Reactive transport and the genesis of kimberlites

When studying the bulk rock analysis of kimberlites a significant correction for visible xenocrysts must be made [1,2]. Such studies concluded that olivine in kimberlite has a xenocrystic core and cognate margin. Previously, material entrained by the melt and dissolved during transport has only been...

Full description

Bibliographic Details
Main Authors: Pilbeam, Llewellyn, Nielsen, Troels, Waight, Tod Earle
Format: Conference Object
Language:English
Published: 2011
Subjects:
Online Access:https://curis.ku.dk/portal/da/publications/reactive-transport-and-the-genesis-of-kimberlites(0c69d2d4-04f3-47e3-85d4-dabff27e263f).html
http://www.the-conference.com/conferences/2011/vmsg2011/abstracts/finalPDFs/A18.pdf
Description
Summary:When studying the bulk rock analysis of kimberlites a significant correction for visible xenocrysts must be made [1,2]. Such studies concluded that olivine in kimberlite has a xenocrystic core and cognate margin. Previously, material entrained by the melt and dissolved during transport has only been qualitatively discussed [3]. We find bulk rock and electron microprobe data strongly support dissolution of orthopyroxene coupled with crystallization of cognate olivine rims in the Majuagaa kimberlite sensu stricto dyke. This reaction was suggested by the topography of phase diagrams for systems containing Enstatite, Forsterite and CO2. The ‘parental’ melt for Majuagaa kimberlite was magnesiocarbonatite which entrained 50wt% harzburgite, consistent with melting experiments. The melt processed ~20wt% SiO2 but at any given time had no more than ~5wt% SiO2. We explain geochemical variations across the southern West Greenland kimberlites sensu lato using similar processes. We may also account for the intercraton variety in kimberlite composition and mineralogy. Mg# and Ni composition was obtained for transects across olivine grains. Pure Rayleigh fractionation did not explain the shape of the nickel transects. Diffusion between homogeneous olivine grains and melt was also eliminated. The data was replicated using a combination of growth and minor later diffusion. Growth was modeled using AFC equations [4]. Later diffusive equilibration between core and margin was minor. Since kimberlite transport time was short magmatic temperatures of ~900oC were supported. ! [1] Nielsen & Sand (2008), Canadian Mineralogist 46, 1043- 1061 [2] Ardnt, Guitreau, Boullier, Le Roex, Tommasi, Cordier & Sobolev (2010), Journal of Petrology 51, 573-602 [3] Mitchell (2008), Journal of Volcanology and Geothermal Research 174, 1-8 [4] De Paulo (1981), Earth and Planetary Science Letters 53, 189-202