The largest diamonds are hosted in iron-rich substrate accreted at the base of the lithosphere

The largest diamonds are hosted in iron-rich substrate accreted at the base of the lithosphere | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Physical Sciences - Article The largest diamonds are hosted in iron-rich substrate accreted at the base of the lithosphere Geoffrey Howarth, Andrea Giuliani, Merrily Tau, Ronghua Cai, Jingao Liu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6808081/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Diamonds larger than 100 carats are some of the most valuable gemstones ever unearthed. They are hosted in mantle-derived magmas called kimberlites, but occur at few locations globally. Beyond their large size and rarity, these diamonds exhibit distinctive attributes such as exceptional clarity and irregular shape, leading to the CLIPPIR acronym1. The carbon isotopes of these diamonds indicate their origin from subducted slab material 2, 3, 4. While the formation of CLIPPIR diamonds in the mantle transition zone (MTZ) appears robustly constrained by the occurrence of majorite inclusions1, the nature of the CLIPPIR diamond substrate remains obscure. Here we show that CLIPPIR diamonds are associated with kimberlites tapping vertically extensive, Fe-rich and deformed domains at the base of the lithosphere. Beyond enrichment in Fe, these domains exhibit light oxygen and heavy Fe isotopes, which indicate a major role of subducted basaltic material that experienced hydrothermal alteration at the Earths surface. The association of CLIPPIR and other sub-lithospheric diamonds with these anomalous Fe-rich domains that are rarely sampled by kimberlites and their similar isotopic anomalies point to a genetic relation. Considerations on kimberlite genesis in the upper convective mantle and partial retrogression of majorite inclusions suggest that the CLIPPIR substrate originally stalled in the MTZ, where the diamonds grew, before being accreted at the base of the lithosphere. The geographic overlap between CLIPPIR diamond locations and the loci of large igneous provinces points to accretion of subducted slab material including dense eclogitic crust via buoyant mantle upwellings. Beyond providing the largest diamonds, these Fe-rich, isotopically anomalous domains contribute to the isotopic heterogeneity of intraplate magmas globally. Earth and environmental sciences/Solid Earth sciences/Petrology Earth and environmental sciences/Solid Earth sciences/Geology Figures Figure 1 Figure 2 Figure 3 Introduction Diamonds larger than several centimetres and >100 carats represent exceptional and highly valuable gems that are sold for millions of dollars 6 . The economic feasibility of some diamond mines such as Karowe (Orapa field) in Botswana and Letseng in northern Lesotho relies entirely on the extraction of these large stones. These diamonds can also provide glimpses of peculiar processes occurring in the deep Earth, which concentrate carbon and foster the growth of unusually large crystals in the deep mantle. Yet, these diamonds are notable components of the diamond population in few localities that are mostly restricted to southern Africa (e.g., Karowe in Botswana, Letseng in Lesotho, Premier in South Africa), and their origin is far from clear. Previous studies have shown that most of these very large diamonds exhibit specific features leading to the CLIPPIR (Cullinan-like, large, inclusion-poor, pure, irregular, resorbed) acronym 1 . The abundant occurrence of Fe-rich metallic inclusions in CLIPPIR diamonds testifies to reducing conditions associated with diamond formation and the rare occurrence of majorite inclusions places crystallisation within the mantle transition zone (MTZ: 410-660 km) 1 . The combination of extremely heavy Fe isotope composition (d 56 Fe = 0.79-0.90‰) and unradiogenic Os isotopes in the metallic inclusions was employed to suggest an origin of CLIPPIR diamonds from subducted serpentinized peridotite 6 . In contrast, the highly variable carbon isotope compositions of CLIPPIR diamonds (d 13 C = -27‰ to +4‰) 1, 2, 3 that overlaps the typical range of lithospheric eclogitic diamonds 7 indicates a crustal carbon source for the diamond-forming fluids. The composition of majorite (low Ti but high Na) and breyite inclusions (low Ti, but containing exsolved CaTiO 3 ) provides more equivocal constraints, which point to both crustal and peridotitic components. Partial retrogression of the majorite inclusions 1 suggest that CLIPPIR diamonds were probably accreted at the base of the lithosphere perhaps within diapirs of buoyant, harzburgitic lithosphere of oceanic ancestry that was subducted to the MTZ, similar to the model proposed for smaller sub-lithospheric diamonds 8 . Kimberlite magmas sourced in the upper convective mantle 9 then entrained and transported these diamonds to the surface. An alternative hypothesis entails crystallisation of CLIPPIR diamonds from kimberlitic melts in the deep lithosphere 4, 10 . This suggestion was based on the common occurrence of CLIPPIR diamonds in kimberlites with abundant Cr-poor megacrysts combined with the size of these diamonds reaching those of lithospheric megacrysts (>1 mm) – these large sizes are at odds with the smaller sizes predicted for convective mantle minerals 11 . This model, supported by recent observations of the relative abundance of megacrystic garnets in southern African kimberlites 12 , however, cannot account for the direct evidence of CLIPPIR diamond formation in the MTZ, namely the inclusions of partly retrogressed majorite 1 . The association between CLIPPIR and sub-lithospheric diamonds with megacrysts might simply indicated preferential sampling of the deep lithosphere rather than a genetic link 12, 13 . The source lithologies of these large diamonds remains obscure and potentially underpin the occurrence of unusual processes and/or peculiar domains in the mantle, which might have far-reaching implications for the formation of compositional heterogeneities in the deep Earth. In this study, we explore the association of CLIPPIR diamonds with Cr-poor megacrysts to resolve the origin of the largest diamonds recovered to date. Using the composition of olivine xenocrysts, we compare the compositional features of the lithospheric mantle entrained by kimberlites in areas containing abundant CLIPPIR and other sub-lithospheric diamonds with those that rarely contain such large stones. A combination of these results with existing oxygen isotope data for olivine and new bulk-kimberlite Fe isotopes suggests that CLIPPIR diamonds are associated with hydrothermally altered, Fe-rich basaltic crust that was subducted to the MTZ and then accreted at the base of the lithosphere as eclogite. Beyond providing abundant sub-lithospheric diamonds, these isotopically anomalous Fe-rich domains contribute to the compositional heterogeneity observed in intraplate magmas globally, including kimberlites. Iron enrichment of the deep lithosphere associated with CLIPPIR and other sub-lithospheric diamonds A global compilation of the composition of olivine in kimberlites 14, 15 shows that the major producers of CLIPPIR diamonds are kimberlites with abundant or dominant Fe-rich olivine, both xenocrystic (olivine cores) and magmatic (olivine rims; Mg#<89) ( Figure 1 and Fig S1 ). Previous work has robustly linked the elevated Fe contents of olivine in these kimberlites with sampling of lithospheric mantle roots containing abundant Cr-poor, Fe-rich megacrysts 14, 16 , consistent with recent working using garnet xenocrysts 12 . In this context, megacrysts represent large grains (>1 cm) that based on isotopic and geochronological similarities with kimberlites 17, 18 are genetically associated with failed pulses of kimberlitic magma interacting with the deep lithosphere 19, 20 . To explore a potential genetic connection between CLIPPIR diamonds and these Fe-rich, deep lithospheric domains, we compare the trace element compositions and sampling depths of olivine xenocrysts from kimberlites in areas where sampling of CLIPPIR and other sub-lithospheric diamonds is well established (Karowe and Damtshaa in the Orapa field; Letseng, Mothae, and Monastery in the northern Lesotho field; Premier kimberlite – new data) with those from localities that do not contain conspicuous populations of sub-lithospheric diamonds (Murowa kimberlite in Zimbabwe – new data; Kimberley kimberlites and Finsch in South Africa, Ekati and Jericho fields in Canada 16, 21, 22 ) ( Figure 2 ). Olivine sampling depths were obtained by combining temperature constraints from the Al-in-olivine thermometer 23 with independent geothermal gradients based on previous xenolith and xenocryst thermobarometry ( Methods ). Kimberlites from the CLIPPIR-bearing Orapa and northern Lesotho fields show a unimodal distribution of olivine sampling depth centred at between 130 and 155 km and limited sampling (51±17 %) within the diamond stability field of the lithosphere. Conversely, olivine in CLIPPIR-poor kimberlites was sampled predominantly (83±14 %) in the diamond stability field including major peaks at depths beyond 150-160 km. Major and trace element concentrations of olivine ( Figure 2 ) indicate that these differences represent a bias due to screening of high-Al and hence hot (~1200-1500°C) olivines enriched in Fe (Mg# <89), V, Zn and other transitional metals ( Dataset 1 ) outside the calibration range of the Al-in-olivine thermometer. These olivine compositions are typical of megacrysts and related sheared peridotites crystallised at pressures of 4.5-7.0 GPa (140-215 km), which are abundant in kimberlites from these CLIPPIR-bearing fields 12, 24, 25, 26, 27 . While 23% of the Premier olivines are megacrystic, the peridotitic olivine disapply unimodal distribution deeper than other CLIPPIR-bearing locations near the LAB (~210 km) but with sampling of predominantly (84% of peridotitic olivine) high-Al olivine from sheared peridotite. These olivine are similar to the high-Al megacrystic olivine but have slightly higher Mg# (>89) and, thus, pass the Al-in-olivine thermometer screening ( Methods ). The trace element composition of deep lithospheric, Al-rich olivine from the CLIPPIR-bearing localities indicates enrichment in Zn, Ti, Mn, and Co ( Figure 3a; Fig S5 ) suggestive of melt infiltration. These observations combined indicate that the lithospheric roots entrained by kimberlites in CLIPPIR-bearing fields are characterised by vertically extensive domains above the lithosphere-asthenosphere boundary (LAB) that contain abundant and locally dominant Fe-rich material with affinity to the megacryst suite and sheared peridotites. Existing oxygen isotope composition for olivine in kimberlites 28, 29 , including olivine megacrysts from Monastery 30 , highlights another key feature of these Fe-rich, deep lithospheric domains, namely low d 18 O values (<5‰; Figure 3b ) compared to typical mantle olivine (5.1 ± 0.3‰) 31 . Iron-rich olivines from the northern Lesotho and the Orapa fields, including Karowe, have d 18 O values as low as 4.0‰ (even lower at the megacryst- and eclogite-rich locality of Kaalvallei in South Africa 29 ). These isotopic compositions cannot be explained by common mantle processes, including metasomatism 29 , and require contribution by subducted crustal material, such as hydrothermally altered oceanic crust 32 or serpentinised lithospheric mantle 33 . Given that Fe-rich olivines ensue from the interaction between precursor kimberlite melts with mantle-like oxygen isotopes 28 and lithospheric mantle wall rocks, the subducted component in the wall rocks must have d 18 O <4‰ 29, 30 . New Fe isotope data for bulk kimberlites ( Methods ) add critical constraints to this picture. Comparison of these data with olivine Mg# (either magmatic, Figure 3c , or xenocrystic) indicates that Fe isotopes in kimberlites become progressively heavier with increasing Fe contents in olivine, with samples from the Premier, Orapa, and northern Lesotho kimberlite fields at the high d 56 Fe end of the measured range. Previous work has shown that the Fe content (or Mg#) of olivine in kimberlites reflects the relative amount of Fe-rich deep lithospheric material, including megacrystic and metasomatised lithologies (e.g., dunites, wehrlites, websterites), that gets entrained and partly assimilated by kimberlites 14 . In other words, heavy Fe isotopes in kimberlites reflect increasing contribution by Fe-rich megacrystic material that must feature heavier Fe isotope compositions compared to typical mantle values (d 56 Fe ~0.00 to 0.05‰) 34 . Mass balance calculations ( Methods ) show that the d 56 Fe of Fe-rich olivine in these deep lithospheric domains is higher than 0.11-0.30‰ compared to typical peridotitic olivine with d 56 Fe <0.05‰ 34, 35, 36 . Provided that the protolith interacted with by kimberlite melts (d 56 Fe ~ 0.09-0.10‰; Figure 3c ) to generate Fe-rich megacrystic olivine, this protolith must have Fe isotopes even heavier than these values. This Fe isotopic signature requires contribution by subducted, altered oceanic crust 37 , perhaps including isotopically heavy pyroxenites 38 , and provides an alternate interpretation for the origin of high d 56 Fe observed in metallic inclusions hosted by CLIPPIR diamonds. However, considering the scarcity of light oxygen isotope data in eclogite xenoliths, including only a few samples with low oxygen isotope values 39, 40, 41 , and almost complete lack of Fe isotope studies for these xenoliths (only five samples measured in one study 42 ), it is difficult to constrain the exact nature of these domains before their interaction with precursor kimberlite melts and conversion to megacrysts and other Fe-rich lithologies (e.g., dunites, wehrlites, websterites 43, 44, 45 ). Iron-rich CLIPPIR-diamond substrate accreted at the base of the lithosphere Above, we have shown that CLIPPIR diamonds are abundant in kimberlites that entrained a peculiar substrate in the deep lithosphere dominated by Fe-rich material with affinity to megacrysts. These domains feature isotopically light oxygen and isotopically heavy Fe, which suggests affinity to subducted, altered oceanic crust that was probably converted to eclogite. Although Fe-rich olivine related to megacrysts and sheared peridotites was generated by interaction with precursor kimberlite melts, it is unlikely that these isotopic features were imparted by precursor kimberlite melts because the convective mantle sources of kimberlites contain only small amounts of subducted material - based on trace element concentrations and Sr-Nd-Hf-Pb-C-O isotopic compositions 14, 28, 29, 46, 47, 48 . The compositional features of the megacrystic domain, including enrichment in Fe, Zn, Ti and other transition metals, light O isotopes and heavy Fe isotopes, were inherited from a subducted crustal substrate. The highly variable C isotope composition of CLIPPIR diamonds, including very low d 13 C (down to -27‰), but also elevated d 13 C of up to +4‰ 2, 3 and the high d 56 Fe (0.79-0.90‰) of their metallic inclusions Smith2021 are consistent with an origin in such a peculiar substrate. Unradiogenic Os isotopes and Ni contents in the metallic inclusions 6 coupled with the low-Cr compositions of included majorites that are intermediate between mafic and peridotitic protoliths 1 , however, suggest involvement of additional peridotitic components, perhaps in a subduction melange. It appears likely that CLIPPIR and sub-lithospheric diamonds with eclogitic affinity represent an integral, although sporadic component, of the Fe-rich isotopically anomalous substrate that was partly converted to Fe-rich megacrysts and other Fe-rich lithologies (e.g., dunites, wehrlites, websterites 43, 44, 45 ) during interaction with precursor kimberlite melts. The process of interaction between lithospheric wall rocks and precursor kimberlite melts, which generates megacrysts and sheared peridotites 19, 20, 27 , might explain several features of CLIPPIR diamonds including deformation, annealing and resorption. These features reflect a highly dynamic environment where CLIPPIR diamonds resided before being entrained and transported to the surface - such as lithospheric wall rocks lining kimberlite magmatic conduits. Here fluid-mediated deformation and recrystallisation are commonly inferred from textural studies of sheared peridotites and recrystallised megacrysts 49, 50 , including samples from CLIPPIR-bearing localities 24, 27 . This hypothesis is supported by the low diamond grades of typical CLIPPIR localities (e.g., 1.5 cpht for Letseng; 20 cpht for Karowe), which has been attributed to diamond destruction associated with extensive infiltrations of the deep lithosphere by failed pulses of kimberlite melt 15, 51 . In this scenario, the unusual abundance of large CLIPPIR diamonds compared to smaller diamonds in some of these localities, such as the Letseng mine, can be explained by near-complete resorption of the smaller diamonds 5 . The remaining question is how this CLIPPIR substrate was exhumed from the MTZ 1 and accreted at the base of the lithosphere. The composition of minerals included in other sub-lithospheric diamonds and the C and N isotope systematics of the diamond hosts, including examples from Monastery and other localities in southern and western Africa, Brazil, Canada and Australia, indicate that a large fraction of sub-lithospheric diamonds derive from oceanic crust that was subducted to the MTZ or deeper 52, 53, 54, 55 . These observations, combined with the almost ubiquitous (partial) retrogression of majorite 56 and the other high-pressure silicate minerals, suggests that accretion of deeply subducted oceanic crust to the bottom of the lithosphere is likely to occur in several localities. Some support to this conclusion comes from previous reports 57, 58 that eclogites are locally concentrated in the lower lithosphere sampled by some kimberlites. These localities include the Orapa kimberlite and Kaalvallei (proximal to the north Lesotho field), where we have documented deep lithosphere with anomalous oxygen and Fe isotope systematics. A possible accretion scenario entails diapiric rising of buoyant, subducted oceanic lithosphere including slivers of oceanic crust 8, 59 . While feasible, this model faces two potential challenges, namely the dominantly lherzolitic and metasomatically enriched nature of the lowermost cratonic lithosphere, where harzburgites and similarly depleted lithologies are not observed 57 , and the paucity of highly depleted and therefore buoyant harzburgite in subducted oceanic lithosphere. An alternative mechanism to accrete dense, Fe-rich eclogitic material of crustal derivation to the bottom of the lithosphere is active mantle upwelling associated with plumes from the core-mantle boundary (or shallower depths). Although early models 60 suggested negligible entrainment of ambient mantle material along the edges of sub-vertical plumes, recent seismic tomography models show substantial deflection and reorganisation of mantle plumes in the mid-mantle region (<1000 km) 61, 62 . Entrainment of crustal material residing in the MTZ by plumes from the lower mantle and/or transport in secondary plumes that are generated at these depths appears possible 63 and has been previously proposed to explain the H 2 O enrichment of some komatiites 64 . Kimberlites containing CLIPPIR and other sub-lithospheric diamonds in southern Africa were either centred above (Orapa field, Premier/Cullinan) or peripheral (northern Lesotho, including