TY - JOUR
AU - H, Williams,
AB - Abstract Long-term carbon cycling is a subject of recent controversy as new mass balance calculations suggest that most carbon is transferred from the slab to the mantle wedge by fluids during subduction, limiting the efficiency of carbon recycling to the deep mantle. Here, we examine the large scale mobility of carbon during subduction using new isotopic tracers sensitive to H–C–O–S–Cl fluids, namely iron and zinc stable isotopes, in samples interpreted to represent residual slab (Queyras, Western Alps) and sub-arc mantle (Kohistan, Himalaya). We show that during subduction there are several stages of carbonate precipitation and dissolution at metasomatic interfaces between metasedimentary and ultramafic rocks in the slab. During the early stages of subduction, before the slab reaches the 300–400°C isotherms, the infiltration of sediment-derived fluids into ultramafic lithologies enhances carbonate precipitation in antigorite-bearing serpentinites. Carbonate storage in serpentinites, therefore, acts as a temporary reservoir of carbon in subduction zones. This episode is accompanied by a decrease in serpentinite iron isotope composition (δ56Fe), due to interaction with low-δ56Fe sediment-derived fluids, and an increase in the concentrations of fluid-mobile elements (e.g. B, Li, As). At higher temperatures (> 400°C), carbonate is leached from the serpentinites by fluids. This is accompanied by a decrease in serpentinite zinc isotope composition (δ66Zn) which we interpret as the release of a carbonate-bearing fluid with an isotopically heavy δ66Zn signature. Thermodynamic modelling shows that the sudden change in fluid carbon mobility is due to a decrease in the aCO2 of the fluids released during slab prograde metamorphism, which shifts from sediment- to serpentinite-dominated dehydration. This demonstrates that slab fluids bearing oxidized carbon (e.g. CO2), associated with isotopically light Fe, heavy Zn and fluid-mobile elements, can be released before the slab reaches eclogite facies P-T conditions. These observations provide strong evidence for the mobility of carbon in fluids during the early stages of subduction. Moreover, the fluids released will act as a potential metasomatic agent for the fore-arc mantle (or slab/mantle interface). The observation of carbonate-bearing metamorphic veins in the Himalayan sub-arc mantle with complementary light δ56Fe and heavy δ66Zn signatures provides further support for the large scale transfer of both sulphate- and carbonate-bearing fluids during the early stages of subduction. This suggests that the fore-arc may have an important role in delivering water, sulfur and carbon to the source of arc-magmas. INTRODUCTION The long-term carbon cycle on Earth has been linked to the rise of O2 in the atmosphere during the Great Oxidation Event at ∼ 2·5 Ga (e.g. Catling & Claire, 2005; Catling, 2012; Lyons et al., 2014; Duncan & Dasgupta, 2017) and to the appearance and evolution of life over geologic time (e.g. Sleep et al., 2011; Schrenk et al., 2013). On modern Earth, the extraction of carbon from the mantle to the surface mainly occurs at mid-oceanic ridges and subduction zones through partial melting and magmatic degassing processes (e.g. Ballhaus & Frost, 1994; Stagno et al., 2013; Poli, 2015). It is widely considered that between 20 to 80% of the carbon is then recycled into the deep mantle by subduction (Gorman et al., 2006; Dasgupta & Hirschmann, 2010; Cook-Kollars et al., 2014;,Collins et al., 2015). However, recent mass balance calculations suggest that most (up to 100%) of the carbon budget of the subducting oceanic lithosphere could be transferred from the slab to the mantle wedge by fluids (Kelemen & Manning, 2015), calling for a re-appraisal of carbon fluxes in subduction systems. Carbon uptake in oceanic lithosphere mainly occurs during its alteration by hydrothermal fluids. Indeed, seafloor lithologies, such as ophicarbonates, sediments and dredged mafic and ultramafic rocks, commonly display high carbonate concentrations (up to 3 wt % carbon) relative to pristine mantle peridotites (30 ppm, Dasgupta & Hirschmann, 2010) and are thus considered to be important reservoirs of carbon in the oceanic lithosphere (e.g. Alt & Teagle, 1999; Delacour et al., 2008; Alt et al., 2013; Kelemen & Manning, 2015). During subduction, numerous studies have shown that carbonates are soluble in aqueous fluids at relatively low temperatures (e.g. Caciagli & Manning, 2003; Frezzotti et al., 2011; Facq et al., 2014; Milesi et al., 2015) suggesting that, providing hydrous fluids are involved, slab decarbonation should be highly efficient and occur at relatively shallow depths (between 60 to 90 km depth; Gorman et al., 2006; Bouilhol et al., 2012; Kelemen & Manning, 2015). However, the existence of carbonated meta-basalts, meta-sediments and meta-ultramafites observed away from zones of intense fluid circulation and deformation (e.g. shear zones, veins, fractures) in Alpine meta-ophiolites suggests that slab-derived lithologies may retain a significant part of their carbon during subduction, at least away from major pathways of slab fluid release (Cook-Kollars et al., 2014; Collins et al., 2015). In addition, the presence of graphite in blueschist facies metasedimentary rocks, which has been interpreted in terms of carbonate reduction reactions taking place in the upper slab (Galvez et al., 2013), suggests that carbon initially released as carbonate in slab fluids may also reprecipitate in the upper part of the slab or at the slab/mantle interface without being transferred directly to the overlying mantle. Although the effect of late retrograde carbonation during massif exhumation remains sometimes unclear (e.g. Piccoli et al., 2016; Vitale-Brovarone et al., 2017), the abundance of marble (Piccoli et al., 2016; Scambelluri et al., 2016) and carbonate rocks (Kerrick & Connolly, 1998; Ague & Nicolesu, 2014; Debret et al., 2016a) in various blueschist or eclogitic meta-ophiolites also suggests that a significant part of the carbon budget of the slab remains immobile during prograde metamorphism. It has, therefore, been proposed that simple decarbonation reactions do not allow a major transfer of carbon from the subducted oceanic lithosphere to the mantle wedge and that other processes, such as the dissolution and leaching of carbonates from the upper slab or the slab/mantle wedge interface by external fluids, must also be considered (Frezzotti et al., 2011). Serpentinites comprise a major part of the subducting oceanic lithosphere hydrated near the seafloor at slow or ultra-slow spreading ridges (Canales et al., 2000; Andreani et al., 2007; Debret et al., 2013b) or during slab-bending (Ranero et al., 2003; Ranero & Sallares, 2004), and are also present as part of the down-dragged slab-wedge interface or mantle wedge that is percolated by aqueous fluids emanating from the dehydrating slab (e.g. Hattori & Guillot, 2007; Reynard, 2013). Serpentinites are thus ubiquitous in subduction zones. Due to their large P–T stability field, they are stable down to depths of ∼120–150 km and are considered to be the major source of aqueous fluids released from the slab to the mantle wedge (Ulmer & Trommsdorff, 1995). Because total carbon solubility increases as fluid CO2 activity (aCO2) decreases (Gorman et al., 2006), the large amount of fluids released during serpentinite dehydration can trigger extensive leaching of carbon-bearing phases, potentially removing all carbon from the downgoing plate and transferring it to the mantle wedge (Kelemen & Manning, 2015). However, little is known about the fate of this mobilized carbon, and whether these C-bearing fluids efficiently separate from the slab and slab/mantle wedge interface. Non-traditional stable isotopes (e.g. Fe, Zn) provide key constraints on elemental mobility and mass balance, as equilibrium stable isotope fractionation between different phases (such as serpentine minerals, Fe-oxides, sulfides, carbonate and fluids) is driven by contrasts in element bonding environment and oxidation state (Polyakov & Mineev, 2000; Schauble et al., 2001; Fujii et al., 2010, 2014). Theory predicts that equilibrium stable isotope fractionation decreases with increasing temperature (1/T2) (Urey, 1947; Schauble, 2004). Nonetheless, high-precision Fe and Zn stable isotope measurements have shown that both of these systems are sensitive to high-temperature petrogenetic processes, such as mantle melting (Williams et al., 2004, 2005, 2009; Weyer et al., 2005; Weyer & Ionov, 2007; Dauphas et al., 2014; Williams & Bizimis, 2014; Nebel et al., 2015; Doucet et al., 2016; Konter et al., 2016; Sossi et al., 2016), igneous differentiation (Teng et al., 2008, 2013; Schuessler et al., 2009; Telus et al., 2012; Chen et al., 2013; Nebel et al., 2015; Sossi et al., 2016) and for Fe, changes in redox state (Williams et al., 2004; Dauphas et al., 2009). More recently, Fe and Zn isotopes have been used to trace the mobility of Fe and oxidizing sulfate (SOX) species during the prograde devolitization of the subducted slab (Debret et al., 2016b; Pons et al., 2016; Inglis et al., 2017). Indeed, theory predicts that Fe mobility, and thus Fe isotopic fractionation, will be driven by the presence of Cl-, SO42- and CO32- anions, which preferentially complex isotopically light iron (Hill & Schauble, 2008, Hill et al., 2010; Fujii et al., 2014; Dauphas et al., 2017). Given that these elements are ubiquitous and these complexes abundant in fluids derived from serpentinites (e.g. Scambelluri et al., 2004, 2015; Debret & Sverjensky, 2017) and sedimentary rocks (e.g. Garofalo, 2012), Fe isotopes can, therefore, be used as potential tracers of volatile cycling in subduction zones. Zinc mobility in fluids is particularly sensitive to the presence of C- and S- bearing fluids, since HS- has an affinity for isotopically light Zn whereas CO32- and SO42- concentrate isotopically heavy Zn isotopes. Hence, the combination of Fe and Zn isotopes can be used as a mean of probing the speciation and oxidation state of carbon and sulfur in slab-derived fluids. In good agreement with this, theoretical calculations (Debret & Sverjensky, 2017) and stable isotope studies (Debret et al., 2016b; Pons et al., 2016) have shown that serpentinite devolatilization is associated with the release of Fe and Zn complexed with Cl- and SO42- anions in the fluids, resulting in significant Fe and Zn stable isotope fractionation. More recently, Inglis et al. (2017) also reported extensive Fe isotope variability in carbonate-bearing metasomatic interfaces between meta-sedimentary rocks and metabasites and suggested that Fe isotopes could also be used as potential tracers of sedimentary rock dehydration and potentially carbon mobility during subduction. In order to better constrain carbon mobility during prograde metamorphism in subduction zones, we have carried out an Fe and Zn stable isotope study of ultramafic rocks and ultramafic–sedimentary interfaces from the Western Alps (Queyras) that have experienced greenschist to blueschist facies metamorphism during the closure of the Piedmont Ocean. We show that these rocks were carbonated by CO2 rich-fluids derived from metasedimentary rocks during the early stages of subduction (< 300°C) and then decarbonated at greater depths. In order to expand our study into a larger framework relating to the carbon cycle in subduction zones, we also analysed a suite of olivines preserved in carbonate-bearing veins from the Kohistan sub-arc mantle (Himalaya). These samples have been interpreted as the products of fluid migration through fore-arc mantle (Bouilhol et al. 2009, 2012) and have distinct Zn isotope compositions (Pons et al., 2016) reflecting this origin. These samples may, therefore, provide a potential end-member of the fluids metasomatizing the mantle wedge, allowing us to track the signature of carbon-bearing fluids at the scale of the Alpine-Himalayan orogeny. GEOLOGICAL SETTINGS AND SAMPLE PETROLOGY The Queyras Schistes Lustrés complex is located in the Piedmont zone of the southwestern Alps (Fig. 1). It is composed of units belonging to the distal European margin and from the nearby oceanic domain (Lemoine et al., 1987) that were juxtaposed during the Alpine subduction and collision in Late Cretaceous to Tertiary times (Tricart, 1984). The complex consists of ∼ 10% of meta-ophiolites embedded in a sedimentary-rich environment consisting of Jurassic to Lower Cretaceous metasedimentary rocks (Lagabrielle et al., 1984; Lemoine et al., 1987). The protoliths of the Schistes Lustrés exposed in the Western Alps are inferred to be carbonate-rich deep-sea sediments with a large degree of heterogeneity, inherent with such types of lithologies (Agard et al., 2002; Bebout et al., 2013; Cook-Kollars et al., 2014). The Queyras meta-ophiolites are mainly composed of mafic and ultramafic olistoliths (referred to as ultramafic boudins hereafter) that are up to 200 m in size and interbedded in metasedimentary rocks (Lagabrielle et al., 2014). Some of the largest olistoliths preserve remnants of primary contacts between ultramafic rocks and their sedimentary cover. Those rocks were highly hydrated and altered at seafloor level before undergoing high-pressure and temperature metamorphism during alpine subduction (Schwartz et al., 2013; Debret et al., 2016a). Fig. 1 View largeDownload slide Simplified geological map of the studied area showing the location of the studied meta-ophiolites (modified after Schwartz et al., 2013). Fig. 1 View largeDownload slide Simplified geological map of the studied area showing the location of the studied meta-ophiolites (modified after Schwartz et al., 2013). Based on their metamorphic grade, three different tectono-metamorphic units have been identified in the Queyras Schistes Lustrés complex (Fig. 1). These record variable P-T conditions during Alpine subduction, increasing from low-temperature blueschist facies conditions (LT-blueschist unit; 320–360°C and 9–11 kbar) in western Queyras to medium-temperature (MT-blueschist unit; 340–390°C and 10–12 kbar) and high-temperature blueschist (HT-blueschist unit; 380–470°C and 12–18 kbar) conditions towards the East (Ballèvre et al., 1990; Agard et al., 2001; Tricart & Schwartz, 2006; Schwartz et al., 2013; Lagabrielle et al., 2014). In this area, previous petrological and geochemical studies of the Queyras meta-ophiolites have identified extensive fluid-rock interactions between mafic/ultramafic and meta-sedimentary rocks occurring at HT–HP during subduction, resulting in the recrystallization and chemical hybridization (metasomatic mixing with hybrid compositions) between mafic/ultramafic and meta-sedimentary rocks at their interface (Debret et al., 2016a). We sampled three different meta-ophiolites across the complex (Fig. 1b) at localities where metasomatic contacts between metasedimentary and ultramafic/mafic rocks are preserved. The Col Peas meta-ophiolite belongs to the LT-blueshist unit (Fig. 1b) and is composed of ultramafic and mafic boudins embedded within highly deformed metasedimentary rocks (Fig. 