MEDDELANDENfrån

STOCKHOLMS UNIVERSITETS INSTITUTIONför

GEOLOGI OCH GEOKEMINo. 323

The Search for Life on Mars – Preparation for Sample Return

Tomas Hode

Stockholm 2005

Department of Geology and GeochemistryStockholm UniversitySE-106 91 Stockholm

Sweden

A dissertation for the degree of doctor of philosophy in Natural Science

Department of Geology and GeochemistryStockholm UniversitySE-106 91 StockholmSweden

Abstract

The purpose of the thesis work has been to develop methods and concepts to aid in the search, detection and assessment of ancient microfossils here on Earth as a guide to the search for ancient life on Mars. The intention has been to identify and characterize environments on Earth that may be considered analo-gous to Martian environments and in which fossil preservation is expected to be good, and to develop and apply methods to characterize the isotopic and chemical composition of possible traces of life in order to assess their biogenicity and biological affinities. An investigation of the Siljan impact structure, Sweden, demonstrated that niches for thermo-philic organisms were created in the associated hydrothermal system. The temperature regimes were favorable for thermophilic life in the outer parts of the structure during the early and main stages of the hydrothermal system, but that these niches moved toward the center of the crater during the final cooling stages. It was demonstrated that the hydrothermal system contains traces of a thermophilic microbial community, represented by fossilized extracellular polymeric substances (EPS). Given the presence of water on Mars, similar impact-induced hydrothermal systems were probably generated on Mars as well. These regions, like those at Siljan, may have supported hyperthermophilic microbial communities on the red planet, emphasizing the relevance of searching for impact-induced hydrothermal deposits for evidence of microbial life on Mars. A method for the determination of stable carbon isotopes with high lateral resolution of TEM (transmission electron microscopy) samples has been developed. The method is based on alpha-particle Rutherford backscattering (RBS), it is non-destructive, and therefore suitable for analysis of extrater-restrial and other rare or irreplaceable material. Also, a novel concept to extract fluid inclusions without ablating the sample has been proposed, and a proof-of-concept has been demonstrated. The purpose is to analyze organic biomarkers trapped in fluid inclusions without risking contamination, and also to extract and analyze single fluid inclusions. The minimized contamination risk and the potential to ex-tract single fluid inclusions could make the method a useful tool in the search for organic biomarkers in early-Earth material, and eventually, in samples returned from Mars.

© Tomas HodeISBN 91-7155-086-0ISSN 1101-1599Cover: Hydrothermal calcite vein in the Siljan impact structure. Photo: Stefan BengtsonPrint: Tryck & Rit AB, Stockholm

The Search for Life on Mars – Preparation for Sample Return

Tomas Hode

Swedish Museum of Natural History, department of Paleozoology, Box 50007, S-10405 Stockholm Sweden

This doctoral thesis consists of a summary and of the following papers:

Paper IHode, T., von Dalwigk, I., Broman, C., (2003). A hydrothermal system associated with the Siljan Impact Structure, Sweden – Implications for the search for fossil life on Mars. Astrobiology 3(2), pp. 271-289. Reprinted with permission.

Paper IIHode, T., von Dalwigk, I., Cady, S.L., Kristiansson, P., Evidence of Ancient Microbial Life in a Large Impact Structure. Submitted to Geology.

Paper IIIKristiansson, P., Hode, T., Auzelyte, V., Elfman, M., Malmqvist, K., Nilsson, C., Pallon, J., Shariff, A., Wegdén, M. (2004). Development of a system for determination of the 13C/12C isotopic ratio with high spatial resolution. Nuclear Instruments and Methods in Physics Research B 219-220, pp. 561-566. Reprinted with permission.

Paper IVHode, T., Kristiansson, P., Hugo, R.C., Cady, S.L., Development of a non-destructive micro-analyti-cal method for stable carbon isotope analysis of Transmission Electron Microscopy (TEM) samples. Submitted to Astrobiology.

Paper VHode, T., Zebühr, Y., Broman, C., Controlled fluid inclusion extraction. Submitted to Planetary and Space Science.

Per Kristiansson co-authored papers III and IV, and took part in the development of the method pre-sented in these papers. He also collaborated on the RBS analysis in paper II. Ilka von Dalwigk took part in the field work, collaborated on data interpretation, and co-authored papers I and II. Curt Broman took part in fluid inclusion studies of paper I, and co-authored papers I and V. Sherry Cady collaborated on data interpretation of paper II, and co-authored papers II and IV. Richard Hugo developed the sample preparation technique mentioned in paper IV. Yngve Zebühr performed GC-MS analysis and co-au-thored paper V.

All other work was carried out by the author alone.

Stockholm, April, 2005

Tomas Hode

3

The Search for Life on Mars – Preparation for Sample Return

Tomas Hode

Swedish Museum of Natural History, department of Paleozoology, Box 50007, S-10405 Stockholm Sweden

This doctoral thesis consists of a summary and of the following papers:

Paper IHode, T., von Dalwigk, I., Broman, C., (2003). A hydrothermal system associated with the Siljan Impact Structure, Sweden – Implications for the search for fossil life on Mars. Astrobiology 3(2), pp. 271-289.

Paper IIHode, T., von Dalwigk, I., Cady, S.L., Kristiansson, P., Evidence of Ancient Microbial Life in a Large Impact Structure. Submitted to Geology.

Paper IIIKristiansson, P., Hode, T., Auzelyte, V., Elfman, M., Malmqvist, K., Nilsson, C., Pallon, J., Shariff, A., Wegdén, M. (2004). Development of a system for determination of the 13C/12C isotopic ratio with high spatial resolution. Nuclear Instruments and Methods in Physics Research B 219-220, pp. 561-566.

Paper IVHode, T., Kristiansson, P., Hugo, R.C., Cady, S.L., Development of a non-destructive micro-analyti-cal method for stable carbon isotope analysis of Transmission Electron Microscopy (TEM) samples. Submitted to Astrobiology.

Paper VHode, T., Zebühr, Y., Broman, C., Controlled fluid inclusion extraction. Submitted to Planetary and Space Science.

Per Kristiansson co-authored papers III and IV, and took part in the development of the method pre-sented in these papers. He also collaborated on the RBS analysis in paper II. Ilka von Dalwigk took part in the field work, collaborated on data interpretation, and co-authored papers I and II. Curt Broman took part in fluid inclusion studies of paper I, and co-authored papers I and V. Sherry Cady collaborated on data interpretation of paper II, and co-authored papers II and IV. Richard Hugo developed the sample preparation technique mentioned in paper IV. Yngve Zebühr performed GC-MS analysis and co-au-thored paper V.

All other work was carried out by the author alone.

Stockholm, April, 2005

Tomas Hode

4

5

The search for life on Mars

The field of astrobiology has been growing sub-stantially in recent years, boosted by the discover-ies of water on Mars, of the probable ocean be-neath the icy surface of Jupiter’s moon Europa, of the existence of extrasolar planets, and of organ-isms living in extreme environments on Earth.

The search for present and past life on Mars in par-ticular has drawn major attention from the scien-tific community, as well as from the national and international space agencies. The reasons for fo-cusing the search on life to Mars is that, apart from being the closest planetary body of major astrobi-ological interest, Mars may have shared a number of environmental features with Earth during the early phases of planetary history (4.5–3.5 Ga; e.g. McKay and Stoker, 1989). The atmosphere was denser and probably similar to the Earth’s atmo-sphere in composition, and liquid water was pres-ent on the surface (Squyres et al., 2004), either as a stable water body, or as frequent flooding events of short duration (Segura et al., 2002). A number of geological processes were probably similar to those occurring on the early Earth, including intense volcanism and a frequent bombardment of large meteorites, implying that hydrothermal systems were widespread on early Mars (e.g., Farmer, 1996). The latter are of particular interest as targets in the search for fossil (and present) life on the red planet (e.g., Farmer, 1998; Nealson, 1999, and references therein), since life on Earth probably adapted early to thermal environments (Stetter, 1996). In addition, hydrothermal systems are capable of preserving biosignatures indicative of microbial life (chemofossils, organosedimen-tary structures, silicified microfossils, biomark-ers, etc.) (e.g. Simoneit et al., 1998; Cady et al., 2003). Altogether, this suggests a conservative search strategy for evidence of life on Mars, us-ing the same principles that have been applied to Precambrian paleontology on Earth (Schopf and Klein, 1992; Walter, 1999).