Monastery) to regions where major mantle plumes impinged the bottom of the lithosphere (Umkondo for Orapa field, Bushveld for Premier, Bushveld and Karoo for northern Lesotho 65 ). The geochemistry and Re-Os isotopes of mantle xenoliths from Premier coupled with seismic tomography of the Kaapvaal craton provide clear indication of accretion of Fe-rich lithosphere associated with the Bushveld plume at ~2.0 Ga 66, 67, 68, 69 . This process has been considered to be of global significance in growing continental lithosphere 70 as well as healing previously scarred cratons 71 . The lithosphere beneath all the major CLIPPIR-bearing localities of southern Africa experienced interaction with strongly buoyant plumes, which were able to transport and potentially accrete dense material to the base of the lithosphere. This mechanism might be responsible for the occurrence of abundant, isotopically anomalous, Fe-rich material of subducted crustal origin in the deep lithosphere at these localities, including CLIPPIR and other sub-lithospheric diamonds. The occurrence of plume-derived He isotopes in sub-lithospheric diamonds from Juina (Brazil) 72 points to a likely plume influence in the transport of sub-lithospheric diamonds from this region too. Finally, although the high temperatures associated with deep mantle plumes can be detrimental towards diamond preservation, previous thermal modelling 73 shows that diamond can survive transient heating events, which is confirmed by the occurrence of Archean diamonds in the lithosphere beneath Premier and the Orapa field 74 . Implications for diamond exploration and global mantle heterogeneities The association between the largest gem diamonds recovered to date and kimberlites that sampled Fe-rich domains with anomalous oxygen and Fe isotope compositions in the deep lithosphere supports a likely genetic relationship. This connection also enlightens an exploration pathfinder for the most precious diamonds where Fe enrichment in olivine and large populations of Fe-rich megacrysts can be employed to target kimberlites with high potential for CLIPPIR diamonds, consistent with suggestions in previous studies 12 . However, as customary for diamonds, this association is not exclusive. This is exemplified by the Karowe mine where erratic kimberlite sampling of the mantle resulted in CLIPPIR diamonds being found in just one of the three volcanic lobes 75 even though the underlying lithosphere is unlikely to be substantially different on such a small scale (<1 km 2 ). Similarly, this study identifies other kimberlites tapping anomalous Fe-rich deep mantle domains (e.g., Kaalvallei) where CLIPPIR diamonds have not yet been reported. The discovery of isotopically anomalous domains enriched in Fe, Zn, Ti and other transition metals bears important implications to understand the genesis of compositional heterogeneities in intraplate magmas globally. Magmatic olivines in kimberlites exhibit a wide range in Mg# where the lowermost values (83-84) have been explained by interaction with Fe-rich material related to precursor kimberlite melts 14 . However, considering the origin of kimberlite melts from unremarkable convective mantle sources (e.g., olivine Mg# ~89) 14 , the assimilated Fe-rich material cannot simply result from kimberlite metasomatism, but requires a pre-existing Fe-rich substrate. Such a substrate is coincident with the Fe-rich isotopically anomalous, presumably eclogitic domains envisaged in this work. Iron-rich domains at the base of the lithospheric mantle, in virtue of their lower melting degrees compared to refractory peridotites, can substantially contribute to the genesis of intraplate magmas. Their influence is clearly noticeable in kimberlites from southern Africa and other Fe-rich localities (e.g., Alto Paranaiba 16, 76 in Brazil which also feature extreme Sr-Nd-Hf isotopic compositions 47, 77 ), and might also contribute to isotopic anomalies of intraplate magmatic provinces elsewhere including the Emeishan large igneous province (LIP) and the extensive intraplate basaltic province of eastern China 78, 79, 80 . Methods Olivine geochemistry Olivine major and minor elements (except for Premier and Murowa) were analysed using a CAMECA SX-100 electron probe microanalyser (EPMA) housed at the University of Johannesburg, South Africa. Analyses were conducted using an acceleration potential of 15 kV, beam current of 20 nA, and beam diameter of 1 μm. Peak counting times varied between 10 s and 60 s depending on measured element. All elements were measured on the Kα line. Reference materials used to calibrate the instrument included jadeite (Na), olivine (Mg), almandine (Al), diopside (Si), wollastonite (Ca), rhodonite (Mn), hematite (Fe), and synthetic Cr, Ni, and Ti oxides. Detection limits are ~ 0.01 wt% for all elements. Data reduction and matrix correction was performed using the ‘X-PHI’ method. Major elements in the Premier and Murowa samples were analysed at Carnegie Institution for Science using a JEOL 8530F Field Emission EPMA, equipped with 5 wavelength dispersive spectrometers. The probe was operated at 15kV and 30nA, and a 1um diameter electron beam. The Probe for EPMA software was used to collect and reduce the data. The standards used were Olivine – San Calros (Mg, Si), Ilmenite (Ti, Fe, Mn), MgCr 2 O 4 (Cr), Anorthite (Al, Ca), and Ni Olivine (Ni). Trace elements analyses in all olivine (except for Premier and Murowa; see below) were performed using a 193 nm Resolution M50 LR Excimer laser ablation system attached to an Agilent 8800 quadrupole inductively coupled plasma mass spectrometer (LA-ICP-MS) at the Stellenbosch University, South Africa. Beam sizes of 100 and 70 µm were used, the former being generally the preferred choice. Background acquisition, ablation and washout times of 15, 35, and 25 seconds were employed, respectively. Laser repetition rate of 7 Hz and energy fluence of 4 J/ cm 2 were applied. Helium was used as the ablation gas and sample transported to the ICP-MS system using an argon gas carrier. Reference materials NIST (National Institute of Standards and Technology) 612 was used as calibration standards. Glass reference materials BCR-2G and BHVO-2G (values from GeoReM 82 ) and olivine 355OL 83 were used for quality control (see Dataset 1 ). Silicon and Mg measured were used as internal standards with nominal values obtained by EMPA. A typical analytical session consisted of 15-20 analyses of unknowns bracketed by analyses of the reference materials. Elements highly incompatible in olivine (including Rb, Sr, Ba, Zr, Nb, Y) were monitored to assess potential contamination by inclusions or material in fractures and/or grain boundaries. Data reduction was performed using the LADR v 1.1.07 software package from Norris Scientific. Trace elements measured in the 355OL olivine were reproduced to better than 90% of the solution values 83 . Trace element analyses for Premier and Murowa olivine were analysed at the Carnegie Institution for Science using the Teledyne Iridia laser connected to an iCAPq mass spectrometer. A beam size of 35 µm was used. A laser repetition rate of 5 Hz and energy fluence of 4 j/cm2 were applied. Background acquisition, ablation and washout times of 30, 40, and 30 seconds were employed, respectively. Argon gas was used with a gas flow rate of 0.9 l/min for transport to the mass spec. Reference materials 355OL and BHVO-2G were used. 355Ol calibration was used for all elements with the exception of Ca where BHVO-2G calibration was used. Silicon was used as an internal standard. Elements highly incompatible in olivine (including Rb, Sr, Ba, Zr, Nb, Y) were monitored to assess potential contamination by inclusions or material in fractures and/or grain boundaries. The composition of the olivine cores from the Orapa field (Karowe, Damtshaa; Fig S2 ), northern Lesotho field (Letseng, Mothae, Monastery; Fig S3 ), and Premier ( Fig S4 ) are presented here to show the range in olivine compositions observed at CLIPPIR-bearing localities. Overall, the olivine xenocrysts display a large range in Mg# from 93.5 to 80.9 and coherent trends between major and trace elements. Letseng, Mothae, and Premier, however, have far fewer olivine cores with Mg <84 relative to Karowe and Damtshaa, indicating that megacrysts at Letseng, Mothae, and Premier rarely evolved to the extremely Fe-rich compositions observed in the Orapa field. This is in contrast to the Monastery kimberlite that has a large population of Fe-rich megacrysts 25, 84 , indicating local variations in the deep lithosphere composition. The mean olivine core composition for each location represents the nature of the lithospheric mantle sampled by the kimberlites ( Fig. S5 ). The olivines are broadly divided into Mg-rich (Mg# >89) and Fe-rich (Mg# 90 (e.g., Ekati, Finsch, Murowa), the kimberlite has predominantly sampled refractory mantle with little to no Fe-rich megacrystic material. In contrast, kimberlites that have low mean olivine core (Mg# <90) sample a high proportion of Fe-rich (Mg# <89) megacrystic olivine 14 . Kimberlites that contain high proportions of Fe-rich megacrystic cores and hence low average Mg# for the cores exhibit low Mg# in their magmatic rims ( Fig. S1 ). This correlation has been previously attributed to the important role of lithospheric mantle assimilation on the composition of kimberlite melts 14 , of which olivine rim Mg# represents an excellent proxy. Thus, the Mg# of mean olivine rims and mean olivine cores can be used interchangeably to approximate the composition of kimberlite magmas and traversed lithospheric mantle. The Mg-rich olivine have constant Ni (~3000 ppm) over their range in Mg# (Mg# 89-93.5) for Karowe and Damtshaa but extend to high Ni values (~3500 ppm) for the most Mg-rich olivine at Letseng, Mothae, and Monastery. For all CLIPPIR-bearing kimberlites analysed here, the Ni concentrations decrease with decreasing Mg# for the Fe-rich cores (i.e., Mg# <89) to values as low as ~1000 ppm due to olivine fractionation 84 and increasing interaction with Ni-poor, Fe-rich substrate during evolution of the megacryst parent melt. Zinc concentrations across the range of olivine core Mg# display a distinct negative correlation. The lowest Zn values are observed for highly refractory, likely harzburgitic or dunitic, olivine cores (Mg# >93) whereas lherzolitic (Mg# 90-93) olivine cores have higher zinc, indicating that Zn concentrations, like Fe, increases during refertilization processes in the SCLM. Zinc in the megacrystic olivine cores behaves as an incompatible element and increasing with decreasing Mg#, consistent with observation in previous studies 16 . In addition, Al and Ca concentrations display a sharp decrease in the early stages of megacryst fractionation (i.e., Mg# 86-89), consistent with previous studies that have suggested concurrent crystallisation of olivine with pyroxene and garnet 84 . A notable observation is the abundant presence of low-Al olivine in equilibrium with spinel-garnet peridotite from the Damtshaa kimberlite relative to other kimberlites, indicating abundant shallow sampling outside of diamond stability. In addition, Damtshaa contains a high abundances of unusual low-Al megacrysts ( Figure 3 and Fig. S5d ), which are also observed at Karowe but in substantially lower abundances. These low-Al megacrysts are interpreted to represent low-temperature, relatively shallow formation compared to the high-Al megacrysts predominant in the northern Lesotho and Karowe kimberlites. The Premier kimberlite is unusual in that it does not sample a significant population of typical granular garnet peridotite olivine ( Fig S4. ) and 85% of the total olivine population is comprised of olivine from Al-rich (i.e., high temperature) megacrystic and sheared peridotite lithologies. This is consistent with the unusual abundance of high temperature sheared peridotite xenoliths at Premier kimberlite 68 . Olivine thermobarometry The Al-in-olivine thermometer 23 has been calibrated for garnet peridotite mantle xenoliths with olivine Mg# >89 and, thus, its application requires screening olivine compositions outside of the calibration (e.g., Fe-rich megacrysts, olivine from spinel peridotites). This screening was done using the following protocol, which is based on the composition of olivine in typical, texturally equilibrated peridotites 23, 85 : Mg# >89; Ni >2300 ppm (or NiO >0.3 wt.%); Mn <1160 ppm (or MnO <0.15 wt.%); Ca <715 ppm (or CaO <0.1 wt.%). Grains that satisfied these criteria were then further screened for garnet vs spinel peridotite origins using Al-V 23 and Al vs Mn 87 relationships. Iterative calculations were then undertaken on olivine in equilibrium with garnet to match the Al-in-olivine temperature with local geothermal gradients for each kimberlite and using 40 mW/m 2 for the Orapa field 88 , 41 mW/m 2 for the northern Lesotho kimberlites 89 , and 39 mW/m 2 for Premier 90 . The depth estimates are plotted as histograms in Fig. S7 and kernel density estimate (KDE) curves in Figure 2 . The Al-in-olivine temperatures calculated in this study are generally >900 o C and outside of the low temperature range where some overestimation of temperatures have been suggested to occur 23 . Thus, this minor calibration issue does not affect the calculated sampling depths in this study. The major limitation of the Al-in-olivine thermometer to reconstruct olivine entrainment depths is in the selection of a local geotherm. For example, the selection of slightly different geothermal gradients in this study resulted in the minor differences in the main peaks of the resultant KDE curves, namely 145-155 km in the Orapa field and 130-145 km in northern Lesotho. These main peaks would overlap completely if the same geotherm was selected. However, the main modes of mantle sampling in the KDE curves of Figure 2 would not change the stark contrast in sampling depths between these CLIPPIR-bearing kimberlites and CLIPPIR-poor kimberlites. While the Al-in-olivine thermometer calibration does not extend to the more Fe-rich range of megacrystic olivine, the Al concentrations in the megacrystic olivine are used here as a qualitative indicator of the likely temperature association of these Fe-rich olivines. The mantle-derived Fe-rich olivines analysed in this study display a positive correlation between Mg# and Al concentrations with the highest Al concentrations observed in the most primitive olivine megacrysts (Mg# 86-89) ( Fig. S2; S3 ), consistent with previous studies 51 . The decrease in Al concentrations with increasing Fe likely reflects a combination of decreasing temperature during crystallisation, interaction with lithospheric mantle, and concurrent garnet co-crystallisation 25, 84 . The Al-in-olivine temperatures for the most primitive (i.e., Mg# = 86-89), Al-rich (i.e., 100-380 ppm) megacrystic olivine are between 1200-1500 o C and, while just qualitative, imply thermal perturbation associated with megacryst crystallisation in the deep lithosphere. Previous studies 26,91 have similarly showed formation of Fe-rich megacrysts and sheared peridotites at elevated temperatures toward the base of the lithosphere. Fe isotopes Iron isotopic analyses were conducted at the State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences (Beijing)-CUGB. Sample powders containing ≥ 250 μg of Fe were dissolved in a mixture of concentrated HF-HNO 3 . After optical check, precipitate-free solutions were evaporated at 160 °C and re-dissolved in aqua regia to remove fluorides and completely oxidize Fe 2+ . The resultant solutions were dried at 90 °C and re-fluxed with concentrated HCl. The samples were finally dissolved in 0.5 mL of 6 N HCl before column chromatography. The solutions were purified twice using AG1-X8 resin to completely remove interference elements. Iron isotopic analyses were conducted on a ThermoFisher Neptune Plus MC-ICP MS using a 57 Fe- 58 Fe double spike technique. The complete analytical procedure is described elsewhere 91 . All Fe isotope data are reported in delta notation against the international standard IRMM-014: δ 56 Fe = [( 56 Fe/ 54 Fe) sample /( 56 Fe/ 54 Fe) IRMM-014 - 1] * 1000 (‰) The reference materials analyzed at CUGB yielded results (BHVO-2: 0.118±0.013‰; BCR-2: 0.093±0.016‰) identical within errors of published values 92, 93 . The whole procedure blank was less than 15 ng. The analytical errors of δ 56 Fe are given as two standard errors (2SE). Bulk-rock d 56 Fe values do not directly provide the composition of kimberlite melts because kimberlites are mixtures of magmatic and mantle-derived xenocrystic components, dominantly olivine. Variation in olivine abundances generate variable MgO contents in bulk kimberlite samples worldwide 48, 94 . Bulk-kimberlite MgO contents hence reflect both the amount and composition of olivine. Assuming primary kimberlite melts derive from similar convective mantle sources globally 14 and, similar to the isotopes of another major element like oxygen 28, 29 , have a relatively restricted range of Fe isotopes, the inverse correlation between bulk-rock d 56 Fe and MgO in Fig. S8 reflects the variable contribution of xenocrystic olivine with variable MgO contents and Fe isotope composition. We estimate the average d 56 Fe of entrained Fe-rich (Mg# <89) olivine xenocrysts in the three samples (Karowe AK-6S, Letseng KB14-01, Wesselton WA-1) with elevated Fe isotopes for which olivine Mg# data are available 9, 86, 95 . The goal is to estimate the Fe isotopic composition of the Fe-rich deep lithosphere which might represent the source of CLIPPIR diamonds (at least at Karowe and Letseng). For these calculations we assume the following relationship: d 56 Fe bulk = F melt d 56 Fe melt + F Mg-Ol d 56 Fe Mg-Ol + F Fe-Ol d 56 Fe Fe-Ol where F is the fraction of each component and melt , Mg-Ol and Fe-Ol refer to magmatic components, Mg-rich xenocrystic olivine (Mg# >89) and Fe-rich xenocrystic olivine (Mg# <89), respectively. The magmatic olivine rims are included in the melt fraction, whereas the abundance of other xenocrystic material is considered negligible, in line with petrographic observations of <5 vol% of garnet, pyroxene, ilmenite, mica and crustal xenocrysts. The fraction of total olivine (F Ol ) is provided by modal analyses adjusted for the density difference between olivine and bulk kimberlite. It follows that: F melt = 1 – F Ol F Mg-Ol = F Ol X Mg-Ol F Fe-Ol = F Ol X Fe-Ol where X Mg-Ol and X Fe-Ol are the relative abundances of Mg-rich and Fe-rich olivine based on published EPMA analyses. Combining the previous equations and resolving for the Fe isotope composition of Fe-rich olivine, we obtain the following expression: d 56 Fe Fe-Ol = d 56 Fe bulk – (1 – F Ol d 56 Fe melt ) + (F Ol X Fe-Ol ) d 56 Fe Mg-Ol )] / (F Ol X Fe-Ol ) where d 56 Fe Mg-Ol = 0.030‰ based on the typical composition of olivine in refractory mantle peridotites 34, 35, 36 , and d 56 Fe melt = 0.092‰ based on the correlation between olivine rim Mg# and bulk-rock d 56 Fe ( Figure 3c ) and assuming primary olivine Mg# of ~89. The role of serpentinisation is not addressed, but considering the limited alteration of these samples, is considered to have a minor effect in the final calculations. The results of these calculations are tabulated in Dataset 3 and show that the average d 56 Fe of Fe-rich olivine from Karowe is 0.11‰ while those for Letseng and Wesselton are substantially higher (0.22‰ and 0.30‰). However, Fe-rich olivine predominantly derives from interaction between early kimberlite liquids in equilibrium with olivine Mg ~89 and an