AbstractIron oxide and hydroxide minerals, likely responsible for Mars' distinctive red color, offer critical insights into the planet's ancient and current climate, as well as its potential habitability. Several previous studies attributed Mars' reddish hue to anhydrous hematite (Fe2O3) and suggested that its formation is a geologically young process. Recent analyses by the Mars Science Laboratory (MSL) rover revealed the presence of volatiles and amorphous materials in the surface fines and dust, but mineralogy remained unresolved. Here, we present evidence that poorly crystalline ferrihydrite (Fe5O8H · nH2O) is responsible for the red color of the Martian dust, as identified through a combination of orbital (CRISM & OMEGA), in-situ (MSL ChemCam, MER Pancam and Pathfinder IMP), and laboratory visible near-infrared spectra. We employ quantitative spectral analyses, which demonstrate that among various iron oxyhydroxides, ferrihydrite is most consistent with the observed Martian dust spectra. In addition, our dehydration experiments show that ferrihydrite does not transform into other more crystalline iron oxide phases when exposed to present-day Martian conditions. The preservation of ferrihydrite until present time is inconsistent with a sustained warm climate after it was formed, since warm conditions would favor transformation into more crystalline hematite and/or goethite. We propose that the formation of abundant ferrihydrite indicates a cool, wet environment in the last stages of early Mars, favorable to oxidative conditions, followed by a transition to a hyper-arid erosional environment that has persisted to the present day.IntroductionIdentifying the dominant iron oxide phases in Martian dust can provide quantitative constraints on the planet’s ancient chemical environments and climate conditions. On Earth iron oxides form under specific environmental conditions including pH, temperature, redox state, and water availability (Cornell & Schwertmann, 2003). The reddish coloration of the Martian surface has been investigated since the early telescopic observations that hinted at the presence of impure iron ore known as limonite, which contains the crystalline iron (oxy)hydroxide mineral goethite (α-FeOOH) (Adams & McCord, 1969; Dollfus, 1957; Sagan et al., 1965; Sharonov, 1961). Subsequent ground-based telescopic and laboratory observations attributed the reddish hue to the presence of pigmentary anhydrous hematite (α-Fe2O3; termed “nanophase NpOx”) dispersed in the surface regolith and/or coating of rocks (Bell et al., 1990; Morris et al., 1989). Based on the lack of water absorption features at near infrared (NIR) wavelengths (1 – 2.5 μm)as determined by ESA’s Observatoire pour la Minéralogie, l’Eau, les Glaces et l’Activité (OMEGA) spectrometer, it was argued that the anhydrous and dusty regions contain ferric oxides, possibly hematite or maghemite (γ-Fe2O3)(Bibring et al., 2006). Furthermore, a widely used mineralogical model (Bibring et al., 2006) proposed that these anhydrous ferric oxides in Martian dust formed by continuous oxidation and weathering under water-poor surface conditions during the Amazonian period, which spans from approximately 3 billion years ago to the present.Early spacecraft observations revealed a distinctive 3 μm hydration feature in the Martian dust spectrum (Murchie et al., 1993; Pimentel et al., 1974) well before the weaker NIR spectral features associated with alteration minerals were identified (Bibring et al., 2006). Later evaluation of the OMEGA data noted that the large 3 μm absorption band is deeper in the observations of bright, dusty regions when compared to dark, less dusty terrains (Jouglet et al., 2007; Milliken et al., 2007). The increased strength of this absorption band in dusty regions was attributed to either higher abundances of water adsorbed on grain surfaces due to the large surface to volume ratio of the dust particles (e.g. Zent & Quinn, 1997) or H2O bound in hydrated minerals in the dust. Audouard et al. (2014), using ten years’ worth of OMEGA data, showed that the 3-µm band can be attributed to tightly bound H2O and/or hydroxyl groups in the mineral structure of the dust. NASA’s Compact Reconnaissance Imaging Spectrometer for Mars (CRISM) also indicated a deep absorption centered at 3 µm (Murchie et al., 2019) in bright, dusty regions. Finally, laboratory reflectance investigations of Martian meteorite ALH 84001 revealed a 3-μm hydration band, which was attributed to H2O, although no bands were observed at 1.4 or 1.9 µm (Bishop et al., 