2a). The contact between metasedimentary rocks and ultramafic boudins is defined by diopside-rich serpentinites (Di-serpentinite) that display a strong foliation and thin (centimetres width) green and white layers (Fig. 2b). The green layers are mainly composed of oriented lamellae of chlorite and antigorite associated with small grains of titanite, sulfides and Fe-rich oxides (magnetite). The oriented lamellae of antigorite and chlorite mark the foliation of the rocks (Fig. 2c). The white layers are made of randomly orientated diopside needles up to 300 µm long with interstitial chlorite with carbonate-rich clusters in their centres (Fig. 2c, d). These observations are interpreted in terms of the post-deformation crystallization of diopside at the expense of carbonate. Towards the serpentinite boudins, the amount of diopside and chlorite decreases and serpentinites are mainly composed of antigorite lamellae associated with accessory Fe-rich oxides (magnetite and mantle spinel) and Ca-carbonate veins. The centres of the serpentinite boudins preserve mesh and bastite textures formed after mantle olivine and pyroxene, respectively, interpreted to have formed during seafloor alteration (Lafay et al., 2013; Schwartz et al., 2013). Rare relicts of mantle clinopyroxene can be preserved in the centre of serpentine textures. The mesh textures are associated with thin magnetite veins ∼50 µm wide (Fig. 2e) while the bastite textures are magnetite-free and associated with titanite grains ∼30 µm wide. Fig. 2 View largeDownload slide Field photographs and photomicrographs of the ultramafic rocks sampled at the Col Peas meta-ophiolite (LT-blueschist unit). (a) Field photograph of an ultramafc boudin embedded in metasedimentary rocks. (b) Metasomatic contact between metasedimentary and ultramafic rocks. The interface between both lithologies is marked by a diopside-bearing serpentinite layer (Di-serpentinite, CP2 sample). The arrows indicate carbonate veins cross-cutting Atg-serpentinites. (c and d) Photomicrographs of a Di-serpentinite in crossed polarized light (CP2 sample). (c) The Di-serpentinites are made of an alternation of antigorite (+/- chlorite, magnetite, chromite and sulfide) and diopside-rich layers. Antigorite lamellae are orientated according to rock foliation. (d) Ca-carbonate cluster preserved in diopside-rich layers. (e) Photomicrograph of a serpentinite boudin core in plane polarized light (CP8 sample). Mesh and bastite textures replace, respectively, olivine and orthopyroxene in serpentinite boudin cores. The mesh textures are associated with abundant magnetite. Fig. 2 View largeDownload slide Field photographs and photomicrographs of the ultramafic rocks sampled at the Col Peas meta-ophiolite (LT-blueschist unit). (a) Field photograph of an ultramafc boudin embedded in metasedimentary rocks. (b) Metasomatic contact between metasedimentary and ultramafic rocks. The interface between both lithologies is marked by a diopside-bearing serpentinite layer (Di-serpentinite, CP2 sample). The arrows indicate carbonate veins cross-cutting Atg-serpentinites. (c and d) Photomicrographs of a Di-serpentinite in crossed polarized light (CP2 sample). (c) The Di-serpentinites are made of an alternation of antigorite (+/- chlorite, magnetite, chromite and sulfide) and diopside-rich layers. Antigorite lamellae are orientated according to rock foliation. (d) Ca-carbonate cluster preserved in diopside-rich layers. (e) Photomicrograph of a serpentinite boudin core in plane polarized light (CP8 sample). Mesh and bastite textures replace, respectively, olivine and orthopyroxene in serpentinite boudin cores. The mesh textures are associated with abundant magnetite. The Echassier meta-ophiolite is located to the south of the Queyras Schistes Lustrés complex and belongs to the MT-blueschist unit (Fig. 1b). As in the Col Peas, it is composed of metre-scale ultramafic and mafic boudins enveloped in highly deformed metasedimentary rocks. At the contact with the meta-sedimentary rocks, the ultramafic rocks are massive and display ophicarbonate-like textures composed of a white vein network cross–cutting a dark blue ultramafic host rock (Fig. 3a). In thin section the white veins are carbonate-free and exclusively composed of lamellar diopside aggregates displaying pseudomorphic textures after olivine and pyroxene (Fig. 3b). Rare olivine relicts are observed in the centre of the diospide pseudomorphic aggregates. The ultramafic part of these rocks is composed exclusively of randomly orientated chlorite lamellae. The rims of the ultramafic boudins are composed of antigorite lamellae associated with chlorite, chromite, Al-spinel and Ca-carbonate, whereas the boudin centre preserves mesh and bastite textures and mantle clinopyroxene partly recrystallized into antigorite (Fig. 3c). The bastite textures are associated with diopside lamellae. Fig. 3 View largeDownload slide Field photographs and photomicrographs of the ultramafic rocks sampled at the Echassier meta-ophiolite (MT-blueschist unit). (a) White vein network observed within the ultramafic rocks (CE14b sample) crystallizing at the metasedimentary rocks contact. (b) Photomicrograph in plane polarized light of a white vein (CE14b sample). These veins consist of diopside lamellae less than 100 µm long with interstitial chlorite and display pseudomorphic textures after olivine (left hand side) and pyroxene (right hand side). (c) Photomicrograph in plane polarized light of serpentinite boudin core. Mesh textures are partly overgrown by antigorite lamellae (CE3 sample). Fig. 3 View largeDownload slide Field photographs and photomicrographs of the ultramafic rocks sampled at the Echassier meta-ophiolite (MT-blueschist unit). (a) White vein network observed within the ultramafic rocks (CE14b sample) crystallizing at the metasedimentary rocks contact. (b) Photomicrograph in plane polarized light of a white vein (CE14b sample). These veins consist of diopside lamellae less than 100 µm long with interstitial chlorite and display pseudomorphic textures after olivine (left hand side) and pyroxene (right hand side). (c) Photomicrograph in plane polarized light of serpentinite boudin core. Mesh textures are partly overgrown by antigorite lamellae (CE3 sample). The Refuge du Viso meta-ophiolite, located in the east, is mostly composed of serpentinites with metagabbroic pods and was metamorphosed at HT-blueschist facies conditions during subduction (Fig. 1b). These serpentinites are themselves enveloped in highly deformed metasedimentary rocks. The contact between the meta-sedimentary envelope and the serpentinites is relatively sharp and is locally defined by an epidote and chlorite-rich layer ∼ 2 m wide. At the contact with this chlorite/epidote layer, the serpentinites are highly deformed and mainly composed of orientated antigorite lamellae with magnetite and olivine crystallizing along shear planes (Fig. 4). The crystallization of olivine in the Refuge du Viso meta-ophiolite has been previously attributed to the onset of serpentinite devolatilization during brucite breakdown (Schwartz et al., 2013). In the centre of this massif, serpentinites are massive and display early mesh and bastite textures partly to fully recrystallized into antigorite and associated with magnetite and chromite. Fig. 4 View largeDownload slide Photomicrograph in crossed polarized light of a sheared serpentinite (RV3) sampled at the contact with metasedimentary rocks in the Refuge du Viso meta-ophiolite (HT-blueschist unit). Olivine and magnetite C-planes cross-cut S-planes highlighted by orientated antigorite lamellae. Fig. 4 View largeDownload slide Photomicrograph in crossed polarized light of a sheared serpentinite (RV3) sampled at the contact with metasedimentary rocks in the Refuge du Viso meta-ophiolite (HT-blueschist unit). Olivine and magnetite C-planes cross-cut S-planes highlighted by orientated antigorite lamellae. In the Queyras Schistes Lustrès complex, the crystallization of lizardite-bearing textures (mesh and bastite) has been previously assigned to sea-floor metamorphism, whereas the crystallization of antigorite at the expense of lizardite is attributed to the influx of SiO2 rich fluids derived from metasedimentary rocks during the onset of subduction (Schwartz et al., 2013; Lafay et al., 2013). In agreement with these studies, our field and petrographic observations of these blueschist units have revealed that the crystallization of antigorite at the expense of lizardite mainly occurs at boudin rims (Figs 2a and 4). In addition, we identify the crystallization products of various stages of carbonation and decarbonation during these fluid-rock interactions (Figs 2d, 3band 5), manifest by the successive precipitation of Ca-carbonate and diopside in the ultramafic rocks formed at the contact with meta-sedimentary rocks. The interface between ultramafic and metasedimentary rocks is defined by Di-serpentinites (Fig. 5). With progressive distance from these contact zones, the mineralogy of the serpentinite boudins evolves from antigorite with Ca-carbonate/diopside at the boudin rims to lizardite (± Atg) -bearing serpentinite within the boudin cores. Fig. 5 View largeDownload slide Schematic sketch summarizing the field and petrographic characteristics of ultramafic rocks in each Queyras meta-ophiolite. (a) Large scale schematic drawing showing ultramafic boudins embedded within metasedimentary rocks. The box shows the studied contact between metasedimentary and ultramafic rocks. (b) In the LT-blueschist unit, the boudin cores preserve mesh and bastite textures mainly made of lizardite. The boudin rims are composed of antigorite with rare carbonate layers. The metasomatic contact is made of antigorite  ±  chlorite and diopside layers. The diopside layers preserve carbonate in their centre. (c) In the MT-blueschist unit, antigorite appears along lizardite grain boundaries in the boudin core whereas the boudin rims are mainly composed of antigorite with rare carbonate. The metasomatic contacts are carbonate free and mainly composed of chlorite and diopside. (d) In the HT-blueschist unit, the boudin core is mainly composed of antigorite with relicts of lizardite-bearing textures. In the boudin rim brucite breakdown results in metamorphic (secondary) olivine crystallization. We observed an epidote-rich layer at the contact with the metasedimentary rocks. Sample names are reported on each schematic drawing. Fig. 5 View largeDownload slide Schematic sketch summarizing the field and petrographic characteristics of ultramafic rocks in each Queyras meta-ophiolite. (a) Large scale schematic drawing showing ultramafic boudins embedded within metasedimentary rocks. The box shows the studied contact between metasedimentary and ultramafic rocks. (b) In the LT-blueschist unit, the boudin cores preserve mesh and bastite textures mainly made of lizardite. The boudin rims are composed of antigorite with rare carbonate layers. The metasomatic contact is made of antigorite  ±  chlorite and diopside layers. The diopside layers preserve carbonate in their centre. (c) In the MT-blueschist unit, antigorite appears along lizardite grain boundaries in the boudin core whereas the boudin rims are mainly composed of antigorite with rare carbonate. The metasomatic contacts are carbonate free and mainly composed of chlorite and diopside. (d) In the HT-blueschist unit, the boudin core is mainly composed of antigorite with relicts of lizardite-bearing textures. In the boudin rim brucite breakdown results in metamorphic (secondary) olivine crystallization. We observed an epidote-rich layer at the contact with the metasedimentary rocks. Sample names are reported on each schematic drawing. In order to decipher better the nature of the fluid(s) circulating in these zones, we sampled metasomatic interfaces (CP2 and CE14b) as well as the rims (CP3, CE10 and RV3) and cores (CP8, CE3 and RV8) of serpentinite boudins composing the Col Peas (LT-blueschist unit), Echassier (MT-blueschist unit) and Refuge du Viso (HT-blueschist unit) meta-ophiolites (Fig. 5b–d). To expand our sample set we have also analysed a suite of olivines (OG, Ogi, OGM and OG1) from the Kohistan sub-arc mantle (Himalaya) for Fe isotopes. These were sampled from calcite–magnetite bearing veins cross-cutting sub-arc mantle peridotites (Bouilhol et al., 2009, 2012). The vein-forming minerals crystallized from a fluid that migrated through the fore-arc mantle, constituting a rare field expression of a potential end-member of the fluids metasomatizing the mantle wedge in subduction zones. In detail, these samples belong to the Sapat complex which represents a part of the sub-arc mantle of the Kohistan–Ladakh Arc, an intra-oceanic island arc relict formed during the subduction of the Tethys Ocean in the Mesozoic (Tahirkheli et al., 1979; Bouilhol et al., 2013). The Sapat sub-arc mantle has been overprinted by slab-derived fluids, in a fore-arc position during subduction, as testified by the formation of olivine–calcite-magnetite bearing veins crossing serpentinized ultramafic rocks (Bouilhol et al., 2009, 2012). These veins are mainly composed of olivine, calcite and magnetite ± Cr-clinochlore, and are often overprinted by serpentine and brucite (Fig. 6). Trace element analyses of vein-bearing minerals, combined with O, C and Sr isotopes indicate that the veins were formed at high temperature (T > 500°C) from H2O–CO2 and fluid-mobile element (e.g. Li, B) rich fluids partly equilibrated with both mantle, slab and/or wedge derived components (Bouilhol et al., 2012). In addition, a recent Zn isotope study (Pons et al., 2016) of the same samples provides evidence for the formation of those veins in equilibrium with oxidized and sulfate-rich fluids derived from the devolatilization of slab serpentinites. These veins, therefore, record a complex history of interactions between sulfate-bearing serpentinite-derived fluids and carbonate-bearing lithologies such as metasedimentary rocks and ophicarbonates derived from the upper slab or the slab/mantle interface, and the sub-arc mantle. Fig. 6 View largeDownload slide Field occurrence of Kohistan gem olivine-bearing veins. (a) Tension gash composed of calcite, magnetite (with brucite and serpentine) and olivine (modified from Bouilhol et al., 2012). (b) Vein fragment containing olivine, calcite and magnetite. Fig. 6 View largeDownload slide Field occurrence of Kohistan gem olivine-bearing veins. (a) Tension gash composed of calcite, magnetite (with brucite and serpentine) and olivine (modified from Bouilhol et al., 2012). (b) Vein fragment containing olivine, calcite and magnetite. METHODS Geochemistry Major and trace element concentrations of the Kohistan gem olivines are provided in Bouilhol et al. (2009, 2012). In situ major element analyses of Queyras serpentinite minerals were performed with a CAMECA SX 100 electron microprobe at the Laboratoire Magmas et Volcans in Clermont-Ferrand (France). Results are reported in Supplementary Data Electronic Appendix A; supplementary data are available for downloading at http://www.petrology.oxfordjournals.org. Bulk-rock