Kimberlite AT-56: a mantle sample from the north central Superior craton, Canada

Kimberlite AT-56, discovered in February 2001, represents the most recent addition to the Attawapiskat kimberlite cluster, located in the James Bay Lowlands of Ontario, Canada. AT-56 is a small kimberlite body with a surface diameter of approximately 40 m and a steep southeastern plunge. It consists of a medium to coarse-grained matrix supported kimberlite with abundant olivine, clinopyroxene, garnet, ilmenite and mica macrocrysts in a green-black to orange-black matrix. The kimberlite is classified as a hypabyssal facies sparsely macrocrystic calcite kimberlite. Heavy mineral concentrates from two representative samples of AT-56 have been analyzed to characterize the mantle sampled by the kimberlite. Both samples yielded large heavy mineral concentrates comprised of roughly equal proportions of Mg-ilmenite, Cr-diopside, high-Cr garnet and low-Cr garnet. Mg-chromite is also present in quantities an order of magnitude less than the other constituents.

The high-Cr peridotitic garnet macrocrysts are only slightly more abundant than the low-Cr varieties, the population being dominated by G9 (lherzolitic) types with only a few (less than 10%) weakly sub-calcic G10 (probable harzburgitic) garnets present. Ni thermometry results for a representative selection of G9 and G10 garnets indicate that the majority equilibrated at temperatures ranging from 1000 to 1250 °C. A significant proportion of the low-Cr garnet population derived from AT-56 is characterized by relatively low-Ti (0.2 to 0.4 wt.% TiO₂) and elevated Na (0.07 to 0.13 wt.% Na₂O) contents characteristic of Group 1, diamond inclusion type eclogite garnets. These sodic garnets have elevated Cr₂O₃ contents (typically 1 to 2 wt.% Cr₂O₃), suggesting they may be websteritic in origin rather than eclogitic. Comparison of AT-56 garnet compositions with published data available for other Attawapiskat kimberlites suggests websteritic mantle has also been sampled by kimberlite bodies elsewhere in the Attawapiskat cluster and it may be an important diamond reservoir in this area.     

A classification scheme for mantle-derived garnets in kimberlite: a tool for investigating the mantle and exploring for diamonds

A new empirical method has been devised for classification of mantle-derived garnets in kimberlite. Simple chemical screens have been developed to distinguish between garnets from different parageneses, based on Mg, Fe, Ca, Cr, Ti and Na values of published analyses of garnets from >2000 ultramafic xenoliths in kimberlite. Although crustal garnets are typically uncommon as xenocrysts in kimberlite, the first step in the classification is to screen these from the mantle population, using data from >600 garnet-bearing crustal rocks. Such a screen may also prove useful in evaluating the source (crust vs. mantle) of garnet in kimberlite exploration samples. Subsequent steps divide mantle garnets into eclogite, peridotite and Cr-poor megacryst groupings, and sub-groups of the peridotite (lherzolite, harzburgite, wehrlite) and eclogite (Groups I and II and A, B, C and grospydite) populations. Important features of this classification include the fact that it is based on distinctions between groups of fundamental geological significance (e.g., peridotite vs. eclogite) and it is based on a large, well-documented and well-understood xenolith database. As it utilizes oxide values and molar ratios of major and minor elements, the rationale for the screens is readily understood and it is simple to use.     

Trace element partitioning between carbonatitic melts and mantle transition zone minerals: Implications for the source of carbonatites

The occurrence of CO₂-rich lavas (carbonatites, kimberlites) and carbonate-rich xenoliths provide evidence for the existence of carbonatitic melts in the mantle. To model the chemical composition of such melts in the deep mantle, we experimentally determined partition coefficients for 23 trace elements (including REE, U-Th, HFSE, LILE) between deep mantle minerals and carbonatite liquids at 20 and 25 GPa and 1600 °C. Under these conditions, majoritic garnet and CaSiO₃ perovskite are the main reservoirs for trace elements. This study used both femtosecond LA-ICP-MS and SIMS techniques to measure reliable trace element concentrations. Comparison of the two techniques shows a general agreement, except for Sc and Ba. Our experimentally determined partition coefficients are consistent with the lattice strain model. The data suggest an effect of melt structure on partition coefficients in this pressure range. For instance, strain-free partition coefficient (D₀) for majorite-carbonatite melts do not follow the order of cation valence, , observed for majorite-CO₂-free silicate melts. The newly determined partition coefficients were combined with trace element composition of majoritic garnets found as inclusions in diamond to model trace element patterns of deep-seated carbonatites. The result compares favorably with natural carbonatites. This suggests that carbonatites can originate from the mantle transition zone.     