However, the actual search for life on Mars is not an easy task, and a number of difficult issues need to be considered: Assuming that life really did evolve on Mars, where exactly should we look for it? Environments likely to have hosted life, and to preserve fossil traces of that life, need to be iden-tified. Should we focus our search on extant or

extinct life? The surface of Mars is a very harsh place, and it is likely that, if present, any Martian lifeforms will dwell in the subsurface. Considering the high costs and difficulties related to reaching that subsurface, the search needs to be focused on fossil life forms preserved in surface rocks. If so, how do we recognize true biosignatures of pos-sible fossilized organisms? How can contamina-tion be avoided? Any samples collected from the Martian surface are very likely to be exposed to contamination of either terrestrial microorgan-isms or organic material, which is a very real and almost unavoidable problem.

Identifying traces of life

Assessing the biogenicity of potential bacterial fossils is an essential but very difficult task, as recently shown by the controversy concerning “Earth’s oldest fossils” (Brasier et al., 2002), and the fierce debate surrounding the claim for possi-ble life in the Martian meteorite ALH84001 (e.g., discussed in Gibson et al., 2001). This is particu-larly crucial when the purported fossil is pushing the boundaries of the life we know, either in time or in space. In the deep history of Earth it is not at all certain that life existed, and when we enter the domains of extreme environments, the presence of life is not granted. On other planetary bodies, life may never have existed, and tests based on known life forms may, furthermore, be inappropriate.

In order to verify that a potential fossil indeed is the product of biologic activity, a number of cri-teria have to be considered (e.g. Gibson et al., 2001):

1. Is the geologic context of the sample known? Is it compatible with life? Age and stratigraphic location of the sample must be known: Is the sample understood well enough to connect it to the geological history?

2. Is there evidence for cellular morphol-ogy or structural remains of colonies or communities within the sample? If so, can these morphological traits be distin-guished from non-biological features and biomorphs?

3. Is there any evidence for biomineraliza-tion (chemical or mineral disequilibria, unusual growth habits) in the sample?

6

4. Is there information about stable isotope fractionation that is characteristic of life?

5. Are there any organic biomarkers pres-ent?

Thus, a thorough investigation of the geological setting, the properties of the potential microfos-sils, as well as an understanding of how the various biological signals are altered during the fossiliza-tion process and post-depositional diagenesis, are essential. New microanalytical methods are there-fore developed and applied in order to meet these demands (e.g., Boyce et al., 2001; House et al., 2000; Toporski et al., 2002a, 2002b; Kristiansson et al., 2004).

Mars analogue studies

Because early Mars and early Earth in many ways probably resembled each other, it is reasonable to argue that ancient-Earth studies will aid in the search of life on Mars as well, even though the nature of the encountered problems may differ somewhat. The main problem with early-Earth studies is the near lack of preserved surface rocks, and the heavy alteration of most preserved rocks. The surface of Mars, on the other hand, is most likely well preserved because of lack of plate tectonics, and in some cases it is probably very ancient. Nevertheless, when an ancient and well-preserved rock from the surface of the Earth is identified, and the search for traces of life is ini-tiated, the problems that are faced are similar to those that can be expected in the search for traces of biologic activity in Martian rocks.

Assuming that life really did emerge on Mars, the question is where to look for it. First of all, it is reasonable to look for areas that show evidence for a hydrological past, such as sedimentary rocks (clastics, calcareous sediments, evaporitic rocks, BIFs, and phosphorites), organic and organically derived sediments, and hydrothermal deposits. Second, it is important to sort out and look for environments that may be expected to preserve biosignatures.

Possible sedimentary traps and environments have been extensively discussed by Komatsu and Ori (2000). Carbonaceous, phosphoritic, and organi-cally derived sediments should be very good ma-terial to start with, but unfortunately such rocks

either have not been identified or are not very likely to exist in reasonable quantities on Mars today. Possible clastic and evaporitic sedimen-tary deposits in fluvial and lacustrine environ-ments have been observed in numerous areas on Mars (e.g., Cabrol and Grin, 1999, and references therein; Squyres et al., 2004). Hydrothermal de-posits are likely to exist, given the obvious ther-mal input and recent evidence for large amounts of water on Mars (Boynton et al., 2002; Feldman et al., 2002; Mitrofanov et al., 2002; Lundin et al., 2004). These environments are thus important targets in the search for life on Mars.

The most important factor for preserving mor-phology and organic matter, or its geochemical signatures, in sedimentary deposits, is burial. Oxidation will destroy any organic matter ef-ficiently, and burial is the best geologic process to avoid the oxidizing environment on the Terran and Martian surfaces. Preservation of organic matter in sediments on Earth is largely a function of subsidence rate, but on Mars, where subsid-ence is a minor component, some other process seems to be responsible for the extensive layering that has been observed by Mars Global Surveyor and MarsExpress. In any case, burial does seem to have occurred, leaving opportunities for puta-tive Martian microorganisms to be preserved at the surface or near-surface regions.

Hydrothermal deposits are of interest for sev-eral reasons. Early life on Earth was probably adapted to hot environments (Stetter, 1996). The often reducing local environment in hydrother-mal systems (e.g. Summons et al., 1996) boosts the chances for preservation of organic matter, and high precipitation rates lead to encrustation of microorganisms (Cady and Farmer, 1996). The two major environments of interest on Mars, with respect to hydrothermal mineralizations, are vol-canic areas and sites of large meteorite impacts. Important conditions are that the hydrothermal system had a temperature interval compatible with life and that it can be accessed on the sur-face of the planet. Hydrothermal areas on Mars should be fairly easy to identify, since volcanoes and large meteorite craters are dominant features on the surface. An additional advantage of study-ing large impact craters on Mars is that paleolakes sometimes may be localized to these craters (e.g. Cabrol et al., 2001).

7

Martian samples

Once a target area has been selected, a number of issues need to be considered. First of all, what kind of mission should be sent to the target? Are robotic missions for in situ research sufficient? Or are robotic sample-return missions, or maybe even manned missions, necessary?

The main advantage with robotic in situ re-search missions is the lower price tag owing to lower technical complexity of one-way mis-sions. Unfortunately, less scientific feedback has to be expected with such missions compared to robotic sample-return missions and manned mis-sions. Robotic in situ missions may well be able to detect areas with shallow-water ice, character-ize mineralogy and maybe some petrology, drill a few centimeters in order to detect organic mol-ecules below the oxidizing surface, and detect a very shallow subsurface biosphere if the biomass is rich, but since there can be no advanced sample preparation or heavy instrumentation, a detailed investigation of rock samples will be difficult to perform.

Thus, in order to have any chance of finding fos-silized microorganisms, a robotic sample return mission will probably be necessary, but perhaps not sufficient. A robotic sample return mission will still have very low flexibility and mobility on the Martian surface, and it will only be able to bring back small amounts of material.

Manned missions, on the other hand, will be char-acterized by high flexibility and mobility, mak-ing it possible to perform a hands-on field review of the chosen samples, increasing the chances to bring back large amounts of highly interesting material such as sedimentary rocks and hydro-thermal mineralizations.

Unfortunately, manned missions will be exceed-ingly expensive and dangerous, so in the mean-time it will be necessary to focus on sample return missions and optimize them as much as possible.

Once a sample is brought back to Earth, an ad-ditional problem has to be faced, namely con-tamination. Any sample brought to a laboratory will likely be exposed to contamination of small amounts of organic matter and terrestrial micro-

organisms. The sample should be transported in a sealed chamber containing Martian atmosphere and pressure, and clean-room handling is vital during the analyses.

An additional way of avoiding contamination may be to study fluid inclusions in Martian rocks. Fluid inclusions are in effect contained in sealed vessels, and until these are opened, the content is insensitive to contamination. Organic matter in such inclusions, if present, is also likely to have been shielded from the oxidizing environment of the Martian atmosphere and would therefore stay a better chance of being preserved, even if the sample is collected from the surface.