major element and volatile (S and C) concentrations were analysed using a PANalytical Axios-Advanced XRF spectrometer and Leco CS230 Carbon/Sulphur Determinator, respectively, at the University of Leicester (UK). Bulk trace element concentrations (Li, Sc, V, Co, Ni, Cu, Zn, As, Rb, Sr, Y, Zr, Nb, Cd, Sb, Cs, Ba, Rare Earth Elements (REE), Hf, Pb, Th, and U) were analysed at the National Oceanography Centre, Southampton using a Thermo X-Series Quadrupole ICP-MS. For these analyses 10 mg of powdered sample were digested using concentrated hydrofluoric and nitric acids. These solutions were then dried down and the residue dissolved in 2% HNO3, which was spiked with 10 ng/mL In and Re to correct for internal drift. The external precision and accuracy of the analyses were assessed by measuring as unknown three rock standards: BHVO and BIR-1 basalts and JA-2 peridotite (Supplementary Data Electronic Appendix B). Our results show good agreement between measured values and expected values for the international standards, and external reproducibility is within 0–5% for Sr, Y, Nb, La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu, Hf, U and Co, and within 5–10% for Li, Sc, Rb, Zr, Cs, Ba, Pb, Th, V, Ni, Cu and Zn. The values obtained for rock standards BHVO, BIR-1 and JA-2 during this study are reported in Supplementary Data Electronic Appendix B. Results of whole-rock analyses are reported in Tables 1 and 2. Table 1 Sample major and trace element composition Sample name CP2 CP3 CP8 CE14b CE10 CE3 RV3 RV8 Lithology Di-serp. Serp. Rim Serp. Core Di-serp. Serp. Rim Serp. Core Serp. Rim Serp. Core SiO2 (wt %) 41·8 39·3 38·2 36·2 40·3 39·2 39·9 39·3 Al2O3 3·7 2·5 3·0 12·6 1·8 2·7 1·4 2·1 Fe2O3 7·0 9·3 8·3 7·8 8·0 7·9 7·1 7·1 MgO 30·7 35·0 37·1 26·6 36·3 36·1 39·3 39·1 CaO 6·6 1·9 0·1 6·0 0·1 0·9 <0·1 <0·1 Na2O <0·1 <0·1 <0·1 0·1 <0·1 <0·1 <0·1 <0·1 K2O <0·1 <0·1 <0·1 <0·1 <0·1 <0·1 <0·1 <0·1 TiO2 0·1 0·1 0·1 0·1 0·1 0·1 0·0 0·0 MnO 0·1 0·1 0·1 0·2 0·1 0·1 0·1 0·1 P2O5 <0·1 <0·1 <0·1 0·1 <0·1 <0·1 <0·1 <0·1 LOI 8·6 10·7 11·9 8·6 11·5 11·2 11·5 11·7 Total 98·7 98·9 98·9 98·3 98·2 98·3 99·3 99·4 Li (ppm) 17·73 6·49 7·26 32·73 9·03 2·82 0·01 0·01 Sc 11·43 13·79 14·72 17·70 12·63 14·26 11·97 10·15 Rb 0·56 0·19 0·21 0·07 0·24 0·08 0·00 0·00 Sr 18·29 9·26 4·39 10·54 3·06 1·39 0·18 0·26 Y 3·60 2·51 3·88 3·31 1·46 2·63 0·67 2·00 Zr 12·64 5·38 4·39 3·94 0·79 1·80 0·22 2·28 Nb 1·06 0·03 0·01 0·04 0·01 0·02 0·01 0·01 Cs 1·23 0·47 1·59 0·18 1·41 0·66 0·00 0·00 Ba 3·60 5·65 7·68 1·79 12·75 2·84 0·21 0·11 La 1·18 0·21 0·06 0·22 0·06 0·04 0·24 0·27 Ce 2·94 0·52 0·36 0·79 0·17 0·15 0·47 0·82 Pr 0·46 0·10 0·09 0·16 0·03 0·03 0·06 0·14 Nd 2·05 0·56 0·60 0·92 0·17 0·24 0·24 0·70 Sm 0·55 0·22 0·27 0·35 0·07 0·14 0·06 0·22 Eu 0·13 0·09 0·12 0·20 0·04 0·06 0·01 0·03 Gd 0·58 0·35 0·44 0·49 0·13 0·26 0·07 0·26 Tb 0·10 0·06 0·08 0·09 0·03 0·05 0·01 0·05 Dy 0·62 0·42 0·58 0·56 0·20 0·40 0·09 0·29 Ho 0·13 0·09 0·13 0·12 0·05 0·09 0·02 0·06 Er 0·40 0·27 0·40 0·34 0·16 0·29 0·07 0·19 Tm 0·06 0·04 0·06 0·05 0·02 0·05 0·01 0·03 Yb 0·41 0·28 0·42 0·37 0·18 0·30 0·09 0·22 Lu 0·06 0·05 0·07 0·06 0·03 0·05 0·02 0·04 Hf 0·38 0·17 0·16 0·13 0·04 0·10 0·01 0·08 Pb 0·20 0·15 0·12 0·06 0·48 0·31 0·08 0·21 Th 0·870 0·004 0·001 0·006 b.d.l. <0·001 <0·001 0·003 U 0·235 0·013 0·001 0·002 0·036 0·001 0·001 0·003 V 54·51 58·82 62·05 86·95 54·83 65·81 46·26 42·80 Cr 2011 2006 1876 n.d. 3631 2479 2057 2448 Co 82·64 98·92 74·75 42·52 119·04 102·48 89·72 127·00 Ni 1601 1938 1880 n.d. 2557 2075 1696 2635 Cu 15·81 13·53 12·62 29·82 18·59 5·15 7·02 13·38 Zn 38·37 38·50 56·98 46·29 39·95 43·24 44·97 48·41 B 26·78 58·47 40·04 1·20 25·79 15·72 14·24 19·96 As 1·14 1·05 0·83 0·84 0·99 1·53 0·94 0·89 Sb 0·03 0·03 0·04 0·16 0·17 0·31 0·03 0·03 Sample name CP2 CP3 CP8 CE14b CE10 CE3 RV3 RV8 Lithology Di-serp. Serp. Rim Serp. Core Di-serp. Serp. Rim Serp. Core Serp. Rim Serp. Core SiO2 (wt %) 41·8 39·3 38·2 36·2 40·3 39·2 39·9 39·3 Al2O3 3·7 2·5 3·0 12·6 1·8 2·7 1·4 2·1 Fe2O3 7·0 9·3 8·3 7·8 8·0 7·9 7·1 7·1 MgO 30·7 35·0 37·1 26·6 36·3 36·1 39·3 39·1 CaO 6·6 1·9 0·1 6·0 0·1 0·9 <0·1 <0·1 Na2O <0·1 <0·1 <0·1 0·1 <0·1 <0·1 <0·1 <0·1 K2O <0·1 <0·1 <0·1 <0·1 <0·1 <0·1 <0·1 <0·1 TiO2 0·1 0·1 0·1 0·1 0·1 0·1 0·0 0·0 MnO 0·1 0·1 0·1 0·2 0·1 0·1 0·1 0·1 P2O5 <0·1 <0·1 <0·1 0·1 <0·1 <0·1 <0·1 <0·1 LOI 8·6 10·7 11·9 8·6 11·5 11·2 11·5 11·7 Total 98·7 98·9 98·9 98·3 98·2 98·3 99·3 99·4 Li (ppm) 17·73 6·49 7·26 32·73 9·03 2·82 0·01 0·01 Sc 11·43 13·79 14·72 17·70 12·63 14·26 11·97 10·15 Rb 0·56 0·19 0·21 0·07 0·24 0·08 0·00 0·00 Sr 18·29 9·26 4·39 10·54 3·06 1·39 0·18 0·26 Y 3·60 2·51 3·88 3·31 1·46 2·63 0·67 2·00 Zr 12·64 5·38 4·39 3·94 0·79 1·80 0·22 2·28 Nb 1·06 0·03 0·01 0·04 0·01 0·02 0·01 0·01 Cs 1·23 0·47 1·59 0·18 1·41 0·66 0·00 0·00 Ba 3·60 5·65 7·68 1·79 12·75 2·84 0·21 0·11 La 1·18 0·21 0·06 0·22 0·06 0·04 0·24 0·27 Ce 2·94 0·52 0·36 0·79 0·17 0·15 0·47 0·82 Pr 0·46 0·10 0·09 0·16 0·03 0·03 0·06 0·14 Nd 2·05 0·56 0·60 0·92 0·17 0·24 0·24 0·70 Sm 0·55 0·22 0·27 0·35 0·07 0·14 0·06 0·22 Eu 0·13 0·09 0·12 0·20 0·04 0·06 0·01 0·03 Gd 0·58 0·35 0·44 0·49 0·13 0·26 0·07 0·26 Tb 0·10 0·06 0·08 0·09 0·03 0·05 0·01 0·05 Dy 0·62 0·42 0·58 0·56 0·20 0·40 0·09 0·29 Ho 0·13 0·09 0·13 0·12 0·05 0·09 0·02 0·06 Er 0·40 0·27 0·40 0·34 0·16 0·29 0·07 0·19 Tm 0·06 0·04 0·06 0·05 0·02 0·05 0·01 0·03 Yb 0·41 0·28 0·42 0·37 0·18 0·30 0·09 0·22 Lu 0·06 0·05 0·07 0·06 0·03 0·05 0·02 0·04 Hf 0·38 0·17 0·16 0·13 0·04 0·10 0·01 0·08 Pb 0·20 0·15 0·12 0·06 0·48 0·31 0·08 0·21 Th 0·870 0·004 0·001 0·006 b.d.l. <0·001 <0·001 0·003 U 0·235 0·013 0·001 0·002 0·036 0·001 0·001 0·003 V 54·51 58·82 62·05 86·95 54·83 65·81 46·26 42·80 Cr 2011 2006 1876 n.d. 3631 2479 2057 2448 Co 82·64 98·92 74·75 42·52 119·04 102·48 89·72 127·00 Ni 1601 1938 1880 n.d. 2557 2075 1696 2635 Cu 15·81 13·53 12·62 29·82 18·59 5·15 7·02 13·38 Zn 38·37 38·50 56·98 46·29 39·95 43·24 44·97 48·41 B 26·78 58·47 40·04 1·20 25·79 15·72 14·24 19·96 As 1·14 1·05 0·83 0·84 0·99 1·53 0·94 0·89 Sb 0·03 0·03 0·04 0·16 0·17 0·31 0·03 0·03 n.d., non determined; b.d.l., below detection limit. Table 1 Sample major and trace element composition Sample name CP2 CP3 CP8 CE14b CE10 CE3 RV3 RV8 Lithology Di-serp. Serp. Rim Serp. Core Di-serp. Serp. Rim Serp. Core Serp. Rim Serp. Core SiO2 (wt %) 41·8 39·3 38·2 36·2 40·3 39·2 39·9 39·3 Al2O3 3·7 2·5 3·0 12·6 1·8 2·7 1·4 2·1 Fe2O3 7·0 9·3 8·3 7·8 8·0 7·9 7·1 7·1 MgO 30·7 35·0 37·1 26·6 36·3 36·1 39·3 39·1 CaO 6·6 1·9 0·1 6·0 0·1 0·9 <0·1 <0·1 Na2O <0·1 <0·1 <0·1 0·1 <0·1 <0·1 <0·1 <0·1 K2O <0·1 <0·1 <0·1 <0·1 <0·1 <0·1 <0·1 <0·1 TiO2 0·1 0·1 0·1 0·1 0·1 0·1 0·0 0·0 MnO 0·1 0·1 0·1 0·2 0·1 0·1 0·1 0·1 P2O5 <0·1 <0·1 <0·1 0·1 <0·1 <0·1 <0·1 <0·1 LOI 8·6 10·7 11·9 8·6 11·5 11·2 11·5 11·7 Total 98·7 98·9 98·9 98·3 98·2 98·3 99·3 99·4 Li (ppm) 17·73 6·49 7·26 32·73 9·03 2·82 0·01 0·01 Sc 11·43 13·79 14·72 17·70 12·63 14·26 11·97 10·15 Rb 0·56 0·19 0·21 0·07 0·24 0·08 0·00 0·00 Sr 18·29 9·26 4·39 10·54 3·06 1·39 0·18 0·26 Y 3·60 2·51 3·88 3·31 1·46 2·63 0·67 2·00 Zr 12·64 5·38 4·39 3·94 0·79 1·80 0·22 2·28 Nb 1·06 0·03 0·01 0·04 0·01 0·02 0·01 0·01 Cs 1·23 0·47 1·59 0·18 1·41 0·66 0·00 0·00 Ba 3·60 5·65 7·68 1·79 12·75 2·84 0·21 0·11 La 1·18 0·21 0·06 0·22 0·06 0·04 0·24 0·27 Ce 2·94 0·52 0·36 0·79 0·17 0·15 0·47 0·82 Pr 0·46 0·10 0·09 0·16 0·03 0·03 0·06 0·14 Nd 2·05 0·56 0·60 0·92 0·17 0·24 0·24 0·70 Sm 0·55 0·22 0·27 0·35 0·07 0·14 0·06 0·22 Eu 0·13 0·09 0·12 0·20 0·04 0·06 0·01 0·03 Gd 0·58 0·35 0·44 0·49 0·13 0·26 0·07 0·26 Tb 0·10 0·06 0·08 0·09 0·03 0·05 0·01 0·05 Dy 0·62 0·42 0·58 0·56 0·20 0·40 0·09 0·29 Ho 0·13 0·09 0·13 0·12 0·05 0·09 0·02 0·06 Er 0·40 0·27 0·40 0·34 0·16 0·29 0·07 0·19 Tm 0·06 0·04 0·06 0·05 0·02 0·05 0·01 0·03 Yb 0·41 0·28 0·42 0·37 0·18 0·30 0·09 0·22 Lu 0·06 0·05 0·07 0·06 0·03 0·05 0·02 0·04 Hf 0·38 0·17 0·16 0·13 0·04 0·10 0·01 0·08 Pb 0·20 0·15 0·12 0·06 0·48 0·31 0·08 0·21 Th 0·870 0·004 0·001 0·006 b.d.l. <0·001 <0·001 0·003 U 0·235 0·013 0·001 0·002 0·036 0·001 0·001 0·003 V 54·51 58·82 62·05 86·95 54·83 65·81 46·26 42·80 Cr 2011 2006 1876 n.d. 3631 2479 2057 2448 Co 82·64 98·92 74·75 42·52 119·04 102·48 89·72 127·00 Ni 1601 1938 1880 n.d. 2557 2075 1696 2635 Cu 15·81 13·53 12·62 29·82 18·59 5·15 7·02 13·38 Zn 38·37 38·50 56·98 46·29 39·95 43·24 44·97 48·41 B 26·78 58·47 40·04 1·20 25·79 15·72 14·24 19·96 As 1·14 1·05 0·83 0·84 0·99 1·53 0·94 0·89 Sb 0·03 0·03 0·04 0·16 0·17 0·31 0·03 0·03 Sample name CP2 CP3 CP8 CE14b CE10 CE3 RV3 RV8 Lithology Di-serp. Serp. Rim Serp. Core Di-serp. Serp. Rim Serp. Core Serp. Rim Serp. Core SiO2 (wt %) 41·8 39·3 38·2 36·2 40·3 39·2 39·9 39·3 Al2O3 3·7 2·5 3·0 12·6 1·8 2·7 1·4 2·1 Fe2O3 7·0 9·3 8·3 7·8 8·0 7·9 7·1 7·1 MgO 30·7 35·0 37·1 26·6 36·3 36·1 39·3 39·1 CaO 6·6 1·9 0·1 6·0 0·1 0·9 <0·1 <0·1 Na2O <0·1 <0·1 <0·1 0·1 <0·1 <0·1 <0·1 <0·1 K2O <0·1 <0·1 <0·1 <0·1 <0·1 <0·1 <0·1 <0·1 TiO2 0·1 0·1 0·1 0·1 0·1 0·1 0·0 0·0 MnO 0·1 0·1 0·1 0·2 0·1 0·1 0·1 0·1 P2O5 <0·1 <0·1 <0·1 0·1 <0·1 <0·1 <0·1 <0·1 LOI 8·6 10·7 11·9 8·6 11·5 11·2 11·5 11·7 Total 98·7 98·9 98·9 98·3 98·2 98·3 99·3 99·4 Li (ppm) 17·73 6·49 7·26 32·73 9·03 2·82 0·01 0·01 Sc 11·43 13·79 14·72 17·70 12·63 14·26 11·97 10·15 Rb 0·56 0·19 0·21 0·07 0·24 0·08 0·00 0·00 Sr 18·29 9·26 4·39 10·54 3·06 1·39 0·18 0·26 Y 3·60 2·51 3·88 3·31 1·46 2·63 0·67 2·00 Zr 12·64 5·38 4·39 3·94 0·79 1·80 0·22 2·28 Nb 1·06 0·03 0·01 0·04 0·01 0·02 0·01 0·01 Cs 1·23 0·47 1·59 0·18 1·41 0·66 0·00 0·00 Ba 3·60 5·65 7·68 1·79 12·75 2·84 0·21 0·11 La 1·18 0·21 0·06 0·22 0·06 0·04 0·24 0·27 Ce 2·94 0·52 0·36 0·79 0·17 0·15 0·47 0·82 Pr 0·46 0·10 0·09 0·16 0·03 0·03 0·06 0·14 Nd 2·05 0·56 0·60 0·92 0·17 0·24 0·24 0·70 Sm 0·55 0·22 0·27 0·35 0·07 0·14 0·06 0·22 Eu 0·13 0·09 0·12 0·20 0·04 0·06 0·01 0·03 Gd 0·58 0·35 0·44 0·49 0·13 0·26 0·07 0·26 Tb 0·10 0·06 0·08 0·09 0·03 0·05 0·01 0·05 Dy 0·62 0·42 0·58 0·56 0·20 0·40 0·09 0·29 Ho 0·13 0·09 0·13 0·12 0·05 0·09 0·02 0·06 Er 0·40 0·27 0·40 0·34 0·16 0·29 0·07 0·19 Tm 0·06 0·04 0·06 0·05 0·02 0·05 0·01 0·03 Yb 0·41 0·28 0·42 0·37 0·18 0·30 0·09 0·22 Lu 0·06 0·05 0·07 0·06 0·03 0·05 0·02 0·04 Hf 0·38 0·17 0·16 0·13 0·04 0·10 0·01 0·08 Pb 0·20 0·15 0·12 0·06 0·48 0·31 0·08 0·21 Th 0·870 0·004 0·001 0·006 b.d.l. <0·001 <0·001 0·003 U 0·235 0·013 0·001 0·002 0·036 0·001 0·001 0·003 V 54·51 58·82 62·05 86·95 54·83 65·81 46·26 42·80 Cr 2011 2006 1876 n.d. 3631 2479 2057 2448 Co 82·64 98·92 74·75 42·52 119·04 102·48 89·72 127·00 Ni 1601 1938 1880 n.d. 2557 2075 1696 2635 Cu 15·81 13·53 12·62 29·82 18·59 5·15 7·02 13·38 Zn 38·37 38·50 56·98 46·29 39·95 43·24 44·97 48·41 B 26·78 58·47 40·04 1·20 25·79 15·72 14·24 19·96 As 1·14 1·05 0·83 0·84 0·99 1·53 0·94 0·89 Sb 0·03 0·03 0·04 0·16 0·17 0·31 0·03 0·03 n.d., non determined; b.d.l., below detection limit. Table 2 Samples volatile (S and C), Fe and Zn isotopic compositions Sample name Lithology Minerals S (ppm) 2sd C (ppm) 2sd n δ56Fe 2sd δ57Fe 2sd n δ66Zn 2sd n Queyras complex (Western Alps) CP2 Di-serp. Atg, Di, Ca ±Chl 32 – 1307 – 1 −0·12 0·04 −0·19 0·07 3 0·19 0·03 2 CP3 Serp. Boudin Rim Atg, Ca 1178 31 507 9 2 −0·05 0·03 −0·09 0·10 3 – – CP8 Serp. Boudin Core Liz ±Chl b·d·l· – 172 – 1 0·01 0·01 0·00 0·02 3 0·22 0·01 4 CE14b Di-serp. Di, Chl b·d·l· – 240 – 1 −0·27 0·07 −0·40 0·21 3 −0·56 0·02 4 CE10 Serp. Boudin Rim Atg, Chl, Ca 663 52 191 12 2 −0·03 0·11 −0·08 0·21 3 0·22 0·01 2 CE3 Serp. Boudin Core Atg, Liz ±Di b·d·l· – 372 101 2 −0·01 0·06 0·00 0·09 3 0·17 0·01 4 RV3 Serp. Boudin Rim Atg, Ol2 390 156 219 27 2 −0·17 0·06 −0·26 0·14 6 0·23 0·02 4 RV8 Serp. Boudin Core Atg, Liz 768 173 354 20 2 −0·07 0·06 −0·11 0·08 6 0·21 0·02 4 Kohistan (Himalaya) OG Gem Olivine – – – – – −0·22 0·08 −0·33 0·07 3 0·79* 0·04 – Ogi Gem Olivine – – – – – −0·36 0·08 −0·54 0·14 3 0·57* 0·05 – OGM Gem Olivine – – – – – −0·32 0·04 −0·50 0·16 3 0·89* 0·05 – OG1 Gem Olivine – – – – – −0·06 0·12 −0·13 0·13 3 0·66* 0·06 – Sample name Lithology Minerals S (ppm) 2sd C (ppm) 2sd n δ56Fe 2sd δ57Fe 2sd n δ66Zn 2sd n Queyras complex (Western Alps) CP2 Di-serp. Atg, Di, Ca ±Chl 32 – 1307 – 1 −0·12 0·04 −0·19 0·07 3 0·19 0·03 2 CP3 Serp. Boudin Rim Atg, Ca 1178 31 507 9 2 −0·05 0·03 −0·09 0·10 3 – – CP8 Serp. Boudin Core Liz ±Chl b·d·l· – 172 – 1 0·01 0·01 0·00 0·02 3 0·22 0·01 4 CE14b Di-serp. Di, Chl b·d·l· – 240 – 1 −0·27 0·07 −0·40 0·21 3 −0·56 0·02 4 CE10 Serp. Boudin Rim Atg, Chl, Ca 663 52 191 12 2 −0·03 0·11 −0·08 0·21 3 0·22 0·01 2 CE3 Serp. Boudin Core Atg, Liz ±Di b·d·l· – 372 101 2 −0·01 0·06 0·00 0·09 3 0·17 0·01 4 RV3 Serp. Boudin Rim Atg, Ol2 390 156 219 27 2 −0·17 0·06 −0·26 0·14 6 0·23 0·02 4 RV8 Serp. Boudin Core Atg, Liz 768 173 354 20 2 −0·07 0·06 −0·11 0·08 6 0·21 0·02 4 Kohistan (Himalaya) OG Gem Olivine – – – – – −0·22 0·08 −0·33 0·07 3 0·79* 0·04 – Ogi Gem Olivine – – – – – −0·36 0·08 −0·54 0·14 3 0·57* 0·05 – OGM Gem Olivine – – – – – −0·32 0·04 −0·50 0·16 3 0·89* 0·05 – OG1 Gem Olivine – – – – – −0·06 0·12 −0·13 0·13 3 0·66* 0·06 – n, number of analyses; sd, standard deviation. * values from Pons et al. (2016). Table 2 Samples volatile (S and C), Fe and Zn isotopic compositions Sample name Lithology Minerals S (ppm) 2sd C (ppm) 2sd n δ56Fe 2sd δ57Fe 2sd n δ66Zn 2sd n Queyras complex (Western Alps) CP2 Di-serp. Atg, Di, Ca ±Chl 32 – 1307 – 1 −0·12 0·04 −0·19 0·07 3 0·19 0·03 2 CP3 Serp. Boudin Rim Atg, Ca 1178 31 507 9 2 −0·05 0·03 −0·09 0·10 3 – – CP8 Serp. Boudin Core Liz ±Chl b·d·l· – 172 – 1 0·01 0·01 0·00 0·02 3 0·22 0·01 4 CE14b Di-serp. Di, Chl b·d·l· – 240 – 1 −0·27 0·07 −0·40 0·21 3 −0·56 0·02 4 CE10 Serp. Boudin Rim Atg, Chl, Ca 663 52 191 12 2 −0·03 0·11 −0·08 0·21 3 0·22 0·01 2 CE3 Serp. Boudin Core Atg, Liz ±Di b·d·l· – 372 101 2 −0·01 0·06 0·00 0·09 3 0·17 0·01 4 RV3 Serp. Boudin Rim Atg, Ol2 390 156 219 27 2 −0·17 0·06 −0·26 0·14 6 0·23 0·02 4 RV8 Serp. Boudin Core Atg, Liz 768 173 354 20 2 −0·07 0·06 −0·11 0·08 6 0·21 0·02 4 Kohistan (Himalaya) OG Gem Olivine – – – – – −0·22 0·08 −0·33 0·07 3 0·79* 0·04 – Ogi Gem Olivine – – – – – −0·36 0·08 −0·54 0·14 3 0·57* 0·05 – OGM Gem Olivine – – – – – −0·32 0·04 −0·50 0·16 3 0·89* 0·05 – OG1 Gem Olivine – – – – – −0·06 0·12 −0·13 0·13 3 0·66* 0·06 – Sample name Lithology Minerals S (ppm) 2sd C (ppm) 2sd n δ56Fe 2sd δ57Fe 2sd n δ66Zn 2sd n Queyras complex (Western Alps) CP2 Di-serp. Atg, Di, Ca ±Chl 32 – 1307 – 1 −0·12 0·04 −0·19 0·07 3 0·19 0·03 2 CP3 Serp. Boudin Rim Atg, Ca 1178 31 507 9 2 −0·05 0·03 −0·09 0·10 3 – – CP8 Serp. Boudin Core Liz ±Chl b·d·l· – 172 – 1 0·01 0·01 0·00 0·02 3 0·22 0·01 4 CE14b Di-serp. Di, Chl b·d·l· – 240 – 1 −0·27 0·07 −0·40 0·21 3 −0·56 0·02 4 CE10 Serp. Boudin Rim Atg, Chl, Ca 663 52 191 12 2 −0·03 0·11 −0·08 0·21 3 0·22 0·01 2 CE3 Serp. Boudin Core Atg, Liz ±Di b·d·l· – 372 101 2 −0·01 0·06 0·00 0·09 3 0·17 0·01 4 RV3 Serp. Boudin Rim Atg, Ol2 390 156 219 27 2 −0·17 0·06 −0·26 0·14 6 0·23 0·02 4 RV8 Serp. Boudin Core Atg, Liz 768 173 354 20 2 −0·07 0·06 −0·11 0·08 6 0·21 0·02 4 Kohistan (Himalaya) OG Gem Olivine – – – – – −0·22 0·08 −0·33 0·07 3 0·79* 0·04 – Ogi Gem Olivine – – – – – −0·36 0·08 −0·54 0·14 3 0·57* 0·05 – OGM Gem Olivine – – – – – −0·32 0·04 −0·50 0·16 3 0·89* 0·05 – OG1 Gem Olivine – – – – – −0·06 0·12 −0·13 0·13 3 0·66* 0·06 – n, number of analyses; sd, standard deviation. * values from Pons et al. (2016). Iron and zinc isotope