Jarosite and Hematite at Meridiani Planum from Opportunity's Mössbauer Spectrometer

G. Klingelhöfer, R. V. Morris, B. Bernhardt, C. Schröder, D. S. Rodionov, P. A. de Souza, A. Yen, R. Gellert, E. N. Evlanov, B. Zubkov, J. Foh, U. Bonnes (+7 others)
2004 Science  
Mössbauer spectra measured by the Opportunity rover revealed four mineralogical components in Meridiani Planum at Eagle crater: jarosite-and hematite-rich outcrop, hematite-rich soil, olivine-bearing basaltic soil, and a pyroxene-bearing basaltic rock (Bounce rock). Spherules, interpreted to be concretions, are hematiterich and dispersed throughout the outcrop. Hematitic soils both within and outside Eagle crater are dominated by spherules and their fragments. Olivine-bearing basaltic soil is
more » ... basaltic soil is present throughout the region. Bounce rock is probably an impact erratic. Because jarosite is a hydroxide sulfate mineral, its presence at Meridiani Planum is mineralogical evidence for aqueous processes on Mars, probably under acid-sulfate conditions. MPssbauer spectrometers provide quantitative information about the distribution of iron among its oxidation states, the identification of iron-bearing phases, and the distribution of iron among those phases. Characterizing the speciation and distribution of iron in martian rock and soil (1) constrains primary rock types; the conditions under which primary minerals crystallize; and the mineralogical composition, process, and extent of alteration and weathering. Hematite (a-Fe 2 O 3 ) was detected in Meridiani Planum from orbital observations before the Mars Exploration Rover (MER) missions (2, 3), and the region was selected as a MER landing site on a scientific basis, because the mineral is a beacon for aqueous processes, and on an engineering basis, because the terrain was considered safe for landing (4). The first MPssbauer spectrum from Meridiani Planum was obtained by the Opportunity rover on 4 February 2004, on soil near the lander at Eagle crater. The MIMOS II (Miniaturized MPssbauer spectrometer) instrument (5) on the Opportunity rover is nearly equivalent to the one on the Spirit rover and is operated in the same manner (6 -8). It is mounted on the robotic arm, which positions the instrument onto surface targets. Physical contact is confirmed by a contact plate. The instrument has backscatter geometry, a drive frequency of È24 Hz, a triangular waveform divided into 512 data channels, and a selectable maximum drive velocity range up to about T20 mm/s (7). Because of the diurnal temperature variation on Mars, the instrument settings were selected to store spectra in 13 separate memory areas that correspond to 11 temperature intervals 10-K wide between 180 and 290 K, plus G180 K and 9290 K windows. The target area and sampling depth of the illuminating 14.4-keV 57 Co gamma radiation (È150 mCi at landing) are È1.5 cm in diameter and È0.2 mm (coherent rock) to È3 mm (air-fall dust) deep, respectively. The MPssbauer contact plate assembly that comes into physical contact with the surface during data acquisition contains a sensor for temperature measurement. We report results for the resonantly scattered 14.4-keV g rays. We followed the data analysis procedures outlined by (6) for MPssbauer spectra from Gusev crater. Briefly, we optimized counting statistics by summing spectra for all temperature intervals for the same target and by summing the same temperature window for different targets whose spectra appear equivalent. Peak parameters (center, width, and area) were calculated by least squares fitting procedures with Lorentzian line shapes. The MPssbauer parameters d (isomer shift relative to metallic iron), DE Q (quadrupole splitting), and B hf (strength of the internal magnetic hyperfine field) were calculated from peak centers. To correct peak areas for differences in recoil-free fractions (the f factor), we used the ratio f(Fe 3þ )/f(Fe 2þ ) 0 1.21 independent of mineralogical composition (9, 10). The centers and widths of low-intensity subspectra (both doublets and sextets) were constrained to the values measured in spectra where their intensities were high. Because we did not observe obvious textural effects, doublet peak areas and widths were always constrained to be equal. For sufficiently intense sextets, two different approaches were used to fit the data: (i) Peak areas were constrained in the proportion 3:x:1:1:x:3, and the widths were unconstrained except for the inner two peaks (which strongly overlap with the doublet peaks), whose widths were constrained to be equal. (ii) Peak areas were constrained in the proportion 3:x:1:1:x:3, and the widths were unconstrained but equal for all six lines.
doi:10.1126/science.1104653 pmid:15576610 fatcat:xtqhrzd3hfehfnhfe3xkyx4jcq