Petrological and geochemical studies of an abyssal periodotite from the Atlantis II fracture zone

Abstract

This thesis investigates the petrology and geochemistry of an abyssal peridotite dredged from the Atlantis II Fracture Zone in the southwestern Indian Ocean Ridge. Texturally, this sample is a serpentinized peridotite with a crosscutting coarse-grained clinopyroxenite vein. One of the alteration veinlets contains rutile and ilmenite in association with plagioclase and amphibole. This veinlet is not related to the pyroxenite vein. In tenns of mineralogy, the composition of the major silicate minerals indicates that this plagioclase llierzolite represents the depleted residue after mantle melting, similar to other abyssal peridotites from this region. In addition to the presence of the unique pyroxenite vein, this sample was earlier shown to be a carrier of 'orphan Sr-87'. Unfortunately, I was unable to find such high Sr isotopic ratios in the magnetic separations of different fractions of the sample. The sulfide mineralogy, together with the whole rock chemistry, suggests that sea water alteration occurs mainly as a result of serpentinization at temperatures higher than 200°C. Since the sample is less than l Ma old, and the low temperature weathering occurred only after the sample was exposed at the sea floor, it is possible that the weathering process was restricted to major alteration veins. This suggests that the alteration process is highly fracture controlled and time dependent Trace element data from clinopyroxene grains in the peridotite shows large variations from grain to grain. The (Ce/Yb )n ranges from 0.17 to 0.54 in the pyroxenite vein, and from 0.75 to 2.35 away from the vein. The tendency for LREE enrichment with the increase of distance from the vein suggests the presence of highly reacted melts. An assimilation-fractional crystallization (AFC) model was derived which supports the idea that the source of the clinopyroxenite vein reacts with the depleted peridotite to form a central reaction zone. Some of the highly reacted melt, after melt-rock reaction, migrates out of the reaction zone, and precipitates some late magmatic phases while being trapped in the country rocks. Since the sulfide is a major Os reservoir in the abyssal peridotites, as shown in leaching experiments, and the melt is saturated in sulfur as a consequence of the reaction process, it is possible to model the heterogeneous distribution of the Os isotopic data by mixing the residual peridotite with 0.2 to 0.5 wt% of sulfides precipitated from the melt. This mixing process can explain most of the heterogeneity from 1.034 to 1.148 for 187Os/186Os. The impact on the peridotite from the melt-rock reaction and impregnation of the late melt is obvious. As evident in Hess Deep gabbroic rocks, conductive heat loss in the transform fault fulfils the physical requirement to create and to preserve such geochemical signature.