analyses were performed on the Queyras samples by multiple-collector inductively coupled plasma mass spectrometry (MC-ICP-MS; Thermo Neptune Plus) at Durham University. Kohistan samples were analysed for Fe isotopes only; their Zn isotopic compositions are provided in Pons et al. (2016). Dissolution, iron or zinc purification and isotopic analyses were undertaken at Durham University using procedures established by Hibbert et al. (2012) and Pons et al. (2016) in the same laboratory for iron and zinc, respectively. Instrumental mass fractionation was corrected by sample-standard bracketing for Fe and external normalization to Cu, coupled with sample-standard bracketing for Zn. In the case of Fe isotope analyses, analysed solutions consisted of 2 ppm natural Fe in 0·1 M HNO3. The standard Fe beam intensities in medium resolution mode typically varied between 25 and 35 V 56Fe for a standard 10-11Ω resistor. Mass dependence, long-term reproducibility and accuracy were evaluated by analysis of an in-house FeCl salt standard (δ56Fe = -0·70 ± 0·05‰; δ57Fe = -1·03 ± 0·09‰ 2sd, n = 24; sd: standard deviation, n: number of analyses) previously analysed in other studies (Williams et al., 2005; Williams & Bizimis, 2014; Weyer & Ionov, 2007; Millet et al., 2012). The USGS standard BIR-1 (Icelandic basalt) was also analysed over the course of this study (δ56Fe = 0·06 ± 0·04‰; δ57Fe = 0·07 ± 0·04‰; 2sd, n = 2). The analyses of BIR are in good agreement with previous studies (e.g. Millet et al., 2012; Sossi et al., 2015; Debret et al., 2016b). The δ56Fe values are reported in Table 2. The total procedural blank contribution was <10 ng of Fe, which is negligible compared to the amount of Fe processed (> 300 μg). In the case of Zn isotope measurements, samples were run at 750 ppb Zn in low resolution mode, for a sensitivity of 10–15 V per ppm of Zn. The long-term reproducibility evaluated by analysis of an in-house Zn standard (Alpha Aesar pure Zn solution, n = 45) is 0·035‰ (2sd). The total external reproducibility of the chemical and analytical procedure on δ66Zn based on repeated analysis of an international rock standard (BCR-2, n = 9) is 0·06‰ (2sd). Our value for BCR-2 is in perfect agreement with previously published studies (Moeller et al., 2012 and references therein). The total procedural blank is <20 ng of Zn, which is negligible compared to the >2 μg of sample Zn processed. Pseudosection modelling In order to assess the role of carbon activity (aCO2) during Di-serpentinite formation, we have constructed T-X(CO2) (Supplementary Data Electronic Appendix C) and T-aCO2 pseudosections for sample CP2 (Fig. 7) in the CFMASH–CO2 (CaO–FeO–MgO–Al2O3–SiO2–H2O–CO2) system using a Gibbs free energy minimization strategy (Perplex; Connolly, 2005). For silicates, we used the thermodynamic database of Holland and Powell (1998, revised in 2003), and their solid solution models for olivine, pyroxenes, talc, brucite and amphibole. Talc and brucite were considered as ideal solid solutions, and amphiboles were modelled between tremolite and Fe-tremolite. The chlorite solid solution model was taken from Lanari et al. (2014), and the antigorite model from Padrón-Navarta et al. (2013). For carbonates we used the solid solution model developed by Franzolin et al. (2011) with an enthalpy of formation of -5·74kj/mol for ordered dolomite (Navrotsky & Capobianco, 1987), which is considered to be in good accordance with natural examples (Scambelluri et al., 2016). Fig. 7 View largeDownload slide T-aCO2 pseudosection at 10 kbar and a water activity of 1 for the CP2 sample. The yellow star shows the inferred conditions of crystallization of the chlorite–antigorite–diopside–Ca-carbonate in the LT-blueschist unit (320–360°C and 9–11 kbar) and white star that of the MT-blueschist unit (340–390°C and 10–12 kbar). Diopside coexists with carbonates serpentine and chlorite over a large range of P–T conditions. The progressive disappearance of Ca-carbonate in Queyras metamorphic terranes is consistent with a decrease of aCO2 in the fluid at low temperature (320 < T < 390°C). The expected T-aCO2 path of Queyras metasomatic contacts is represented by a grey arrow. Atg, antigorite; Cpx, clinopyroxene (diospide); Tr, Amphibole (tremolite); Tlc, Talc; Qz, Quartz; Cb, carbonates; Ol,Olivine; Brc, Brucite; Chl, Chlorite. Fig. 7 View largeDownload slide T-aCO2 pseudosection at 10 kbar and a water activity of 1 for the CP2 sample. The yellow star shows the inferred conditions of crystallization of the chlorite–antigorite–diopside–Ca-carbonate in the LT-blueschist unit (320–360°C and 9–11 kbar) and white star that of the MT-blueschist unit (340–390°C and 10–12 kbar). Diopside coexists with carbonates serpentine and chlorite over a large range of P–T conditions. The progressive disappearance of Ca-carbonate in Queyras metamorphic terranes is consistent with a decrease of aCO2 in the fluid at low temperature (320 < T < 390°C). The expected T-aCO2 path of Queyras metasomatic contacts is represented by a grey arrow. Atg, antigorite; Cpx, clinopyroxene (diospide); Tr, Amphibole (tremolite); Tlc, Talc; Qz, Quartz; Cb, carbonates; Ol,Olivine; Brc, Brucite; Chl, Chlorite. RESULTS Queyras meta-ophiolites Metasomatic interfaces The metasomatic interfaces are primarily composed of Ca-carbonate, diopside, magnetite, antigorite and/or chlorite (Di-serpentinite). Samples CP2 (from the LT-blueschist, Col Peas) and CE14b (from the MT-blueschist, Echassier) are Di-serpentinites. In both samples, diopside lamellae and needles have low Al2O3 (<0·1–2·3 wt %) and Cr2O3 (< 0·1 wt %) contents relative to those expected for mantle clinopyroxene (Al2O3 > 3 wt %, Cr2O3 > 1 wt %, e.g. Debret et al., 2013b). Antigorite Al2O3 and FeO contents range from 1·5 to 3·8 wt % and from 4·3 to 6·3 wt %, respectively. Chlorite compositions range from Cr-clinochlore to chamosite end-members (Al2O3 = 10·8–17·7 wt %, FeO = 3·5–15·7 wt % and Cr2O3 = < 0·1–8·2 wt %). Importantly, the difference between these two samples is marked by the absence of Ca-carbonates in CE14b, otherwise present in CP2. The bulk-rock compositions of Di-serpentinites is characterized by low MgO (26·6–30·7 wt %) and high Al2O3 (3·7–12·6 wt %) and CaO (6–6·6 wt %) contents relative to abyssal serpentinites (Fig. 8). The Di-serpentinites display a large range in C (240–1307 ppm) and have low S (< 32 ppm) contents. Normalized to the primitive mantle, their trace element patterns are relatively flat (Cen/Ybn = 0·6–2) with an alkali enrichment (Csn/Cen = 50–470) and positive Li, As and Sb anomalies (Fig. 9a). The Di-serpentinites display high fluid-mobile element (FME: B, Li, As, Sb, Cs and Sr) concentrations and highly fractionated δ56Fe values, ranging from -0·27 ± 0·07‰ (2sd, n = 3) to -0·12 ± 0·04‰ (2sd, n = 3), relative to those of seafloor serpentinites, which have δ56Fe values that typically range from -0·05 to + 0·05‰ (Craddock et al., 2013; Debret et al., 2016b; Fig. 8). The CE14b sample (δ66Zn = -0·56 ± 0·02‰, 2sd, n = 4) is characterized by an extremely light δ66Zn value relative to that of the CP2 sample (δ66Zn = 0·19 ± 0·03‰, 2sd, n = 2) and seafloor serpentinites (+ 0·12 to + 0·28‰; Pons et al., 2011; Fig. 8). Fig. 8 View largeDownload slide Variations in (a) CaO (wt %), (b) C (wt %) concentrations, (c) Li/Y, (d) Sr/Y, (e) δ56Fe and (f) δ66Zn values of ultramafic rocks formed at the contact with the metasedimentary rocks (this study and Lafay et al., 2013). The C and CaO concentrations and Li/Y and Sr/Y ratios progressively increase from serpentine boudin cores to metasomatic contacts in the LT-blueschist unit. These variations are accompanied by a decrease in δ56Fe. In MT- and HT-blueschist units, similar trends are observed for CaO, Li/Y, Sr/Y and δ56Fe values, whereas the studied samples display relatively low C contents. The δ66Zn values of ultramafic rocks are constant in the different studied meta-ophiolites, with one carbonate-free Di-serpentinite displaying a light δ66Zn value. The grey field corresponds to the mean value of abyssal serpentinites for C (Kelemen & Manning, 2015), CaO (Deschamps et al., 2013), δ56Fe (Craddock et al., 2013) and δ66Zn (Pons et al., 2016) and the black field to sediment δ56Fe values (Inglis et al., 2017). Fig. 8 View largeDownload slide Variations in (a) CaO (wt %), (b) C (wt %) concentrations, (c) Li/Y, (d) Sr/Y, (e) δ56Fe and (f) δ66Zn values of ultramafic rocks formed at the contact with the metasedimentary rocks (this study and Lafay et al., 2013). The C and CaO concentrations and Li/Y and Sr/Y ratios progressively increase from serpentine boudin cores to metasomatic contacts in the LT-blueschist unit. These variations are accompanied by a decrease in δ56Fe. In MT- and HT-blueschist units, similar trends are observed for CaO, Li/Y, Sr/Y and δ56Fe values, whereas the studied samples display relatively low C contents. The δ66Zn values of ultramafic rocks are constant in the different studied meta-ophiolites, with one carbonate-free Di-serpentinite displaying a light δ66Zn value. The grey field corresponds to the mean value of abyssal serpentinites for C (Kelemen & Manning, 2015), CaO (Deschamps et al., 2013), δ56Fe (Craddock et al., 2013) and δ66Zn (Pons et al., 2016) and the black field to sediment δ56Fe values (Inglis et al., 2017). Fig. 9 View largeDownload slide Primitive mantle-normalized whole-rock trace element patterns of the different lithologies composing the Queyras meta-ophiolites; normalizing values are from Sun & McDonough (1989). The grey field corresponds to the seafloor lizardite-bearing serpentinites from the Chenaillet Massif (Debret et al., 2016b). Symbol free lines for comparisons from Lafay et al. (2013). Fig. 9 View largeDownload slide Primitive mantle-normalized whole-rock trace element patterns of the different lithologies composing the Queyras meta-ophiolites; normalizing values are from Sun & McDonough (1989). The grey field corresponds to the seafloor lizardite-bearing serpentinites from the Chenaillet Massif (Debret et al., 2016b). Symbol free lines for comparisons from Lafay et al. (2013). Serpentinites boudin cores and rims The serpentinite boudin rims and cores are mainly composed of lizardite and antigorite with minor amounts of chlorite and secondary olivine (Fig. 5). Although serpentine-bearing assemblages display a large range in major element compositions (Al2O3 = 0·5–5·2 wt %; FeO = 1·7–10·5 wt %), that can be attributed to various substitution processes (see for example Beard & Frost, 2016), no systematic variations are observed between the different studied areas. Chlorites display a restricted range of compositions and correspond to clinochlore (Al2O3 = 13·2–13·6 wt %, FeO = 4·2–7·3 wt % and Cr2O3 < 0·2 wt %). Secondary olivines from HT-blueschist terrains have high MnO contents (1·4–1·5 wt %) and XMg (Mg /(Mg + Fe) ∼ 92) relative to those expected from their mantle protoliths (MnO < 0·5 wt %; XMg = 0·88–0·92; e.g. Debret et al., 2013b). Such observations are in agreement with previous studies of secondary olivines generated as a consequence of antigorite and brucite breakdown reactions (e.g. Almirez massif, Spain, Trommsdorff et al., 1998; Padrón-Navarta et al., 2011). The major and trace element bulk-rock compositions of the studied serpentinites are consistent with previous studies of Queyras serpentinites (e.g. Lafay et al., 2013) and abyssal peridotites/serpentinites (e.g. Deschamps et al., 2013). They display harzburgitic to depleted lherzolitic-like compositions (MgO = 35–39·3 wt %, Al2O3 = 1·4–3 wt % and CaO = < 0·1–1·9 wt %) and high C (170–510 ppm) and variable S (< 50–1180 ppm) contents compared to mantle peridotites (C = 30 ppm; and S = 100–250 ppm;Dasgupta & Hirschmann, 2010; Alt et al., 2013). However, the carbon contents of the Queyras serpentinite boudins are low relative to the Di-serpentinites (Fig. 8). In the LT- and MT-units, the trace element compositions of serpentinites are characterized by depletion in LREE (Light Rare Earth Elements; Cen/Ybn = 0·13–0·25) and a slight increase from MREE (middle rare earth elements) to HREE (heavy rare earth elements; Gdn/Ybn = 0·71–0·87) as well as positive U (Un/Thn = 3–14) and Pb (Pbn/Cen = 3–28) anomalies and strong alkali enrichment relative to LREE (Csn/Cen = 50– 470, Fig. 9). In the HT-unit, the trace element patterns of serpentinites are flat (Cen/Ybn = 1·1–1·4) and characterized by low alkali concentrations (Csn/Cen < 0·1) and negative anomalies in Sr (Srn/Ndn < 0·1) and Eu (Eu* = 0·4–0·5; Fig. 9). The FME (fluid-mobile element) concentrations significantly decrease in serpentinites passing from LT- or MT- blueschist units to HT-blueschist unit (Figs 8 and 10). Fig. 10 View largeDownload slide Comparisons of ultramafic rocks compositions in the Western Alps and Himalayan (meta)ophiolites. (a) Plots of Li* (= Lin / [(Dyn + Yn) / 2]) vs Cs/Y. (b) As vs Sb. Data from the Queyras meta-ophiolites are in large symbols (bordered and coloured symbols: this study; non-bordered coloured symbols: Lafay et al., 2013). Queyras serpentinites display high FME concentrations relative to other Western Alps serpentinites. Their composition is close to those reported from the Himalayan mantle wedge. Data for other Western Alps meta-ophiolites (Chenaillet, Monte Maggiore, Lanzo, Monviso) are from Lafay et al. (2013) and Debret et al. (2013a, 2016b). Himalayan data are from Bouilhol et al. (2009), Hattori & Guillot (2007) and Deschamps et al. (2013). Fig. 10 View largeDownload slide Comparisons of ultramafic rocks compositions in the Western Alps and Himalayan (meta)ophiolites. (a) Plots of Li* (= Lin / [(Dyn + Yn) / 2]) vs Cs/Y. (b) As vs Sb. Data from the Queyras meta-ophiolites are in large symbols (bordered and coloured symbols: this study; non-bordered coloured symbols: Lafay et al., 2013). Queyras serpentinites display high FME concentrations relative to other Western Alps serpentinites. Their composition is close to those reported from the Himalayan mantle wedge. Data for other Western Alps meta-ophiolites (Chenaillet, Monte Maggiore, Lanzo, Monviso) are from Lafay et al. (2013) and Debret et al. (2013a, 2016b). Himalayan data are from Bouilhol et al. (2009), Hattori & Guillot (2007) and Deschamps et al. (2013). In LT- and MT-blueschist units, serpentinite δ56Fe values decrease from ultramafic boudin cores to rims. The δ56Fe values of the boudin cores range from -0·01 ± 0·06‰ (2sd, n = 3) to +0·01 ± 0·06‰ (2sd, n = 3) and are similar to Fe isotope data reported for abyssal peridotites and serpentinites (δ56Fe = +0·05 to -0·05‰, Craddock et al., 2013, non-weathered rocks, Fig. 8). The δ56Fe values of boudin rims range from -0·03 ± 0·11‰ (2sd, n = 3) to -0·05 ± 0·03‰ (2sd, n = 3). Such values are lower than those reported for antigorite schists in other Western Alps meta-ophiolites (δ56Fe = +0·14 to -0·01‰, Debret et al., 2016b, Atg/Liz- and Atg-serpentinites). In the HT-unit, the serpentinites δ56Fe values are shifted to lower values and range from -0·07 ± 0·06‰ (2sd, n = 6) for boudin cores to -0·17 ± 0·06‰ (2sd, n = 6) at boudin rims (Fig. 8). In all the studied meta-ophiolites, the serpentinites display