosion process of ordinary quartz crystals can produce activated quartz grains, which sequester methane by forming covalent Si-C bonds. If this process is operational on Mars, which our recent preliminary studies on olivine indicate could be the case then it can explain the observed fast destruction of methane. Of course, there is a possibility that the methylated material is a temporary sink which may undergo chemical reactions with surface regolith, perhaps triggered by the abundant UV radiation. Therefore, we propose that in future Mars explorations it will be of high interest to look for chemical evidence of methylated materials – or downstream reaction products- in the Martial regolith and dust.
9
Figure 1: Schematic drawing of the tumbling apparatus.
The flask (borosilicate,Simax 3.3) contains about 10 grams of SiO2 grains, which have been sieved to extract the fraction between 125 and 1000 µm. This fraction was ultrasonic treated and washed several times to remove all attached finer particles. The gas in the flask is
13
C-enriched CH4 at a
pressure of approximately 600 mbar. The tumbling rate is 30 RPM. The whole tumbling apparatus is enclosed in a box to avoid any potential interference from photochemistry.
10
Figure 2: 13C{1H} CP/MAS NMR spectra.
Spectra of the quartz sample, agitated in an atmosphere of 99 % 13C-enriched CH4 for 115 days at room temperature. The spectra are acquired at room temperature and at 7.05 T using a CP contact time τCP = 3.0 ms. (A) Spectrum of the original sample using a spinning frequency νR = 10.0 kHz and acquisition of 18944 scans. (B) Spectrum of the original sample, after heat-treatment at 250 oC for 1 hr, using a spinning frequency νR = 3.0 kHz and acquisition of 1024 scans.
11
Figure 3: 29Si MAS and CP/MAS NMR spectra.
Spectra of the quartz sample, agitated in an atmosphere of 99 % 13C-enriched CH4 for 115 days at room temperature. The spectra are acquired at room temperature and at 9.39 T. The different highlighted environments for the central Si atom refer to their 29Si chemical shifts discussed in the text. (A) 29Si{1H} CP/MAS NMR spectrum acquired using a spinning frequency νR = 4.0 kHz, a CP contact time τCP = 4.0 ms, a relaxation delay of 4 s , and 30720 scans. (B) Single-pulse 29Si MAS NMR spectrum acquired using a spinning frequency νR = 6.0 kHz, a relaxation delay of 120 s, and 644 scans.
12
References Albert and Bayer, 1991 K. Albert, E. Bayer Characterization of bonded phases by solid-state NMR spectroscopy J. Chromatogr. 544 (1991), pp. 345-370
Chastain and Chevrier, 2007 B.K. Chastain, V. Chevrier Methane clathrate hydrates as a potential source for Martian atmospheric methane Planet. Space Sci. 55 (2007), pp. 1246–1256 Etiope et al., 2007 G. Etiope, T. Fridriksson, F Italiano, W. Winiwarter, J. Theloke Natural emissions of methane from geothermal and volcanic sources in Europe J. Volcanol. Geoth. Res. 165 (2007), pp. 76-86 Formisano et al., 2004 V. Formisano, S. Atreya, T. Encrenaz, N. Ignatiev, M. Giuranna Detection of methane in the atmosphere of Mars Science 306 (2004), pp. 1758-1761 Gellert et al., 2004 R. Gellert, et al. Chemistry of rocks and soils in Gusev Crater from the Alpha Particle X-ray spectrometer Science 305 (2004), pp. 829-832 Geminale et al., 2008 A. Geminale, V. Formisano, M. Giuranna Methane in Martian atmosphere: Average spatial, diurnal, and seasonal behavior Planet. Space Sci. 56 (2008), pp. 1194-2003 Geminale et al., 2011 A. Geminale, V. Formisano, G. Sindoni Mapping methane in Martian atmosphere with PFS-MEX data Space Sci. 59 (2011), pp. 137-148 Holstein-Rathlou et al., 2010 C. Holstein-Rathlou, et al. Winds at the Phoenix landing site J. Geophys. Res., 115 (2010), p. E00E16 Kellberg et al., 1993 L. Kellberg, P. Zeuthen, H. J. Jakobsen Deactivation of HDT calatalysts by formation of silica gels from silicone oil. Characterization of spent catalysts from HDT of coker naphta using 29Si and 13C CP/MAS NMR J. Catal. 143 (1993), pp. 45-51 Krasnopolsky et al., 2004 V. Krasnopolsky, J. Maillard, T. Owen Detection of methane in the Martian atmosphere: evidence for life? Icarus 172 (2004), pp. 537-547 Krasnopolsky, 2006 V. Krasnopolsky Some problems related to the origin of methane on Mars Icarus 180 (2006), pp. 359-367 Krasnopolsky, 2012 V. Krasnopolsky Search for methane and upper limits to ethane and SO2 on Mars Icarus 217 (2012), pp. 144-152
