Kimberlites from Source to Surface: Insights from Experiments

Abstract

High-pressure experiments are unconvincing in explaining kimberlites as direct melts of carbonated peridotite because the appropriate minerals do not coexist stably at the kimberlite liquidus. High-pressure melts of peridotite with CO2 and H2O have compositions similar to kimberlites only at pressures where conditions are insufficiently oxidizing to stabilize CO2: they do not replicate the high K2O/Na2O of kimberlites. Kimberlite melts may begin their ascent at ≈300 km depth in reduced conditions as melts rich in MgO and SiO2 and poor in Na2O. These melts interact with modified, oxidized zones at the base of cratons where they gain CO2, CaO, H2O, and K2O and lose SiO2. Decreasing CO2 solubility at low pressures facilitates the incorporation of xenocrystic olivine, resulting in kimberlites’ characteristically high MgO/CaO.

Mapping global kimberlite potential from reconstructions of mantle flow over the past billion years

“…Recent geochemical fingerprinting of kimberlites revealed that some kimberlite erupted above basal mantle structures could be derived from a primitive, depleted mantle source, whereas kimberlite erupted away from basal mantle structures could be derived from a source that contains deeply subducted crustal material [ 80 ]. Even though upwelling of hot mantle from the deep Earth may provide the source of heat leading to kimberlite eruption, the physical processes that could link basal mantle structures [>2000 km deep; 26] and kimberlite eruption from ~120–300 km depth [ 12 , 13 ] are not yet firmly established.…”

“…Kimberlite eruptions are rapid magmatic events with magma ascending at speeds up to 20 m s -1 [ 9 , 10 ], which is geologically instantaneous compared to average plate motion at ~1.26 × 10 −9 m s -1 (4 cm yr -1 ) over the last 200 Myr [ 11 ]. Kimberlites originate from depths in excess of 120–150 km [ 12 ], forming from melts that may pool from up to ~300 km depth [ 13 ]. Some rare superdeep (sub-lithospheric) kimberlites may have ascended from as deep as 800 km [ 14 – 17 ], although these diamonds could have been transported to the base of the lithosphere before being entrained by kimberlite magmas [ 18 ].…”

“…It is generally accepted that major LIPs are linked to deep mantle plumes [ 33 , 34 ], and the synchronicity of Mesozoic-Cenozoic kimberlite blooms with periods of LIP eruption suggests a potentially common heat source for kimberlite and LIP eruptions [ 25 , 35 ]. However, the physical process connecting kimberlite magmatism [from 120–300 km depth; 12 , 13 ] to LLSVPs [between 2000 km and 2500 km depth; 26 ] remains to be established. The spatial link between LLSVPs and reconstructed kimberlite eruption locations implies that hot basal mantle structures could have been stationary and rigid over time [ 25 ].…”

Geodynamic and Isotopic Constraints on the Genesis of Kimberlites, Lamproites and Related Magmas From the Finnish Segment of the Karelian Craton

“…Although there are some limitations associated with the perovskite and monticellite oxybarometers (Bellis & Canil, 2007; Canil & Bellis, 2007; Le Pioufle and Canil, 2012), the relative differences in f O 2 rather than absolute values are of importance here. Metasomatized mantle lithosphere is considered to be oxidized (e.g., Foley & Fischer, 2017; Foley et al., 2019; Woodland et al., 1996; Yaxley et al., 2017). Therefore it follows, to a first order approximation, that progressive assimilation of enriched SCLM material increases oxidation levels in kimberlite melts.…”

Sulfur Isotope Constraints on the Petrogenesis of the Kimberley Kimberlites

“…The formation of sulfates requires prevailing relatively oxidising conditions in the later stage of kimberlite melt crystallisation and/or during subsolidus hydrothermal alteration. Previous work has indeed demonstrated that kimberlite melt oxygen fugacity progressively increases during crystallisation, based on experimental (e.g., Foley et al, 2019) as well as natural evidence, such as decreasing V/Sc in magmatic olivine rims (Giuliani, 2018;Howarth & Taylor, 2016); increasing melt Mg/Fe 2+ during groundmass crystallisation (Bussweiler et al, 2015;Soltys et al, 2018a;Soltys, Giuliani, Phillips, & Kamenetsky, 2020); and increasing Fe 3+ /ΣFe in garnet in peridotite xenoliths in kimberlites occurring shortly before kimberlite emplacement (Hanger et al, 2015). The formation of native Ni and Cu locally associated with other products of sulfide alteration does not contradict this scenario because these phases form under relatively oxidising conditions (i.e., log (fO 2 ) > quartz-fayalite-magnetite buffer).…”