homogeneous δ66Zn values ranging from 0·17 ± 0·01‰ (2sd, n = 4) to 0·23 ± 0·02‰ (2sd, n = 4). These values are in good agreement with previous analyses of Western Alps serpentinites for which δ66Zn values range from 0·12 ± 0·04‰ to 0·26 ± 0·04‰ (Pons et al., 2016, Atg/Liz- and Atg-serpentinites). Kohistan olivines The major and trace elements concentrations and the Zn isotopic compositions of the Kohistan gem olivines are from Bouilhol et al. (2009, 2012) and Pons et al. (2016), respectively; their Fe isotope compositions were obtained as part of this study. The Kohistan gem olivines show a wide range of XMg (0·89–0·97), MnO (0·1–0·25 wt %) and NiO (0·22–0·88 wt %) contents. Compared to mantle olivine, their trace element patterns display highly fractionated HREE slopes and high concentrations of HREE and High Field Strength Elements (HFSE; Nd, Ta, Zr, Hf, Ti; Bouilhol et al., 2009, 2012). Such features are commonly observed in metamorphic olivines crystallizing in equilibrium with metamorphic fluids (e.g. Garrido et al., 2005; Debret et al., 2013a) and should be considered as characteristic of olivines produced as a consequence of fluid-rock reactions. Moreover, concentrations of Li (2·1–11 ppm), B (19–121 ppm) and chalcophile elements (Cu = 0·8–207 ppm; Zn = 17–152 ppm) in the Kohistan olivines exceed those of typical mantle olivines (mantle olivine: Li < 3 ppm; B < 1 ppm; Cu < 3 ppm, Zn = 28–81 ppm; e.g. Vils et al., 2008; De Hoog et al., 2010). The Kohistan gem olivines display highly fractionated δ56Fe and δ66Zn values ranging from -0·06 ± 0·12‰ (2sd, n = 3) to -0·36 ± 0·08‰ (2sd, n = 3) and from 0·57 ± 0·05‰ to 0·89 ± 0·05‰ (Table 2), respectively. DISCUSSION (De)carbonation reactions and carbon leaching during the early stages of subduction In each Queyras tectono-metamorphic unit, the ultramafic boudin cores preserve low temperature assemblages, such as lizardite-bearing mesh and bastite textures (Figs 2e, 3c and 5), whereas the boudin rims and metasomatic interfaces display higher temperature and pressure (HT–HP) assemblages, such as antigorite and secondary olivine (Fig. 4). The preferential crystallization of HT–HP assemblages at the boudin rims and metasomatic interfaces suggests the existence of a free fluid phase that could have accelerated (de)serpentinization and enhanced (de)carbonation reactions in these zones during subduction. In support of this hypothesis, previous experimental studies have shown that the presence of Si, Al, Ca or CO2 in fluids can increase the kinetics of serpentinization reactions (e.g. Martin & Fyfe, 1970; Andreani et al., 2013; Pens et al., 2016). In this scenario, the contact between metasedimentary and ultramafic rocks may correspond to a reaction front where fluids promote mass transfer and affect reaction kinetics and local equilibriums during subduction. In the LT-blueschist unit, the interface between metasedimentary and ultramafic rocks is mainly composed of metamorphic diopside, Ca-carbonate, antigorite, chlorite and magnetite (Di-serpentinites, Fig. 5b) reflecting a two-stage crystallization process. The Di-serpentinite foliation is marked by orientated lamellae of antigorite (±chlorite) whereas diopside needles are randomly orientated (Fig. 2c), showing that diopside crystallization post-dates deformation and antigorite crystallization. The presence of Ca-carbonate clusters preserved within diopside layers (Fig. 2d) suggests an early paragenesis composed of Ca-carbonate and antigorite. The crystallization of both Ca-carbonate and antigorite required the percolation of Ca, C- and SiO2-rich fluids, triggering the formation of carbonate and antigorite at the expense of lizardite (e.g. Evans, 2004). The early Ca-carbonate crystallization is also evidenced by the high CaO (6·6 wt %) and C (1307 ppm) concentrations of Di-serpentinite relative to the rims of the ultramafic boudins (Ca = 1·9 wt %; C = 507 ppm), cores (Ca = 0·1 wt %; C = 172 ppm) and abyssal serpentinites (Fig. 8). In the MT-blueschist unit, the metasomatic contacts show evidence of extensive decarbonation. They are carbonate free and exclusively composed of diopside, chlorite and magnetite (Figs 3a, b and 5c). Previous P–T estimates have shown that the MT-blueschist unit records a peak around 340–390°C and 10–12 kbar (Schwartz et al., 2013). At these conditions antigorite crystallizes at the expense of lizardite in serpentinites (Fig. 3c) but carbonate remains stable (Fig. 7). The absence of carbonate in these rocks cannot be explained by a simple increase in P–T conditions and alternative processes must be considered. Fluid circulation in metasomatic fronts can promote chemical exchange and change local mineral equilibrium (e.g. Malaspina et al., 2009; Tumiati et al., 2015). A change of fluid composition in these zones, potentially driven by the dehydration of other slab lithologies during subduction, will affect local mineral equilibrium and may lead to carbonate leaching in fluids. Thermodynamic calculations show that dolomite, quartz and Ca-carbonate (aragonite or calcite) are progressively replaced by talc, amphibole and finally diopside at decreasing aCO2 (Supplementary Data Electronic Appendix C). Diopside crystallization and full carbonate dissolution is achieved in an ultramafic fluid-saturated system at very low aCO2. (Fig. 7). The observed absence of carbonate may, therefore, reflect the percolation of a fluid with extremely low aCO2, resulting in extensive carbon leaching and diopside crystallization. A low aCO2 could be achieved either by a change of fluid source or extensive depletion of carbon in the fluid source prior to MT- or HT-blueschist facies metamorphism. In agreement with these scenarios, the Di-serpentinites forming the MT-metasomatic contacts preserve high CaO contents (between 6 to 6·6 wt %, Fig. 8), far greater than that observed in peridotites and serpentinites in orogenic or abyssal contexts (between 0 and 4 wt %, e.g. Godard et al., 2008; Bodinier & Godard, 2013), suggesting that they experienced percolation by carbonate-bearing fluids, which would have enhanced carbonate crystallization. However, these samples also display low C contents relative to LT-metasomatic contacts, implying subsequent leaching of carbon during diopside crystallization (Fig. 8). The Queyras Schistes Lustrés complex is considered to be an open system in which fluids released during sediment devolatilization have infiltrated the surrounding rocks during the early stages of subduction (Lafay et al., 2013; Schwartz et al., 2013; Cook-Kollars et al., 2014; Debret et al., 2016a), promoting the formation of metasomatic contacts. Several studies have demonstrated the mobility of many elements, including Si, Cs, B, Li, As, Sb, Ba, Rb, Ce, S and Cu, in metasediment-derived fluids (Busigny et al., 2003; Bebout et al., 2007, 2013; Garofalo, 2012; Penniston-Dorland et al., 2012; Lafay et al., 2013; Cannaò et al., 2016; Debret et al., 2016a; Peters et al., 2017). In good agreement with these studies, serpentinites from LT- and MT-blueschist units display high Li, As, Sb, Ba, Rb and Cs concentrations relative to slab serpentinites from other Western Alps meta-ophiolites (Fig. 10), indicating significant metasomatism by sediment-derived fluids during the early stages of subduction (e.g. Peters et al., 2017). Recent studies have shown that the thermal decomposition of carbonate (i.e. siderite) into CO2-rich fluids and magnetite occurs in aqueous fluids at relatively low temperature, ranging from 200 to 300°C (Milesi et al., 2015). This suggests that the Queyras metasedimentary rocks experienced an early decarbonation stage, before the slab reaches the 300°C isotherm, generating a fluid with a significant amount of dissolved CO2. The infiltration of these fluids into the ultramafic lithologies will subsequently promote carbonate precipitation and the storage of fluid-mobile elements and carbon in antigorite-bearing serpentinites. The ultramafic rocks from the HT-blueschist unit are characterized by the presence of secondary olivine, indicative of serpentinite devolatilization, which, under these P–T conditions, would be related to brucite breakdown. This process leads to a significant release of FME-rich fluids, as testified by the low FME concentrations of the ultramafic rocks forming the HT-blueschist meta-ophiolites (Fig. 10), and is likely to be associated with a significant release of aqueous fluids (about 2 wt % of H2O lost, Padrón-Navarta et al., 2013). These fluids would be virtually CO2-free, and their percolation would thus cause a decrease in aCO2 (Fig. 7). This would promote the leaching of the previously crystallized carbonates and the re-crystallization of the metasomatic contact into a carbonate-free zone, which would in turn lead to the substantial transfer of CO2-bearing fluids from the slab to the slab/mantle wedge interface and/or the sub-arc mantle (see also Kelemen & Manning, 2015; Scambelluri et al., 2016). It however remains unclear whether the breakdown of brucite alone would produce enough fluid to leach all the carbon. Indeed, other sources of fluid (e.g. amphibole breakdown in mafic lithologies; e.g. Debret et al., 2016a) and reaction pathways (e.g. stabilization of graphite in metasedimentary rocks; e.g. Galvez et al., 2013) may also lead to a decrease of aCO2 in fluids. Tracking fluid migration in the subducting slab using stable isotopes Recent studies have successfully used Fe and Zn stable isotopes to track the composition and redox state of fluids circulating in subduction zones (Debret et al.,2016b; Pons et al., 2016) as ab initio calculations predict that these isotopic systems are particularly sensitive to H–C–O–S–Cl fluids (Hill & Schauble, 2008; Fujii et al., 2010, 2011, 2014; Hill et al., 2010). In the Queyras Schistes Lustrés Complex, Fe and Zn display contrasting behaviour. In each Queyras tectono-metamorphic unit, δ56Fe progressively decreases from boudin cores to boudin rims and Di-serpentinites (Fig. 8). In contrast, the δ66Zn values of the Queyras ultramafic rocks are homogeneous and overlap with those of previous studies of Western Alps or mid-oceanic ridge serpentinites (Pons et al., 2016). Only a carbonate free Di-serpentinite from the MT-blueschist unit displays an isotopically light δ66Zn value (Fig. 8). Previous studies have shown that Fe behaves conservatively during ocean floor serpentinization and that no net Fe stable isotope fractionation takes place during this process (Craddock et al., 2013; Debret et al., 2016b). The range of δ56Fe values at mid-oceanic ridges is thus interpreted as reflecting preexisting protolith Fe isotope heterogeneity, with pyroxene-rich and less depleted peridotites displaying isotopically heavier compositions (Williams et al., 2005; Williams & Bizimis, 2014). In Alpine ophiolites, serpentinite protolith fertility can be assessed using fluid-immobile elements whose ratios are unaffected by serpentinization (e.g. Al2O3/SiO2, Zr/Nb; see Debret et al., 2016b), with dunites displaying relatively low Al2O3/SiO2 and light δ56Fe values and lherzolites displaying high Al2O3/SiO2 and heavy δ56Fe values (see Supplementary Data Electronic Appendix C). Although the δ56Fe values of the ultramafic boudin cores overlap with those of abyssal peridotites and display a broadly similar trend with regards to Al2O3/SiO2 (Supplementary Data Electronic Appendix D), the δ56Fe values of the boudin rims and metasomatic interfaces do not correlate with indices of peridotite fertility. In fact, the metasomatic contacts display the highest Al2O3/SiO2 ratios and the lightest δ56Fe values (from -0·12 to -0·27‰). Critically, these observations demonstrate that the isotopically light δ56Fe values of Queyras ultramafic rocks (boudin rims and metasomatic interfaces) are not controlled by preexisting protolith heterogeneities (e.g. dunite vs lherzolite) and must have been generated by the mobilization of Fe during prograde subduction metamorphism. Previous Fe isotope studies of subducted serpentinites samples from Western Alps meta-ophiolites have shown that blueschist (+ 0·07 ± 0·07‰, 2σ) and eclogite (+ 0·08 ± 0·11‰) facies slab serpentinites display heavy Fe isotope signatures relative to those of abyssal (+ 0·01 ± 0·08‰) and low metamorphic grade serpentinites (-0·02 ± 0·08‰; Debret et al., 2016b). This has been interpreted to reflect the progressive loss of isotopically light Fe from the slab with increasing prograde metamorphism. This scenario is consistent with the release of sulfate-rich and/ or hypersaline fluids, which preferentially complex isotopically light Fe in the form of Fe(II)–SOX or Fe(II)–Cl2 species. The release of fluids in which Zn was complexed by sulfate during serpentinite devolatilization has also been confirmed by a recent Zn isotope study of the same serpentinite samples (Pons et al., 2016). Therefore, the isotopically light (δ56Fe = - 0·27 ± 0·01) compositions of the Queyras ultramafic rocks cannot be explained by a simple scenario of serpentinite devolatilization, as this should be accompanied by the release of isotopically light Fe which would generate isotopically heavy residues, which is opposite to what is observed. An alternative process is, therefore, required. The observed Fe isotopic fractionation in the metasomatic interfaces could, therefore, result from: (1) the addition of light δ56Fe and/or (2) the leaching of a heavy δ56Fe component by external fluids. Critically, the influence of kinetic effects, related to the preferential mobility of isotopically light species during fluid-rock interaction or diffusion processes, can be ruled out. Indeed, the chemical potential gradient between solid phases and fluids is such that the direction of diffusive transport from solid (Zn- and Fe-rich) to fluid (Zn- and Fe-poor) would result in heavy δ66Zn and δ56Fe in the residue and, therefore, generate positive co-variations between δ66Zn and δ56Fe in the studied samples, which are not observed. The progressive decrease of δ56Fe values in the LT- and MT- units, from boudin cores to metasomatic interfaces, is correlated with an increase in whole-rock CaO content and other indices of metasediment-derived fluids, such as FME (Fig. 8). This observation provides evidence for overprinting by sediment-derived fluids. Here we discuss and model the transfer of isotopically light Fe during the overprinting of ultramafic rocks by sediment-derived fluids. Metasedimentary rocks in the Queyras display δ56Fe values of ∼ 0·1‰ (Inglis et al., 2017) and cannot be distinguished from abyssal deep-sea sediments which display a large range of δ56Fe values varying from about -0·74 to 0·15‰ (e.g. Rouxel et al., 2003). However, it should be noted that carbonated deep-sea sediments are expected to display extremely light δ56Fe values (Rouxel et al., 2003). In that sense, the relatively heavy δ56Fe values of Queyras metasedimentary rocks may, therefore, be compatible with the dissolution of Fe-bearing carbonates and the release of isotopically light Fe-bearing fluids, as the process would drive the residual metasediments to isotopically heavy values. Furthermore, siderite (FeCO3) and