Interaction between protokimberlite melts and mantle lithosphere: Evidence from mantle xenoliths from the Dalnyaya kimberlite pipe,Yakutia (Russia)

The Dalnyaya kimberlite pipe(Yakutia,Russia) contains mantle peridotite xenoliths(mostly Iherzolites and harzburgites) that show both sheared porphyroclastic(deformed) and coarse granular textures,together with ilmenite and clinopyroxene megacrysts.Deformed peridotites contain high-temperature Fe-rich clinopyroxenes,sometimes associated with picroilmenites,which are products of interaction of the lithospheric mantle with protokimberlite related melts.The orthopyroxene-derived geotherm for the lithospheric mantle beneath Dalnyaya is stepped similar to that beneath the Udachnaya pipe.Coarse granular xenoliths fall on a geotherm of 35 mWm-2 whereas deformed varieties yield a 45 mWm-2)geotherm in the 2-7.5 GPa pressure interval.The chemistry of the constituent minerals including garnet,olivine and clinopyroxene shows trends of increasing Fe~#(=Fe/(Fe+Mg))with decreasing pressure.This may suggest that the interaction with fractionating protokimberlite melts occurred at different levels.Two major mantle lithologies are distinguished by the trace element patterns of their constituent minerals,determined by LA-ICP-MS.Orthopyroxenes,some clinopyroxenes and rare garnets are depleted in Ba,Sr,HFSE and MREE and represent relic lithospheric mantle.Re-fertilized garnet and clinopyroxene are more enriched.The distribution of trace elements between garnet and clinopyroxene shows that the garnets dissolved primary orthopyroxene and clinopyroxene.Later high temperature clinopyroxenes related to the protokimberlite melts partially dissolved these garnets.Olivines show decreases in Ni and increases in Al,Ca and Ti from Mg-rich varieties to the more Fe-rich,deformed and refertilized ones.Minerals showing higher Fe~#(0.11-0.15) are found within intergrowths of low-Cr ilmenite-clinopyroxene-garnet related to the crystallization of protokimberlite melts in feeder channels.In P-f(O_2) diagrams,garnets and Cr-rich clinopyroxenes indicate reduced conditions at the base of the lithosphere at-5 log units below a FMQ buffer.However,Cr-poor clinopyroxenes,together with ilmenite and some Fe-Ca-rich garnets,demonstrate a more oxidized trend in the lower part of lithosphere at-2 to 0 log units relative to FMQ.Clinopyroxenes from xenoliths in most cases show conditions transitional between those determined for garnets and megacrystalline Cr-poor suite.The relatively low diamond grade of Dalnyaya kimberlites is explained by a high degree of interaction with the oxidized protokimberlite melts,which is greater at the base of the lithosphere.     

