Eur. Phys. J. Plus (2017) 132: 346 DOI 10.1140/epjp/i2017-11637-y

THE EUROPEAN PHYSICAL JOURNAL PLUS

Review

Did Viking discover life on Mars? Cesare Guaitaa GAT/Milano Planetarium - Cs Venezia 67, 20121 Milano, Italy Received: 10 July 2017 c Societ` Published online: 10 August 2017 –  a Italiana di Fisica / Springer-Verlag 2017 Abstract. The last analytical results of SAM laboratory onboard of Curiosity offer a new insight on the interpretation of the Label Release (LR) experiment performed on Mars by the two Viking landers about 40 years ago. The fundamental action of perchlorate salt, able to decompose all organic compounds at high temperature (> 300 ◦ C) is discussed.

1 Introduction First direct search for traces of present or past biological life on Mars, was set up by the two Viking landers in the 1970 s [1]. The results were so contentious that, after more that 35 years, no unambiguous interpretation was found. The GC-MS instruments (Gas Chromatograph-Mass Spectrometer) onboard of both Viking landers [2,3] were tasked with detecting organic compounds. GC-MS instruments heated many samples of Martian soil, but they did not detect any trace of complex organic molecules, even if they detected an amount of 0.1–1% H2 O and 50–500 ppm CO2 , respectively, (table 1), and the enigmatic release of about 15 ppb of CH3 Cl (chloro-methane) and up to 20–30 ppb of CH2 Cl2 (methylene chloride) (fig. 1) [4]. At that time, the two light chloro-derivatives, being released together with some traces of a solvent of sure terrestrial origin, such as freon-E, were considered as a terrestrial contamination, ruling out the occurrence of any form of Martian life [4]. In the meantime, the H2 O and CO2 release upon heating were explained as thermal decomposition of hydrous silicates and carbonates, respectively. The measured ratio, 37 Cl/35 Cl = 0.319, similar to that of the terrestrial chloride [4] supported this interpretation. However, Sharp [5] found that the ratio 37 Cl/35 Cl is quite constant all over the Solar System: actually its value is the same on Earth, in the chloro-salts enclosed inside the carbonaceous chondrites and also on some Martian meteorites. Moreover, a suitable inquiry proved that CH3 Cl and/or CH2 Cl2 were never used during the Viking assembly, so any onboard trace was impossible [6]. We could rediscuss the conclusion taken from the Viking GC-MS results on the basis of two reasons. The first reason is linked to the fact that the Viking LR (Labeled Release) experiment, made at ambient temperature (10–15 ◦ C), gave a substantially positive biological result [7,8]. However, it was the only experiment with a clearly positive response, whereas the other two “biological” experiments, i.e. the Gas Exchange Esperiment [9] and the Carbon Assimilation Experiment [10], gave dubious results, suggesting a lack of biological reactions. The LR experiment was based on the well-known concept that all terrestrial microorganisms metabolize organic substances releasing CO2 . In the Viking LR experiment, the landers collected samples of Martian soil by means of their robotic arm, injected them with a drop of dilute nutrient solution containing alanine, formic acid, glycine, glycolic acid and lactic acid, and then monitored the air above the soil for signs of metabolic by-products. Since the nutrients were tagged with radioactive carbon-14, if microorganisms in the soil metabolized the nutrients, they would be expected to produce radioactive by-products, such as radioactive carbon dioxide (14 CO2 ). 14 CO2 was indeed released when an aqueous solution of 14C labelled amminoacids was added, but a much lower amount (i.e. by one order of magnitude) was released in the case of terrain samples sterilised at 160 ◦ C (fig. 2). To rule out the possibility that the strong ultraviolet radiation on Mars might be causing positive results, the landers collected also soil buried underneath a rock, which again tested positive. The control tests also worked, with the 160 ◦ C sterilization control, yielding negative results [11].  a

Contribution to the Focus Point on “Highlights of Planetary Science in Italy” edited by P. Cerroni, E. Dotto, P. Paolicchi. e-mail: [email protected]

Page 2 of 9

Eur. Phys. J. Plus (2017) 132: 346

Table 1. Results of GC-MS onboard of Viking landers, under heating of Martian soil samples up to 500 ◦ C (adapted from [4]). Sampling site