Aim of the project

The main thrust of the project has been to develop methods and concepts to aid in the search, detec-tion and assessment of ancient microfossils here on Earth as a guide to the search for ancient life on Mars. The intention has been to identify and characterize environments on Earth that may be considered analogous to Martian environments, and in which fossil preservation is expected to be good, and to develop and apply methods to char-acterize the isotopic and chemical composition of possible traces of life in order to assess their bio-genicity and biological affinities.

Thus, the project has been divided into two parts:

Papers I and II:

Investigation of the hydrothermal system associ-ated with the Siljan Impact Structure as an ana-logue to Martian impacts, with the purpose to as-sess whether such impact-induced hydrothermal systems may be suitable targets in the search for fossil life on Mars.

Papers III, IV, and V:

Development of new analytical methods and concepts to be applied on samples returned from Mars. The development of a non-destructive method to analyze stable carbon isotopes of very thin samples is presented, and a novel concept to extract (single) fluid inclusions without risking contamination is demonstrated.

8

Summary of papers

PAPER I: A HYDROTHERMAL SYSTEM ASSOCIATED WITH THE SILJAN IMPACT STRUCTURE, SWEDEN – IMPLICATIONS FOR THE SEARCH FOR FOSSIL LIFE ON MARS.

Large impact craters may for several reasons be of interest in Martian exploration: There is evidence for the formation of paleolakes with associated sedimentary deposits (Cabrol and Grin, 1999, 2001, and references therein), and formation of local hydrothermal systems associated with the impacts has been discussed (e.g., Newsom, 1980; Newsom et al., 1996). The latter is of particular in-terest in the search for fossil life on Mars (Walter and Des Marais, 1993; Farmer and Des Marais, 1993). Impact-related hydrothermal systems on Earth may therefore be a suitable analogue to those on Mars in the search for fossil (and pres-ent) life (Nealson, 1999).

In order to test the hypothesis that large impacts generate suitable habitats for thermophilic com-munities, we have studied the Siljan Impact Structure situated in southern central Sweden. The diameter of the structure has been deter-mined to 65 km (Kenkmann and von Dalwigk, 2000), which makes it the largest known impact structure of Western Europe, and it has been dated to 368±1.1Ma by the 40Ar/39Ar method (Grieve, 1987). It is a complex crater, displaying several structural units that include a central uplifted re-gion surrounded by a ring-shaped depression. Associated with the crater, there is evidence of a low-temperature hydrothermal system with hy-drothermal mineralizations located either within the ring depression, or at the border between the central uplift and the surrounding ring depres-sion.

The hydrothermal system was most likely formed by the impact. Had the hydrothermal minerals been formed prior to the impact, the fluid inclu-sions present in the examined vein samples would most likely have been destroyed. Actually, such empty pre-impact inclusions have been reported in fractured granite at a depth of approximately 1 km below the present surface of the central Siljan structure (Smith, 1996), thus indicating that the hydrothermal minerals containing preserved fluid inclusions were formed after the impact. Also,

Komor et al. (1988) demonstrated that trails of fluid inclusions were cross-cutting planar defor-mation features at oblique angles, demonstrat-ing that the hydrothermal system was formed after the impact. On the other hand, if the hydro-thermal minerals had been formed by a regional metamorphic event during, or after, the impact, it would have been expected that hydrothermal mineralizations were present in fractured bedrock on a regional scale, which is not the case. Finally, fluid-inclusion studies of the impact-induced hy-drothermal mineralizations revealed a very low salt content (0–2.4 eq. wt. % NaCl), indicating a meteoric origin of the fluids in which the minerals precipitated, whereas fluid inclusions in hydro-thermal minerals that were formed as a result of the Caledonian orogeny (440-390 Ma; Högdahl, et al., 2001) display significantly higher salt con-tent (25 eq. wt. % NaCl+CaCl

2; Lindblom, 1986;

Sturkell et al., 1998). This indicates that the hy-drothermal mineral assemblages associated with the Siljan impact structure were formed indepen-dently of the Caledonian orogeny. Since no other regional metamorphic event has occurred after the impact, this strongly supports an impact-induced origin of the hydrothermal system.

Fluid-inclusion studies of the mineralizations re-vealed homogenization temperatures that ranged from below 75ºC up to 137ºC (variations depend-ing on mineral assemblage), and with an estimated erosional unloading of about 1 km, the formation temperatures probably were not more than 10-15ºC higher. The lowest formation temperatures were found in quartz veins and breccia fillings, evidenced by single-phase fluid inclusions indi-cating temperatures below 75ºC. Slightly higher formation temperatures were observed in calcite and fluorite veins (90-120ºC and 100-130ºC re-spectively, including pressure correction), whereas the highest formation temperatures (up to 340ºC) were found in quartz veins in the central area of the structure (Lindblom and Wickman, 1985).

Field observations and formation temperatures indicate a quartz-calcite-fluorite-pyrite-galena-sphalerite-quartz paragenetic sequence for the studied samples. The outlined paragenetic se-quence suggests that during early and main stages of the impact-induced hydrothermal system, the niches for thermophilic organisms were located toward the rim of the crater (figure 1), but that

9

these niches moved toward the center of the crater during the final cooling stages.

Based on these observations, we confirm the as-sumption that large impacts can generate hydro-thermal systems with temperature intervals that are suitable for thermophilic communities. This is consistent with similar studies of other impact structures (e.g. Osinski et al., 2001) on Earth, thus indicating that the Siljan impact structure is not unique in this context. Given the recent evidence for water on Mars, it is reasonable to assume that large impacts generated hydrothermal systems on Mars as well, supporting the idea that impact structures are suitable places to look for fossil thermophilic organisms on the red planet.

PAPER II: EVIDENCE OF ANCIENT MICROBIAL LIFE IN A LARGE IMPACT STRUCTURE

In order to search for ancient microbial life in a particular geological setting, it is essential that the local environment, in which the purported micro-organisms were to have existed, is shown to be consistent with the possibility to support living organisms. In the case of hydrothermal systems, the critical factor for survival of organisms is temperature, i.e. the temperature of the percolat-

Figure 1. Fluid inclusion studies revealed that the tempera-tures of the hydrothermal system peaked at the center of the crater (327-342ºC), and declined towards the outer parts of the structure (<130ºC).

Figure 2. Hydrothermal calcite vein in the downfaulted Or-dovician limestone in the Siljan impact structure.

ing fluid should not exceed the upper limits under which life is sustainable.

Several studies of large impact craters demon-strate that the temperature interval of the associ-ated hydrothermal systems indeed were favorable for thermophilic communities (e.g. Hode et al., 2003). However, no evidence of fossilized micro-bial communities associated with impact-induced hydrothermal systems has yet been presented, despite the facts that several studies have inves-tigated preservation processes of microorganisms in modern hydrothermal environments (Cady and Farmer, 1996; Renaut et al., 1998; Konhauser et al., 2001; Toporski et al., 2002b) and that other ancient hydrothermal deposits have been shown to contain fossilized microorganisms (Logan et al., 2001; Walter, 1996; Rasmussen, 2000; Westall et al., 2001). Thus, in order to confirm the hy-pothesis that large impact structures are suitable targets in the search for ancient life on Mars, it is necessary to identify biosignatures in such sys-tems on Earth.

Samples were therefore collected from a hydro-thermal calcite vein (figure 2) in the downfaulted sedimentary rocks in the Siljan impact structure. Fluid inclusion microthermometry revealed that the formation temperature of this particular vein ranged between 90ºC and 110ºC (Hode et al., 2003), indicating that the temperature interval was suitable for possible thermophilic organisms when the minerals formed. The hydrothermal so-lutions were anoxic, rich in hydrocarbon content, and given the various sulphide mineralizations there was probably a significant amount of re-

10

duced sulphur present in the circulating solutions, all of which is consistent with the possibility of survival for an ancient anaerobic thermophilic community (e.g. Logan et al., 2001).

The collected material was prepared for SEM studies in three different ways; etching with 2% EDTA for 30 minutes, etching by 30% HCl for one minute, and no etching at all. The samples were studied with a Hitachi S-4300 field emission scan-ning microscope (FESEM), and element analysis was performed on the Nuclear Microprobe facility at the Lund Institute of Technology, Sweden. The SEM studies revealed a large number of possible preserved microbial features (figure 3) that were not visible with an optical microscope. The largest concentration of the fossilized material was found in samples that were etched with EDTA, whereas a significantly lower concentration was found in samples etched with HCl, and no preserved mate-rial was found in the non-etched samples.

