Formation of jarosite-bearing deposits through aqueous oxidation of pyrite at Meridiani Planum, Mars

Authors


Abstract

[1] The discovery of sulfate-rich layered deposits with hematite spherules at the landing site of the Opportunity rover is consistent with mineral deposition in an aqueous environment. We evaluate conditions responsible for the formation of a jarosite-goethite-gypsum assemblage with speciation calculations. The results show that the assemblage could have precipitated from acidic solutions formed through near-surface aqueous oxidation of pyrite. Our hypothesis is that regional heating in the Meridiani Planum caused a release of sulfide-rich hydrothermal waters, leading to formation of pyrite-rich regional deposits in a depression. Aqueous oxidation of these deposits by atmospheric O2 created an acidic environment that allowed formation of sulfates and goethite. Partial neutralization of the solution caused further goethite precipitation and conversion of jarosite to goethite, leading to formation of goethite concretions. Subsequent dehydration of goethite to coarse-grained hematite would also have been facilitated by regional heating.

1. Introduction

[2] Remote sensing studies of the hematite-rich region at the Meridiani Planum on Mars are consistent with deposition of ferric oxyhydroxides in a water-filled basin over an area of about 0.15 Mkm2 followed by their dehydroxylation to hematite in a thin (<200 m) sedimentary layer [Christensen and Ruff, 2004]. Deposition in a lake or near-surface groundwater environment is supported by the discovery of layered deposits that contain 30–40 wt. % of Ca-, Mg-, and Fe-sulfates with imbedded hematite spherules (Burns formation) at the landing site of the Opportunity rover that landed in the hematite-rich region [Squyres et al., 2004; McLennan et al., 2005]. The detection of jarosite, (K, Na, H3O)FeIII3(SO4)2(OH)6, in these rocks may indicate acidic and oxidizing aqueous environments [Klingelhöfer et al., 2004], based on corresponding terrestrial mineral assemblages and related fluid chemistry [Dutrizac and Jambor, 2000]. Burns and Fisher [1990] suggested that martian jarosite is likely to be formed through aqueous oxidation of Fe sulfides. Here we evaluate conditions that could have been responsible for the formation of a jarosite-goethite-gypsum (JGG) assemblage, estimate changes in solution chemistry that could have caused mineral precipitation, dissolution and replacement, and discuss the origin of the mineralogy of the Burns formation.

2. Models for Water-Mineral Interactions at Martian Conditions

[3] Conditions of mineral precipitation can be determined by calculating equilibrium compositions among solutions, dissolved gases, and solid phases in hypothesized martian environments. We explored variations in the ranges of water to rock (W/R) mass ratio, pH, K content, and oxidation state in speciation models involving potassium jarosite, a JGG assemblage, and pyrite. The calculations of thermochemical equilibria among non-ideal aqueous solutions, dissolved gases, and one-component solids were performed with the GEOCHEQ code [Mironenko et al., 2000] using thermodynamic data for solutes and jarosite from Shock et al. [1989, 1997] and Stoffregen et al. [2000], respectively. Bulk composition of the water-rock system was varied as an input constraint. Amounts, concentrations and activities (a) of solutes and amounts of formed solids were among the output data. Activity of H+ and fugacity (f) of gases were either set (open systems) or calculated (closed systems). Low solubilities of major minerals allowed us to use the Debye-Hückel model for activities of ions (a consideration of highly soluble Mg sulfates would require the Pitzer model). In open system models for surface waters, solutions were allowed to saturate with respect to O2 and CO2 at the fugacities of the present atmosphere. Since CO2 had no effect on our results we do not discuss carbon speciation. All calculations are performed for 1 bar total pressure.

2.1. Jarosite Stability in Solutions

[4] Aqueous compositions that accompany jarosite precipitation and dissolution in contact with the martian atmosphere were calculated for the jarosite-H2O system over a range of W/R mass ratios, where R represents jarosite. The results show that at W/R < 115, 25°C, and fO2 = 10−5 jarosite coexists with goethite and an acidic solution of a uniform composition (Table 1, column 1). The solution chemistry is controlled by equilibria among jarosite and goethite given by the reaction