ankerite (Ca(Fe, Mg, Mn)(CO3)2) are the most common Fe-bearing carbonates crystallizing in seafloor sediments and can display highly fractionated δ56Fe values down to -3·9‰ (e.g. Belshaw et al., 2000; Johnson et al., 2008). These minerals are also particularly soluble at temperatures ranging from 200 to 300°C (Milesi et al., 2015) and their dissolution would contribute to the formation of C-bearing fluids with a light Fe isotope signature. Therefore, the correlation between indices of metasediment-derived fluids (e.g. CaO, Li/Y or Sr/Y) and the δ56Fe values of the surrounding lithologies (Fig. 8), as well as the observation of carbonate clusters within Di-serpentinites (Fig. 2d), strongly supports the isotopically light Fe signature being inherited from the metasediment-derived fluids. The percolation of such fluids through serpentinites would enhance the recrystallization of serpentinites with high carbon contents and isotopically light δ56Fe. This scenario is further corroborated by a simple geochemical model, in which the ultramafic rocks are metasomatized by the surrounding metasediments, as discussed below. The exact nature of Fe complexation in sediment-derived fluids remains poorly constrained. Nonetheless, natural (Debret et al., 2016b; Inglis et al., 2017) and theoretical (Schauble, 2004; Hill & Schauble, 2008, Hill et al., 2010; Fujii et al., 2014) studies have shown that Cl-, SO42- and CO32- anions will preferentially complex isotopically light Fe in fluids, thus being potential vectors of Fe transfer from sedimentary rocks to serpentinites. However, if CO2 and to a minor extent S are released in sediment derived-fluids during prograde metamorphism (e.g. Garofalo, 2012), Cl is likely to be retained in sedimentary rocks during devolatilization (Selverstone & Sharp, 2015 and reference therein). Here, we modelled the transfer of Fe from sedimentary rocks to serpentinites from 200 to 300°C during siderite dissolution using the ab initio calculations of Blanchard et al. (2009), Schauble (2004) and Fujii et al. (2014) compilation from Dauphas et al. (2017) for siderite and [FeIICl4]2-, FeIICl2(H2O)4, FeIICO3(H2O)4 and FeIISO4(H2O)5 complexes (Fig. 8; see Supplementary Data Electronic Appendices E and F for details). The fluid was modelled using a simple Rayleigh distillation model during the dissolution of siderite according to the following equations: 103ln(αfluid−siderite)=103ln(βfluid)–103ln(βSiderite) δ56Fefluid=(1000+δ56FeSiderite) x (F(α−1)−1)+δ56FeSiderite−103ln(αfluid−siderite) where α is the fractionation factor between siderite and a fluid phase complexing Fe as [FeIICl4]2-, FeIICl2(H2O)4, FeIICO3(H2O)4 or FeIISO4(H2O)5; F is the variable defined by the fraction of Fe remaining, ranging from 1 (unreacted) to 0 (all of Fe lost to the fluid phase), and was set to 0·1 to model an advanced stage of siderite dissolution. The δ56FeSiderite were taken from Belshaw et al. (2000) and Johnson et al. (2008). It should be noted that among the different models, FeIICO3(H2O)4 and FeIISO4(H2O)5 complexes generate fluids with the lightest isotopic compositions. The isotopic compositions of the metasomatic contacts were then assumed to be a binary mixture between this sediment-derived fluid and the serpentinites: δ56Femixture=(Naδ56Fea+ Nbδ56Feb)/(Na+ Nb) where N is the abundance of Fe in the fluid and in the serpentinite. The whole dataset of Queyras serpentinite δ56Fe values can be explained through the transfer of 0·1 to 0·4 wt % of Fe in fluids, whatever the considered model. This low Fe amount is consistent with the relatively constant whole-rock Fe concentrations (Fe2O3 varying from 7 to 9·3 wt %) and estimates of Fe solubility in aqueous fluids (e.g. Hermann et al., 2013). Zinc isotopes are particularly sensitive to carbonate ions in fluids, as theory predicts that carbonate, as well as sulfate, preferentially complexes isotopically heavy zinc in fluids relative to reduced complexes, such as Zn(HS)42- (Black et al., 2011; Fujii et al., 2011) and can thus induce isotopic fractionation during metamorphic processes (Pons et al., 2016). This is confirmed by the analysis of carbonated deep-sea sediments which display heavy values relative to mafic and ultramafic rocks (e.g. Pichat et al., 2003; Bentahila et al., 2008). However, surprisingly, the δ66Zn and Zn concentrations (39–57 ppm, Table 1) of the Queyras ultramafic rocks are relatively invariant in the different studied ophiolites (Fig. 8), irrespective of Ca or C concentrations. This observation suggests that the carbonation of Queyras ultramafic rocks by sediment-derived CO2-bearing fluids occurred without significant transfer of zinc. In contrast, at higher P–T conditions (MT-unit, T > 300°C), the decarbonation of ultramafic rocks seems to be accompanied by a significant release of isotopically heavy zinc as attested by the light δ66Zn value of a single carbonate-free Di-serpentinite. Although compositional and isotopic heterogeneity is inevitable in metasomatic zones, this result suggests that the transfer of heavy zinc in subduction zones is restricted to higher temperatures (T > 300°C) during the main stage of serpentinite devolatilization which, in terms of Zn isotopes, is dominated by the release of sulfate-bearing fluids associated with slab sulfide breakdown (Pons et al., 2016). Carbonate fluid transfer to the serpentinized mantle wedge The subduction of oceanic lithosphere and overlying sediments initiates a continuum of metamorphic reactions and dehydration/rehydration events between the downgoing slab, the slab/mantle wedge interface and/or the mantle wedge, which are strongly controlled by temperature (Fig. 10; Manning, 2004; Magni et al., 2014; Spandler et al., 2014). At shallow depths, the sedimentary rocks that compose the top of the slab and/or the slab/mantle interface experience compaction, deformation and metamorphic reactions (e.g. clay minerals breakdown) resulting in the release of FME-rich fluids (Bebout, 1997, 2014; Hattori & Guillot, 2007; Garofalo, 2012; Penniston-Dorland et al., 2012; Bebout et al., 2013; Lafay et al., 2013; Barnes et al., 2014; Cannaò et al., 2016; Debret et al., 2016a, b) and the transfer of these fluids to the overlying fore-arc mantle (Fig. 11). This serpentinized fore-arc mantle displays high FME concentrations relative to slab serpentinites (Fig. 10) and mixed isotopic signatures between depleted mantle and a ‘sedimentary’ component (Savov et al., 2007; Bouilhol et al., 2009; Deschamps et al., 2010; Scambelluri et al., 2014; Cannaò et al., 2016). Our study further characterizes these sediment-derived fluids as being isotopically light in Fe, which is most likely complexed with carbon. As Fe-bearing carbonates are highly soluble between 200–300°C (Milesi et al., 2015), Fe and C are expected to be released from the slab before it reaches the 400°C isotherm (Fig. 11b). This process, and the transfer of such Fe- and C-bearing fluids to the overlying mantle, may enhance the precipitation of C-bearing phases, such as calcite or graphite, within the fore-arc mantle. This serpentinization of mantle wedge peridotite has the potential to be associated with significant Fe oxidation as the result of Fe3+-bearing serpentine or magnetite precipitation and may, therefore, provide a means of sequestering reduced carbon into the fore-arc mantle according the following equation: 3 FeO+1/2 CO2 (aq)=Fe3O4+½1/2C Fig. 11 View largeDownload slide Sketches illustrating the changes in the nature of slab-derived fluids across subduction zones. (a) Subduction system with two main episodes of metasomatism. At low pressure and temperature, sediment compaction and devolatilization enhances the slab-mantle interface and induces mantle wedge serpentinization. This episode is accompanied by significant transfer of carbon and its storage within the slab-mantle interface and mantle wedge. At greater depths, serpentinite dehydration releases a huge amount of hydrous and sulphate-bearing fluids. (b) Zoom of the deep part of the subduction zone. The release of fluids is mainly controlled by subduction isotherms; the warmer part of subduction (top of the slab, slab-mantle interface and mantle wedge) will release fluids earlier than the serpentinized oceanic lithosphere. This involves a change in the nature of slab-derived fluids that pass from carbonate-dominated to sulphate-dominated through subduction processes producing the complex signature of fluids in the sub-arc mantle region. Fig. 11 View largeDownload slide Sketches illustrating the changes in the nature of slab-derived fluids across subduction zones. (a) Subduction system with two main episodes of metasomatism. At low pressure and temperature, sediment compaction and devolatilization enhances the slab-mantle interface and induces mantle wedge serpentinization. This episode is accompanied by significant transfer of carbon and its storage within the slab-mantle interface and mantle wedge. At greater depths, serpentinite dehydration releases a huge amount of hydrous and sulphate-bearing fluids. (b) Zoom of the deep part of the subduction zone. The release of fluids is mainly controlled by subduction isotherms; the warmer part of subduction (top of the slab, slab-mantle interface and mantle wedge) will release fluids earlier than the serpentinized oceanic lithosphere. This involves a change in the nature of slab-derived fluids that pass from carbonate-dominated to sulphate-dominated through subduction processes producing the complex signature of fluids in the sub-arc mantle region. At present, this reaction is speculative due to the absence of Fe and C redox state data for mantle wedge serpentinites. Nonetheless, in this model, the early stages of sediment devolatilization promote the serpentinization of fore-arc mantle, allowing it to act as a temporary reservoir for FME and carbon during subduction. The fore-arc mantle: a mixed source for arc magmas? The Kohistan gem olivine data are used here as a first-order proxy for slab-derived fluids that have percolated and reacted with the mantle wedge. The olivines are found in veins considered to have formed in a fore-arc setting near the 500°C isotherm (Bouilhol et al., 2009, 2012). The Kohistan gem olivines display light δ56Fe values (-0·06 to -0·36‰) relative to that of mantle olivine (e.g. San Carlos olivine δ56Fe is about 0·01‰; Sossi et al., 2015) that complement their heavy Zn isotope compositions (Pons et al., 2016) and support the hypothesis that they crystallized through reaction with an isotopically fractionated fluid. Although Fe is considered to be relatively immobile in aqueous fluids, the large amount of chlorine and sulfur released during serpentinite dehydration can increase Fe solubility and mobility (e.g. Wykes et al., 2008; Hill et al., 2010) as testified by the analysis of Fe-rich fluid inclusions trapped in dehydrated serpentinites (up to 2 wt %; Scambelluri et al., 2015). This provides evidence for the transfer of isotopically light Fe across subduction zones, in which Fe is complexed by chorine or sulfate anions (Hill et al., 2010; Debret et al., 2016b). High chalcophile element (Cu and Zn) concentrations in some Kohistan gem olivines, in conjunction with isotopically heavy δ66Zn signatures, also provide evidence for metasomatism by sulfate-bearing fluids (Pons et al., 2016). However, the presence of carbonate associated with these gem olivines as well as their FME concentrations suggests that carbonate-rich fluids with sediment-like signatures are also important metasomatic agents. This is further supported by the Sr (0·705626 > 87Sr/86Sr > 0·705449), C (-0·54 > δ13CVPDB > -3·98‰) and O (12·02 > δ18OSMOW > 9·45‰) isotopic compositions of carbonates associated with the olivines (Bouilhol et al., 2012). These geochemical characteristics point to a vein-forming fluid with a mixed signature between sulfate and carbonate. Indeed, recent findings of eclogitic garnet (Bardane, Norway; Rielli et al., 2017), formed through the interaction of slab-derived fluids and depleted mantle, or diamond-bearing fluid inclusions (Western Alps, Italy; Frezzotti et al., 2011) containing both sulfur- and carbon-bearing phases, suggests that the transfer of sulfate and carbonate from the slab to the mantle wedge could be concomitant in subduction zones. Such complex fluid signatures suggest that there are at least two stages of fluid transfer from the slab to the sub-arc mantle: the first resulting in the storage of carbon within the cold parts of the subduction zone (e.g. fore-arc mantle and/or slab mantle interface) and the second in the destabilization and release of C-bearing fluids in conjunction with sulfate-bearing fluids at greater depths (Fig. 11b). CONCLUSIONS Our study provides evidence for multiple episodes of carbon-bearing fluid transfer and the involvement of the fore-arc mantle as a temporary storage location for carbon. The ultramafic rocks of the Queyras Schistes Lustres complex record successive stages of carbonation and decarbonation during the onset of subduction. These are: (1) at LT, the percolation of C-bearing fluids from metasedimentary into ultramafic rocks results in the crystallization of Ca-carbonates associated with diopside, chlorite, antigorite and magnetite in the LT-blueschist unit. This episode is accompanied with a significant gain in FME, carbon and a decrease in δ56Fe value in ultramafic rocks while no Zn isotope fractionation is observed; and (2) in the MT- and HT-blueschist units the progressive disappearance of Ca-carbonates suggests a switch from a carbon to a H2O-dominated system. This process results in the remobilization of carbon, FME and isotopically heavy Zn by fluids without significant modification of ultramafic rock δ56Fe. Our results suggest extensive mobility of carbon, isotopically light Fe and FME during the early (LT) stage of the slab prograde path (i.e. before the slab reaches the 400°C isotherm). Such fluids are a potential metasomatic agent for the fore-arc mantle, which then acts as a temporary volatile reservoir. If dragged down to greater depth, the devolatilization of this area may supply both carbonate- and sulphate-bearing fluids to the source of arc melts, making the fore-arc mantle an important source for both elements in subduction systems. ACKNOWLEDGMENTS We thank C. Nicollet (LMV, Clermont-Ferrand, France) for his help in the field; J.L. Devidal (LMV, Clermont-Ferrand, France) and G. Nowell (Durham University, UK) for technical support during microprobe and MC-ICP-MS analyses, respectively; E. Inglis (Durham University, UK) and M. Cooper (National Oceanography Centre, Southampton, UK) for trace element analyses; A. Vitale-Brovarone (IPMC, Paris, France) for instructive discussions. We also thank P. Sossi and two anonymous reviewers for critical comments on an earlier version of this article and careful editorial handling by J. Hermann. FUNDING This work was supported by an ERC Starting Grant (HabitablePlanet; 306655) and a NERC Deep Volatiles Consortium Grant (NE/M0003/1) awarded to HW. BD acknowledges support from his F.N.R.S. (Fonds National de la Recherche Scientifique) chargé de