Ninth International Conference on 2019 (LPI Contrib. No. 2089) 6336.pdf

PHOSPHORUS RESULTS FROM THE APXS IN CRATER: OCCURRENCE AND EVIDENCE OF IN SITU MOBILITY. J. A. Berger1, R. Gellert1, P. L. King2, D. W. Ming3, C. D. O’Connell-Cooper4, M. E. Schmidt5, J. G. Spray4, L. M. Thompson4, S. J. V. VanBommel6, A. S. Yen7. 1University of Guelph, Guelph, CAN; 2Australian National University, Canberra, AUS; 3Johnson Space Center, Houston, USA; 4University of New Brunswick, Freder- icton, CAN; 5Brock University, St. Catharines, CAN; 6Washington University, St. Louis, USA; 7JPL-Caltech, Pasa- dena, USA.

Introduction: The mobility of phosphorus on the 3b). In the haloes, P2O5 is up to 2.51 wt%, which repre- is fundamental to assessing habitability sents a 185% enrichment relative to the less-altered ad- because the element is considered essential for prebiotic jacent bedrock [2]. TiO2 has low mobility in diagenetic and biotic chemical reactions and biological structures. fluids under most Eh-pH conditions, and the increase of Phosphates can also constrain ancient aqueous condi- molar P/Ti from 1.1 ± 0.1 in the less altered Stimson tions because the stabilities of different phosphate bedrock to 1.6-1.9 in the altered haloes indicates a net phases are sensitive to pH. The Alpha Particle X-ray addition of P during formation of the haloes. Spectrometer (APXS) on the rover Curiosity measures P concentrations (reported as P2O5) above ~300 ppm, Bradbury group Mt. Sharp group Soil and the distribution of P in Gale Crater materials has 4 Sand JM Sheepbed Nova been established along the >20 km traverse. Here, we Stimson report the occurrence of P2O5 in Gale and discuss evi- 3 Bradbury dence of in situ P mobility and possible implications. Bradbury-like (wt%) Lower Murray 5 2 Upper Murray Phosphorus Occurrence in Gale Crater: Over the O 2 first 700 sols, Curiosity traversed the Bradbury group, P which consists of basaltic and alkali-rich units with a 1 relatively limited range of P2O5 concentrations (median 0 = 0.88 ± 0.22 wt%; Figure 1). This range is within that 025050075010001250150017502000 of basaltic soil, i.e., average Mars crust (0.93 ± 0.05 Sol wt%). One exception is a unique, Ca- and P-rich float Figure 1: P2O5 versus sol. The target Jones Marsh (JM; target named Nova (sol 687), which is interpreted to be denoted by arrow) with high P2O5 (7.6 ± 0.4 wt%) an apatite-rich clast with an igneous composition simi- is not shown. The brown horizontal line denotes the me- lar to the local mugearitic float in the Bradbury group. dian soil concentration; the red vertical line indicates Excluding Nova, the alkali-rich rocks of the Bradbury the transition between the Bradbury group (sols 0-700) group have lower P2O5 (median = 0.80 ± 0.16 wt%) than and Mt. Sharp group (sols 700-2300+). the soil-like basaltic rocks, including the Sheepbed member (median = 0.92 ± 0.09 wt%). (a) JM (b) Soil 4 4 Stimson The Mt. Sharp group, in contrast, has much greater 3 Lower Murray Upper Murray P2O5 variance than the Bradbury group (Figures 1, 2). 3 3 The bedrock of the Pahrump Hills member of the Mur- 1 (wt%)

(wt%) 1

5 Haloes 5 2 2 O ray fm. (lower Murray; sols ~700-1060) has higher P2O5 O 2 2 (median = 1.25 ± 0.33 wt%) than the overlying Murray P P 1 1 units (upper Murray; median = 0.93 ± 0.52 wt%). In the Bedrock 2 Murray fm., localized features occur throughout the 0 0 15 25 35 45 55 65 75 15 25 35 45 55 65 75 traverse with P2O5 enrichments >1.5 wt%. Some of SiO (wt%) SiO (wt%) 2 2 these P-rich features have corresponding Mn enrich- Figure 2: P2O5 versus silica in (a) Murray fm. and (b) ments (1-4 wt%; Figure 3). The Mn- and P-rich features Stimson fm. rocks. The black arrows and corresponding are gray nodules and gray patches on bedrock (e.g., numbers indicate three trends: (1) Bedrock with ele- Timber Point, sol 1716; Maple Spring, sol 1679; Cliff- vated SiO2 + P2O5, which occurs primarily in the lower side Bridge, sol 1681; and Jones Marsh, sol 1727). An- Murray fm. units and Stimson fm. silica-rich haloes. (2) other occurrence of P enrichment is in the Garden City Bedrock with prominent Ca-sulfate veins in the APXS vein in the Pahrump member, where the target Kern field of view. (3) Localized, P-rich diagenetic features Peak, interpreted to be altered vein-hosting bedrock [1], in the upper Murray units. Blue arrow denotes Jones has elevated P2O5 (2.72 ± 0.14 wt%). Phosphorus en- Marsh (JM). richment was also found in silica-rich fracture haloes lo- cated in the soil-like basaltic Stimson fm. (Figure 2b, Ninth International Conference on Mars 2019 (LPI Contrib. No. 2089) 6336.pdf