Fe-rich substrate, and olivine preferentially partition the light isotopes of Fe compared to coexisting silicate minerals 34, 35 . Therefore, it appears likely that the Fe-rich deep lithosphere before kimberlite metasomatism had even heavier although unconstrained Fe isotope composition. It is possible that this lithospheric substrate locally approached the very high d 56 Fe (0.79-0.90‰) observed in the Fe-rich metallic inclusions hosted by CLIPPIR diamonds 6 , an hypothesis which requires further scrutiny using natural samples including megacrysts and eclogite xenoliths. Declarations Acknowledgments Most samples for this study were source from the John J. Gurney Mantle collection housed at the University of Cape Town with additional samples collected by AG with thanks to Petra diamonds and Rio Tinto. GHH acknowledges funding from the DSI-NRF Centre of Excellence (CoE) for Integrated Mineral and Energy Resource Analysis (DSI-NRF CIMERA) towards this research is hereby acknowledged. Opinions expressed and conclusions arrived at, are those of the author(s) and are not necessarily to be attributed to the CoE. Christian Reinke is thanked for help with microprobe analyses at the University of Johannesburg. Riana Rossouw is thanked for help with laser ablation analyses at the University of Stellenbosch. Discussion with several colleagues including Peng Ni, Graham Pearson, Steve Shirey, Evan Smith, Thomas Stachel, Suzette Timmermann, Mike Walter and Qiwei Zhang substantially improved the contents of this work even though they do not necessarily reflect the opinions of these colleagues. Author contributions G.H.H and A.G. conceptualized the project and wrote the original draft. G.H.H., A.G., and M.M.T. collected EPMA and laser ablation data for olivine. J.L. and R.C. were responsible for Fe isotope data collection. G.H.H. drafted all figures. G.H.H. and M.M.T. compiled online datasets. All authors contributed to the writing, editing, and reviewing the manuscript drafts. Competing interests . The authors declare no competing interests. References Smith, E.M., Shirey, S.B., Nestola, F., Bullock, E.S., Wang, J., Richardson, S.H., Wang, W., 2016. Large gem diamonds from metallic liquid in Earth’s deep mantle. Science 354(6318), 1403-1405. Milledge, H.J., Mendelssohn, M.J., Seal, M., Rouse, J.E., Swart, P.K., Pillinger, C.T., 1983. Carbon isotopic variation in spectral type II diamonds. Nature 303(5920), 791-792. doi.org/10.1038/303791a0. Banas, A., Stachel, T., Stern, R.A., Allan, A., Freeman, L., 2017. Can microdiamonds be used to predict the distribution of large Type IIa macrodiamonds? A case study at the Letseng mine. International Kimberlite Conference: Extended Abstracts 11. doi.org/10.29173/ikc4017. Moore, A., Helmstaedt, H., 2023. Origin of framesite revisited: Possible implications for the formation of CLIPPIR diamonds. Earth-Science Reviews 241, 104434. doi.org/https://doi.org/10.1016/j.earscirev.2023.104434. Bowen, D.C., Ferraris, R.D., Palmer, C.E., Ward, J.D., 2009. On the unusual characteristics of the diamonds from Letšeng-la-Terae kimberlites, Lesotho. Lithos 112, 767-774. doi.org/https://doi.org/10.1016/j.lithos.2009.04.026. Smith, E.M., Ni, P., Shirey, S.B., Richardson, S.H., Wang, W., Shahar, A., 2021. Heavy iron in large gem diamonds traces deep subduction of serpentinized ocean floor. Science Advances 7(14), eabe9773. doi.org/10.1126/sciadv.abe9773. Stachel, T., Cartigny, P., Chacko, T., & Pearson, D. G. 2022. Carbon and nitrogen in mantle-derived diamonds. Reviews in Mineralogy and Geochemistry , 88 (1), 809-875. Timmerman, S., Stachel, T., Koornneef, J.M., Smit, K.V., Harlou, R., Nowell, G.M., Thomson, A.R., Kohn, S.C., Davies, J.H.F.L., Davies, G.R., Krebs, M.Y., Zhang, Q., Milne, S.E.M., Harris, J.W., Kaminsky, F., Zedgenizov, D., Bulanova, G., Smith, C.B., Cabral Neto, I., Silveira, F.V., Burnham, A.D., Nestola, F., Shirey, S.B., Walter, M.J., Steele, A., Pearson, D.G., 2023. Sublithospheric diamond ages and the supercontinent cycle. Nature 623(7988), 752-756. doi.org/10.1038/s41586-023-06662-9. Giuliani, A., Schmidt, M.W., Torsvik, T.H., Fedortchouk, Y., 2023. Genesis and evolution of kimberlites. Nature Reviews Earth & Environment 4(11), 738-753. doi.org/10.1038/s43017-023-00481-2. Moore, A.E., 2009. Type II diamonds: Flamboyant Megacrysts? South African Journal of Geology 112(1), 23-38. doi.org/10.2113/gssajg.112.1.23. Solomatov, V.S., Reese, C.C., 2008. Grain size variations in the Earth's mantle and the evolution of primordial chemical heterogeneities. Journal of Geophysical Research: Solid Earth 113(B7). doi.org/https://doi.org/10.1029/2007JB005319. Zhang, Q., Hardman, M. F., Stachel, T., Chinn, I., Seller, M., Kjarsgaard, B., & Pearson, D. G. (2025). A machine learning approach to single garnet geothermometry and application to tracing the fingerprint of superdeep diamonds. Geochemistry, Geophysics, Geosystems , 26 (3), e2024GC012124. Zhang, Q., Malakhova, I., Stachel, T., Elliott, B., Chinn, I., Seller, M., ... & Pearson, D. G. (2025). A link between enhanced sampling of Ti–rich mantle garnets and superdeep diamonds: Insights from the DO-27 kimberlite, Northern Canada. Mineralogy and Petrology , 1-17. Giuliani, A., Pearson, D.G., Soltys, A., Dalton, H., Phillips, D., Foley, S.F., Lim, E., Goemann, K., Griffin, W.L., Mitchell, R.H., 2020. Kimberlite genesis from a common carbonate-rich primary melt modified by lithospheric mantle assimilation. Science Advances 6(17), eaaz0424. doi.org/10.1126/sciadv.aaz0424. Giuliani, A., Phillips, D., Pearson, D.G., Sarkar, S., Müller, A.A., Weiss, Y., Preston, R., Seller, M., Spetsius, Z., 2023. Diamond preservation in the lithospheric mantle recorded by olivine in kimberlites. Nature Communications 14(1), 6999. doi.org/10.1038/s41467-023-42888-x. Howarth, G.H., Giuliani, A., Soltys, A., Bussweiler, Y., 2022. Compositional Variations in Primitive Kimberlite Melts and Entrained Mantle Cargo from a Global Survey of Trace Element Compositions in Kimberlite Olivine. Journal of Petrology 63(8), egac062. doi.org/10.1093/petrology/egac062. Nowell, G.M., Pearson, D.G., Bell, D.R., Carlson, R.W., Smith, C.B., Kempton, P.D., Noble, S.R., 2004. Hf Isotope Systematics of Kimberlites and their Megacrysts: New Constraints on their Source Regions. Journal of Petrology 45(8), 1583-1612. Woodhead, J., Hergt, J., Giuliani, A., Phillips, D., Maas, R., 2017. Tracking continental-scale modification of the Earth’s mantle using zircon megacrysts. Geochemical Perspectives Letters 4, 1-6. doi.org/http://dx.doi.org/10.7185/geochemlet.1727. Moore, A., Belousova, E., 2005. Crystallization of Cr-poor and Cr-rich megacryst suites from the host kimberlite magma: implications for mantle structure and the generation of kimberlite magmas. Contrib Mineral Petrol 149(4), 462-481. Nkere, B.J., Janney, P.E., Tinguely, C., 2021. Cr-poor and Cr-rich clinopyroxene and garnet megacrysts from southern African Group 1 and Group 2 kimberlites: Clues to megacryst origins and their relationship to kimberlites. Lithos 396-397, 106231. doi.org/https://doi.org/10.1016/j.lithos.2021.106231. Veglio, C., Lawley, C.J.M., Pearson, D.G., Kjarsgaard, B.A., Petts, D.C., Jackson, S.E., 2022. Olivine xenocrysts reveal carbonated mid-lithosphere in the northern Slave craton. Lithos 414-415, 106633. doi.org/https://doi.org/10.1016/j.lithos.2022.106633. Tau, M. M., Howarth, G. H., & Giuliani, A. (2024, July). Lithospheric mantle sampling of kimberlites and lamproites using Al-in-olivine thermometry. In International Kimberlite Conference: Extended Abstracts (Vol. 12). Bussweiler, Y., Brey, G.P., Pearson, D.G., Stachel, T., Stern, R.A., Hardman, M.F., Kjarsgaard, B.A., Jackson, S.E., 2017. The aluminum-in-olivine thermometer for mantle peridotites — Experimental versus empirical calibration and potential applications. Lithos 272–273, 301-314. doi.org/http://dx.doi.org/10.1016/j.lithos.2016.12.015. Boyd, F.R., Nixon, P.H., 1975. Origins of the ultramafic nodules from some kimberlites of northern Lesotho and the Monastery Mine, South Africa. Physics and Chemistry of the Earth 9, 431-454. doi.org/http://dx.doi.org/10.1016/0079-1946(75)90032-4. Gurney, J.J., Jakob, W.R.O., Dawson, J.B., 1979. Megacrysts from the Monastery kimberlite pipe, in: Boyd, F.R., Meyer, H.O.A. (Eds.), The mantle sample. 2nd International Kimberlite Conference. American Geophysical Union, Washington, DC, pp. 227-243. Tappe, S., Massuyeau, M., Smart, K.A., Woodland, A.B., Gussone, N., Milne, S., Stracke, A., 2021. Sheared Peridotite and Megacryst Formation Beneath the Kaapvaal Craton: a Snapshot of Tectonomagmatic Processes across the Lithosphere–Asthenosphere Transition. Journal of Petrology 62(8). doi.org/10.1093/petrology/egab046. Heckel, C., Woodland, A.B., Linckens, J., Gibson, S.A., Seitz, H.-M., 2023. Sheared Peridotites from Northern Lesotho: Metasomatism-Induced Deformation and Craton Destabilization. Journal of Petrology 64(10), egad076. doi.org/10.1093/petrology/egad076. Giuliani, A., Martin, L.A.J., Soltys, A., Griffin, W.L., 2019. Mantle-like oxygen isotopes in kimberlites determined by in situ SIMS analyses of zoned olivine. Geochimica et Cosmochimica Acta 266, 274-291. doi.org/https://doi.org/10.1016/j.gca.2019.03.032. Xu, J.-Y., Giuliani, A., Li, Q.-L., Lu, K., Melgarejo, J.C., Griffin, W.L., 2021. Light oxygen isotopes in mantle-derived magmas reflect assimilation of sub-continental lithospheric mantle material. Nature Communications 12(1), 6295. doi.org/10.1038/s41467-021-26668-z. van Blerk, J., Howarth, G.H., Harris, C., Janney, P.E., 2025. Oxygen isotope constraints on proto-kimberlite melt modification through assimilation of low δ18O recycled crust in the deep lithosphere. Geochimica et Cosmochimica Acta. doi.org/https://doi.org/10.1016/j.gca.2024.12.028. Mattey, D., Lowry, D., Macpherson, C., 1994. Oxygen isotope composition of mantle peridotite. Earth and Planetary Science Letters 128(3‚Äì4), 231-241. doi.org/http://dx.doi.org/10.1016/0012-821X(94)90147-3. Taylor, H.P., Sheppard, S.M.F., 1986. Igneous rocks I: processes of isotopic fractionation and isotope systematics. Reviews in Mineralogy 16, 227-272. Wenner, D.B., Taylor Jr, H.P., 1974. D/H and O18/O16 studies of serpentinization of ultramaflc rocks. Geochimica et Cosmochimica Acta 38(8), 1255-1286. doi.org/http://dx.doi.org/10.1016/0016-7037(74)90120-3. Sossi, P.A., Nebel, O., Foden, J., 2016. Iron isotope systematics in planetary reservoirs. Earth and Planetary Science Letters 452, 295-308. doi.org/http://dx.doi.org/10.1016/j.epsl.2016.07.032. Huang, F., Zhang, Z., Lundstrom, C.C., Zhi, X., 2011. Iron and magnesium isotopic compositions of peridotite xenoliths from Eastern China. Geochimica et Cosmochimica Acta 75(12), 3318-3334. doi.org/https://doi.org/10.1016/j.gca.2011.03.036. An, Y., Huang, J.-X., Griffin, W.L., Liu, C., Huang, F., 2017. Isotopic composition of Mg and Fe in garnet peridotites from the Kaapvaal and Siberian cratons. Geochimica et Cosmochimica Acta 200, 167-185. doi.org/http://dx.doi.org/10.1016/j.gca.2016.11.041. Rouxel, O., Dobbek, N., Ludden, J., Fouquet, Y., 2003. Iron isotope fractionation during oceanic crust alteration. Chemical Geology 202(1), 155-182. doi.org/https://doi.org/10.1016/j.chemgeo.2003.08.011. Williams, H.M., Bizimis, M., 2014. Iron isotope tracing of mantle heterogeneity within the source regions of oceanic basalts. Earth and Planetary Science Letters 404, 396-407. doi.org/https://doi.org/10.1016/j.epsl.2014.07.033. Radu, I.-B., Harris, C., Moine, B.N., Costin, G., Cottin, J.-Y., 2019. Subduction relics in the subcontinental lithospheric mantle evidence from variation in the δ18O value of eclogite xenoliths from the Kaapvaal craton. Contrib Mineral Petrol 174(3), 19. doi.org/10.1007/s00410-019-1552-z. Shu, Q., Brey, G.P., Hoefer, H.E., Zhao, Z., Pearson, D.G., 2016. Kyanite/corundum eclogites from the Kaapvaal Craton: subducted troctolites and layered gabbros from the Mid- to Early Archean. Contrib Mineral Petrol 171(2), 1-24. doi.org/10.1007/s00410-015-1225-5. Korolev, N.M., Melnik, A.E., Li, X.-H., Skublov, S.G., 2018. The oxygen isotope composition of mantle eclogites as a proxy of their origin and evolution: A review. Earth-Science Reviews 185, 288-300. doi.org/https://doi.org/10.1016/j.earscirev.2018.06.007. Williams, H.M., Nielsen, S.G., Renac, C., Griffin, W.L., O'Reilly, S.Y., McCammon, C.A., Pearson, N., Viljoen, F., Alt, J.C., Halliday, A.N., 2009. Fractionation of oxygen and iron isotopes by partial melting processes: Implications for the interpretation of stable isotope signatures in mafic rocks. Earth and Planetary Science Letters 283(1), 156-166. doi.org/https://doi.org/10.1016/j.epsl.2009.04.011. Rehfeldt, T., Foley, S.F., Jacob, D.E., Carlson, R.W., Lowry, D., 2008. Contrasting types of metasomatism in dunite, wehrlite and websterite xenoliths from Kimberley, South Africa. Geochimica et Cosmochimica Acta 72(23), 5722-5756. Rehfeldt, T., Jacob, D.E., Carlson, R.W., Foley, S.F., 2007. Fe-rich Dunite Xenoliths from South African Kimberlites: Cumulates from Karoo Flood Basalts. Journal of Petrology 48(7), 1387-1409. doi.org/ 10.1093/petrology/egm023 . Arndt, N., Cordier, C., Boullier, A.-M., Batanova, V., Magnin, V., Findling, N., 2022. Olivine in Kimberlites and Related Rocks – Macrocrysts, Dunitic Nodules and Megacrysts from the Same Metasomatized Source. Journal of Petrology 63(8), egac063. doi.org/10.1093/petrology/egac063. Smith, C.B., 1983. Pb, Sr and Nd isotopic evidence for sources of southern African Cretaceous kimberlites. Nature 304, 51-54. Woodhead, J., Hergt, J., Giuliani, A., Maas, R., Phillips, D., Pearson, D.G., Nowell, G., 2019. Kimberlites reveal 2.5-billion-year evolution of a deep, isolated mantle reservoir. Nature 573(7775), 578-581. doi.org/10.1038/s41586-019-1574-8. Giuliani, A., Dalton, H., Pearson, D.G., 2025. Kimberlites: The deepest geochemical probes of Earth, in: Anbar, A., Weis, D. (Eds.), Treatise on Geochemistry (Third edition). Elsevier, Oxford, pp. 159-230. doi.org/https://doi.org/10.1016/B978-0-323-99762-1.00064-4. Mercier, J.-C.C., 1979. Peridotite Xenoliths and the Dynamics of Kimberlite Intrusion, The Mantle Sample: Inclusion in Kimberlites and Other Volcanics. American Geophysical Union, pp. 197-212. doi.org/10.1029/SP016p0197. Heckel, C., Woodland, A. B., & Gibson, S. A. (2025). Cretaceous thinning of the Kaapvaal craton and diamond resorption: Key insights from a highly-deformed and metasomatized ilmenite-dunite xenolith. Lithos , 494 , 107901. Howarth, G.H., Giuliani, A., Bussweiler, Y., Tau, M., Hashibi, S., Janney, P.E., Nowicki, T.E., 2024. Kimberlite pre-conditioning of the lithospheric mantle and implications for diamond survival: a case study of olivine and mantle xenocrysts from the Koidu mine (Sierra Leone). Mineralium Deposita. doi.org/10.1007/s00126-024-01312-0. Tappert, R., Stachel, T., Harris, J.W., Muehlenbachs, K., Ludwig, T., Brey, G.P., 2005. Subducting oceanic crust: The source of deep diamonds. Geology 33(7), 565-568. Palot, M., Pearson, D.G., Stachel, T., Stern, R.A., Le Pioufle, A., Gurney, J.J., Harris, J.W., 2017. The transition zone as a host for recycled volatiles: Evidence from nitrogen and carbon isotopes in ultra-deep diamonds from Monastery and Jagersfontein (South Africa). Chemical Geology 466, 733-749. doi.org/https://doi.org/10.1016/j.chemgeo.2017.07.023. Nestola, F., Korolev, N., Kopylova, M., Rotiroti, N., Pearson, D.G., Pamato, M.G., Alvaro, M., Peruzzo, L., Gurney, J.J., Moore, A.E., Davidson, J., 2018. CaSiO3 perovskite in diamond indicates the recycling of oceanic crust into the lower mantle. Nature 555, 237-240. doi.org/10.1038/nature25972. Shirey, S.B., Pearson, D.G., Stachel, T., Walter, M.J., 2024. Sublithospheric Diamonds: Plate Tectonics from Earth’s Deepest Mantle Samples. Annual Review of Earth and Planetary Sciences 52(Volume 52, 2024), 249-293. doi.org/https://doi.org/10.1146/annurev-earth-032320-105438. Harte, B., Cayzer, N., 2007. Decompression and unmixing of crystals included in diamonds from the mantle transition zone. Physics and Chemistry of Minerals 34(9), 647-656. doi.org/10.1007/s00269-007-0178-2. Griffin, W.L., O'Reilly, S.Y., 2007. Cratonic lithospheric mantle: Is anything subducted? Episodes 30(1), 43-53. Hardman, M.F., Stachel, T., Pearson, D.G., Stern, R.A., Lai, M.Y., 2025. Constraints on the distribution, composition, and lithology of eclogite and pyroxenite from the central Slave craton. Mineralogy and Petrology. doi.org/10.1007/s00710-024-00884-z. Perchuk, A. L., Gerya, T. V., Zakharov, V. S., & Griffin, W. L. (2020). Building cratonic keels in Precambrian plate tectonics. Nature , 586 (7829), 395-401. Farnetani, C.G., Legras, B., Tackley, P.J., 2002. Mixing and deformations in mantle plumes. Earth and Planetary Science Letters 196(1–2), 1-15. doi.org/http://dx.doi.org/10.1016/S0012-821X(01)00597-0. Dongmo Wamba, M., Montagner, J.-P., Romanowicz, B., 2023. Imaging deep-mantle plumbing beneath La Réunion and Comores hot spots: Vertical plume conduits and horizontal ponding zones. Science Advances 9(4), eade3723. doi.org/10.1126/sciadv.ade3723. Munch, F.D., Romanowicz, B., Mukhopadhyay, S., Rudolph, M.L., 2024. Deep mantle plumes feeding periodic alignments of asthenospheric fingers beneath the central and southern Atlantic Ocean. Proceedings of the National Academy of Sciences 121(46), e2407543121. doi.org/10.1073/pnas.2407543121. Doucet, L. S., Li, Z. X., & Gamal El Dien, H. (2021). Oceanic and super-deep continental diamonds share a transition zone origin and mantle plume transportation. Scientific Reports , 11 (1), 16958. Sobolev, A.V., Asafov, E.V., Gurenko, A.A., Arndt, N.T., Batanova, V.G., Portnyagin, M.V., Garbe-Schönberg, D., Krasheninnikov, S.P., 2016. Komatiites reveal a hydrous Archaean deep-mantle reservoir. Nature 531(7596), 628-632. doi.org/10.1038/nature17152 Ashwal, L.D., 2021. Sub-lithospheric mantle sources for overlapping southern African Large Igneous Provinces. South African Journal of Geology 124(2), 421-442. doi.org/10.25131/sajg.124.0023. Hoal, K.O., 2003. Samples of Proterozoic iron-enriched mantle from the Premier kimberlite. Lithos 71(2-4), 259-272. Fouch, M. J., Silver, P. G., Bell, D. R., & Lee, J. N. (2004). Small-scale variations in seismic anisotropy near Kimberley, South Africa. Geophysical Journal International , 157 (2), 764-774. Zhang, Q., Morel, M.L.A., Liu, J., Legros, H., Luguet, A., Viljoen, K.S., Davies, G.R., Pearson, D.G., 2022. Re-healing cratonic mantle lithosphere after the world's largest igneous intrusion: Constraints from peridotites erupted by the Premier kimberlite, South Africa. Earth and Planetary Science Letters 598, 117838. doi.org/https://doi.org/10.1016/j.epsl.2022.117838. Linzmeyer, T., Day, J.M.D., Alonso-Perez, R., Phahla, T., Wolmarans, A., Rogers, A., 2024. A 187Re-187Os and highly siderophile element study of diamondiferous kimberlite melt-mantle interactions and the inferred age of continental lithosphere. Geochimica et Cosmochimica Acta 386, 157-172. doi.org/https://doi.org/10.1016/j.gca.2024.09.026. Aulbach, S., 2012. Craton nucleation and formation of thick lithospheric roots. Lithos 149(0), 16-30. Liu, J., Pearson, D.G., Wang, L.H., Mather, K.A., Kjarsgaard, B.A., Schaeffer, A.J., Irvine, G.J., Kopylova, M.G., Armstrong, J.P., 2021. Plume-driven recratonization of deep continental lithospheric mantle. Nature 592(7856), 732-736. doi.org/10.1038/s41586-021-03395-5. Timmerman, S., Honda, M., Burnham, A.D., Amelin, Y., Woodland, S., Pearson, D.G., Jaques, A.L., Le Losq, C., Bennett, V.C., Bulanova, G.P., Smith, C.B., Harris, J.W., Tohver, E., 2019. Primordial and recycled helium isotope signatures in the mantle transition zone. Science 365(6454), 692. doi.org/10.1126/science.aax5293. Ernst, R.E., Davies, D.R., Jowitt, S.M., Campbell, I.H., 2018. When do mantle plumes destroy diamonds? Earth and Planetary Science Letters 502, 244-252. doi.org/https://doi.org/10.1016/j.epsl.2018.08.058. Howell, D., Stachel, T., Stern, R.A., Pearson, D.G., Nestola, F., Hardman, M.F., Harris, J.W., Jaques, A.L., Shirey, S.B., Cartigny, P., Smit, K.V., Aulbach, S., Brenker, F.E., Jacob, D.E., Thomassot, E., Walter, M.J., Navon, O., 2020. Deep carbon through time: Earth’s diamond record and its implications for carbon cycling and fluid speciation in the mantle. Geochimica et Cosmochimica Acta 275, 99-122. doi.org/https://doi.org/10.1016/j.gca.2020.02.011. Chinn, I., Krug, M., Minnie, W., Rikhotso, C., 2008. Relationships between diamond populations and geology from the AK06 kimberlite, Botswana. International Kimberlite Conference: Extended Abstracts 9(0). doi.org/10.29173/ikc3486. Lim, E., Giuliani, A., Phillips, D., Goemann, K., 2018. Origin of complex zoning in olivine from diverse, diamondiferous kimberlites and tectonic settings: Ekati (Canada), Alto Paranaiba (Brazil) and Kaalvallei (South Africa). Mineralogy and Petrology 112(2), 539-554. doi.org/10.1007/s00710-018-0607-6. Araujo, A., Carlson, R.W., Gaspar, J.C., Bizzi, L.A., 2001. Petrology of kamafugites and kimberlites from the Alto Paranaíba Alkaline Province, Minas Gerais, Brazil. Contrib Mineral Petrol 142(2), 163-177. doi.org/10.1007/s004100100280. Wang, X.