The aim of this thesis was to better understand the origin and speciation of carbon in metasomatic fluids that give rise to diamonds. It included an analysis of diamonds from a same xenolith , polycrystalline diamonds and three suites of sulfide-bearing inclusion diamonds. This work was conducted in close collaboration with, among others , Dr. Jeff Harris ( University of Glasgow ) for the selection and characterization of samples and Dr. Marc Chaussidon and Dr Claire Rollion Bard ( CRPG -Nancy ) for implementation of the first analyses of 33S/32S and 34S/32S isotopic compo sitions on the 1270-irms at CRPG.
This work has led to several conclusions. First, the existence of reduced mantle fluids (ie methane-bearing) has been demonstrated. Finally, several opposite conclusions derived from previous studies have been clarified by the systematic analyses of C-, N, S- isotopic compositions of diamonds and their sulfide inclusion.
This work has confirmed the superficial origin of sulfur (Δ33S ≠ 0 per mille) of sulfide inclusions yet with diamonds mantle signatures ( δ13C and δ15N ~ -5 per mille). The data support a model of metasomatic formation of diamond by precipitation of mantle fluids including a pre- existing sulphide.
The main implications of this work relates to our understanding of diamond crystallisation ages. Metasomatic diamond formation encapsulating a pre-existing mineral implies
that diamond genesis ages inferred from the the study of diamond inclusions may not necessarily reflect the age of diamond precipitation.
Furthermore , this study shows significant differences between diamonds with an inclusion of sulfide compared to those with a silicate inclusion , suggesting that the “ages” of diamond formation inferred from either silicate or sulfide inclusion can not be generalized to diamonds from this locality.