No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
The influence of melt infiltration on the Li and Mg isotopic composition of the Horoman Peridotite Massif
Abstract
We have analysed the Li and Mg isotope ratios of a suite of samples from the Horoman peridotite massif. Our results show that most Li and all Mg isotopic compositions of the Horoman peridotites are constant over 100 metres of continuous outcrop, yielding values for pristine mantle of δ7Li = 3.8 ± 1.4 ‰ (2SD, n = 9), δ25Mg = -0.12 ± 0.02 ‰ and δ26Mg = -0.23 ± 0.04 ‰ (2SD, n = 17), in keeping with values for undisturbed mantle xenoliths. However, there are also some anomalously low δ7Li values (-0.2 to 1.6 ‰), which coincide with locations that show enrichment of incompatible elements, indicative of the prior passage of small degree melts. We suggest Li diffused from the infiltrating melts with high [Li] into the low [Li] minerals and kinetically fractionated 7Li/6Li as a result. Continued diffusion after the melt flow had ceased would have resulted in the disappearance of this isotopically light signature in less than 15 Ma. In order to preserve this feature, the melt infiltration must have been a late stage event and the massif must have subsequently cooled over a maximum of ∼0.3 Ma from peak temperature (950°C, assuming the melts are hydrous) to Li closure temperature (700°C), likely during emplacement. The constant δ26Mg values of Horoman peridotites suggest that chemical potential gradients caused by melt infiltration were insufficient to drive associated δ26Mg fractionation greater than our external precision of 0.03 ‰.
... Much effort has already been expended in constraining this value through peridotite analyses (Handler et al., 2009;Yang et al., 2009;Bourdon et al., 2010;Teng et al., 2010;Huang et al., 2011;Liu et al., 2011;Pogge von Strandmann et al., 2011;Xiao et al., 2013Xiao et al., , 2016Lai et al., 2015;Wang et al., 2016;An et al., 2017;Stracke et al., 2018;Hu et al., 2020) and these sample-standard bracketing measurements yield a typical value of δ 26 Mg DSM (the relative difference of 26 Mg/ 24 Mg ratio of sample from reference material DSM-3, Galy et al., 2003) around − 0.25‰ for 'normal' mantle (Teng, 2017). The improved reproducibility achievable using critical mixture double-spiking Hin et al., 2017;He et al., 2022), however, prompts a reassessment of mantle Mg isotopic compositions. ...
... Interpreting the information carried in the Mg isotopic compositions of mantle derived melts further requires accurate and precise fractionation factors between mantle phases (e.g., Soderman et al., 2022). Measurements of mineral separates from peridotites have provided vital empirical support of magnesium isotopic fractionation factors determined from theory (Handler et al., 2009;Yang et al., 2009;Liu et al., 2011;Huang et al., 2011;Pogge von Strandmann et al., 2011;Xiao et al., 2013Xiao et al., , 2016Lai et al., 2015;Hu et al., 2016;An et al., 2017;Liu et al., 2017;Zhao et al., 2017;Chen et al., 2018;Stracke et al., 2018;Hu et al., 2020). Yet the inter-mineral fractionations are subtle compared to the typical analytical reproducibility of sample-standard bracketing analysis, so the ab initio calculations have not currently been calibrated to appropriate precision. ...
... The Horoman peridotite has been argued to be formed from ancient (0.8-1 Ga) partial melting residues of a MORB source mantle (Takazawa et al., 1996;Yoshikawa and Nakamura, 2000;Saal et al., 2001;Malaviarachchi et al., 2008Malaviarachchi et al., , 2010, which then reacted with percolating melt at ~23 Ma, possibly in a mantle wedge environment, locally causing the enrichment in LREE and the rare formation of phlogopite (Yoshikawa et al., 1993;Takazawa et al., 1996). Lai et al. (2015) identified some Horoman samples that have kinetically perturbed Li isotopic ratios. We avoided analysing these samples for high precision Mg isotope measurements. ...
In order to better constrain the Mg isotopic composition of the mantle, we have analysed twenty-eight samples of both oceanic and continental peridotite using a high-precision, critical mixture double spiking approach. The unaltered samples show no variability δ26Mg in outside analytical uncertainty and yield a value of − 0.236 ± 0.006‰ (2 s.e.) for the accessible mantle, substantiating its non-chondritic composition. We have also deter- mined inter-mineral Mg isotopic fractionations for a sub-set of samples. We document small but significant differences in δ26Mg between olivine and pyroxenes, Δ26/24Mg ol/cpx = − 0.118 ± 0.018‰ and Δ26/24Mg ol/opx = − 0.056 ± 0.018‰, in excellent agreement with ab initio calculations for temperatures ~1000 ◦C, as recorded by mineral thermometry in the peridotites. The differences in δ26Mg between olivine and spinel (Δ26/24Mg ol/sp) are more variable and generally higher than theoretical calculations at corresponding temperatures, likely due to incomplete Fe-Mg diffusive exchange during post-eruptive cooling of the xenoliths.
Using these data, together with a recently determined olivine-melt fractionation factor for Mg isotopes, we show that partial melting has a negligible influence on the δ26Mg of residual peridotites. This helps account for the minimal variability of δ26Mg in fresh, mantle peridotites. However, the δ26Mg of primary mantle melts are predicted to be discernibly higher than their sources (Δ26Mg ~ 0.06‰ and ~0.123‰ for representative partial melts of peridotitic and pyroxenitic sources respectively) across a wide range of melting conditions. Such elevated δ26Mg values are not generally observed in the current dataset of mantle derived melts. We propose that this inconsistency is likely a consequence of diffusive fractionation during partial re-equilibration between low Mg/Fe melts migrating through high Mg/Fe mantle en route to the surface.
... 2 of 25 +5.6‰; H. R. Marschall et al., 2017;Tomascak et al., 2008). Given that MORB likely represent a well-averaged melt extracted from the upper mantle (Langmuir et al., 1992), the δ 7 Li value of MORB well approximates the δ 7 Li value of the upper mantle (+1.4 to +5.2‰), postulated by the analyses of pristine mantle rocks (Jeffcoate et al., 2007;Lai et al., 2015;Pogge von Strandmann et al., 2011;Seitz et al., 2004). Accordingly, the similarity of δ 7 Li between MORB and arc volcanic rocks has been interpreted to reflect the prevalence of the upper mantle as a source of Li in arc magmas (Tomascak et al., 2002). ...
... The range of δ 7 Li values in Shikoku Basin basalts is assumed to be the same as the variation in global seafloor basalts (i.e., altered oceanic crusts) presented in Penniston-Dorland et al. (2017). δ 7 Li value of mantle wedge is assumed to be +3.5 ± 0.5‰, which is the mean δ 7 Li value of pristine peridotites (Jeffcoate et al., 2007;Lai et al., 2015;Pogge von Strandmann et al., 2011;Seitz et al., 2004). Uncertainty of δ 7 Li values for the Chugoku district samples is ±0.6‰ (2σ external reproducibility). ...
... The δ 7 Li value of the mantle would have varied little since peridotite represents a major source of Li in the mantle. We thus assume the δ 7 Li value of the wedge mantle to be the same as the mean δ 7 Li value (+3.5 ± 1.0‰) of pristine peridotites (Jeffcoate et al., 2007;Lai et al., 2015;Pogge von Strandmann et al., 2011;Seitz et al., 2004). It is noted that the δ 7 Li value postulated for the wedge mantle is identical to the mean δ 7 Li value of unaltered MORB (+3.6 ± 2.0‰) within analytical uncertainty. ...
Dehydration of subducting oceanic lithosphere (slab) induces Li‐isotope fractionation between the fluid and the slab, suggested by the δ⁷Li variation (∼10‰) in exhumed subduction complexes. Given that arc magmas represent melt of the supraslab mantle, a large δ⁷Li variation is anticipated for arc volcanic rocks. However, the δ⁷Li values in these rocks are mostly homogeneous within the range of mid‐ocean ridge basalts (+1.6 to +5.6‰). The lack of a subduction‐related δ⁷Li signature has been explained by (a) homogenization by mixing of different magma sources, (b) loss of Li from the slab via dehydration, or (c) homogenization by diffusive exchange of slab‐derived Li and the mantle. The Chugoku district in SW Japan is an ideal place to study the process responsible for Li‐isotope variation in arc magmas, since the Chugoku volcanic rocks show large δ⁷Li variation (−1.9 to +7.4‰). High δ⁷Li values (+6.3 to +7.4‰) are found in some high‐Sr andesites and dacites (adakites) whereas low δ⁷Li values (−1.0 to −0.1‰) are found in high‐Mg andesites. The parental magmas of these rocks have been sourced from subducted oceanic crust and sediments, respectively, with various extents of the interaction with wedge mantle. The limited extents of Li isotope modification are indicated by the similarity of the δ⁷Li values of these rocks and their supposed sources. The models for a slab dehydration and a diffusive exchange between slab‐derived melt and mantle demonstrate that the δ⁷Li signatures of the sources can be preserved in the adakites if they ascent rapidly in mantle.
... As a promising isotopic proxy, Li isotope systematics (expressed as δ 7 Li) have so far been rarely used to trace recycled sedimentary carbonate (e.g., Halama et al., 2008;Tan et al., 2022), despite the large isotopic contrast between sedimentary carbonate and the depleted mantle (Tomascak et al., 2016 and references therein). The depleted mantle displays relatively homogeneous δ 7 Li values of 3.8 ± 1.4 ‰ (2SD, Seitz et al., 2004;Jeffcoate et al., 2007;Magna et al., 2008;Lai et al., 2015). Although large Li isotopic fractionation has been reported during the magma degassing process (Neukampf et al., 2022;Vlastélic et al., 2011), most magmatic processes do not significantly fractionate Li isotopes (Jeffcoate et al., 2007;Schuessler et al., 2009;Tomascak et al., 1999). ...
... Marschall et al. (2017). Depleted mantle data are from Lai et al. (2015). The positions of AOC (altered oceanic crust) are beyond the scales of the plots due to its high δ 7 Li value (10-20‰) and Li concentration (~19 μg/g) (data from Chan et al., 2002;Kelley et al., 2003). . ...
... MORB (mid-ocean ridge basalt) data are from Marschall et al. (2017) and Sun and Sun and McDonough (1989). Depleted mantle data are from Lai et al. (2015) and Workman and Hart (2005). The positions of AOC (altered oceanic crust) are beyond the scales of the plots due to its high δ 7 Li value (10-20‰) and Li/Y (~0.75) (data from Chan et al., 2002;Kelley et al., 2003). ...
... The Li isotopic compositions of minerals can be easily affected by melt/ fluid activities in rocks (Lai et al., 2015;Parkinson et al., 2007;Pogge von Strandmann et al., 2011;Rudnick & Ionov, 2007). Given the freshness of the dunite enclaves (Figures 2b-2e), the Li isotopic features of olivine grains in the dunite enclaves could not have been notably affected by serpentinization. ...
... This suggests that the closed-system diffusion of Li only affected a few grains, and many others, especially the inner parts of larger grains (e.g., 3-4 mm, Figures 2b-2d), were unlikely affected by the diffusion, so that their Li isotopic features obtained during the open-system reaction are well preserved, explaining why the δ 7 Li values of most analytical spots fall into the narrow range (∼+10‰ to +15‰; Figures 4b-4c). Moreover, because of the rapid diffusivity of Li, large Li isotopic variations in mantle rocks cannot survive for long under high temperature conditions (Gao et al., 2011;Halama et al., 2009;Lai et al., 2015;P. P. Liu et al., 2020;Marschall & Tang, 2020;Parkinson et al., 2007). ...
The compositions of chromitites and dunites from Moho transition zone (MTZ) of the Coto block of the Zambales ophiolite, Philippines, are used to investigate the geodynamic transition from anhydrous to hydrous magmatism during subduction initiation (SI). Chromite grains in the chromitites have Cr# values [100 × Cr/(Cr + Al)] and TiO2 contents ∼35–50 and 0.05–0.30 wt.%, respectively, intermediate between those of chromite in typical MORB‐like lavas (Cr#, ∼20–60; TiO2, ∼0.6–1.7 wt.%) and boninites (Cr#, ∼70–85; TiO2, <0.4 wt.%). Olivine grains in the dunites have δ⁷Li values varying from ∼−2‰ to +21‰ with most between +10‰ and +15‰, beyond that of normal mantle (+4 ± 2‰) but comparable to those of some arc lavas (up to +12‰). The data set indicates that parental magmas of the high‐Al chromitites originated from hydrated harzburgitic mantle sources and formed temporally between MORB‐like and boninitic magmatism during SI, resulting from the early stage of flux melting in the Zambales proto‐forearc mantle. Modeling of Li diffusion reveals that the MTZ cooled down at a minimum rate of 0.1°C/yr in order to preserve the large δ⁷Li variation of olivine in the dunites, comparable to the thermal conditions below ultra‐slow to slow spreading ridges. Such a stage of transitional magmatism, although displaying notable slab contributions, took place at a sluggish period of slab rollback and asthenospheric upwelling, leading to a trough level of heat flow and magma production during the entire course of SI.
... In contrast, unweathered oceanic mantle peridotites (abyssal or ophiolitic) worldwide have Hu et al. (2020). Literature data of relatively fresh oceanic mantle peridotites (Table S3) (de Obeso et al., 2021;Eom et al., 2022;Lai et al., 2015;Liu et al., 2017;Su et al., 2015) are plotted for comparison. The yellow area in Fig. a represents average Mg isotopic compositions of mantle peridotites (δ 26 Mg = − 0.25 ± 0.04‰, Teng et al., 2010). ...
... This isotopic heterogeneity may be caused by plate subduction that brings sediments, altered oceanic crust, and abyssal peridotites with different MgO contents and Mg isotopic signatures (e.g., Hu et al., 2017;Huang et al., 2018;Liu et al., 2017) into the asthenospheric mantle and recycles them into the global oceanic mantle by mantle convection (Liu et al., 2022a;Liu et al., 2008;Stracke, 2012). Hence, the recognizable Mg isotope difference (Table S3 and S4) of unweathered oceanic mantle peridotites (de Obeso et al., 2021;Eom et al., 2022;Lai et al., 2015;Liu et al., 2017;Su et al., 2015) and other mantle wedge peridotites (solid orange circles in Fig. c) (Hu et al., 2020) are plotted for comparison. between the oceanic mantle and mantle wedge indicates that Mg isotopes can be a useful tool for identifying mantle heterogeneity and tracing geochemical cycling in the deep earth. ...
The serpentinized mantle wedge is critical for the geochemical cycling of water, volatiles, and fluid-mobile elements in the subduction zone. It is also a major reservoir of magnesium (Mg) in subduction zones, but its Mg isotopic compositions are still not well constrained. To investigate Mg isotopic fractionation during mantle wedge serpentinization, and better understand Mg isotopic compositions of the mantle wedge, we studied Mg isotopes of mantle wedge serpentinites/serpentinized peridotites exhumed by Mariana forearc serpentinite mud volcanoes. Whole-rock δ26Mg values of these samples vary from -0.29 to -0.03‰. Some serpentinite/serpentinized peridotite samples have significantly elevated δ26Mg values up to -0.03‰, which is caused by seafloor weathering after their exhumation by the mud volcanoes. In contrast, the unweathered serpentinized peridotites have homogeneous δ26Mg values of -0.29 to -0.27‰ (mean δ26Mg = -0.28 ± 0.01‰, 2SD, n=3), which represent the primary Mg isotopic compositions of the Mariana forearc mantle wedge peridotite. However, the unweathered mature serpentinites (i.e., completely serpentinized) have slightly heavier Mg isotopic compositions (δ26Mg= -0.29 to -0.21‰, mean δ26Mg = -0.24 ± 0.05‰, 2SD, n=16) than the serpentinized peridotites, indicating that Mg isotopes are fractionated during the late-stage mantle wedge serpentinization. It is probably due to the leach of isotopically light Mg by the infiltrating slab fluids after the complete consumption of olivine.
By compiling Mg isotopic data of mantle wedge and oceanic mantle peridotites, we find that mantle wedge peridotites have Mg isotopic compositions (δ26Mg = -0.27 ± 0.04‰) that are identical to subcontinental lithospheric mantle peridotites (-0.25 ± 0.04‰), which are more uniform and lighter than those of oceanic mantle peridotites (-0.21 ± 0.12‰). The Mg isotope difference between them is most probably caused by mantle heterogeneity, not by different degrees of early partial melting as indicated by the incremental batch melting modeling.
... Significantly lower δ 26 Mg-values compared to the average peridotitic mantle (− 0.25‰ ±0.07) (Teng et al., 2010;Teng et al., 2007) are typically attributed to carbonated (dolomite-magnesite bearing) lithologies in the mantle source. δ 26 Mg isotope values of peridotites range from − 0.48 to − 0.06‰ (Bourdon et al., 2010;Handler et al., 2009;Huang et al., 2011;Lai et al., 2015;Pogge von Strandmann et al., 2011;Teng et al., 2010;Xiao et al., 2013;Yang et al., 2009). Mg isotope fractionation during mantle melting and magmatic differentiation is considered to be limited (e.g., Dai et al., 2017;Huang et al., 2015;Teng et al., 2010;Teng et al., 2007;Wang et al., 2016). ...
... with an average of − 0.319‰ (Table 2, Fig. 5). The δ 26 Mg values of these highly alkaline samples may be considered to be within the range of isotope variability measured for peridotites, within the ranges of − 0.48 to − 0.06‰ (Bourdon et al., 2010;Handler et al., 2009;Huang et al., 2011;Lai et al., 2015;Pogge von Strandmann et al., 2011;Teng et al., 2010;Xiao et al., 2013;Yang et al., 2009). The average value of the Brunlanes lavas (− 0.319 ± 0.052[2SD]) is slightly lower than the δ 26 Mg of the normal peridotitic mantle (− 0.25 ± 0.07‰; Hu et al., 2016;Teng et al., 2007). ...
... Mg isotopic data for Pitcairn basalts are from Wang et al. (2018). The Fe and Mg isotopic compositions of the subcontinental lithospheric mantle (SCLM) are assumed to be those of the average mantle (Lai et al., 2015;Sossi et al., 2016). The Fe and Mg isotopic compositions of the average MORB are from Sossi et al. (2016) and Teng (2017), respectively. ...
... Second, high MgO and low K 2 O and SiO 2 contents for the low-δ 57 Fe endmember resemble those of melts derived from the garnet peridotite ( Figure 4). Third, the widely accepted δ 57 Fe and δ 26 Mg of normal mantle are 0.05 ± 0.04‰ (Sossi et al., 2016) and 0.23 ± 0.04‰ (Lai et al., 2015), respectively, which is similar to those of the low-δ 57 Fe endmember. Finally, Liu et al. (2016) have found that the chemical variation ( 87 Sr/ 86 Sr, K 2 O/ Na 2 O, MgO, and Rb/Y) of the potassic basalts are well correlated to the lithospheric thickness, indicating the various degree of melt-rock interaction for each potassic volcanic field. ...
The origin of EM1 (Enriched Mantle 1) reservoir, initially defined by the ocean island basalts (OIBs) with extremely low ¹⁴³Nd/¹⁴⁴Nd and ²⁰⁶Pb/²⁰⁴Pb, has been long debated, because melting of the ambient refractory peridotite along with the EM1 component will dilute the “EM1 fingerprints” recorded in these rocks. Comparing to the OIBs, Cenozoic potassic basalts from northeast China, the typical EM1‐type basalts in continental region, are formed at a lower‐degree melting, and therefore have the chance to preserve more information of the EM1 component. Here high‐precision Fe isotopes of these potassic basalts are reported to constrain the source lithology. Their δ⁵⁷Fe (0.15–0.28‰) are positively correlated with the K2O, SiO2, K/U, and Rb/Y, and negatively correlated with the εNd and δ²⁶Mg, forming binary mixing arrays. One endmember is the inferred EM1 reservoir, whereas the other is the local lithospheric mantle. Major elemental compositions of the melts released from the EM1 component resemble those sediment‐derived experimental melts. Combining with their heavier Fe isotopes and higher Zn/Fe ratios relative to those mid‐ocean ridge basalts (MORBs), an eclogitic source of these potassic basalts is therefore proposed to account for these features. Differing from the most conventional thinking of the metasomatized, phlogopite‐bearing lithospheric mantle, we argue that the EM1 component in the source of continental potassic basalts are composed of ancient subducted crustal materials (i.e., recycled sediment ± oceanic crust). This deep EM1 component will transform into eclogite and release high‐SiO2 potassic melts when ascending to the shallow asthenosphere.
... The Li isotopic compositions of minerals can be easily affected by melt/ fluid activities in rocks (Lai et al., 2015;Parkinson et al., 2007;Pogge von Strandmann et al., 2011;Rudnick & Ionov, 2007). Given the freshness of the dunite enclaves (Figures 2b-2e), the Li isotopic features of olivine grains in the dunite enclaves could not have been notably affected by serpentinization. ...
... This suggests that the closed-system diffusion of Li only affected a few grains, and many others, especially the inner parts of larger grains (e.g., 3-4 mm, Figures 2b-2d), were unlikely affected by the diffusion, so that their Li isotopic features obtained during the open-system reaction are well preserved, explaining why the δ 7 Li values of most analytical spots fall into the narrow range (∼+10‰ to +15‰; Figures 4b-4c). Moreover, because of the rapid diffusivity of Li, large Li isotopic variations in mantle rocks cannot survive for long under high temperature conditions (Gao et al., 2011;Halama et al., 2009;Lai et al., 2015;P. P. Liu et al., 2020;Marschall & Tang, 2020;Parkinson et al., 2007). ...
Here we present new occurrences of amphibole in a suite of chromitites, dunites, and harzburgites from the mantle sequence of the Lycian ophiolite in the Tauride Belt, southwest Turkey. The amphibole occurs both as interstitial grains among the major constituent minerals and as inclusions in chromite grains. The interstitial amphibole shows generally decreasing trends in Na2O and Al2O3 contents from the chromitites (0.14–1.54 wt% and 0.04–6.67 wt%, respectively) and the dunites (0.09–2.37 wt%; 0.12–11.9 wt%) to the host harzburgites (<0.61 wt%; 0.02–5.41 wt%). Amphibole inclusions in chromite of the amphibole-bearing harzburgites are poorer in Al2O3 (1.12–8.86 wt%), CaO (8.47–13.2 wt%), and Na2O (b.d.l.–1.38 wt%) than their counterparts in the amphibole-bearing chromitites (Al2O3 = 6.13–10.0 wt%; CaO = 12.1–12.9 wt%; Na2O = 1.11–1.91 wt%). Estimated crystallization temperatures for the interstitial amphibole grains and amphibole inclusions range from 706 to 974 °C, with the higher values in the latter. A comparison of amphibole inclusions in chromite with interstitial grains provides direct evidence for the involvement of water in chromitite formation and the presence of hydrous melt/fluid metasomatism in the peridotites during initial subduction of Neo-Tethyan oceanic lithosphere. The hydrous melts/fluids were released from the chromitites after being collected on chromite surfaces during crystallization. Different fluid/wall rock ratios are thought to have controlled the crystallization and composition of the Lycian amphibole and the extent of modification of the chromite and pyroxene grains in the peridotites. Considering the wide distribution of podiform chromitites in this ophiolite, the link between chromitite formation and melt/fluid metasomatism defined in our study may be applicable to other ophiolites worldwide.