-C., Wilde, S.A., Li, Q.-L., Yang, Y.-N., 2015. Continental flood basalts derived from the hydrous mantle transition zone. Nature Communications 6, 7700. doi.org/10.1038/ncomms8700. He, Y., Meng, X., Ke, S., Wu, H., Zhu, C., Teng, F.-Z., Hoefs, J., Huang, J., Yang, W., Xu, L., Hou, Z., Ren, Z.-Y., Li, S., 2019. A nephelinitic component with unusual δ56Fe in Cenozoic basalts from eastern China and its implications for deep oxygen cycle. Earth and Planetary Science Letters 512, 175-183. doi.org/https://doi.org/10.1016/j.epsl.2019.02.009. Wu, T.-H., Liu, S.-A., 2024. Heavy iron isotopes reveal mantle oxidation in addition to pyroxenite sources for intraplate basalts. Chemical Geology 670, 122432. doi.org/https://doi.org/10.1016/j.chemgeo.2024.122432. Kostrovitskiy, S.I., Solov’eva, L.V., Gornova, M.A., Alymova, N.V., Yakovlev, D.A., Ignat’ev, A.V., Velivetskaya, T.A., Suvorova, L.F., 2012. Oxygen isotope composition in minerals of mantle parageneses from Yakutian kimberlites. Doklady Earth Sciences 444(1), 579-584. doi.org/10.1134/S1028334X12050030. Jochum, K.P., Nohl, U., Herwig, K., Lammel, E., Stoll, B., Hofmann, A.W., 2005. GeoReM: A New Geochemical Database for Reference Materials and Isotopic Standards. Geostandards and Geoanalytical Research 29(3), 333-338. doi.org/10.1111/j.1751-908X.2005.tb00904.x. Bussweiler, Y., Giuliani, A., Greig, A., Kjarsgaard, B.A., Petts, D., Jackson, S.E., Barrett, N., Luo, Y., Pearson, D.G., 2019. Trace element analysis of high-Mg olivine by LA-ICP-MS – Characterization of natural olivine standards for matrix-matched calibration and application to mantle peridotites. Chemical Geology 524, 136-157. doi.org/https://doi.org/10.1016/j.chemgeo.2019.06.019. Howarth, G.H., 2018. Olivine megacryst chemistry, Monastery kimberlite: Constraints on the mineralogy of the HIMU mantle reservoir in southern Africa. Lithos 314-315, 658-668. doi.org/https://doi.org/10.1016/j.lithos.2018.07.007. Giuliani, A., 2018. Insights into kimberlite petrogenesis and mantle metasomatism from a review of the compositional zoning of olivine in kimberlites worldwide. Lithos 312-313, 322-342. doi.org/https://doi.org/10.1016/j.lithos.2018.04.029. Soltys, A., Giuliani, A., Phillips, D., Kamenetsky, V.S., 2020. Kimberlite Metasomatism of the Lithosphere and the Evolution of Olivine in Carbonate-rich Melts — Evidence from the Kimberley Kimberlites (South Africa). Journal of Petrology 61(6), egaa062. doi.org/10.1093/petrology/egaa062. De Hoog, J.C.M., Gall, L., Cornell, D.H., 2010. Trace-element geochemistry of mantle olivine and application to mantle petrogenesis and geothermobarometry. Chemical Geology 270(1‚Äì4), 196-215. doi.org/http://dx.doi.org/10.1016/j.chemgeo.2009.11.017. Motsamai, T., Harris, J.W., Stachel, T., Pearson, D.G., Armstrong, J., 2018. Mineral inclusions in diamonds from Karowe Mine, Botswana: super-deep sources for super-sized diamonds? Mineralogy and Petrology 112(1), 169-180. doi.org/10.1007/s00710-018-0604-9. Lahermaa, P. (2022). Indicator Mineral Chemistry and Diamond Potential of the Liqhobong Kimberlites, Lesotho, Southern Africa. MSc thesis. University of Helsinki, Finland. Nimis, P., Preston, R., Perritt, S. H., & Chinn, I. L. (2020). Diamond's depth distribution systematics. Lithos , 376 , 105729. Moore, A.E., Lock, N.P., 2001. The origin of mantle-derived megacrysts and sheared peridotites-evidence from kimberlites in the northern Lesotho - Orange Free State (South Africa) and Botswana pipe clusters. South African Journal of Geology 104(1), 23-38. Li, X., He, Y., Ke, S., Sun, A.Y., Zhang, Y.C., Wang, Y., Yang, R.Y., Wang, P.J., 2022. High-precision Fe Isotope Analysis on MC-ICPMS Using a 57Fe-58Fe Double Spike Technique. Atomic Spectroscopy 43(2), 164-173. Chen, K., Bao, Z., Yuan, H., Lv, N., 2022. Direct measurement of Fe isotope compositions in iron-dominated minerals without column chromatography using MC-ICP-MS. Journal of Analytical Atomic Spectrometry 37(2), 249-263. doi.org/10.1039/D1JA00381J. le Roex, A.P., Bell, D.R., Davis, P., 2003. Petrogenesis of Group I Kimberlites from Kimberley, South Africa: Evidence from Bulk-rock Geochemistry. Journal of Petrology 44(12), 2261-2286. doi.org/ 10.1093/petrology/egg077 . Stamm, N., Schmidt, M.W., Szymanowski, D., von Quadt, A., Mohapi, T., Fourie, A., 2018. Primary petrology, mineralogy and age of the Letšeng-la-Terae kimberlite (Lesotho, Southern Africa) and parental magmas of Group-I kimberlites. Contrib Mineral Petrol 173(9), 76. doi.org/10.1007/s00410-018-1502-1. Bussweiler, Y., Foley, S.F., Prelević, D., Jacob, D.E., 2015. The olivine macrocryst problem: New insights from minor and trace element compositions of olivine from Lac de Gras kimberlites, Canada. Lithos 220–223, 238-252. doi.org/http://dx.doi.org/10.1016/j.lithos.2015.02.016. Katayama, I., Suyama, Y., Ando, J.-i., Komiya, T., 2009. Mineral chemistry and P–T condition of granular and sheared peridotite xenoliths from Kimberley, South Africa: Origin of the textural variation in the cratonic mantle. Lithos 109(3–4), 333-340. doi.org/http://dx.doi.org/10.1016/j.lithos.2008.05.004. Heckel, C., Woodland, A.B., Linckens, J., Gibson, S.A., Seitz, H.-M., 2022. Sheared Peridotites from Kimberley (Kaapvaal Craton, RSA): Record of Multiple Metasomatic Events Accompanied with Deformation. Journal of Petrology 63(10), egac096. doi.org/10.1093/petrology/egac096. Menzies, A., Westerlund, K., Grütter, H., Gurney, J., Carlson, J., Fung, A., Nowicki, T., 2004. Peridotitic mantle xenoliths from kimberlites on the Ekati Diamond Mine property, N.W.T., Canada: major element compositions and implications for the lithosphere beneath the central Slave craton. Lithos 77(1–4), 395-412. doi.org/http://dx.doi.org/10.1016/j.lithos.2004.04.013. Additional Declarations There is NO Competing Interest. Supplementary Files Dataset1Olivinemajorandtrace.xlsx Dataset 1 Dataset2Feisotopes.xlsx Dataset 2 Dataset3Feisotcalculations.xlsx Dataset 3 SupplementaryData.docx Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6808081","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Physical Sciences - Article","associatedPublications":[],"authors":[{"id":465761923,"identity":"233b8317-01f6-4ec0-b986-13b8474db6d1","order_by":0,"name":"Geoffrey Howarth","email":"data:image/png;base64,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","orcid":"","institution":"University of Cape Town","correspondingAuthor":true,"prefix":"","firstName":"Geoffrey","middleName":"","lastName":"Howarth","suffix":""},{"id":465761924,"identity":"466e3eda-c6a4-4a60-9274-f48274cc179c","order_by":1,"name":"Andrea Giuliani","email":"","orcid":"https://orcid.org/0000-0002-6823-2807","institution":"Carnegie Institution for Science","correspondingAuthor":false,"prefix":"","firstName":"Andrea","middleName":"","lastName":"Giuliani","suffix":""},{"id":465761925,"identity":"c5f63ac9-6343-4c91-8b4c-13b7f57b7897","order_by":2,"name":"Merrily Tau","email":"","orcid":"","institution":"University of Cape Town","correspondingAuthor":false,"prefix":"","firstName":"Merrily","middleName":"","lastName":"Tau","suffix":""},{"id":465761926,"identity":"2b31b419-d610-4c60-afbf-c73f065e6972","order_by":3,"name":"Ronghua Cai","email":"","orcid":"","institution":"China University of Geoscience","correspondingAuthor":false,"prefix":"","firstName":"Ronghua","middleName":"","lastName":"Cai","suffix":""},{"id":465761927,"identity":"8923ccaf-0bb7-4810-81fe-daf0d4c406dd","order_by":4,"name":"Jingao Liu","email":"","orcid":"https://orcid.org/0000-0002-8800-5898","institution":"China University of Geosciences (Beijing)","correspondingAuthor":false,"prefix":"","firstName":"Jingao","middleName":"","lastName":"Liu","suffix":""}],"badges":[],"createdAt":"2025-06-03 07:16:53","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6808081/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6808081/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":84055654,"identity":"25170829-e317-42de-a94b-786653c4c024","added_by":"auto","created_at":"2025-06-06 09:16:28","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":35891,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eOlivine composition in kimberlites containing CLIPPIR diamonds.\u003c/strong\u003e The inverse correlation between mean Mg# (= atomic proportions of Mg/(Mg+Fe)) of magmatic olivine rims and the proportion of olivine xenocrystic cores with Mg# \u0026lt;90 in kimberlites worldwide indicates the effects of lithospheric mantle (olivine xenocryst) assimilation on the composition of kimberlite melts\u003csup\u003e14\u003c/sup\u003e. Olivine cores with Mg# \u0026lt;90 are predominantly megacrystic\u003csup\u003e84\u003c/sup\u003e or from similar Fe-rich lithologies \u003csup\u003e45,85\u003c/sup\u003e and reflect kimberlite sampling of anomalous Fe-rich lithospheric sources. Kimberlites containing CLIPPIR diamonds are characterised by high proportions of megacrystic olivine suggesting a potential link between these diamonds and this Fe-rich mantle substrate. Data sources: \u003csup\u003e9, 14\u003c/sup\u003e.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-6808081/v1/1ad46a2866a322e20bacfc90.png"},{"id":84055658,"identity":"5eef111f-aa3e-473d-88c4-e70e7b7e51c2","added_by":"auto","created_at":"2025-06-06 09:16:28","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":183036,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eXenocrystic olivine compositions in kimberlites containing CLIPPIR diamonds.\u003c/strong\u003e a) Mg# vs Al (ppm) and b) Al vs V (ppm) covariation diagrams for peridotitic (Mg# \u0026gt;89) and megacrystic (Mg# \u0026lt;89) olivine cores for kimberlites of the Orapa field (Karowe, Damtshaa), northern Lesotho field (Letseng, Mothae, Monastery), and Premier. Temperature dashed lines and peridotite classification after Bussweiler et al\u003csup\u003e23\u003c/sup\u003e. These CLIPPIR-bearing kimberlites are characterised by a high proportion of high-Al (\u0026gt;100 ppm), high-temperature olivine relative to kimberlites without CLIPPIR diamonds where high-Al cores are scarce to absent (see \u003cstrong\u003eFig. S6\u003c/strong\u003e). The high-Al concentrations of the predominantly Fe-rich olivine (grey box) reflect transient high temperatures in the deep mantle associated with metasomatism by precursor kimberlite liquids. c) and d) Kernel density estimation (KDE) curves for the calculated sampling depth of peridotitic olivines from kimberlites containing and lacking CLIPPIR diamonds, respectively. Depths were calculated iteratively using the Al-in-olivine thermometer and known geothermal gradients for each location (see \u003cstrong\u003eMethods\u003c/strong\u003e). The absence of substantial deep-lithosphere sampling (i.e. in the diamond stability field) for CLIPPIR-bearing kimberlites is a bias due to the high-Al megacrystic olivine falling outside the calibration range of the Al-in-olivine thermometer\u003csup\u003e23\u003c/sup\u003e. This deep, Fe-rich region of the lithosphere is interpreted to be the likely host of CLIPPIR diamonds. The exception being Premier where high-Al cores are slightly more Mg-rich than those from the Orapa field and northern Lesotho, which pass Al-in-olivine filters. Data sources for olivine in CLIPPIR-bearing kimberlites and Finsch: this study; Kimberley and Ekati:\u003csup\u003e16, 95\u003c/sup\u003e; Jericho field including Jericho, Muskox, and Voyageur kimberlites\u003csup\u003e21\u003c/sup\u003e.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-6808081/v1/08cd2c30d514f6f3092b4038.png"},{"id":84056501,"identity":"98c0822c-1b5a-494d-83da-d976ae75565c","added_by":"auto","created_at":"2025-06-06 09:24:28","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":88860,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRelation between enrichment in Fe and transition metals, and oxygen-Fe isotopes in Fe-rich deep lithospheric domains. \u003c/strong\u003ea) Olivine core mean Mg# vs. Zn (ppm). Data compiled from Howarth et al\u003csup\u003e16\u003c/sup\u003e. b) Olivine core δ\u003csup\u003e18\u003c/sup\u003eO vs olivine core Mg# for xenocrystic olivine previously measured by SIMS \u003csup\u003e28, 29\u003c/sup\u003e and megacrysts analysed by laser fluorination \u003csup\u003e30,\u003c/sup\u003e \u003csup\u003e81\u003c/sup\u003e. Fe-rich olivines are characterised by δ\u003csup\u003e18\u003c/sup\u003eO below the typical mantle range of 5.18 ± 0.28‰ \u003csup\u003e31\u003c/sup\u003e, which reflects interaction between isotopically anomalous, Fe-rich, deep lithosphere and precursor kimberlite melts with mantle-like oxygen isotopes\u003csup\u003e28\u003c/sup\u003e. c) Mean Mg# of magmatic olivine rims vs bulk-rock δ\u003csup\u003e56\u003c/sup\u003eFe. The inverse correlation between kimberlite melt Fe contents exemplified by olivine Mg# (data sources Giuliani et al. \u003csup\u003e14, 15\u003c/sup\u003e) and bulk-kimberlite Fe isotopes (\u003cstrong\u003eDataset 2\u003c/strong\u003e) is due to increasing assimilation of Fe-rich deep lithosphere with isotopically heavy Fe isotopes. The average Mg# of olivine rims is used here rather than that of the olivine cores because these two variables are highly correlated\u003csup\u003e14\u003c/sup\u003e but using the rims, which feature almost invariant Mg#, reduce the potential sampling bias arising from measurements of randomly selected cores of largely variable composition. These data combined show that kimberlites with Fe-rich olivine entrain material from the deep lithosphere which feature anomalous oxygen and Fe isotopes and could correspond to the source of CLIPPIR diamonds.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-6808081/v1/d0f8c1230fc43c313c684475.png"},{"id":84421264,"identity":"07f34547-324f-4431-a257-7abc661f6a4e","added_by":"auto","created_at":"2025-06-11 18:15:56","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1215665,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6808081/v1/59fa3b77-3e87-4265-8a6f-14eeaba91e7f.pdf"},{"id":84055659,"identity":"fcc865cb-18fa-4ce0-ac9d-55973e25fd52","added_by":"auto","created_at":"2025-06-06 09:16:28","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":345473,"visible":true,"origin":"","legend":"Dataset 1","description":"","filename":"Dataset1Olivinemajorandtrace.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-6808081/v1/4026ad388028184ee12c4df6.xlsx"},{"id":84055657,"identity":"c36a6a2b-e12c-4e1d-8687-929b482c6712","added_by":"auto","created_at":"2025-06-06 09:16:28","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":11632,"visible":true,"origin":"","legend":"\u003cp\u003eDataset 2\u003c/p\u003e","description":"","filename":"Dataset2Feisotopes.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-6808081/v1/8ebdee401c587dcc5fde0e15.xlsx"},{"id":84055655,"identity":"73473cd8-6983-401a-ba79-f2209a3b5e24","added_by":"auto","created_at":"2025-06-06 09:16:28","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":13782,"visible":true,"origin":"","legend":"Dataset 3","description":"","filename":"Dataset3Feisotcalculations.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-6808081/v1/c68bfa67f2fc1acef077d3d5.xlsx"},{"id":84055660,"identity":"3e9d42ca-7e3c-4eed-94f2-ab66d7e5a8ed","added_by":"auto","created_at":"2025-06-06 09:16:28","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":7885065,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryData.docx","url":"https://assets-eu.researchsquare.com/files/rs-6808081/v1/373c8c1ac6eeb4754c5ff9e9.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"The largest diamonds are hosted in iron-rich substrate accreted at the base of the lithosphere","fulltext":[{"header":"Introduction","content":"\u003cp\u003eDiamonds larger than several centimetres and \u0026gt;100 carats represent exceptional and highly valuable gems that are sold for millions of dollars\u003csup\u003e6\u003c/sup\u003e. The economic feasibility of some diamond mines such as Karowe (Orapa field) in Botswana and Letseng in northern Lesotho relies entirely on the extraction of these large stones. These diamonds can also provide glimpses of peculiar processes occurring in the deep Earth, which concentrate carbon and foster the growth of unusually large crystals in the deep mantle. Yet, these diamonds are notable components of the diamond population in few localities that are mostly restricted to southern Africa (e.g., Karowe in Botswana, Letseng in Lesotho, Premier in South Africa), and their origin is far from clear.\u003c/p\u003e\n\u003cp\u003ePrevious studies have shown that most of these very large diamonds exhibit specific features leading to the CLIPPIR (Cullinan-like, large, inclusion-poor, pure, irregular, resorbed) acronym\u003csup\u003e1\u003c/sup\u003e. The abundant occurrence of Fe-rich metallic inclusions in CLIPPIR diamonds testifies to reducing conditions associated with diamond formation and the rare occurrence of majorite inclusions places crystallisation within the mantle transition zone (MTZ: 410-660 km)\u003csup\u003e1\u003c/sup\u003e. The combination of extremely heavy Fe isotope composition (d\u003csup\u003e56\u003c/sup\u003eFe = 0.79-0.90\u0026permil;) and unradiogenic Os isotopes in the metallic inclusions was employed to suggest an origin of CLIPPIR diamonds from subducted serpentinized peridotite\u003csup\u003e6\u003c/sup\u003e. In contrast, the highly variable carbon isotope compositions of CLIPPIR diamonds (d\u003csup\u003e13\u003c/sup\u003eC = -27\u0026permil; to +4\u0026permil;)\u003csup\u003e1, 2, 3\u003c/sup\u003e that overlaps the typical range of lithospheric eclogitic diamonds\u003csup\u003e7\u003c/sup\u003e indicates a crustal carbon source for the diamond-forming fluids. The composition of majorite (low Ti but high Na) and breyite inclusions (low Ti, but containing exsolved CaTiO\u003csub\u003e3\u003c/sub\u003e) provides more equivocal constraints, which point to both crustal and peridotitic components. Partial retrogression of the majorite inclusions\u003csup\u003e1\u003c/sup\u003e suggest that CLIPPIR diamonds were probably accreted at the base of the lithosphere perhaps within diapirs of buoyant, harzburgitic lithosphere of oceanic ancestry that was subducted to the MTZ, similar to the model proposed for smaller sub-lithospheric diamonds\u003csup\u003e8\u003c/sup\u003e. Kimberlite magmas sourced in the upper convective mantle\u003csup\u003e9\u003c/sup\u003e then entrained and transported these diamonds to the surface.