Links between carbonatite and kimberlite melts in chloride-carbonate-silicate systems: experiments and application to natural assemblages

Recent data suggest an active role for chloride-bearing alkali carbonatitic melts in the formation and evolution of kimberlites, whereas experiments indicate that chlorides could be responsible for liquid immiscibility during melting of carbonated mantle rocks. This study considers melting trends in...

Full description

Bibliographic Details
Published in:Journal of Petrology
Main Authors: Safonov, OG, Kamenetsky, VS, Perchuk, LL
Format: Article in Journal/Newspaper
Language:English
Published: Oxford Univ Press 2011
Subjects:
Online Access:https://doi.org/10.1093/petrology/egq034
http://ecite.utas.edu.au/74070
Description
Summary:Recent data suggest an active role for chloride-bearing alkali carbonatitic melts in the formation and evolution of kimberlites, whereas experiments indicate that chlorides could be responsible for liquid immiscibility during melting of carbonated mantle rocks. This study considers melting trends in kimberlite-related chloride- carbonate-silicate systems at a pressure of 5.5 GPa.The direction of these trends largely depends on the chlorine concentration in the system. All trends start with a Cl-rich carbonatitic liquid coexisting with crystalline phases. Melting of a peridotite-carbonate-chloride system containing 4.4 wt % Cl results in a gradual transition from a carbonatitic melt (5-7 wt % SiO2, MgO/CaO= 0.5-0.6, ~2 wt%Cl) at 1000-1100 oC through a Cl-rich carbonate-silicate melt (12-15 wt % SiO2, MgO/CaO =0.6, and up to 14 wt % Cl) at 1360-1400 oC towards a Cl-bearing ultrabasic carbonate- silicate melt (UCSM), i.e. kimberlite-like (26-29 wt % SiO2, MgO/CaO=1.5-2.8, and 6-4 wt % Cl), at 1500-1600 oC.This trend results from the specific behavior of chlorides, which are found to be stable crystalline phases up to about 300 oC above the solidus. In addition, the trend touches a miscibility gap at about 1450 oC, although immiscibility does not significantly influence the melt evolution. In contrast, the melting trend of a peridotite system with about 1.3 wt%Cl does not intersect the miscibility gap and proceeds from a carbonatitic melt (~5 wt % SiO2, MgO/CaO= 0.4-0.5, ~2wt % Cl) at 1100-1150 oC toward the UCSM (~25 wt % SiO2, MgO/CaO= 1.7-1.8, and 0.6-10 wt % Cl) at 1500- 1600 oC. Chloride-carbonate-silicate systems containing 17 wt % Cl show an abrupt transition from the chloride-carbonate liquid toward the UCSM because of the immiscibility gap between carbonate- silicate and chloride-carbonate melts. Melting relations in all studied chloride-carbonate-silicate systems are exclusively regulated by peritectic reactions between the silicate phases and carbonate constituents (mostly CaCO3) of the melts. Chlorides decrease the solidus temperatures of carbonated peridotites, but do not reduce the temperature interval for the transition from carbonate- to silicate-dominated melts. The experimental results suggest that Cl-rich carbonatite liquids preserved in diamonds and the Cl-rich kimberlites of the Udachnaya-East kimberlite pipe (Yakutia) could represent a linked system of chloride-rich liquids that evolved over a wide temperature interval during incipient melting and evolution of kimberlite magma from a carbonated mantle source containing 3-4 wt % Cl.