Petrochemistry of eclogites from the Koidu Kimberlite Complex,Sierra Leone

Petrography, mineral and bulk chemistry of upper mantle-derived eclogites (garnet and clinopyroxene) from the Koidu Kimberlite Complex, Sierra Leone, are presented in the first comprehensive study of these xenoliths from West Africa. Although peridotite-suite xenoliths are generally more common in kimberlites, the upper mantle sample preserved in Pipe Number 1 at Koidu is exclusively eclogitic, making this the fifth locality in which eclogite is the sole polymineralic xenolith in kimberlite. Over 2000 xenoliths were collected, of which 47 are described in detail that include diamond, graphite, kyanite, corundum, quartz after coesite, and amphibole eclogites. Grossular-pyrope-almandine garnets are chromium-poor (<0.72 wt% Cr₂O₃) and fall into two distinct groups based on magnesium content. High-MgO garnets have an average composition of Pyr₆₇Alm₂₂Gross₁₁, low-MgO garnets are grossular- and almandine-rich with an average composition of Gross₃₄Pyr₃₃Alm₃₃. Clinopyroxenes are omphacitic with a range in jadeite contents from 7.7 to 70.1 mol%. Three eclogites contain zoned and mantled garnets with almandine-rich cores and pyrope-rich rims, and zoned clinopyroxenes with diopside-rich cores and jadeite-rich rims, and are among a very rare group of eclogites reported on a world-wide basis. The bulk compositions of eclogites have ranges comparable to that of basalts. High-MgO eclogites (16–20 wt% MgO) have close chemical affinities to picrites, whereas low-MgO eclogites (6–13 wt% MgO) are similar to alkali basalts. High-MgO eclogites contain high-MgO garnets and jadeiterich clinopyroxenes. Low-MgO eclogites contain low-MgO garnets, diopside and omphacite, and the group of primary accessory phases (diamond, graphite, quartz after coesite, kyanite, and corundum); grospydites are peraluminous. Estimated temperatures and pressures of equilibration of diamond-bearing eclogites, using the diamond-graphite stability curve and the Ellis and Green (1979) geothermometer, are 1031°–1363° C at 45–50 kb.K _D values of Fe-Mg in garnet and clinopyroxene range from 2.3 to 12.2. Diamonds in eclogites are green, yellow, and clear, and range from cube to octahedral morphologies; the entire spectrum in color and morphology is present in a single metasomatized eclogite with zoned garnet and clinopyroxene. Ages estimated from Sm-Nd mineral isochrons range from 92–247 Ma. _Nd values range from +4.05 to 5.23. Values of specific gravity range from 3.06–3.60 g/cc, with calculated seismic Vp of 7.4–8.7 km/s. Petrographie, mineral, and bulk chemical data demonstrate an overall close similarity between the Koidu xenolith suite and upper mantle eclogites from other districts in Africa, Siberia and the United States. At least two origins are implied byP-T, bulk chemistry and mineral compositions: low-MgO eclogites, with diamond and other accessory minerals, are considered to have formed from melts trapped and metamorphically equilibrated in the lithosphere; high-MgO eclogites are picritic and are the products of large degrees of partial melting, with equilibration in the asthenosphere. Fluid or diluted melt metasomatism is pervasive and contributed here and elsewhere to the LIL and refractory silicate incompatible element signature in kimberlites and lamproites, and to secondary diamond growth.     

Enrichment processes at the base of the Archean lithospheric mantle: observations from trace element characteristics of pyropic garnet inclusions in diamonds

Trace element concentrations of peridotitic garnet inclusions in diamonds from two Chinese kimberlite pipes were determined using the ion microprobe. Garnet xenocrysts from the same two kimberlite pipes were also analyzed for comparison. In contrast to their extremely refractory major element compositions, all harzburgitic garnets showed enrichment in light rare earth elements (REE) relative to chondrite, resulting in sinuous REE patterns. Both normal and sinuous REE patterns were observed from the lherzolitic garnets. Concentrations of REE in garnets changed significantly from diamond to diamond and no specific correlations were observed with their major element compositions. Analyses of randomly selected two to three points within every grain of a large number of garnet inclusions by the ion microprobe demonstrated that there was no evident compositional heterogeneity, and multiple grains of one phase from a single diamond host also exhibit very similar compositions. This implies that the trace element heterogeneity within one grain or among multiple inclusions from the same diamond host, as reported from Siberian diamonds, is not a common feature for these Chinese diamonds. Concentrations of Na, Ti, and Zr tend to decrease when garnets become more refractory, but variations of Sr and Li are more complex. Compositions rich in light REE and relatively poor in high field strength elements (HFSE) of the harzburgitic garnet inclusions in diamonds are generally consistent with metasomatism by carbonatite melts. The trace element features observed from the garnet inclusions in Chinese diamonds may be caused by carbonatite melt infiltration and partial melt extraction. Spatial and temporal gradients in melt/rock ratio and temperature are the main reasons for the large variations of REE patterns and other trace element concentrations. Received: 27 April 1999 / Accepted: 1 March 2000     

Metasomatic origin of garnet xenocrysts from the V. Grib kimberlite pipe,Arkhangelsk region,NW Russia