recherche fellowship. PB acknowledges support from his Auvergne fellowship (ANR-10LABX-0006; Clervolc contribution #256) and an ERC Starting Grant 279828 awarded to J. van Hunen (Durham University, UK). REFERENCES Agard P. , Vidal P. , Goffé B. ( 2001 ). Interlayer and Si content of phengite in HP-LT carpholite-bearing metapelites . Journal of Metamorphic Geology 19 , 479 – 495 . Google Scholar Crossref Search ADS Agard P. , Monie P. , Jolivet L. , Goffe B. ( 2002 ). Exhumation of the Schistes Lustres complex: in situ laser probe 40Ar/39Ar constraints and implications for the Western Alps . Journal of Metamorphic Geology 20 , 599 – 618 . Google Scholar Crossref Search ADS Ague J. J. , Nicolescu S. ( 2014 ). Carbon dioxide released from subduction zones by fluid-mediated reactions . Nature Geoscience 7 , 355 – 360 . Google Scholar Crossref Search ADS Alt J. C. , Teagle D. A. H. ( 1999 ). Uptake of carbon during alteration of oceanic crust . Geochimica et Cosmochimica Acta 63 , 1527 – 1535 . Google Scholar Crossref Search ADS Alt J. , Schwarzenbach E. M. , Früh-Green G. L. , Shanks W. C. S. , Bernasconi S. M. , Garrido C. J. , Crispini L. , Gaggero L. , Padrón-Navarta J. A. , Marchesi C. ( 2013 ). The role of serpentinites in cycling of carbon and sulfur: seafloor serpentinization and subduction metamorphism . Lithos 178 , 40 – 54 . Google Scholar Crossref Search ADS Andreani M. , Mével C. , Boullier A. M. , Escartin J. ( 2007 ). Dynamic control on serpentine crystallization in veins: constraints on hydration processes in oceanic peridotites . Geochemistry, Geophysics, Geosystems 8 , Q02012 . Google Scholar Crossref Search ADS Andreani M. , Muñoz M. , Marcaillou C. , Delacour A. ( 2013 ). μXANES study of iron redox state in serpentine during oceanic serpentinization . Lithos 178 , 70 – 83 . Google Scholar Crossref Search ADS Ballèvre M. , Lagabrielle Y. , Merle O. ( 1990 ). Tertiary ductile normal faulting as a consequence of lithospheric stacking in the western Alps . Société Géologique de France, Mémoires 156 , 227 – 236 . Ballhaus C. , Frost B. R. ( 1994 ). The generation of oxidized CO2-bearing basaltic melts from reduced CH4-bearing upper mantle sources . Geochimica et Cosmochimica Acta 58 , 4931 – 4940 . Google Scholar Crossref Search ADS Barnes J. D. , Beltrando M. , Lee C. T. A. , Cisneros M. , Loewy S. , Chin E. ( 2014 ). Geochemistry of Alpine serpentinites from rifting to subduction: A view across paleogeographic domains and metamorphic grade . Chemical Geology 389 , 29 – 47 . Google Scholar Crossref Search ADS Beard J. S. , Frost B. R. ( 2016 ). The stoichiometric effects of ferric iron substitutions in serpentine from microprobe data . International Geology Review 58 , 1 – 547 . Google Scholar Crossref Search ADS Bebout G. E. ( 1997 ). Nitrogen isotope tracers of high-temperature fluid–rock interactions: case study of the Catalina Schist, California . Earth and Planetary Science Letters 151 , 77 – 90 . Google Scholar Crossref Search ADS Bebout G. E. ( 2014 ). Chemical and isotopic cycling in subduction zones. In: Holland H. D. , Turekian K. K. (eds) Treatise on Geochemistry , Vol. 4 , 2 nd edn. Oxford : Elsevier , pp. 703 – 747 . Bebout G. E. , Bebout A. E. , Graham C. M. ( 2007 ). Cycling of B, Li, and LILE (K, Cs, Rb, Ba, Sr) into subduction zones: SIMS evidence from micas in high-P/T metasedimentary rocks . Chemical Geology 239 , 284 – 304 . Google Scholar Crossref Search ADS Bebout G. E. , Agard P. , Kobayashi K. , Moriguti T. , Nakamura E. ( 2013 ). Devolatilization history and trace element mobility in deeply subducted sedimentary rocks: evidence from Western Alps HP/UHP suites . Chemical Geology 342 , 1 – 20 . Google Scholar Crossref Search ADS Belshaw N. S. , Zhu X. K. , Guo Y. , O’Nions R. K. ( 2000 ). High precision measurement of iron isotopes by plasma source mass spectrometry . International Journal of Mass Spectrometry 197 , 191 – 195 . Google Scholar Crossref Search ADS Bentahila Y. , Ben Othman D. , Luck J. M. ( 2008 ). Strontium, lead and zinc isotopes in marine cores as tracers of sedimentary provenance: a case study around Taiwan orogeny . Chemical Geology 248 , 62 – 82 . Google Scholar Crossref Search ADS Black J. R. , Kavner A. , Schauble E. A. ( 2011 ). Calculation of equilibrium stable isotope partition function ratios for aqueous zinc complexes and metallic zinc . Geochimica Cosmochimica Acta 75 , 769 – 783 . Google Scholar Crossref Search ADS Blanchard M. , Poitrasson F. , Méheut M. , Lazzeri M. , Mauri F. , Balan E. ( 2009 ). Iron isotope fractionation between pyrite (FeS2), hematite (Fe2O3) and siderite (FeCO3): A first-principles density functional theory study . Geochimica Cosmochimica Acta 73 , 6565 – 6578 . Google Scholar Crossref Search ADS Bodinier J. L. , Godard M. ( 2013 ). Orogenic, ophiolitic, and abyssal peridotites, In: Carlson R. W. (ed.) Treatise on Geochemistry , Volume 2 , The Mantle and Core, 2 nd edn. The Netherlands : Elsevier , pp. 103 – 151 . Bouilhol P. , Jagoutz O. , Hanchar J. M. , Dudas F. O. ( 2013 ). Dating the India–Eurasia collision through arc magmatic records . Earth and Planetary Science Letters 366 , 163 – 175 . Google Scholar Crossref Search ADS Bouilhol P. , Burg J. P. , Bodinier J. L. , Schmidt M. W. , Bernasconi M. , Dawood H. ( 2012 ). Gem olivine and calcite mineralization precipitated from subduction derived fluids in the Kohistan arc-mantle (Pakistan) . The Canadian Mineralogist 50 , 1291 – 1304 . Google Scholar Crossref Search ADS Bouilhol P. , Burg J. P. , Bodinier J. L. , Schmidt M. W. , Dawood H. , Hussain S. ( 2009 ). Magma and fluid percolation in arc to fore-arc mantle: evidence from Sapat (Kohistan, Northern Pakistan) . Lithos 107 , 17 – 37 . Google Scholar Crossref Search ADS Busigny V. , Cartigny P. , Philippot P. , Ader M. , Javoy M. ( 2003 ). Massive recycling of nitrogen and other fluid-mobile elements (K, Rb, Cs, H) in a cold slab environment: evidence from HP to UHP oceanic metasediments of the Schistes Lustrés nappe (western Alps, Europe) . Earth and Planetary Science Letters 215 , 27 – 42 . Google Scholar Crossref Search ADS Caciagli N. C. , Manning C. E. ( 2003 ). The solubility of calcite in water at 6–16kbar and 500–800°C . Contribution to Mineralogy and Petrology 146 , 275 – 285 . Google Scholar Crossref Search ADS Canales J. P. , Collins J. A. , Escartin J. , Detrick R. S. ( 2000 ). Seismic structure across the rift valley of the Mid-Atlantic ridge at 23°20’N (MARK area): implications for crustal accretion processes at slow-spreading ridges . Journal of Geophysical Research: Solid Earth 105 , 28411 – 28425 . Google Scholar Crossref Search ADS Cannaò E. , Scambelluri M. , Agostini S. , Tonarini S. , Godard M. ( 2016 ). Linking serpentinite geochemistry with tectonic evolution at the subduction plate-interface: the Voltri Massif case study (Ligurian Western Alps, Italy) . Geochimica et Cosmochimica Acta 190 , 115 – 133 . Google Scholar Crossref Search ADS Catling D. C. ( 2012 ). The Great oxidation event transition. In: Canfield D. , Farquhar J. , Kasting J. F. (eds) The Atmosphere—History, Vol. 13, Treatise on Geochemistry , 2 nd edn. Elsevier , 177 – 195 . Catling D. C. , Claire M. W. ( 2005 ). How Earth’s atmosphere evolved to an oxic state: a status report . Earth and Planetary Science Letters 237 , 1 – 20 . Google Scholar Crossref Search ADS Chen H. , Savage P. S. , Teng F. Z. , Helz R. T. , Moynier F. ( 2013 ). Zinc isotope fractionation during magmatic differentiation and the isotopic composition of the bulk Earth . Earth and Planetary Science Letters 369–370 , 34 – 42 . Google Scholar Crossref Search ADS Collins N. C. , Bebout G. E. , Angiboust S. , Agard P. , Scambelluri M. , Crispini L. , John T. ( 2015 ). Subduction-zone metamorphic pathway for deep carbon cycling: II. Ev-idence from HP/UHP metabasaltic rocks and ophicarbonates . Chemical Geology 412 , 132 – 150 . Google Scholar Crossref Search ADS Connolly J. A. D. ( 2005 ). Computation of phase equilibria by linear programming: atool for geodynamic modeling and its application to subduction zone decar-bonation . Earth and Planetary Science Letters 236 , 524 – 541 . Google Scholar Crossref Search ADS Cook-Kollars J. , Bebout G. E. , Collins N. C. , Angiboust S. , Agard P. ( 2014 ). Subduction-zone metamorphic pathway for deep carbon cycling: evidence from HP/UHP metasedimentary rocks, Italian Alps . Chemical Geology 386 , 31 – 48 . Google Scholar Crossref Search ADS Craddock P. R. , Warren J. M. , Dauphas N. ( 2013 ). Abyssal peridotites reveal the near-chondritic Fe isotopic composition of the Earth . Earth and Planetary Science Letters 365 , 63 – 76 . Google Scholar Crossref Search ADS Dasgupta R. , Hirschmann M. M. ( 2010 ). The deep carbon cycle and melting in Earth’s interior . Earth and Planetary Science Letters 298 , 1 – 13 . Google Scholar Crossref Search ADS Dauphas N. , John S. G. , Rouxel O. ( 2017 ). Iron isotopes systematics . Reviews in Mineralogy & Geochemistry 82 , 415 – 510 . Google Scholar Crossref Search ADS Dauphas N. , Craddock P. R. , Asimow P. D. , Bennett V. C. , Nutman A. P. , Ohnenstetter D. ( 2009 ). Iron isotopes may reveal the redox conditions of mantle melting from Archean to present . Earth and Planetary Science Letters 288 , 255 – 267 . Google Scholar Crossref Search ADS Dauphas N. , Roskosz M. , Alp E. E. , Neuville D. R. , Hu M. Y. , Sio C. K. , Tissot F. L. H. , Zhao J. , Tissandier L. , Médard E. , Cordier C. ( 2014 ). Magma redox and structural controls on iron isotope variations in Earth’s mantle and crust . Earth and Planetary Science Letters 398 , 127 – 140 . Google Scholar Crossref Search ADS Debret B. , Sverjensky D. ( 2017 ). Highly oxidising fluids generated during serpentinite breakdown in subduction zones . Scientific Reports 7 , doi:10.1038/s41598-017-09626-y. Debret B. , Andreani M. , Godard M. , Nicollet C. , Schwartz S. , Lafay R. ( 2013a ). Trace element behaviour during serpentinization/deserpentinization of an eclogitized oceanic lithosphere: a LA-ICPMS study of the Lanzo ultramafic massif (Western Alps) . Chemical Geology 357 , 117 – 133 . Google Scholar Crossref Search ADS Debret B. , Nicollet C. , Andreani M. , Schwartz S. , Godard M. ( 2013b ). Three steps of serpentinization in an eclogitized oceanic serpentinization front (Lanzo Massif—Western Alps) . Journal of Metamorphic Geology 31 , 165 – 186 . Google Scholar Crossref Search ADS Debret B. , Koga K. T. , Cattani F. , Nicollet C. , Van den Bleeken G. , Schwartz S. ( 2016a ). Volatile (Li, B, F and Cl) mobility during amphibole breakdown in subduction zones . Lithos 244 , 165 – 181 . Google Scholar Crossref Search ADS Debret B. , Millet M. A. , Pons M. L. , Bouilhol P. , Inglis E. , Williams H. ( 2016b ). Isotopic evidence for iron mobility during subduction . Geology 44 , 215 – 218 . Google Scholar Crossref Search ADS De Hoog J. C. M. , Gall L. , Cornell D. H. ( 2010 ). Trace-element geochemistry of mantle olivine and application to mantle petrogenesis and geothermobarometry . Chemical Geology 270 , 196 – 215 . Google Scholar Crossref Search ADS Delacour A. , Früh-Green G. L. , Bernasconi S. M. , Schaeffer P. , Kelley D. S. ( 2008 ). Carbon geochemistry of serpentinites in the Lost City Hydrothermal system . Geochimica et Cosmochimica Acta 72 , 3681 – 3702 . Google Scholar Crossref Search ADS Deschamps F. , Guillot S. , Godard M. , Chauvel C. , Andreani M. , Hattori K. ( 2010 ). In situ characterization of serpentinites from forearc mantle wedges: timing of serpentinization and behavior of fluid-mobile elements in subduction zones . Chemical Geology 269 , 262 – 277 . Google Scholar Crossref Search ADS Deschamps F. , Godard M. , Guillot S. , Hattori K. ( 2013 ). Geochemistry of subduction zone serpentinites: a review . Lithos 178 , 96 – 127 . Google Scholar Crossref Search ADS Doucet L. S. , Mattielli N. , Ionov D. A. , Debouge W. , Golovin A. V. ( 2016 ). Zn isotopic heterogeneity in the mantle: a melting control? Earth and Planetary Science Letters 451 , 232 – 240 . Google Scholar Crossref Search ADS Duncan M. S. , Dasgupta R. ( 2017 ). Rise of Earth’s atmospheric oxygen controlled by efficient subduction of organic carbon . Nature Geoscience 10 , 387 . Google Scholar Crossref Search ADS Evans B. W. ( 2004 ). The serpentinite multisystem revisited: chrysotile is metastable . International Geology Review 46 , 479 – 506 . Google Scholar Crossref Search ADS Facq S. , Daniel I. , Montagnac G. , Cardon H. , Sverjensky D. A. ( 2014 ). In situ Raman study and thermodynamic model of aqueous carbonate speciation in equilibrium with aragonite under subduction zone conditions . Geochimica et Cosmochimica Acta 132 , 375 – 390 . Google Scholar Crossref Search ADS Franzolin E. , Schmidt M. W. , Poli S. ( 2011 ). Ternary Ca–Mg–Fe carbonates: subsolidus phase relations at 3.5GPa and a thermodynamic solid solution model including order/disorder . Contributions to Mineralogy and Petrology 161 , 213 – 227 . Google Scholar Crossref Search ADS Frezzotti M. L. , Selverstone J. , Sharp Z. D. , Compagnoni R. ( 2011 ). Carbonate dissolution during subduction revealed by diamond-bearing rocks from the Alps . Nature Geoscience 4 , 703 – 706 . Google Scholar Crossref Search ADS Fujii T. , Moynier F. , Blichert-Toft J. , Albarède F. ( 2014 ). Density functional theory estimation of isotope fractionation of Fe, Ni, Cu, and Zn among species relevant to geochemical and biological environments . Geochimica et Cosmochimica Acta 140 , 553 – 576 . Google Scholar Crossref Search ADS Fujii T. , Moynier F. , Pons M. L. , Albarède F. ( 2011 ). The origin of Zn isotope fractionation in sulfides . Geochimica et Cosmochimica Acta 75 , 7632 – 7643 . Google Scholar Crossref Search ADS Fujii T. , Moynier F. , Telouk P. , Abe M. ( 2010 ). Experimental and theoretical investigation of isotope fractionation of zinc between aqua, chloro, and macrocyclic complexes . The Journal of Physical Chemistry A 114 , 2543 – 2552 . Google Scholar Crossref Search ADS PubMed Galvez M. , Beyssac O. , Martinez I. , Benzerara K. , Chaduteau C. , Malvoisin B. , Malavieille J. ( 2013 ). Graphite formation by carbonate reduction during subduc-tion . Nature Geoscience 6 , 473 – 477 . Google Scholar Crossref Search ADS Garofalo P. S. ( 2012 ). The composition of Alpine marine sediments (Bündnerschiefer Formation, W Alps) and the mobility of their chemical components during orogenic metamorphism . Lithos 128–131 , 55 – 72 . Google Scholar Crossref Search ADS Garrido C. J. , López-Sánchez-Vizcaíno V. , Gómez-Pugnaire M. T. , Trommsdorff V. , Alard O. , Bodinier J. L. , Godard M. ( 2005 ). Enrichment of HFSE in chlorite-harzburgite produced by high-pressure dehydration of antigorite-serpentinite: implications for subduction magmatism . Geochemistry, Geophysics, Geosystems 6 , Q01J15 . Google Scholar Crossref Search ADS Godard M. , Lagabrielle Y. , Alard O. , Harvey J. ( 2008 ). Geochemistry of the highly depleted peridotites drilled at ODP Sites 1272 and 1274 (Fifteen-Twenty Fracture Zone, Mid-Atlantic Ridge): Implications for mantle dynamics beneath a slow spreading ridge . Earth and Planetary Science Letters 267 , 410 – 425 . Google Scholar Crossref Search ADS Gorman P. J. , Kerrick D. M. , Connolly J. A. D. ( 2006 ). Modeling open system metamor-phic decarbonation of subducting slabs . Geochemistry, Geophysics, Geosystems 7 , Q04007 . Google Scholar Crossref Search ADS Hattori K. H. , Guillot S. ( 2007 ). Geochemical character of serpentinites associated with high- to ultrahigh-pressure metamorphic rocks in the Alps, Cuba, and the Himalayas: recycling of elements in subduction zones . Geochemistry, Geophysics, Geosystems 8 , Q09010 . Google Scholar Crossref Search ADS Hermann J. , Zheng Y.