equation image

Although solution chemistry remains unaffected until jarosite completely dissolves at W/R = 116, the jarosite/goethite ratio decreases as the W/R increases (Figure 1a). Once all jarosite converts to goethite, concentrations of solutes (except O2) decrease because of dilution. In other words, addition of liquid water causes conversion of jarosite to goethite while H2O consumption (e.g., through evaporation, hydration or freezing) favors precipitation of jarosite. A similar pattern of mineral substitution is observed for the JGG assemblage (Figure 1b). If acidity is controlled by the solution itself and aH+ is fixed, jarosite is only stable in a range of low pH (Figure 2). Lower W/R ratios make jarosite stable over a broader pH range. Goethite forms together with jarosite, except at very low pH, and the jarosite/goethite ratio decreases as the pH increases. Although solution chemistry is affected by the pH, W/R ratio, and mineral precipitation, major solutes in equilibrium with jarosite remain the same, except Fe3+ (Figure 2c).

Figure 1.

Volumes of minerals formed through reactions of jarosite-bearing rocks with water as functions of the W/R ratio at 25°C, and fO2 = 10−5. The “rock” (1 kg) is represented by (a) pure jarosite or (b) a jarosite-goethite-gypsum mixture with mole ratios of 2:1:1. In Figure 1a, goethite is present in negligible amounts at low W/R ratios.

Figure 2.

The (a) stability fields, (b) volumes of minerals, and (c) solution composition for the water-jarosite system as functions of pH and W/R ratio at 25°C, and fO2 = 10−5. Figures 2b and 2c are for W/R = 100, and the vertical dotted lines show the boundaries of mineral stability.

Table 1. Equilibrium Concentrations of Major Solutes (mol (kg H2O)−1) and Volumes of Precipitated Minerals (cm3) at Assumed Martian Environmental Conditionsa
 123456
  • a

    Col. 1, Aqueous solution in equilibrium with jarosite. Col. 2, Solutions formed through dissolution of pyrite in closed system. Col. 3–4, products of pyrite oxidation at different water/pyrite ratios. Col. 5, solution (without Cl) and minerals formed through addition of 0.01 mol (kg H2O)−1 KCl to the solution presented in col. 3. Col. 6, aqueous solution formed through interaction of jarosite (2 moles), goethite (1 mole), and gypsum (1 mole) with water at the fixed pH of 1.6. All data, except col. 2, represent hypothetical martian surface conditions. Concentrations less that 10−10 are not shown. Jar, jarosite; gth, goethite; gyp, gypsum. Volumes of minerals are in parentheses.

T, °C250002525
W/R1–1150.1–10001001000-30
Log fO2−5.0−77.4−5.0−5.0−5.0−5.0
H+3.2 × 10−23.2 × 10−88.7 × 10 −22.6 × 10−20.143.2 × 10−2
HSO41.9 × 10−2-8.1 × 10 −26.7 × 10−20.112.2 × 10−2
K+1.6 × 10−2---2.0 × 10−41.4 × 10−2
SO42−1.5 × 10−21.8 × 10−98.6 × 10 −21.0 × 10−23.5 × 10−21.1 × 10−2
KSO47.8 × 10−4---1.3 × 10−57.8 × 10−4
Fe3+1.1 × 10−4-5.5 × 10−23.7 × 10−41.7 × 10−21.4 × 10−4
FeOH2+9.3 × 10−6-1.7 × 10−49.4 × 10−62.1 × 10−49.9 × 10−6
KHSO4, aq6.9 × 10−8---3.7 × 10−96.5 × 10−8
FeO+6.7 × 10−8-7.4 × 10−82.2 × 10−82.8 × 10−76.7 × 10−8
OH--3.6 × 10−7----
Fe2+-7.1 × 10−91.0 × 10−9-1.0 × 10−8-
HS-6.8 × 10−9----
H2S, aq-5.6 × 10−9----
Ca2+-----1.2 × 10−2
CaSO4-----4.0 × 10−3
O2,aq1.3 × 10−88 × 10−812.2 × 10−82.2 × 10−81.3 × 10−8-
pH1.67.51.21.71.01.6
Ionic strength0.06<0.010.500.040.280.098
jarpyritegth (57)gth (166)jar (1.5)jar (222)
gth---gth (0.73)gth (57)
-----gyp (30)

[5] These models illustrate that relative abundances and substitution patterns among jarosite and goethite can be used to reconstruct solution chemistry, amounts of water, and pH, as well as temporal changes in these parameters. The stability of jarosite at low pH and low W/R ratios is consistent with the presence of jarosite in terrestrial acid soils and weathering crusts, acid mine drainage, and in evaporites of acid saline lakes [Dutrizac and Jambor, 2000].

