2014 GSA Annual Meeting in Vancouver, British Columbia (19–22 October 2014)

Paper No. 104-1
Presentation Time: 8:15 AM

SIMS APPLICATIONS TO DIAMOND RESEARCH


STACHEL, Thomas1, STERN, Richard A.2, PETTS, Duane1, NICHOLS, Katie1 and CHACKO, Tom3, (1)Earth and Atmospheric Sciences, University of Alberta, 1-24 ESB, Edmonton, AB T6G 2E3, Canada, (2)Canadian Centre for Isotopic Microanalysis, University of Alberta, 1-26 Earth Sciences Building, University of Alberta, Edmonton, AB T6G 2E3, Canada, (3)Earth and Atmospheric Sciences, University of Alberta, Edmonton, AB T6G 2E3, Canada

Over the past 40 years tremendous progress has been made in constraining the environment (source paragenesis) and the physical conditions of diamond formation. Significant gaps, however, still remain with respect to the processes that lead to diamond precipitation. It is generally assumed that diamond forms during reduction of carbonate or oxidation of methane contained in mantle fluids or melts. But there is no consensus as to which of these fluids/melts is more prevalent and what typical fluid-rock ratios would be.

High resolution multi-collector ion microprobes now allow rapid and high-precision in-situ analysis of carbon (δ13C), nitrogen (δ15N) isotopic compositions and nitrogen concentration ([N]) across diamond growth zones previously imaged with cathodoluminescence. These data allow modelling of the composition and evolution of diamond-forming fluids and provide novel insights into the complexity and protracted history of diamond growth.

Using experimentally and theoretically constrained carbon isotope fraction factors during diamond growth and observed co-variations among δ13C - δ15N -[N] in individual growth layers , we have derived both nitrogen partition coefficients and nitrogen isotope fractionation factors between diamond and fluid. We find that nitrogen is strongly compatible in diamond and that, relative to the growth medium, diamond is highly depleted in 15N (difference of ~ -5 ‰ in δ15N). Applying this information to large data sets (in situ analyses of δ13C - δ15N -[N]) derived from individual diamond occurrences reveals that δ13C actually is a very poor tracer of diamond forming processes. Compared to [N] and δ15N, δ13C only shows significant evolution (beyond the analytical uncertainty of ~0.2‰) in very dilute (carbon poor) diamond-forming fluids or large degrees of diamond crystallization. Thus, an apparent decoupled behaviour of carbon (little variation) and nitrogen isotopes (strong variation) in diamond is largely a consequence of diamond precipitation from a growth medium with high carbon concentration and/or low nitrogen content. So far our studies have focussed on diamonds of eclogitic source paragenesis and for these data sets the co-variations among δ13C - δ15N -[N] suggest a dominance of oxidized (carbonate-bearing) diamond-forming fluids.