... Of particular interest to this study is the ability of Fe and Mg isotopes to record metasomatic processes in the mantle. The effects of metasomatism on the Mg isotope composition of mantle rocks depends on the metasomatizing agent, with silicate-melt metasomatism generating a range of δ 26 Mg values from lower to higher than mantle values (Yang et al., 2009;Huang et al., 2011;Pogge von Strandmann et al., 2011;Xiao et al., 2013;Hu et al., 2015;Lai et al., 2015;Su et al., 2015;Yu et al., 2021), carbonatemelt metasomatism generating values that are typically lighter than mantle values (Yang et al., 2012;Hu et al., 2015;Wang et al., 2016b;Sun et al., 2021;Yu et al., 2021), and metasomatism by slab-derived fluids generating values that are typically heavier than mantle values (Pogge von Strandmann et al., 2011;Li et al., 2018) (Fig. 2c). In contrast to Mg isotopes, the effects of metasomatism on Fe isotopes are more challenging to discern because they also fractionate as a result of mantle partial melting and fractional crystallization (Teng et al., 2008;Dauphas et al., 2009;Schuessler et al., 2009;Sossi et al., 2012). ...
... The δ 56 Fe values of whole-rock samples and mineral separates are from the Baima layered mafic intrusions (Chen et al., 2014;Liu et al., 2014), the Panzhihua layered mafic intrusion (Tian et al., 2020), the Windimurra Igneous Complex (Nebel et al., 2020), and the Bushveld Igneous Complex (Bilenker et al., 2017). The δ 26 Mg values of metasomatized rocks are from Yang et al. (2009), Huang et al. (2011), Pogge von Strandmann et al. (2011), Yang et al. (2012, Xiao et al. (2013), Hu et al. (2015), Huang et al. (2015), Lai et al. (2015), Su et al. (2015), Chen et al. (2016), Wang et al. (2016b), Li et al. (2018), Sun et al. (2021), and Yu et al. (2021). The δ 56 Fe values of metasomatized rocks are from Weyer & Ionov (2007), Zhao et al. (2010), Huang et al. (2011), Zhao et al. (2012, Poitrasson et al. (2013), Su et al. (2015), Debret et al. (2016), Debret et al. (2018), and Sun et al. (2020). ...
Mafic intrusions in the Coldwell Complex have previously been interpreted as forming from a metasomatized mantle source. To build upon our understanding of this metasomatism, the Mg–Fe isotope compositions of these rocks have been determined, and variations are assessed with respect to the magmatic processes that could have occurred at different stages of their formation. The mineralized Marathon Series (δ26Mg = -0.28‰ to -0.19‰), associated metabasalt (δ26Mg = -0.24‰ to -0.23‰), and the Geordie Lake gabbro (δ26Mg = -0.31‰ to -0.22‰) are characterized by δ26Mg values that are within the range of mantle values, whereas the unmineralized Layered Series (δ26Mg = -0.2‰ to -0.05‰) is heavier than mantle. In contrast, the δ56Fe values of all the Coldwell basaltic–gabbroic rocks (δ56Fe = 0.07 ± 0.08‰) are heavier than mantle, but within the range of terrestrial basalts and mafic–ultramafic layered intrusions. We propose that the Mg–Fe isotope compositions of these rocks was not significantly modified by processes such as partial melting or garnet retention/fractionation in the mantle, fractional crystallization, or contamination during ascent through the crust, as the isotope values do not correlate with proxies for these processes (e.g., La/Sm and La/Yb, Gd/Yb, MgO–CaO–TiO2, and Th/Nb and Th/La, respectively). Their isotope compositions are, therefore, proposed to reflect the compositions of their metasomatized mantle sources. We conclude that metasomatism was not caused by a carbonate melt, subduction-altered oceanic crust and sediments, or an evolved silicate melt, as these processes generate light δ26Mg, variably fractionated δ56Fe, and heavy δ56Fe values, respectively, which are not observed in our dataset for the Coldwell Complex. The agent that metasomatized the mantle beneath the Coldwell Complex was likely slab-derived fluids characterized by isotopically heavy δ26Mg and basaltic δ56Fe values. This scenario can account for the lack of Fe isotope fractionation from basaltic values in all of the Coldwell rocks. The variably heavier δ26Mg of the Layered Series (-0.20 ± 0.01‰ to -0.05 ± 0.05‰) relative to the mantle (-0.25 ± 0.07‰) suggests that the magmas for the Coldwell rocks were derived by tapping of an isotopically heterogeneous mantle source that had undergone variable degrees of metasomatism. The distinctive geochemistry of mafic sequences in the Coldwell and numerous mafic dykes located in the northeast shoulder of the Midcontinent Rift suggests the presence of a variably metasomatized mantle source beneath a large area of the rift.
... Isotope disequilibrium during the Fe-Mg exchange also has the chance to produce Mg-Fe isotope arrays with a negative slope; however, such proposal cannot explain the formation of Pitcairn basalts for two reasons. On the one hand, diffusion is suggested to be very fast at a high temperature (Richter et al., 2008;Lai et al., 2015), and thus the negative correlation of Fe-Mg isotopes caused by Fe-Mg exchange is difficult to preserve at such condition. On the other hand, the Fe isotopic compositions of the Pitcairn basalts are also well correlated with Nd and Hf isotopes, indicating similar genesis for these isotopes; but to our knowledge, diffusion cannot modify the Nd and Hf isotopes of basalts. ...
... Mg isotope data of Pitcairn basalts and Louisville basalts are from Wang et al. (2018) and Zhong et al. (2017), respectively. The Fe, Nd, and Mg isotopic compositions of peridotitic mantle are from Sossi et al. (2016), Chauvel and Blichert-Toft (2001) and Lai et al. (2015). Other details of calculation can be seen in Table S4 proportions can explain the observed correlations shown by the Pitcairn basalts. ...
Subducted oceanic crust can transform into eclogite in the upper mantle, and generate chemical heterogeneity of mantle plumes, as recorded by elemental and radiogenic isotopic variations in oceanic island basalts (OIBs). The secondary pyroxenite produced by the reaction between eclogite-derived melt and peridotite is increasingly regarded as a major source component of OIBs, as well as peridotite. However, it remains unclear whether eclogite can be a direct source component of OIBs. To test this possibility, we present high-precision whole-rock Fe isotopes and the chemical compositions of olivine phenocrysts from well-characterized EM1-type basalts from Pitcairn Island. The Pitcairn basalts are characterized by moderate ⁸⁷Sr/⁸⁶Sr, low ¹⁴³Nd/¹⁴⁴Nd and ²⁰⁶Pb/²⁰⁴Pb isotopic ratios, and lowest δ²⁶Mg values among OIBs, suggesting a contribution from recycled ancient crustal components (oceanic crust plus sediment). For comparison, we also report the Fe isotope compositions of FOZO-type basalts (with low ⁸⁷Sr/⁸⁶Sr, moderately high ¹⁴³Nd/¹⁴⁴Nd and moderate ²⁰⁶Pb/²⁰⁴Pb ratios) from the Louisville hot spot track, which were suggested to be a typical peridotite-derived OIB. The Louisville basalts have MORB-like δ⁵⁷Fe values (0.06‰–0.15‰), whereas the Pitcairn lavas have substantially heavier Fe isotopic compositions (δ⁵⁷Fe = 0.17‰–0.31‰) than MORBs. Quantitative evaluations suggest that magmatic differentiation, partial melting, and elevated oxygen fugacity in the source are insufficient to generate the heavy Fe isotopic compositions of the Pitcairn basalts.
A good correlation between δ⁵⁷Fe and εNd(i) (or δ²⁶Mg) values in the Pitcairn basalts indicates binary mixing, and the melts derived from the EM1 endmember have unusually heavy Fe and light Mg isotopic compositions. In order to explain the origin of the Fe and Mg isotopic compositions, we calculated the Fe, Mg, and Nd isotopic compositions of partial melts of eclogite, secondary pyroxenite and peridotite, respectively. The results indicate that eclogite is the only suitable candidate to generate melts with both heavy Fe and light Mg isotopes. This inference is strengthened by the major and minor elemental compositions of olivine phenocrysts from the Pitcairn basalts, which show low Fo (73.2–82.5) and Ni contents (620–1949 ppm), and low Mn/Fe (0.011–0.015) and Ni/(Mg/Fe) (524–1041) ratios. These characteristics are markedly different from those of olivines from MORBs and the Koolau lavas from Hawaii, which represent phenocrysts that equilibrated with peridotite-derived and secondary pyroxenite-derived melts, respectively. We therefore argue that eclogite is the source lithology of the EM1 endmember of the Pitcairn basalts. Binary mixing between our modelled eclogite- and peridotite-derived melts produced magmas with relatively low Mg# value, Mg/Fe ratios and moderate Ni contents characteristics, which are preserved in the low-Fo Pitcairn olivines. Our results highlight that, in addition to peridotite and secondary pyroxenite, eclogite may survive in mantle plumes at shallow depth and make a substantial contribution to the source of OIBs.
... This isotopic fractionation could have been controlled by mineral structures Roskosz et al. 2015), changes in the oxidation state of Fe (Williams et al. 2004(Williams et al. , 2005Dauphas et al. 2009Dauphas et al. , 2014Sossi et al. 2012), melt extraction (Williams et al. 2004(Williams et al. , 2005Weyer and Ionov 2007;Teng et al. 2008;Schuessler et al. 2009;Hibbert et al. 2012), the effect of metasomatism (Beard and Johnson 2004;Williams et al. 2005;Weyer and Ionov 2007;Dauphas et al. 2009;Zhao et al. 2010Zhao et al. , 2012Zhao et al. , 2015Poitrasson et al. 2013) and/or kinetic isotopic fractionation caused by diffusion Weyer and Ionov 2007;Huang et al. 2011;Weyer and Seitz 2012;Zhao et al. 2012Zhao et al. , 2015Zhao et al. , 2017Poitrasson et al. 2013). By contrast, analysis of most oceanic basalts and mantle peridotites has shown that the terrestrial mantle has a homogeneous Mg isotopic composition (average δ 26 Mg = −0.25 ± 0.07, 2SD) (Teng et al. 2007(Teng et al. , 2017Handler et al. 2009;Yang et al. 2009;Bourdon et al. 2010;Dauphas et al. 2010;Bizzarro et al. 2011;Huang et al. 2011;Pogge von Strandmann et al. 2011;Liu et al. 2011;Xiao et al. 2013;Lai et al. 2015) and that the fractionation of Mg isotopes during partial melting and magma-differentiation processes is limited in the mantle (<0.07‰ for δ 26 Mg, Teng et al. 2007Teng et al. , 2010Teng et al. , 2017. However, recent studies have shown that chemical diffusion generates Fe and Mg isotope fractionation in komatiite, zoned olivine and strongly metasomatized mantle xenoliths that exceeds potential equilibrium isotope fractionation by an order of magnitude Teng et al. 2011;Weyer and Seitz 2012;Sio et al. 2013;Huang et al. 2011;Zhao et al. 2012Zhao et al. , 2015Hu et al. 2016). ...
... Here, we have shown that considerable Fe isotopic heterogeneity exists at both the mineral and bulkrock scales in the strongly metasomatized garnet pyroxenites and phlogopite clinopyroxenite from Hannuoba (Fig. 3, Fig. 5). Previous analyses of most oceanic basalts and mantle peridotites yielded consistent δ 26 Mg values (−0.25 ± 0.07) within the uncertainty of current analytical methods and hence suggest a homogeneous mantle (Handler et al. 2009;Yang et al. 2009;Bourdon et al. 2010;Teng et al. 2010;Huang et al. 2011;Liu et al. 2011;Lai et al. 2015). However, more recent studies on strongly metasomatized peridotites, particularly wehrlites, found that δ 26 Mg values deviated from the nominal mantle range Pogge von Strandmann et al. 2011;Xiao et al. 2013), indicating that the mantle is not completely homogeneous in terms of Mg isotopes. ...
... This isotopic fractionation could have been controlled by mineral structures Roskosz et al. 2015), changes in the oxidation state of Fe (Williams et al. 2004(Williams et al. , 2005Dauphas et al. 2009Dauphas et al. , 2014Sossi et al. 2012), melt extraction (Williams et al. 2004(Williams et al. , 2005Weyer and Ionov 2007;Teng et al. 2008;Schuessler et al. 2009;Hibbert et al. 2012), the effect of metasomatism (Beard and Johnson 2004;Williams et al. 2005;Weyer and Ionov 2007;Dauphas et al. 2009;Zhao et al. 2010Zhao et al. , 2012Zhao et al. , 2015Poitrasson et al. 2013) and/or kinetic isotopic fractionation caused by diffusion Weyer and Ionov 2007;Huang et al. 2011;Weyer and Seitz 2012;Zhao et al. 2012Zhao et al. , 2015Zhao et al. , 2017Poitrasson et al. 2013). By contrast, analysis of most oceanic basalts and mantle peridotites has shown that the terrestrial mantle has a homogeneous Mg isotopic composition (average δ 26 Mg = −0.25 ± 0.07, 2SD) (Teng et al. 2007(Teng et al. , 2017Handler et al. 2009;Yang et al. 2009;Bourdon et al. 2010;Dauphas et al. 2010;Bizzarro et al. 2011;Huang et al. 2011;Pogge von Strandmann et al. 2011;Liu et al. 2011;Xiao et al. 2013;Lai et al. 2015) and that the fractionation of Mg isotopes during partial melting and magma-differentiation processes is limited in the mantle (<0.07‰ for δ 26 Mg, Teng et al. 2007Teng et al. , 2010Teng et al. , 2017. However, recent studies have shown that chemical diffusion generates Fe and Mg isotope fractionation in komatiite, zoned olivine and strongly metasomatized mantle xenoliths that exceeds potential equilibrium isotope fractionation by an order of magnitude Teng et al. 2011;Weyer and Seitz 2012;Sio et al. 2013;Huang et al. 2011;Zhao et al. 2012Zhao et al. , 2015Hu et al. 2016). ...
... Here, we have shown that considerable Fe isotopic heterogeneity exists at both the mineral and bulkrock scales in the strongly metasomatized garnet pyroxenites and phlogopite clinopyroxenite from Hannuoba (Fig. 3, Fig. 5). Previous analyses of most oceanic basalts and mantle peridotites yielded consistent δ 26 Mg values (−0.25 ± 0.07) within the uncertainty of current analytical methods and hence suggest a homogeneous mantle (Handler et al. 2009;Yang et al. 2009;Bourdon et al. 2010;Teng et al. 2010;Huang et al. 2011;Liu et al. 2011;Lai et al. 2015). However, more recent studies on strongly metasomatized peridotites, particularly wehrlites, found that δ 26 Mg values deviated from the nominal mantle range Pogge von Strandmann et al. 2011;Xiao et al. 2013), indicating that the mantle is not completely homogeneous in terms of Mg isotopes. ...
... Diffusion-driven fractionation of Li isotopes is the most important mechanism for inter-mineral Li isotopic disequilibrium, which may result from closed-system Li-redistribution, Li diffusion from melt/host magma into entrained xenoliths or metasomatism (Jeffcoate et al., 2007;Rudnick and Ionov, 2007;Tang et al., 2007;Wagner and Deloule, 2007;Aulbach et al., 2008;Aulbach and Rudnick, 2009;Su et al., 2012Su et al., , 2014Su et al., , 2018Weyer and Seitz, 2012;Lai et al., 2015). The diffusion coefficients of Li in olivine range from 1 × 10 −11 m 2 s −1 in olivine (Dohmen et al., 2010), while Coogan et al. (2005) determined the diffusion coefficient of Li in clinopyroxene which is found to range between 5 × 10 −15 and 4 × 10 −12 m 2 s −1 at temperatures of 800-1100°C. ...
... However, closed-system Liredistribution would only cause low Li concentration and high δ 7 Li value in olivine, in combination with high Li concentration in coexisting clinopyroxene, which is different from that in our studies. On the other hand, the faster diffusion rate in clinopyroxene can also induce the observed isotopically heavy cores with lower Li concentrations than the light rims in minerals especially clinopyroxene from host magma (Jeffcoate et al., 2007;Rudnick and Ionov, 2007;Wagner and Deloule, 2007;Aulbach et al., 2008;Aulbach and Rudnick, 2009;Lai et al., 2015;Gu et al., 2016). However, the Li concentration in the host magma ranges from 3.90 to 6.20 ppm in the Niutoushan basalts, from 9.10 to 12.7 ppm in the Mingxi basalts and from 14.7 to 18.0 ppm in the Xilong basalts (Huang et al., 2017), all of which are much lower than those in clinopyroxene from these peridotites (Fig. 1b and e). ...
... The exceptionally high diffusivity of Li, about 3% differential diffusion of 6 Li than 7 Li and faster diffusion in clinopyroxene than in olivine at mantle conditions mean that the Li isotope composition of mantle material can be affected by diffusive processes (Richter et al., 2003;Tomascak et al., 2016). Thus, high temperature diffusion-driven fractionation of Li isotopes could be used to explain the significant δ 7 Li variations in minerals from peridotites (e.g., Lundstrom et al., 2005;Jeffcoate et al., 2007;Rudnick and Ionov, 2007;Ionov and Seitz, 2008;Gao et al., 2011;Pogge von Strandmann et al., 2011;Lai et al., 2015;Xiao et al., 2015;Gu et al., 2018), phenocrysts in lavas (e.g., Parkinson et al., 2007), peridotites (Turner et al., 2018), subduction-related metamorphic rocks (e.g., Marschall et al., 2007;Penniston-Dorland et al., 2010), an eclogite alteration envelope (John et al., 2012), and contact metamorphic aureoles adjacent to plutons (e.g., Teng et al., 2006;Liu et al., 2010;Ireland and Penniston-Dorland, 2015). ...
... In the Sr\ \Nd diagram (Fig. 4a) (Chung et al., 2005;Chu et al., 2006;Wang et al., 2016;Wei et al., 2017) and fluids derived from the subducted slab could have significantly influenced the isotopic composition of the overlying SCLM (Tatsumi et al., 1986). Tian et al. (2018) suggested that this SCLM has similar Sr-Nd-Pb-Li isotopic compositions ( Gao et al., 2011;Pogge von Strandmann et al., 2011;Lai et al., 2015). However, Krienitz et al. (2012) suggested that those lavas with the EMII geochemical characteristics have δ 7 Li values of about +4.5‰ to +6.3‰. ...
Post-collisional K-rich volcanic rocks (KVRs) can provide an opportunity to constrain the architecture of the lithosphere and the mechanisms of plateau uplift. However, their petrogenesis and geodynamic setting remain in dispute. Lithium concentrations and isotopic compositions of 87 potassic, ultrapotassic and Mg-rich potassic volcanic rocks (PVRs, UPVs, and MPRs, respectively) in SW Tibet, along with new Pb–Sr–Nd isotope data and whole-rock analyses, are used to constrain their mantle source and genesis. These rocks are characterized by very similar δ⁷Li values: PVRs vary from −4.9‰ to +3.2‰, UPVs from −3.9‰ to +1.7‰, and MPRs from −1.2‰ to +3.5‰. They can be classified into two groups: Group I (19 out of 87 samples) with heavier δ⁷Li values (+1.0‰ to +3.5‰) similar to those reported for mid-ocean-ridge and ocean-island basalts (MORBs and OIBs, respectively), and Group II (68 out of 87 samples) with lighter values (−4.9‰ to +1.0‰) similar to those of Indian lower crust. These variable isotopic compositions may record the isotopic signature of the early-middle Miocene subcontinental lithospheric mantle (SCLM). This paper demonstrates the existence of isotopically light mantle domains beneath the Lhasa terrane, which were ascribed to the interaction with fluids/melts derived from the subducted Indian lower crust. The modeling curves of Indian lower crust with a metasomatized mantle composition fully account for compositional variations in the PVRs, UPVs, and MPRs. They were generated by the partial melting of SCLM, which was metasomatized by fluids/melts derived from the subducted Indian lower crust (ca. 4–14%, ca. 4–10%, and ca. 6–10% for the PVRs, UPVs, and MPRs, respectively). The Li isotopic data indicate that the Indian lower crust was subducted beneath the central Lhasa subterrane, and this sheds new light on the formation of the Tibet Plateau.
... However, the difference in the Mg isotope composition of bulk silicate Earth (δ 26 Mg = -0.23 ‰) and that of the continental runoff (δ 26 Mg = -1.09 ‰) indicates significant fractionation that can be attributed to the different dissolution rates of isotopically distinct Mg-bearing minerals and to formation of solid weathering product phases, such as clay minerals (Hin et al., 2017;Lai et al., 2015;Tipper et al., 2006a, b;Teng et al., 2010b). ...
... The δ 26 Mg value of the least altered serpentinite from the study area is slightly higher than the composition of pristine mafic and ultramafic rocks Hin et al., 2017;Lai et al., 2015) but in agreement with antigorite separates from Norwegian ophiolite sequences (Beinlich et al., 2014). Decreasing MgO concentrations and increasing degree of weathering, as expressed by the proxy MgO/SiO2* ( Fig. 3; Liu et al., 2017;Snow and Dick, 1995), are correlated with increasing δ 26 Mg of the serpentine weathering products (Fig. 6a, b). ...
The Mg-isotope systematics of peridotite weathering and low-temperature carbonation have not yet been thoroughly investigated, despite their potential to provide insights into reaction pathways and mechanisms of lithosphere-hydrosphere transfer of Mg and sequestration of CO2 in carbonate minerals. Here, we present new observations of the evolution of Mg isotope ratios during subtropical ultramafic rock weathering and associated magnesite formation, including the lowest δ²⁶Mg of magnesite reported so far. At the investigated field sites in eastern Australia, the proximity of the ultramafic Mg source rocks and associated magnesite deposits provides boundary conditions that constrain Mg isotope fractionation during low-temperature alteration. Saprolite samples from Attunga, New South Wales, show that weathering of serpentinite is accompanied by Mg loss and formation of secondary Mg-bearing clay minerals. Furthermore, Mg isotope ratios increase systematically with weathering intensity, indicating that incorporation of ²⁶Mg into clay mineral structures controls Mg isotope fractionation during ultramafic rock weathering. The Mg-bearing clay formed by decomposition of serpentine minerals has a δ²⁶Mg value of ∼0.35‰, which is up to ∼0.6‰ heavier than the ultramafic precursor. In contrast, nodular magnesite hosted in ultramafic rock shows δ²⁶Mg values between −3.26‰ and −2.55‰ that are significantly lower than those of magnesite and dolomite formed by hydrothermal alteration of peridotite at higher temperature (δ²⁶Mg = −0.69‰ and −0.62‰). The strong enrichment of ²⁴Mg in nodular magnesite does not reconcile with simple fractionation during direct precipitation from ultramafic host rock buffered meteoric fluids and instead suggests multiple formation steps involving dissolution and re-precipitation of pre-existing carbonate accompanied by fractionation between species of dissolved Mg. Our data highlight the potential of Mg isotope studies for distinguishing the formation pathways of low temperature magnesite and for tracing Mg in low temperature alteration processes based on the distinct signatures of secondary silicate and carbonate minerals.
... Since no significant Mg isotopic fractionation occurs during magmatic differentiation (Lai et al., 2015;Liu et al., 2010;Xiao et al., 2016) and metamorphic processes Wang et al., 2014) on the whole-rock scale, both igneous rocks (such as granites and gabbros) and metamorphic rocks (such as gneisses) (accounting for about 80% of the upper crust volume), as the main rock types in the upper crust, have similar Mg isotopic compositions to their Table 1 Mg isotopic compositions of different reservoirs on the Earth. (Azmy et al., 2013;Fantle and Higgins, 2014;Galy et al., 2002;Hu et al., 2017aHu et al., , 2017bHuang et al., 2015aHuang et al., , 2015bHuang et al., , 2015cImmenhauser et al., 2010;Tang et al., 2018). ...
... The relative consistency of Mg isotopes in the mantle, basalts, and granites of the upper crust indicates that high-temperature magmatic processes are in-efficient in fractionating Mg isotopes (Lai et al., 2015;Liu et al., 2010;Xiao et al., 2016). Partial melting and subsequent crystallization differentiation are main processes of magmatic evolution. ...
