Formation of diamondiferous kyanite–eclogite in a subduction mélange

Diamond- and kyanite-bearing eclogites from the Lace kimberlite on the Kaapvaal craton have common picritic to gabbroic oceanic protoliths with bimineralic eclogites, lying on arrays of Eu∗ and ∑REE that are consistent with accumulation and fractionation of plagioclase and olivine. However, they als...

Full description

Saved in:
Bibliographic Details
Published in:Geochimica et cosmochimica acta Vol. 179; pp. 156 - 176
Main Authors: Aulbach, S., Gerdes, A., Viljoen, K.S.
Format: Journal Article
Language:English
Published: Elsevier Ltd 15-04-2016
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Diamond- and kyanite-bearing eclogites from the Lace kimberlite on the Kaapvaal craton have common picritic to gabbroic oceanic protoliths with bimineralic eclogites, lying on arrays of Eu∗ and ∑REE that are consistent with accumulation and fractionation of plagioclase and olivine. However, they also show significant compositional differences, such as more grossular-rich garnet and aluminous clinopyroxene (cpx), which require the operation of additional processes. Their nature is elucidated using mineral major- and trace-element compositions, as well as Sr isotope ratios determined by in situ techniques. Highly variable major-element compositions across the co-genetic eclogite suites exert a strong effect on the trace-element distribution between garnet and cpx, whereby Sc, Ge, Sr, Y, Cd, REE, Th and U partition more strongly into garnet with increasing grossular-content. Thus, significant differences between the trace-element compositions of garnet can ensue from crystal-chemical effects alone, making their use as petrogenetic indicators potentially ambiguous. After correcting for these compositional effects, garnet in kyanite-/diamond eclogites, and in eclogites devoid of accessory minerals but with similar signatures, shows depletion (or dilution) in Sc, Ge, Y, In, Zr, Hf and the HREE, and enrichment in the LREE and Th compared to garnet in bimineralic eclogites. This is interpreted as the signature of a pelite-derived melt, which was transferred by addition of aluminous cpx that later exsolved kyanite and garnet, as observed in other aluminous eclogite suites. Continental input can explain initial (at 2.9Ga) 87Sr/86Sr⩾0.714 measured in cpx in eleven samples with low 87Rb/86Sr (<0.01). The association of diamond with kyanite suggests that diamond formation is also linked to this event, possibly due to diamond formation by oxidation of reduced carbon, such as methane, and attendant reduction of Fe3+ in garnet. This model of sediment melt–oceanic crust interaction reconciles evidence for both low- and high-pressure igneous processes in some aluminous eclogites. We suggest that a subduction mélange is a favourable setting for the transfer of a sediment-derived signature into oceanic crust, leading to formation of diamondiferous kyanite–eclogites from bimineralic eclogites. Diapirism, fluxed by the presence of partial melt, may have facilitated dispersal of the eclogites in the lithosphere column, consistent with their widely varying equilibration pressures ranging from ∼5 to 8GPa.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0016-7037
1872-9533
DOI:10.1016/j.gca.2016.01.038