The shape and complexity of the features indicate a biological origin rather than abiological organic films. They occur as randomly distributed patches attached to surfaces or partly embedded in min-eral matrices. The preserved material was found exclusively inside the hydrothermal calcite vein, i.e. none of the features were discovered in the etched parts of the carbonate rocks surrounding the vein. No preferred alignment of the various features inside the vein was evident; the fossilized material appears inside calcite crystals as well as attached between crystal borders and attached to non-calcitic material in the vein. In several cases the etching has only partly uncovered the features (figure 3a), leaving some of it still embedded in the calcite matrix, indicating that it is not a recent contamination. This is also supported by the fact that none of the features were found outside of the vein, or in the non-etched samples.

Special attention was given to a biofilm-like rem-nant shown in figure 4a. The curved film, which stretches nearly 30 µm, is attached to grains at discrete positions and is partly embedded in the calcite grain on the left. The right-hand side of the film is draped over underlying crystals and elon-gate objects, and wrapped around them. Though the film is mineralized (see below), the elliptical shape of two holes within it indicates that it was flexible and contracted slightly prior to mineral-

Figure 3. SEM photomicrograph images that illustrate the thread and polymer-like objects associated with the fossil-ized biofilm remnants identified in this study in the Siljan impact hydrothermal structure. (a) Threads partly embed-ded in a calcite matrix. The white rectangle marks the en-larged area on the right-hand side of the image, showing de-tail of the partly embedded threads. (b) A twisted film with fibrils attached to calcite and clay (lower right corner). (c) A bundle of threads situated on top of calcite and clay. The threads are broken on the lower right portion of the bundle, indicating brittleness and partial mineralization.

11

ization. These morphological attributes match those of modern partly mineralized hyperthermo-philic biofilms that form on sinter surfaces and among debris that accumulates at the bottom of active near-boiling hot spring pools and outflow channels (figure 5).

Rutherford backscattering (RBS) element analy-sis of the film demonstrated that the carbon con-tent was approximately three times as high in the preserved features relative to the surrounding cal-cite (figure 4b and 4c). Since the RBS method is depth-sensitive, it was possible to determine that the higher carbon concentration was confined to the biofilm-matrix structure and did not extend into the calcite grain situated beneath the feature. The RBS spectra also revealed that the film is partly calcified, evidenced by the lack of shift in energy between the calcium peaks in the analyzed feature and the surrounding calcite. All data were

analyzed with the SIMNRA software developed at the Max-Planck-Institute of Plasma Physics.

These findings meet all of the criteria proposed to distinguish the preserved biofilm remnants, re-ferred to collectively as extracellular polymeric substances (EPS), from their abiotic counterparts (Westall et al., 2000). Such criteria include the composition, structure, texture, distribution and morphological complexity of the material, and the occurrence of such objects in geological settings that would have supported a microbial commu-nity:

1) RBS analysis indicates that the biogenic-like features, though calcified, are carbonaceous. The enriched carbon content is not consistent with mineral artefacts, and the partly calcified nature of the features indicates that they are not modern contaminants.

Figure 4. Rutherford backscattering (RBS) spectra of purported microbial feature and surrounding calcite grains. (a) SEM image showing a smooth but curved film-like feature stretched between calcite (beneath and around the upper portion of the feature) and pyrite (below the feature) crystals. (b) RBS spectrum of the calcite situated above the film-like feature. Carbon, oxygen and calcium peaks are labelled in the spectrum. The x-axis shows the energy of the backscattered alpha-particles, and the y-axis shows the number of counts per channel. (c) RBS spectrum of the film-like feature showing the elevated (3x) carbon content. Data analysis revealed that the higher carbon concentration in the biofilm remnant did not extend into the calcite grain situated beneath it. The lack of energy shift between the calcium peaks (starting at 1700 KeV) in Figures 1b and 1c demonstrates that the film-like feature is calcified.

12

2) The features are structurally and texturally complex. These attributes, together with the part-ly calcified nature, the character of the attachment points, and the size range of the features clearly distinguish them from organic films that accumu-lated or were trapped between or within grains.

3) The mode of occurrence and the partly em-bedded nature of the features indicate that they formed while the hydrothermal system was ac-tive. Morphological evidence also shows that the

features were attached to the substratum in situ, demonstrating that they are not biofilm remnants that were carried along with the fluids to become trapped inside the vein minerals.

4) The geological setting would have supported a thermophilic community in the hydrothermal system while it was active. The temperature of fluid that moved through this particular vein is es-timated to have ranged between 90ºC and 110ºC. The mineral assemblage within the vein indicates that the fluid was anoxic and reducing, and rich in organics and reduced sulphur phases, conditions that could have supported ancient subsurface thermophilic microbial communities and possibly sulphur-oxidizing populations

Our findings indicate, for the first time, that im-pact-induced hydrothermal systems were colo-nized by thermophilic microbial communities. Though compelling cellular remains were not found, the discovery of mineralized biofilm rem-nants supports the hypothesis that – given the right context and mode of occurrence – fossilized microbial polymeric substances (FPS) may rep-resent an equally convincing biosignature as pre-served cellular remains.

We therefore support the suggestion that impact-induced hydrothermal deposits should be consid-ered as possible targets in the search for fossil life on Mars. Impact-induced hydrothermal mineral assemblages are exposed at the surface after an impact (Osinski et al., 2001), making it possible to locate remotely the outer regions of the depos-its that could have become colonized by microbial communities, and narrowing down the search for fossil microbial life to highly specific areas on the red planet. PAPER III: DEVELOPMENT OF A SYSTEM FOR DETER-MINATION OF THE 13C/12C ISOTOPIC RATIO WITH HIGH SPATIAL RESOLUTION

Stable carbon isotopes are important indicators of biological activity, particularly in the fossil re-cord. Fractionation is associated with the fixation of CO2 in autotrophic organisms (Hayes, 2001). The fractionation is dominated by the 13C deple-tion associated with the uptake of CO2, but also with enzymatic discrimination in the synthesis of biomass from inorganic carbon (Hayes, 2001).

Figure 5. Examples of remnants of modern hyperthermo-philic biofilms in near-boiling hot springs, Yellowstone National Park, USA. The remnant consists primarily of ex-tracellular polymeric substances that comprise the biofilm matrix and polymers that retain adherence of the biofilm to the mineral substratum . (a) Extracellular matrix appears to have shrunk and contracted, pulling away from the substra-tum except at those attachment points that remain intact. The flexible biofilm matrix tends to curl back around itself along some of its edges, enclosing cells and granular debris. (b) Biofilm matrix attached to underlying substratum with several convergent and distinct polymeric “threads” char-acterized by a range of thicknesses that overlap, but are not constricted by, those of viable filamentous cells.

13

The isotopic composition of preserved organic carbon in potential microfossils may therefore be useful in the assessment of the biogenicity, as well as in the interpretation of the biological activity. It is important to note that carbon isotope ratios in themselves are not sufficient to prove bioge-nicity, particularly if the isotope content of, e.g., paleo-atmospheric CO2 is poorly known, such as with Mars. However, isotopic information may be a very useful paleontological and astrobiological tool when combined with geological and chemi-cal data.

Presently, the most popular tool for carbon iso-tope microanalysis of individual bacterial fossils is the secondary ion mass spectrometry (SIMS) method (e.g., House et al., 2000). However, as with other mass spectrometry methods, it is lim-ited by matrix effects and by being slightly de-structive, making it essentially non-applicable to biological and transmission electron microscopy (TEM) samples. TEM is a widely used method of imaging microscopic and submicroscopic biolog-ical and geological samples (e.g. Thomas-Keprta, 2000), but owing to the extremely thin nature of TEM samples (~100 nm), isotope and quantified element analyses have been difficult to perform.