This paper presents new major and trace element data from 150 garnet xenocrysts from the V. Grib kimberlite pipe located in the central part of the Arkhangelsk diamondiferous province (ADP). Based on the concentrations of Cr₂O₃, CaO, TiO₂ and rare earth elements (REE) the garnets were divided into seven groups: (1) lherzolitic “depleted” garnets (“Lz 1”), (2) lherzolitic garnets with normal REE patterns (“Lz 2”), (3) lherzolitic garnets with weakly sinusoidal REE patterns (“Lz 3”), (4) lherzolitic garnets with strongly sinusoidal REE patterns (“Lz 4”), (5) harzburgitic garnets with sinusoidal REE patterns (“Hz”), (6) wehrlitic garnets with weakly sinusoidal REE patterns (“W”), (7) garnets of megacryst paragenesis with normal REE patterns (“Meg”). Detailed mineralogical and geochemical garnet studies and modeling results suggest several stages of mantle metasomatism influenced by carbonatite and silicate melts. Carbonatitic metasomatism at the first stage resulted in refertilization of the lithospheric mantle, which is evidenced by a nearly vertical CaO-Cr₂O₃ trend from harzburgitic (“Hz”) to lherzolitic (“Lz 4”) garnet composition. Harzburgitic garnets (“Hz”) have probably been formed by interactions between carbonatite melts and exsolved garnets in high-degree melt extraction residues. At the second stage of metasomatism, garnets with weakly sinusoidal REE patterns (“Lz 3”, “W”) were affected by a silicate melt possessing a REE composition similar to that of ADP alkaline mica-poor picrites. At the last stage, the garnets interacted with basaltic melts, which resulted in the decrease CaO-Cr₂O₃ trend of “Lz 2” garnet composition. Cr-poor garnets of megacryst paragenesis (“Meg”) could crystallize directly from the silicate melt which has a REE composition close to that of ADP alkaline mica-poor picrites. P-T estimates of the garnet xenocrysts indicate that the interval of ～60–110 km of the lithospheric mantle beneath the V. Grib pipe was predominantly affected by the silicate melts, whereas the lithospheric mantle deeper than 150 km was influenced by the carbonatite melts.     

Cr-rich megacrysts of clinopyroxene and garnet from Lac de Gras kimberlites,Slave Craton,Canada – implications for the origin of clinopyroxene and garnet in cratonic lherzolites

Kimberlites from the Diavik and Ekati diamond mines in the Lac de Gras kimberlite field contain abundant large (>1 cm) clinopyroxene (Cr-diopside) and garnet (Cr-pyrope) crystals. We present the first extensive mineral chemical dataset for these megacrysts from Diavik and Ekati and compare their compositions to cratonic peridotites and megacrysts from the Slave and other cratons. The Diavik and Ekati Cr-diopside and Cr-pyrope megacrysts are interpreted to belong to the Cr-rich megacryst suite. Evidence for textural, compositional, and isotopic disequilibrium suggests that they constitute xenocrysts in their host kimberlites. Nevertheless, their formation may be linked to extensive kimberlite magmatism and accompanying mantle metasomatism preceding the eruption of their host kimberlites. It is proposed that the formation of megacrysts may be linked to failed kimberlites. In this scheme, the Cr-rich megacrysts are formed by progressive interaction of percolating melts with the surrounding depleted mantle (originally harzburgite). As these melts percolate outwards, they may contribute to the introduction of clinopyroxene and garnet into the depleted mantle, thereby forming lherzolite. This model hinges on the observation that lherzolitic clinopyroxenes and garnets at Lac de Gras have compositions that are strikingly similar to those of the Cr-rich megacrysts, in terms of major and trace elements, as well as Sr isotopes. As such, the Cr-rich megacrysts may have implications for the origin of clinopyroxene and garnet in cratonic lherzolites worldwide.

     

Major element composition of concentrate garnets in Proterozoic kimberlites from the Eastern Dharwar Craton,India: Implications on sub-continental lithospheric mantle

This study examines the major element composition of mantle-derived garnets recovered from heavy mineral concentrates of several Proterozoic kimberlites of the diamondiferous Wajrakarur Kimberlite Field (WKF) and the almost barren Narayanpet Kimberlite Field (NKF) in the Eastern Dharwar Craton of southern India. Concentrate garnets are abundant in the WKF kimberlites, and notably rare in the NKF kimberlites. Chemical characteristics of the pyropes indicate that the lithology of the sub-continental lithospheric mantle (SCLM) beneath both the kimberlite fields was mainly lherzolitic at the time of kimberlite eruption. A subset of green pyropes from the WKF is marked by high CaO and Cr₂O₃ contents, which imply contribution from a wehrlitic source. The lithological information on SCLM, when studied alongside geobarometry of lherzolite and harzburgite xenoliths, indicates that there are thin layers of harzburgite within a dominantly lherzolitic mantle in the depth interval of 115–190 km beneath the WKF. In addition, wehrlite and olivine clinopyroxenite occur locally in the depth range of 120–130 km. Mantle geotherm derived from xenoliths constrains the depth of graphite–diamond transition to 155 km beneath the kimberlite fields. Diamond in the WKF thus could have been derived from both lherzolitic and harzburgitic lithologies below this depth. The rarity of diamond and garnet xenocrysts in the NKF strongly suggest sampling of shallower (<155 km depth) mantle, and possibly a shallower source of kimberlite magma than at the WKF.     