As a major element of the lithosphere and minor element of the hydrosphere, magnesium (Mg) migrates and transforms during various geological processes, especially in the low-temperature processes. Magnesium has three stable isotopes, 24Mg, 25Mg, and 26Mg, with the relative abundances of 78.99%, 10.00% and 11.01%, respectively. Mass differences between 24Mg and 26Mg are more than 8%, thus significant mass fractionation could occur under changing chemical-physical conditions in different geological environments. As such, ratios between different Mg isotopes could be powerful indicators and tracers for geological processes. In this review, we summarize geochemical behaviors of Mg isotopes during magmatic, metamorphic, weathering, and diagenetic processes, to constrain the geological cycle of Mg isotopes. Mg isotopic fractionation occurs strongly during weathering process and precipitation and dissolution of carbonates and moderately during magmatism. There is limited Mg isotopic fractionation during metamorphism. A summary of Mg isotopic distribution and geochemical behaviors in the Earth system allow us to establish a Mg isotopic cycling model, and the geological applications to petrology, mineralogy, and mineralization processes are further reviewed. As a promising geothermometer, Mg isotopes are useful to calculate the temperature of different geological processes. Mg isotopes could be used to trace the magmatic history, petrogenesis, material circulation and mantle metasomatism, and genesis of ore deposits. Generally, the Mg isotopic composition of mantle is relatively homogeneous, which is similar to that of igneous rocks and metamorphic rocks. Some basalts and granites show heterogeneous Mg isotopic ratios resulting from incorporation of surficial materials. Carbonatites and carbonate rocks have very large variations in Mg isotopic compositions and are the main sink of light Mg isotopes in the Earth system. Siliceous sediments and sedimentary rocks contain clay minerals and have relatively heavy Mg isotopic compositions than the mantle. Soils and rivers from different regions have heterogeneous Mg isotopic compositions, whereas seawater has generally homogenous Mg isotopic composition. Surficial processes and chemical diffusion play a significant role in producing extremely variable Mg isotopic compositions.
... However, (Sun et al., 2021), potassic/ultrapotassic rocks in Northeast China Tian et al., 2016;Wang et al., 2017a), sodic rocks in Eastern China (He et al., 2019;Huang and Xiao, 2016;Huang et al., 2015;Jin et al., 2020;Li et al., 2017;Yang et al., 2012;Yu et al., 2019). Mantle peridotite and pyroxenite massifs/xenoliths (Hu et al., 2016;Huang et al., 2011;Lai et al., 2015;Liu et al., 2011;Pogge Von Strandmann et al., 2011;Shen et al., 2018;Su et al., 2019;Teng et al., 2010;Wang et al., 2016;Xiao et al., 2013;Yang et al., 2009;Young et al., 2009) Fig. 6b and c) and Ce/Pb ratios (not shown) and no correlation is observed, which indicates that the Mg and Zn isotopic anomaly together with their low Nb/U and Ce/Pb ratios are not caused by wall-rock assimilation, but reflect their source heterogeneity. Furthermore, the presence of entrained spinel lherzolite xenoliths within these lamproites (Sheraton and Cundari, 1980;Collerson and McCulloch, 1983) also indicates they formed from magmas that underwent insignificant crustal contamination. ...
Lamproites are rare mantle-derived peralkaline ultrapotassic rocks, and they are commonly geographically associated with the ultramafic lamprophyres and kimberlites. Their unique geochemistry and mineralogy make determining their mantle source and origin important because of the significance for inferring specific geodynamic processes. In this study, we further examine lamproite petrogenesis using new Mg and Zn isotopic data for the typical Gaussberg lamproites, Antarctica, the source of which were thought to be contributed by recycled crustal materials. Results show that these lamproites have lower δ26 Mg (-0.44‰ to-0.39‰) and higher δ66 Zn (0.36‰ to 0.39‰) values than terrestrial mantle (δ26 Mg =-0.25±0.04‰, δ66 Zn = 0.18±0.05‰). The post-magmatic alteration and crustal contamination as well as fractional crystallization and partial melting cannot account for these anomalous Mg and Zn isotopic values. By contrast, the involvement of sedimentary carbonates which are characterized by light δ26 Mg (average approximately-2.0‰) and heavy δ66 Zn (average ~+0.91‰) values in their mantle source can explain these Mg and Zn isotopic anomalies. Quantitative modelling suggests that addition of 10-15% subducted dolomite into the source of Gaussberg lamproites can well reproduce their Mg and Zn isotopic values. The source component with light Mg and heavy Zn isotopic compositions can either be sub-continental lithospheric mantle metasomatized by carbonate melts or residue of subducted carbonate-bearing sediments after deep melting in the mantle transition zone. A lithospheric mantle contribution is indeed required to explain their strongly enriched radiogenic isotopic compositions. However, in terms of carbonate component, their positive Zr-Hf anomalies (Hf/Hf* = 1.28-2.19), and extremely high K/U (~40, 000) and Ba/Th (~400) ratios lead us to favor the latter deep recycling model in which the recycled carbonate-bearing sediments subducted as K-hollandite and majorite underwent partial melting within the mantle transition zone.
... δ 26 Mgresidual melt = δ 26 Mginitial + 1000 Di ln(F) ln( 26/24 αsolid/melt) c l i and c 0 i are MgO contents in residual and initial melts; F is the liquid mass fraction; Di is Mg distribution coefficient between the solid phase and melt (4.44 for olivine-melt, Beattie, 1994;2.11 for Cpx-melt, Adam and Green, 2006 (Handler et al., 2009;Yang et al., 2009;Huang et al., 2011;Liu et al., 2011;Pogge van Strandmann et al., 2011;Xiao et al., 2013Xiao et al., , 2016Lai et al., 2015;Liu et al., 2017;Stracke et al., 2018). Given the dominant control of olivine on removal of Mg from the evolving liquid and its greater fractionation relative to melt compared to clinopyroxene, the calculations are most strongly controlled by our new value of 26/24 MgOl/melt. ...
... Notably, inferred proximal peridotites (at Seefeld/M) also display markedly lower δ 7 Li values and generally higher Li contents than distal peridotites, which are similar in composition to average depleted mantle (Fig. 4H). Similar patterns were previously reported for other oro-genic peridotites (e.g., Horoman peridotite at Hokkaido, Japan: Lai et al., 2015) and possibly reflect metasomatic input rather than scavenging of Li by dehydration fluids (Marschall and Tang, 2020). Lithium is highly fluid-mobile and quickly diffuses in most mantle minerals, so infiltration of a Li-rich fluid into peridotite should generate a chemical potential gradient that drives diffusion of Li from the fluid into peridotite minerals (Aulbach and Rudnick, 2009;Marschall and Tang, 2020). ...
Metasomatized mantle wedge peridotites exhumed within high-pressure terranes of continental collision zones provide unique insights into crust-mantle interaction and attendant mass transfer, which are critical to our understanding of terrestrial element cycles. Such peridotites occur in high-grade gneisses of the Ulten Zone in the European Alps and record metasomatism by crustal fluids at 330 Ma and high-pressure conditions (2.0 GPa, 850 °C) that caused a transition from coarse-grained, garnet-bearing to fine-grained, amphibole-rich rocks. We explored the effects of crustal fluids on canonically robust Lu-Hf peridotite isotope signatures in comparison with fluid-sensitive trace elements and Nd-Li isotopes. Notably, we found that a Lu-Hf pseudo-isochron is created by a decrease in bulk-rock 176Lu/177Hf from coarse- to fine-grained peridotite that is demonstrably caused by heavy rare earth element (HREE) loss during fluid-assisted, garnet-consuming, amphibole-forming reactions accompanied by enrichment in fluid-mobile elements and the addition of unradiogenic Nd. Despite close spatial relationships, some peridotite lenses record more intense fluid activity that causes complete garnet breakdown and high field strength element (HFSE) addition along with the addition of crust-derived unradiogenic Hf, as well as distinct chromatographic light REE (LREE) fractionation. We suggest that the observed geochemical and isotopic provinciality between peridotite lenses reflects different positions relative to the crustal fluid source at depth. This interpretation is supported by Li isotopes: inferred proximal peridotites show light δ7Li due to strong kinetic Li isotope fractionation (–4.7–2.0‰) that accompanies Li enrichment, whereas distal peridotites show Li contents and δ7Li similar to those of the depleted mantle (1.0–7.2‰). Thus, Earth’s mantle can acquire significant Hf-Nd-Li-isotopic heterogeneity during locally variable ingress of crustal fluids in continental subduction zones.
... With respect to pristine oceanic lithosphere (mantle-like δ 26 Mg; e.g., Lai et al. (2015)), the SCS lithosphere is expected to have a slightly elevated δ 26 Mg value which is equal to those Site U1433 MORBs (average δ 26 Mg = −0.20 ± 0.06‰, 2SD) , and could be the high-δ 26 Mg endmember of Site U1431 basalts. The Site U1431 sample, 10R-4-W 87/90, has the highest δ 26 Mg of −0.11 ± 0.03‰ (2SD) that exceed the range of high-δ 26 Mg endmember. ...
The post-spreading seamount chain formed along the fossil ridge of the South China Sea (SCS) consists of carbonated silicate melts and alkaline basalts. These temporally and spatially linked samples give an intriguing case where CO2-rich magmas and associated basaltic rocks occurred upon the ultra-thin oceanic lithosphere, and thus they could carry robust information on the generation of alkaline magma via melting of carbonated mantle source. Here we present magnesium isotopic data (δ²⁶Mg), and elemental and radiogenic isotope data for the lavas from the on-axis SCS seamount chain. Both the early- and late-stage samples classified based on their formation ages exhibit remarkable δ²⁶Mg variations (−0.53‰ to −0.11‰ and − 0.38‰ to −0.23‰, respectively). The low-δ²⁶Mg signatures found in two groups of the SCS samples indicate a recycled origin for the source carbon inventory. However, the low-δ²⁶Mg endmember of the early-stage samples is characterized by low SiO2 and enrichment in rare earth elements (REEs), whereas the late-stage samples display a low-δ²⁶Mg endmember that produced high-SiO2 and REE-depleted melts. The Mg-Nd isotope systematics suggest that the early-stage lavas were yielded by CO2-enhanced melting of sediment-rich oceanic crust (εNd = 4.2), whereas the late-stage lavas have sourced the residual and carbon-deficient oceanic crust (εNd = 8.7). Our model demonstrates that the melting of carbonated component can produce compositionally distinct carbonatitic and silicate melts in sequence due to different solidus temperatures. Owe to the relatively thin lithosphere, these two types of melt could individually appear along the fossil ridge of the SCS.
... = serpentinite; Ol = olivine, Opx = orthopyroxene, Cpx = clinopyroxene, Ant = antigorite, Serp = serpentine (unspecified), Brc = brucite, Mgs = magnesite, Hmgs = hydromagnesite, Ht = hydrotalcite.As illustrated inFig. 3, Mg isotopic compositions of both bulk tailings and bedrock serpentinite are consistent with pristine terrestrial basalts (e.g.Huang et al., 2015) and peridotites (e.g.Yang et al., 2009;Beinlich et al., 2014;Pogge von Strandmann et al., 2015), as well as bulk silicate Earth (Fig. 3;Teng et al., 2010;Lai et al., 2015;Hin et al., 2017).Mineral separates of olivine, orthopyroxene, clinopyroxene and antigorite from partly serpentinised ultramafic rocks have similar δ 26 Mg as bulk peridotites and serpentinites (Fig. 3) and there is no relationship between Mg isotopic composition and degree of serpentinisation(Beinlich et al., 2014;Liu et al., 2017). This suggests that serpentinisation does not significantly affect mineral or bulk rock Mg isotopic composition(Beinlich et al., 2014;Liu et al., 2017), and, consequently, olivine, pyroxene and serpentine in the tailings can be considered to have similar Mg isotopic compositions.Brucite is a minor (< 3 %, Turvey et al., 2017) but highly reactive component of the mine tailings. ...
In this study we present the first Mg isotope data that record the fate of Mg during mineralisation of atmospheric CO2 in ultramafic mine tailings. At the Woodsreef Asbestos Mine, New South Wales, Australia, weathering of ultramafic mine waste sequesters significant amounts of CO2 in hydromagnesite [Mg5(CO3)4(OH)2·4H2O]. Mineralisation of CO2 in above-ground, sub-aerially stored tailings is driven by the infiltration of rainwater dissolving Mg from bedrock minerals present in the tailings. Hydromagnesite, forming on the surface of the tailings, has lower δ²⁶Mg (δ²⁶MgHmgs = -1.48 ± 0.02 ‰) than the serpentinised harzburgite bedrock (δ²⁶MgSerpentinite = -0.10 ± 0.06 ‰), the bulk tailings (δ²⁶MgBulk tailings = -0.29 ± 0.03 ‰) and weathered tailings containing authigenic clay minerals (δ²⁶MgWeathered tailings = +0.28 ± 0.06 ‰). Dripwater (δ²⁶MgDripwater = -1.79 ± 0.02 ‰) and co-existing hydromagnesite (δ²⁶MgHmgs = -2.01 ± 0.09 ‰), forming in a tunnel within the tailings, and nodular bedrock magnesite [MgCO3] (δ²⁶MgMgs = -3.26 ± 0.10 ‰) have lower δ²⁶Mg than surficial fluid (δ²⁶Mg = -0.36 ‰) and hydromagnesite.
Complete dissolution of source minerals, or formation of Mg-poor products during weathering, is expected to transfer Mg into solution without significant alteration of the Mg isotopic composition. Aqueous geochemical data and modelling of saturation indices, along with Rayleigh distillation and mixing calculations, indicate that the ²⁶Mg-depletion in the drip water, relative to surficial water, is the result of brucite dissolution and/or precipitation of secondary Mg-bearing silicates and cannot be assigned to bedrock magnesite dissolution. Our results show that the main mineral sources of Mg in the tailings (silicate, oxide/hydroxide and carbonate minerals) are isotopically distinct and that the Mg isotopic composition of fluids and thus of the precipitating hydromagnesite reflects both isotopic composition of source minerals and precipitation of Mg-rich secondary phases. The consistent enrichment and depletion of ²⁶Mg in secondary silicates and carbonates, respectively, underpins the use of the presented hydromagnesite and fluid Mg isotopic compositions as a tracer of Mg sources and pathways during CO2 mineralisation in ultramafic rocks.
... fractionations (e.g., $10‰ for d 44/40 Ca, see method part for the definition of d 44/40 Ca) can occur during chemical diffusion (e.g., Richter et al., 1999;Richter et al., 2009;Watkins et al., 2014), thermal diffusion (e.g., Huang et al., 2010a;Richter et al., 2009), and self-diffusion in silicate liquids (Watkins et al., 2014). In terrestrial rocks, large diffusiondriven Li isotope fractionations (>20‰ in d 7 Li) have been documented on the micron scale in minerals from peridotites (e.g., Aulbach et al., 2008;Ionov and Seitz, 2008;Lai et al., 2015;Pogge von Strandmann et al., 2011;Rudnick and Ionov, 2007) and at the meter to tens of meters' scale in peridotites adjacent to magmatic conduits (Lundstrom et al., 2005). Such fractionations could be accounted by the chemical diffusion between mineral and grain boundary melt/fluid (Rudnick and Ionov, 2007) or that driven by changing partitioning behavior during cooling of the rock (Ionov and Seitz, 2008). ...
This study reports Ca-Fe isotope compositions of mineral separates in a set of peridotite xenoliths including ten lherzolites, one harzburgite, and one wehrlite from Hainan Island, south China. Iron isotope compositions show a small variation (<0.15‰ of δ56/54Fe) among olivine (ol), orthopyroxene (opx), and clinopyroxene (cpx). Δ44/40Caopx-cpx (δ44/40Caopx-δ44/40Cacpx) show a large range from 0.00‰ to 1.23‰. Together with data in previous studies, Δ44/40Caopx-cpx in mantle xenoliths worldwide show a negative correlation with the apparent temperatures, which cannot be explained by equilibrium fractionation. Instead, the observed inter-mineral Ca isotope fractionation between opx and cpx (>2‰) in previous studies should be dominated by disequilibrium sub-solidus Ca redistribution between cpx and opx, which is driven by the changes of mantle temperature and chemical composition in pyroxene. The wehrlite, which was formed by reaction of peridotite with an evolved, silica-undersaturated melt, has δ44/40Ca of 3.22‰ and δ56/54Fe of 0.22‰, significantly heavier than the value of fertile mantle. Such heavy isotope signatures in the wehrlite are best explained by diffusion during the metasomatic melt percolating the refractory mantle. In summary, this study reveals that significant kinetic Ca isotope fractionations could occur in the mantle by inter-mineral Ca diffusion during re-equilibration upon cooling and Ca diffusion from melt into peridotite during melt percolation.
... Magnesium isotopes as a geological tracer attract more and more attention. Given usually limited Mg isotope fractionation during silicate magmatic processes (e.g., partial melting and magmatic differentiation) if no abundant spinel and garnet are involved, the majority of silicate igneous rocks have relatively homogeneous Mg isotopic compositions, indistinguishable from the upper mantle (δ 26 Mg = −0.25 ± 0.06‰, 2SD) within the current analytical precision (e.g., Handler et al., 2009;Huang et al., 2011;Lai et al., 2015;Liu et al., 2010;Pogge von Strandmann et al., 2011;Teng et al., 2007, 2010a, Teng, 2017Yang et al., 2009;Young et al., 2009). Meanwhile, sediments yield highly variable Mg isotopic compositions (e.g., δ 26 Mg of carbonates down to −5.57‰; and that of clastic sediments up to +0.92‰, Teng, 2017; reference therein) due to large isotope fractionation during weathering and chemical precipitation. ...
... Furthermore, the majority of the peridotites have been previously measured for their Li isotope composition (Brooker et al., 2004;Pogge von Strandmann et al., 2011;Lai et al., 2015) so that we can explicitly avoid samples affected by kinetic fractionation resulting from diffusional disturbance during xenolith entrainment or as a result of melt percolation, which might possibly affect Ni isotope systematics in a similar manner to Mg (Pogge von Strandmann et al., 2011;see Fig. 1). ...
Nickel is a major element in the Earth. Due to its siderophile nature, 93% of Ni is hosted in the core and the Ni isotope composition of the bulk silicate Earth might inform on the conditions of terrestrial core formation. Whether Earth’s mantle is fractionated relative to the chondritic reservoir, and by inference to the core, is a matter of debate that largely arises from the uncertain Ni isotope composition of the mantle. We address this issue through high-precision Ni isotope measurements of fertile- to melt-depleted peridotites and compare these data to chondritic meteorites. Terrestrial peridotites that are free from metasomatic overprint display a limited range in δ60/58Ni (deviation of 60Ni/58Ni relative to NIST SRM 986) and no systematic variation with degree of melt depletion. The latter is consistent with olivine and orthopyroxene buffering the Ni budget and isotope composition of the refractory peridotites. As such, the average Ni isotope composition of these peridotites (δ60/58Ni = 0.115 ± 0.011‰) provides a robust estimate of the δ60/58Ni of the bulk silicate Earth. Peridotites with evidence for melt metasomatism range to heavier Ni isotope compositions where the introduction of clinopyroxene appears to drive an increase in δ60/58Ni. This requires a process where melts do not reach isotopic equilibrium with buffering olivine and orthopyroxene, but its exact nature remains obscure. Chondritic meteorites have variability in δ60/58Ni due to heterogeneity at the sampling scale. In particular, CI1 chondrites are displaced to isotopically lighter values due to sorption of Ni onto ferrihydrite during parent body alteration. Chondrites less extensively altered than the CI1 chondrites show no systematic differences in δ60/58Ni between classes and yield average δ60/58Ni = 0.212 ± 0.013‰, which is isotopically heavier than our estimate of the bulk silicate Earth. The notable isotopic difference between the bulk silicate Earth and chondrites likely results from the segregation of the terrestrial core. Our observations potentially provide a novel constraint on the conditions of terrestrial core formation but requires further experimental calibration.
... For instance, the Horoman massif peridotites from Japan are interpreted as samples of an ancient mantle wedge infiltrated by slab-derived fluids/melts, leading to enriched geochemical compositions and formation of secondary Opx (e.g., Yoshikawa and Nakamura, 2000). Nevertheless, these massif peridotites preserve mantle-like d 26 Mg values of À0.27 to À0.19‰ (Lai et al., 2015), similar to those of the Avacha sub-arc peridotites analyzed in this study. In comparison, Su et al. (2015) reported a slightly wider d 26 Mg variation (À0.28 to À0.14‰) coupled with a negative trend with d 56 Fe for a suite of Tibetan ophiolitic peridotites and hinted at a role of sub-arc metasomatism. ...
Dehydration of subducting oceanic crust and underlying serpentinized mantle is potentially the most important source of fluids for element cycling between surface-processed materials and the interior of the Earth. Magnesium (Mg) isotopes -have been proposed as a promising tracer of dehydration of serpentinites because of their high Mg contents and distinctive Mg isotopic signatures relative to the mantle, which is difficult to identify using other isotopes. To constrain the primary controls on Mg isotope cycling in subduction zones, here we analyze a suite of well-characterized sub-arc mantle xenoliths from the Avacha volcano in southern Kamchatka arc, Russia. They are comparable in modal and/or chemical composition to serpentinized fore-arc harzburgites and sub-arc harzburgite xenoliths from the western Pacific. Despite evidence for ubiquitous slab-related fluid metasomatism, the 23 spinel harzburgite xenoliths and four pyroxenite metasomatic veins display mantle-like δ26Mg values from −0.30 to −0.21‰. Mineral separates have similar δ26Mg values, with an average of −0.26 ± 0.04‰ (2SD, n = 17) for olivine, −0.23 ± 0.04‰ (2SD, n = 17) for orthopyroxene, and −0.24 ± 0.10‰ (2SD, n = 5) for late-stage interstitial clinopyroxene. The lack of a clear slab-derived δ26Mg signature in the Avacha sub-arc peridotites is probably due to their high MgO contents and low fluid/rock mass ratios during flux melting and metasomatism. We also compare published δ26Mg data for four volcanic arcs distributed worldwide. This inter-arc comparison reveals that the thermal structure of a subduction zone appears to exert a strong control on the efficiency of Mg isotope cycling from the downgoing slab to the mantle wedge and subsequently to the arc magmas; those arcs that permit large-scale dehydration of Mg-rich serpentinized slab mantle, coupled with available channelized fluid pathways, are more likely to produce isotopically diverse arc magmas.
... Magnesium (Mg) is a major element in the silicate Earth and can potentially offer unique insights into the source of metamorphic fluids in subduction zones Teng, 2017). Previous work has indicated that Mg isotopes are homogeneous in peridotitic mantle and their fractionation is limited during mantle melting, magma differentiation, and crustal dehydration during subduction zone metamorphism (Teng et al., 2007(Teng et al., , 2010aLiu et al., 2010a,b;Wang et al., 2015;Lai et al., 2015). In contrast, Mg isotopes are significantly fractionated during chemical weathering (Tipper et al., 2006;Teng et al., 2010b;Huang et al., 2012;Liu et al., 2014), the dehydration of ultramafic rocks, especially the dehydration of serpentinities and the dissolution of carbonates (Chen et al., 2018). ...