Here we present the development of an alternative microanalytical method to analyze stable carbon isotopes with high lateral resolution. The meth-od is based on Rutherford Backscattering (RBS) with a nuclear microprobe instrument (NMP). Traditional RBS measures the number and energy of ions that backscatter after colliding with atomic nuclei in a sample. The probability that an ion will backscatter is roughly proportional to the square of the atomic number of the target atom. A beam particle hitting the target will lose energy, and the ratio of post-collision to pre-collision energy (the kinematic factor) is measured. At the energy of 2.751 MeV, there is an energy resonance interval that will enhance the cross section for 13C rela-tive to 12C by a factor of about 40 because of de-viations from the pure Rutherford cross section. This substantially increases the 13C cross section relative to the pure Rutherford cross section, and simultaneously the 12C cross section decreases. A set of experiments was conducted in order to evaluate whether it was possible to separate be-tween 13C and 12C and whether good statistics was achievable.

The following conclusions were drawn:

1) A separation between 13C and 12C appears to be achievable (figure 6a). However, unless puri-fied carbon samples are used, there will be a large disturbing background under the 13C peak from elements with higher cross section. The conclu-sion was therefore that thin samples should be used, i.e. thin enough to eliminate the major part of the background but thick enough not to loose yield from the 13C resonance. Standard transmis-sion electron microscopy (TEM) samples were therefore prepared in order to evaluate whether the interfering background from heavier elements could be reduced.

2) The following experiment confirmed that TEM sections are thin enough to reduce the background (figure 6b) from oxygen. However, the TEM sec-tions had been placed on Cu-grids, and it was evi-dent that a background signal from Cu interfered with the spectrum. It was therefore concluded that Be grids had to be used to eliminate the back-ground from the grid.

3) As a consequence of the very thin nature of TEM samples (~100 nm), the count rate was sig-nificantly reduced. In order for the method to be useful in reality, it would be required that the sta-tistical precision is in order of ± 2 ‰, something that requires at least 250.000 counts of 13C assum-ing Poisson distribution statistics. The 13C yield largely depends on the incoming alpha-particle energy, but also by the scattering angle. Thus, to optimize the statistics, it is necessary to measure the backscattered particles over the entire solid angle for which there is a resonant behavior.

An experiment was therefore conducted with the aim to determine the angular interval in which 13C acts with a resonant behavior. The conclusion was that the angular interval to be used should cover at least 150-180°. However, the energy difference between the recoils from 13C and 12C in the an-gular interval 150-180 degrees is about 85 keV and the energy variation for the recoils from 13C is about 100 keV. As a consequence, there will be an energy overlap between recoils from 12C at 180° and 13C at 150°, so to avoid this, it is necessary to segment the angular interval by using multiple de-tectors. A new detector system, consisting of four annular surface barrier detectors placed around

14

the incoming alpha beam, was therefore designed. The detectors are placed in such a way that the kinematic effects are reduced and the resolution could be kept sufficiently high.

As a conclusion, it was demonstrated that the sys-tem should be capable of collecting enough statis-tics for the required precision.

This method of determining ratios of isotopes in TEM-samples has advantages that will make it a valuable compliment to other micro/nano-analyti-cal methods:

(1) It is a non-destructive non-sputtering method, and therefore particularly suited for micro/nano-analysis of extraterrestrial and other rare or irre-placeable material.

(2) Because of its non-destructive nature, the method can be applied to very thin samples (~100 nm), thereby avoiding the problems with matrix effects that tend to occur with sputtering mass-

spectrometry methods. Thus, any type of solid material can be analyzed, including biological material.

PAPER IV: DEVELOPMENT OF A NON-DESTRUCTIVE MI-CRO-ANALYTICAL METHOD FOR STABLE CARBON ISOTOPE ANALYSIS OF TRANSMISSION ELECTRON MICROSCOPY (TEM) SAMPLES

This paper describes the second part in the devel-opment of an RBS-based method to determine the 13C/12C ratio of TEM samples with high lateral resolution. To test the experimental setup pre-sented in Kristiansson et al. (2004), and to further evaluate whether RBS is applicable on stable car-bon isotope analysis, the energy interval at which there is a resonant behavior was investigated, data were collected, and data handling routines were developed.

As discussed in Kristiansson et al. (2004), using very thin samples may not necessarily decrease the 13C yield since the resonance energy interval is very narrow. If a thicker sample is used, the al-pha-particles will lose energy as they pass through the sample, and slip out of the resonance interval somewhere below the surface, which means that the resonance effect will only occur in the surface volume of the sample. Since very thin samples will act as that surface volume where electron stopping is negligible, there will be no interfer-ing background from heavier elements. However, if the thickness of the sample is smaller than the depth interval at which there is a resonant behav-ior, the total 13C yield will be decreased, hence invalidating the statistical approach needed to maintain the precision needed for this application of the technique.

Thus, in order to evaluate whether TEM sections are sufficiently thick, it was necessary to investi-gate the energy interval in which there is a reso-nant behavior. By gradually shifting the beam energy while monitoring the acquired spectra, the resonance energy interval was determined to be in the order of ~10 keV, i.e. TEM samples are thick enough not to compromise the 13C yield.

Once an optimal experimental approach had been achieved, an experiment was conducted in order to gather enough statistics to evaluate wheth-er a determination of the 13C content with high

Figure 6. Rutherford backscattering spectra of carbon-rich samples. (a) Analysis of a graphite rod demonstrating a clear separation between 13C and 12C. (b) Analysis of a TEM sec-tion indicated that the background from heavier elements may be reduced with thin samples.

15

enough precision was possible with the given setup. Analysis was performed on a graphite stan-dard with a known δ13C value (–25.6 ± 0.2 ‰ vs. PDB), and on cultured Chloroflexus cells with a δ13C value of approximately – 49 ‰ vs. PDB. The samples were prepared according to protocols de-scribed by Hugo and Cady (2004).

A total of eight spectra were collected, four for each sample, i.e. one spectrum for each of the four detectors in the detector system described by Kristiansson et al. (2004). Regression analysis was performed on the different spectra, where the 12C and 13C peaks were described as functions of normal distributions, and the background was as-sumed to behave exponentially. R2 values of the fitted spectra ranged between 0.998 and 0.999, in-dicating that a good fit was achieved. It was there-fore concluded that the total number of 13C counts indeed could be calculated accurately.

However, inasmuch as it was demonstrated that the number of 13C counts could be determined ac-curately, it was also evident that a consistent way of determining 12C could not be achieved with the given setup. Since it is the difference between the relative 13C/12C ratio of the standard and the ana-lyzed sample that needs to be investigated, a con-sistent way of measuring 12C is needed in order to acquire a δ13C value of the sample of interest.

It will therefore be necessary to develop a con-sistent way of determining 12C of the standard and the analyzed sample before the method can be used. This includes further development of the sample preparation techniques by placing the standard and the sample on the same beryllium grid, and removal of additional background by using a forward-backward coincidence technique. Finally, by using the scanning transmission ion microscopy (STIM) technique to determine the thickness of the samples with high accuracy, it will be possible to ensure a consistent measure of 12C versus 13C.

Once finalized, the implications of this non-de-structive method are of particular interest in the Martian exploration research, or, more specifically, for analysis of samples returned from Mars. Any material delivered from Mars must be considered extremely valuable, and this is particularly true in material containing possible microfossils.

PAPER V: CONTROLLED FLUID INCLUSION EXTRAC-TION

Fluid inclusions in rocks and minerals may hold valuable information about the temperature and composition of the fluids in which the minerals precipitated. Low temperature mineralizations in fluids containing organic phases should be of spe-cial interest for astrobiologists since they may act as sealed containers of non-contaminated organic matter with a defined minimum age.

Most techniques concerning the study of these in-clusions are indirect, and involve methods such as freezing and heating the sample (in order to study homogenization temperatures and composition of the fluid), Raman analysis of solid phases, and fluorescence spectroscopy of organic phases. To perform (often needed) direct analysis of the fluid content, it has been necessary either to crush the sample or to heat it, with subsequent decrepita-tion of the inclusions, in order to extract the fluids. However, a disadvantage with crushing or heating is that a mean composition of all fluid inclusions in the sample is analyzed, and no distinction is made between primary and secondary inclusions. The only way so far to open fluid inclusions in a controlled manner has been laser ablation, but that method has many disadvantages, particularly when it concerns organic matter, and it has there-fore not been used very frequently.

