Spectral study of Fe-rich vermiculite bearing rocks and implications for mineralogy of Oxia Planum, ExoMars 2022 Rover landing site

In 2022 rover, ESA-Roscosmos will launch rover to land at Oxia Planum on Mars. Oxia Planum is a wide, Noachian-age, phyllosilicate-bearing plain located between Mawrth and Ares Valles. The bedrock phyllosilicate deposits at Oxia Planum is Fe,Mg-rich and one of the largest exposures of this type on Mars, with a thickness of more than 10 m [1]. When characterized in the near infrared by CRISM and OMEGA instruments, the bedrock deposits exhibit absorptions that suggest the presence of Fe,Mg-rich phyllosilicates with significant amounts of Fe2+ in octahedral sites (i.e., trioctahedral) [2]. The typical spectrum is best matched by saponite or vermiculite, but an accurate spectral match is lacking so far. Because precise mineral composition of the bedrock is not fully understood, limited conclusions can be drawn regarding the aqueous evolution and habitability potential of Oxia – the main scientific aim of the ExoMars 2022 mission. To fill this gap, and to better prepare for in-situ analyses by the ExoMars2022 rover, we performed a survey of potential terrestrial analogue rocks. We have identified terrestrial deposits of Fe-rich, trioctahedral vermiculite and perform spectral comparisons with the Martian substrate. Analogues from two terrestrial sites were obtained: (1) vermiculitized chlorite-schists from Blue Spur, Otago, New Zealand and [3] (2) basaltic tuffs and pyroclastics from Granby, Massachusetts, USA, with Fe-rich clays filling amygdales [4]. The samples were characterized by X-ray diffraction to understand chemistry and structure of clay minerals and by NIR spectroscopy in order to provide further structural details and to assess the match to Oxia Planum bedrock clays. XRD was performed on bulk rock samples as well as for separated clay fraction minerals. Near Infrared (NIR) spectra for each sample were collected using a reflectance spectrometer in the near-infrared (0.8–4.2 μm) mode. Analysis was performed on powdered samples, under ambient temperature and pressure conditions. Both analogues have been added to a newly built Planetary Terrestrial Analogue Library (PTAL) rock collection and spectral library. Deposits in Blue Spur formed by erosion and deposition of greenschist-facies quartzo-feldspathic schists rich in Fe-rich chlorite. Due to rapid erosion and non-oxidizing or mild oxidative conditions [3] chlorite remained unoxidized when buried. Later interactions with groundwater caused formation of trioctahedral vermiculite and progressive alteration of deposit led to illitization and kaolinitization For our purposes we have collected samples of pristine schist clasts, greenish (unoxidized) vermiculitized material and pale, illitized clasts [3]. The XRD patterns of oriented clay fraction separates suggest the presence of chlorite in pristine material, trioctahedral vermiculite in greenish samples that underwent alteration in anoxic conditions, and interstratified illite-vermiculite in more oxidized samples.NIR spectroscopy analysis of Otago samples tends to indicate that the nature and overall Fe content is broadly similar to the phyllosilicates at Oxia. Furthermore, illitization of the vermiculite seen spectrally in Otago is likely limited for Oxia. This means that vermiculite at Oxia is a rather pure, one-phase mineral that underwent only minor (if any) post-depositional alterations toward illite. The Granby Tuff consists of basaltic flows and volcanic ash units. In places, it contains zones of vesicular basalts that have amygdales filled with calcite and dark-brown clays. The clays filling the amygdales are of apparent hydrothermal origin, that is precipitated in gas vesicles from solutions generated by hydrothermal alteration of the basaltic tuff. The clays were reported to be either saponite or vermiculite [4]. The XRD patterns of the oriented clay fraction separates from these tuffs reveal the presence of well-crystallized, trioctahedral vermiculite and various amounts of saponite in different samples. In NIR spectroscopy, Granby samples show quite a good spectral match to Oxia and indicate that saponite, present in Granby, is absent from Oxia spectra. Oxia deposits are likely pure vermiculite rather than a mixture of saponite and vermiculite. Additionally, other spectral features suggest that the Granby clays have different Fe/Mg ratios than minerals of Oxia Planum, or a different (plausibly lower) oxidation state of Fe in the clay structure. Our spectral analogue study shows that rocks containing Fe2+-rich vermiculite, locally with octahedral Al and (oxidized) Fe3+, are likely the main component of bedrock phyllosilicates at Oxia Planum. In terms of clay structure and its di- versus trioctahedral nature, Oxia deposits are matched best by a vermiculite-saponite from hydrothermally-derived settings (Granby Tuffs), although the contribution of saponite must be minor at Oxia. Comparison of Otago vermiculite to Oxia shows high Fe content in Oxia and a similar oxidation state. Spectral inconsistencies between Otago and Oxia related to presence of Al in the clay structure allow us to conclude that Oxia bedrock deposits were likely not oxidized nor illitized after formation. Acknowledgments: This project has received funding from the EU Horizon 2020 Research and Innovation Program (687302, PTAL), and the Research Council of Norway (Centres of Excellence funding 223272). References: [1] Quantin-Nataf C. et al. 2020: Astrobiology, accepted. [2] Carter J. et al. 2016. 47th Lunar and Planetary Science Conference. Abstract no. 2064. [3] Craw D. 1984. Clay Minerals 19: 509-520. [4] April R.H. and Keller D.M. 1992. Clays and Clay Minerals 40: 22-31.