The discovery of garnets closely related to diamonds in the Finsch pipe,South Africa

Magnesium-rich, calcium-poor, lilac coloured garnets have been found in the heavy mineral concentrate of the Finsch kimberlite pipe. Some of these garnets contain sufficient chromium to place them within the compositional field of the garnets previously only reported as inclusions in diamonds.These lilac garnets are considered to have formed in equilibrium with the minerals found as inclusions in diamond and hence with the diamond itself. Their presence in the kimberlite should be diagnostic of the presence of diamond, but it is not known if there is any quantitative relationship. The garnets are considered to have a deeper provenance than the magnesian garnets commonly found as xenocrysts in kimberlite and in garnet peridotite xenoliths. The mantle composition at their depths of origin must be more refractory in nature than the peridotite xenoliths. The garnets having a higher magnesium and chromium content, a higher Mg/Fe ratio and lower calcium, aluminium and titanium than those found in the xenoliths.     

Nd-isotopes in selected mantle-derived rocks and minerals and their implications for mantle evolution

The Sm-Nd systematics in a variety of mantle-derived samples including kimberlites, alnoite, carbonatite, pyroxene and amphibole inclusions in alkali basalts and xenolithic eclogites, granulites and a pyroxene megacryst in kimberlites are reported. The additional data on kimberlites strengthen our earlier conclusion that kimberlites are derived from a relatively undifferentiated chondritic mantle source. This conclusion is based on the observation that the _Nd values of most of the kimberlites are near zero. In contrast with the kimberlites, their garnet lherzolite inclusions show both time-averaged Nd enrichment and depletion with respect to Sm. Separated clinopyroxenes in eclogite xenoliths from the Roberts Victor kimberlite pipe show both positive and negative _Nd values suggesting different genetic history. A whole rock lower crustal scapolite granulite xenolith from the Matsoku kimberlite pipe shows a negative _Nd value of -4.2, possibly representative of the base of the crust in Lesotho. It appears that all inclusions, mafic and ultramafic, in kimberlites are unrelated to their kimberlite host.The above data and additional Sm-Nd data on xenoliths in alkali basalts, alpine peridotite and alnoite-carbonatites are used to construct a model for the upper 200 km of the earth's mantle — both oceanic and continental. The essential feature of this model is the increasing degree of fertility of the mantle with depth. The kimberlite's source at depths below 200 km in the subcontinental mantle is the most primitive in this model, and this primitive layer is also extended to the suboceanic mantle. However, it is clear from the Nd-isotopic data in the xenoliths of the continental kimberlites that above 200 km the continental mantle is distinctly different from their suboceanic counterpart.     

Nd isotopic ages of crust formation and metamorphism in the Precambrian of eastern and southern Mexico

Sm-Nd ages for garnets in the three Precambrian exposures of eastern and southern Mexico demonstrate that they belong to the Grenville tectonothermal event. The Sm-Nd garnet ages of 0.95 Ga for the Oaxacan Complex and 0.90 Ga for the Huiznopala Gneiss, Molango and the Novillo Gneiss, Ciudad Victoria, are postdated 75 Ma by Rb-Sr ages on biotites. Both sets of data document a cooling history following Grenville metamorphism at or before 1.0 Ga ago. Our garnet data are consistent with a blocking temperature for Sm-Nd in that mineral around 600° C suggested by Humphries and Cliff (1982).The three Precambrian occurrences have Nd chemical ages of separation from depleted mantle (T_DM) grouped in the range 1.40–1.60 Ga. This may result from derivation of the rocks from actual crustal protoliths which had been separated from the mantle 0.5 Ga before the Grenville Orogeny. It is much more likely, however, that crustal materials of 1.7 Ga or older age were mixed with mantle-derived products during Grenville events to produce intermediate T_DM ages and _Nd values around zero 1.0 Ga ago.     

