51st Lunar and Planetary Science Conference (2020) 1146.pdf

TRANSFORMATION OF IRON(III) OXIDE MINERALS IN MARS-RELEVANT FLUIDS: IMPLICATIONS FOR DIAGENETIC HEMATITE FORMATION. A. L. Knight, K. Mitra, and J. G. Catalano, Department of Earth & Planetary Sciences, Washington University, Saint Louis, MO 63130 USA ([email protected])

Introduction: The distinctive reddish hue of the bearing phases (e.g., goethite [5, 19-21], ferrihydrite, Martian surface is attributed to abundant iron akageneite, magnetite [22]) or coarsening of the fine- (oxyhydr)oxide minerals. These phases have been grained red hematite [22] that is pervasive throughout identified in numerous rock, soil, sediment, and red units in the VRR as well as the underlying Murray meteorite samples across Mars [1, 2] and are possible formation. mineralogical indicators of past aqueous and In this study we explore transformation of goethite atmospheric conditions on Mars. Hematite (α-Fe2O3) is and “red” hematite into more crystalline “gray” thermodynamically the most stable Fe(III)-oxide hematite by plausible Mars-relevant diagenetic fluids. mineral on Mars [3]. Most other Fe(III) minerals exist The effect of these diagenetic fluids on the mineralogy metastably on the Martian surface and can serve as and spectral characteristics of the mineral products are precursor minerals that can undergo diagenetic investigated. Analyses specifically focused on assessing transformation to stable hematite [4-7]. Hematite has crystallinity and reflectance properties after aging at been reported in various locations on Mars including elevated temperature in potential diagenetic fluids. Meridiani Planum and Gale crater [8-10]. Methods: Laboratory experiments were conducted The Mars Science Laboratory (MSL) Curiosity to investigate the transformation of Fe(III) oxide rover recently investigated the Vera Rubin Ridge minerals in Mars-relevant fluids at acidic and neutral (VRR), a topographically-distinct hematite-bearing pH. Goethite (α-FeOOH) and hematite were prepared as sedimentary ridge on Aeolis Mons [11]. VRR yielded starting minerals using standard laboratory preparation strong spectral signatures of red crystalline hematite in techniques [5]. These minerals were dried, ground in a Compact Reconnaissance Imaging Spectrometer for mortar and pestle, and analyzed using X-ray diffraction Mars (CRISM) [8, 12] images prior to MSL exploration. (XRD) to confirm the mineralogy. Experimental Detailed in situ analysis of the VRR formation by MSL suspensions of 10 mL volume were prepared containing inferred this unit primarily consisted of fine-grained, red 20 g/L of the respective minerals in 1 M MgSO4 or and gray (to blue) lacustrine mudstones with thin, MgCl2 background salt solutions. Sulfate and chloride parallel stratification. While red hematite is dispersed fluids have been proposed to be the dominant types on throughout the VRR bedrock, the decameter-scale gray Mars [23]. The pH values of the suspensions were set to to blue zones are dominantly found in the Jura Member ~7 or ~3 using small additions of hydrochloric acid and [13]. The gray color variation crosscuts strata in the sodium hydroxide. The suspensions were sealed in otherwise red-colored VRR, indicating a post- PTFE-lined bombs (Parr Instrument Co.) and heated in depositional (i.e., diagenetic) origin. Grain-size digital convection ovens. The bombs were aged at 98C measurements point towards possible coarsening and 200C for ~50 and ~30 days, respectively. Two associated with gray hematite [13, 14]. The abundant different temperature conditions were investigated to iron-related diagenetic features throughout the VRR and explore diagenesis at elevated temperature and to the possible correlation between grain-size and spectral accelerate the transformation process. reflectance of the hematite could indicate either syn- or At the end of the experiments, the bombs were post-depositional diagenesis. Gray hematite on Mars removed from the ovens and allowed to cool before has also been found in Sinus Meridiani, Valles extracting the minerals via syringe filtration using 0.22 Marineris, and Aram Chaos [15]. μm MCE filters. The filtered minerals were rinsed with Unequivocal evidence of diagenesis on the Martian ultrapure water and dried by heating in an oven at 70C surface has been documented by various orbiter and for ~1 hour. The washed samples were then analyzed rover studies [16, 17]. The sediments in Gale crater are using XRD and VNIR spectroscopy. A portion of the lacustrine and have experienced multiple episodes of recovered minerals were ground in a mortar and pestle post-depositional diagenesis. [9, 10, 18]. The chemical to analyze using XRD. Quantitative mineral abundances composition and nature of the diagenetic fluids that potentially altered the Gale crater sediments, including were determined via Rietveld refinement using Profex the VRR, are unclear. Diagenetic fluids that [24]. The remainder was not ground, with the recrystallized pre-existing Fe(III)-bearing phases under reflectance spectra of the intact minerals acquired using ambient to hydrothermal conditions may explain the a lab-based Analytical Spectral Devices (ASDs) observed features in the VRR. Coarse-grained gray portable VNIR spectroradiometer. hematite that cross-cuts strata exposed on the VRR Results and Discussion: The experiments could reflect transformation of precursor Fe(III)- containing goethite as the starting mineral showed 51st Lunar and Planetary Science Conference (2020) 1146.pdf