Metamorphic fluids produced by dehydration of subducting crust transport mass and energy at the slab-mantle interface in subduction channels. It is commonly assumed that fluid flow is significant in oceanic subduction channels but insignificant in continental subduction channels. This assumption is challenged by a combined study of whole-rock geochemistry, Mg and O isotopes, zircon U-Pb ages and trace elements in coesite-bearing jadeite quartzites from the Dabie orogen, China. Although the target samples were collected from different outcrops in an area of ∼50 km², zircon U-Pb dating yields similar discordia lines with not only consistent upper intercept ages of 1.9–2.0 Ga but also consistent lower intercept ages of 224–235 Ma. This indicates the same Paleoproterozoic protolith and the same Triassic metamorphism for these jadeite quartzites. The O isotope analysis of mineral separates and whole-rock yields variable δ¹⁸O values from 6.3‰ to 9.4‰, indicating involvement of supracrustal components. Except for one outlier at −0.43‰, all the other rocks give variable δ²⁶Mg values from −0.16‰ to 0.61‰, much higher than normal mantle values. The whole-rock Mg isotopes show significant positive correlations not only with MgO contents but also with Rb/La, Rb/Gd and Rb/Nb ratios, but a negative correlation with Na2O contents. These observations indicate that the middle Paleoproterozoic protolith of jadeite quartzites was weathered to produce a kind of sedimentary rocks in a passive continental margin and then underwent significant metasomatism by metamorphic fluids with high δ²⁶Mg values during the continental subduction in the Triassic. The metamorphic fluids were produced by the breakdown of biotite in the metasedimentary rocks during their subduction to subarc depths for ultrahigh-pressure metamorphism. They would have acquired their geochemical compositions not only from the biotite breakdown but also through leaching reaction with the TTG provenance. In view of the spatial occurrences of the target samples, the metamorphic fluids would have flowed inside the continental subduction channel on a large scale. This is the first report of the large-scale fluid flow in the continental subduction zone and therefore demonstrates that fluid flow can be significant in continental subduction channels.
... ment fluid may not adequately explain the Aleutian data. Of particular issue are the lower than MORB-range of d 7 Li values of several samples, which would require depleted mantle d 7 Li values of $2‰, which have not been observed (e.g., d 7 Li values in unaltered peridotite xenoliths range from $+2.5‰ to $+4.9‰; Brooker et al., 2004;Seitz et al., 2004;Magna et al., 2006b;Jeffcoate et al., 2007;Magna et al., 2008;Gao et al., 2011;Pogge von Strandmann et al., 2011;Lai et al., 2015, with an average of 3.8 ± 0.7, Penniston-Dorland et al., 2017. A more likely scenario may be the presence of a sediment melt component that can skew the d 7 Li values of MORB-range samples to slightly lower values than if sediment fluid was the only slab influence (Fig. 9). ...
We report [Li] and δ⁷Li values for a well-characterized suite of 52 geographically (165–184°W), compositionally (SiO2 = 46–70 wt.%), and temporally (0–38 Ma) diverse lavas and intrusive samples. The δ⁷Li in these rocks range from −0.7‰ to +14.2‰, with 32 of the 35 lavas and 12 of the 17 intrusive samples falling within the depleted mantle range (δ⁷Li +1.6 to +5.6‰), as sampled by mid-ocean ridge basalts (MORB). The δ⁷Li values of Aleutian lavas do not exhibit the spatial trends observed in other slab component tracers, nor do δ⁷Li values correlate with any slab component indicators, such as radiogenic isotopes, oxygen isotopes, or trace element ratios such as Cs/La and Th/La. The δ⁷Li values in Aleutian intrusions also do not exhibit temporal trends, however, an overall positive relationship exists between δ⁷Li and Th/Nd. Mixing models for δ⁷Li and ¹⁴³Nd/¹⁴⁴Nd values suggest that Aleutian samples within or above the MORB δ⁷Li range can be explained by addition of <1–2% sediment-derived aqueous fluid and ≤3% sediment melt to depleted mantle; both are required to explain the range in δ⁷Li that is observed. Sediment-derived fluid exerts a stronger control on Aleutian samples having higher δ⁷Li values than the MORB range, while sediment melt skews the Li isotopic compositions of MORB-range samples to slightly lower values than if sediment fluid was the only slab influence. Our study demonstrates that a slab signature may be deciphered via modeling even in arcs where spatial trends in δ⁷Li values and correlations with slab component indicators are lacking.
... Mg sodic basalts from eastern China (gray diamonds) and Chaihe basalts are shown for comparison Huang & Xiao, 2016;Li et al., 2017;Qian et al., 2015;Sun, Teng, et al., 2017;Tang et al., 2006;Wei, 2014;Xue et al., 2019;Yang et al., 2011;Yang et al., 2012). Yellow dots represent new measured δ Lai et al. (2015) and Teng (2017). Modeled pyroxenite melting and peridotite melting trends are modified from Zhong et al. (2017) with newly calculated La/Sm. ...
Low-δ26Mg basalts are commonly interpreted to represent melts derived from carbonated mantle sources. The mantle domain feeding low-δ26Mg Cenozoic basalts in eastern China overlaps the so-called Big Mantle Wedge (BMW) above the stagnant Pacific slab in the mantle transition zone, which indicates that the BMW is an important carbon reservoir generated by the slab. However, Mg isotopic composition in the nearby mantle beyond the BMW, and thus the spatial extent of carbonated components in the mantle beneath eastern Asia have not yet been extensively characterized. Therefore, it remains largely unconstrained if additional or alternative carbon reservoirs exist. Here we carried out a geochemical study on Cenozoic Huihe nephelinites, which crop out ~500 km west of the present-day BMW. These rocks are characterized by negative K, Zr, Hf, and Ti anomalies, high Zr/Hf, Ca/Al ratios and low δ26Mg values, which suggest that they are derived from a carbonated mantle source. The composition of the nephelinites demonstrates that low δ26Mg mantle components exist at significant distances from the present-day BMW, which highlights that in addition to the stagnant Pacific slab, other oceanic slab(s) also contribute carbonate-bearing crustal materials into the mantle sources of eastern Asia’s Cenozoic volcanism.
... Again, this model is in conflict with the high Zn concentrations and d 66 Zn ratios in Eastern China basalts (Fig. 3), since melts from the pyroxenites with higher d 66 Zn ratios, if formed via diffusion, should have relatively lower Zn concentrations in the absence of a significant difference in melting degrees. On the other hand, since diffusion will be very fast at high temperatures [101,106], it is unclear whether the possible diffusion-driven isotope effects of such sources could be preserved at mantle temperatures. Future simulation studies are required in order to quantitatively evaluate whether the predicted, isotopically abnormal sources could be preserved over the interval between pyroxenite formation and subsequent melting to produce basaltic melts. ...
The subduction of marine carbonates and carbonated oceanic crust to the Earth’s interior and the return of recycled carbon to the surface via volcanism may play a pivotal role in governing Earth’s atmosphere, climate, and biosphere over geologic time. Identifying recycled marine carbonates and evaluating their fluxes in Earth’s mantle are essential in order to obtain a complete understanding of the global deep carbon cycle (DCC). Here, we review recent advances in tracing the DCC using stable isotopes of divalent metals such as calcium (Ca), magnesium (Mg), and zinc (Zn). The three isotope systematics show great capability as tracers due to appreciable isotope differences between marine carbonate and the terrestrial mantle. Recent studies have observed anomalies of Ca, Mg, and Zn isotopes in basalts worldwide, which have been interpreted as evidence for the recycling of carbonates into the mantle, even into the mantle transition zone (410–660 km). Nevertheless, considerable challenges in determining the DCC remain because other processes can potentially fractionate isotopes in the same direction as expected for carbonate recycling; these processes include partial melting, recycling of carbonated eclogite, separation of metals and carbon, and diffusion. Discriminating between these effects has become a key issue in the study of the DCC and must be considered when interpreting any isotope anomaly of mantle-derived rocks. An ongoing evaluation on the plausibility of potential mechanisms and possible solutions for these challenges is discussed in detail in this work. Based on a comprehensive evaluation, we conclude that the large-scale Mg and Zn isotope anomalies of the Eastern China basalts were produced by recycling of Mg- and Zn-rich carbonates into their mantle source. Keywords: Deep carbon cycle, Calcium isotopes, Magnesium isotopes, Zinc isotopes
... Lithium elemental and isotopic data suggest that diffusion is an important mechanism controlling Li abundances and isotopic distribution in the lithospheric mantle (e.g., Su et al., 2014;Tomascak et al., 2016). This heterogeneity could be caused by kinetic Li isotope fractionation due to inter-mineral Li diffusion (e.g., Tang et al., 2007;Su et al., 2012Su et al., , 2014 or Li diffusion from melt into rocks (e.g., Nishio et al., 2004;Jeffcoate et al., 2007;Rudnick and Ionov, 2007;Weyer and Seitz, 2012;Lai et al., 2015). This mechanism can account for the observations of heterogeneous compositional and isotopic signatures in olivine from the clinopyroxene-bearing chromitites (KZ14-38 and KZ15-38). ...
Chromitites in the Kızıldağ ophiolite mostly occur in the mantle harzburgites and some of them in crustal cumulate dunites. Olivine in both the dunites and chromitites is highly magnesian (Fo = 90.5–92.8 and 90.2–96.1, respectively), while the coexisting chromite varies widely in composition (Cr# = 52.9–68.6 and 57.1–76.2, respectively), placing it between high-Al and high-Cr varieties. Concentrations of Ni, V and Ga in the chromite (Ni: 741–1310 ppm; V: 521–1227 ppm; Ga: 20–51 ppm) are also highly variable, straddling the range of high-Al and high-Cr chromite. Olivine in the dunites and chromitites has a large range of Li concentrations (0.27–2.47 ppm) and δ⁷Li values (−7.20–29.90‰). Based on the calculated Al2O3 (12.1–17.2 wt.%) contents, FeO/MgO values (0.92–1.57) and trace element concentrations of the parental magmas of the dunites and chromitites, it is suggested that they are transitional between MORB and boninitic melts, and should have been derived by mixing of geochemically and spatially distinct magmas generated by partial melting of different sources. The Li isotopic compositions of olivine from the dunites and chromitites (except KZ14-38 and KZ15-38) are neither controlled by diffusion nor fractional crystallization, and thus may be intrinsic features of their parental melts. The range of δ⁷Li values of olivine in the dunites and chromitites is analogous to that of arc lavas and marine sediments, indicating that the chromitites record various degrees of melt penetration. The wide range of Li isotopic variations in olivine and chemistry in chromite from the Kızıldağ chromitites suggest that the parental melts were originated in a proto-forearc mantle during subduction initiation.
... Therefore, the bulkrock Li isotopic system of LBS-05 can be considered to be closed after refertilization. Given that Li isotopes can work as effective geo-speedometers due to their rapid diffusivities (Coogan et al., 2005;Gao et al., 2011;Pogge von Strandmann et al., 2011;Lai et al., 2015), the compositions of olivine and clinopyroxene in the sample LBS-05 are selected for modeling the inter-mineral Li diffusion in order to quantitatively constrain the temporal evolution of the Luobusa ophiolite. Although relevant modeling has been done before (Parkinson et al., 2007;Gallagher and Elliott, 2009;Halama et al., 2009;Gao et al., 2011), further improvements are still needed. ...
The Li-O isotopes of olivine and pyroxene from the mantle sequence of the Luobusa ophiolite, a fragment of Neo-Tethyan forearc lithosphere in Tibet, reveal a series of magmatic processes and geochemical evolution in a nascent mantle wedge during subduction initiation. Olivine grains from the ophiolitic mantle sequence display δ¹⁸O values similar to those of normal mantle (+5.18 ± 0.2‰), but have large δ⁷Li variations, ∼+4‰ to +13‰ in harzburgites and ∼+1‰ to +11‰ in high-Cr chromitites. Clinopyroxene grains in cpx-bearing harzburgites have finger-like protrusions and LREE-depleted patterns comparable to those of clinopyroxene in abyssal peridotites, and also show high Li concentrations (∼2–8 μg/g) and negative δ⁷Li values (∼−15‰ to −8‰), which are beyond the ranges of normal upper mantle (<1.3 μg/g and +4 ± 2‰, respectively). Our dataset suggests that the nascent Luobusa harzburgitic mantle wedge was penetrated by limited amounts of slab-derived fluids at an early stage, leading to decoupling of high δ⁷Li but normal mantle-like δ¹⁸O values of the mantle rocks. The hydrated harzburgites were later refertilized by depleted mantle-derived melts. This process generated the cpx-bearing harzburgites with light-δ⁷Li clinopyroxene, and also imposed normal mantle-like Li isotopic signatures on some previously hydrated high-δ⁷Li mantle areas at deeper levels. Subsequent melting in both the high-δ⁷Li and normal mantle-like Li isotopic mantle regimes generated the parental magmas of the high-Cr chromitites with high-Ca boninitic affinities, accounting for the large δ⁷Li variation of olivine in the chromitites. Slab rollback below the nascent mantle wedge induced asthenospheric upwelling and a high geothermal gradient, which played a more direct role than slab dehydration in triggering the magmatic events of refertilization and formation of the high-Cr chromitites. Numerical modeling of inter-mineral Li diffusion in cpx-bearing harzburgites suggests an interval of <1000 years between the two magmatic events and rapid slab rollback before the generation of the high-Cr chromitites. Intra-oceanic emplacement of the Luobusa proto-forearc lithosphere was possibly finished within 10⁶ years after the chromitite formation.
... The δ 26 Mg values of Tedside basalts range from −0.40 to −0.31‰, clearly lower than those of the late-stage Pitcairn lavas (−0.30 to −0.16‰), peridotitic mantle [δ 26 Mg = −0.23 ± 0.04‰ (20) or δ 26 Mg = −0.25 ± 0.04‰ (21)], and other typical OIBs (Hawaii, Louisville, and Rarotonga) (Fig. 2). The lower end of the Pitcairn range is the lowest value observed so far in OIBs (see the compilation in ref. 21) and is close to the upper bound of the continental intraplate EM1 end member basalts from Wudalianchi and Erkeshan [hereafter referred to as "intracontinental EM1 basalts"; δ 26 Mg = −0.57 ...
Significance
Lavas from Pitcairn Island are the best candidates for exploring the origin of the enigmatic EM1 component found in some mantle plumes because they show the most extreme isotopic compositions of Sr, Nd, Hf, and Pb that define the EM1 component. We find that these lavas have the lowest δ ²⁶ Mg values so far recorded in oceanic basalts. Subducted late Archean dolomite-bearing sediments are the most plausible source of the low-δ ²⁶ Mg feature of Pitcairn lavas. This requires that an ancient, originally sedimentary component has been emplaced near the core–mantle boundary to ultimately become part of the Pitcairn plume source.
... Magnesium isotopic compositions of the bulk rocks and mineral grains including clinopyroxene, orthopyroxene and magnetite are listed in Table 3. The δ 26 Mg values of the andesites range from −0.26 ± 0.05‰ to −0.15 ± 0.04‰, with the average value of −0.18 ± 0.05‰, slightly heavier than that of the normal mantle (−0.25 ± 0.07‰; Huang et al., 2011;Lai et al., 2015;Teng et al., 2010a), but similar to that of the oceanic arc lavas (average δ 26 Mg = −0.18 as reported by Teng et al. (2016) , Fig. 9a). The clinopyroxene and orthopyroxene shows δ 26 Mg values of −0.27 ± 0.03‰ to −0.20 ± 0.04‰ with an average value of −0.24 ± 0.04‰, and − 0.24 ± 0.06‰ to −0.18 ± 0.03‰ with an average value of −0.22 ± 0.04‰, respectively, whereas the δ 26 Mg values of magnetite range from +0.09 ± 0.06‰ to +0.33 ± 0.06‰ with an average value of +0.23 ± 0.06‰. ...
Continental crust can dramatically modify the geochemical and isotopic compositions (e.g., Sr-Nd, Pb) of mantle-derived lavas, and has important implications in understanding magmatic processes in continental arcs which involve subducted materials. In this paper, we report the Mg isotopic compositions of continental arc andesites from El Laco in northern Chile, and evaluate the contribution of the subducted slab to the formation of continental arc lavas. The andesites in the El Laco volcanic complex (ELVC) display relatively high (⁸⁷Sr/⁸⁶Sr)i ratios and negative age-corrected εNd(t) (t = 1.6 Ma) values. The δ²⁶Mg values of the ELVC andesites range from −0.26 ± 0.05‰ to −0.15 ± 0.04‰ (average δ²⁶Mg = −0.18 ± 0.05‰), slightly heavier than that of the primitive mantle. The δ²⁶Mg values of clinopyroxene, orthopyroxene and magnetite grains separated from andesites are −0.27 ± 0.03‰ to −0.20 ± 0.04‰ −0.24 ± 0.06‰ to −0.18 ± 0.03‰ and +0.09 ± 0.06‰ to +0.33 ± 0.06‰ respectively. Our results suggest that Mg isotope fractionation occurred during the fractional crystallization of El Laco andesite, and imply complex Mg isotopic fractionation at high-temperature conditions than previously inferred. Combing the Mg isotopes with the geochemical and Sr-Nd isotopic data, we confirm that neither the deep process of partial melting nor the shallow process of crust contamination contributed to the heavier Mg isotopes of the andesites. The heavier Mg isotopic features of these lavas were mainly inherited from aqueous fluids derived from the subducted slab, which reacted with the primitive mantle wedge to produce a metasomatic mantle source for the continental arc lavas.
GD Anadolu Orojenik Kuşağı Ofiyolitlerinin Manto Peridotitleri
Bileşimleri ve Petrolojisi: Ana Element, İz Element Jeokimyası, Mineral
Kimyası ve Fe, Mg ve Os İzotopları
Petrogenesis of two-mica granites and the associated Li mineralization from the Jiajika super-large Li deposit in Sichuan Province, southwest China, remains equivocal. This study analyzed Li contents and isotopic compositions of metasedimentary rocks (schists and slates) and two-mica granites in this deposit to investigate the issue. The metasedimentary rocks show large variations in Li (71.7 ppm to 1792 ppm) and δ⁷Li (–14.1 ‰ to +7.9 ‰), while the two-mica granites have Li contents from 60.2 ppm to 388 ppm and δ⁷Li values from –3.1 ‰ to +1.9 ‰. Integrated with bulk geochemistry and available Nd–Hf isotopes from this deposit, the two-mica granites probably originated from partial melting of the middle-upper crustal pelitic metasediments. The lower average δ⁷Li value of metasedimentary rocks (–2.9 ‰) than that of two-mica granites (–0.4 ‰) is possibly attributed to ⁶Li-rich garnets entering residuum during incongruent melting. We speculate that Li isotopic fractionation might have occurred during crustal anatexis. The partial melting that contributed to the transfer of volatiles and rare-metal elements into the granitic melts played an important role in Li mineralization. Li isotopes can potentially guide Li deposit prospecting as Li-bearing veins are likely to occur in regions with light ⁶Li.
Lithium isotopes are increasingly used as a tracer for the sources of mafic-ultramafic intrusive rocks, but the validity of such approach has never been evaluated directly. We have addressed this issue by analyzing mineral separates from an arc-type mafic-ultramafic complex (namely Xiarihamu) in the Tibetan plateau for Li isotopes. The Xiarihamu complex consists of a gently-dipping, ultramafic chonolith in the center and older gabbro in the margins. The chonolith was interpreted to be a feeder for olivine-charged magma by previous studies. Our new data show minor to strong isotopic disequilibrium between coexisting minerals in the chonolith, with lower δ⁷Li in clinopyroxene than in coexisting orthopyroxene by up to 7‰. The total range of δ⁷Li in the pyroxene and olivine separates together is from 5.0‰ to −16.2‰, which is significantly different from that of known arc lavas in the world (3.6 ± 1.2 ‰). Samples from a subvertical drill core of ~190 m in length penetrating the chonolith show a concentric decrease of δ⁷Li in orthopyroxene from −16.2‰ in the middle to +1.5‰ and + 3.8‰ in both ends. The δ⁷Li values of mineral separates show no correlation with whole-rock SrNd isotopes and incompatible trace element ratios, as well as olivine forsterite contents and orthopyroxene Mg/(Mg + Fe) molar ratios, indicating that the observed great variability of δ⁷Li in the minerals was not caused by variable subduction inputs or magma differentiation alone. A positive correlation exists between orthopyroxene δ⁷Li and whole-rock LOI, illustrating that serpentinization-talc alteration caused δ⁷Li in the residual mineral to increase, contributing to the great variability of δ⁷Li on the mineral scales. Subducting sediment input during magma generation and crustal contamination together cannot adequately account for the observed extremely low δ⁷Li (−2 to −16.2‰) in the minerals from many of the rock samples as well as the unusually low average δ⁷Li (−3.5‰) of all of the samples, because the δ⁷Li of the crust and subducting sediments are much higher. Kinetic fractionation of Li isotopes in response to a temperature gradient across the chonolith as well as to subsequent diffusion of Li from interstitial melt to the mineral phases is the best explanation. Since the kinetic fractionation is independent of tectonic settings, such effect is inevitably present in all mafic-ultramafic intrusions on Earth, illustrating that the effect of kinetic Li isotope fractionation must be considered before using Li isotopic data to constrain mantle sources and magmatic evolution, especially if mineral rather than whole-rock data are employed.
Titanium isotopes are a promising novel tracer for magmatic processes. The Ti isotope composition of the Earth's mantle is key for using this tracer on a planetary scale, however, recent studies point to potential Ti isotope variations in mantle rocks. This study presents stable Ti isotope data for well-characterised peridotites from the Horoman massif (Japan) as well as for primitive, orogenic (Variscan and Alpine belts) and South African ultrapotassic rocks. The peridotites originate from a continuous section across the layered Horoman peridotite body and comprise highly depleted harzburgites to fertile lherzolites as well as metasomatically overprinted peridotites. Unlike most primitive mantle-derived magmas, which display very limited Ti isotope variations, δ⁴⁹Ti (deviation of ⁴⁹Ti/⁴⁷Ti from the OL-Ti standard) values of the Horoman peridotites vary from –1.523 ± 0.029 to 0.547 ± 0.015 ‰ (2SD), spanning a total δ⁴⁹Ti range of 2.07 ‰. Heavy, continental crust-like Ti isotope compositions are correlated with Nb/Th and are likely related to metasomatism of the mantle wedge above the Hidaka subduction zone. In such a scenario, isotopically heavy Ti was inherited from the subducted terrigenous sediments and might have been further enhanced by Ti isotope fractionation during mobilisation from the slab. Significant δ⁴⁹Ti differences between coexisting clino- and orthopyroxene (Δ⁴⁹Tiopx-cpx = 0.16–0.29 ‰) and notably light Ti isotope compositions (δ⁴⁹Ti = -1.523 ± 0.029 to -0.677 ± 0.036 ‰, 2SD) recorded in highly refractory peridotites might be related to kinetic diffusion-driven isotope fractionation during fluid/melt percolation. The analysed ultrapotassic rocks display distinctly high δ⁴⁹Ti values (0.114 ± 0.040 to 0.290 ± 0.030 ‰, 2SD) compared to the majority of primitive mantle-derived magmas. The observed enrichment in heavy Ti isotopes likely originates from recycled continental material characterised by high δ⁴⁹Ti, in line with their derivation from strongly metasomatised mantle sources. Our results demonstrate the presence of significant small-scale Ti isotope heterogeneity within the upper mantle.
The foundations of stable isotope geochemistry were laid in 1947 by Urey’s classic paper on the thermodynamic properties of isotopic substances and by Nier’s development of the isotope ratio mass spectrometer (IRMSIsotope Ratio Mass Spectrometer (IRMS)). Before discussing details of the naturally occurring variations in stable isotope ratios, it is useful to describe some generalities that are pertinent to the field of non-radiogenic isotope geochemistry as a whole.
Mantle peridotites exposed on the Earth's surface are “fossilized” mantle materials. Records of pressure-temperature-deformation histories (P-T-d-t paths) preserved in mantle peridotites reveal dynamic motion within the mantle with association of deformation, changes in pressure and temperature, and magma-related processes. Reconstruction of such P-T-d-t paths for mantle peridotites backward from the moment of “fossilization” through exhumation as far back in time as possible, will expand the current understanding of the long-term dynamic behavior of the mantle. The Horoman peridotite complex in the southern Hidaka metamorphic belt of Hokkaido, Japan, has long been examined using various approaches to decode past P-T-d-t paths recorded in the peridotite and mafic rocks. Here we critically review previous work related to reconstructions of P-T-d-t paths for the Horoman peridotite complex, and use the data to outline the dynamic history of the complex from the oldest events until their final ascent leading to emplacement within the crust.