T (◦ C)

Carrier gas

H2 O (%)

CO2 (ppm)

VL-I* Oven 1 (cruise)

500

13

CO2

< 0.1

Sandy Flats

200

13

CO2

< 0.1

500

13

CO2

0.1–1.0

350

13

CO2

0.1–1.0

500

13

CO2

0.1–1.0

500

13

CO2

0.1–1.0

Rocky Flats

VL-2 Bonneville

200

H2

0.05

< 50

350

H2

0.3

50–500

500

H2

1.0

50–500

CO2

0.25

500

13

Under

50

H2

< 0.01

< 50

Badger Rock

200

H2

0.2

50–500

350

H2

0.3

40–400

H2

0.8

70–700

CO2

0.6

500 500

13

In 2002 a possible circadian fashion (i.e. having the same periodicity of the Martian day) of 14 CO2 release was found, which may be a typical biological signature [12]. A complex statistical analysis [13] reached the same conclusion. In any case, it is important to underline that Levin and Straat, the principal investigators of the LR experiment, observed that the release of radioactive CO2 could be due also to non-biological reactants [8]; a real possibility discussed also by Klein [14]. On this subject some lab tests were performed, assuming that some Fe superoxides are built up in the Martian soil by the strong UV radiation. This Fe superoxide could decompose (with release of 14 CO2 ) the carbon molecules of the LR nutrient solution directly [15] or through the formation of H2 O2 [16]. The other reason why we re-discuss the negative response of the two GC-MSs onboard of the Viking landers is the following: the Martian soil, in any case, should be enriched by organic (not biological) molecules that could be taken by comets and carbonaceous chondrites. The recent discovery of simple and polymeric organic substances inside ten Martian meteorites could be evidence in this regard [17]. A possible explanation of the results of the Viking analyses was found by the Wet Chemistry Laboratory on the Phoenix Mars lander which, in the summer of 2008, discovered up to 0.6% of magnesium perchlorate [Mg-(ClO4 )2 ] in the North polar sands of Mars [18]. This salt is inert at low temperature, but at high temperature becomes a strong oxidant able to decompose all carbon compounds [18]. So, if we suppose that also the soil sampled by Viking were rich in perchlorates, the GC-MS analysis, being performed at 500 ◦ C, possibly could result in the demolition of all organic molecules (biological or not) during the same analytical process [19]. The assumption of the presence of perchlorates at the Viking landing sites might seem a little hasty, because Hechte [18] stated that perchlorate may form preferably at high latitudes, whereas the Viking 1 landing site was at equatorial latitudes, and the Viking 2 one at intermediate latitudes. However, we could not exclude this possibility [20], specially after the discovery of perchlorates inside the Gale Crater (Lat= 5.24◦ S), landing site of Curiosity [21], and the discovery of perchlorates also in some Martian meteorites [22]. In the case of analyses performed at low temperature, perchlorates are totally inert and, so, a positive response, as observed by the Viking LR experiment, may really suggest the presence of organic substances.

Eur. Phys. J. Plus (2017) 132: 346

Page 3 of 9

Fig. 1. Results of GC-MS experiments on board the Viking landers. On the x-axis the so-called retention time is plotted, i.e. the time (in seconds) that each molecule takes to get out of GC column. On the y-axis the electronic response of the MS (proportional to the amount of a single molecule eluted by the GC) is plotted. Under heating of the Martian soil at 200 ◦ C in a suitable oven, the GC-MS of Viking landers showed release of CHCl3 (molecular weight = 50) from the soil sample taken at Cryse (Viking 1) and CH2 Cl2 (molecular weight = 84) from the soil sample taken at Utopia (Viking 2). These chloro-compounds are of unclear origin. A GC-MS run on empty oven did not show any release of chloro-compounds (see the top panel), so demonstrating that CHCl3 and CH2 Cl2 do not arise from some kind of terrestrial contamination.