Discussion: The enrichment of P2O5 in haloes and still present in some Murray fm. samples [5]. We pro- localized diagenetic features is evidence that P phases pose that apatite inclusions in feldspar could be pro- were precipitated from a P-bearing solution. This indi- tected from interaction with acidic diagenetic fluids, re- cates that P was soluble and mobile during and/or after sulting in the preservation of a fraction of the primary sediment emplacement. At low temperatures, phosphate apatite. is more soluble and thus more likely to be mobile in 1 10 10 1 acidic fluids with pH < 6.5 [3]. However, in low pH flu- (a) JM (b) Soil Stimson ids with high ionic strength, Fe-phosphate (strengite and Lower Murray nanophase Fe-phosphate) has low solubility [4]. It is Upper Murray plausible that a more dilute, moderately-low pH fluid (wt%)

(wt%) Haloes 5 5 O (~5-6) dissolved phosphate-bearing phases (primarily O 2 2 P 0 P apatite) in the precursor material, and then phosphate 10 10 0 was subsequently concentrated as the fluid evolved into Bedrock

a high ionic strength fluid as water was removed by 10 -2 10 -1 10 0 10 1 10 -2 10 -1 10 0 10 1 evaporation and/or freezing. Phosphate was then finally MnO (wt%) MnO (wt%) precipitated as an insoluble Fe-phosphate in locally en- Figure 3: Log P2O5 versus log MnO in the (a) Murray riched features. This model is consistent with that pro- fm. and (b) Stimson fm. Phosphorus is enriched in lo- posed by Rampe et al. [5], who suggested that evidence calized features in the Murray fm., and some of the fea- including apparent cation leaching (Fe, Mn, Ni, Zn) and tures also have Mn enrichments (e.g., Jones Marsh; the CheMin detection of jarosite (1-3 wt%) supported JM). In contrast, alteration haloes in the Stimson fm. alteration of Murray fm. rock by acid-sulfate fluids. Fe- are depleted in Mn and enriched in P. phosphate was not detected by CheMin XRD; however, X-ray amorphous Al- and Fe-phosphates could be pre- sent [e.g., 4]. A key line of evidence might contradict the hypoth- esis that phosphates were mobilized in acidic fluids in the Murray fm. Fluorapatite, which is soluble in low pH fluid, was detected (1-2 wt%) in the same samples that contain jarosite, which only forms at pH < ~4 [5]. The two minerals are a disequilibrium assemblage because Figure 4: Thin section micrographs of (a) weathered the low pH conditions under which the secondary sul- mugearite from Kohala altered in a circumneutral re- fate mineral jarosite forms would likely have caused the gime (MK15-3c) [6] and (b) acid-sulfate altered ha- primary igneous mineral fluorapatite to dissolve. A pos- waiitic tephra from Maunakea (HWMK959) [7]. Ar- sible explanation for this disequilibrium is that the fluo- rows in (a) indicate representative acicular apatite in- rapatite occurs as inclusions in feldspar, which is rela- clusions in plagioclase. Arrows in (b) indicate repre- tively resistant to low pH alteration. Armoring of apatite sentative plagioclase laths in an altered matrix contain- by feldspar could therefore inhibit the complete disso- ing alunite and jarosite. lution of phosphates in a rock that interacts with acidic fluids. For example, some hawaiitic and mugearitic References: [1] Berger, J. A. et al. (2017). Journal flows on Hawai’i (samples collected by [6]) have acic- of Geophysical Research: Planets, 2017JE005290. [2] ular apatite inclusions in feldspar, and the feldspar is re- Yen, A. S. et al. (2017). Earth and Planetary Science sistant to alteration in circumneutral pedogenic condi- Letters. [3] Adcock, C. T.et al. (2013). Nature Geosci- tions as well as acid sulfate conditions (Figure 4). Apa- ence, 6(10), 824–827. [4] Tu et al. (2014). American tite inclusions in feldspar are plausible for martian ma- Mineralogist, 99(7), 1206–1215. [5] Rampe, E. B. et al. terials because they have been found in martian meteor- (2017). Earth and Planetary Science Letters. [6] Berger, ites [8]. J. A. (2017). Electronic Thesis and Dissertation Repos- Conclusion: Elevated P2O5 detected by the APXS itory. 4674. [7] Morris, R. V. et al. (2000). Journal of in diagenetic features and fracture-associated haloes is Geophysical Research: Planets, 105(E1), 1757–1817. evidence that phosphorus was mobile during and/or af- [8] McCubbin, F. M., & Nekvasil, H. (2015). American ter emplacement of the sediment in Gale Crater. The Mineralogist, 93(4), 676–684. mobility of P indicates fluids with pH < ~6 interacted Acknowledgements: The APXS is managed and fi- with the sediment. If igneous apatite was the primary nanced by the Canadian Space Agency. We appreciate source of mobile phosphate via chemical weathering, the support of the dedicated engineers at JPL. then dissolution was incomplete because fluorapatite is