Despite numerous studies, the origin of intraplate basalts from eastern Asia is still elusive. This includes the relative importance of subducted oceanic crust and metasomatized lithospheric mantle in generation of these lavas. To address this important issue, we have carried out an integrated Mg-Sr-Nd isotope study of Jiaohe garnet pyroxenite xenoliths and Cenozoic basalts from the Changbaishan-baoqing volcanic belt (CVB), NE China. The Jiaohe garnet pyroxenite xenoliths exhibit low Hf/Hf* ((HfN/(SmN × NdN)0.5, 0.7–0.9), depleted SrNd isotopic compositions (⁸⁷Sr/⁸⁶Sri = 0.7035–0.7041; εNd = +2.9 − +8.6) and low δ²⁶Mg (−0.61 to −0.39‰). These features reflect both protolith heterogeneity and the effect of isotopic exchange with carbonates during subduction. The CVB basalts display variable Mg isotopic compositions. Low-Si basalts are characterized by low δ²⁶Mg (−0.27 to −0.44‰), variable Hf/Hf* (0.7–1.4) and depleted SrNd isotopic compositions (⁸⁷Sr/⁸⁶Sri = 0.7039–0.7050; εNd = +1.0 − +4.7), which were attributed to partial melting of upwelling asthenospheric mantle containing recycled oceanic crustal components. In contrast, high-Si basalts display low Hf/Hf* (0.6–0.9), EM1-like SrNd isotopic compositions (⁸⁷Sr/⁸⁶Sri = 0.7047–0.7054; εNd = −2.4 - -0.2) and normal mantle-like δ²⁶Mg values (−0.18 to −0.33‰). Such features can be explained by participation of metasomatized lithospheric mantle in their source. The transition from low-Si melts with low-δ²⁶Mg to high-Si melts with normal mantle-like δ²⁶Mg for the CVB basalts are genetically related to continental rifting, asthenospheric mantle upwelling and mechanical-chemical erosion of basal metasomatized lithosphere induced by subduction of the Pacific oceanic slab. Recycled oceanic crust and basal lithospheric mantle both played a fundamental role in generation of the intraplate basalts in eastern Asia.
Garnet-bearing peridotites commonly occur in the deeper parts of mature or thickened oceanic lithosphere, and are rarely exhumed and emplaced onto the seafloor. The Purang ophiolitic peridotites in south Tibet contain rare symplectite pseudomorphs after garnet, offering a unique window into the still poorly understood evolution of the deep oceanic lithosphere. Here, integrated petrologic and Os-Li isotopic data are used to constrain the evolution and dynamics of emplacement for these garnet peridotite protoliths. The Purang peridotites show wide variations of chemical compositions (spinel Cr#: 0.2−0.8) and Os model ages (up to 2.0 Ga), thus representing a piece of heterogeneous oceanic mantle lithosphere. Dunite channels show two distinctive groups of Cr# of spinels and Os-isotope compositions, with the low- to medium-Cr# (0.2−0.6) and high-Cr# (0.7−0.8) dunites reflecting the reaction of host lherzolites/harzburgites with percolating mid-ocean ridge basalt−like and boninitic melts, respectively. This confirms recent subduction initiation-related melt percolation in the Purang peridotites. Coexisting olivines and pyroxenes in the peridotites show systematic Li elemental and isotopic disequilibrium, suggesting fast cooling of the peridotites to Li closure temperature shortly after the melt percolations, likely during exhumation of the peridotites onto the seafloor. This supports a close link between subduction initiation and tectonic emplacement of the Purang peridotites. Combined with other geological evidence, we suggest the Purang peridotites may originate from the deep part of old, thick oceanic lithosphere of the Neo-Tethys. This thick oceanic lithosphere was progressively weakened and thinned likely during widespread plume-lithosphere interaction, triggering the transformation of garnet peridotite protoliths to spinel peridotites. Subsequently, initiation of a new subduction zone along the lithospheric weakness caused rapid ascent and emplacement of the Purang peridotites at a nascent forearc.
This study presents lithium and magnesium isotope ratios of soils and their drainage waters from a well-characterised weathering experiment with two soil cores, one with olivine added to the surface layer, and the other a control core. The experimental design mimics olivine addition to soils for CO2 sequestration and/or crop fertilisation, as well as natural surface addition of reactive minerals such as during volcanic deposition. More generally, this study presents an opportunity to better understand how isotopic fractionation records weathering processes. At the start of the experiment, waters draining both cores have similar Mg isotope composition to the soil exchangeable pool. The composition in the two cores evolve in different directions as olivine dissolution progresses. Mass balance calculations show that the water δ²⁶Mg value is controlled by congruent dissolution of carbonate and silicates (the latter in the olivine core only), plus an isotopically fractionated exchangeable pool. For Li, waters exiting the base of the cores initially have the same isotope composition, but then diverge as olivine dissolution progresses. For both Mg and Li, the transport down-core is significantly retarded and fractionated by exchange with the exchangeable pool. This observation has implications for the monitoring of enhanced weathering using trace elements or isotopes, because dissolution rates and fluxes will be underestimated during the time when the exchangeable pool evolves towards a new equilibrium.
Like many continental flood basalt (CFB) provinces in the world, the source mantle compositions and melting conditions for generation of the Emeishan CFB are highly debated. We have carried out an integrated Sr-Nd-O-Mg isotopic study of the Emeishan picritic lavas to evaluate possible effects of mantle heterogeneity on the magma composition. Moreover, we have used PRIMELT3 (Herzberg and Asimow, 2015) to calculate the primary magma compositions and melting conditions for these picritic lavas. Picrites from Maoniuping, Tanglanghe and Wuguijing exhibit depleted Sr-Nd isotopic compositions (⁸⁷Sr/⁸⁶Sri = 0.7038–0.7050; εNd = 0.27–3.89), normal mantle-like δ²⁶Mg (−0.27 to −0.2‰) and δ¹⁸O (5.0–5.4‰), which can be explained by partial melting of plume materials with negligible contribution of sedimentary carbonates. The Wulongba picrites are characterized by depleted Sr-Nd isotopic compositions (⁸⁷Sr/⁸⁶Sri = 0.7041–0.7045; εNd = 2.57–2.85), slightly lower δ²⁶Mg (−0.36 to −0.23‰) and higher δ¹⁸O (5.6–5.7‰), which were attributed to involvement of minor sedimentary carbonates (magnesite) in their mantle source. The calculated mantle potential temperatures for the Emeishan picrites range from 1516 to 1596 °C. The initial melting pressures range from 4 to 5 GPa. The final melting pressures for the Wulongba picrites (2–3 GPa) are lower than those for the Tanglanghe, Maoniuping and Wuguijing picrites (>3 GPa). This suggests that the Wulongba picrites were generated in shallower garnet-spinel transition zone of the upper mantle. Picrites from other three localities are entirely generated in garnet stability field of the upper mantle.
Nucleosynthetic production of Fe in massive AGB stars (and in supernovae) generated five Fe isotopes, namely ⁵⁴Fe, ⁵⁶Fe, ⁵⁷Fe, ⁵⁸Fe, and ⁶⁰Fe, with long enough half-lives to be extant during the formation of the Solar System.
23.0±1.2 Ma was obtained by Rb-Sr method for a phlogopite-bearing spinel lherzolite from the Horoman peridotite complex in the Hidaka metamorphic belt, Hokkaido, Japan. The age is essentially identical to the oldest ages of the country metamorphic rocks reported so far in the Hidaka metamorphic belt, and may indicate the time of a metasomatic event which occurred in the wedge mantle. This metasomatic event may have occurred during uplift of this mantle fragment under mantle conditions contemporaneously with elevation of the Hidaka metamorphic belt resulting from collision between the Eurasian and North American plates.
An investigation of the petrology and geochemistry of peridotites and gabbros in the Horoman massif, Hokkaido, Japan was undertaken to constrain geochemical processes in the upper mantle.Two types of sample were studied: one type comprises peridotites and gabbros forming thin layers varying from a few millimeters to centimeters in scale (thin-layer peridotites and gabbros); the other comprises thick layers (>1m scale; massive peridotites and gabbros). There is no clear trace element evidence for metasomatism in the thin-layer peridotites. Instead, they have melt-rock reaction textures interpreted in terms of the formation of secondary pyroxene at the expense of primary porphyroclastic olivine and dissolution of primary porphyroclastic pyroxene to form secondary olivine. The thin-layer gabbros also exhibit no metasomatic effects; they have incompatible element depleted trace element characteristics and mid-ocean ridge basalt (MORB)-like isotopic signatures consistent with the presence of a new type of gabbro that previously has not been described from the Horoman Massif.The whole-rock chemistry of the thin-layer peridotites and thin-layer gabbros can be explained by melt-peridotite reactions between isotopically highly depleted MORB mantle (represented by the thin-layer peridotites) and melt with geochemical affinity to Pacific MORB (represented by the thin-layer gabbros). Sm-Nd and Lu-Hf isotope systematics suggest that these reactions might have occurred at ~300 Ma. Some of the plagioclase lherzolites and all of the spinel lherzolites and harzburgites within the massive peridotites show enrichment in incompatible trace elements and more radiogenic Hf-Nd-Pb isotopic compositions than the incompatible-element depleted thin-layer peridotites. The analyzed massive gabbros are interpreted as subduction-related magmas formed in a MORB-source mantle wedge, which have subsequently interacted with a fluid or melt in the Hidaka subduction zone. Hf-Nd-Pb isotope systematics reveal that this interaction may have occurred at an age younger than~50 Ma. Meltand fluid-related processes occurring in the upper mantle are systematically identified from the samples of the Horoman Massif based on petrography, major and trace element, and Sr-Nd-Pb-Hf isotope geochemistry. These processes occurred in different tectonic settings such as the convecting oceanic mantle and supra-subduction zone mantle wedge and have variably modified the original chemistry of residual mantle protolith, formed by partial melting of a depleted MORB source mantle at ~1Ga. © The Author 2010. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: [email protected]
/* */
Lithium elemental and isotopic compositions of 33 glass and whole-rock samples from nine oceanic island regions were determined
to characterize the Li inventory of the deep mantle. The Li contents of the investigated lavas range from 1·5 to 13·3 μg g−1, whereas δ7Li ranges from 2·4 to 4·8‰. There are weak co-variations between the Li/Y, δ7Li, and Sr–Nd–Pb isotope compositions of the lavas, indicating that the Li elemental and isotopic characteristics of ocean
island basalt to some extent reflect mantle source heterogeneity. In detail, HIMU-type lavas are characterized by δ7Li values (up to 4·8‰) slightly heavier than those for average normal mid-ocean ridge basalt (3·4 ± 1·4‰) and by comparatively
low Li contents; EM1-type lavas are characterized by isotopically light Li (average 3·2‰) and relative Li enrichment, whereas
EM2-type lavas tend to heavier δ7Li values (up to 4·4‰) with high Li concentrations. The Li contents and isotope characteristics of HIMU-type lavas are consistent
with recycling of altered and dehydrated oceanic crust, whereas those of the EM1-type lavas can be attributed to sediment
recycling. The Li characteristics of EM2-type lavas may reflect reworking of mantle wedge material that has been infiltrated
by fluids derived from the subducting plate.
We report novel methods for the chemical purification of Mg from silicate rocks by ion-exchange chromatography, and high-precision analysis of Mg-isotopes by high-resolution multiple collector inductively coupled plasma source mass spectrometry (HR-MC-ICPMS). Using these methods, we have measured the relative and absolute Mg-isotope composition of a number of terrestrial and extraterrestrial materials, including international reference rock standards as well as pure Mg standards, olivine crystals separated from a mantle-derived spinel lherzolite (J12 olivine), one enstatite chondrite, a martian shergottite and sea water samples. Repeated analyses of terrestrial and extraterrestrial samples demonstrate that it is possible to routinely measure the relative Mg-isotope composition of silicate materials with an external reproducibility of 2.5 and 20 ppm for the μ26Mg* and μ25Mg values, respectively (μ notation is the per 106 deviation from a reference material). Analyses of bulk mantle-derived rocks as well as a martian shergottite and an enstatite chondrite define a restricted range in μ25Mg of −120 ± 28 ppm (2sd) relative to the DSM-3 reference standard (μ25,26Mg = 0), suggesting that the Mg-isotope composition of inner solar system bulk planetary materials is uniform within the resolution of our analyses. We have determined the absolute Mg-isotope composition of the J12 olivine, two CI chondrites as well as the DSM-3 and Cambridge-1 reference standards using a mixed 26Mg-24Mg double-spike. The differences between the absolute 25Mg/24Mg ratios of the various materials analyzed relative to the DSM-3 standard are in excellent agreement with results obtained by the sample-standard bracketing method. Based on the averages obtained for the J12 olivine separates, we estimate the absolute Mg-isotope composition for Earth's mantle – and hence that of the bulk silicate Earth – to be 25Mg/24Mg = 0.126896 ± 0.000025 and 26Mg/24Mg = 0.139652 ± 0.000033. Given the restricted range of μ25Mg obtained for bulk planetary material by the sample-standard bracketing technique and the excellent agreement between the data obtained by the relative and absolute methods, we propose that these new values represent the absolute Mg-isotope composition of the bulk inner solar system. Using the absolute Mg-isotope composition of the J12 olivine, we calculate the isotopic abundances of Mg as 24Mg = 0.789548 ± 0.000026, 25Mg = 0.100190 ± 0.000018, and 26Mg = 0.110261 ± 0.000023. Based on this result, we have calculated an atomic weight for Mg of 24.305565 ± 0.000045, which is marginally heavier than previous estimates but a factor of 10 more precise.
Re-Os isotopic data for 20 samples from a well-characterized 140 m section across a layered sequence, ranging from plagioclase lherzolite through lherzolite to harzburgite, of the Horoman peridotite show: (1) a range in Os-187/Os-188 ratios (from 0.1158 to 0.1283) similar to that reported for other peridotitic massifs, thereby suggesting that the processes responsible for the Re-Os isotopic variation at the meter-scale and the whole-massif scale are similar; (2) that the Os isotopic ratio is controlled by the Re content through radiogenic ingrowth over a period of similar to0.9 Gy. The ultramafic and some of the mafic rocks (Type I layers)from the Horoman massif define an 'apparent age' of 1.12 +/- 0.24 Ga in the Re-OS isochron diagram, within error of the previously reported age of 833 +/- 78 Ma based on Sm-Nd isotopes. Although the Re-Os isotopic data do not define an isochron, the consistency of the similar to 900 Ma age defined by both isotopic systems suggests that this age has a geologic meaning and that mafic (Type I layers) and ultramafic rocks are genetically related, A plausible explanation for the genetic relationship between the mafic and ultramafic rocks, the meter-scale compositional variations from lherzolite to plagioclase lheriolite, the suprachondritic Re-187/Os-188 ratios in some fertile peridotites, and the oldest RE depletion model age of similar to1.86 Ga obtained for Horoman rocks is a refertilization process involving reaction of a mid-ocean ridge basalt-like magma with depleted lithospheric mantle at similar to 900 Ma.
Two major types of mafic granulite layers occur within the Horoman peridotite, an 8 km × 10 km × 3 km orogenic lherzolite
exposed in the high-T and low-P Hidaka metamorphic belt of Hokkaido, Japan. The mineral assemblages and textures of these
layers reflect subsolidus reactions occurring during uplift from the upper mantle to the crust. Nevertheless, their whole-rock
compositions can be used to infer the primary mineralogy of these layers, and a genetic relationship to melts geochemically
similar to mid-ocean ridge basalts (MORB). The intralayer compositional variation of Type I layers (Al–Ti augite type mafic
granulites) shows that the centers formed as garnet clinopyroxenites in equilibrium with an incompatible element depleted
melt that crystallized to form the margins. In contrast, the Type II layers (Cr-diopside type mafic granulites) formed at
relatively shallow depths and are much older, ∼830 Ma, than the Type I garnet pyroxenites, which formed at ∼80 Ma. The temporal
sequence supports the hypothesis that the Horoman peridotite represents shallow MORB-related oceanic mantle that had subsided
to deeper mantle depths before crustal emplacement.
Major and trace element compositions and isotopic ratios of Sr and Nd were determined for bulk rocks and their constituent clinopyroxenes from the Horoman peridotite complex, Japan. Al2O3, CaO, and heavy rare earth elements (HREE) contents of peridotites generally decrease from plagioclase Iherzolite through spinel lherzolite to spinel harzburgite, indicating simple melt extraction from a single source. However, the extremely large variations in isotopic (87Sr/86Sr=0.7019 to 0.7066, εNd=+110 to -10) and trace element compositions ([Ce/Yb]N=0.006 to 4.0) cannot be explained by a simple melt extraction mechanism. The samples can be divided into two groups: one suite has depleted isotopic and light REE (LREE) characteristics (DP), while the other suite shows enriched isotopic and LREE signatures (EP). Sm-Nd isotope systematics of whole-rock DP samples yield an isochron age of 833+/-78Ma with an initial 143Nd/144Nd ratio of 0.5119+/-2, which is identical to the isotopic composition of mid-ocean ridge basalt (MORB) source mantle at that time. The relationship between MgO and Yb abundances of whole rocks shows that melt extraction was initiated at pressures near the garnet and spinel lherzolite transition. Peridotites that formed at different depths presently occur in close proximity to each other, sometimes within tens of meters. The chemical and isotopic signatures of the EP samples can be explained by mixing between mantle residue and an isotopically and more incompatible element enriched fluid derived from a subducted slab. These observations suggest that the small-scale compositional layering observed in the complex may have formed in a wedge mantle by water-enhanced thinning and folding of metasomatized peridotites which had previously developed large-scale simple stratification as a result of melt extraction beneath a mid-ocean ridge.
Basalts at mid-ocean ridges are generated by partial melting of the Earth's upper mantle. As a result of this process, the upper mantle has become depleted over time in elements that are preferentially removed by melting. Although mid-ocean-ridge basalts have traditionally been thought to reflect the chemical composition of such depleted mantle, recent work has revealed the existence of domains in the upper mantle that are apparently not sampled by the basalts. Here we present the lead (Pb), neodymium (Nd) and hafnium (Hf) isotope compositions of peridotites from the Horoman orogenic massif in Japan, which is considered to represent the residues of melting of the upper mantle. These peridotites exhibit the lowest Pb isotope ratios reported from any known mantle material, along with high Nd and Hf isotope ratios. These data suggest that chemical depletion of the peridotites occurred around a billion years ago, and that they represent ancient mantle domains that have escaped convective stirring and homogenization. We suggest that such domains-if abundant in the mantle-may constitute a hitherto unrecognized reservoir with highly unradiogenic lead.
A potential driving force for cumulate compaction and melt segregation in magmas arises from the change in crystal solubility with temperature. Spatial thermal variation in a cumulus crystal pile sets up a gradient in interstitial melt composition, which in turn provides a diffusion potential for silicate liquid species. Mass transport in response to this potential, referred to as thermal migration, drives adcumulus mineral growth in the cooler regions of a cumulus crystal pile and segregation of interstitial melt toward warmer regions. Heat loss through the cumulate mush zone to the country rock promotes expulsion of intercumulus liquid back toward the magma reservoir. Heat loss from the mush to cooler convective draughts within the chamber promotes trapping of intercumulus liquid. Liquid state Soret diffusion alters this mechanism only in minor chemical detail except when Soret fractionation produces changes in liquid chemistry that exceed those produced by crystal fractionation. This can cause solubility curves to be intersected at high temperature rather than low T, burying the intercumulus liquid in the boundary layer and preventing its escape back into the magma. Quartz saturation in silicic systems may be affected in this way. We demonstrate this process with laboratory experiments in the systems Mg2SiO4-SiO2 and mid-ocean ridge basalt. -from Authors
To investigate the origin of the HIMU (high-μ) reservoir in the mantle, we measured Li, Sr, and Nd isotopic composition of several oceanic island basalts (OIBs) from the Polynesian region. We used a recently developed multiple-collector inductively coupled plasma mass spectrometry method that allows precise and accurate Li isotopic determinations. This study presents the first Li isotopic data on HIMU OIBs. The measured whole-rock δ7Li values (δ 7Li = [ [7Li/6Li]sample/ [7Li/6Li]L-SVECstandard - 1] × 1000) of the Polynesian HIMU OIBs (Mangaia, Tubuai, and Rurutu) range from +5.0‰ to +7.4‰, which are higher than those of fresh normal mid-ocean ridge basalt (N-MORB) glasses (ca.+3‰). The simultaneously measured K/Rb, Ba/Rb, and 87Sr/86Sr ratios indicate that the analyzed HIMU OIBs are free from significant posteruption alteration. These results suggest that the δ7Li value of the Polynesian HIMU source is never lower than those of the N-MORBs. Among the numerous models for the origin of the HIMU source, the most widely accepted model is that it involves subducted (dehydrated) oceanic crust. For this HIMU-origin model, our new Li isotopic results exclude the highly altered portion, that is, the uppermost part of the oceanic crust, because the δ7Li value of subducted highly altered MORB should be extremely low (δ7Li < fresh MORB). For these reasons, we propose that the Polynesian HIMU source is the relatively less-altered oceanic crust underlying the highly altered crust. Whereas Pb, Sr, and Nd isotopic signatures dominantly indicate the involvement of sediments in a source, the Li isotopic signature is more sensitive to the degree of alteration experienced by the basaltic crust and thus can be used to distinguish what part of the crust was trapped in the OIB magma. It therefore provides information complementary to that provided by the radiogenic isotopes.
This study presents a new methodology to obtain highly precise measurements (+/-0.10/00) of magnesium isotope ratios in very small samples of foraminiferal carbonate (40-50 mug). Here this technique is used to examine Mg isotopic variation among different species of core top foraminifera over a range of different ambient conditions. Despite the high degree of temperature control on the abundance of elemental Mg incorporated into foraminiferal tests, analyses of surface dwelling Globigerinoides ruber and G. sacculifer from five cores, with associated sea surface temperatures ranging from 20 to 31°C, show no significant temperature-dependent variations in their Mg isotope ratios. Analyses of different size fractions of G. sacculifer show an increase in Mg/Ca with test size but no variation of Mg isotope ratio. In all, nine planktic and benthic species were analyzed; all show identical Mg isotope ratios with a mean of delta 26Mg = -4.720/00, apart from small differences in three species, namely O. universa, G. sacculifer (which are both ~0.40/00 lighter than the average), and P. obliquiloculata (which is ~0.40/00 heavier). These results highlight the constancy of foraminiferal Mg isotope ratios, despite changing environmental conditions which dominate Mg/Ca variation and arguably affect Ca isotope fractionation. This is an important observation which needs to be included in any model of foraminiferal calcification. The insusceptibility of delta 26Mg values to external parameters makes Mg isotopes ideally suited to constraining past variations in the Mg isotope budget of the oceans and the information this carries about the history of oceanic dolomitization, continental weathering, and hydrothermal behavior.
Constraining the timescales for the assembly and differentiation of planetary bodies in our young solar system is essential for a complete understanding of planet-forming processes. This is best achieved through the study of the daughter products of extinct radionuclides with short half-lives, as they provide unsurpassed time resolution as compared to long-lived chronometers. Here we report high-precision Mg isotope measurements of bulk samples of basalt, gabbro, and pyroxenite meteorites obtained by multiple-collector inductively coupled plasma mass spectrometry (MC-ICP-MS). All samples from the eucrite and mesosiderite parent bodies (EPB and MPB) with suprachondritic Al/Mg ratios have resolvable 26Mg excesses compared to matrix-matched samples from the Earth, the Moon, Mars, and chondrites. Basaltic magmatism on the EPB and MPB thus occurred during the life span of the now-extinct 26Al nuclide. Initial 26Al/27Al values range from (1.26 ± 0.37) × 10-6 to (5.12 ± 0.81) × 10-6 at the time of magmatism on the EPB and MPB, and are among the highest 26Al abundances reported for igneous meteorites. These results indicate that widespread silicate melting and differentiation of rocky bodies occurred within 3 million years of solar system formation, when 26Al and 60Fe were extant enough to induce planetesimal melting. Finally, thermal modeling constrains the accretion of these differentiated asteroids to within 1 million years of solar system formation, that is, prior to the accretion of chondrite parent bodies.
