Igneous cooling history of olivine‐phyric shergottite Yamato 980459 constrained by dynamic crystallization experiments

Abstract

Abstract-Dynamic crystallization experiments were performed on a liquid having the bulk composition of olivine-phyric shergottite Yamato 980459, to constrain the igneous thermal history of this meteorite. Key characteristics of the meteorite's mineralogy and texture, including several morphologically distinct olivine and pyroxene crystal populations and a glassy mesostasis devoid of plagioclase, were replicated upon cooling from 1435 to 909°C at 1 atmosphere under reducing conditions. Three sequential cooling ramps are required to produce synthetic samples with textures and compositions matching Yamato 980459. Olivine phenocrysts formed at <1°C h À1 , presumably at depth in the Martian crust. Pyroxene phenocrysts formed mainly at~10°C h À1 , consistent with crystallization within a lava flow at depths of 25-45 cm. Increased cooling rate (~100°C h À1 ) in a third stage suppressed the formation of plagioclase and produced groundmass crystals, consistent with crystallization at lava flow depths of 5-7 cm. Although Y 980459 is unique among Martian meteorites (i.e., preserving a primary glassy mesostasis), its emplacement did not require unique physical conditions. Rather, the second and third cooling stages may reflect cooling within the core of a p ahoehoe-like flow and subsequent breakout on the surface of Mars.

Controlled Cooling‐Rate Experiments on Olivine‐Hosted Melt Inclusions: Chemical Diffusion and Quantification of Eruptive Cooling Rates on Hawaii and Mars

“…Cooling rates from zoned melt inclusions in Yamato 980459 range from 85 to 1,047 °C/hr with a mean and 1σ of the average cooling rate calculated for each inclusion of 383±43 °C/hr ( n = 8). These rates overlap with estimates for cooling associated with stage 3 of the inferred thermal history of Y980450 (First & Hammer, ; Greshake et al, ) and support the hypothesis that the sample erupted at or near the Martian surface and was rapidly cooled through the glass transition…”

“…Cooling rates calculated from MgO diffusion in melt inclusions correspond to stage 3 of the proposed cooling history of Y980459, which produced the vitrophyric texture in the rock. The distribution of rates recovered from the eight inclusions (ranging from 85 to 1,047 °C/hr) overlaps with the 150 °C/hr estimate based on crystal size distributions from the cooling experiments of First and Hammer () and is within a factor of 2 of the estimates of 1,450 to 1,890 °C/hr from Greshake et al (); Figure c). The very low H 2 O contents measured in two melt inclusions from Y980459 (146–251 ppm and 458–841 ppm; Usui et al, ) indicate that the range in cooling rates is likely not due to the effect of variable H 2 O on diffusion in the inclusion liquids.…”

“…Thin sections of Y980459 have relatively large (up to ~100 μm longest dimension) glassy magmatic inclusions contained in olivine (Greshake et al, ; Ikeda, ; Ikeda, ). A multistage magmatic thermal history of Y980459 has been proposed to explain (1) large olivine phenocrysts formed during slow cooling from a liquidus temperature of 1,440 °C at rates <1 °C/hr (First & Hammer, ); (2) the crystal size distributions of pyroxene phenocrysts, interpreted as requiring cooling rates <10 °C/hr (First & Hammer, ; Lentz & McSween, ); and (3) the vitrophyric groundmass texture, interpreted as evidence of a late stage of rapid cooling with rate estimates ranging from ~150 °C/hr (First & Hammer, ) up to 1,450–1,890 °C/hr (Greshake et al, ; Figure ). Although the melt inclusions were likely trapped during the growth of phenocrysts early in the cooling history of the magma, the proposed cooling rates from stages 1 and 2 are too low to preserve measurable diffusion profiles in melt inclusion based on our results.…”

“…For example, a forward model of an inclusion cooling from 1,150 to 1,050 °C predicts that the MgO gradient will be less than the 2σ uncertainty in the microprobe measurements (<0.15 wt %) for inclusions <50 μm in radius. The initial slow cooling stages were likely at temperatures of 1,440 °C down to ~1,115 °C (First & Hammer, ) where diffusion would have been even more efficient at homogenization of inclusions than in the example given in the previous sentence. Accordingly, the inverse model developed in section is only applicable to cooling stage 3, the final cooling interval, which produced the vitrophyric matrix.…”

Chemical layering in the upper mantle of Mars: Evidence from olivine‐hosted melt inclusions in Tissint

“…The late‐stage crystallization of Tissint PM is more readily similar to the Y‐98059 as both cooled very fast (>5°C h −1 ; Basu Sarbadhikari et al. ; First and Hammer ) than the early to intermediate stage of crystallization and had severely depleted plagioclase component. The difference between presence of plagioclase (17–20 vol%; Basu Sarbadhikari et al.…”