2 Viking and perchlorates A further support to a positive interpretation of the Viking LR experiment was given by Navarro-Gonzales [23]. In summary, a sample of a Mars-like soil of the driest core of the Atacama desert in Northern Chile (the Yungay area), containing very low organic concentration (32 ppm), was subjected to a thermal volatilization process. The released gases and volatiles have been then measured by a GC-MS similar to the Viking ones (fig. 3). At 500 ◦ C a clear emission of organic substances, such as benzene, toluene, formic acid, was observed. But when the same soil was heated at 500 ◦ C after the addition of 1% of Mg perchlorate, the mentioned organic substances were no longer observed, whereas a release of CO2 and H2 O and, amazingly, also of CH3 Cl and CH2 Cl2 , was measured. According to Navarro-Gonzales [23] the CH3 Cl and CH2 Cl2 release was ascribed to a reaction between perchlorate and organics. However, Biemann and Bada [24] strongly questioned this conclusion for many reasons, such as the doubtful presence of perchlorate at the Viking landing sites, the use, by Navarro-Gonzales, of equipment and mode of operation quite different from that of Viking, which precluded the detection of some chlorinated aromatics (such as chlorobenzene and chlorotoluene), which should have substantiate the thermal action of perchlorate [25]. The research of Navarro-Gonzales [23] seems to demonstrate that the emission of CH3 Cl and CH2 Cl2 may be an organic signal, linked to the presence of perchlorate salts in the two Viking landing sites, even if Biemann and Bada paper [24] gives a different interpretation. Unfortunately, the two Vikings were not able to search for perchlorates. But it is possible to “read” the potential presence of these salts in some meaningful clues. However, the RXFS (X-Ray Fluorescence Spectrometer) onboard of the Viking lander was suitable to search for Cl (chlorine) in Martian soil, finding similar values: Viking 1 found 0.8% of Cl on the landing site of Cryse (22.7◦ N, 48.2◦ W) and Viking 2 found about 0.4% of Cl on the landing site of Utopia (48.3◦ N, 226◦ W) [24,26]. Furthermore

Page 4 of 9

Eur. Phys. J. Plus (2017) 132: 346

Fig. 2. Example of the results of Viking LR (Labeled Release) experiment, on active (top) and sterilized (bottom) Martian soil. On the x-axis Martian days (sol) are plotted, on the y-axis the electronic response of the detector (counts/min) to the radioactive 14 CO2 , released during the experiment is plotted. Take very well into account the scale difference between the two plot.

Pathfinder (1997) found up to 1% of Cl on Ares Valley (19.3◦ N, 33.6◦ W) [27], Spirit found about 0.5% of Cl inside the Gusev crater (14.6◦ N, 175.5◦ E) [28] and Opportunity found up to 1% of Cl on Meridiani Planum (1.9◦ S, 354.5◦ E) [29]. The kind of compound containing Cl should be investigated. After the unexpected discovery of 0.4–0.6% perchlorate ions by mass, with Mg, Na and Ca being the dominant cations, on the Martian polar soil (68.3◦ N, 127.0◦ W) performed by the Phoenix lander [18], with only traces (0.02%) of other salt containing Cl, the quite abundant amount of Cl found by Vikings appears as a strong indication of the presence of perchlorates. The mechanism of Martian perchlorate production is still being debated. It has been suggested that production pathways for perchlorate on Mars are similar to those on Earth, primarily photochemically in the upper atmosphere via oxidation of chlorine by ozone [30]. But, because of the low amount of ozone in the Martian atmosphere, mechanisms involving surface components are probable [31]. For example, perchlorates may form from the radiolysis of surface component caused by galactic cosmic rays, causing a sublimation of chlorine oxide in the atmosphere, where final oxidation to perchloric acid is performed by some sources of active oxigen (i.e. O3 and/or CO2 photolysis) [32]. Moreover, in the presence of a suitable catalist, such as TiO2 , the strong Martian UV illumination could oxidize chloride ions to perchlorate also in aqueous solutions [33]. The permanence of perchlorates (very soluble in water) on the Martian soil is made possible by the strong ambient dryness: possibly, Mars lacks rains able to dissolve perchlorates for millions of years. The logical interpretation of the Navarro-Gonzales results [23] on the Atacama soil starts from the well-known decomposition of Mg(ClO4 )2 at temperature > 400 ◦ C [34], with release of O2 and Cl: 2Mg(ClO4 )2 → MgO + MgCl2 + 15/2O2 + 2Cl. O2 and Cl react with organics compounds, releasing, on one side, H2 O and CO2 , and, on the other side, the chlorine compounds observed by the Viking GCMS. The results of Phoenix and Atacama analyses, suggest to reconsider methods for searching carbon molecules on Mars, taking into account the significant risk arising from the thermal methods.