The mantle is Earth's largest reservoir of Mg containing > 99% of Earth's Mg inventory. However, no consensus exists on the stable Mg isotope composition of the Earth's mantle or how variable it is and, in particular, whether the mantle has the same stable Mg isotope composition as chondrite meteorites. We have determined the Mg isotope composition of olivine from 22 mantle peridotites from eastern Australia, west Antarctica, Jordan, Yemen and southwest Greenland by pseudo-high-resolution MC-ICP-MS on Mg purified to > 99%. The samples include fertile lherzolites, depleted harzburgites and dunites, cryptically metasomatised (‘dry’) peridotites and modally metasomatised apatite ± amphibole-bearing harzburgites and wehrlites. Olivine from these samples of early Archaean through to Permian lithospheric mantle have δ25MgDSM-3 = − 0.22 to − 0.08‰. These data indicate the bulk upper mantle as represented by peridotite olivine is homogeneous within current analytical uncertainties (external reproducibility ≤ ± 0.07‰ [2 sd]). We find no systematic δ25Mg variations with location, lithospheric age, peridotite fertility, or degree or nature of mantle metasomatism. Although pyroxene may have slightly heavier δ25Mg than coexisting olivine, any fractionation between mantle pyroxene and olivine is also within current analytical uncertainties with a mean Δ25Mgpyr-ol = +0.06 ± 0.10‰ (2 sd; n = 5). Our average mantle olivine δ25MgDSM-3 = − 0.14 ± 0.07‰ and δ26MgDSM-3 = − 0.27 ± 0.14‰ (2 sd) are indistinguishable from the average of data previously reported for terrestrial basalts, confirming that basalts have stable Mg isotope compositions representative of the mantle. Olivine from five pallasite meteorites have δ25MgDSM-3 = − 0.16 to − 0.11‰ that are identical to terrestrial olivine and indistinguishable from the average δ25Mg previously reported for chondrites. These data provide no evidence for measurable heterogeneity in the stable Mg isotope composition of the source material in the proto-planetary disc from which Earth and chondrite and pallasite parent bodies accreted.
First published as an Advance Article on the web 16th October 2003 Multicollector ICP-MS has been used for the precise measurement of variations in the isotopic composition of the isotopic standard of magnesium (SRM980) provided by the National Institute of Standards and Technology (Gaithersburg, MD, USA). The SRM980 consists of metal chips weighing between 1 and 50 mg and each unit delivered by the National Institute of Standards and Technology corresponds to a bottle containing about 0.3 g. Height units were analysed. Variations in sample 25 Mg/ 24 Mg, and 26 Mg/ 24 Mg ratios are expressed as d 25 Mg and d 26 Mg units, respectively, which are deviations in parts per 10 3 from the same ratio in a standard solution. The differences in d 25 Mg and d 26 Mg of the SRM980 are up to 4.20 and 8.19%, respectively, while the long-term repeatability of d 25 Mg and d 26 Mg are 0.09 and 0.16%, respectively, at 95% confidence. However, when plotted in a three-isotope diagram, all the data fall on a single mass fractionation line. Overall limits of error of the SRM980 reported here fall within the previously reported overall limits of error. The isotopic heterogeneity not only corresponds to differences among units but has been found at the chip-size level. This result, due to the precision of the MC-ICP-MS technique, makes the SRM980 inappropriate for the international isotopic standard of magnesium. The SRM980 can still be used to report the excess of 26 Mg, which is defined by the deviation from the mass-dependent relationship between 25 Mg/ 24 Mg, and 26 Mg/ 24 Mg ratios. Two large batches (around 10 g of Mg in each) of pure Mg solutions (in 0.3 M HNO 3) have been prepared and characterised. These 2 solutions (DSM3 and Cambridge 1) are suitable reference material because they are immune to heterogeneity. DSM3 and Cambridge 1 are isotopically different (by 1.3% per u) and are available upon request from the first author. In addition, DSM3 has an isotopic composition very similar to the Mg-isotopic composition of carbonaceous chondrites (Orgueil and Allende). Because of the lack of heterogeneity and the cosmochemical and geochemical significance of DSM3, we urge the use of DSM3 as the primary isotopic reference material to report Mg-isotopic variations.
Contents: The diffusion equations; methods of solution when the diffusion coefficient is constant; Infinite and semi-infinite media; Diffusion in a plane sheet; Diffusion in a cylinder; Diffusion in a sphere; Concentration-dependent diffusion-methods of solution; numerical methods; Some calculated results for variable diffusion coefficients; The definition and measurement of diffusion coefficients; Non-Fickian diffusion; Diffusion in heterogeneous media; Moving boundaries; Diffusion and chemical reaction; Simultaneous diffusion of heat and moisture.
A precise and simple method for the determination of lithium concentrations in small amounts of silicate sample was developed by applying isotope dilution-inductively coupled plasma-mass spectrometry (ID-ICP-MS). Samples plus a Li spike were digested with HF-HClO4, dried and diluted with HNO3, and measured by ICP-MS. No matrix effects were observed for 7Li/6Li in rock solutions with a dilution factor (DF) of ≥ 97 at an ICP power of 1.7 kW. By this method, the determination of 0.5 μg g-1 Li in a silicate sample of 1 mg can be made with a blank correction of < 1%. Lithium contents of ultrabasic to acidic silicate reference materials (JP-1, JB-2, JB-3, JA-1, JA-2, JA-3, JR-1 and JR-2 from the Geological Survey of Japan, and PCC-1 from the US Geological Survey) and chondrites (three different Allende and one Murchison sample) of 8 to 81 mg were determined. The relative standard deviation (RSD) was typically < 1.7%. Lithium contents of these samples were further determined by isotope dilution-thermal ionisation mass spectrometry (ID-TIMS). The relative differences between ID-ICP-MS and ID-TIMS were typically < 2%, indicating the high accuracy of ID-ICP-MS developed in this study.
Lithium concentrations and isotopic compositions in the country rocks (amphibolites and schists) of the Tin Mountain pegmatite show systematic changes with distance to the contact. Both Li and delta Li-7 decrease dramatically along a similar to 10 m traverse from the pegmatite into amphibolite, with Li concentration decreasing from 471 to 68 ppm and delta Li-7 decreasing from +7.6 to -19.9. Rubidium and Cs also decrease from the pegmatite contact into the country rock, but only within the first 2 m of the contact, after which their concentrations remain constant. Neither mixing between pegmatite fluids and amphibolite, nor Li isotope fractionation by Rayleigh distillation during fluid infiltration is a likely explanation of these observations, due to the extremely light isotopic composition required for the amphibolite end-member in the mixing model (delta Li-7=-20) and the similarly extreme isotopic fractionation required in a Rayleigh distillation model. Rather, these variations are likely due to isotopic fractionation accompanying Li diffusion from the Li-rich pegmatite (Li=450 to 730ppm) into amphibolites (Li=20ppm). The fact that other alkali element concentrations vary only within 2 in of the contact reflects the orders of magnitude faster diffusion of Li relative to heavier elements. Quartz mica schists in contact with the pegmatite also show large variations in both Li and 67 Li as a function of distance from contact (similar to 1 wt.% to similar to 70ppm and +10.8 to -18.6, respectively), but over a longer distance of > 30m. Lithium concentrations of the schist decrease from similar to 1 wt.% adjacent to the contact to similar to 70ppm 300m from the contact; the latter is a typical concentration in metapelites. The nature of the delta Li-7 variations in the schists is different than in the amphibolites. Schists within the first 2m of the contact have nearly identical delta Li-7 of + 10, which mimics that of the estimated bulk pegmatite (+8 to + 11). At a distance of 30m the 67 Li reaches the lowest value in the schists of -18.6 (similar to the lowest amphibolite measured). At a distance of 300m the 6 7 Li climbs back to +2.5, which is within the range of delta Li-7 of other schists in the region and metapelites worldwide. The behavior of Li in the schists can also be modeled by Li diffusion, with the effective diffusion coefficient in the schist being similar to 10 times greater than that in the amphibolite. The effective diffusion coefficients of Li in the amphibolite and schist are > 2 orders of magnitude greater that those in minerals, which implicates the importance of fluid-assistcd grain-boundary diffusion over solid-state diffusion in transporting Li through these rocks. (c) 2006 Elsevier B.V. All rights reserved.
Several recent studies found large lithium isotopic fractionations correlated with concentration gradients in pyroxene minerals from lava flows and mantle nodules that were interpreted as indicating diffusion of lithium into the grains. Motivated by these findings experiments were undertaken in which powdered spodumene (LiAlSi2O6) or Li2SiO3 was used to diffuse lithium into Templeton augite or Dekalb diopside grains at 900 °C and oxygen fugacity ranging from log fO2 = −17 to log fO2 = −12. The purpose of these experiments was to determine the diffusion coefficient of lithium in pyroxene minerals and to measure the isotopic fractionation of lithium in the diffusion boundary layer due to the relative mobility of 6Li compared to 7Li. The diffusion profiles of lithium that had not yet reached the center of Templeton augite grains were in most cases sharp steps propagating in from each boundary. In one case a more usual profile with smoothly decreasing lithium concentration with distance from the grain boundary was found. A model in which lithium occupies two different sites – one being fast diffusing interstitial lithium, the other much less mobile lithium in a metal site, reproduced both types of profiles. The step-like profiles arise in the model when interstitial lithium diffusing into the grain is strongly partitioned into abundant metal sites and thus does not penetrate further into the grain until all the metal sites at a given distance become filled. While the rate of propagation of the concentration step can be used to calculate an effective diffusivity for the penetration of lithium into the augite grains, the multiple speciation of lithium precludes making a separate precise determination of the diffusion coefficient of the interstitial lithium. Isotopic fractionations of 7Li/6Li of about 30‰ were found in the step-like diffusion boundary layers, which translate into a ratio of the isotope diffusion coefficients D7Li/D6Li=0.9592D7Li/D6Li=0.9592 (i.e., (6/7)β with β = 0.27). The same value of β = 0.27 was also able to fit the isotopic fractionation data from the experiment with the smoothly decreasing lithium concentration profile. The laboratory experiments confirmed that diffusion of lithium produces large kinetic isotopic fractionations and thus highlight the importance of isotopic measurements for discriminating when a particular instance of chemical zoning in minerals was the result of diffusion or some other process such crystal growth from an evolving melt. The experiments also showed that, contrary to conventional wisdom, isotopic gradients do not dissipate faster than gradients in the parent element, indeed the contrary is seen in several of the Templeton experiments where very large lithium isotopic fractionations persisted after the lithium concentration had become effectively homogenized. Published data of lithium isotopic fractionation of zoned augite grains from a Martian meteorite and in clinopyroxene grains from both lava flows on the Solomon Islands and from a San Carlos mantle xenolith were modeled in terms of lithium diffusion with an isotope fractionation parameter β between 0.25 and 0.30, which is very similar to that derived from the laboratory experiments.
Laboratory experiments are used to document isotopic fractionation of magnesium by chemical diffusion in a silicate melt and the results compared to the magnesium isotopic composition across contacts between igneous rocks of different composition in natural settings. The natural samples are from transects from felsic to mafic rocks at Vinal Cove in the Vinalhaven Intrusive Complex, Maine and from the Aztec Wash pluton in Nevada. Two laboratory diffusion couples made by juxtaposing melts made from powders of the felsic and mafic compositions sampled at Vinal Cove were annealed at about 1500 °C for 22.5 and 10 h, respectively. The transport of magnesium in the diffusion couples resulted in easily measured magnesium isotopic fractionations at the interface (δ26Mg∼1.5‰). These isotopic fractionations provide a distinctive isotopic “fingerprint” that we use to determine whether chemical gradients in natural settings where melts of different composition were juxtaposed were due to chemical diffusion. The magnesium isotopic fractionation along one profile at Vinal Cove is exactly what one would expect based on the fractionations found in the laboratory experiments. This is an important result in that it shows that the isotope fractionation by chemical diffusion found in highly controlled laboratory experiments can be found in a natural setting. This correspondence implies that chemical diffusion was the dominant process responsible for the transport of magnesium across this particular contact at Vinal Cove. A second Vinal Cove profile has a very similar gradient in magnesium concentration but with significantly less magnesium isotopic fractionation than expected. This suggests that mass transport at this location was only partly by diffusion and that some other mass transport mechanism such as mechanical mixing must have also played a role. The magnesium isotopic composition of samples from Aztec Wash shows no resolvable isotopic fractionation across the contact between the mafic and felsic rocks. The different degrees of magnesium isotopic fractionation associated with otherwise similar composition gradients in natural settings show that kinetic isotope fractionations provide a key discriminator for establishing whether or not molecular diffusion was the process responsible for an observed elemental gradient. In the one case of a contact at Vinal Cove where we are confident that the magnesium elemental and isotopic gradients were produced by diffusion, we deduced a cooling rate of about 1.5 °C per day.
A thirteen kilogram sample of highly purified Li2CO3 prepared from virgin ores has been assayed to have an absolute 6Li/7Li abundance ratio of 0.0832±0.0002. This value was determined by a mass spectrometric comparison with a primary standard blended from specially prepared, highly enriched, well-characterized isotopic materials. The abundance ratio of the primary standard was 0.083656±0.000003. The instrumental bias of the mass spectrometer was determined from isotopic ratio measurements of four primary standards having 6Li abundances in the 7–8% range.
The significant relative mass difference (c. 16%) between the two stable isotopes of Li (approximately 6Li 7.5%, 7Li 92.5%), coupled with broad elemental dispersion in Earth and planetary materials, makes this a system of considerable interest in fingerprinting geochemical processes, determining mass balances, and in thermometry. Natural mass fractionation in this system is responsible for c. 6% variation among materials examined to date (Fig. 1⇓). Although the “modern era” of Li isotope quantification has begun, there are still many questions about the Li isotopic compositions of fundamental materials and the nature of fractionation by important mechanisms that are unanswered (e.g., Hoefs 1997).
Figure 1.
Summary of lithium isotopic compositions of Earth and planetary materials. Filled bars are solid samples, open bars are liquids. See text for references and details.
The purpose of this chapter is to summarize the current understanding of Li isotopes in geo- and cosmochemical systems and to indicate (1) where Li isotopes have a high probability of adding new understanding of these systems; (2) where some of the more significant deficits in knowledge exist. The small but burgeoning Li isotope community has not yet compiled the volume of peer-reviewed literature needed to adequately assess even that which has been studied to date. As a result, significant portions of this chapter are based on data reported in abstracts, and as such are more than normally subject to revisions over time. This chapter is anticipated to serve as a starting point for those interested in research incorporating Li isotope geochemistry, or in understanding the state of extant research.
### Experiments in Li isotope fractionation
Knowledge of significant Li isotopic fractionation during basic chemical processes is long standing. The early experiments by Taylor and Urey (1938), in which Li isotopes were fractionated by incomplete extraction of an aqueous solution from a zeolite exchange …
Li isotopic compositions of magmatic rocks have gained considerable attention recently as probes of mantle-scale processes. However, the concentrations and isotopic composition of Li in mantle minerals from mid-ocean ridges remain relatively unconstrained. This is largely because of the general presence of seawater alteration in abyssal peridotites. Lithium elemental and isotopic compositions for mineral separates of coexisting olivine, clinopyroxene, orthopyroxene and bulk rocks of serpentine-free Gakkel Ridge peridotites were investigated. Bulk rocks have Li contents of 1.6 to 2.7 ppm and δ7Li values of 3 to 5‰, which fall within the range of reported normal pristine “MORB mantle” values. Lithium concentrations vary in the order cpx (2.1-4.7 ppm) > opx (0.9-1.7 ppm) ≥ olivine (0.4-0.9 ppm), the opposite found in “equilibrated” mantle peridotite xenoliths (Seitz and Woodland, 2000). The Li isotopic compositions indicate a systematic mineral variation with δ7Liolivine (7.14‰-15.09‰) > δ7Liopx (1.81‰-3.66‰) > δ7Licpx (−2.43‰ − −0.39‰). The δ7Li values of cpx are negatively correlated with their Li concentrations with the lightest value for the most enriched cpx grains. There is a first order negative linear correlation between Δolivine-cpx (δ7Liolivine − δ7Licpx) and ol/cpxD (Liolivine/Licpx).
An ion-microprobe-based technique has been used to measure lithium tracer-diffusion coefficients (DLi) in an alkali-basaltic melt at 1300, 1350 and 1400°C. The results can be expressed in the form: DLi=7.5 ×10-2exp(-27,600/RT)cm2S-1 The results show significantly faster diffusion rates than those previously recorded for other monovalent, divalent and trivalent cations in a tholeiitic melt. Consequently, diffusive transport of ions acting over a given time in a basaltic melt can produce a wider range of transport distance values than hitherto supposed. Hence, it is concluded that great care should be exercised when applying diffusion data to petrological problems.
To investigate the magnitude and mechanism of inter-mineral Mg isotope fractionation at mantle temperatures, we measured Mg isotopic compositions of coexisting olivine (Ol), orthopyroxene (Opx), clinopyroxene (Cpx), phlogopite (Phl) and spinel (Spl) from harzburgite, lherzolite and clinopyroxenite xenoliths in the North China craton. These xenoliths are well-characterized and formed over a wide temperature range from ~800 to 1150°C. The coexisting Opx and Ol have constant and indistinguishable Mg isotopic compositions, with δ26Mg ranging from −0.29 to −0.22‰ in Ol and from −0.28 to −0.22‰ in Opx (Δ26MgOpx–Ol=δ26MgOpx−δ26MgOl=−0.04 to +0.04‰; n=11). By contrast, Mg isotopic compositions of Cpx and Phl are variable and slightly heavier than coexisting Ol (Δ26MgCpx–Ol=0 to +0.13‰, n=13; Δ26MgPhl-Ol=+0.11 to +0.20‰, n=3). Isotope fractionations between coexisting Cpx and Ol are correlated with temperatures, implying equilibrium isotope fractionation. The degree and direction of isotope fractionations among these mantle silicates agree with theoretical predictions, suggesting that inter-mineral Mg isotope fractionation is primarily controlled by the MgO bond strength, with stronger bonds favoring heavier Mg isotopes. Cpx and Phl have shorter and thus stronger MgO bonds, and hence are isotopically heavier than coexisting Ol. Compared with coexisting silicates, δ26Mg values of spinels are more variable and much heavier, ranging from +0.03 to +0.28‰. The Δ26MgSpl–Ol values vary significantly from +0.25 to +0.55‰ (n=10) and show an excellent, positive, linear correlation with 106/T2(K) [Δ26MgSpl–Ol=0.63(±0.12)×106/T2(K)−0.03(±0.08)], indicating equilibrium Spl–Ol isotope fractionation. The absence of intra-mineral isotopic variation and quantitative diffusion calculations further confirm isotope exchange equilibrium between coexisting Spl and Ol. Our results demonstrate the existence of measurable Mg isotope fractionation between mantle minerals and suggest that the large high-temperature equilibrium Spl–Ol Mg isotope fractionation in peridotite xenoliths can potentially be used as a geothermometer in mantle geochemistry.
Tracer diffusion coefficients for Li in glasses of albite, orthoclase, and obsidian composition have been determined by a method involving deposition of a thin source on polished glass wafers, anneal under controlled temperature conditions (300-900°C), and ion-microprobe determination of the concentration profile. All results conform to an Arrhenius-type relationship, D = D0 exp(-Q/RT), where Q is 23, 17, and 22 kcal mol-1 D0 is 0.2, 0.003, and 0.03 cm2s-1 for albite and orthoclase glasses, and obsidian respectively. Lithium is thus a fast diffusing ion and behaves similarly to sodium in the same glasses. A mechanism involving jumps of the diffusing ions through oxygen hexagonal rings is suggested by consideration of ionic radii ratio of alkali (H, Li, Na, K, Rb, and Cs) ions to the oxygen anions.
Lithium concentrations and isotopic compositions in the country rocks (amphibolites and schists) of the Tin Mountain pegmatite show systematic changes with distance to the contact. Both Li and δ7Li decrease dramatically along a ∼10 m traverse from the pegmatite into amphibolite, with Li concentration decreasing from 471 to 68 ppm and δ7Li decreasing from + 7.6 to − 19.9. Rubidium and Cs also decrease from the pegmatite contact into the country rock, but only within the first 2 m of the contact, after which their concentrations remain constant. Neither mixing between pegmatite fluids and amphibolite, nor Li isotope fractionation by Rayleigh distillation during fluid infiltration is a likely explanation of these observations, due to the extremely light isotopic composition required for the amphibolite end-member in the mixing model (δ7Li = − 20) and the similarly extreme isotopic fractionation required in a Rayleigh distillation model. Rather, these variations are likely due to isotopic fractionation accompanying Li diffusion from the Li-rich pegmatite (Li = 450 to 730 ppm) into amphibolites (Li = 20 ppm). The fact that other alkali element concentrations vary only within 2 m of the contact reflects the orders of magnitude faster diffusion of Li relative to heavier elements.
Episodic emplacement and cooling of lavas and dikes at mid-ocean ridges leads to large fluctuations in hydrothermal fluxes and biological activity. However, the processes operating beneath the seafloor during these transient events such as permeability creation and dike cooling are poorly understood. We have developed a new approach to determine the cooling rate of the sheeted dike complex based on the extent of diffusion of lithium from plagioclase into clinopyroxene during cooling. We have calibrated this Li-geospeedometer using new high-temperature experiments to determine both the temperature dependence of the partitioning of Li between plagioclase and clinopyroxene and the diffusion coefficient for Li in clinopyroxene. Application of this method to lavas and dikes from ODP Hole 504B shows that cooling rates vary dramatically with depth in the upper oceanic crust. Extremely rapid cooling rates (> 450 °C hr- 1) in the upper part of the sheeted dike complex are sufficient to power hydrothermal megaplume formation within the overlying water column.
Whole rock major and trace element abundances of the Horoman peridotites were used to understand processes forming lithological and compositional variations in the upper mantle. Similar to other orogenic peridotites, Horoman peridotites range from fertile lherzolites (3 to 4% Al2O3 and CaO) to depleted harzburgites (∼0.5% Al2O3 and CaO). Abundances of major oxides and compatible to moderately incompatible elements vary systematically with variations in MgO content. Such trends are commonly interpreted as indicating that the peridotites formed as residues from varying degrees of partial melting. The fertile end of these trends coincides with estimates of primitive mantle composition. Because of a mismatch between experimental melting trends for spinel peridotite, especially the Na2O-MgO trend, the compositional variations of Horoman peridotites are not consistent with formation as residues from partial melting of spinel peridotite. Non-linear trends in minor and trace element versus major element abundance diagrams also preclude a two-component mixing model. Recent melting experiments on garnet peridotite demonstrate that at 3 GPa the near-solidus peridotite has a large amount of subcalcic clinopyroxene (ca. 27%) coexisting with small amount of garnet (ca. 2%). Residues from polybaric melting of such garnet peridotite are consistent with the abundance variations of major and moderately incompatible elements, such as Na and heavy rare-earth elements, in the Horoman peridotites. A similar conclusion is applicable to other orogenic peridotites such as the Ronda peridotite because their major element compositional variations are similar to the Horoman peridotite.
