Oxygen Fugacity of the Martian Mantle from Pyroxene/melt Partitoning of Ree

Introduction: This study is part of an ongoing effort [1-3] to calibrate the pyroxene/melt REE oxybarometer for conditions relevant to the martian meteorites. Redox variations have been reported among the shergottites [4-6]. Wadhwa [4] used the Eu and Gd augite/melt partitioning experiments of [7], designed for the LEW86010 angrite, to infer a range of fo2 for the shergottites. [5,6] inferred fo2 using equilibria between Fe-Ti oxides. There is fairly good agreement between the Fe-Ti oxide determinations [5,6] and the estimates from Eu anomalies [4] in terms of which meteorites are more or less oxidized. The Eu anomaly technique and the Fe-Ti oxide technique both essentially show the same trend, with Shergotty and Zagami being the most oxidized and QUE94201 and DaG 476 being the most reduced. Thus, the variation in fo2 appears to be both real and substantive. However, although the redox trends indicated by the two techniques are similar, there is as much as a two log unit offset between the results of [5,6] and [4]. One explanation for this offset is that the Eu calibration used for the shergottites was actually designed for the LEW86010 angrite, a silica-undersaturated basalt whose pyroxene (diopside) compositions are rather extreme. To correct this, [1-3] have conducted experiments on the redox relationship of Eu partitioning relative to Sm and Gd for pyroxene/melt compositions more relevant to Martian meteorites. We report here preliminary results for experiments on pigeonite/melt partitioning as a function of fO2. Experimental: Pigeonite/melt partitioning experiments were run in a gas-mixing DelTec furnace at fO2 ranging from IW-2 to IW+4 (QFM+0.5). The simplified (i.e. CMAS + Fe, Na) starting composition was designed to produce a martian composition on the liquidus. This starting composition (Table1) was derived by mixing 65% glass from melting experiments on QUE 94201 [8] that is in equilibrium with the first pigeonite to crystallize, 30% of the experimental pigeonite (PX4), and 5% silica (to suppress olivine crystallization). This composition was doped with 1% each of Sm, Eu and Gd. Samples were suspended from a Re wire loop in the furnace. The samples were first raised to super liquidus temperature (1300°C) and held there for 30 minutes then ramped down at 1°/min. to run temperature (1240°C) and held for up to 24 hrs and then quenched in water. Run products were analysed for all elements with the SX100 EMP at NASA/JSC. Table 1: Starting composition (Sh3)