\u003c/p\u003e\n\u003cp\u003eAn alternative hypothesis entails crystallisation of CLIPPIR diamonds from kimberlitic melts in the deep lithosphere\u003csup\u003e4, 10\u003c/sup\u003e. This suggestion was based on the common occurrence of CLIPPIR diamonds in kimberlites with abundant Cr-poor megacrysts combined with the size of these diamonds reaching those of lithospheric megacrysts (\u0026gt;1 mm) \u0026ndash; these large sizes are at odds with the smaller sizes predicted for convective mantle minerals\u003csup\u003e11\u003c/sup\u003e. This model, supported by recent observations of the relative abundance of megacrystic garnets in southern African kimberlites\u003csup\u003e12\u003c/sup\u003e, however, cannot account for the direct evidence of CLIPPIR diamond formation in the MTZ, namely the inclusions of partly retrogressed majorite\u003csup\u003e1\u003c/sup\u003e. The association between CLIPPIR and sub-lithospheric diamonds with megacrysts might simply indicated preferential sampling of the deep lithosphere rather than a genetic link\u003csup\u003e12, 13\u003c/sup\u003e. The source lithologies of these large diamonds remains obscure and potentially underpin the occurrence of unusual processes and/or peculiar domains in the mantle, which might have far-reaching implications for the formation of compositional heterogeneities in the deep Earth.\u003c/p\u003e\n\u003cp\u003eIn this study, we explore the association of CLIPPIR diamonds with Cr-poor megacrysts to resolve the origin of the largest diamonds recovered to date. Using the composition of olivine xenocrysts, we compare the compositional features of the lithospheric mantle entrained by kimberlites in areas containing abundant CLIPPIR and other sub-lithospheric diamonds with those that rarely contain such large stones. A combination of these results with existing oxygen isotope data for olivine and new bulk-kimberlite Fe isotopes suggests that CLIPPIR diamonds are associated with hydrothermally altered, Fe-rich basaltic crust that was subducted to the MTZ and then accreted at the base of the lithosphere as eclogite. Beyond providing abundant sub-lithospheric diamonds, these isotopically anomalous Fe-rich domains contribute to the compositional heterogeneity observed in intraplate magmas globally, including kimberlites.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIron enrichment of the deep lithosphere associated with CLIPPIR and other sub-lithospheric diamonds\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA global compilation of the composition of olivine in kimberlites\u003csup\u003e14, 15\u003c/sup\u003e shows that the major producers of CLIPPIR diamonds are kimberlites with abundant or dominant Fe-rich olivine, both xenocrystic (olivine cores) and magmatic (olivine rims; Mg#\u0026lt;89) (\u003cstrong\u003eFigure 1 and Fig S1\u003c/strong\u003e). Previous work has robustly linked the elevated Fe contents of olivine in these kimberlites with sampling of lithospheric mantle roots containing abundant Cr-poor, Fe-rich megacrysts\u003csup\u003e14, 16\u003c/sup\u003e, consistent with recent working using garnet xenocrysts\u003csup\u003e12\u003c/sup\u003e. In this context, megacrysts represent large grains (\u0026gt;1 cm) that based on isotopic and geochronological similarities with kimberlites \u003csup\u003e17, 18\u003c/sup\u003e are genetically associated with failed pulses of kimberlitic magma interacting with the deep lithosphere \u003csup\u003e19, 20\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eTo explore a potential genetic connection between CLIPPIR diamonds and these Fe-rich, deep lithospheric domains, we compare the trace element compositions and sampling depths of olivine xenocrysts from kimberlites in areas where sampling of CLIPPIR and other sub-lithospheric diamonds is well established (Karowe and Damtshaa in the Orapa field; Letseng, Mothae, and Monastery in the northern Lesotho field; Premier kimberlite \u0026ndash; new data) with those from localities that do not contain conspicuous populations of sub-lithospheric diamonds (Murowa kimberlite in Zimbabwe \u0026ndash; new data; Kimberley kimberlites and Finsch in South Africa, Ekati and Jericho fields in Canada \u003csup\u003e16, 21, 22\u003c/sup\u003e) (\u003cstrong\u003eFigure 2\u003c/strong\u003e). Olivine sampling depths were obtained by combining temperature constraints from the Al-in-olivine thermometer\u003csup\u003e23\u003c/sup\u003e with independent geothermal gradients based on previous xenolith and xenocryst thermobarometry (\u003cstrong\u003eMethods\u003c/strong\u003e). Kimberlites from the CLIPPIR-bearing Orapa and northern Lesotho fields show a unimodal distribution of olivine sampling depth centred at between 130 and 155 km and limited sampling (51\u0026plusmn;17 %) within the diamond stability field of the lithosphere. Conversely, olivine in CLIPPIR-poor kimberlites was sampled predominantly (83\u0026plusmn;14 %) in the diamond stability field including major peaks at depths beyond 150-160 km. Major and trace element concentrations of olivine (\u003cstrong\u003eFigure 2\u003c/strong\u003e) indicate that these differences represent a bias due to screening of high-Al and hence hot (~1200-1500\u0026deg;C) olivines enriched in Fe (Mg# \u0026lt;89), V, Zn and other transitional metals (\u003cstrong\u003eDataset 1\u003c/strong\u003e) outside the calibration range of the Al-in-olivine thermometer. These olivine compositions are typical of megacrysts and related sheared peridotites crystallised at pressures of 4.5-7.0 GPa (140-215 km), which are abundant in kimberlites from these CLIPPIR-bearing fields\u003csup\u003e12,\u003c/sup\u003e \u003csup\u003e24, 25, 26, 27\u003c/sup\u003e. While 23% of the Premier olivines are megacrystic, the peridotitic olivine disapply unimodal distribution deeper than other CLIPPIR-bearing locations near the LAB (~210 km) but with sampling of predominantly (84% of peridotitic olivine) high-Al olivine from sheared peridotite. These olivine are similar to the high-Al megacrystic olivine but have slightly higher Mg# (\u0026gt;89) and, thus, pass the Al-in-olivine thermometer screening (\u003cstrong\u003eMethods\u003c/strong\u003e). The trace element composition of deep lithospheric, Al-rich olivine from the CLIPPIR-bearing localities indicates enrichment in Zn, Ti, Mn, and Co (\u003cstrong\u003eFigure 3a; Fig S5\u003c/strong\u003e) suggestive of melt infiltration. These observations combined indicate that the lithospheric roots entrained by kimberlites in CLIPPIR-bearing fields are characterised by vertically extensive domains above the lithosphere-asthenosphere boundary (LAB) that contain abundant and locally dominant Fe-rich material with affinity to the megacryst suite and sheared peridotites.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eExisting oxygen isotope composition for olivine in kimberlites \u003csup\u003e28, 29\u003c/sup\u003e, including olivine megacrysts from Monastery\u003csup\u003e30\u003c/sup\u003e, highlights another key feature of these Fe-rich, deep lithospheric domains, namely low\u0026nbsp;d\u003csup\u003e18\u003c/sup\u003eO values (\u0026lt;5\u0026permil;; \u003cstrong\u003eFigure 3b\u003c/strong\u003e) compared to typical mantle olivine (5.1 \u0026plusmn; 0.3\u0026permil;)\u003csup\u003e31\u003c/sup\u003e. Iron-rich olivines from the northern Lesotho and the Orapa fields, including Karowe, have\u0026nbsp;d\u003csup\u003e18\u003c/sup\u003eO values as low as 4.0\u0026permil; (even lower at the megacryst- and eclogite-rich locality of Kaalvallei in South Africa\u003csup\u003e29\u003c/sup\u003e). These isotopic compositions cannot be explained by common mantle processes, including metasomatism\u003csup\u003e29\u003c/sup\u003e, and require contribution by subducted crustal material, such as hydrothermally altered oceanic crust\u003csup\u003e32\u003c/sup\u003e or serpentinised lithospheric mantle\u003csup\u003e33\u003c/sup\u003e. Given that Fe-rich olivines ensue from the interaction between precursor kimberlite melts with mantle-like oxygen isotopes\u003csup\u003e28\u003c/sup\u003e and lithospheric mantle wall rocks, the subducted component in the wall rocks must have\u0026nbsp;d\u003csup\u003e18\u003c/sup\u003eO \u0026lt;4\u0026permil; \u003csup\u003e29, 30\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eNew Fe isotope data for bulk kimberlites (\u003cstrong\u003eMethods\u003c/strong\u003e) add critical constraints to this picture. Comparison of these data with olivine Mg# (either magmatic, \u003cstrong\u003eFigure 3c\u003c/strong\u003e, or xenocrystic) indicates that Fe isotopes in kimberlites become progressively heavier with increasing Fe contents in olivine, with samples from the Premier, Orapa, and northern Lesotho kimberlite fields at the high\u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe end of the measured range. Previous work has shown that the Fe content (or Mg#) of olivine in kimberlites reflects the relative amount of Fe-rich deep lithospheric material, including megacrystic and metasomatised lithologies (e.g., dunites, wehrlites, websterites), that gets entrained and partly assimilated by kimberlites\u003csup\u003e14\u003c/sup\u003e. In other words, heavy Fe isotopes in kimberlites reflect increasing contribution by Fe-rich megacrystic material that must feature heavier Fe isotope compositions compared to typical mantle values (d\u003csup\u003e56\u003c/sup\u003eFe ~0.00 to 0.05\u0026permil;)\u003csup\u003e34\u003c/sup\u003e. Mass balance calculations (\u003cstrong\u003eMethods\u003c/strong\u003e) show that the\u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe of Fe-rich olivine in these deep lithospheric domains is higher than 0.11-0.30\u0026permil; compared to typical peridotitic olivine with\u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe \u0026lt;0.05\u0026permil; \u003csup\u003e34, 35, 36\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eProvided that the protolith interacted with by kimberlite melts (d\u003csup\u003e56\u003c/sup\u003eFe ~ 0.09-0.10\u0026permil;; \u003cstrong\u003eFigure 3c\u003c/strong\u003e) to generate Fe-rich megacrystic olivine, this protolith must have Fe isotopes even heavier than these values. This Fe isotopic signature requires contribution by subducted, altered oceanic crust\u003csup\u003e37\u003c/sup\u003e, perhaps including isotopically heavy pyroxenites\u003csup\u003e38\u003c/sup\u003e, and provides an alternate interpretation for the origin of high\u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe observed in metallic inclusions hosted by CLIPPIR diamonds. However, considering the scarcity of light oxygen isotope data in eclogite xenoliths, including only a few samples with low oxygen isotope values \u003csup\u003e39, 40, 41\u003c/sup\u003e, and almost complete lack of Fe isotope studies for these xenoliths (only five samples measured in one study\u003csup\u003e42\u003c/sup\u003e), it is difficult to constrain the exact nature of these domains before their interaction with precursor kimberlite melts and conversion to megacrysts and other Fe-rich lithologies (e.g., dunites, wehrlites, websterites \u003csup\u003e43, 44, 45\u003c/sup\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIron-rich CLIPPIR-diamond substrate accreted at the base of the lithosphere\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAbove, we have shown that CLIPPIR diamonds are abundant in kimberlites that entrained a peculiar substrate in the deep lithosphere dominated by Fe-rich material with affinity to megacrysts. These domains feature isotopically light oxygen and isotopically heavy Fe, which suggests affinity to subducted, altered oceanic crust that was probably converted to eclogite. Although Fe-rich olivine related to megacrysts and sheared peridotites was generated by interaction with precursor kimberlite melts, it is unlikely that these isotopic features were imparted by precursor kimberlite melts because the convective mantle sources of kimberlites contain only small amounts of subducted material - based on trace element concentrations and Sr-Nd-Hf-Pb-C-O isotopic compositions \u003csup\u003e14, 28, 29,\u003c/sup\u003e \u003csup\u003e46, 47, 48\u003c/sup\u003e. The compositional features of the megacrystic domain, including enrichment in Fe, Zn, Ti and other transition metals, light O isotopes and heavy Fe isotopes, were inherited from a subducted crustal substrate. The highly variable C isotope composition of CLIPPIR diamonds, including very low\u0026nbsp;d\u003csup\u003e13\u003c/sup\u003eC (down to -27\u0026permil;), but also elevated\u0026nbsp;d\u003csup\u003e13\u003c/sup\u003eC of up to +4\u0026permil; \u003csup\u003e2, 3\u003c/sup\u003e and the high\u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe (0.79-0.90\u0026permil;) of their metallic inclusions \u003csup\u003eSmith2021\u003c/sup\u003e are consistent with an origin in such a peculiar substrate. Unradiogenic Os isotopes and Ni contents in the metallic inclusions\u003csup\u003e6\u003c/sup\u003e coupled with the low-Cr compositions of included majorites that are intermediate between mafic and peridotitic protoliths\u003csup\u003e1\u003c/sup\u003e, however, suggest involvement of additional peridotitic components, perhaps in a subduction melange. It appears likely that CLIPPIR and sub-lithospheric diamonds with eclogitic affinity represent an integral, although sporadic component, of the Fe-rich isotopically anomalous substrate that was partly converted to Fe-rich megacrysts and other Fe-rich lithologies (e.g., dunites, wehrlites, websterites \u003csup\u003e43, 44, 45\u003c/sup\u003e) during interaction with precursor kimberlite melts.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe process of interaction between lithospheric wall rocks and precursor kimberlite melts, which generates megacrysts and sheared peridotites \u003csup\u003e19, 20, 27\u003c/sup\u003e, might explain several features of CLIPPIR diamonds including deformation, annealing and resorption. These features reflect a highly dynamic environment where CLIPPIR diamonds resided before being entrained and transported to the surface - such as lithospheric wall rocks lining kimberlite magmatic conduits. Here fluid-mediated deformation and recrystallisation are commonly inferred from textural studies of sheared peridotites and recrystallised megacrysts \u003csup\u003e49, 50\u003c/sup\u003e, including samples from CLIPPIR-bearing localities \u003csup\u003e24, 27\u003c/sup\u003e. This hypothesis is supported by the low diamond grades of typical CLIPPIR localities (e.g., 1.5 cpht for Letseng; 20 cpht for Karowe), which has been attributed to diamond destruction associated with extensive infiltrations of the deep lithosphere by failed pulses of kimberlite melt\u003csup\u003e\u0026nbsp;15, 51\u003c/sup\u003e. In this scenario, the unusual abundance of large CLIPPIR diamonds compared to smaller diamonds in some of these localities, such as the Letseng mine, can be explained by near-complete resorption of the smaller diamonds\u003csup\u003e5\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe remaining question is how this CLIPPIR substrate was exhumed from the MTZ\u003csup\u003e1\u003c/sup\u003e and accreted at the base of the lithosphere. The composition of minerals included in other sub-lithospheric diamonds and the C and N isotope systematics of the diamond hosts, including examples from Monastery and other localities in southern and western Africa, Brazil, Canada and Australia, indicate that a large fraction of sub-lithospheric diamonds derive from oceanic crust that was subducted to the MTZ or deeper\u003csup\u003e52, 53, 54, 55\u003c/sup\u003e. These observations, combined with the almost ubiquitous (partial) retrogression of majorite\u003csup\u003e56\u003c/sup\u003e and the other high-pressure silicate minerals, suggests that accretion of deeply subducted oceanic crust to the bottom of the lithosphere is likely to occur in several localities. Some support to this conclusion comes from previous reports\u003csup\u003e57, 58\u003c/sup\u003e that eclogites are locally concentrated in the lower lithosphere sampled by some kimberlites. These localities include the Orapa kimberlite and Kaalvallei (proximal to the north Lesotho field), where we have documented deep lithosphere with anomalous oxygen and Fe isotope systematics.\u003c/p\u003e\n\u003cp\u003eA possible accretion scenario entails diapiric rising of buoyant, subducted oceanic lithosphere including slivers of oceanic crust\u003csup\u003e\u0026nbsp;8, 59\u003c/sup\u003e. While feasible, this model faces two potential challenges, namely the dominantly lherzolitic and metasomatically enriched nature of the lowermost cratonic lithosphere, where harzburgites and similarly depleted lithologies are not observed\u003csup\u003e57\u003c/sup\u003e, and the paucity of highly depleted and therefore buoyant harzburgite in subducted oceanic lithosphere. An alternative mechanism to accrete dense, Fe-rich eclogitic material of crustal derivation to the bottom of the lithosphere is active mantle upwelling associated with plumes from the core-mantle boundary (or shallower depths). Although early models\u003csup\u003e60\u003c/sup\u003e suggested negligible entrainment of ambient mantle material along the edges of sub-vertical plumes, recent seismic tomography models show substantial deflection and reorganisation of mantle plumes in the mid-mantle region (\u0026lt;1000 km) \u003csup\u003e61, 62\u003c/sup\u003e. Entrainment of crustal material residing in the MTZ by plumes from the lower mantle and/or transport in secondary plumes that are generated at these depths appears possible\u003csup\u003e63\u0026nbsp;\u003c/sup\u003eand has been previously proposed to explain the H\u003csub\u003e2\u003c/sub\u003eO enrichment of some komatiites\u003csup\u003e64\u003c/sup\u003e. Kimberlites containing CLIPPIR and other sub-lithospheric diamonds in southern Africa were either centred above (Orapa field, Premier/Cullinan) or peripheral (northern Lesotho, including Monastery) to regions where major mantle plumes impinged the bottom of the lithosphere (Umkondo for Orapa field, Bushveld for Premier, Bushveld and Karoo for northern Lesotho\u003csup\u003e65\u003c/sup\u003e). The geochemistry and Re-Os isotopes of mantle xenoliths from Premier\u0026nbsp;coupled with seismic tomography of the Kaapvaal craton provide clear indication of accretion of Fe-rich lithosphere associated with the Bushveld plume at ~2.0 Ga \u003csup\u003e66,\u0026nbsp;67, 68, 69\u003c/sup\u003e. This process has been considered to be of global significance in growing continental lithosphere\u003csup\u003e70\u003c/sup\u003e as well as healing previously scarred cratons\u003csup\u003e71\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThe lithosphere beneath all the major CLIPPIR-bearing localities of southern Africa experienced interaction with strongly buoyant plumes, which were able to transport and potentially accrete dense material to the base of the lithosphere. This mechanism might be responsible for the occurrence of abundant, isotopically anomalous, Fe-rich material of subducted crustal origin in the deep lithosphere at these localities, including CLIPPIR and other sub-lithospheric diamonds. The occurrence of plume-derived He isotopes in sub-lithospheric diamonds from Juina (Brazil)\u003csup\u003e72\u003c/sup\u003e points to a likely plume influence in the transport of sub-lithospheric diamonds from this region too. Finally, although the high temperatures associated with deep mantle plumes can be detrimental towards diamond preservation, previous thermal modelling\u003csup\u003e73\u003c/sup\u003e shows that diamond can survive transient heating events, which is confirmed by the occurrence of Archean diamonds in the lithosphere beneath Premier and the Orapa field\u003csup\u003e74\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImplications for diamond exploration and global mantle heterogeneities\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe association between the largest gem diamonds recovered to date and kimberlites that sampled Fe-rich domains with anomalous oxygen and Fe isotope compositions in the deep lithosphere supports a likely genetic relationship. This connection also enlightens an exploration pathfinder for the most precious diamonds where Fe enrichment in olivine and large populations of Fe-rich megacrysts can be employed to target kimberlites with high potential for CLIPPIR diamonds, consistent with suggestions in previous studies\u003csup\u003e12\u003c/sup\u003e. However, as customary for diamonds, this association is not exclusive. This is exemplified by the Karowe mine where erratic kimberlite sampling of the mantle resulted in CLIPPIR diamonds being found in just one of the three volcanic lobes\u003csup\u003e75\u003c/sup\u003e even though the underlying lithosphere is unlikely to be substantially different on such a small scale (\u0026lt;1 km\u003csup\u003e2\u003c/sup\u003e). Similarly, this study identifies other kimberlites tapping anomalous Fe-rich deep mantle domains (e.g., Kaalvallei) where CLIPPIR diamonds have not yet been reported.