Kimberlite emplacement and mantle sampling through time at A154N kimberlite volcano,Diavik Diamond Mine: lessons from the deep

The Diavik Diamond Mine in the NWT of Canada has produced in excess of 100 million carats from 3 kimberlite pipes since mining commenced in 2002. Here, we present new findings from deep (>400 m below surface) mining, sampling and drilling work in the A154N kimberlite volcano that require a revision of previous geological and emplacement models and provide a window into how the sub-continental lithospheric mantle (SCLM) below Diavik was sampled by kimberlite magmas through time. Updated internal geological models feature two volcanic packages interpreted to represent two successive cycles of explosive eruption followed by active and passive sedimentation from a presumed crater-rim, both preceded and followed by intrusions of coherent kimberlite. Contact relationships apparent among the geological units allow for a sequential organization of as many as five temporally-discrete emplacement events. Representative populations of mantle minerals extracted from geological units corresponding to four of the emplacement events at A154N are analyzed for major and trace elements, and provide insights into the whether or not kimberlites randomly sample from the mantle. Two independent geothermometers using clinopyroxene and garnet data indicate similar source depths for clinopyroxenes and G9 garnets (130–160 km), and suggest deeper sampling with time for both clinopyroxene and garnets. Harzburgite is limited to 110–160 km, and appears more prevalent in early, low-volume events. Variable ratios of garnet parageneses from the same depth horizons suggest random sampling by passing magmas, but deeper garnet sampling through time suggests early preferential sampling of shallow/depleted SCLM. Evaluations of Ti, Zr, Y and Ga over the range of estimated depths support models of the SCLM underlying the central Slave terrane.

     

The trace element composition of silicate inclusions in diamonds: a review

On a global scale, peridotitic garnet inclusions in diamonds from the subcratonic lithosphere indicate an evolution from strongly sinusoidal REE_N, typical for harzburgitic garnets, to mildly sinusoidal or “normal” patterns (positive slope from LREE_N to MREE_N, fairly flat MREE_N–HREE_N), typical for lherzolitic garnets. Using the Cr-number of garnet as a proxy for the bulk rock major element composition it becomes apparent that strong LREE enrichment in garnet is restricted to highly depleted lithologies, whereas flat or positive LREE–MREE slopes are limited to less depleted rocks. For lherzolitic garnet inclusions, there is a positive relation between equilibration temperature, enrichment in MREE, HREE and other HFSE (Ti, Zr, Y), and decreasing depletion in major elements. For harzburgitic garnets, relations are not linear, but it appears that lherzolite style enrichment in MREE–HREE only occurs at temperatures above 1150–1200 °C, whereas strong enrichment in Sr is absent at these high temperatures. These observations suggest a transition from melt metasomatism (typical for the lherzolitic sources) characterized by fairly unfractionated trace and major element compositions to metasomatism by CHO fluids carrying primarily incompatible trace elements. Melt and fluid metasomatism are viewed as a compositional continuum, with residual CHO fluids resulting from primary silicate or carbonate melts in the course of fractional crystallization and equilibration with lithospheric host rocks.

Eclogitic garnet inclusions show “normal” REE_N patterns, with LREE at about 1× and HREE at about 30× chondritic abundance. Clinopyroxenes approximately mirror the garnet patterns, being enriched in LREE and having chondritic HREE abundances. Positive and negative Eu anomalies are observed for both garnet and clinopyroxene inclusions. Such anomalies are strong evidence for crustal precursors for the eclogitic diamond sources. The trace element composition of an “average eclogitic diamond source” based on garnet and clinopyroxene inclusions is consistent with derivation from former oceanic crust that lost about 10% of a partial melt in the garnet stability field and that subsequently experienced only minor reenrichment in the most incompatible trace elements. Based on individual diamonds, this simplistic picture becomes more complex, with evidence for both strong enrichment and depletion in LREE.

Trace element data for sublithospheric inclusions in diamonds are less abundant. REE in majoritic garnets indicate source compositions that range from being similar to lithospheric eclogitic sources to strongly LREE enriched. Lower mantle sources, assessed based on CaSi–perovskite as the principal host for REE, are not primitive in composition but show moderate to strong LREE enrichment. The bulk rock LREE_N–HREE_N slope cannot be determined from CaSi–perovskites alone, as garnet may be present in these shallow lower mantle sources and then would act as an important host for HREE. Positive and negative Eu anomalies are widespread in CaSi–perovskites and negative anomalies have also been observed for a majoritic garnet and a coexisting clinopyroxene inclusion. This suggests that sublithospheric diamond sources may be linked to old oceanic slabs, possibly because only former crustal rocks can provide the redox gradients necessary for diamond precipitation in an otherwise reduced sublithospheric mantle.     