In this paper we present a novel concept to ex-tract selected fluid inclusions, without ablating the sample. The method is based on the illumina-tion of selected fluid inclusions with a laser of a wavelength that is absorbed by water and organic material, but not by the minerals encapsulating the fluid. Thus, when the inclusion is illuminated by the laser, the fluid expands and the inclusion decrepitates. Since the sample is placed in a vacu-um chamber, the fluid evaporates, and may subse-quently be collected in a cold finger.

As a proof-of-concept, an instrument was de-signed, and an experiment was conducted in order to extract and analyze hydrocarbons trapped in fluid inclusions. The basic parts of the instrumen-tal design consisted of an ErYAG laser (λ=2940 nm), and a vacuum chamber with a quartz lid, a cold finger, a vacuum pump, and a vacuum meter (figure 7). The cold finger was designed so that an

16

exchangeable GC-MS (gas chromatography mass spectrometer) vial constituted the tip of the finger that was lowered into the liquid nitrogen. By do-ing so, any extracted material could be analyzed directly with the GC-MS instrument without a transfer of the collected material being necessary, thus reducing potential contamination during the handling of the collected hydrocarbons. A calcite sample, collected in the Siljan impact structure, was prepared as a doubly polished thin section, and placed in the chamber. The sample, which was known to contain hydrocarbon-rich fluid inclu-sions (Hode et al., 2003), was thoroughly cleaned with acetone, hexane and de-ionized water, before the experiment was conducted. Also, a reference sample was prepared by collecting hydrocarbons from a crushed piece of the original sample. The purpose of the reference sample was to compare any hydrocarbons extracted by the laser with the original composition of the oil. The chamber was cleaned for several hours in a bath of hydrogen peroxide and hydrochloric acid, and subsequently washed with acetone, hexane and finally de-ion-ized water.

The experiment was divided into two parts: 1) A reference blank was produced, and 2) the actual extraction and hydrocarbon collection was per-formed. The reference blank was produced by performing the exact same operation as with the extraction procedure, but without actually firing

the laser on the fluid inclusions. This means that possible contaminants, present on the surfaces of the sample and chamber walls, that may have in-terfered with the final result, would also show up in the reference blank. The extraction process was basically performed in three steps:

1) The pressure in the chamber was reduced to 2x10-2 mbar, and a valve sealed the chamber from the ambient atmosphere.

2) The cold finger was lowered into liquid ni-trogen.

3) The laser was fired through the quartz lid upon the sample. In order to increase the extraction yield, the laser was gradually moved closer to the sample while firing. By increasing the temperature, more fluid in-clusions decrepitated, but this also increased the risk of cracking the hydrocarbons in the fluid inclusions.

Since the pressure in the chamber gradually in-creased because of leakage, steps 1-3 were repeat-ed several times. The decrepitation of the fluid in-clusions could be monitored on the vacuum meter: as soon as an inclusion was opened, the pressure in the chamber increased, only to decrease again immediately as the fluid condensed in the cold finger. Once the experiment was finished, the ref-erence blank and the extracted sample were taken to analysis by the GC-MS.

Figure 7. The instrument designed to extract and collect fluid inclusions from a doubly polished thin section. The instrument consists of a sample chamber, a cold finger, and a vacuum meter. The chamber is connected to a vacuum pump, and sealed with valves of stainless steel. The laser extracting the fluid inclusions is fired through the quartz lid above the sample.

17

The fundamental limitation with the presented setup was the relatively low sensitivity of the GC-MS instrument. The volume of a single fluid in-clusion with a size in the range of 10-30 µm is less than a few picoliters, which means that a large number of inclusions had to be extracted in order to exceed the detection limit of the GC-MS. Thus, the temperature had to be gradually increased, in order to extract as many fluid inclusions as pos-sible, by moving the laser (which was not colli-mated) closer to the sample when the extraction took place.

The GC-MS analysis revealed that hydrocarbons were extracted by the laser and collected in the cold finger. Alkane and isoprenoid biomarkers, that matched the analysis of the crushed reference, were detected in the extracted sample, whereas no hydrocarbons could be detected in the reference blank, indicating that no contamination had oc-curred.

The presented paper may therefore be considered as a proof-of-concept for the outlined method. However, in order extract and analyze single fluid inclusions, which is the ultimate purpose of the method, it will be necessary to combine the ex-traction device with a high-sensitivity analytical instrument, such as the ToF-SIMS (Time of Flight Secondary Ion Mass Spectrometry). It has been demonstrated that the ToF-SIMS is capable of analyzing organic biomarkers (Steele et al., 2001; Toporski and Steele, 2004), and the next step in the development of the outlined concept will therefore be to implement the extraction device with a ToF-SIMS instrument.

The ability to analyze the content of single, or a few selected, fluid inclusions may have profound implications for early Earth studies. Molecular fossils, such as organic biomarkers, represent very convincing evidence of ancient life, not only by being unquestionable traces of life, but also by being diagnostic of what type of organism they are derived from. However, hydrocarbons from the deep history of Earth are often only present inside fluid inclusions, and the ability to analyze and separate between different fluid inclusion generations, may provide new insights of the ear-ly record of life on Earth. An additional advantage with the method is the potential to avoid contami-nation of the analyzed hydrocarbons. Since fluid

inclusions act as sealed vessels, the chemical com-position of the fluid stays intact until the vessel has been opened. The sample can thus be handled and prepared in any desired way (often as doubly polished thin sections) without contaminating the fluid before the extraction takes place.

The method should therefore be ideal for analy-sis of extraterrestrial material: since the sample preparation and extraction procedure takes place without exposing the content of the inclusions to the terrestrial atmosphere, analysis of very small amounts of hydrocarbons may be performed with-out risking contamination.

Major conclusions

1) An investigation of the Siljan impact structure demonstrated that niches for thermophilic organ-isms were created in the associated hydrothermal system. The temperature regimes were favorable for thermophilic life in the outer parts of the struc-ture during the early and main stages of the hy-drothermal system, but that these niches moved toward the center of the crater during the final cooling stages. Given the presence of water on Mars, similar impact-induced hydrothermal sys-tems were probably generated on Mars as well. Thus, if hydrothermal mineralizations were to be found in the rims of Martian craters, it is likely that these minerals precipitated in a hydrothermal system with a temperature interval favorable for thermophilic life.

2) The hydrothermal mineralizations in the Siljan impact structure contain preserved traces of an-aerobic, and possibly sulfur reducing, thermo-philic microorganisms. These traces are repre-sented by fossilized EPS (extracellular polymeric substances), and the shape, complexity and at-tachment points of the FPS (fossilized polymeric substances) show strong similarity with EPS in modern hydrothermal settings. The FPS is partly embedded in the calcite matrix, it is carbonaceous but also partly calcified, all of which indicating that these features are not modern contamina-tions, nor that they represent secondary organic films trapped in the minerals. These findings con-firm the relevance of searching impact-induced hydrothermal deposits for evidence of microbial life on Mars.

18

3) A method for the determination of stable car-bon isotopes with high lateral resolution of TEM sections is currently being developed. The method is based on RBS (Rutherford backscattering), it is non-destructive, and therefore suitable for analy-sis of extraterrestrial and other rare or irreplace-able material. The results, so far, demonstrate that a separation between 13C and 12C is achievable with RBS, that the designed system is capable of collecting enough statistics for the required pre-cision, and that the developed data-handling rou-tines are capable of determining the number of 13C counts accurately. The remaining work to be per-formed before the method is finalized, includes the implementation of well investigated methods to eliminate the remaining spectral background.

4) A novel concept to extract fluid inclusions with-out ablating the sample has been proposed, and a proof-of-concept has been demonstrated. The purpose is to analyze organic biomarkers trapped in fluid inclusions without risking contamination, and also to potentially extract and analyze single fluid inclusions. The minimized contamination risk and the potential ability to extract single fluid inclusions could make the method a useful tool in the search for organic biomarkers in early-Earth material, and eventually, in samples returned from Mars.