13
Lefèvre and Forget, 2009 F. Lefèvre, F. Forget Observed variations of methane on Mars unexplained by known atmospheric chemistry and physics Nature 460 (2009), pp. 720-723 Levine, 2002 I. R. Levine Physical Chemistry McGraw-Hill (2002), pp. 903 Lippmaa et al., 1980 E. Lippmaa, M. Mägi, A. Samoson, G. Engelhardt and A. R. Grimmer Structural studies of silicates by solid-state high-resolution 29Si NMR J. Am. Chem. Soc., 102 (1980), pp. 4889-4893 McSween, 2004 H. Y. McSween Basaltic rocks analyzed by the Spirit Rover in Gusev Crater Science 305 (2004), pp. 842-845 Merrison et al., 2010 J. P. Merrison, H. P. Gunnlaugsson, S. J. Knak Jensen, P. Nørnberg Mineral alteration induced by sand transport; a source for the reddish color of Mars Icarus, 205(2) (2010), pp. 716-718 Merrison, 2012 J. P. Merrison Sand Transport, Erosion and Granular Electrification Aeolian Research 4 (2012), pp. 1–16 Mumma et al., 2009 M. J. Mumma, et al. Strong release of methane on Mars in Northern Summer Science 323 (2009), pp. 1041 -1045 Oze and Sharma, 2005 C. Oze, M. Sharma Have olivine, will gas: Serpentinization and the abiogenic production of methane on Mars Geophys. Res. Lett. 32 (2005) L10203 Reid et al., 2006 I.N. Reid et al., Terrestrial models for extraterrestrial life: methanogens and halophiles at Martian temperatures Int. J. Astrobio. 5 (2006) 89-97 Smith and Blackwell, 1983 J. V. Smith, C.S Blackwell Nuclear magnetic resonance of silica polymorphs Nature 303 (1983), pp. 223-225 ten Kate, 2010 Inge L. ten Kate Organics on Mars? Astrobiology 10 (2010), pp. 589-603 Villanueva et al., 2013 G. L. Villanueva, M. J. Mumma, R. E. Novak, Y. L. Radeva, H. U. Käufl, A. Smette, A. Tokunagaf, A. Khayat. T. Encrenaz, P. Hartogh A sensitive search for organics (CH4, CH3OH, H2CO, C2H6, C2H2, C2H4), hydroperoxyl (HO2), nitrogen compounds (N2O, NH3, HCN) and chlorine species (HCl, CH3Cl) on Mars using ground-based high-resolution infrared spectroscopy Icarus 223 (2013), pp. 11-27 Webster et al., 2013 C. R. Webster et al. Low Upper Limit to Methane Abundance on Mars Science 342 (2013), pp. 355-356
14
Acknowledgments: This research was supported by Villum Kann Rasmussen Foundation, the Danish National Research Council, the Danish Council for Independent Research, Natural Sciences and the Carlsberg Foundation. The authors wish to thank Dr. Morten Bo Madsen, University of Copenhagen, for useful discussions.
Supplementary Information. The solid-state
13
C and
29
Si MAS (magic-angle spinning) and CP/MAS NMR experiments were
performed on Varian INOVA 300 MHz (7.05 T) and 400 MHz (9.39 T) spectrometers, employing home-built CP/MAS NMR probes for 5 and 7 mm o.d. zirconia (PSZ) rotors. The
13
C{1H}
CP/MAS NMR spectra (7.05 T) were acquired with a 5-mm probe using spinning speeds of either νR = 3.0 kHz or 10.0 kHz, a 4-s relaxation delay, and rf field strengths of γB1/2π ≈ γB2/2π = 40 kHz for 13C and 1H during the CP contact and γB2/2π = 75 kHz for the initial 90o 1H pulse and 1H TPPM decoupling. The
13
C MAS NMR spectra acquired for the quantitative analysis used a
relaxation delay of 60 s. The 29Si MAS and 29Si{1H} CP/MAS NMR spectra (9.39 T) employed a 7mm probe, spinning speeds of νR = 6.0 kHz and 4.0 kHz, respectively, and γB1/2π ≈ γB2/2π = 36 kHz for the 29Si and 1H pulses in both types of experiments. 13C and 29Si chemical shifts are relative to neat Si(CH3)4 for both nuclei.
15
Mars wind erosion is simulated by tumbling quartz (SiO2) and olivine along with CH4 Tumbling of SiO2 and olivine along with CH4 leads to consumption of CH4 Solid-state 13C/29Si NMR of tumbled SiO2 material shows formation of Si-CH3 bonds The results offer an explanation for the fast disappearance of CH4 on Mars