Bellsbank Kimberlite Research Articles

Constraints on Archean crust recycling and the origin of mantle redox variability from the δ44/40Ca – δ18O – fO2 signatures of cratonic eclogites

The nature of the deep calcium geochemical cycle through time is unresolved, in part due to the dearth of information about the calcium isotope composition of Archean recycled oceanic crust. Remnants of such ancient oceanic crust are preserved in the form of cratonic mantle eclogites, brought to surface as xenoliths in kimberlite magma eruptions. The δ44/40Ca of fresh mantle-derived eclogite xenoliths (i.e., garnet and omphacite mineral separates) from the Bellsbank kimberlite on the Kaapvaal craton in South Africa are presented here in combination with their trace element compositions, garnet Fe3+ contents and δ18O values.The studied Bellsbank eclogite xenoliths have geochemical compositions that indicate oceanic crustal protoliths, with bulk Al2O3 from 15 to 27 wt.%, Eu anomalies from 0.8 to 2.6 and, significantly, garnet δ18O values from +2.7 to +6.2‰. Garnet Fe3+/ΣFe contents yield logfO2(ΔFMQ) values between −4.0 and −1.2 for a depth range of 110–180 km, recording strong redox heterogeneity of the eclogite component within the Archean Kaapvaal mantle lithosphere. Reconstructed bulk eclogite MgO contents correlate negatively with fO2, suggesting that the redox compositions are related to magmatic differentiation during oceanic crust formation, excluding secondary metasomatic overprints. These data may thus emphasize that Archean basaltic oceanic crust had a similarly variable redox composition to modern MORB-type crust.Reconstructed bulk δ44/40Ca values for the Bellsbank eclogites range from +0.28 to +1.56‰. Although some of the xenoliths have δ44/40Ca values that overlap with the average mantle composition and modern MORB (+0.94 ± 0.1 and +0.83 ± 0.05‰), half of our dataset shows excursions to more extreme Ca isotopic compositions. Both higher and lower δ44/40Ca relative to mantle compositions are recorded by the eclogites, with a general negative correlation with δ18O suggestive of seawater-alteration of oceanic crust. The combined low δ44/40Ca (+0.28‰) and δ18O (+3.4‰) measured for one eclogite xenolith may record a subtle imprint by carbonate-rich mantle melts, which are known to contain isotopically light calcium contributed by recycled sediments. In contrast, the high δ44/40Ca of up to +1.56‰ for some eclogite xenoliths, coupled with strong LREE depletion, can be explained by calcium isotope fractionation during partial melting.The protracted history of recycled oceanic crust as probed by cratonic mantle eclogites is recorded by their highly variable δ44/40Ca–δ18O–fO2 signatures. Whereas some of this heterogeneity can be linked to processes that operated on the Archean ocean floor such as seawater-alteration of basaltic crust, other sources of compositional variability are introduced by loss and addition of melts during subduction recycling and mantle residence. The observed δ44/40Ca complexity of ancient recycled oceanic crust components at the scale of a single mantle-derived eclogite xenolith suite implies that mantle plume sourced intraplate magmas should reveal similarly strong calcium isotope variations contributed by apparently essential recycled crust components – as observed in the global oceanic island basalt record.

Kyanite/corundum eclogites from the Kaapvaal Craton: subducted troctolites and layered gabbros from the Mid- to Early Archean

An oceanic crustal origin is the commonly accepted paradigm for mantle-derived eclogites. However, the significance of the aluminous members of the eclogite suite, containing kyanite and corundum, has long been underrated and their role neglected in genetic models of cratonic evolution. Here, we present a geochemical and petrological study of a suite of kyanite- and corundum-bearing eclogites from the Bellsbank kimberlite, S. Africa, which originate from depths between 150 and 200 km. Although clearly of high-pressure provenance, these rocks had a low-pressure cumulative origin with plagioclase and olivine as major cumulate phases. This is shown by the very pronounced positive Eu anomalies, low REE abundances, and δ 18O values lower than the Earth’s mantle. Many chemical features are identical to modern-day troctolitic cumulates including a light REE depletion akin to MORB, but there are also distinguishing features in that the eclogites are richer in Na, Fe, and Ni. Two of the eclogites have a minimum age of ~3.2 Ga, defined by the extremely unradiogenic 87Sr/86Sr (0.7007) in clinopyroxene. Phase equilibria indicate that the parent melts were formed by partial melting below an Archean volcanic center that generated (alkali-)picritic to high-alumina tholeiitic melts from a mantle whose oxygen fugacity was lower than today. Fractional crystallization produced troctolites with immiscible sulfide melt droplets within the mafic crust. Instability of the mafic crust led to deep subduction and re-equilibration at 4–6 GPa. Phase relationships plus the presence of a sample with appreciable modal corundum but no Eu anomaly suggest that kyanite- and corundum-bearing eclogites may also originate as plagioclase-free, higher pressure cumulates of highly aluminous clinopyroxene, spinel, and olivine. This is consistent with the crystallizing phase assemblage from an olivine tholeiitic to picritic magma deeper in the Archean oceanic crust or uppermost mantle. We postulate that the magmatic and subduction processes driving modern plate tectonics already existed in the Meso- to Early Archean.