1998). Basaltic volcanic glasses also typically include a broad 3-µm band due to H2O without the weaker 1.4 or 1.9 µm (e.g. Bishop, 2019).Data collected by the MIMOSII Mössbauer instrument (MB) on the Mars Exploration Rovers (MER) showed the existence of coarse-grained hematite and goethite in specific rock outcrops as well as the ubiquitous presence of undetermined iron oxide phase (“nanophase NpOx”) in the fine dust (Morris et al., 2006). While MER MB data can be used to determine the Fe oxidation state (Fe3+/FeT) it is more difficult to distinguish the mineralogy of ferric iron present in the Martian dust (Morris & Klingelhöfer, 2008). This difficulty arises because in the microcrystal range, the distinct characteristics of different iron oxides gradually disappear as particle size and crystallinity decrease, resulting in broad and diffuse spectral lines (Coey, 1974; Murad & Schwertmann, 1980). Further, characterization of nanophase components is difficult in mixtures. However, data from MERs showed that the iron concentration in the fine dust is positively correlated with sulfur and chlorine abundances, while dark olivine-rich soils contained lower abundances of these elements, suggesting that iron in the dust is a product of chemical alteration (Ming et al., 2008; Morris et al., 2006; Yen et al., 2005). The MERs were also equipped with a series of magnet arrays designed to analyze airfall dust. The analysis of the magnetic targets using MB spectral and imaging systems identified two distinct ferric iron endmembers in the dust: one comprising strongly magnetic and dark-colored magnetite, and the other an unidentified bright-colored (oxy)hydroxide exhibiting weak magnetic properties (Goetz et al., 2005; Madsen et al., 2009). Earlier results from the Mars Pathfinder mission (Madsen et al., 1999), which utilized five magnets of varying strengths, indicated that the magnetic properties of Martian soil are likely due to small amounts of maghemite present in intimate association with silicate particles, suggesting that the dust particles are composites containing both magnetic and non-magnetic components.NASA’s Mars Science Laboratory (MSL) rover provided several key chemistry and mineralogy measurements of Martian dust and soils. The Chemistry and Camera (ChemCam) instrument utilized its laser-induced breakdown spectroscopy (LIBS) capability to analyze the composition of airfall dust. In each of the initial laser shots from a series of 50 shots on dusty rock surfaces and calibration targets that collected dust over the years, ChemCam consistently detected a hydrogen signal that exhibited no diurnal variation, suggesting that hydrogen is chemically bound within the dust particles (Lasue et al., 2018; Meslin et al., 2013). Samples from the dust covered sand dune known as “Rocknest” were measured with the Chemistry and Mineralogy (CheMin) X-ray diffraction instrument. These measurements revealed that up to scooped soil is X-ray amorphous and that ~20 wt. % of the amorphous component consists of iron oxides (Bish et al., 2013; Blake et al., 2013). In addition, the Alpha Particle X-ray Spectrometer (APXS) instrument analyzed air fall dust on the science observation tray. These measurements (Berger et al., 2016) indicated that the dust is compositionally similar to the bulk basaltic Mars crust (Gellert & Yen, 2019; McLennan & Taylor, 2008), but is enriched in SO3, Cl and Fe, which is in agreement with MER observations (Goetz et al., 2005). Both APXS and ChemCam measurements suggested that the amorphous iron oxide component observed at “Rocknest” soils may be linked to dust (Berger et al., 2016; Lasue et al., 2018). The Sample Analysis at Mars (SAM) instrument, which includes a gas chromatograph and a quadrupole mass spectrometer, detected volatile species (H2O, SO2, CO2 & O2) when the ’Rocknest’ sample was heated to ~835 °C (Leshin et al., 2013). This finding suggested that H2O is bound to the amorphous component of the sample, as the CheMin instrument did not detect any crystalline phyllosilicate minerals in this sample (Leshin et al., 2013).Here we report the spectral detection of ferrihydrite (Fe5O8H · nH2O) – a poorly crystalline X-ray amorphous and hydrated iron oxide mineral – using a combination of orbital, in-situ and laboratory visible near-infrared (VNIR) spectra. In addition, we show that ferrihydrite is stable under simulated present-day Martian conditions (UV irradiation, 6 mbar pressure, CO2 atmosphere). We then discuss its importance and implications for the past climate and habitability on Mars.