Eur. Phys. J. Plus (2017) 132: 346

Page 5 of 9

Fig. 3. Release of gases at different temperatures on the experiment of Navarro-Gonzales et al. [23] on a Martian analogue (Yungay soil) with a CG-MS similar to the Viking one. Under heating up to 500 ◦ C with a GC-MS similar to that of Viking, the Yungay soil (Atacama desert) showed (top) an emission of many organic substances, possibly from bacterial decomposition (blue curves) and no chloro-compounds (red curves). After the addition of perchlorate (bottom) the organic substances (blue curves) are not longer observed, and CHCl3 e CH2 Cl2 were released (red curves).

3 Curiosity and SAM results The first chance for to this new approach came with the Curiosity mission [35], a rover of 900 kg that landed successfully on August 6, 2012 inside the Martian Gale crater (5.4◦ S 137.8◦ E) at a lower latitude than Viking (Cryse at 22.7◦ N and Utopia at 48.3◦ N). SAM (Sample Analysis at Mars) is aimed at analyzing of soil samples on board of Curiosity. It includes an improved and more sensitive (up to 100 times) version of the Viking GC-MS [36], and a laser infrared spectrometer (TLS) [37] able to analyse any gaseous substance from both Martian atmosphere and GC-MS. The TLS is able to detect the IR absorption band of CH4 at 3.27 micron with a sensitivity of 1 ppb (part per billion) and was aimed at confirming the existence and the seasonal cycle of methane discovered by terrestrial telescopes [38] and possibly confirmed by PFS spectromenter onboard of Mars Express [39]. The task of TLS is to continue the search for methane in order to establish its source (geological or biological). Until April 2013 SAM made two complete analyses on the Gale crater soil, the first in October 2012 and the second in February/March 2013 [40], probably discovering for the first time Martian organic molecules. Between sol 56 and 100 (October 2 to November 16, 2012) Curiosity reached the sandy terrain of Rocknest, located about 550 meters away from the landing site (fig. 4). The APXS instrument (Alpha Particle X-rays spectrometer) [41] detected, on Rocknest, a little amount of S and Cl [42]. The sandy texture of the soil was suitable to be easily transferred inside the SAM. Under the heating of the sample up to 800 ◦ C, many kinds of gaseous substances were released [43]. The release of molecular oxigen (O2 ) at 300–400 ◦ C (fig. 4) was very important: together with the presence of Cl, this emission is a suggestion of Ca (ClO4 )2 (calcium perchorate), a salt that decomposes under heat just to this temperature. Laboratory-based TGA (thermal gravimetrical analysis performed by the author on a Perkin-Elmer TGA 7 instrument) on synthetic perchlorates clearly shows that the calcium perchlorate starts to release molecular oxigen at 350 ◦ C, leaving a main residue of calcium chloride (CaCl2 ) (fig. 5). Therefore, after the discovery of perchlorate at high latitude by Phoenix, SAM demonstrated an occurrence of perchlorate also at equatorial latitude: so its occurrence also at midlatitude (i.e. Viking landing sites)

Page 6 of 9

Eur. Phys. J. Plus (2017) 132: 346

Fig. 4. The SAM-GC-MS analyses on Rocknest soil (four sample taken, as shown by the numbers at left). On the y-axis the electronic response of the MS is plotted.