2.2. Oxidation of Pyrite and Formation of Acidic Solutions

[6] Interactions of pyrite with deep subsurface waters and surface solutions were modeled in closed and open systems, respectively. In the closed system, inefficient low-temperature dissolution of pyrite leads to dilute alkaline and reduced solutions with a uniform composition over a broad range of water/pyrite ratios (Table 1, col. 2). Results for the open system show that increases in fO2 cause dissolution of pyrite followed by oxidation of Fe2+ to Fe3+ and precipitation of ferrihydrite, Fe(OH)3, that converts to stable goethite. After pyrite dissolves, highly acidic solutions form, in which H+, HSO4, SO42−, Fe2+ and/or Fe3+ are the major ions. These results are illustrated in Figure 3 and are generally consistent with experimental studies [Williamson and Rimstidt, 1994; Kamei and Ohmoto, 2000]. Although the water/pyrite ratio has a mild effect on solution composition, concentrations of major solutes and H+ are higher at low water contents (Table 1, col. 3, 4).

Figure 3.

Equilibrium concentrations of (a) solutes and (b) volumes of minerals formed in the H2O-FeS2 system as functions of log fO2 at W/R = 100, and 0°C. The left dotted line represents pyrite dissolution in the closed system. The right margin of the plot stands for the composition saturated with atmospheric O2.

[7] Investigations of acid mine drainage environments [e.g., Langmuir, 1997; Nordstrom et al., 2000] show that the formation of low-pH fluids is accounted for by the reactions

equation image
equation image
equation image

In addition, the presence of H2S(aq) and oxidation of H2S(aq) by O2 leads to low-pH solutions. Although oxidation of pyrite, H2S(aq), and Fe2+ are commonly bio-mediated, the abiotic processes occur more slowly [Williamson and Rimstidt, 1994; Millero and Hershey, 1989; Millero et al., 1987]. On Earth, halflives of abiotically oxidized Fe2+ and H2S(aq) in surface waters are on the order of minutes and hours, respectively. On Mars, the oxidation rates would be slower by a factor consistent with the value fO2, Earth (fO2, Mars)−1 = 2 × 104 [Burns, 1993]. Lower oxidation rates on Mars imply that reduced solutes (Fe2+, H2S, HS) would exist for some time in surface waters, especially in solutions recently released from the subsurface. High salinity and low temperatures should also slow oxidation. Note however, that slow oxidation of Fe2+, which is a limiting stage in the oxidation of pyrite, does not prevent formation of acidic solutions via equation (2). Even very low fO2 in martian solutions would allow dissolution of pyrite and release of H+ (Figure 3). Although the low mass of O2 in the atmosphere of Mars would have prevented rapid formation of abundant ferric species [Catling and Moore, 2003], slow and/or episodic oxidation could have occurred through re-supply of photochemically formed O2 and peroxides. In addition, oxidation by water would have been possible and facilitated by escape of H2: FeS2 + 8H2O → Fe2+ + 2SO42− + 2H+ + 7H2.

2.3. Formation and Transformations of the Jarosite-Gypsum-Goethite Assemblage

[8] Formation of low-pH fluids in the vicinity of the martian surface should have led to dissolution of silicates and carbonates. Dissolution of plagioclase would release Ca, Na, and K into the solution followed by deposition of sparingly soluble sulfates (gypsum), which can precipitate together with ferric hydroxide. By analogy to terrestrial acidic environments, the amount of K in the solution could be a limiting factor that permits formation of K-jarosite. Addition of K (in form of KCl) to a solution formed through oxidation of FeS2 causes deposition of jarosite and an increase in the jarosite to goethite ratio until jarosite becomes the only precipitate (Figure 4). These calculations illustrate that the conversion of goethite to jarosite in sediments could be caused by changes in dissolved alkali content.

Figure 4.

Equilibrium concentrations of (a) solutes and (b) volumes of minerals formed in the Fe-S-K-Cl-H2O system as functions of bulk KCl content at 25°C, and log fO2 = −5. The model represents addition of K to the solution formed through oxidation of pyrite at 0°C, log fO2 = −5, at W/R = 100 (see Figure 3a and Table 1, col. 3).