Processes occurring in the Earth's upper mantle are important in controlling evolution of the crust-mantle system. The effects of multiple igneous and metamorphic processes are recorded in upper-mantle peridotites, such as the Horoman Peridotite in Hokkaido, Japan. Geochemical studies of these peridotites and their minerals, combined with determination of the spatial geochemical variations, can be used to understand the sequence of processes that affected the perioditite. In this study we show that compositional zoning patterns of major and trace elements in clinopyroxene porphyroclasts reflect a sub-solidus, closed-system transition from garnet periodotite, equilibrated at 20–24 kbar and 1040–1160°C, to plagioclase periodotite equilibrated at ∼ 7 kbar and 850–950°C. The preservation of compositionally zoned pyroxenes indicates that this transition was a relatively recent process that probably occurred as the Horoman Peridotite was emplaced into the Hidaka metamorphic belt. The clinopyroxene compositions also show that: (1) harzburgites and lherzolites reacted with a fluid/melt that resulted in relative enrichment of highly incompatible elements, such as the light rare-earth elements; and (2) this enrichment process preceded the sub-solidus breakdown of garnet.
Experiments were conducted at 1.5 GPa and temperatures between 700 °C and 1100 °C in order to assess the equilibrium distribution of lithium between olivine and diopside in the upper mantle. Lithium in olivine and diopside from natural mantle xenoliths displays a broad array of apparent partition coefficients ranging from ~0.2 to 10. In addition, a strikingly large range of lithium isotope ratios is observed in olivine and diopside from mantle xenoliths, with Δ7Liol–di (=δ7Liol−δ7Lidi) ranging from nearly zero to greater than 20‰. Both of these observations might be explained if the distribution of Li between olivine and diopside is strongly temperature dependent at mantle conditions such that a change in temperature, i.e. cooling upon exhuma- tion, initiates diffusive re-equilibration of Li between phases in the xenolith. Accompanying dynamic frac- tionation of 6Li from 7Li due to differing diffusion rates of the two isotopes could then be permanently recorded in the xenolith if its temperature drops below the closure temperature before a new equilibrium is reached.
ol/di
The results of this study indicate a partition coefficient for Li between olivine and diopside (DLi ) of 2.0 ± 0.2
that is independent of temperature (within the error of our analyses) over the range 700 °C to 1100 °C. This lack of temperature dependence holds true when data from previous experiments at temperatures as high as 1375 °C are considered. Thus it appears that closed-system diffusion of Li between olivine and diopside in response to changing temperature is not an appropriate explanation for the observed range of elemental and isotopic distributions in natural xenoliths. Other possible explanations include Li redistribution in re- sponse to changing oxygen fugacity in the system, or diffusive addition or subtraction of Li during open- system interaction with an infiltrating melt or fluid.
Ultraslow-spreading ridges (full spreading rate less than 20 mm per
year) comprise about one third (lengthwise) of the global ridge system
today, and might also have been important in the geologic past,
particularly after breakup of large continents. Geochemical studies of
abyssal peridotites show that large local- scale (single-dredge to hand
specimen scales) variations in isotopic composition, modal abundance of
clinopyroxene (CPX), and trace element abundance in CPX are distinct
characteristics of ultraslow-spreading ridge peridotites. It is
suggested that these local-scale variabilities are produced by melt-rock
reaction during melt migration in the mantle (e.g., Warren et al.,
2006). We explored using these geochemical characteristics to identify
ancient ultraslow-spreading ridge peridotites among "orogenic lherzolite
massifs". The Horoman peridotite massif in Hokkaido, Japan, possesses
DMM-like initial Nd and Sr isotopic characteristics at the time of melt
extraction, and displays large variations in modal CPX (13.2 - 5.2%,
only for lherzolites and plagioclase lherzolites) and in trace elements
in CPX (e.g., Nd ranges from 0.3 to 10 times C1 chondrite) over a
sampling scale length of 150 m. Additionally, there is ample evidence
for chromatographic melt-rock reaction in the presence of garnet
(Takazawa et al., 1996; Yoshikawa and Nakamura, 2000). These geochemical
characteristics are very similar to those observed for the present-day
ultraslow-spreading ridge peridotites such as those from the SWIR and
Gakkel ridge. A whole-rock Sm-Nd isochron age for the plagioclase
lherzolites of 830+/-78 Ma (Yoshikawa and Nakamura, 2000) and a
suggestion by Saal et al. (2001) that refertilization of depleted
peridotites by a MORB-like magma occurred around 900 Ma, based on the
Re-Os systematics, indicate that melt extraction and melt-rock reaction
formed the Horoman peridotite massif at 800 to 900 Ma. This time is
similar to that, 750 Ma, estimated for formation of the Paleopacific
ocean by rifting of the Rodinia supercontinent (e.g., Torsvik, 2003). We
suggeset that the Horoman massif could represent abyssal peridotite
beneath the Paleopacific ridge when its spreading rate was ultraslow.
Saal et al. (2001) J. Petrol., 42, 25-37; Takazawa et al. (1996) Chem.
Geol., 134, 3-26; Torsvik (2003) Science, 300, 1379-1381; Yoshikawa and
Nakamura (2000) JGR, 105, 2879-2901; Warren et al. (2006) Goldschmidt
Conf. Abstract.
Occurrence and ClassificationFragments of the Earth's mantle are
frequently transported to the surface via volcanic rocks that are
dominantly alkaline in nature. These fragments range up to sizes in
excess of 1 m across. The term "mantle xenoliths" or "mantle nodules" is
applied to all rock and mineral inclusions of presumed mantle derivation
that are found within host rocks of volcanic origin. The purpose of this
contribution is to review the geochemistry of mantle xenoliths. For
detailed petrological descriptions of individual locations and suites,
together with their geological setting, the reader is referred to the
major reference work by Nixon (1987).Despite peridotite xenoliths in
basalts being recognized for several centuries and comparisons being
made to lherzolite massifs (Lacroix, 1893), it was not until work on
garnet peridotites and diamonds in kimberlites by Fermor (1913) and
Wagner (1914) that such xenoliths were conceptually associated with a
peridotite zone in the Earth beneath the crust, i.e., the zone that we
now identify as the mantle. Mantle xenoliths provide snapshots of the
lithospheric mantle beneath particular regions at the time of their
eruption and hence are crucial direct evidence of the nature of the
mantle beneath regions where no samples have been exposed by tectonic
activity. As such, xenoliths are an essential compliment to tectonically
exposed bodies of mantle (orogenic peridotites and ophiolites) that
occur at plate boundaries (see Chapter 2.04). One obvious contrast
between the mantle samples provided by xenoliths and those provided by
peridotite massifs is the lack of field relationships available for
xenoliths. Other drawbacks include the small size of many xenoliths.
This makes accurate estimation of bulk compositions difficult and
accentuates modal heterogeneities. The frequent infiltration of the host
magma also complicates their chemical signature. Despite these
drawbacks, xenoliths are of immense value, being the only samples of
mantle available beneath many areas. Because they are erupted rapidly,
they freeze in the mineralogical and chemical signatures of their depth
of origin, in contrast to massifs which tend to re-equilibrate
extensively during emplacement into the crust. In addition, many
xenolith suites, particularly those erupted by kimberlites, provide
samples from a considerably greater depth range than massifs. Over 3,500
mantle xenolith localities are currently known. The location and nature
of many of these occurrences are summarized by Nixon (1987). A
historical perspective on their study is given by Nixon (1987) and
Menzies (1990a). Mantle xenoliths from any tectonic setting are most
commonly described from three main igneous/pyroclastic magma types
(where no genetic relationships are implied):(i) Alkalic basalts
sensu-lato (commonly comprising alkali basalt-basanites and more evolved
derivatives), nephelinites and melilitites.(ii) Lamprophyres and related
magmas (e.g., minettes, monchiquites, and alnoites) and lamproites.(iii)
The kimberlite series (Group I and Group II or orangeites; Mitchell,
1995).Although mantle xenoliths most commonly occur in primitive members
of the above alkaline rocks, rare occurrences have been noted in more
evolved magmas such as phonolites and trachytes (e.g., Irving and Price,
1981).To simplify matters and to circumvent the petrographic
complexities of alkaline volcanic rocks in general, we will use the term
"alkalic and potassic mafic magmas" to include alkalic basalts,
nephelinites, melilitites, and lamprophyres. Occurrence of xenoliths in
such magmas can be compared to those occurring in kimberlites and
related rocks. As a general rule, the spectrum of mantle xenoliths at a
given location varies with host rock type. In particular, alkalic and
potassic mafic magmas tend to erupt peridotites belonging predominantly
to the spinel-facies, whereas kimberlites erupt both spinel and
garnet-facies peridotites (Nixon, 1987; Harte and Hawkesworth,
1989).Even within either "group" of volcanic rocks the variety of
possible xenolith types is great. Table 1 presents a summary of the most
common mantle xenolith groups that are found in kimberlitic hosts and
within the alkalic and potassic mafic magmas. The significance and
abundance of these groups will be discussed below. Table 1. Major groups
of mantle xenoliths in kimberlite-related and alkali basalt series
volcanic rocks (after Harte and Hawkesworth, 1989). Textural
classification follows that of Harte (1977). Terminology for
phlogopite-rich mafic mantle xenoliths from Gregoire et al. (2002). For
supplementary data and classifications see Nixon (1987), table 62
TypeCharacteristicsExamplesMg# olivine (A) Cratonic/circum-cratonic
xenoliths erupted by Kimberlite-related volcanics AI: Coarse Mg-rich,
low-T peridotitesOften abundant. Mostly harzburgites and lherzolites
with varying but low modal diopside and garnet. Wide range of
orthopyroxene abundance, Kaapvaal examples notably enriched. Crystals
typically 0.2 mm with equant or tabular shapes, irregular grain
boundaries, rarely granoblastic (Harte, 1977). Bulk compositions
typically highly depleted in Fe, Ca, and Al, enriched in Mg. Mineralogy:
Cr-rich pyrope, Cr-diopside. Orthopyroxene in garnet facies
characterized by >1.0 wt.% Al2O3. Cr-spinel sometimes evident. Minor
phlogopite common grading into type VIII phlogopite peridotites.
Phlogopite often surrounds garnet and is strongly correlated with the
presence of diopside. Estimated equilibration temperatures less than
1,100 °C. Equilibration pressures can vary widely within a pipe and
range from c. 2 GPa to >6 GPa. Rarely diamondiferous (e.g., Dawson
and Smith, 1975), more commonly contain graphite ( Pearson et al.,
1994).N. Lesotho (Nixon and Boyd, 1973a), Kaapvaal craton ( Gurney and
Harte, 1980; Boyd and Nixon, 1978; Boyd and Mertzman, 1987; Nixon,
1987), Siberia ( Sobolev, 1974; Boyd et al., 1993); Jericho Slave craton
( Kopylova et al., 1999)Av 92.8 (91-95) Subcalcic garnet (high
Cr-pyrope; knorringitic) bearing harzburgite varieties scarce but can
contain diamond and graphite. Can be megacrystalline. Textures similar
to type I. Equilibration temperatures and pressures intermediate between
low-T and high-T lherzolites, i.e., 1,150 °C, 5-6 GPa, but vary
widely.Udachnaya, Siberia (Sobolev et al., 1973; Pokhilenko et al.,
1993), Kaapvaal ( Boyd et al., 1993)92-95.5 Spinel facies widespread but
less abundant. Textures as for garnet variety, spinel texture
symplectitic or irregular. Equilibration temperatures <800 °C.
Can also be orthopyroxene enriched, like garnet facies. Spinel
composition can vary widely in Cr# but mostly aluminous. Cr-rich spinels
coexist with garnet. Orthopyroxenes in spinel facies have >1.0 wt.%
Al2O3. Similar range in bulk composition to garnet facies.Kaapvaal
craton (Carswell et al., 1984; Boyd et al., 1999)91.5-94 AII: coarse,
Fe-rich low-T peridotites and pyroxenitesWidespread, normally rare but
locally abundant. Mainly garnet lherzolites and garnet websterites but
also clinopyroxenites and orthopyroxenites ("bronzitites"). Ilmenite can
be present in pyroxenites. Coarse grained to "megacrystalline" (at
Jericho). Textures and equilibration temperatures as for type I.
Sometimes modally layered. Wide ranging bulk and mineral compositions,
with high Fe, Ca, Al, and Na relative to type I. Rare fine-grained
"quench textured" ilmenite/garnet pyroxenites.Matsoku, Kaapval craton
(Gurney et al., 1975); Jericho, Slave craton ( Kopylova et al., 1999);
Mzongwana, SE margin Kaapvaal craton ( Boyd et al., 1984a)83-89 AIII:
dunitesWidespread, locally common. Two varieties: (i) Highly depleted,
coarse to ultracoarse >50 mm olivine (megacrystalline) dunites, often
containing chromite or sub calcic high-Cr pyrope and frequently
diamondiferous. (ii) Often fine to medium grained more Fe-rich dunites,
mineral zoning indicates "metasomatism." Mostly deformed textures.
Orthopyroxene, garnet, phlogopite, diopside, chromite present.Siberia,
notably Udachnaya (Pokhilenko et al., 1993)Kimberley ( Boyd et al.,
1983; Dawson et al., 1981)93-9585-93 AIV: deformed low-T peridotites and
pyroxenitesWidespread, locally common. Porphyroclastic or
mosaic-porphyroclastic textures. Modal abundances, chemical
characteristics and P-T equilibration conditions similar to those of
type I.Jericho, Slave craton (Kopylova et al., 1999)91-95 AV: deformed
high-T peridotitesWidespread but variable abundance in group I
kimberlites, absent/scarce in group II kimberlites. Commonly deformed;
porphyroclastic and mosaic-porphyroclastic textures with fine neoblasts
of olivine. Although generally more depleted than pyrolite, bulk rocks
and minerals generally enriched in Fe and Ti compared to type I (low-T)
and significant compositional overlap of minerals with megacrysts (type
X). Equilibration temperatures 1,100 °C to >1,500 °C,
equilibration pressures generally 4.5 GPa to >6.5 GPa. Garnets and
pyroxenes frequently zoned.N. Lesotho (Nixon and Boyd, 1973a);
Jagersfontein, Kaapvaal craton ( Burgess and Harte, 1999); Siberia (
Sobolev, 1974; Boyd et al., 1993); Slave ( Kopylova et al., 1999);
Somerset Island, Churchill Province ( Schmidberger and Francis,
1999)87-92 AVI: phlogopite-rich mafic mantle xenolithsWidespread and
locally common. Olivine poor/absent rocks. Two main subdivisions of this
group (Gregoire et al., 2002) are: (i) MARID suite
(mica-amphibole-rutile-ilmenite-diopside) with accessory zircon common.
Probable genetic link to group II kimberlites. Medium to coarse grained,
undeformed to deformed, sometimes modal banding. Amphibole always
K-richterite. (ii) PIC suite (phlogopite-ilmenite-clinopyroxene) with
minor rutile. Diopside or Al- and Ti-poor augites. Probable genetic link
to group I kimberlites. K-richterite is absent; grade to glimmerites as
phlogopite mica reaches >90%. Coarse grained, variably
deformed.Kimberley (Dawson and Smith, 1977; Gregoire et al.,
2002)Kimberley ( Gregoire et al., 2002)NANA AVII: pyroxenite sheets rich
in Fe and TiRestricted to Matsoku. Orthopyroxene and clinopyroxene rich
rocks with widely variable olivine and garnet compositions, often with
ilmenite and phlogopite (the IRPS suite: see type VIII). Bulk
compositions Fe and Ti rich. Form magmatic intrusions (<16 cm thick)
into type I rocks which become metasomatized.Matsoku (Gurney et al.,
1975; Harte et al., 1975, 1987) AVIII: modally metasomatized
peridotitesWide spread, variable abundance. Mostly metasomatized
variants of type I. Diverse mineralogies, Two most commonly recognized
groups are phlogopite peridotites (PP) and
phlogopite-K-richterite-Peridotites (PKP) of Erlank et al. (1987). Can
be harzburgite or lherzolite, typically coarse grained, undeformed but
some display porphyroclastic textures. Assemblages vary with location.
Cr-titanate "LIMA" minerals (Lindsleyite-Mathiasite) relatively common
at Bultfontein; edenite-phlogopite association at Jagersfontein;
ilmenite-rutile-phlogopite-sulfide (IRPS) suite at Matsoku associated
with pyroxenitic sheets (type VII). Metasomatic clinopyroxene link to
type AI.Matsoku (Gurney et al., 1975); Kimberley pipes ( Erlank et al.,
1987; Gregoire et al., 2002); Jagersfontein ( Winterburn et al.,
1990)Same as type I, to more Fe-rich. AIX: eclogites, grospydites,
alkremites, and variantsVery widespread, rare to locally abundant.
Eclogites (omphacite and pyrope-almandine garnet). Garnet composition
widely variable, in grospydites garnet has a large grossular component.
At some locations (e.g., Jagersfontein), unusual assemblages of
garnet+spinel (Alkremites), garnet+corundum (Corganites) and
corundum+garnet+spinel (Corgaspinites) occur (Mazzone and Haggerty
1989). Accessory phases in eclogites include kyanite, corundum,
ilmenite, rutile, sanidine, coesite, sulfides, graphite and diamond.
Eclogites classified on texture: group I large subhedral to rounded
garnets in matrix of omphacite. High Cr, Ca, Fe, and Mn in omphacite.
Garnets more Na (avg. 0.1 wt.% Na2O) and Mg rich. Group II have
interlocking texture of anhedral garnet and omphacite and are less
altered. Garnets are lower in Na2O (0.05 wt.%). Common hosts for
diamond, especially group I. Not all eclogites of obvious mantle origin
and some grade into garnet granulites and pyroxenites of crust
origin.Roberts Victor (MacGregor and Carter, 1970; McCandless and
Gurney, 1989); Jagersfontein ( Nixon et al., 1978; Mazonne and Haggerty,
1989); Orapa ( Robinson et al., 1984), all Kaapvaal craton. Udachnaya,
Siberian craton ( Sobolev, 1974; Ponomarenko, 1975); Koidu, W. African
craton ( Tompkins and Haggerty, 1984; Hills and Haggerty, 1989)NA AX:
megacrysts (discrete nodules)Single crystals or monominerallic
polycrystalline aggregates (sometimes exsolved) weighing up to 15 kg.
Rare mutual lamellar or granular intergrowths. Large range in Mg#, Cr,
and Ti in a given suite. Cr-poor variety: widespread, locally abundant
(e.g., Monastery). Garnets, clino- and orthopyroxenes, phlogopite and
ilmenite most common, zircon and olivine rarer. Debatable whether
phlogopite and olivine are members of Cr-poor suite. Wide range in
chemistry but Cr-poor, Fe-Ti-rich relative to type I (low-T) peridotite
minerals. Mineral chemistry and estimated equilibration P/Ts overlap
those of type V (high-T) lherzolites. Some Slave craton "Cr-poor
megacrysts" show mineral chemistry links to type II megacrystalline
pyroxenite xenoliths. See review of Schulze (1987).N. Lesotho (Nixon and
Boyd 1973b); Monastery ( Gurney et al., 1979), Jagersfontein ( Hops et
al., 1992), Kaapvaal craton; The Malaita megacryst suite ( Nixon and
Boyd, 1979), occurs in an ocean plateau alnoite, but has many
similarities with the kimberlitic low-Cr suite Cr-rich variety: (i) A
suite comprising garnet plus ortho- and clinopyroxene, mostly restricted
to kimberlites from Colorado-Wyoming. Mineralogically similar to type I
lherzolites. (ii) "Granny Smith" diopsides; bright green Cr-diopside,
may contain blebs/intergrowths of ilmenite and phlogopite. Can be
polycrystalline.Colorado-Wyoming craton (Eggler et al., 1979)Kimberley
and Jagersfontein ( Boyd et al., 1984b) Miscellaneous: mostly garnets
and pyroxenes with no clear paragenetic association or links to other
megacryst suites. May represent disrupted
peridotites/eclogites/pyroxenites. AXI: polymict aggregatesPolymict
aggregates of peridotite, eclogite and megacrysts, of variable grain
size, some containing quenched melt. Mineral assemblages not in
elemental or isotopic equilibrium.Bultfontein, De Beers and Premier
mines, Kaapvaal (Lawless et al., 1979). MalaitaHighly variable AXII:
diamond and inclusions in diamondsWidespread and closely related to
cratons. Abundance varies from <1 ppm to 100 ppm by weight. Size
<<0.1 g to c. 750 g. Type I diamonds contain abundant N, type II
low N (Harris, 1987).All cratons (Harris, 1987; Meyer, 1987)93-96
Inclusion suites divided into peridotitic (P-type) and eclogitic
(E-type) parageneses. P-type inclusions: high-Cr, low Ca garnet,
Cr-diopside, Fo-rich olivine, orthopyroxene, chromite, wustite, Ni-rich
sulfide, have restricted, high Mg, high Ni chemistry. Equilibration
temperatures 900-1,100 °C. E-Type inclusions: pyrope-almandine, high
Na garnet (>0.1 wt.%), omphacite, coesite, low-Ni sulfide. AXIII:
ultra-deep peridotitesRare and restricted to Jagersfontein (Kaapvaal
Craton) and Koidu (W. African craton). Four-phase garnet lherzolite.
Close association of pyrope-garnet (˜70% py; 2 wt.% Cr2O3) and
jadeite-rich clinopyroxene (20% Jd, & 4% Di). Clinopyroxene forms
either orientated rods in garnet host or as discrete grains attached to
garnet in cuspate contact. Both pyroxenes exsolved from garnet at
100-150 km depth. Recombination of garnet gives original depths of
derivation of 300-400 km. Discrete garnets and "lherzolites" with
eclogitic affinities also found (Sautter et al., 1991).All samples so
far from the Jagersfontein kimberlite, S. Africa (Haggerty and Sautter,
1990; Sautter et al., 1991) and Koidu, Sierra Leone ( Deines and
Haggerty, 2000)91.6 (B): Non-cratonic xenoliths erupted by alkalic and
potassic mafic magmas sensu latoa BI: Cr-diopside lherzolite groupVery
widespread and common in a variety of tectonic settings, off-craton.
Dominantly spinel-facies (Al or Cr-spinel) lherzolites but can be
garnet-facies and garnet-spinel facies (e.g., Vitim). Coarse grained,
commonly little deformed, sometimes show preferred orientation. Include
harzburgites, orthopyroxenites, clinopyroxenites, websterites, and
wehrlites. Pargasite and phlogopite may also be common. Both low TiO2
and high TiO2 amphiboles can occur at the same locality. Accessory
apatite, can be common locally (e.g., Bullenmerri, Victoria).
Interstitial silicate glass can be present. Garnet and spinel facies
significantly more olivine-rich and orthopyroxene poor than peridotites
from cratons such as Kaapvaal and Siberia. Bulk rocks less depleted in
Ca, Al, Fe, and lower in Mg than cratonic peridotites. Minerals
generally higher Mg# and Cr# and lower Na and Ti than those of the
Al-Augite group. Can be subdivided into type IA (LREE depleted
clinopyroxene) and type IB (LREE enriched clinopyroxene).Victoria, SE
Australia (Frey and Green, 1974); Vitim ( Ionov et al., 1993a); San
Carlos and other W. USA localities ( Frey and Prinz, 1978; Wilshire and
Shervais, 1975); Eifel ( Stosch and Seck, 1980); Hawaii ( Jackson and
Wright, 1970); Scotland ( Menzies and Halliday, 1988)Garnet facies:
Thumb, Navajo field ( Ehrenberg, 1982a, b); Pali-Aike, Patagonia ( Stern
et al., 1989); Vitim, S. Siberia ( Ionov, 1993a, b)>0.85, Avg.
˜90 BII: Al-augite wehrlite-pyroxenite groupWidespread and common.
Frequently clinopyroxene-rich rocks but widely variable: wehrlites,
clinopyroxenites, dunites, websterites, lherzolites, lherzites, gabbros.
Al-spinel is the typical aluminous phase but may contain plagioclase.
Kaersutite common along with apatite, Fe-Ti oxides, and phlogopite. Some
igneous and metamorphic textures. Composite xenoliths relatively common
(in contrast to kimberlite-related xenoliths). Cross-cutting
pyroxene-rich veins and layers may occur in olivine-rich hosts.