Magnesium isotopic variations in cratonic eclogites: Origins and implications

Cratonic eclogites play an important role in the formation and dynamic evolution of the subcontinental lithospheric mantle. However, their origin, whether as fragments of subducted oceanic crust or high-pressure mantle cumulates, remains controversial. Here, we report Mg isotopic compositions (δ26Mg) for cratonic eclogites from Kaalvallei and Bellsbank kimberlite pipes, South Africa. We find that clinopyroxene is 0.375±0.069 to 0.676±0.075‰ heavier than coexisting garnet, which reflects equilibrium isotope fractionation between these phases, primarily driven by the difference in Mg coordination between clinopyroxene and garnet. Bulk eclogites have strikingly variable Mg isotopic compositions, which range from −0.797±0.075 to −0.139±0.061‰, values that are significantly lighter than the range displayed by global mantle peridotites to date (−0.25±0.07‰, 2SD). As significant Mg isotope fractionation is only known to occur during low-temperature water–rock interaction, our results provide further evidence for the derivation of cratonic eclogites from subducted altered oceanic crust. In addition, the lack of correlation between Δ26Mg and Δ57Fe provides evidence for redox control on equilibrium inter-mineral Fe isotope fractionation.

Thallium isotopes as a potential tracer for the origin of cratonic eclogites

Cratonic eclogites are inferred to originate either from subducted ocean crust or mantle melts accreted onto the roots of continents. These models have different implications for the growth of continents, but it is currently difficult to determine the origin of individual eclogite suites. Upper ocean crust altered at low temperatures and marine sediments both display high thallium (Tl) concentrations and strongly fractionated Tl isotope signatures relative to the ambient upper mantle. In this study we carry out the first examination of the suitability of Tl isotopes as a tracer for an ocean-crust origin of cratonic eclogites. We have analysed the Tl isotope composition of clinopyroxene and garnet in six eclogites from the Kaalvallei and Bellsbank kimberlite pipes in South Africa. Minerals were pre-cleaned with an HCl leaching technique and the leachates display variably light Tl isotope ratios. These most likely reflect low-temperature hydrothermal alteration occurring after eruption of the kimberlite that carried the eclogites to the surface. The leached mineral pairs all display identical Tl isotope ratios, strongly suggesting that the source of the analysed Tl is identical for each mineral pair. It is, however, not possible to exclude the possibility that the analysed Tl originates from kimberlitic material that was not removed by the cleaning procedure. Only one of the six samples exhibits a Tl isotope composition different from ambient mantle. Assuming that the Tl isotope signatures indeed represent the eclogite minerals and not any form of contamination, the Tl isotope composition in this sample is consistent with containing a minor component (<3%) of ocean crust altered at low temperatures. Thallium isotopes may become one of the most sensitive indicators for the presence of low-T altered ocean crust because of the stark contrast in Tl concentration and isotopic composition between the mantle and altered ocean crust. In fact, no other chemical or isotopic tracer could have provided an indication that any of the samples studied here had a subduction origin. However, much work is still required before it becomes clear if Tl isotope measurements are a viable means to establish the origin of cratonic eclogites.

Graphite- and diamond-bearing eclogite xenoliths from the Bellsbank kimberlites, Northern Cape, South Africa

One diamond-bearing and eight graphite-bearing eclogite xenoliths are described from the Bellsbank kimberlites, Cape Province, South Africa. Graphite mostly occurs as discrete grains which are commonly in the form of tabular prisms. Diamond is octahedral. Both Group I and Group II eclogite varieties are represented by the graphite-bearing specimens, while the single diamond-bearing eclogite is of the Group I variety. The carbon isotopic composition of the graphite varies from δ13C=−7‰ to δ13C=−2.8‰. This is within the range of carbon isotopic compositions for inclusion-free diamonds in kimberlite from this locality, suggesting that the carbon for the eclogites as well as some of the kimberlite diamonds are derived from the same source. The present day Nd isotopic compositions of clinopyroxene from three graphite-bearing xenoliths are slightly higher than the bulk earth estimate. Sr isotopic compositions of the clinopyroxene in these xenoliths vary from 87Sr/86Sr=0.703 to 87Sr/86Sr=0.706. This could be due to derivation of the xenoliths from a protolith with variable 87Sr/86Sr isotopic composition or could be the result of mixing between a low-Sr, high 87Sr/86Sr component and a high Sr, low 87Sr/86Sr component.