changes in mineralogy at the end of the experiments at potentially showed changes in reflectivity. However, both 98 and 200C (Fig. 1). The 98C experiments the volume of minerals recovered from the experiments yielded lesser amounts of hematite when starting with was inadequate to fill the field of view of the ASD goethite compared to the corresponding 200C studies. spectrometer, causing artifacts in the data. A detailed This provides a qualitative assessment of the impact of relationship between the changes seen in XRD and the temperature on transformation timescales during reflectance properties of the solid has thus not yet been diagenesis. As hematite is thermodynamically stable at established. Replicate experiments are being conducted both temperatures, and thus should be the sole iron with greater initial mineral mass to permit improved oxide present, it indicates that under the fluid conditions spectral characterization. Additionally, experiments are studied prolonged heating at elevated temperature is underway to investigate the transformation of required to drive its formation from a goethite precursor. nanophase hematite, ferrihydrite, akaganeite, jarosite, The experiments at 200C demonstrated substantial and magnetite in various fluid compositions. variations in mineralogy with pH and fluid composition. Conclusion: The rate of transformation of goethite Acidic solutions showed near-quantitative conversion to hematite is influenced by fluid-type, solution pH, and of goethite into hematite in both chloride and sulfate temperature. The coarsening of fine-grained red fluids. In contrast, pH 7 solutions yielded only partial hematite to gray hematite may also show similar transformation to hematite, with a greater extent of variations with fluid-type, solution pH, and conversion in sulfate fluids. These studies indicate that temperature. Diagenesis in SO4-rich solutions at low pH fluid composition affects the rate of iron oxide conditions appear most favorable for driving diagenetic transformation during diagenesis. It also suggests that recrystallization to gray hematite. However, further other chemical species in aqueous fluids may accelerate study is necessary to establish the spectral properties of diagenetic transformations, motivating future study. hematite produced under such conditions and All experiments with “red” hematite showed minor recrystallization in alternative fluid compositions needs sharpening of hematite peaks in XRD. However, these to be explored. were still broader than peaks from hematite generated References: [1] Morris R. V. et al. (2006) JGRP, by conversion from goethite, indicating a larger 111, E02S13. [2] Bridges J. C. et al. (2001) Space Sci. coherent domain size in the latter experiments. The Rev., 96, 365-392. [3] Gooding J. L. (1978) Icarus, 33, VNIR spectra of thermally-transformed hematite 483-513. [4] Burns R. G. and Fisher D. S. (1993) JGRP, 98, 3365-3372. [5] Cornell R. M. and Schwertmann U. (2003) Wiley and Sons. [6] Glotch T. D. and Kraft M. (2008) Phys. Chem. Miner., 35, 569-581. [7] Golden D. C. et al. (2008) Am. Miner., 93, 1201-1214. [8] Fraeman A. A. et al. (2013) Geology, 41, 1103-1106. [9] Rampe E. B. et al. (2017) EPSL, 471, 172-185. [10] Yen A. S. et al. (2017) EPSL, 471, 186-198. [11] Heydari E. et al. (2018) LPS XLIX, Abstract #1817. [12] Fraeman A. A. et al. (2016) JGRP, 121, 1713-1736. [13] Horgan B. H. N. et al. (2019) submitted to JGRP. [14] Bennett K. A. et al. (2018) LPS XLIX, Abstract #1769. [15] Christensen P. R. et al (2001) JGRP, 106, 23873-23885. [16] Grotzinger J. P. et al. (2015) Science, 350, aac7575. [17] McLennan S. M. et al. (2005) EPSL, 240, 95-131. [18] Hurowitz J. A. et al. (2017) Science, 356, eaah6849. [19] Zolotov M. Y. and Shock E. L. (2005) Geo. Res. Lett., 32, L21203. [20] Goss C. J. (1987) Mineral. Mag., 51, 437-451. [21] Christensen P. R. and Ruff S. W. (2004) JGRP, 109, E08003. [22] Walker T. R. et al. (1981) JGR, 86, 317-333. [23] Newsom H. E. et al (1999) JGRP, 104, 8717-8728. [24] Doebelin N.

Figure 1: Final mineral proportions of precipitates and Kleeberg R. (2015) J. App. Cryst., 48, 1573-1580. obtained by the transformation of goethite to hematite in Acknowledgements: Ray Arvidson and Paul 10-1 M MgCl2 and MgSO4 at initial pH 3 and 7 at 98 and Carpenter are thanked for their assistance with mineral 200°C determined via Rietveld refinement. characterization.