Olivine-rich aggregates also found in pyroxene-rich xenoliths. Minerals
generally lower Mg# and Cr#, higher Ti than those of the type I
(Cr-diopside group).Victoria, SE Australia (Frey and Green, 1974); San
Carlos and other W. USA localities ( Frey and Prinz, 1978; Wilshire and
Shervais, 1975; Irving, 1980); Hawaii ( Jackson and Wright, 1970;
Irving, 1980)<0.85 BIII: garnet pyroxenite groupWidespread but not
abundant. Garnet clinopyroxenites and websterites plus clinopyroxenites
and websterites where pyroxenes commonly show exsolution of garnet
and/or spinel as well as Ca-rich or Ca-poor pyroxene. Accessory ilmenite
and sometimes apatite. Coarse grained, undeformed textures, sometimes
layered. "Basaltic" bulk compositions.Delegate, SE Australia (Lovering
and White, 1969; Irving, 1974), Salt Lake Crater, Hawaii ( Beeson and
Jackson, 1970)NA BIV: modal metasomatic groupWidespread varieties of the
above groups showing evidence for modal (or "patent") metasomatism.
Wehrlite-clinopyroxenites with mica, glimmerites. Typical metasomatic
phases include pargasite/kaersutite, phlogopite, apatite, and
grain-boundary oxides e.g., rutile. Apatite only in some cases. Silicate
glass as melting product of amphibole, clinopyroxene, or phlogopite
common. Composite xenoliths occur.Nunivak, Alaska (Francis, 1976), SE
Australia ( O'Reilly et al., 1991), Menzies and Murthy, 1980a, Vitim (
Ionov et al., 1993a), Loch Roag and Fife Scotland ( Menzies et al.,
1989)NA BV: megacrystsWidespread with variable abundance. Usually large
(>1 cm) single crystals. Large range in Mg#, Cr, and Ti in a given
suite.SE Australia (Binns et al., 1970; Irving and Frey, 1984; Schulze,
1987), Loch Roag, Scotland ( Menzies et al., 1989)NA Group A: Al-augite,
Al-bronzite, olivine, kaersutite, pyrope, pleonaste, plagioclase; some
of which may have crystallized from the host magma Group B:
Anorthoclase, Ti-mica, Fe-Na salite, apatite, magnetite, ilmenite,
zircon, rutile, sphene, and corundum, all of which are likely
xenocrysts. Some coarse crystals are undoubtedly derived from
disaggregated type I and type II xenoliths. a Based on Harte and
Hawkesworth (1989) with nomenclature from Frey and Green (1974),
Wilshire and Shervais (1975), Frey and Prinz (1978), Irving (1980), and
Menzies (1983). Although widespread in the literature, classification of
xenoliths on the basis of their host magma is not fully informative. It
is more geologically useful to subdivide xenoliths in terms of their
tectonic setting. A basic subdivision of xenolith occurrences is between
those erupted in oceanic settings and those erupted in continental
settings. The continental occurrences far outnumber the oceanic
occurrences. The continental occurrences can be further subdivided
depending on age of the crust and the tectonic history of the area being
sampled. Xenoliths from stable cratonic and circum-cratonic regions are
distinctly different in petrology from those occurring in areas that
have experienced significant lithospheric rifting, generally in
noncratonic crust, in the recent geological past. As such, we will
utilize the terms cratonic/circum-cratonic to refer to xenoliths
occurring on and around craton margins and the term noncratonic in
referring to mantle sampled away from cratons, often in areas that have
experienced recent lithospheric thinning. There is a link back to the
host rock in that, as a general rule, cratonic and circum-cratonic
xenoliths are erupted by kimberlites and noncratonic xenoliths are
erupted by alkalic and potassic mafic magmas.2.05.1.1.1. Mantle
xenoliths in continental volcanic rocksXenoliths found in Archean
cratonic regions are characterized by the lithological types reported in
Table 1(a). Garnet-facies peridotites dominate the peridotite xenolith
inventory in these locations. In contrast, away from cratons, there is a
scarcity of garnet-facies peridotites ( Table 1(b)). In addition,
cratonic xenolith suites contain samples derived from depths ranging
from crustal levels to >200 km, whereas noncratonic xensoliths come
from less than 140 km deep. There can be distinct differences between
xenoliths erupted on craton and those erupted in stable areas of
Proterozoic crust marginal to cratons. For instance, subcalcic-garnet
harzburgites occur on most cratons but do not occur in circum-cratonic
suites ( Boyd et al., 1993). In addition, the maximum depths of
equilibration of circum-cratonic peridotite suites are less than for
cratonic peridotite suites (e.g., Finnerty and Boyd, 1987). These
differences warrant the distinction between "cratonic" and
"circum-cratonic" xenoliths. In addition, young rift-related magmatism,
marginal to cratons, samples very thin lithosphere compared to cratonic
and circum-cratonic lithosphere. The xenoliths sampled in this
environment fall into the loose category of "noncratonic" xenoliths. A
more detailed and complex tectonic classification is provided by Griffin
et al. (1999a).2.05.1.1.2. Mantle xenoliths in oceanic volcanic rocksThe
nature of the suboceanic mantle is largely constrained from geochemical
studies of its partial melts (see Chapter 2.08) because occurrences of
mantle xenoliths in the ocean basins are much rarer than on the
continents. The host rocks for these xenoliths are exclusively alkalic
and potassic mafic magmas. The xenolith suite of the Hawaiian volcanic
chain is perhaps the best characterized (Jackson and Wright, 1970) of
the ocean islands, while extensive suites have also been found in the
Canary Islands ( Neumann et al., 1995), Samoa ( Hauri et al., 1993),
Grande Comore ( Coltorti et al., 1999), and Tahiti. Most of these
occurrences sample the oceanic lithosphere directly below the islands
and those of type I ( Table 1) are proposed to be residues of partial
melting that have been variably metasomatized, with carbonatite-like
fluids frequently being invoked ( Hauri et al., 1993; Coltorti et al.,
1999). The Hawaiian suite is more complex. Pyroxenites of type II and
type III are common and iron-rich peridotites, some with garnet, are
thought to be physical mixtures of spinel lherzolites with the
pyroxenite suite ( Sen and Leeman, 1991).Some xenolith localities sample
the mantle lithosphere beneath oceanic plateaux. The most extensive and
varied xenolith suite in this regard is that from Malaita (Solomon
Islands) on the margin of the Ontong Java Plateau (Nixon and Boyd,
1979). This locality is hosted by an alnoite and contains both garnet
and spinel-facies lherzolites together with a spectacular megacryst
suite. Although in an oceanic setting, the variety of the xenolith suite
provided by the Malaita alnoite, in particular the megacrysts, show
strong similarities to suites from kimberlites ( Nixon and Boyd,
1979).2.05.1.1.3. Mantle xenoliths in subduction zone
environmentsAlthough xenoliths from subduction-related tectonic settings
have been known for sometime, their detailed relationship to the
subduction zone system has been a matter of debate. Most samples are
type-I spinel lherzolites and modal metasomatic variants of type IV,
most commonly kaersutite and phlogopite. Among the best-known examples
are from Itinome-Gata, Japan (Aoki, 1968) and Simcoe, NW, USA ( Brandon
et al., 1999), although spinel lherzolites from Grenada, Lesser
Antilles, also occur. It is not well established whether these xenoliths
actually represent parts of the metasomatized mantle wedge above the
subduction zone, or simply mantle lithosphere not intimately related to
the subduction zone process. McInnes and Cameron (1994) have reported
xenoliths from the Tabar-Lihir-Tanga-Feni arc, Papua New Guinea, that
are purported to be mantle wedge compositions.
### Geological relevance of grain boundary diffusion
As a rule of thumb, diffusion along grain boundaries is several orders of magnitude faster than within minerals and hence it could enable mass transport on much larger spatial scales within geological time scales. Therefore Earth scientists are interested in the transport properties of grain and interphase boundaries (and other short circuit diffusion paths such as dislocations or a fluid network) since they may provide very efficient transport paths controlling a variety of processes operating on very different spatial scales. For example, on a micron scale, corona formation around minerals is a common phenomenon that is typically controlled by grain boundary diffusion. Similarly, element and isotopic exchange between a physically separated mineral pair (used, for example, as a geothermometer) requires transport along grain boundaries. Metasomatic reaction zones on the centimeter to meter scale are not always simply related to fluid flow but are also induced by fluid assisted diffusive fluxes between two different rock types imposed on each other due to tectonic processes (e.g., Miller et al. 2009). It has been also argued that the observed strong gradual fractionation of Li isotopes on the meter scale was a result of diffusion, where a strong contribution comes from grain boundary diffusion within these rocks (Teng et al. 2006). Further, new data on grain boundary diffusion of siderophile elements and carbon in polycrystalline MgO suggest that grain boundary diffusion provides an efficient mechanism to exchange these elements between the mantle and the core even on the kilometer scale within reasonable time scales (Hayden and Watson 2007, 2008). In addition to the significant role of grain boundary diffusion for element transfer in rocks, grain boundary diffusion can also dominate the bulk physical properties of rocks, e.g., their viscosity (Coble creep) or electrical conductivity, which is for example relevant for the interpretation of …
THE segregation of melts from the Earth's upper mantle into the crust is an important process in the chemical evolution of the crust–mantle system. The processes of melt formation and migration in the upper mantle are inadequately understood, but some important characteristics of these processes can be inferred from upper-mantle rocks exposed at the Earth's surface. The Horoman peridotite body in northern Japan is a layered upper-mantle rock. The major-element compositions of the layers are consistent with their formation as residues from varying extents of melting; however, abundances of rare-earth elements (REE) require additional processes to have occurred1, such as post-melting enrichment (metasomatism) resulting from reaction with a migrating fluid phase. We report here that chondrite-normalized REE patterns in clinopyroxenes show abrupt changes in slope, which vary with stratigraphic position and rock type. These data can be modelled by chromatographic fractionation as melts migrated through and interacted with peridotite, creating compositional heterogeneities in the upper mantle. In the Horoman peridotite these heterogeneities occur on a scale length of tens of metres.
MELT segregation from a solid matrix in the upper mantle is the first step in the formation of magmas. The mechanisms of melt segregation proposed so far have been based on two different physical models: the percolation of interstitial melt driven by a density difference between melt and matrix, associated with compaction and deformation of the matrix1; or the suction of interstitial melt into fractures from a surrounding porous matrix2–4. Although the latter process was originally invoked to explain the occurrence of dunite and pyroxenite–gabbro dykes with zones depleted in melt component on both sides2,5–7, most of these zones are discordant, small in scale and clearly formed by reaction between peridotite and basaltic melt8. I have suggested9 that thick, concordant dunite zones in the most depleted peridotite layers of the Horoman peridotite complex formed by suction of partial melt towards fractures later filled by dunite. Here I report a detailed mineralogical variation across the dunite which, when set in its geological and petrographic context, provides clearer evidence for melt segregation by dynamic forcing3, rather than passive percolation. This mechanism of melt segregation may play an important role in the formation of primary magmas with low production rate and large geochemical variability.
The Li isotope ratios of four international rock reference materials, USGS BHVO-2, GSJ JB-2, JG-2, JA-1 and modern seawater (Mediterranean, Pacific and North Atlantic) were determined using multi-collector inductively coupled plasma-mass spectrometry (MC-ICP-MS). These reference materials of natural samples were chosen to span a considerable range in Li isotope ratios and cover several different matrices in order to provide a useful benchmark for future studies. Our new analytical technique achieves significantly higher precision and reproducibility (< ± O.3%o 2s) than previous methods, with the additional advantage of requiring very low sample masses of ca. 2 ng of Li.
Les rapports isotopiques du Li de 4 matériaux de référence, de provenance Internationale, BHVO-2, JB-2, JG-2, JA-1 et d'eau de mer (Méditerranée, Pacifique et Atlantique Nord) ont été déterminés par MC-ICP-MS (spectrométrie de masse avec source à plasma induit à multicollection). Ces matériaux de référence naturels ont été choisis car ils balaient un large champ des rapports isotopiques du Lithium et couvrent différentes matrices afin de fournir un point de repère utile pour les études futures. Notre nouvelle technique analytique permet d'atteindre une précision et une reproductibilité (< ± 0.3%. 2s) nettement supérieures à celles des méthodes précédemment utilisées et présente I'avantage de pouvoir travailler avec des échantillons de petite masse, ∼ 2 ng de Li.
A precise and simple method for the determination of lithium concentrations in small amounts of silicate sample was developed by applying isotope dilution-inductively coupled plasma-mass spectrometry (ID-ICP-MS). Samples plus a Li spike were digested with HF-HClO4, dried and diluted with HNO3, and measured by ICP-MS. No matrix effects were observed for Li-7/Li-6 in rock solutions with a dilution factor (DF) of a >= 97 at an ICP power of 1.7 W. By this method, the determination of 0.5 mu g g(-1) Li in a silicate sample of 1 mg can be made with a blank correction of < 1%. Lithium contents of ultrabasic to acidic silicate reference materials (JP-1, JB-2, JB-3, JA-1, JA-2, JA-3, JR-1 and JR-2 from the Geological Survey of Japan, and PCC-1 from the US Geological Survey) and chondrites (three different Allende and one Murchison sample) of 8 to 81 mg were determined. The relative standard deviation (RSD) was typically < 1.7%. Lithium contents of these samples were further determined by isotope dilution-thermal ionisation mass spectrometry (ID-TIMS). The relative differences between ID-ICP-MS and ID-TIMS were typically < 2%, indicating the high accuracy of ID-ICP-MS developed in this study.
In order to provide a more substantial foundation for interpreting the behaviour of B, Be, and Li during the production and early crystallization of primitive igneous rocks, we have measured olivine-, clinopyroxene-, orthopyroxene-, and amphibole-melt partition coefficients for these elements involving broadly basaltic-andesitic melt compositions. Experiments were conducted at both one atmosphere and 1.0–1.5 GPa and employed a time-temperature history that yielded large crystals with minimal compositional zoning. Experiment temperatures ranged from 1000 to 1350°C and were selected to minimize the total crystal fraction in a given experiment. Partition coefficients for olivine and clinopyroxene were found to be independent of run duration or total concentration of B, Be, or Li suggesting that crystal-liquid equilibrium was closely approached.
Two types of laboratory experiments were used to quantify magnesium isotopic fractionations associated with chemical and thermal (Soret) diffusion in silicate liquids. Chemical diffusion couples juxtaposing a molten natural basalt (SUNY MORB) and a molten natural rhyolite (Lake County Obsidian) were run in a piston cylinder apparatus and used to determine the isotopic fractionation of magnesium as it diffused from molten basalt to molten rhyolite. The thermal diffusion experiments were also run in a piston cylinder apparatus but with a sample made entirely of molten SUNY MORB displaced from the hotspot of the assembly furnace so that the sample would have a temperature difference of about 100–200 °C from one end to the other. The chemical diffusion experiments showed fractionations of 26Mg/24Mg by as much as 7‰, which resulted in an estimate for the mass dependence of the self-diffusion coefficients of the magnesium isotopes corresponding to D26Mg/D24Mg=(24/26)β with β = 0.05. The thermal diffusion experiments showed that a temperature difference of about 100 °C resulted in the MgO, CaO, and FeO components of the basalt becoming slightly enriched by about 1 wt% in the colder end while SiO2 was enriched by several wt% in the hotter end. The temperature gradient also fractionated the magnesium isotopes. A temperature difference of about 150 °C produced an 8‰ enrichment of 26Mg/24Mg at the colder end relative to the hotter end. The magnesium isotopic fractionation as a function of temperature in molten basalt corresponds to 3.6 × 10−2‰/°C/amu.
We present high-precision measurements of Mg and Fe isotopic compositions of olivine, orthopyroxene (opx), and clinopyroxene (cpx) for 18 lherzolite xenoliths from east central China and provide the first combined Fe and Mg isotopic study of the upper mantle. δ56Fe in olivines varies from 0.18‰ to −0.22‰ with an average of −0.01 ± 0.18‰ (2SD, n = 18), opx from 0.24‰ to −0.22‰ with an average of 0.04 ± 0.20‰, and cpx from 0.24‰ to −0.16‰ with an average of 0.10 ± 0.19‰. δ26Mg of olivines varies from −0.25‰ to −0.42‰ with an average of −0.34 ± 0.10‰ (2SD, n = 18), opx from −0.19‰ to −0.34‰ with an average of −0.25 ± 0.10‰, and cpx from −0.09‰ to −0.43‰ with an average of −0.24 ± 0.18‰. Although current precision (∼±0.06‰ for δ56Fe; ±0.10‰ for δ26Mg, 2SD) limits the ability to analytically distinguish inter-mineral isotopic fractionations, systematic behavior of inter-mineral fractionation for both Fe and Mg is statistically observed: Δ56Feol–cpx = −0.10 ± 0.12‰ (2SD, n = 18); Δ56Feol–opx = −0.05 ± 0.11‰; Δ26Mgol–opx = −0.09 ± 0.12‰; Δ26Mgol–cpx = −0.10 ± 0.15‰. Fe and Mg isotopic composition of bulk rocks were calculated based on the modes of olivine, opx, and cpx. The average δ56Fe of peridotites in this study is 0.01 ± 0.17‰ (2SD, n = 18), similar to the values of chondrites but slightly lower than mid-ocean ridge basalts (MORB) and oceanic island basalts (OIB). The average δ26Mg is −0.30 ± 0.09‰, indistinguishable from chondrites, MORB, and OIB. Our data support the conclusion that the bulk silicate Earth (BSE) has chondritic δ56Fe and δ26Mg.
We report high-precision MC-ICPMS measurements of 25Mg/24Mg and 26Mg/24Mg isotope ratios for 14 meteorites, both primitive and differentiated, 5 lunar basalts and breccia samples, 5 terrestrial basalts, 1 dunite and 3 peridotites. We estimate the δ26Mg value of the bulk silicate Earth (BSE) at − 0.54 ± 0.04 (2SE) w.r.t. DSM3. Carbonaceous and ordinary chondrites show an average δ26Mg of − 0.52 ± 0.04 (2SE), martian meteorites show a narrow range in δ26Mg with an average δ26Mg of − 0.57 ± 0.02 (2SE), average δ26Mg of pallasite olivines is − 0.54 ± 0.04 (2SE) whereas that of lunar samples, including ilmenite basalts, olivine basalt and one crystalline breccia is − 0.51 ± 0.03 (2SE). Our data demonstrate that the overall stable Mg isotopic composition of the inner Solar System (average δ25Mg = − 0.273, δ26Mg = − 0.535) is homogeneous to within ± 0.04‰ for 26Mg/24Mg. This value is indistinguishable from that of the BSE and lunar samples demonstrating that the Mg isotopic composition of the Earth-Moon system is chondritic. Accurate determination of the Mg isotopic composition of the rocky planets is important for Mg isotope studies of surface processes on Earth as well as for using Mg isotope data to understand evaporation and condensation effects in the solar nebula.
In order to further investigate inter-mineral Mg isotope fractionation at mantle temperatures and to better constrain the Mg isotopic composition of the terrestrial mantle, we have studied a set of well-characterized mantle peridotite xenoliths from Sanyitang and Beiyan, North China craton. The Sanyitang and Beiyan peridotites, which have diverse origins with different mineralogy, chemical composition and degree of partial melting and metasomatism, display a small variation in Mg isotopic composition, with δ 26 Mg varying from − 0.48 to − 0.12 and an average value of − 0.29 ± 0.19 (2SD, n = 21) in olivines, from − 0.27 to − 0.10 and an average value of − 0.21 ± 0.09 (2SD, n = 12) in orthopyroxenes and from − 0.35 to − 0.08 and an average value of − 0.22 ± 0.14 (2SD, n = 15) in clinopyroxenes. The Mg isotopic compositions of the coexisting olivine, orthopyroxene and clinopyroxene in all peridotites are identical within our external precision (~ ± 0.1‰, 2SD), suggesting that Mg isotope fractionation between olivine and pyroxenes at the temperatures of > 900 °C is insignificant. These results are thus consistent with the absence of Mg isotope fractionation during basalt differentiation. The ~ 0.4‰ Mg isotopic variations in these peridotite samples, which are larger than our external precision, might result from thermal diffusion-driven isotope fractionation or melt-rock interactions. Overall, the δ 26 Mg of the mantle, based on peridotite minerals analyzed here, is estimated to be −0.26± 0.16 (2SD). This value is in excellent agreement with recent studies based on peridotites and oceanic basalts (Handler et al., 2009; Teng et al., 2007). It is also similar to δ 26 Mg values of two additional dunite standards (DTS-1 and DTS-2) and two carbonaceous chondrites (Allende and Murchison) analyzed in this study as well as δ 26 Mg of all published chondrites (~−0.3). The silicate Earth thus has a chondritic Mg isotopic composition.
The magnesium isotopic composition of Earth is not yet well constrained despite significant advances in methods for measuring Mg isotope ratios in rocks. One impediment to establishing 25 Mg/ 24 Mg and 26 Mg/ 24 Mg of Earth is the lack of constraints on inter-mineral Mg isotope fractionations at high temperatures. Advances in computational chemistry afford the capacity to predict quantitatively Mg isotope fractionations among high-temperature minerals. High-precision MC-ICPMS measurements in turn provide the opportunity to test these predictions in well-characterized samples. Toward this end, we present new high-precision 25 Mg/ 24 Mg and 26 Mg/ 24 Mg measurements of mantle minerals and compare these ratios with predictions for temperature-dependent inter-mineral fractionations. Our results for two San Carlos volcanic field xenoliths show that there is measurable and systematic fractionation in Mg isotope ratios between constituent minerals that are consistent with theoretical predictions. The observed order from highest to lowest 25 Mg/ 24 Mg is spinel > clinopyroxene > orthopyroxene > olivine. The fractionation between spinel and olivine suggests an equilibration temperature of 814° +/− 60 °C based on the temperature dependence obtained from ab initio calculations. This temperature is consistent with independent T indicators involving spinel, suggesting that spinel and olivine are in Mg isotopic equilibrium in these mantle rocks, and lending credence to the accuracy of the results. Pyroxene, on the other hand, is apparently not in Mg isotopic equilibrium with spinel and olivine if the predicted temperature-dependent fractionations are correct. Consideration of the influences of modal abundances and inter-mineral fractionations on the Mg isotopic compositions of mantle minerals, and comparisons to new meteorite data reported herein, strengthen previous suggestions that the Earth may be different from carbonaceous chondrites in 25 Mg/ 24 Mg.
Despite the growing interest for Li and B as geochemical tracers, especially for material transfer from subducting slabs to overlying peridotites, little is known about the behaviour of these two elements during partial melting of mantle sources. In particular, mineral/melt partition coefficients for B and to a lesser extent Li are still a matter of debate. In this work, we re-equilibrated a synthetic basalt doped with ~10 ppm B and ~6 ppm Li with an olivine powder from a spinel lherzolite xenolith at 1 GPa–1,330°C, and we analyzed Li and B in the run products by secondary ion mass spectrometry (SIMS). In our experiment, B behaved as a highly incompatible element with mineral/melt partition coefficients of the order of 10−2 (D
ol/melt = 0.008 (0.004–0.013); D
opx/melt = 0.024 (0.015–0.033); D
cpx/melt = 0.041 (0.021–0.061)), and Li as a moderately incompatible element (D
ol/melt = 0.427 (0.418–0.436); D
opx/melt = 0.211 (0.167–0.256); D
cpx/melt = 0.246 (0.229–0.264)). Our partition coefficients for Li are in good agreement with previous determinations. In the case of B, our partition coefficients are equal within error to those reported by Brenan et al. (1998) for all the mineral phases analyzed, but are lower than other coefficients from literature for some of the phases (up to 5 times for cpx). Our measurements complement the data set of Ds for modelling partial melting of the upper mantle and basalt generation, and confirm that, in this context, B is more incompatible than previously anticipated.