Eclogites with oceanic crustal and mantle signatures from the Bellsbank kimberlite, South Africa, part 2: Sr, Nd, and O isotope geochemistry

The Sr, Nd, and O isotopic compositions of garnet and clinopyroxene mineral separates from nine eclogite xenoliths from the Bellsbank kimberlite (erupted at 120 Ma) define three groups. Group A eclogites, considered to be mantle cumulates, are characterized by δ 18O , and 87Sr 86Sr values typical of mantle-derived materials (+5.1 to +5.6‰ and 0.7042−0.7046, respectively), and very low Sm/Nd ratios, (apparent) Rb/Sr ratios and ε Nd[120] values (0.057–0.078, 0.00005–0.00136 and −14 to −16 respectively). The REE and isotopic data for these eclogites can be modelled in terms of crystallization from a Group II kimberlite magma at ∼ 1–1.5 Ga. Group B and C eclogites, believed to be the metamorphosed products of ancient subducted oceanic crust, are characterized by low δ 18O (+2.9 to +4.7), extremely high ε Nd[120] (∼ +40 to +219), and radiogenic 87Sr/ 86Sr ratios (0.708–0.710). The Sm/Nd ratios of the Group B eclogites are very high (up to 1.6). The data for Group B and C eclogites define a linear correlation on Sm/Nd and 1/Nd vs. ε Nd[120] diagrams. These relationships are consistent with mixing of the Bellsbank kimberlite ( ε Nd[120] = -10; Sm/Nd = 0.10) with a depleted eclogite end-member ( ε Nd[120] +219; Sm/Nd = 1.6) during a cryptic metasomatic event. The Sr isotopic variations in Group B and C eclogites cannot be generated by simple two-component mixing. The Sr, Nd, and O isotope data for Group B and C eclogites probably reflect a complex sequence of depletion and enrichment events, in both crust and mantle settings. Enrichments which possibly affected the Group B and C eclogites include seawater-alteration of a MORB-like protolith, which lowered the δ 18O and raised the 87Sr 86Sr ratio, but left the Nd isotopic compositions unchanged, and cryptic metasomatism by the magmatism that produced the Bellsbank kimberlite. The high Sm/Nd ratio of the depleted eclogite end-member cannot be generated by extraction of a melt from a modern MORB composition. Rather, it is argued that such high Sm/Nd ratios are produced as a result of partitioning during the recrystallization of a MORB-like component to eclogite during subduction. Nd model ages suggest that this process occurred ∼ 2.3–2.4 Ga.

Eclogites with Oceanic Crustal and Mantle Signatures from the Bellsbank Kimberlite, South Africa, Part I: Mineralogy, Petrography, and Whole Rock Chemistry

Three groups of eclogite xenoliths have been identified from the DeBruyn and Martin Mine of the Bellsbank kimberlite, South Africa. These eclogites are divided into the three groups on the basis of petrography, clinopyroxene and garnet mineral chemistry, and rare earth element (REE) contents of the whole rocks, clinopyroxenes, and garnets. Abundances of the REE and garnet-clinopyroxene Kd's are consistent with a petrogenesis by fractional crystallization for Group A eclogites, but not Groups and We suggest, on the basis of bulk-rock major-element and REE analyses and reconstructed REE patterns, that eclogites from Groups and are the metamorphosed products of ancient subducted oceanic crust. Group eclogites have major element chemistry similar to Archean basalts, and Group eclogites have high abundances. Metasomatism has affected all groups of eclogites, producing phlogopite, feldspar, amphibole, and a breakdown of primary clinopyroxene. On the basis of large ion lithophile (LIL) elements, this metasomatism can be traced to the host kimberlite. Our study demonstrates that petrogenesis by fractional crystallization cannot account for all eclogite xenolith compositions found in kimberlite.