[9] Stability calculations for the JGG assemblage as a function of pH show that all three minerals coexist only in a narrow range of pH (Figure 5). Overall, the results presented in Figures 1b and 5 demonstrate that the JGG assemblage is stable at low pH and low W/R. If all these minerals are present, an increase in W/R would lead to conversion of jarosite into goethite and can cause dissolution of gypsum. A decrease in the pH causes dissolution of goethite and gypsum but can increase the amount of jarosite, until it too dissolves at very low pH. An increase in pH makes goethite and gypsum stable but dissolves jarosite.

Figure 5.

Equilibrium concentrations of (a) solutes and (b) volumes of minerals formed in the Fe-S-K-Ca-H2O system as functions of the pH at 25°C, log fO2 = −5, and W/R = 50. The “rock” composition is as in Figure 1.

[10] Although genetic relationships among jarosite and hematite in the Meridiani rocks are unknown, formation of FeIII-rich spherules implies an excess of Fe3+ at the time of their formation. One possibility is that partial dissolution of jarosite via equation (1) caused precipitation of the spherules. Dilution of solution by water and/or an increase in the pH can drive this reaction (Figure 6). In particular, even moderate neutralization of solution through dissolution of silicates would have caused precipitation of ferrihydrite/goethite and partial conversion of jarosite to goethite and gypsum. Water consumption through evaporation and hydration can also contribute to an increase in concentrations of cations leading to higher pH and formation of corresponding sulfate minerals together with goethite (path III in Figure 6).

Figure 6.

The stability fields of minerals in the Fe-S-K-Ca-H2O system as functions of the pH and W/R ratio at 25°C and log fO2 = −5. The “rock” is as in Figure 1. The grey arrows show paths of jarosite-goethite conversion through dilution (I, II) and neutralization (II, III). Path III also represents consumption of solution. Paths I and II cause gypsum dissolution. The composition of the starting point for the paths is from Table 1, col. 6.

3. An Origin of the Burns Formation

[11] On Earth, acidic surface waters exist as local phenomena related either to dissolution and oxidation of volcanic gases (H2S, HCl, HF, and SO2) in caldera lakes [Varekamp et al., 2000] and hot springs or to oxidation of iron sulfides [Langmuir, 1997; Nordstrom et al., 2000]. Since the hematite-rich site at Meridiani does not resemble a volcanic caldera in size and shape, aqueous oxidation of iron sulfides (pyrite, pyrrhotite) is the likely source of acidity. By analogy to Earth, regional sulfide-rich deposits at Meridiani could have formed through sulfate reduction in organic-rich sediments (unlikely), accumulation of sulfides in magma chambers or lava flows [Burns and Fisher, 1990] (consistent with elevated Ni content in Meridiani rocks and soils), and hydrothermal processes developed above a magma body or an impact melt. In hydrothermal fluids, reducing conditions and low solubilities of Ca sulfates should have caused preferential transport of aqueous sulfides (H2S, HS) followed by rapid deposition of FeS2 from cooling fluids closer to the surface, and even at the bottom of a lake sourced by hydrothermal fluids. Although aqueous oxidation of H2S(aq), HS, and FeS2 could have occurred during hydrothermal activity, pyrite oxidation could also have been caused by another water release after a dry period.

[12] Our hypothesis is that regional heating of the ice-bearing crust in Meridiani Planum caused a release of subsurface waters rich in aqueous sulfides, rapid precipitation of pyrite and formation of regional pyrite-rich deposits in the near-surface, slow oxidation of pyrite and aqueous sulfides in near-surface aqueous environments, and precipitation of sulfates and Fe(OH)3 (which converts to goethite). Elevated salinities and pH values of about 1–3 were necessary for jarosite deposition. Neutralization of solution through dissolution of silicates caused further formation of goethite, partial conversion of jarosite to goethite, and favored formation of FeIII-rich concretions. Finally, consumption of solution was followed by dehydration of minerals to phases that are stable at the dry equatorial conditions (e.g., goethite to hematite, epsomite to kieserite). The regional heating that caused water release might have enhanced dehydration, especially hematite formation, including course-grained crystalline hematite [cf., Catling and Moore, 2003], which may have been inhibited otherwise.

Ancillary