Acknowledgements

First of all I whish to thank my friends and su-pervisors Stefan Bengston (sorry, Bengtson) and Sherry Cady, the former for countless hours of stimulating discussions and for always taking time to patiently answering my questions, and the lat-ter for putting up with my heavy email bombard-ment and for always being positive and encourag-ing me whenever we meet. I am also grateful to Curt Broman for guiding me through the world of inclusions and for always being positive and kind, to Per Kristiansson who actually managed to keep up a good face when talking to a nonphysicist, to Ilka von Dalwigk for invaluable help about im-pact geology and for putting up with me during numerous field trips, and to Johan Söderhielm for always being interested and for supplying me with thin sections despite often very short notice. I am very happy to have spent countless hours of quality time at numerous pub.., I mean confer-ences, with Jan Toporski, and it has significantly

increased the joy of doing science. I also thank Alasdair Skelton for very stimulating discussions, Yngve Zebühr for helping me out with the GC-MS during various night-shifts, the NMP group in Lund for taking such an interest in the developed techniques, the staff at the Paleozoology depart-ment at the Swedish Museum of Natural History and the staff at the department of Geology and Geochemistry for interesting and fruitful discus-sions about paleontology and geology in general. This thesis has been made possible through fi-nancial support from the Swedish National Space Board, for which I am very grateful. Finally, but most importantly, I whish to thank Jaana for very stimulating discussions and for always loving and supporting me, and my family and friends for un-flagging support. Our cats did not care very much about this thesis.

References

Boyce, C.K., Hazen, R.M., Knoll, A.H. (2001). Nondestructive, in situ, cellular-scale mapping of elemental abundances including organic carbon in permineralized fossils. PNAS 98(11), pp. 5970-5974.

Boynton, W.V., Feldman, W.C., Squyres, S.W., Prettyman, T.H., Brückner, J., Evans, L.G., Reedy, R.C., Starr, R., Arnold, J.R., Drake, D.M., Englert, P.A.J., Metzger, A.E., Mitrofanov, I., Trombka, J.I., d’Uston, C., Wänke, H., Gasnault, O., Hamara, D.K., Janes, D.M., Mircialis, R.L., Maurice, S., Mikheeva, I., Taylor, G.J., Tokar, R., Shinohara, C. (2002). Distribution of Hydrogen in the Near-Surface of Mars: Evidence for Subsurface Ice Deposits. Science 297, pp. 81-85.

Brasier, B.D., Green, O.R., Jephcoat, A.P., Kleppe, A.K., Van Kranendonk, M.J., Lindsay, J.F., Steele, A., Grassineau, N.V. (2002). Questioning the evi-dence for Earth’s Oldest Fossils, Nature 416., pp. 76-81.

Cabrol, N.A., Grin, E.A. (1999). Distribution, Classification, and Ages of Martian Impact Crater Lakes. Icarus 142, pp. 160-172.

Cabrol, N.A., Grin, E.A. (2001). The Evolution of Lacustrine Environments on mars: Is Mars Only Hydrologically Dormant? Icarus 149, pp. 291-328.

19

Cabrol, N.A., Wynn-Williams, D.D., Crawford, D.A., Grin, E.A. (2001). Recent Aqueous Environments in Martian Impact Craters: An Astrobiological Perspective. Icarus 154, pp. 98-112.

Cady, S.L., Farmer, J.D. (1998). Fossilization Processes in Siliceous Thermal Springs: Trends in Preservation Along Thermal Gradients. In Bock, G., and Goode, J. (eds.), Evolution of Hydrothermal Ecosystems on Earth (and Mars?), Chichester, John Wiley & Sons, pp. 150-173.

Cady, S.L., Farmer, J.D., Grotzinger, J.P., Schopf, J.W., Steele, A., (2003). Morphological Biosignatures and the Search for Life on Mars. Astrobiology 3, pp. 351-368.

Farmer, J.D. (1996) Hydrothermal systems on Mars: an assessment of present evidence. In Bock, G., and Goode, J. (eds.), Evolution of Hydrothermal Ecosystems on Earth (and Mars?), Chichester, John Wiley & Sons, pp. 273–299.

Farmer, J.D. (1998) Thermophiles, early bio-sphere evolution, and the origin of life on Earth; implications for the exobiological exploration of Mars. Journal of Geophysical Research 103E, pp. 28457–28461.

Feldman, W.C., Boynton, R.L., Tokar, R.L., Prettyman, T.H., Gasnault, O., Squyres, S.W., Elphic, R.C., Lawrence, D.J., Lawson, S.L., Maurice, S., McKinney, G.W., Moore, K.R., Reedy, R.C. (2002). Global Distribution of Neutrons from Mars: Results from Mars Odyssey. Science 297, pp. 75-78.

Gibson, E.K., McKay, D.S., Thomas-Keprta, K.L., Wentworth, S.J., Westall, F., Steele, A., Romanek, C.S., Bell, M.S., Toporski, J. (2001). Life on Mars: evaluation of the evidence within Martian meteorites ALH84001, Nakhla, and Shergotty. Precambrian Research 106, pp. 15-34.

Grieve, R.A.F. (1987). Terrestrial impact struc-tures. Annual Review of Earth and Planetary Sciences 15, pp. 245-270.

Hayes, J.M. (2001). Fractionation of Carbon and Hydrogen Isotopes in Biosynthetic Processes. In Valley, J.W. and Cole, D. (Eds.), Stable Isotope

Geochemistry, Reviews in Mineralogy 43, pp. 225-277.

Hjelmqvist, S. (1966) Beskrivning till berggrund-skarta över Kopparbergs län. Swedish Geological Survey Ca 40, 1–127.

Hode, T., von Dalwigk, I., Broman, C.A. (2003). Hydrothermal system associated with the Siljan impact structure, Sweden - Implications for the search for fossil life on Mars. Astrobiology 3, pp. 271–289.

House, C.H., Schopf, J.W., McKeegan, K.D., Coath, C.D., Harrison, T.M., Stetter, K.O. (2000). Carbon isotopic composition of individ-ual Precambrian microfossils. Geology 28(8), pp. 707-710.Hugo, R.C., Cady, S.L. (2004). Preparation of Geological and Biological TEM Specimens by Embedding in Sulfur. Microscopy Today 12(5), pp 28-30.

Högdahl, K., Gromet, P.L., Broman C. (2001). Low P-T Caledonian resetting of U-rich Paleoproterozoic zircons, central Sweden. American Mineralogist 86, pp. 534-546.

Kenkmann, T., von Dalwigk, I. (2000). Radial transpression ridges: A new structural feature of complex impact craters. Meteoritics and Planetary Science 35(6), pp. 1189-1201.

Komatsu, G., Ori, G.G. (2000). Exobiological implications of potential sedimentary deposits on Mars. Planetary and Space Science 48, pp. 1043-1052.

Komor, S.C., Valley, J.W., Brown, P.E., Collini, B. (1988). Fluid Inclusions in Granite from the Siljan Ring Impact Structure and Surrounding Regions. In Bodén, A., and Eriksson, K.G. (eds.), Deep drilling in crystalline bedrock; Volume 1, The deep gas drilling in the Siljan impact struc-ture, Sweden and astroblemes, proceedings of the international symposium: New York, Springer Verlag, pp. 181–208.

Konhauser, K.O., Phoenix, V.R., Bottrell, S.H., Adams, D.G., Head, I.M. (2001). Microbial-silica interactions in Icelandic hot spring sinter: Possible analogues for some Precambrian siliceous stro-

20

matolites. Sedimentology 48(2), pp. 415-433.

Kresten, P., Aaro, S. (1987). Bedrock maps 13F Falun NV, 13F Falun NO, 13F Falun SV, and 13F Falun SO. Swedish Geological Survey Serial Ai15–Ai18.

Kresten, P., Aaro, S., Karis, L. (1991). Bedrock maps 14F Rättvik NV, 14F Rättvik NO, 14F Rättvik SV, 14F Rättvik SO, 14E Mora NO, and 14E Mora SO. Swedish Geological Survey Serial Ai46–Ai51.

Kristiansson, P., Hode, T., Auzelyte, V., Elfman, M., Malmqvist, K., Nilsson, C., Pallon, J., Shariff, A., Wegdén, M. (2004). Development of a system for determination of the 13C/12C isotopic ratio with high spatial resolution. Nuclear Instruments and Methods in Physics Research Section B 219-220, pp. 561-566.

Lindblom, S. (1986). Textural and fluid inclusion evidence for ore deposition in the Pb-Zn deposit at Laisvall, Sweden. Economic Geology 81(1), pp. 46-65.