\u003c/p\u003e\n\u003cp\u003eThe discovery of isotopically anomalous domains enriched in Fe, Zn, Ti and other transition metals bears important implications to understand the genesis of compositional heterogeneities in intraplate magmas globally. Magmatic olivines in kimberlites exhibit a wide range in Mg# where the lowermost values (83-84) have been explained by interaction with Fe-rich material related to precursor kimberlite melts\u003csup\u003e14\u003c/sup\u003e. However, considering the origin of kimberlite melts from unremarkable convective mantle sources (e.g., olivine Mg# ~89)\u003csup\u003e14\u003c/sup\u003e, the assimilated Fe-rich material cannot simply result from kimberlite metasomatism, but requires a pre-existing Fe-rich substrate. Such a substrate is coincident with the Fe-rich isotopically anomalous, presumably eclogitic domains envisaged in this work. Iron-rich domains at the base of the lithospheric mantle, in virtue of their lower melting degrees compared to refractory peridotites, can substantially contribute to the genesis of intraplate magmas. Their influence is clearly noticeable in kimberlites from southern Africa and other Fe-rich localities (e.g., Alto Paranaiba \u003csup\u003e16,\u003c/sup\u003e \u003csup\u003e76\u0026nbsp;\u003c/sup\u003ein Brazil which also feature extreme Sr-Nd-Hf isotopic compositions\u003csup\u003e47,\u003c/sup\u003e \u003csup\u003e77\u003c/sup\u003e), and might also contribute to isotopic anomalies of intraplate magmatic provinces elsewhere including the Emeishan large igneous province (LIP) and the extensive intraplate basaltic province of eastern China \u003csup\u003e78, 79, 80\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eOlivine geochemistry\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOlivine major and minor elements (except for Premier and Murowa) were analysed using a CAMECA SX-100 electron probe microanalyser (EPMA) housed at the University of Johannesburg, South Africa.\u0026nbsp;Analyses were conducted using an acceleration potential of 15 kV, beam current of 20 nA, and beam diameter of 1 \u0026mu;m. Peak counting times varied between 10 s and 60 s depending on measured element. All elements were measured on the K\u0026alpha; line. Reference materials used to calibrate the instrument included jadeite (Na), olivine (Mg), almandine (Al), diopside (Si), wollastonite (Ca), rhodonite (Mn), hematite (Fe), and synthetic Cr, Ni, and Ti oxides. Detection limits are ~ 0.01 wt% for all elements. Data reduction and matrix correction was performed using the \u0026lsquo;X-PHI\u0026rsquo; method.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMajor elements in the Premier and Murowa samples were analysed at Carnegie Institution for Science using a JEOL 8530F Field Emission EPMA, equipped\u0026nbsp;with 5 wavelength dispersive\u0026nbsp;spectrometers. The probe was operated at 15kV and 30nA, and a 1um diameter electron beam. The Probe for EPMA software was used to collect and reduce the data. The standards used were Olivine \u0026ndash; San Calros (Mg, Si), Ilmenite (Ti, Fe, Mn), MgCr\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e4\u003c/sub\u003e (Cr), Anorthite (Al, Ca), and Ni Olivine (Ni).\u003c/p\u003e\n\u003cp\u003eTrace elements analyses in all olivine (except for Premier and Murowa; see below) were performed using a 193 nm Resolution M50 LR Excimer laser ablation system attached to an Agilent 8800 quadrupole inductively coupled plasma mass spectrometer (LA-ICP-MS) at the Stellenbosch University, South Africa. Beam sizes of 100 and 70 \u0026micro;m were used, the former being generally the preferred choice. Background acquisition, ablation and washout times of 15, 35, and 25 seconds were employed, respectively. Laser repetition rate of 7 Hz and energy fluence of 4 J/ cm\u003csup\u003e2\u003c/sup\u003e were applied. Helium was used as the ablation gas and sample transported to the ICP-MS system using an argon gas carrier. Reference materials NIST (National Institute of Standards and Technology) 612 was used as calibration standards. Glass reference materials BCR-2G and BHVO-2G (values from GeoReM\u003csup\u003e82\u003c/sup\u003e) and olivine 355OL\u003csup\u003e83\u0026nbsp;\u003c/sup\u003ewere used for quality control (see \u003cstrong\u003eDataset 1\u003c/strong\u003e). Silicon and Mg measured were used as internal standards with nominal values obtained by EMPA. A typical analytical session consisted of 15-20 analyses of unknowns bracketed by analyses of the reference materials. Elements highly incompatible in olivine (including Rb, Sr, Ba, Zr, Nb, Y) were monitored to assess potential contamination by inclusions or material in fractures and/or grain boundaries. Data reduction was performed using the LADR v 1.1.07 software package from Norris Scientific. Trace elements measured in the 355OL olivine were reproduced to better than 90% of the solution values\u003csup\u003e83\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eTrace element analyses for Premier and Murowa olivine were analysed at the Carnegie Institution for Science using the Teledyne Iridia laser connected to an iCAPq mass spectrometer. A beam size of 35 \u0026micro;m was used. A laser repetition rate of 5 Hz and energy fluence of 4 j/cm2 were applied. Background acquisition, ablation and washout times of 30, 40, and 30 seconds were employed, respectively. Argon gas was used with a gas flow rate of 0.9 l/min for transport to the mass spec. Reference materials 355OL and BHVO-2G were used. 355Ol calibration was used for all elements with the exception of Ca where BHVO-2G calibration was used. Silicon was used as an internal standard. Elements highly incompatible in olivine (including Rb, Sr, Ba, Zr, Nb, Y) were monitored to assess potential contamination by inclusions or material in fractures and/or grain boundaries.\u003c/p\u003e\n\u003cp\u003eThe composition of the olivine cores from the Orapa field (Karowe, Damtshaa; \u003cstrong\u003eFig S2\u003c/strong\u003e), northern Lesotho field (Letseng, Mothae, Monastery; \u003cstrong\u003eFig S3\u003c/strong\u003e), and Premier (\u003cstrong\u003eFig S4\u003c/strong\u003e) are presented here to show the range in olivine compositions observed at CLIPPIR-bearing localities. Overall, the olivine xenocrysts display a large range in Mg# from 93.5 to 80.9 and coherent trends between major and trace elements. Letseng, Mothae, and Premier, however, have far fewer olivine cores \u0026nbsp;with Mg \u0026lt;84 relative to Karowe and Damtshaa, indicating that megacrysts at Letseng, Mothae, and Premier rarely evolved to the extremely Fe-rich compositions observed in the Orapa field. This is in contrast to the Monastery kimberlite that has a large population of Fe-rich megacrysts \u003csup\u003e25, 84\u003c/sup\u003e, indicating local variations in the deep lithosphere composition.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe mean olivine core composition for each location represents the nature of the lithospheric mantle sampled by the kimberlites (\u003cstrong\u003eFig. S5\u003c/strong\u003e). The olivines are broadly divided into Mg-rich (Mg# \u0026gt;89) and Fe-rich (Mg# \u0026lt;89) groups (\u003cstrong\u003eFigs S2; S3\u003c/strong\u003e) to discriminate peridotitic from anomalous sources such as megacrysts, sheared peridotites and Fe-rich dunites \u003csup\u003e16, 45, 85, 86\u003c/sup\u003e. In cases where the mean olivine Mg# is \u0026gt;90 (e.g., Ekati, Finsch, Murowa), the kimberlite has predominantly sampled refractory mantle with little to no Fe-rich megacrystic material. In contrast, kimberlites that have low mean olivine core (Mg# \u0026lt;90) sample a high proportion of Fe-rich (Mg# \u0026lt;89) megacrystic olivine\u003csup\u003e14\u003c/sup\u003e. Kimberlites that contain high proportions of Fe-rich megacrystic cores and hence low average Mg# for the cores exhibit low Mg# in their magmatic rims (\u003cstrong\u003eFig. S1\u003c/strong\u003e). This correlation has been previously attributed to the important role of lithospheric mantle assimilation on the composition of kimberlite melts\u003csup\u003e14\u003c/sup\u003e, of which olivine rim Mg# represents an excellent proxy. Thus, the Mg# of mean olivine rims and mean olivine cores can be used interchangeably to approximate the composition of kimberlite magmas and traversed lithospheric mantle.\u003c/p\u003e\n\u003cp\u003eThe Mg-rich olivine have constant Ni (~3000 ppm) over their range in Mg# (Mg# 89-93.5) for Karowe and Damtshaa but extend to high Ni values (~3500 ppm) for the most Mg-rich olivine at Letseng, Mothae, and Monastery. For all CLIPPIR-bearing kimberlites analysed here, the Ni concentrations decrease with decreasing Mg# for the Fe-rich cores (i.e., Mg# \u0026lt;89) to values as low as ~1000 ppm due to olivine fractionation\u003csup\u003e84\u003c/sup\u003e and increasing interaction with Ni-poor, Fe-rich substrate during evolution of the megacryst parent melt. Zinc concentrations across the range of olivine core Mg# display a distinct negative correlation. The lowest Zn values are observed for highly refractory, likely harzburgitic or dunitic, olivine cores (Mg# \u0026gt;93) whereas lherzolitic (Mg# 90-93) olivine cores have higher zinc, indicating that Zn concentrations, like Fe, increases during refertilization processes in the SCLM. Zinc in the megacrystic olivine cores behaves as an incompatible element and increasing with decreasing Mg#, consistent with observation in previous studies\u003csup\u003e16\u003c/sup\u003e. In addition, Al and Ca concentrations display a sharp decrease in the early stages of megacryst fractionation (i.e., Mg# 86-89), consistent with previous studies that have suggested concurrent crystallisation of olivine with pyroxene and garnet\u003csup\u003e84\u003c/sup\u003e. A notable observation is the abundant presence of low-Al olivine in equilibrium with spinel-garnet peridotite from the Damtshaa kimberlite relative to other kimberlites, indicating abundant shallow sampling outside of diamond stability. In addition, Damtshaa contains a high abundances of unusual low-Al megacrysts (\u003cstrong\u003eFigure 3 and Fig. S5d\u003c/strong\u003e), which are also observed at Karowe but in substantially lower abundances. These low-Al megacrysts are interpreted to represent low-temperature, relatively shallow formation compared to the high-Al megacrysts predominant in the northern Lesotho and Karowe kimberlites. The Premier kimberlite is unusual in that it does not sample a significant population of typical granular garnet peridotite olivine (\u003cstrong\u003eFig S4.\u003c/strong\u003e) and 85% of the total olivine population is comprised of olivine from Al-rich (i.e., high temperature) megacrystic and sheared peridotite lithologies.\u0026nbsp;This is consistent with the unusual abundance of high temperature sheared peridotite xenoliths at Premier kimberlite\u003csup\u003e68\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOlivine thermobarometry\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe Al-in-olivine thermometer\u003csup\u003e23\u003c/sup\u003e has been calibrated for garnet peridotite mantle xenoliths with olivine Mg# \u0026gt;89 and, thus, its application requires screening olivine compositions outside of the calibration (e.g., Fe-rich megacrysts, olivine from spinel peridotites). This screening was done using the following protocol, which is based on the composition of olivine in typical, texturally equilibrated peridotites \u003csup\u003e23, 85\u003c/sup\u003e: Mg# \u0026gt;89; Ni \u0026gt;2300 ppm (or NiO \u0026gt;0.3 wt.%); Mn \u0026lt;1160 ppm (or MnO \u0026lt;0.15 wt.%); Ca \u0026lt;715 ppm (or CaO \u0026lt;0.1 wt.%). Grains that satisfied these criteria were then further screened for garnet vs spinel peridotite origins using Al-V \u003csup\u003e23\u003c/sup\u003e and Al vs Mn \u003csup\u003e87\u0026nbsp;\u003c/sup\u003erelationships. Iterative calculations were then undertaken on olivine in equilibrium with garnet to match the Al-in-olivine temperature with local geothermal gradients for each kimberlite and using 40 mW/m\u003csup\u003e2\u003c/sup\u003e for the Orapa field\u003csup\u003e88\u003c/sup\u003e, 41 mW/m\u003csup\u003e2\u003c/sup\u003e for the northern Lesotho kimberlites\u003csup\u003e89\u003c/sup\u003e, and 39 mW/m\u003csup\u003e2\u003c/sup\u003e for Premier\u003csup\u003e90\u003c/sup\u003e. The depth estimates are plotted as histograms in \u003cstrong\u003eFig. S7\u003c/strong\u003e and kernel density estimate (KDE) curves in \u003cstrong\u003eFigure 2\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003eThe Al-in-olivine temperatures calculated in this study are generally \u0026gt;900 \u003csup\u003eo\u003c/sup\u003eC and outside of the low temperature range where some overestimation of temperatures have been suggested to occur\u003csup\u003e23\u003c/sup\u003e. Thus, this minor calibration issue does not affect the calculated sampling depths in this study. The major limitation of the Al-in-olivine thermometer to reconstruct olivine entrainment depths is in the selection of a local geotherm. For example, the selection of slightly different geothermal gradients in this study resulted in the minor differences in the main peaks of the resultant KDE curves, namely 145-155 km in the Orapa field and 130-145 km in northern Lesotho. These main peaks would overlap completely if the same geotherm was selected. However, the main modes of mantle sampling in the KDE curves of \u003cstrong\u003eFigure 2\u003c/strong\u003e would not change the stark contrast in sampling depths between these CLIPPIR-bearing kimberlites and CLIPPIR-poor kimberlites.\u003c/p\u003e\n\u003cp\u003eWhile the Al-in-olivine thermometer calibration does not extend to the more Fe-rich range of megacrystic olivine, the Al concentrations in the megacrystic olivine are used here as a qualitative indicator of the likely temperature association of these Fe-rich olivines. The mantle-derived Fe-rich olivines analysed in this study display a positive correlation between Mg# and Al concentrations with the highest Al concentrations observed in the most primitive olivine megacrysts (Mg# 86-89) (\u003cstrong\u003eFig. S2; S3\u003c/strong\u003e), consistent with previous studies\u003csup\u003e51\u003c/sup\u003e. The decrease in Al concentrations with increasing Fe likely reflects a combination of decreasing temperature during crystallisation, interaction with lithospheric mantle, and concurrent garnet co-crystallisation\u003csup\u003e25, 84\u003c/sup\u003e. The Al-in-olivine temperatures for the most primitive (i.e., Mg# = 86-89), Al-rich (i.e., 100-380 ppm) megacrystic olivine are between 1200-1500 \u003csup\u003eo\u003c/sup\u003eC and, while just qualitative, imply thermal perturbation associated with megacryst crystallisation in the deep lithosphere. Previous studies \u003csup\u003e26,91\u0026nbsp;\u003c/sup\u003ehave similarly showed formation of Fe-rich megacrysts and sheared peridotites at elevated temperatures toward the base of the lithosphere.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFe isotopes\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIron isotopic analyses were conducted at the State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences (Beijing)-CUGB. Sample powders containing \u0026ge; 250 \u0026mu;g of Fe were dissolved in a mixture of concentrated HF-HNO\u003csub\u003e3\u003c/sub\u003e. After optical check, precipitate-free solutions were evaporated at 160 \u0026deg;C and re-dissolved in\u0026nbsp;\u003cem\u003eaqua regia\u003c/em\u003e to remove fluorides and completely oxidize Fe\u003csup\u003e2+\u003c/sup\u003e. The resultant solutions were dried at 90 \u0026deg;C and re-fluxed with concentrated HCl. The samples were finally dissolved in 0.5 mL of 6 N HCl before column chromatography. The solutions were purified twice using AG1-X8 resin\u0026nbsp;to completely remove interference elements. Iron isotopic analyses were conducted on a ThermoFisher Neptune Plus MC-ICP MS using a \u003csup\u003e57\u003c/sup\u003eFe-\u003csup\u003e58\u003c/sup\u003eFe double spike technique. The complete analytical procedure is described elsewhere \u003csup\u003e91\u003c/sup\u003e. All Fe isotope data\u0026nbsp;are reported in delta notation against\u0026nbsp;the international standard IRMM-014:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026delta;\u003csup\u003e56\u003c/sup\u003eFe = [(\u003csup\u003e56\u003c/sup\u003eFe/\u003csup\u003e54\u003c/sup\u003eFe) \u003csub\u003esample\u003c/sub\u003e/(\u003csup\u003e56\u003c/sup\u003eFe/\u003csup\u003e54\u003c/sup\u003eFe) \u003csub\u003eIRMM-014\u0026nbsp;\u003c/sub\u003e- 1]\u0026nbsp;*\u0026nbsp;1000\u0026nbsp;(\u0026permil;)\u003c/p\u003e\n\u003cp\u003eThe reference materials analyzed at CUGB yielded results (BHVO-2: 0.118\u0026plusmn;0.013\u0026permil;; BCR-2: 0.093\u0026plusmn;0.016\u0026permil;) identical within errors of published values \u003csup\u003e92, 93\u003c/sup\u003e. The whole procedure blank was less than 15 ng. The analytical errors of\u0026nbsp;\u0026delta;\u003csup\u003e56\u003c/sup\u003eFe are given as two standard errors (2SE).