On the origin of silicate-bearing diamondites

Garnets and clinopyroxenes, intergrown with diamonds in 37 diamondites (“bort”, “polycrystalline diamond aggregates”, “polycrystalline diamond”, “framesite”), presumably from southern Africa, were analyzed for trace element contents by LA-ICP-MS. The intimate diamond-silicate intergrowths suggest that both precipitated from the same fluids during the same crystallization events. In this study we distinguish 5 chemical garnet groups: “peridotitic” (P), intermediate (I) and 3 “eclogitic” groups (E1, E2 and E3). Chondrite-normalized trace element patterns for the garnet groups roughly correlate with major element abundances. Most of P garnets show complex, mildly sinusoidal REE_N patterns with relatively flat HREE_N-MREE_N, a small hump at Sm-Nd and depleted LREE_N, and have relatively high contents of Nb, Ta, U, and Th. The REE_N abundance patterns of E garnets differ by showing a continuous increase from LREE to HREE and depletion in LREE and highly incompatible elements relative to the P garnets. Of all garnet groups, E3 garnets are the poorest in highly incompatible trace elements and in Mg. Model equilibrium fluids for P garnets suggest crystallization from magnesian carbonate-bearing fluids/melts, which were very rich in incompatible trace elements — similar to kimberlites. Hypothetical equilibrium melts for E1 and E2 garnets are also magnesian and poorer in LREE and highly incompatible elements relative to typical kimberlitic or carbonatitic melts. Fluids that crystallized the P and most of the E garnets have similar mg numbers indicating a peridotitic source for both. The differences in Cr and highly incompatible element contents can be the result of differences in fluid formation and/or evolution rather than different source rock. The positive correlation of Cr₂O₃ and mg with the abundances of highly incompatible elements in garnets indicate fluid-rock fractionation processes rather than igneous fractional crystallization processes being responsible for the evolution of the diamondite-forming fluids.     

Origins of subcalcic garnets and their relation to diamond forming fluids—Case studies from Ekati (NWT-Canada) and Murowa (Zimbabwe)

Subcalcic, high-Cr (G10) garnets are found as inclusions within diamonds and in peridotitic xenoliths. The strong spatial associations between G10 garnets and diamond make them an important tool in the investigation of diamond genesis. We present an integrated study of the major and trace element composition and oxygen-Sr-Nd-Hf isotopic ratios of eight G10 garnets from the Ekati mine (NWT-Canada) and four from the Murowa mine (Zimbabwe) in an attempt to determine their petrogenetic evolution and to further examine a possible relationship between the metasomatic agents responsible for G10 garnet signatures and diamond forming fluids.All garnets display sinusoidal to mildly sinusoidal REE patterns and have negative Ti, Sr and positive U anomalies. They have variably radiogenic ⁸⁷Sr/⁸⁶Sr (0.703261-0.731191) and non-radiogenic ε_Nd values (−8.1 to −27.1), except for one sample from Murowa that has a positive ε_Nd of 2.5. One Ekati sample has an extremely low ε_Hf value of −61.6. The Ekati garnets we have studied all appear to come from a single depth in the Slave lithospheric mantle. On the base of Cr-Ca relations they have crystallized at 4.9 GPa and display dunitic Ca intercept values. Their δ¹⁸O values range between +5.23‰ and +5.42‰.The Ekati G10 garnets record a complex, multi-stage metasomatic history involving the interaction of several components during their genesis. One metasomatic agent was enriched in HFSE, LREE, Sr, and depleted in Nb. This agent had the least radiogenic Sr. Another metasomatic agent had highly radiogenic Sr, and was enriched in LREE, Sr, Nb, Th and U.The G10 garnets have very low ε_Nd and ε_Hf values combined with radiogenic Sr, thus, they require an early lithospheric mantle enrichment event at some stage during their genesis or during the evolution of any precursor material that they formed from. The only Hf isotope composition measurable from the Ekati suite is so unradiogenic (ε_Hf = −61) that it yields a Lu/Hf model age of 3521 Ma. This indicates that the lithospheric enrichment event seen by the Ekati garnets or their precursors may have occurred in the early stages of the craton stabilization, during the diamond forming event [Westerlund K., Shirey S., Richardson S., Carlson R., Gurney J. and Harris J. (2006) A subduction wedge origin for Paleoarchean peridotitic diamonds and harzburgites from the Panda kimberlite, Slave craton: evidence from Re-Os isotope systematics. Contrib. Mineral. Petrol.152(3), 275-294]. Although our data cannot unequivocally discriminate between a variety of models for the genesis of subcalcic garnets it is clear that the host peridotite originated via melting at shallow depths followed by subduction and that the observed geochemical fingerprint of the garnets is strongly influenced by diamond forming fluids. Diamond forming fluids sampled from fibrous diamonds, have steep REE patterns, negative Ti and Sr anomalies and very low Sm/Nd ratios that are very similar to G10 garnet characteristics. These diamond forming fluids have been recently shown to have extreme Sr and Nd isotopic compositions [Klein-BenDavid O., Pearson D. G., Nowell G. M. and Cantigny P. (2008) Origins of diamond forming fluids—constraints from a coupled Sr-Nd isotope and trace element approach. Extended abstracts to the 9th International Kimberlite Conference, Frankfurt, Germany, 9IKC-A-00118.] that are closely concordant with G10 garnets. The fluids are also rich in LREE, P, K and water, sharing these features with mica-rich metasomes. These similarities suggest that ancient lithospheric metasomes could either provide a source region for, or be a product of diamond forming fluids. Diamond forming fluids appear to be intimately involved in the evolution of G10 garnets in the lithospheric mantle, either acting as a metasomatic agent, or being integral to triggering or enhancing garnet growth in a Cr-rich protolith. Such a link explains the strong association between G10 garnets and diamonds.     