Lindblom, S., Wickman, F.E. (1985). Fluid inclu-sions in quartz from a quartz breccia in the Siljan Ring structure, central Sweden. GFF 107, pp. 53-58.

Logan, G.A., Hinman, M.C., Walter, M.R., Summons, R.E. (2001). Biogeochemistry of the 1640 Ma McArthur River (HYC) lead–zinc ore and host sediments, Northern Territory, Australia. Geochimica et Cosmochimica Acta 65, pp. 2317–2336.

Lundin, R., Barabash, S., Andersson, H., Holmström, M., Grigoriev, A., Yamauchi, M., Sauvaud, J.-A., Fedorov, A., Budnik, E., Thocaven, J.-J., Winningham, D., Frahm, R., Scherrer, J., Sharber, J., Asamura, K., Hayakawa, H., Coates, A., Linder, D.R., Curtis, C., Hsieh, K.C., Sandel, B.R., Grande, M., Carter, M., Reading, D.H., Koskinen, H., Kallio, E., Riihela, P., Schmidt, W., Säles, T., Kozyra, J., Krupp, N., Woch, J., Luhmann, J., McKenna-Lawler, S., Cerulli-Irelli, R., Orsini, S., Maggi, M., Mura, A., Milillo, A., Roelof, E., Williams, D., Livi, S., Brandt, P., Wurz, P., Bochsler P. (2004). Solar Wind–Induced Atmospheric Erosion at Mars: First Results from

ASPERA-3 on Mars Express. Science 305, pp. 1933-1936.

McKay, C.P. Stoker, C.R. (1989). The early en-vironment and its evolution on Mars; implica-tions for life. Reviews of Geophysics 27(2), pp. 189–214.

Mitrofanov, I., Anfimov, D., Kozyrev, A., Litvak, M., Sanin, A., Tret’yakov, V., Krylov, A., Shvetsov, V., Boynton, W.V., Shinohara, C., Hamara, D., Saunders, R.S. (2002). Maps of Subsurface Hydrogen from the High-Energy Neutron Detector, Mars Odyssey. Science 297, pp.78-81.

Nealson, K.H. (1999). Post-viking microbiology: New approaches, new data, new insights. Origins of Life and Evolution of the Biosphere 29(1), pp. 73–93.

Newsom, H.E. (1980). Hydrothermal Alteration of Impact Melt Sheets with Implications for Mars. Icarus 44, pp. 207-216.

Newsom, H.E., Brittelle, G.E., Hibbitts, C.A. (1996). Impact crater lakes on Mars. Journal of Gephysical Research 101, pp. 14951-14955.

Osinski, G.R., Spray, J.G., Lee, P. (2001). Impact-induced hydrothermal activity within the Haughton impact structure, arctic Canada: Generation of a transient, warm, wet oasis. Meteoritics and Planetary Science 36, pp. 731-745.

Rasmussen, B. (2000). Filamentous microfossils in a 3,235-million-year-old volcanogenic massive sulphide deposit. Nature 405, pp. 676-679.

Renaut, R.W., Jones, B., Tiercelin, J.J. (1998). Rapid in situ silicification of microbes at Loburu hot springs, Lake Bogira, Kenya Rift Valley. Sedimentology 45, pp. 1083-1103.

Schopf, J.W. and Klein, C. (eds.) (1992). The Proterozoic Biosphere. Cambridge U.P., Cambridge., 1348 pp.

Segura, T.L., Toon, O.B., Colaprete, A., Zahnle, K. (2002). Environmental Effects of Large Impacts on Mars. Science 298, pp. 1977-1980.

21

Simoneit, B.R.T., Summons, R.E., Jahnke, L.L. (1998). Biomarkers as tracers for life on early Earth and mars. Origins of Life and Evolution of the Biosphere 28(4–6): 475–483.

Smith, M.P. (1996). Fluid inclusion volatile well logs of the Gravberg #1 Well, Siljan Ring, Sweden. In Brown, P.E. and Hagemann, S.G. (eds.) Sixth biennial Pan-American conference on research on fluid inclusions. Program and abstracts, pp. 122.

Squyres, S.W., Grotzinger, J.P., Arvidson, R.E., Bell III, J.F., Calvin, W., Christensen, P.R., Clark, B.C., Crisp, J.A., Farrand, W.H., Herkenhoff, K.E., Johnson, J.R., Klingelhöfer, G., Knoll, A.H., McLennan, S.M., McSween Jr., H.Y., Morris, R.V., Rice Jr., J.W., Rieder, R., Soderblom L.A. (2004). In Situ Evidence for an Ancient Aqueous Environment at Meridiani Planum, Mars. Science 306, pp. 1709-1714.

Steele, A., Toporski, J.K.W., Avci, R., Guidry, S., McKay, D.S. (2001). Time of flight secondary ion mass spectrometry (ToFSIMS) of a number of hopanoids. Organic Geochemistry 32, pp. 905-911.

Stetter, K.O. (1996). Hyperthermophiles in the history of life. In Bock, G., and Goode, J. (eds.), Evolution of Hydrothermal Ecosystems on Earth (and Mars?), Chichester, John Wiley & Sons, pp.1–23.

Sturkell, F.F.E., Broman, C., Forsberg, P., Torssander, P. (1998). Impact-related hydrothermal activity in the Lockne impact structure, Jämtland, Sweden. European Journal of Mineralogy 10, pp. 589-606.

Summons, R.E., Jahnke, L.L., Simoneit, B.R.T. (1996). Lipid biomarkers for bacterial ecosys-tems: studies of cultured organisms, hydrothermal environments and ancient sediments. In Bock, G., and Goode, J. (eds.), Evolution of Hydrothermal Ecosystems on Earth (and Mars?), Chichester, John Wiley & Sons, pp. 174–194.

Thomas-Keprta, K.L., Bazylinski, D.A., Kirschvink, J.L., Clemett, S.J., McKay, D.S., Wentworth, S.J., Vali, H., Gibson, E.K. Jr., Romanek, C.S. (2000). Elongated prismatic mag-netite crystals in ALH84001 carbonate globules:

Potential Martian magnetofossils. Geochimica et Cosmochimica Acta 64(23), pp. 4049-4081.

Toporski, J.K.W., Steele, A., Westall, F., Avci, R., Martill, D.M., McKay, D.S. (2002a). Morphologic and spectral investigation of exceptionally well-preserved bacterial biofilms from the Oligocene Enspel formation, Germany. Geochimica et Cosmochimica Acta 66 (10), pp. 1773-1791,

Toporski, J.K.W., Steele, A., Westall, F., Thomas-Keprta, K.L., McKay, D.S. (2002b). The Simulated Silicification of Bacteria – New Clues to the Modes and Timing of Bacterial Preservation and Implications for the Search for Extraterrestrial Microfossils. Astrobiology 2(1), pp. 1-26.

Toporski, J.K.W., Steele, A. (2004). Characterization of purified biomarker com-pounds using time of flight-secondary ion mass spectrometry (ToF-SIMS). Organic Geochemistry 35, pp. 793-811.

Walter, M.R., Des Marais, D.J. (1993). Preservation of biological information in thermal spring depos-its: Developing a strategy for the search of a fossil record on Mars. Icarus 101, pp. 129–143.

Walter, M.R. (1996). Ancient hydrothermal eco-systems on Earth: a new palaeobiological frontier. In Bock, G., and Goode, J. (eds.), Evolution of Hydrothermal Ecosystems on Earth (and Mars?), Chichester, John Wiley & Sons, pp.112-127.

Walter, M. (1999). The Search for Life on Mars. Perseus, Cambridge, MA, 170 pp.

Westall, F., Steele, A., Toporski, J., Walsh, M., Allen, C., Guidry, S., McKay, D.S., Gibson, E., Chafetz, H.S. (2000). Polymeric substances and biofilms as biomarkers in terrestrial materials; im-plications for extraterrestrial samples. Journal of Geophysical Research E105, pp. 24511–24527.

Westall, F., de Wit, M.J., Dann, J., de Ronde, C.E.J., Gaast, S.J., Gerneke, D. (2001). Early Archean fossil bacteria and biofilms in hydrother-mally influenced sediments from the Barberton greenstone belt, South Africa. Precambrian Research 106, pp. 93–116.