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eBulk-rock\u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe values do not directly provide the composition of kimberlite melts because kimberlites are mixtures of magmatic and mantle-derived xenocrystic components, dominantly olivine. Variation in olivine abundances generate variable MgO contents in bulk kimberlite samples worldwide \u003csup\u003e48,\u003c/sup\u003e \u003csup\u003e94\u003c/sup\u003e. Bulk-kimberlite MgO contents hence reflect both the amount and composition of olivine. Assuming primary kimberlite melts derive from similar convective mantle sources globally\u003csup\u003e14\u003c/sup\u003e and, similar to the isotopes of another major element like oxygen \u003csup\u003e28, 29\u003c/sup\u003e, have a relatively restricted range of Fe isotopes, the inverse correlation between bulk-rock\u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe and MgO in \u003cstrong\u003eFig. S8\u003c/strong\u003e reflects the variable contribution of xenocrystic olivine with variable MgO contents and Fe isotope composition.\u003c/p\u003e\n\u003cp\u003eWe estimate the average\u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe of entrained Fe-rich (Mg# \u0026lt;89) olivine xenocrysts in the three samples (Karowe AK-6S, Letseng KB14-01, Wesselton WA-1) with elevated Fe isotopes for which olivine Mg# data are available \u003csup\u003e9, 86,\u003c/sup\u003e \u003csup\u003e95\u003c/sup\u003e. The goal is to estimate the Fe isotopic composition of the Fe-rich deep lithosphere which might represent the source of CLIPPIR diamonds (at least at Karowe and Letseng). For these calculations we assume the following relationship:\u003c/p\u003e\n\u003cp\u003ed\u003csup\u003e56\u003c/sup\u003eFe\u003csub\u003ebulk\u003c/sub\u003e = F\u003csub\u003emelt\u003c/sub\u003e \u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe\u003csub\u003emelt\u003c/sub\u003e + F\u003csub\u003eMg-Ol\u003c/sub\u003e \u0026nbsp;\u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe\u003csub\u003eMg-Ol\u003c/sub\u003e + F\u003csub\u003eFe-Ol\u003c/sub\u003e \u0026nbsp;\u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe\u003csub\u003eFe-Ol\u003c/sub\u003e\u003c/p\u003e\n\u003cp\u003ewhere F is the fraction of each component and \u003cem\u003emelt\u003c/em\u003e, \u003cem\u003eMg-Ol\u003c/em\u003e and \u003cem\u003eFe-Ol\u003c/em\u003e refer to magmatic components, Mg-rich xenocrystic olivine (Mg# \u0026gt;89) and Fe-rich xenocrystic olivine (Mg# \u0026lt;89), respectively. The magmatic olivine rims are included in the melt fraction, whereas the abundance of other xenocrystic material is considered negligible, in line with petrographic observations of \u0026lt;5 vol% of garnet, pyroxene, ilmenite, mica and crustal xenocrysts. The fraction of total olivine (F\u003csub\u003eOl\u003c/sub\u003e) is provided by modal analyses adjusted for the density difference between olivine and bulk kimberlite. It follows that:\u003c/p\u003e\n\u003cp\u003eF\u003csub\u003emelt\u003c/sub\u003e = 1 \u0026ndash; F\u003csub\u003eOl\u003c/sub\u003e\u003c/p\u003e\n\u003cp\u003eF\u003csub\u003eMg-Ol\u003c/sub\u003e = F\u003csub\u003eOl\u003c/sub\u003e \u0026nbsp;\u0026nbsp;X\u003csub\u003eMg-Ol\u003c/sub\u003e\u003c/p\u003e\n\u003cp\u003eF\u003csub\u003eFe-Ol\u003c/sub\u003e = F\u003csub\u003eOl\u003c/sub\u003e \u0026nbsp;\u0026nbsp;X\u003csub\u003eFe-Ol\u003c/sub\u003e\u003c/p\u003e\n\u003cp\u003ewhere X\u003csub\u003eMg-Ol\u003c/sub\u003e and X\u003csub\u003eFe-Ol\u0026nbsp;\u003c/sub\u003eare the relative abundances of Mg-rich and Fe-rich olivine based on published EPMA analyses. Combining the previous equations and resolving for the Fe isotope composition of Fe-rich olivine, we obtain the following expression:\u003c/p\u003e\n\u003cp\u003e\u0026nbsp; \u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe\u003csub\u003eFe-Ol\u003c/sub\u003e =\u0026nbsp;\u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe\u003csub\u003ebulk\u003c/sub\u003e \u0026ndash; (1 \u0026ndash; F\u003csub\u003eOl\u0026nbsp;\u003c/sub\u003e d\u003csup\u003e56\u003c/sup\u003eFe\u003csub\u003emelt\u003c/sub\u003e) + (F\u003csub\u003eOl\u003c/sub\u003e \u0026nbsp;\u0026nbsp;X\u003csub\u003eFe-Ol\u003c/sub\u003e)\u0026nbsp;\u0026nbsp;\u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe\u003csub\u003eMg-Ol\u003c/sub\u003e)] / (F\u003csub\u003eOl\u003c/sub\u003e \u0026nbsp;\u0026nbsp;X\u003csub\u003eFe-Ol\u003c/sub\u003e)\u003c/p\u003e\n\u003cp\u003ewhere\u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe\u003csub\u003eMg-Ol\u003c/sub\u003e = 0.030\u0026permil; based on the typical composition of olivine in refractory mantle peridotites \u003csup\u003e34,\u003c/sup\u003e \u003csup\u003e35, 36\u003c/sup\u003e, and\u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe\u003csub\u003emelt\u003c/sub\u003e = 0.092\u0026permil; based on the correlation between olivine rim Mg# and bulk-rock\u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe (\u003cstrong\u003eFigure 3c\u003c/strong\u003e) and assuming primary olivine Mg# of ~89. The role of serpentinisation is not addressed, but considering the limited alteration of these samples, is considered to have a minor effect in the final calculations.\u003c/p\u003e\n\u003cp\u003eThe results of these calculations are tabulated in \u003cstrong\u003eDataset 3\u003c/strong\u003e and show that the average\u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe of Fe-rich olivine from Karowe is 0.11\u0026permil; while those for Letseng and Wesselton are substantially higher (0.22\u0026permil; and 0.30\u0026permil;). However, Fe-rich olivine predominantly derives from interaction between early kimberlite liquids in equilibrium with olivine Mg ~89 and an Fe-rich substrate, and olivine preferentially partition the light isotopes of Fe compared to coexisting silicate minerals \u003csup\u003e34, 35\u003c/sup\u003e. Therefore, it appears likely that the Fe-rich deep lithosphere before kimberlite metasomatism had even heavier although unconstrained Fe isotope composition. It is possible that this lithospheric substrate locally approached the very high\u0026nbsp;d\u003csup\u003e56\u003c/sup\u003eFe (0.79-0.90\u0026permil;) observed in the Fe-rich metallic inclusions hosted by CLIPPIR diamonds\u003csup\u003e6\u003c/sup\u003e, an hypothesis which requires further scrutiny using natural samples including megacrysts and eclogite xenoliths.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMost samples for this study were source from the John J. Gurney Mantle collection housed at the University of Cape Town with additional samples collected by AG with thanks to Petra diamonds and Rio Tinto. GHH acknowledges funding from the DSI-NRF Centre of Excellence (CoE) for Integrated Mineral and Energy Resource Analysis (DSI-NRF CIMERA) towards this research is hereby acknowledged. Opinions expressed and conclusions arrived at, are those of the author(s) and are not necessarily to be attributed to the CoE. Christian Reinke is thanked for help with microprobe analyses at the University of Johannesburg. Riana Rossouw is thanked for help with laser ablation analyses at the University of Stellenbosch. Discussion with several colleagues including Peng Ni, Graham Pearson, Steve Shirey, Evan Smith, Thomas Stachel, Suzette Timmermann, Mike Walter and Qiwei Zhang substantially improved the contents of this work even though they do not necessarily reflect the opinions of these colleagues.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eG.H.H and A.G. conceptualized the project and wrote the original draft. G.H.H., A.G., and M.M.T. collected EPMA and laser ablation data for olivine. J.L. and R.C. were responsible for Fe isotope data collection. G.H.H. drafted all figures. G.H.H. and M.M.T. compiled online datasets. All authors contributed to the writing, editing, and reviewing the manuscript drafts.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eSmith, E.M., Shirey, S.B., Nestola, F., Bullock, E.S., Wang, J., Richardson, S.H., Wang, W., 2016. Large gem diamonds from metallic liquid in Earth\u0026rsquo;s deep mantle. Science 354(6318), 1403-1405. \u003c/li\u003e\n\u003cli\u003eMilledge, H.J., Mendelssohn, M.J., Seal, M., Rouse, J.E., Swart, P.K., Pillinger, C.T., 1983. Carbon isotopic variation in spectral type II diamonds. Nature 303(5920), 791-792. doi.org/10.1038/303791a0. \u003c/li\u003e\n\u003cli\u003eBanas, A., Stachel, T., Stern, R.A., Allan, A., Freeman, L., 2017. Can microdiamonds be used to predict the distribution of large Type IIa macrodiamonds? A case study at the Letseng mine. International Kimberlite Conference: Extended Abstracts 11. doi.org/10.29173/ikc4017. \u003c/li\u003e\n\u003cli\u003eMoore, A., Helmstaedt, H., 2023. Origin of framesite revisited: Possible implications for the formation of CLIPPIR diamonds. Earth-Science Reviews 241, 104434. doi.org/https://doi.org/10.1016/j.earscirev.2023.104434. \u003c/li\u003e\n\u003cli\u003eBowen, D.C., Ferraris, R.D., Palmer, C.E., Ward, J.D., 2009. On the unusual characteristics of the diamonds from Let\u0026scaron;eng-la-Terae kimberlites, Lesotho. Lithos 112, 767-774. doi.org/https://doi.org/10.1016/j.lithos.2009.04.026. \u003c/li\u003e\n\u003cli\u003eSmith, E.M., Ni, P., Shirey, S.B., Richardson, S.H., Wang, W., Shahar, A., 2021. Heavy iron in large gem diamonds traces deep subduction of serpentinized ocean floor. Science Advances 7(14), eabe9773. doi.org/10.1126/sciadv.abe9773. \u003c/li\u003e\n\u003cli\u003eStachel, T., Cartigny, P., Chacko, T., \u0026amp; Pearson, D. G. 2022. Carbon and nitrogen in mantle-derived diamonds. \u003cem\u003eReviews in Mineralogy and Geochemistry\u003c/em\u003e, \u003cem\u003e88\u003c/em\u003e(1), 809-875.\u003c/li\u003e\n\u003cli\u003eTimmerman, S., Stachel, T., Koornneef, J.M., Smit, K.V., Harlou, R., Nowell, G.M., Thomson, A.R., Kohn, S.C., Davies, J.H.F.L., Davies, G.R., Krebs, M.Y., Zhang, Q., Milne, S.E.M., Harris, J.W., Kaminsky, F., Zedgenizov, D., Bulanova, G., Smith, C.B., Cabral Neto, I., Silveira, F.V., Burnham, A.D., Nestola, F., Shirey, S.B., Walter, M.J., Steele, A., Pearson, D.G., 2023. Sublithospheric diamond ages and the supercontinent cycle. Nature 623(7988), 752-756. doi.org/10.1038/s41586-023-06662-9. \u003c/li\u003e\n\u003cli\u003eGiuliani, A., Schmidt, M.W., Torsvik, T.H., Fedortchouk, Y., 2023. Genesis and evolution of kimberlites. Nature Reviews Earth \u0026amp; Environment 4(11), 738-753. doi.org/10.1038/s43017-023-00481-2. \u003c/li\u003e\n\u003cli\u003eMoore, A.E., 2009. Type II diamonds: Flamboyant Megacrysts? South African Journal of Geology 112(1), 23-38. doi.org/10.2113/gssajg.112.1.23. \u003c/li\u003e\n\u003cli\u003eSolomatov, V.S., Reese, C.C., 2008. Grain size variations in the Earth\u0026apos;s mantle and the evolution of primordial chemical heterogeneities. Journal of Geophysical Research: Solid Earth 113(B7). doi.org/https://doi.org/10.1029/2007JB005319. \u003c/li\u003e\n\u003cli\u003eZhang, Q., Hardman, M. F., Stachel, T., Chinn, I., Seller, M., Kjarsgaard, B., \u0026amp; Pearson, D. G. (2025). A machine learning approach to single garnet geothermometry and application to tracing the fingerprint of superdeep diamonds. \u003cem\u003eGeochemistry, Geophysics, Geosystems\u003c/em\u003e, \u003cem\u003e26\u003c/em\u003e(3), e2024GC012124.\u003c/li\u003e\n\u003cli\u003eZhang, Q., Malakhova, I., Stachel, T., Elliott, B., Chinn, I., Seller, M., ... \u0026amp; Pearson, D. G. (2025). A link between enhanced sampling of Ti\u0026ndash;rich mantle garnets and superdeep diamonds: Insights from the DO-27 kimberlite, Northern Canada. \u003cem\u003eMineralogy and Petrology\u003c/em\u003e, 1-17.\u003c/li\u003e\n\u003cli\u003eGiuliani, A., Pearson, D.G., Soltys, A., Dalton, H., Phillips, D., Foley, S.F., Lim, E., Goemann, K., Griffin, W.L., Mitchell, R.H., 2020. Kimberlite genesis from a common carbonate-rich primary melt modified by lithospheric mantle assimilation. Science Advances 6(17), eaaz0424. doi.org/10.1126/sciadv.aaz0424. \u003c/li\u003e\n\u003cli\u003eGiuliani, A., Phillips, D., Pearson, D.G., Sarkar, S., M\u0026uuml;ller, A.A., Weiss, Y., Preston, R., Seller, M., Spetsius, Z., 2023. Diamond preservation in the lithospheric mantle recorded by olivine in kimberlites. Nature Communications 14(1), 6999. doi.org/10.1038/s41467-023-42888-x. \u003c/li\u003e\n\u003cli\u003eHowarth, G.H., Giuliani, A., Soltys, A., Bussweiler, Y., 2022. Compositional Variations in Primitive Kimberlite Melts and Entrained Mantle Cargo from a Global Survey of Trace Element Compositions in Kimberlite Olivine. Journal of Petrology 63(8), egac062. doi.org/10.1093/petrology/egac062. \u003c/li\u003e\n\u003cli\u003eNowell, G.M., Pearson, D.G., Bell, D.R., Carlson, R.W., Smith, C.B., Kempton, P.D., Noble, S.R., 2004. Hf Isotope Systematics of Kimberlites and their Megacrysts: New Constraints on their Source Regions. Journal of Petrology 45(8), 1583-1612. \u003c/li\u003e\n\u003cli\u003eWoodhead, J., Hergt, J., Giuliani, A., Phillips, D., Maas, R., 2017. Tracking continental-scale modification of the Earth\u0026rsquo;s mantle using zircon megacrysts. Geochemical Perspectives Letters 4, 1-6. doi.org/http://dx.doi.org/10.7185/geochemlet.1727. \u003c/li\u003e\n\u003cli\u003eMoore, A., Belousova, E., 2005. Crystallization of Cr-poor and Cr-rich megacryst suites from the host kimberlite magma: implications for mantle structure and the generation of kimberlite magmas. Contrib Mineral Petrol 149(4), 462-481. \u003c/li\u003e\n\u003cli\u003eNkere, B.J., Janney, P.E., Tinguely, C., 2021. Cr-poor and Cr-rich clinopyroxene and garnet megacrysts from southern African Group 1 and Group 2 kimberlites: Clues to megacryst origins and their relationship to kimberlites. Lithos 396-397, 106231. doi.org/https://doi.org/10.1016/j.lithos.2021.106231. \u003c/li\u003e\n\u003cli\u003eVeglio, C., Lawley, C.J.M., Pearson, D.G., Kjarsgaard, B.A., Petts, D.C., Jackson, S.E., 2022. Olivine xenocrysts reveal carbonated mid-lithosphere in the northern Slave craton. Lithos 414-415, 106633. doi.org/https://doi.org/10.1016/j.lithos.2022.106633. \u003c/li\u003e\n\u003cli\u003eTau, M. M., Howarth, G. H., \u0026amp; Giuliani, A. (2024, July). Lithospheric mantle sampling of kimberlites and lamproites using Al-in-olivine thermometry. In \u003cem\u003eInternational Kimberlite Conference: Extended Abstracts\u003c/em\u003e (Vol. 12).\u003c/li\u003e\n\u003cli\u003eBussweiler, Y., Brey, G.P., Pearson, D.G., Stachel, T., Stern, R.A., Hardman, M.F., Kjarsgaard, B.A., Jackson, S.E., 2017. The aluminum-in-olivine thermometer for mantle peridotites \u0026mdash; Experimental versus empirical calibration and potential applications. Lithos 272\u0026ndash;273, 301-314. doi.org/http://dx.doi.org/10.1016/j.lithos.2016.12.015. \u003c/li\u003e\n\u003cli\u003eBoyd, F.R., Nixon, P.H., 1975. Origins of the ultramafic nodules from some kimberlites of northern Lesotho and the Monastery Mine, South Africa. Physics and Chemistry of the Earth 9, 431-454. doi.org/http://dx.doi.org/10.1016/0079-1946(75)90032-4. \u003c/li\u003e\n\u003cli\u003eGurney, J.J., Jakob, W.R.O., Dawson, J.B., 1979. Megacrysts from the Monastery kimberlite pipe, in: Boyd, F.R., Meyer, H.O.A. (Eds.), The mantle sample. 2nd International Kimberlite Conference. American Geophysical Union, Washington, DC, pp. 227-243.\u003c/li\u003e\n\u003cli\u003eTappe, S., Massuyeau, M., Smart, K.A., Woodland, A.B., Gussone, N., Milne, S., Stracke, A., 2021. Sheared Peridotite and Megacryst Formation Beneath the Kaapvaal Craton: a Snapshot of Tectonomagmatic Processes across the Lithosphere\u0026ndash;Asthenosphere Transition. Journal of Petrology 62(8). doi.org/10.1093/petrology/egab046. \u003c/li\u003e\n\u003cli\u003eHeckel, C., Woodland, A.B., Linckens, J., Gibson, S.A., Seitz, H.-M., 2023. Sheared Peridotites from Northern Lesotho: Metasomatism-Induced Deformation and Craton Destabilization. Journal of Petrology 64(10), egad076. doi.org/10.1093/petrology/egad076. \u003c/li\u003e\n\u003cli\u003eGiuliani, A., Martin, L.A.J., Soltys, A., Griffin, W.L., 2019. Mantle-like oxygen isotopes in kimberlites determined by in situ SIMS analyses of zoned olivine. Geochimica et Cosmochimica Acta 266, 274-291. doi.org/https://doi.org/10.1016/j.gca.2019.03.032. \u003c/li\u003e\n\u003cli\u003eXu, J.-Y., Giuliani, A., Li, Q.-L., Lu, K., Melgarejo, J.C., Griffin, W.L., 2021. Light oxygen isotopes in mantle-derived magmas reflect assimilation of sub-continental lithospheric mantle material. Nature Communications 12(1), 6295. doi.org/10.1038/s41467-021-26668-z. \u003c/li\u003e\n\u003cli\u003evan Blerk, J., Howarth, G.H., Harris, C., Janney, P.E., 2025. Oxygen isotope constraints on proto-kimberlite melt modification through assimilation of low \u0026delta;18O recycled crust in the deep lithosphere. Geochimica et Cosmochimica Acta. doi.org/https://doi.org/10.1016/j.gca.2024.12.028. \u003c/li\u003e\n\u003cli\u003eMattey, D., Lowry, D., Macpherson, C., 1994. Oxygen isotope composition of mantle peridotite. Earth and Planetary Science Letters 128(3\u0026sbquo;\u0026Auml;\u0026igrave;4), 231-241. doi.org/http://dx.doi.org/10.1016/0012-821X(94)90147-3. \u003c/li\u003e\n\u003cli\u003eTaylor, H.P., Sheppard, S.M.F., 1986. Igneous rocks I: processes of isotopic fractionation and isotope systematics. Reviews in Mineralogy 16, 227-272. \u003c/li\u003e\n\u003cli\u003eWenner, D.B., Taylor Jr, H.P., 1974. D/H and O18/O16 studies of serpentinization of ultramaflc rocks. Geochimica et Cosmochimica Acta 38(8), 1255-1286. doi.org/http://dx.doi.org/10.1016/0016-7037(74)90120-3. \u003c/li\u003e\n\u003cli\u003eSossi, P.A., Nebel, O., Foden, J., 2016. Iron isotope systematics in planetary reservoirs. Earth and Planetary Science Letters 452, 295-308. doi.org/http://dx.doi.org/10.1016/j.epsl.2016.07.032. \u003c/li\u003e\n\u003cli\u003eHuang, F., Zhang, Z., Lundstrom, C.C., Zhi, X., 2011. Iron and magnesium isotopic compositions of peridotite xenoliths from Eastern China. Geochimica et Cosmochimica Acta 75(12), 3318-3334. doi.org/https://doi.org/10.1016/j.gca.2011.03.036. \u003c/li\u003e\n\u003cli\u003eAn, Y., Huang, J.-X., Griffin, W.L., Liu, C., Huang, F., 2017. Isotopic composition of Mg and Fe in garnet peridotites from the Kaapvaal and Siberian cratons. Geochimica et Cosmochimica Acta 200, 167-185. doi.org/http://dx.doi.org/10.1016/j.gca.2016.11.041. \u003c/li\u003e\n\u003cli\u003eRouxel, O., Dobbek, N., Ludden, J., Fouquet, Y., 2003. Iron isotope fractionation during oceanic crust alteration. Chemical Geology 202(1), 155-182. doi.org/https://doi.org/10.1016/j.chemgeo.2003.08.011. \u003c/li\u003e\n\u003cli\u003eWilliams, H.M., Bizimis, M., 2014. Iron isotope tracing of mantle heterogeneity within the source regions of oceanic basalts. Earth and Planetary Science Letters 404, 396-407. doi.org/https://doi.org/10.1016/j.epsl.2014.07.033. \u003c/li\u003e\n\u003cli\u003eRadu, I.-B., Harris, C., Moine, B.N., Costin, G., Cottin, J.-Y., 2019. Subduction relics in the subcontinental lithospheric mantle evidence from variation in the \u0026delta;18O value of eclogite xenoliths from the Kaapvaal craton. Contrib Mineral Petrol 174(3), 19. doi.org/10.1007/s00410-019-1552-z. \u003c/li\u003e\n\u003cli\u003eShu, Q., Brey, G.P., Hoefer, H.E., Zhao, Z., Pearson, D.G., 2016. Kyanite/corundum eclogites from the Kaapvaal Craton: subducted troctolites and layered gabbros from the Mid- to Early Archean. Contrib Mineral Petrol 171(2), 1-24. doi.org/10.1007/s00410-015-1225-5. \u003c/li\u003e\n\u003cli\u003eKorolev, N.M., Melnik, A.E., Li, X.