Diamond formation by carbon saturation in C–O–H fluids during cold subduction of oceanic lithosphere

Abstract

Microdiamonds in garnet of graphite-free ultrahigh pressure metamorphic (UHPM) rocks from Lago di Cignana (western Alps, Italy) represent the first occurrence of diamond in a low-temperature subduction complex of oceanic origin (T = ∼600 °C; P ⩾ 3.2 GPa). The presence of diamonds in fluid inclusions provides evidence for carbon transport and precipitation in an oxidized H2O-rich C-O-H crustal fluid buffered by mineral equilibria at sub-arc mantle depths. The structural state of carbon in fluid-precipitated diamonds was analyzed with 514 nm excitation source confocal Raman microspectroscopy. The first order peak of sp3-bonded carbon in crystalline diamonds lies at 1331 (±2) cm−1, similar to diamonds in other UHPM terranes. The analysis of the spectra shows additional Raman features due to sp2 carbon phases indicating the presence of both hydrogenated carbon (assigned to trans-polyacetylene segments) in grain boundaries, and graphite-like amorphous carbon in the bulk, i.e. showing a structural disorder much greater than that found in graphite of other UHPM rocks. In one rock sample, disordered microdiamonds are recognized inside fluid inclusions by the presence of a weaker and broader Raman band, downshifted from 1332 to 1328 cm−1. The association of sp3- with sp2-bonded carbon indicates variable kinetics during diamond precipitation. We suggest that precipitation of disordered sp2 carbon acted as a precursor for diamond formation outside the thermodynamic stability field of crystalline graphite. Diamond formation started when the H2O-rich fluid reached the excess concentration of C required for the spontaneous nucleation of diamond. The interplay of rock buffered fO2fO2 and the prograde P-T path at high pressures controlled carbon saturation. Thermodynamic modeling confirms that the C-O-H fluids from which diamond precipitated must have been water rich (0.992<XH2O<0.9970.992<XH2O<0.997), assuming that fO2fO2 is fixed by the EMOD equilibrium.