REE abundances in exsolved garnet-bearing clinopyroxene megacryst from Bellsbank kimberlite (South Africa)

REE abundances of subcalcic clinopyroxene megacryst and its exsolved phases have been determined by mass spectrometry. The megacryst is characterized by rare-earth element (REE) concentrations higher than chondrite by a factor of 1.5 to 2.4 and a rather flat chondrite-normalized pattern. REE data indicate that the clinopyroxene megacryst is in equilibrium with an alkali picrite magma, but not with a kimberlite. Partitioning of major elements among coexisting diopside, enstatite and garnet, which have exsolved from the megacryst, is the same as that of the granular-type garnet peridotite xenoliths, whereas partitioning of REE among the phases is rather similar to the sheared peridotites. The data indicate that chemical equilibrium of major elements and REE was not maintained among the exsolved phases during exsolution and subsequent recrystallization, after the formation of the original homogeneous clinopyroxene megacryst.

Equilibration conditions of upper-mantle eclogites: implications for kyanite-bearing and diamondiferous varieties

AbstractNew mineral data for kyanite-bearing eclogites supplement existing data for eclogite nodule suites from the Roberts Victor and Bellsbank kimberlite intrusions, South Africa. Calculation of equilibration temperatures using the refined KD calibration of Ellis and Green (1979) indicates that the majority of these eclogites equilibrated over a considerably narrower range (940–1185 °C) than implied by earlier estimates which did not take account of the influence of the Ca-component of garnet on KD. Most of the Roberts Victor diamondiferous eclogites appear to have equilibrated in the lower part of this range, whilst the equilibration temperatures for the Roberts Victor diamond-free eclogites extend to higher values than those from Bellsbank. The petrogenetic implications of the (calculated) temperature and pressure equilibration conditions for these eclogite suites are discussed with particular reference to possible uppermantle geotherms and available data on the graphitediamond transition boundary.

The significance of the inter-element relationships of some minor and trace elements in south african kimberlites

Fifty-eight kimberlite samples from three localities: Premier, Koffyfontein-Ebenhaezer and Bellsbank, have been analysed for Ti, Nb, Ta, Zr, Hf, Sc, Cr, P and Y by modern instrumental techniques (XRF and INAA). From a comparison of the data obtained with that of crustal rocks, it is concluded that these elements are not grossly affected by crustal contamination. Abundance data are similar to data avilable for kimberlites from other localities. The micaceous kimberlites are characterized by higher concentrations of Nb, Ta, Zr, Hf, Sc and P (in addition to Sr, REE, Th and U) when compared with non-micaceous kimberlites. The large range in concentration of elements such as Nb, Ta and P in the Bellsbank kimberlites cannot be accounted for by dilution with crustal carbonate. The formation of stable complexes in the presence of ions like OH −, CO 3 2− and F − is considered to be a possible explanation for these high concentrations. The concentration of incompatible elements in kimberlite can only in part be accounted for by phlogopite in the source rock, and the presence of other minor phases such as apatite and rutile are suggested by the inter-element relationships found.

Aspects of the geochemistry of kimberlites from the Premier mine, and other selected South African occurrences with particular reference to the rare earth elements

Comparative major element, as well as trace element data for REE, Sr, Ba, Th, U, Rb and Cs for three South African kimberlite groups: Premier, Koffyfontein—Ebenhaezer and Bellsbank kimberlites are presented. These kimberlites have different ages, modes of emplacement, degree and type of crustal contamination. Major element data indicates such contamination and requires correction if Si Mg > 0.9 . Loss of volatiles in the relative order Cs > Rb > K near wall rock contacts and intrusive dykes are evident in Premier kimberlites. High and variable Th U generally indicates loss of U by groundwater leaching in kimberlites. The refractory elements, Ti, Nb, Ta, Zr, Hf and the REE as well as P, are considered to be insignificantly affected by crustal contamination. It is suggested that a study of their geochemistry may elucidate the genetic relationship of upper mantle derived material. The close and linear association of some of these elements observed in kimberlites suggests the presence of a carbonated-phosphate phase, probably apatite, as a host mineral for LREE, Th, U, Sr and volatiles (e.g. F −, Cl −, H 2O) in fertile upper mantle assemblages. REE in South African kimberlites show different degrees of enrichment and LREE/HREE fractionation previously only reported in carbonatites. South African kimberlites show varying degrees of Eu depletion, with anomalies most conspicuous in carbonate-rich micaceous kimberlites. More “primitive” Premier kimberlites exhibit weak negative Yb anomalies attributed to garnet fractionation. REE whole rock patterns in kimberlites are influenced by mineralogical effects. REE enrichment and fractionation in kimberlitic magmas is considered to be dominantly controlled by the selective formation and removal of mobile REE-carbonate complexes into carbonatitic fluids which separate from the silica-richer complementary “kimberlite” magma during emplacement. A close genetic relationship between kimberlites and carbonatites is implied.