-H., Skublov, S.G., 2018. The oxygen isotope composition of mantle eclogites as a proxy of their origin and evolution: A review. Earth-Science Reviews 185, 288-300. doi.org/https://doi.org/10.1016/j.earscirev.2018.06.007. \u003c/li\u003e\n\u003cli\u003eWilliams, H.M., Nielsen, S.G., Renac, C., Griffin, W.L., O\u0026apos;Reilly, S.Y., McCammon, C.A., Pearson, N., Viljoen, F., Alt, J.C., Halliday, A.N., 2009. Fractionation of oxygen and iron isotopes by partial melting processes: Implications for the interpretation of stable isotope signatures in mafic rocks. Earth and Planetary Science Letters 283(1), 156-166. doi.org/https://doi.org/10.1016/j.epsl.2009.04.011. \u003c/li\u003e\n\u003cli\u003eRehfeldt, T., Foley, S.F., Jacob, D.E., Carlson, R.W., Lowry, D., 2008. Contrasting types of metasomatism in dunite, wehrlite and websterite xenoliths from Kimberley, South Africa. Geochimica et Cosmochimica Acta 72(23), 5722-5756. \u003c/li\u003e\n\u003cli\u003eRehfeldt, T., Jacob, D.E., Carlson, R.W., Foley, S.F., 2007. Fe-rich Dunite Xenoliths from South African Kimberlites: Cumulates from Karoo Flood Basalts. Journal of Petrology 48(7), 1387-1409. doi.org/\u003cu\u003e10.1093/petrology/egm023\u003c/u\u003e. \u003c/li\u003e\n\u003cli\u003eArndt, N., Cordier, C., Boullier, A.-M., Batanova, V., Magnin, V., Findling, N., 2022. Olivine in Kimberlites and Related Rocks \u0026ndash; Macrocrysts, Dunitic Nodules and Megacrysts from the Same Metasomatized Source. Journal of Petrology 63(8), egac063. doi.org/10.1093/petrology/egac063. \u003c/li\u003e\n\u003cli\u003eSmith, C.B., 1983. Pb, Sr and Nd isotopic evidence for sources of southern African Cretaceous kimberlites. Nature 304, 51-54. \u003c/li\u003e\n\u003cli\u003eWoodhead, J., Hergt, J., Giuliani, A., Maas, R., Phillips, D., Pearson, D.G., Nowell, G., 2019. Kimberlites reveal 2.5-billion-year evolution of a deep, isolated mantle reservoir. Nature 573(7775), 578-581. doi.org/10.1038/s41586-019-1574-8. \u003c/li\u003e\n\u003cli\u003eGiuliani, A., Dalton, H., Pearson, D.G., 2025. Kimberlites: The deepest geochemical probes of Earth, in: Anbar, A., Weis, D. (Eds.), Treatise on Geochemistry (Third edition). Elsevier, Oxford, pp. 159-230. doi.org/https://doi.org/10.1016/B978-0-323-99762-1.00064-4.\u003c/li\u003e\n\u003cli\u003eMercier, J.-C.C., 1979. Peridotite Xenoliths and the Dynamics of Kimberlite Intrusion, The Mantle Sample: Inclusion in Kimberlites and Other Volcanics. American Geophysical Union, pp. 197-212. doi.org/10.1029/SP016p0197.\u003c/li\u003e\n\u003cli\u003eHeckel, C., Woodland, A. B., \u0026amp; Gibson, S. A. (2025). Cretaceous thinning of the Kaapvaal craton and diamond resorption: Key insights from a highly-deformed and metasomatized ilmenite-dunite xenolith. \u003cem\u003eLithos\u003c/em\u003e, \u003cem\u003e494\u003c/em\u003e, 107901.\u003c/li\u003e\n\u003cli\u003eHowarth, G.H., Giuliani, A., Bussweiler, Y., Tau, M., Hashibi, S., Janney, P.E., Nowicki, T.E., 2024. Kimberlite pre-conditioning of the lithospheric mantle and implications for diamond survival: a case study of olivine and mantle xenocrysts from the Koidu mine (Sierra Leone). Mineralium Deposita. doi.org/10.1007/s00126-024-01312-0. \u003c/li\u003e\n\u003cli\u003eTappert, R., Stachel, T., Harris, J.W., Muehlenbachs, K., Ludwig, T., Brey, G.P., 2005. Subducting oceanic crust: The source of deep diamonds. Geology 33(7), 565-568. \u003c/li\u003e\n\u003cli\u003ePalot, M., Pearson, D.G., Stachel, T., Stern, R.A., Le Pioufle, A., Gurney, J.J., Harris, J.W., 2017. The transition zone as a host for recycled volatiles: Evidence from nitrogen and carbon isotopes in ultra-deep diamonds from Monastery and Jagersfontein (South Africa). Chemical Geology 466, 733-749. doi.org/https://doi.org/10.1016/j.chemgeo.2017.07.023. \u003c/li\u003e\n\u003cli\u003eNestola, F., Korolev, N., Kopylova, M., Rotiroti, N., Pearson, D.G., Pamato, M.G., Alvaro, M., Peruzzo, L., Gurney, J.J., Moore, A.E., Davidson, J., 2018. CaSiO3 perovskite in diamond indicates the recycling of oceanic crust into the lower mantle. Nature 555, 237-240. doi.org/10.1038/nature25972. \u003c/li\u003e\n\u003cli\u003eShirey, S.B., Pearson, D.G., Stachel, T., Walter, M.J., 2024. Sublithospheric Diamonds: Plate Tectonics from Earth\u0026rsquo;s Deepest Mantle Samples. Annual Review of Earth and Planetary Sciences 52(Volume 52, 2024), 249-293. doi.org/https://doi.org/10.1146/annurev-earth-032320-105438. \u003c/li\u003e\n\u003cli\u003eHarte, B., Cayzer, N., 2007. Decompression and unmixing of crystals included in diamonds from the mantle transition zone. Physics and Chemistry of Minerals 34(9), 647-656. doi.org/10.1007/s00269-007-0178-2. \u003c/li\u003e\n\u003cli\u003eGriffin, W.L., O\u0026apos;Reilly, S.Y., 2007. Cratonic lithospheric mantle: Is anything subducted? Episodes 30(1), 43-53. \u003c/li\u003e\n\u003cli\u003eHardman, M.F., Stachel, T., Pearson, D.G., Stern, R.A., Lai, M.Y., 2025. Constraints on the distribution, composition, and lithology of eclogite and pyroxenite from the central Slave craton. Mineralogy and Petrology. doi.org/10.1007/s00710-024-00884-z. \u003c/li\u003e\n\u003cli\u003ePerchuk, A. L., Gerya, T. V., Zakharov, V. S., \u0026amp; Griffin, W. L. (2020). Building cratonic keels in Precambrian plate tectonics. \u003cem\u003eNature\u003c/em\u003e, \u003cem\u003e586\u003c/em\u003e(7829), 395-401.\u003c/li\u003e\n\u003cli\u003eFarnetani, C.G., Legras, B., Tackley, P.J., 2002. Mixing and deformations in mantle plumes. Earth and Planetary Science Letters 196(1\u0026ndash;2), 1-15. doi.org/http://dx.doi.org/10.1016/S0012-821X(01)00597-0. \u003c/li\u003e\n\u003cli\u003eDongmo Wamba, M., Montagner, J.-P., Romanowicz, B., 2023. Imaging deep-mantle plumbing beneath La R\u0026eacute;union and Comores hot spots: Vertical plume conduits and horizontal ponding zones. Science Advances 9(4), eade3723. doi.org/10.1126/sciadv.ade3723. \u003c/li\u003e\n\u003cli\u003eMunch, F.D., Romanowicz, B., Mukhopadhyay, S., Rudolph, M.L., 2024. Deep mantle plumes feeding periodic alignments of asthenospheric fingers beneath the central and southern Atlantic Ocean. Proceedings of the National Academy of Sciences 121(46), e2407543121. doi.org/10.1073/pnas.2407543121. \u003c/li\u003e\n\u003cli\u003eDoucet, L. S., Li, Z. X., \u0026amp; Gamal El Dien, H. (2021). Oceanic and super-deep continental diamonds share a transition zone origin and mantle plume transportation. \u003cem\u003eScientific Reports\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(1), 16958.\u003c/li\u003e\n\u003cli\u003eSobolev, A.V., Asafov, E.V., Gurenko, A.A., Arndt, N.T., Batanova, V.G., Portnyagin, M.V., Garbe-Sch\u0026ouml;nberg, D., Krasheninnikov, S.P., 2016. Komatiites reveal a hydrous Archaean deep-mantle reservoir. Nature 531(7596), 628-632. doi.org/10.1038/nature17152\u003c/li\u003e\n\u003cli\u003eAshwal, L.D., 2021. Sub-lithospheric mantle sources for overlapping southern African Large Igneous Provinces. South African Journal of Geology 124(2), 421-442. doi.org/10.25131/sajg.124.0023. \u003c/li\u003e\n\u003cli\u003eHoal, K.O., 2003. Samples of Proterozoic iron-enriched mantle from the Premier kimberlite. Lithos 71(2-4), 259-272. \u003c/li\u003e\n\u003cli\u003eFouch, M. J., Silver, P. G., Bell, D. R., \u0026amp; Lee, J. N. (2004). Small-scale variations in seismic anisotropy near Kimberley, South Africa. \u003cem\u003eGeophysical Journal International\u003c/em\u003e, \u003cem\u003e157\u003c/em\u003e(2), 764-774.\u003c/li\u003e\n\u003cli\u003eZhang, Q., Morel, M.L.A., Liu, J., Legros, H., Luguet, A., Viljoen, K.S., Davies, G.R., Pearson, D.G., 2022. Re-healing cratonic mantle lithosphere after the world\u0026apos;s largest igneous intrusion: Constraints from peridotites erupted by the Premier kimberlite, South Africa. Earth and Planetary Science Letters 598, 117838. doi.org/https://doi.org/10.1016/j.epsl.2022.117838. \u003c/li\u003e\n\u003cli\u003eLinzmeyer, T., Day, J.M.D., Alonso-Perez, R., Phahla, T., Wolmarans, A., Rogers, A., 2024. A 187Re-187Os and highly siderophile element study of diamondiferous kimberlite melt-mantle interactions and the inferred age of continental lithosphere. Geochimica et Cosmochimica Acta 386, 157-172. doi.org/https://doi.org/10.1016/j.gca.2024.09.026. \u003c/li\u003e\n\u003cli\u003eAulbach, S., 2012. Craton nucleation and formation of thick lithospheric roots. Lithos 149(0), 16-30. \u003c/li\u003e\n\u003cli\u003eLiu, J., Pearson, D.G., Wang, L.H., Mather, K.A., Kjarsgaard, B.A., Schaeffer, A.J., Irvine, G.J., Kopylova, M.G., Armstrong, J.P., 2021. Plume-driven recratonization of deep continental lithospheric mantle. Nature 592(7856), 732-736. doi.org/10.1038/s41586-021-03395-5. \u003c/li\u003e\n\u003cli\u003eTimmerman, S., Honda, M., Burnham, A.D., Amelin, Y., Woodland, S., Pearson, D.G., Jaques, A.L., Le Losq, C., Bennett, V.C., Bulanova, G.P., Smith, C.B., Harris, J.W., Tohver, E., 2019. Primordial and recycled helium isotope signatures in the mantle transition zone. Science 365(6454), 692. doi.org/10.1126/science.aax5293. \u003c/li\u003e\n\u003cli\u003eErnst, R.E., Davies, D.R., Jowitt, S.M., Campbell, I.H., 2018. When do mantle plumes destroy diamonds? Earth and Planetary Science Letters 502, 244-252. doi.org/https://doi.org/10.1016/j.epsl.2018.08.058. \u003c/li\u003e\n\u003cli\u003eHowell, D., Stachel, T., Stern, R.A., Pearson, D.G., Nestola, F., Hardman, M.F., Harris, J.W., Jaques, A.L., Shirey, S.B., Cartigny, P., Smit, K.V., Aulbach, S., Brenker, F.E., Jacob, D.E., Thomassot, E., Walter, M.J., Navon, O., 2020. Deep carbon through time: Earth\u0026rsquo;s diamond record and its implications for carbon cycling and fluid speciation in the mantle. Geochimica et Cosmochimica Acta 275, 99-122. doi.org/https://doi.org/10.1016/j.gca.2020.02.011. \u003c/li\u003e\n\u003cli\u003eChinn, I., Krug, M., Minnie, W., Rikhotso, C., 2008. Relationships between diamond populations and geology from the AK06 kimberlite, Botswana. International Kimberlite Conference: Extended Abstracts 9(0). doi.org/10.29173/ikc3486. \u003c/li\u003e\n\u003cli\u003eLim, E., Giuliani, A., Phillips, D., Goemann, K., 2018. Origin of complex zoning in olivine from diverse, diamondiferous kimberlites and tectonic settings: Ekati (Canada), Alto Paranaiba (Brazil) and Kaalvallei (South Africa). Mineralogy and Petrology 112(2), 539-554. doi.org/10.1007/s00710-018-0607-6. \u003c/li\u003e\n\u003cli\u003eAraujo, A., Carlson, R.W., Gaspar, J.C., Bizzi, L.A., 2001. Petrology of kamafugites and kimberlites from the Alto Parana\u0026iacute;ba Alkaline Province, Minas Gerais, Brazil. Contrib Mineral Petrol 142(2), 163-177. doi.org/10.1007/s004100100280. \u003c/li\u003e\n\u003cli\u003eWang, X.-C., Wilde, S.A., Li, Q.-L., Yang, Y.-N., 2015. Continental flood basalts derived from the hydrous mantle transition zone. Nature Communications 6, 7700. doi.org/10.1038/ncomms8700. \u003c/li\u003e\n\u003cli\u003eHe, Y., Meng, X., Ke, S., Wu, H., Zhu, C., Teng, F.-Z., Hoefs, J., Huang, J., Yang, W., Xu, L., Hou, Z., Ren, Z.-Y., Li, S., 2019. A nephelinitic component with unusual \u0026delta;56Fe in Cenozoic basalts from eastern China and its implications for deep oxygen cycle. Earth and Planetary Science Letters 512, 175-183. doi.org/https://doi.org/10.1016/j.epsl.2019.02.009. \u003c/li\u003e\n\u003cli\u003eWu, T.-H., Liu, S.-A., 2024. Heavy iron isotopes reveal mantle oxidation in addition to pyroxenite sources for intraplate basalts. Chemical Geology 670, 122432. doi.org/https://doi.org/10.1016/j.chemgeo.2024.122432. \u003c/li\u003e\n\u003cli\u003eKostrovitskiy, S.I., Solov\u0026rsquo;eva, L.V., Gornova, M.A., Alymova, N.V., Yakovlev, D.A., Ignat\u0026rsquo;ev, A.V., Velivetskaya, T.A., Suvorova, L.F., 2012. Oxygen isotope composition in minerals of mantle parageneses from Yakutian kimberlites. Doklady Earth Sciences 444(1), 579-584. doi.org/10.1134/S1028334X12050030. \u003c/li\u003e\n\u003cli\u003eJochum, K.P., Nohl, U., Herwig, K., Lammel, E., Stoll, B., Hofmann, A.W., 2005. GeoReM: A New Geochemical Database for Reference Materials and Isotopic Standards. Geostandards and Geoanalytical Research 29(3), 333-338. doi.org/10.1111/j.1751-908X.2005.tb00904.x. \u003c/li\u003e\n\u003cli\u003eBussweiler, Y., Giuliani, A., Greig, A., Kjarsgaard, B.A., Petts, D., Jackson, S.E., Barrett, N., Luo, Y., Pearson, D.G., 2019. Trace element analysis of high-Mg olivine by LA-ICP-MS \u0026ndash; Characterization of natural olivine standards for matrix-matched calibration and application to mantle peridotites. Chemical Geology 524, 136-157. doi.org/https://doi.org/10.1016/j.chemgeo.2019.06.019. \u003c/li\u003e\n\u003cli\u003eHowarth, G.H., 2018. Olivine megacryst chemistry, Monastery kimberlite: Constraints on the mineralogy of the HIMU mantle reservoir in southern Africa. Lithos 314-315, 658-668. doi.org/https://doi.org/10.1016/j.lithos.2018.07.007. \u003c/li\u003e\n\u003cli\u003eGiuliani, A., 2018. Insights into kimberlite petrogenesis and mantle metasomatism from a review of the compositional zoning of olivine in kimberlites worldwide. Lithos 312-313, 322-342. doi.org/https://doi.org/10.1016/j.lithos.2018.04.029. \u003c/li\u003e\n\u003cli\u003eSoltys, A., Giuliani, A., Phillips, D., Kamenetsky, V.S., 2020. Kimberlite Metasomatism of the Lithosphere and the Evolution of Olivine in Carbonate-rich Melts \u0026mdash; Evidence from the Kimberley Kimberlites (South Africa). Journal of Petrology 61(6), egaa062. doi.org/10.1093/petrology/egaa062. \u003c/li\u003e\n\u003cli\u003eDe Hoog, J.C.M., Gall, L., Cornell, D.H., 2010. Trace-element geochemistry of mantle olivine and application to mantle petrogenesis and geothermobarometry. Chemical Geology 270(1\u0026sbquo;\u0026Auml;\u0026igrave;4), 196-215. doi.org/http://dx.doi.org/10.1016/j.chemgeo.2009.11.017. \u003c/li\u003e\n\u003cli\u003eMotsamai, T., Harris, J.W., Stachel, T., Pearson, D.G., Armstrong, J., 2018. Mineral inclusions in diamonds from Karowe Mine, Botswana: super-deep sources for super-sized diamonds? Mineralogy and Petrology 112(1), 169-180. doi.org/10.1007/s00710-018-0604-9. \u003c/li\u003e\n\u003cli\u003eLahermaa, P. (2022). Indicator Mineral Chemistry and Diamond Potential of the Liqhobong Kimberlites, Lesotho, Southern Africa. MSc thesis. University of Helsinki, Finland.\u003c/li\u003e\n\u003cli\u003eNimis, P., Preston, R., Perritt, S. H., \u0026amp; Chinn, I. L. (2020). Diamond\u0026apos;s depth distribution systematics. \u003cem\u003eLithos\u003c/em\u003e, \u003cem\u003e376\u003c/em\u003e, 105729.\u003c/li\u003e\n\u003cli\u003eMoore, A.E., Lock, N.P., 2001. The origin of mantle-derived megacrysts and sheared peridotites-evidence from kimberlites in the northern Lesotho - Orange Free State (South Africa) and Botswana pipe clusters. South African Journal of Geology 104(1), 23-38. \u003c/li\u003e\n\u003cli\u003eLi, X., He, Y., Ke, S., Sun, A.Y., Zhang, Y.C., Wang, Y., Yang, R.Y., Wang, P.J., 2022. High-precision Fe Isotope Analysis on MC-ICPMS Using a 57Fe-58Fe Double Spike Technique. Atomic Spectroscopy 43(2), 164-173. \u003c/li\u003e\n\u003cli\u003eChen, K., Bao, Z., Yuan, H., Lv, N., 2022. Direct measurement of Fe isotope compositions in iron-dominated minerals without column chromatography using MC-ICP-MS. Journal of Analytical Atomic Spectrometry 37(2), 249-263. doi.org/10.1039/D1JA00381J. \u003c/li\u003e\n\u003cli\u003ele Roex, A.P., Bell, D.R., Davis, P., 2003. Petrogenesis of Group I Kimberlites from Kimberley, South Africa: Evidence from Bulk-rock Geochemistry. Journal of Petrology 44(12), 2261-2286. doi.org/\u003cu\u003e10.1093/petrology/egg077\u003c/u\u003e. \u003c/li\u003e\n\u003cli\u003eStamm, N., Schmidt, M.W., Szymanowski, D., von Quadt, A., Mohapi, T., Fourie, A., 2018. Primary petrology, mineralogy and age of the Let\u0026scaron;eng-la-Terae kimberlite (Lesotho, Southern Africa) and parental magmas of Group-I kimberlites. Contrib Mineral Petrol 173(9), 76. doi.org/10.1007/s00410-018-1502-1. \u003c/li\u003e\n\u003cli\u003eBussweiler, Y., Foley, S.F., Prelević, D., Jacob, D.E., 2015. The olivine macrocryst problem: New insights from minor and trace element compositions of olivine from Lac de Gras kimberlites, Canada. Lithos 220\u0026ndash;223, 238-252. doi.org/http://dx.doi.org/10.1016/j.lithos.2015.02.016. \u003c/li\u003e\n\u003cli\u003eKatayama, I., Suyama, Y., Ando, J.-i., Komiya, T., 2009. Mineral chemistry and P\u0026ndash;T condition of granular and sheared peridotite xenoliths from Kimberley, South Africa: Origin of the textural variation in the cratonic mantle. Lithos 109(3\u0026ndash;4), 333-340. doi.org/http://dx.doi.org/10.1016/j.lithos.2008.05.004. \u003c/li\u003e\n\u003cli\u003eHeckel, C., Woodland, A.B., Linckens, J., Gibson, S.A., Seitz, H.-M., 2022. Sheared Peridotites from Kimberley (Kaapvaal Craton, RSA): Record of Multiple Metasomatic Events Accompanied with Deformation. Journal of Petrology 63(10), egac096. doi.org/10.1093/petrology/egac096. \u003c/li\u003e\n\u003cli\u003eMenzies, A., Westerlund, K., Gr\u0026uuml;tter, H., Gurney, J., Carlson, J., Fung, A., Nowicki, T., 2004. Peridotitic mantle xenoliths from kimberlites on the Ekati Diamond Mine property, N.W.T., Canada: major element compositions and implications for the lithosphere beneath the central Slave craton. Lithos 77(1\u0026ndash;4), 395-412. doi.org/http://dx.doi.org/10.1016/j.lithos.2004.04.013. \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6808081/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6808081/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Diamonds larger than 100 carats are some of the most valuable gemstones ever unearthed. They are hosted in mantle-derived magmas called kimberlites, but occur at few locations globally. Beyond their large size and rarity, these diamonds exhibit distinctive attributes such as exceptional clarity and irregular shape, leading to the CLIPPIR acronym1. The carbon isotopes of these diamonds indicate their origin from subducted slab material 2, 3, 4. While the formation of CLIPPIR diamonds in the mantle transition zone (MTZ) appears robustly constrained by the occurrence of majorite inclusions1, the nature of the CLIPPIR diamond substrate remains obscure. Here we show that CLIPPIR diamonds are associated with kimberlites tapping vertically extensive, Fe-rich and deformed domains at the base of the lithosphere. Beyond enrichment in Fe, these domains exhibit light oxygen and heavy Fe isotopes, which indicate a major role of subducted basaltic material that experienced hydrothermal alteration at the Earths surface. The association of CLIPPIR and other sub-lithospheric diamonds with these anomalous Fe-rich domains that are rarely sampled by kimberlites and their similar isotopic anomalies point to a genetic relation. Considerations on kimberlite genesis in the upper convective mantle and partial retrogression of majorite inclusions suggest that the CLIPPIR substrate originally stalled in the MTZ, where the diamonds grew, before being accreted at the base of the lithosphere. The geographic overlap between CLIPPIR diamond locations and the loci of large igneous provinces points to accretion of subducted slab material including dense eclogitic crust via buoyant mantle upwellings. Beyond providing the largest diamonds, these Fe-rich, isotopically anomalous domains contribute to the isotopic heterogeneity of intraplate magmas globally.","manuscriptTitle":"The largest diamonds are hosted in iron-rich substrate accreted at the base of the lithosphere","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-06 09:16:23","doi":"10.21203/rs.3.rs-6808081/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"423c9ec4-9c14-40cb-9314-8dd71e8174fe","owner":[],"postedDate":"June 6th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":49444402,"name":"Earth and environmental sciences/Solid Earth sciences/Petrology"},{"id":49444403,"name":"Earth and environmental sciences/Solid Earth sciences/Geology"}],"tags":[],"updatedAt":"2025-06-25T16:50:44+00:00","versionOfRecord":[],"versionCreatedAt":"2025-06-06 09:16:23","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6808081","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6808081","identity":"rs-6808081","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}