could not be excluded. Actually, between 200 and 500 ◦ C, the soil of Rocknest released water and two peaks of CO2 (i.e. two releases at two different temperatures) (fig. 4). The origin of this water and carbon dioxide is doubtful. Being released at more than 200 ◦ C, the water cannot be free, but bound to soil minerals as water of crystallization. In addition, a lab simulation shows that the two peaks of CO2 could arise from the thermal decomposition of Mg and Fe carbonate [43]. But alternative hypotheses could have been given. The water and carbon dioxide seen by SAM could be breakdown products of organic substances under the action of perchlorates. This claim results from another discovery of GC-MS onboard of SAM: the detection of simple chlorinated molecules, such as CH3 Cl and a minor amount of CH2 Cl2 and CHCl3 [21]. At the end of February 2013 the SAM made a second series of analyses on a powdered sample of a sedimentary terrain named John Klein, located about 50 meters away from Rocknest (fig. 6), confirming results of first analysis, i.e. emission of CO2 and H2 O, of O2 over 250 ◦ C (probably generated by perclorates dissociation), and release of CH3 Cl + CH3 Cl2 [40]. Therefore, the SAM and Viking GC-MS results could be in agreement and a sufficient amount of perchlorates could mask occurrence of organics. To overcome this issue, a 15% of SAM analyses (i.e. on a set of 9 sample thimbles, over a total of 68) can be performed in the next months at lower temperature after a silanization of the supposed Martian organic substances [44]. Silanization is a derivatization reaction that increases so much the volatility, that a GC-MS analysis at low temperature becomes possible, so avoiding the interference of perchlorates. In this case, the silanizing agent is an organic compound containing florine and silicium, named MTBSTFA (N-tert-butyldimethylsilyl- N-methyltrifluoroacetamide), able to instantly replace active hydrogens on OH and NH2 (carboxylic acid, amine, amino-acid) with a N-tert-butyldimethylsilyl group: this non-polar moiety increases the volatility of the original compound by removing its polar nature, resulting in a much lower temperature needed for a GC-MS analysis [45].

Eur. Phys. J. Plus (2017) 132: 346

Page 7 of 9

Fig. 5. A Laboratory-based TGA (thermal gravimetric analysis) on two different perchlorates, performed by the author by a Perkin-Elmer TGA 7 instrument. The Instrument measures the loss of weight (y-axis) of a sample under heating up to 600 ◦ C (x-axis). A loss of weight with release of O2 begins at a temperature that is distinctive of each perchlorate, as clearly indicated by the blue arrows.

Fig. 6. The SAM-GC-MS analyses on John Klein soil. On the y-axis the electronic response of the MS, proportional to the amount of each molecule, is plotted. On the x-axis of the bottom plot the retention time is reported, i.e. the time (here in seconds) each molecule takes to go out off the CG column, whereas, on the x-axis of the top plot, temperature is reported.

Page 8 of 9

Eur. Phys. J. Plus (2017) 132: 346

Fig. 7. The SAM-GC-MS analyses on three sample of Cumberland soil. The release of Chloro-benzene (bottom) was a clear indication of local organic material. The meaning of the x- and y-axis is the same as in figs. 4 and 6.

During the examination of the results, the SAM obtained on Rocknest and John Klein, the SAM team discovered that a vial of MTBSTFA was broken, so polluting all the analytical system. The problem was that MTBSTFA, being itself an organic compound, reacts under heat with perchlorates, giving the same kind of light chloro-derivatives (CH3 Cl and CH2 Cl2 ) found by the SAM on the Martian samples! From here a dreadful doubt that the origin of the “positive” results obtained so far by the SAM could be “terrestrial” and not Martian [21]. Only after having removed all traces of MTBSTFA, the SAM team started again its analytical work, on a soil sample of the site of Cumberland, that was taken on May 2013, not far from John Klein. The SAM results were crucial [46]: aside from the usual light chloro-derivatives, also an abundant release (about 250 ppb) of chloro-benzene was detected (fig. 7). Lab tests demonstrated that chloro-benzene, an organic compound containing 6 carbon atoms, could not be formed from the heating of MTBSTFA in the presence of perchlorates [47], but only when various types of organic materials (from bacteria or from a meteoric extract) are pyrolysed (i.e. heated at high temperature) in the presence of a chlorine source.

4 Conclusions The discovery of organic molecules on Mars is a necessary but not sufficient condition for the demonstration of some present or past form of Martian bacterial life. Comets and the carbonaceous chondrites are indeed copious sources of organic material. But the positive results of the “famous” Viking LR experiment are intriguing, even because it is important taking into account that Viking LR was performed at low temperature, a condition where possible perchlorates do not show any oxidant effects. The biological explanation of the positive LR results might allow associating the emission of methane detected by the TLS onboard SAM [48], currently not ascribed to a known source, to some kind of methanogenic bacteria.