Rare and unusual mineral inclusions in diamonds from Mwadui, Tanzania

Syngenetic diamond inclusions from the Mwadui kimberlite reveal that an unusually fertile section of lithospheric mantle beneath the Central African Craton was sampled. This is shown by a very high ratio of lherzolitic to harzburgitic garnet inclusions (1:2) and low Mg/Fe-ratios in olivine and orthopyroxene. Geothermometry applied to the peridotitic inclusions indicates disequilibrium between non-touching inclusion pairs to be common. Disequilibrium between garnet-olivine and garnet-orthopyroxene pairs suggests successive iron enrichment during diamond formation, e.g. leading to the presence of harzburgitic garnet and lherzolitic olivine in the same diamond. Apart from the dominant peridotitic inclusion suite (88%), rare eclogitic inclusions occur (2%) and a number of uncertain paragenesis. Two diamonds, one with eclogitic garnets with moderate pyroxene solid solution and the other with a single ferro-periclase inclusion, suggest the contribution of a small sub-lithospheric component. The finding of the association Fe-FeO-Fe₃O₄ in one single diamond indicates diamond formation over a large range of f _O2 conditions, possibly along redox fronts. Steep compositional gradients may also be reflected by the joint occurrence of harzburgitic garnet and a SiO₂-phase in the same diamond. Alternatively the formation of the SiO₂-phase may be due to extreme carbonation of the peridotitic source. Further unusual findings include the exsolution of a silicate phase from magnetite inclusions, (i.e. primary solution of γ-olivine) and an ilmenite inclusion with an eskolaite (Cr₂O₃) component of 14.5 mol%, the latter together with harzburgitic paragenesis silicate inclusions. Received: 23 August 1997 / Accepted: 7 January 1998     

Shrilankite Inclusions in Garnets from Kimberlite Bodies and Diamondiferous Volcanic–Sedimentary Rocks of the Yakutian Kimberlite Province,Russia

Pyrope–almandine garnets (Mg# = 28.3–44.9, Ca# = 15.5–21.3) from a heavy mineral concentrate of diamondiferous kimberlites of the largest diamond deposit, the Yubileinaya pipe, along with kimberlite- like rocks and diamondiferous volcano–sediments of the Laptev Sea coast, have been found to contain polymineral, predominantly acicular inclusions, composed of aggregates of shrilankite (Ti₂ZrO₆), rutile, ilmenite, clinopyroxene, and apatite. The presence of shrilankite as an inclusion in garnets from assumed garnet–pyroxene rocks of the lower crust, lifted up by diamond-bearing kimberlite, allows it to be considered as an indicator mineral of kimberlite, which expands the possibilities when searching for kimberlite in the Arctic.     

Trace element abundance patterns of garnet inclusions in peridotite-suite diamonds

Individual, sub-calcic, chrome-pyrope crystals from Finsch and Kimberley diamonds, Finsch and Bultfontein kimberlite heavy mineral concentrate, and from diamondiferous harzburgite-dunite xenoliths from the Udachnaya kimberlite pipe were analyzed for rare earth elements (REE), Sc, Ti, and Zr with the ion microprobe. The abundances and abundance ratios of these trace elements including LREE enrichment and low Ti, together with high and variable Cr contents, are inconsistent with a simple equilibrium relationship between peridotite-suite garnet and silicate and carbonate liquids. It is suggested that the trace element abundance patterns represent a signature of ancient mantle metasomatism which preceded the formation of peridotite-suite garnet and diamond.