Eur. Phys. J. Plus (2017) 132: 346

Page 9 of 9

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48.

G.A. Soffen, J. Geophys. Res. 82, 3959 (1977). K. Biemann et al., Science 194, 72 (1976). H. Klein et al., The search for extant life on Mars, in Mars (University of Arizona Press, 1992) pp. 1221–1233. K. Biemann et al., J. Geophys. Res. 82, 4641 (1977). Z.D. Sharp et al., Nature 446, 1062 (2007). W. Bains, Martian Methyl Chloride. A lesson in uncertainty, arXiv:1304.4429 (2013). G. Levin, P. Straat, Origins Life 7, 293 (1976). G. Levin, P. Straat, Science 194, 1322 (1976). V. Oyama, B. Bendahi, J. Geophys. Res. 82, 4669 (1977). N.H. Horowitz et al., J. Geophys. Res. 82, 4659 (1977). G. Levin, P. Straat, J. Geophys. Res. 82, 4663 (1977). J.D. Miller et al., SPIE Proc. 4495, 96 (2002). G. Bianciardi et al., Int. J. Aeronaut. Space Sci. 13, 14 (2012). H. Klein, J. Geophys. Res. 82, 4677 (1977). A.S. Yen et al., Science 289, 1909 (2000). C. Ponnamperuma et al., Science 197, 455 (1977). A. Steele et al., Science 337, 221 (2012). M. Hecht et al., Science 325, 64 (2009). C. Guaita, Phoenix e la scoperta dei perclorati, in Astronomia, N. 6 (UAI, 2009) pp. 4-8. K. Biemann, Proc. Natl. Acad. Sci. 104, 10310 (2007). D.P. Glavin et al., J. Geophys. Res. Planets 118, 1 (2013). S.P. Kounaves et al., Icarus 229, 206 (2014). R. Navarro-Gonzales et al., J. Geophys. Res. 115 (2010) DOI: 10.1029/2010JE003599. K. Bieamnn, J.L. Bada, J. Geophys. Res. 116, E12001 (2011). H. Steininger et al., Planet. Space Sci. 71, 9 (2012). B.C. Clark et al., J. Geophys. Res. 87, 10059 (1982). C.N. Foley et al., J. Geophys. Res. 108, 8096 (2003). R. Gellert et al., The Alpha-Particle-X-ray-Spectrometer (APXS) for the Mars Science Laboratory (MSL) Rover Mission, in 40th Lunar and Planetary Science Conference (LPI, 2009) ID: 2364. R. Rieder et al., Science 306, 1746 (2004). D.C. Catling et al., J. Geophys. Res. 115, E00E11 (2010). B.L. Carrier, S.P. Kounaves, Geophys. Res. Lett. 42, 3739 (2015). E. Wilson et al., J. Geophys. Res. Planets 121, 1472 (2016). J.D. Schuttlefield et al., J. Am. Chem. Soc. 133, 17521 (2011). G.G. Marvin, L.B. Woolaver, Ind. Eng. Chem. Anal. 17, 474 (1945). NASA, Mars Science Laboratory Press Kit (2012) pp. 1–61. P.R. Mahaffy et al., Space Sci. Rev. 170, 401 (2012). C.G. Tarsitano, C.R. Webster, Appl. Opt. 46, 6923 (2007). M.J. Mumma et al., Science 323, 1041 (2009). V. Formisano et al., Science 306, 1758 (2004). D.W. Ming et al., Science 343, 1245267 (2014). R. Gellert et al., Science 305, 829 (2004). D.F. Blake et al., Science 341, 1239505 (2013). L.A. Leshin et al., Science 341, 1238937 (2013). P. Voosen, Science 355, 444 (2017). H. Kataoka, J. Cromarogr. 733, 19 (1996). C. Freissinet et al., J. Geophys. Res. Planets 120, 495 (2015). K.E. Miller et al., J. Geophys. Res. Planets 121, 296 (2016). C. Webster et al., Science 347, 417 (2015).