-F. , Rubatto D. ( 2013 ). Deep fluids in subducted continental crust . Elements 9 , 281 – 287 . Google Scholar Crossref Search ADS Hibbert K. E. J. , Williams H. M. , Kerr A. C. , Puchtel I. S. ( 2012 ). Iron isotopes in an-cient and modern komatiites: evidence in support of an oxidised mantle from Archean to present . Earth and Planetary Science Letters 321–322 , 198 – 207 . Google Scholar Crossref Search ADS Hill P. S. , Schauble E. A. ( 2008 ). Modeling the effects of bond environment on equilibrium iron isotope fractionation in ferric aquo-chloro complexes . Geochimica et Cosmochimica Acta 72 , 1939 – 1958 . Google Scholar Crossref Search ADS Hill P. S. , Schauble E. A. , Young E. D. ( 2010 ). Effects of changing solution chemistry on Fe3+/Fe2+ isotope fractionation in aqueous Fe–Cl solutions . Geochimica et Cosmochimica Acta 74 , 6669 – 6689 . Google Scholar Crossref Search ADS Holland T. J. B. , Powell R. ( 1998 ). An internally consistent thermodynamic data set for phases of petrological interest . Journal of Metamorphic Geology 16 , 309 – 343 . Google Scholar Crossref Search ADS Holland T. J. B. , Powell R. ( 2003 ). Activity-composition relations for phases in petro-logical calculations: an asymmetric multicomponent formulation . Contributions to Mineralogy and Petrology 145 , 492 – 501 . Google Scholar Crossref Search ADS Inglis E. C. , Debret B. , Burton K. W. , Millet M. A. , Pons M. L. , Dale C. W. , Bouilhol P. , Cooper M. , Nowell G. M. , McCoy-West A. , Williams H. M. ( 2017 ). The behaviour of iron and zinc stable isotopes accompanying the subduction of mafic oceanic crust: a case study from Western Alpine ophiolites . Geochemistry, Geophysics, Geosystems 18 , 2562 - 2579 . Google Scholar Crossref Search ADS Johnson C. M. , Beard B. L. , Klein C. , Beukes N. J. , Roden E. E. ( 2008 ). Iron isotopes constrain biologic and abiologic processes in Banded Iron Formation genesis . Geochimica et Cosmochimica Acta 72 , 151 – 169 . Google Scholar Crossref Search ADS Kelemen P. B. , Manning C. E. ( 2015 ). Reevaluating carbon fluxes in subduction zones, what goes down, mostly comes up . Proceedings of the National Academy of Sciences of the United States of America 112 , E3997 – E4006 . Google Scholar Crossref Search ADS PubMed Kerrick D. M. , Connolly J. A. D. ( 1998 ). Subduction of ophicarbonates and recycling of CO2and H2O . Geology 26 , 375 – 378 . Google Scholar Crossref Search ADS Konter J. G. , Pietruszka A. J. , Hanan B. B. , Finlayson V. A. , Craddock P. R. , Jackson M. G. , Dauphas N. ( 2016 ). Unusual d56Fe values in Samoan rejuvenated lavas generated in the mantle . Earth and Planetary Science Letters 450 , 221 – 232 . Google Scholar Crossref Search ADS Lafay R. , Deschamps F. , Schwartz S. , Guillot S. , Godard M. , Debret B. , Nicollet C. ( 2013 ). High-pressure serpentinites, a trap-and-release system controlled by metamorphic conditions: example from the Piedmont zone of the western Alps . Chemical Geology 343 , 38 – 54 . Google Scholar Crossref Search ADS Lagabrielle Y. , Vitale Brovarone A. , Ildefonse B. ( 2014 ). Fossil Oceanic Core Complexes recognized in the blueschist ophiolitic units of Western Alps and Corsica . Earth Science Reviews 141 , 1 . Google Scholar Crossref Search ADS Lagabrielle Y. , Polino R. , Auzende J. M. , Blanchet R. , Caby R. , Fudral S. , Lemoine M. , Mével C. , Ohnenstetter M. , Robert D. , Tricart P. ( 1984 ). Les témoins d’une tectonique intra-océanique dans le domaine téthysien: analyse des rapports entre les ophiolites et leurs couvertures métasédimentaires dans la zone piémontaise des Alpes franco-italiennes . Ofioliti 9 , 67 – 88 . Lanari P. , Wagner T. , Vidal O. ( 2014 ). A thermodynamic model for di-trioctahedral chlorite from experimental and natural data in the system MgO–FeO–Al2O3–SiO2–H2O: applications to P–T sectionsand geothermometry . Contributions to Mineralogy and Petrology 167 , 968 . Google Scholar Crossref Search ADS Lemoine M. , Tricart P. , Boillot G. ( 1987 ). Ultramafic and gabbroic ocean floor of the Ligurian Tethys (Alps, Corsica, Apennines): in search of a genetic model . Geology 15 , 622 – 625 . Google Scholar Crossref Search ADS Lyons T. W. , Reinhard C. T. , Planavsky N. J. ( 2014 ). The rise of oxygen in Earth’s early ocean and atmosphere . Nature 506 , 307 – 315 . Google Scholar Crossref Search ADS PubMed Magni V. , Bouilhol P. , van Hunen J. ( 2014 ). Deep water recycling through time . Geochemistry, Geophysics, Geosystems ; doi:10.1002/2014GC005525. Malaspina N. , Poli S. , Fumagalli P. ( 2009 ). The oxidation state of metasomatized mantle wedge: insights from C–O–H-bearing garnet peridotite . Journal of Petrology 50 , 1533 – 1552 . Google Scholar Crossref Search ADS Manning C. E. ( 2004 ). The chemistry of subduction-zone fluids . Earth and Planetary Science Letters 223 , 1 – 16 . Google Scholar Crossref Search ADS Martin B. , Fyfe W. S. ( 1970 ). Some experimental and theoretical observations on the kinetics of hydratation reactions with particular reference to serpentinisation . Chemical Geology 6 , 185 – 202 . Google Scholar Crossref Search ADS Milesi V. , Guyot F. , Brunet F. , Richard L. , Recham N. , Benedetti M. , Dairou J. , Prinzhofer A. ( 2015 ). Formation of CO2, H2 and condensed carbon from siderite dissolution in the 200–300 °C range and at 50 MPa . Geochimica et Cosmochimica Acta 154 , 201 – 211 . Google Scholar Crossref Search ADS Millet M. A. , Baker J. A. , Payne C. E. ( 2012 ). Ultra-precise stable Fe isotope measurements by high resolution multiple-collector inductively coupled plasma mass spectrometry with a 57 Fe–58 Fe double spike . Chemical Geology 304–305 , 18 – 25 . Google Scholar Crossref Search ADS Moeller K. , Schoenberg R. , Pedersen R.-B. , Weiss D. , Dong S. ( 2012 ). Calibration of the new certified reference materials ERM-AE633 and ERM-AE647 for copper and IRMM-3702 for zinc isotope amount ratio determinations . Geostandards and Geoanalytical Research 36 , 177 – 199 . Google Scholar Crossref Search ADS Navrotsky A. , Capobianco C. ( 1987 ). Enthalpies of formation of dolomite and magne-sian calcites . American Mineralogist 72 , 782 – 787 . Nebel O. , Sossi P. A. , Bénard A. , Wille M. , Vroon P. Z. , Arculus R. J. ( 2015 ). Redox-variability and controls in subduction zones from an iron-isotope perspec-tive . Earth and Planetary Science Letters 432 , 142 – 151 . Google Scholar Crossref Search ADS Padrón-Navarta J. A. , López Sánchez-Vizcaíno V. , Garrido C. J. , Gómez-Pugnaire M. T. ( 2011 ). Metamorphic record of high-pressure dehydration of antigorite serpentinite to chlorite harzburgite in a subduction setting (Cerro del Almirez, Nevado-Filabride Complex, Southern Spain) . Journal of Petrology 52 , 2047 – 2078 . Google Scholar Crossref Search ADS Padrón-Navarta J. A. , López Sánchez-Vizcaíno V. , Hermann J. , Connolly J. A. D. , Garrido C. J. , Gómez-Pugnaire M. T. , Marchesi C. ( 2013 ). Tschermak’s substitution in antigorite and consequences for phase relations and water liberation in high-grade serpentinites . Lithos 178 , 186 – 196 . Google Scholar Crossref Search ADS Penniston-Dorland S. C. , Bebout G. E. , Pogge von Strandmann P. A. E. , Elliott T. , Sorensen S. S. ( 2012 ). Lithium and its isotopes as tracers of subduction zone fluids and metasomatic processes: evidence from the Catalina Schist, California, USA . Geochimica et Cosmochimica Acta 77 , 530 – 545 . Google Scholar Crossref Search ADS Pens M. , Andreani M. , Daniel I. , Perrillat J. P. , Cardon H. ( 2016 ). Contrasted effect of aluminum on the serpentinization rate of olivine and orthopyroxene under hydrothermal conditions . Chemical Geology 441 , 256 – 264 . Google Scholar Crossref Search ADS Peters D. , Bretscher A. , John T. , Scambelluri M. , Pettke T. ( 2017 ). Fluid-mobile elements in serpentinites: Constraints on serpentinisation environments and element cycling in subduction zones . Chemical Geology 466 , 654 – 666 . Google Scholar Crossref Search ADS Piccoli F. , Vitale Brovarone A. , Beyssac O. , Martinez I. , Ague J. J. , Chaduteau C. ( 2016 ). Carbonation by fluid–rock interactions at high-pressure conditions: implications for carbon cycling in subduction zones . Earth and Planetary Science Letters 445 , 146 – 159 . Google Scholar Crossref Search ADS Pichat S. , Douchet C. , Albarède F. ( 2003 ). Zinc isotope variations in deep-sea carbonates from the eastern equatorial Pacific over the last 175 ka . Earth and Planetary Science Letters 210 , 167 – 178 . Google Scholar Crossref Search ADS Poli S. ( 2015 ). Carbon mobilized at shallow depths in subduction zones by carbonatitic liquids . Nature Geoscience 8 , 633 – 636 . Google Scholar Crossref Search ADS Polyakov V. B. , Mineev S. D. ( 2000 ). The use of Mössbauer spectroscopy in stable isotope geochemistry . Geochimica et Cosmochimica Acta 64 , 849 – 865 . Google Scholar Crossref Search ADS Pons M. L. , Debret B. , Bouilhol P. , Delacour A. , Williams H. ( 2016 ). Zinc isotope evidence for sulfate-rich fluid transfer across subduction zones . Nature Communications 7 , 13794 . Google Scholar Crossref Search ADS PubMed Pons M. L. , Quitté G. , Fujii T. , Rosing M. T. , Reynard B. , Moynier F. , Douchet C. , Albarède F. ( 2011 ). Early archean serpentine mud volcanoes at Isua, Greenland, as a niche for early life . Proceedings National Academy Sciences of the United States of America 108 , 17639 – 17643 . Google Scholar Crossref Search ADS Ranero C. R. , Phipps Morgan J. , McIntosh K. , Reichert C. ( 2003 ). Bending-related faulting and mantle serpentinization at the Middle America trench . Nature 425 , 367 – 373 . Google Scholar Crossref Search ADS PubMed Ranero C. R. , Sallares V. ( 2004 ). Geophysical evidence for hydration of the crust and mantle of the Nazca plate during bending at the north Chile trench . Geology 32 , 549 – 552 . Google Scholar Crossref Search ADS Reynard B. ( 2013 ). Serpentine in active subduction zones . Lithos 178 , 171 – 185 . Google Scholar Crossref Search ADS Rielli A. , Tomkins A. G. , Nebel O. , Brugger J. , Etschmann B. , Zhong R. , Yaxley G. M. , Paterson D. ( 2017 ). Evidence of sub-arc mantle oxidation by sulphur and carbon . Geochemical Perspective Letters 3 , 124 – 132 . Google Scholar Crossref Search ADS Rouxel O. , Dobbek N. , Ludden J. , Fouquet Y. ( 2003 ). Iron isotope fractionation during oceanic crust alteration . Chemical Geology 202 , 155 – 182 . Google Scholar Crossref Search ADS Savov I. P. , Ryan J. G. , D’Antonio M. , Fryer P. ( 2007 ). Shallow slab fluid release across and along the Mariana arc-basin system: insights from geochemistry of serpentinized peridotites from the Mariana fore arc . Journal of Geophysical Research 112, B09205 . Google Scholar Crossref Search ADS Scambelluri M. , Pettke T. , Cannao E. ( 2015 ). Fluid-related inclusions in Alpine high-pressure peridotite reveal trace element recycling during subduction-zone dehydration of serpentinized mantle (Cima di Gagnone, Swiss Alps) . Earth and Planetary Science Letters 429 , 45 – 59 . Google Scholar Crossref Search ADS Scambelluri M. , Müntener O. , Ottolini L. , Pettke T. T. , Vannucci R. ( 2004 ). The fate of B, Cl and Li in the subducted oceanic mantle and in the antigorite breakdown fluids . Earth and Planetary Science Letters 222 , 217 – 234 . Google Scholar Crossref Search ADS Scambelluri M. , Pettke T. , Rampone E. , Godard M. , Reusser E. ( 2014 ). Petrology and trace element budgets of high-pressure peridotites indicate subduction dehydration of serpentinized mantle (Cima di Gagnone, Central Alps, Switzerland) . Journal of Petrology 55 , 459 – 498 . Google Scholar Crossref Search ADS Scambelluri M. , Bebout B. E. , Belmonte D. , Gilio M. , Campomenosi N. , Collins N. , Crispini L. ( 2016 ). Carbonation of subduction-zone serpentinite (high-pressure ophicarbonate; Ligurian Western Alps) and implications for the deep carbon cycling . Earth and Planetary Science Letters 441 , 155 – 166 . Google Scholar Crossref Search ADS Schauble E. A. ( 2004 ). Applying stable isotope fractionation theory to new systems . Reviews in Mineralogy and Geochemistry 55 , 65 – 102 . Google Scholar Crossref Search ADS Schauble E. A. , Rossman G. R. , Taylor H. P. ( 2001 ). Theoretical estimates of equilibrium Fe-isotope fractionations from vibrational spectroscopy . Geochimica et Cosmochimica Acta 65 , 2487 – 2497 . Google Scholar Crossref Search ADS Schrenk M. O. , Brazelton W. J. , Lang S. Q. ( 2013 ). Serpentinization, carbon, and deep life . Reviews in Mineralogy & Geochemistry 75 , 575 – 606 . Google Scholar Crossref Search ADS Schuessler J. A. , Schoenberg R. , Sigmarsson O. ( 2009 ). Iron and lithium isotope systematics of the Hekla volcano, Iceland—evidence for Fe isotope fractionation during magma differentiation . Chemical Geology 258 , 78 – 91 . Google Scholar Crossref Search ADS Schwartz S. , Guillot S. , Reynard B. , Lafay R. , Debret B. , Nicollet C. , Lanari P. , Auzende A. L. ( 2013 ). Pressure-temperature estimates of the lizardite/antigorite transition in high pressure serpentinites . Lithos 178 , 197 – 210 . Google Scholar Crossref Search ADS Selverstone J. , Sharp Z. D. ( 2015 ). Chlorine isotope behavior during prograde metamorphism of sedimentary rocks . Earth and Planetary Science Letters 417 , 120 – 131 . Google Scholar Crossref Search ADS Sleep N. H. , Bird D. K. , Pope E. C. ( 2011 ). Serpentinite and the dawn of life . Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 366 , 2857 – 2869 . Google Scholar Crossref Search ADS PubMed Sossi P. A. , Nebel O. , Foden J. ( 2016 ). Iron isotope systematics in planetary reservoirs . Earth and Planetary Science Letters 452 , 295 – 308 . Google Scholar Crossref Search ADS Sossi P. A. , Halverson P. A. , Nebel O. , Eggins S. M. ( 2015 ). Combined separation of Cu, Fe and Zn from rock matrices and improved analytical protocols for stable isotope determination . Geostandards and Geoanalytical Research 39 , 129 – 149 . Google Scholar Crossref Search ADS Spandler C. , Pettke T. , Hermann J. ( 2014 ). Experimental study of trace element release during ultrahigh-pressure serpentinite dehydration . Earth and Planetary Science Letters 391 , 296 – 306 . Google Scholar Crossref Search ADS Stagno V. , Ojwang D. O. , McCammon C. A. , Frost D. J. ( 2013 ). The oxidation state of the mantle and the extraction of carbon from Earth/’s interior . Nature 493 , 84 – 88 . Google Scholar Crossref Search ADS PubMed Sun S. , McDonough W. F. ( 1989 ). Chemical and Isotopic Systematics of Oceanic Basalts: Implications for Mantle Composition and Processes. In: Saunders, A. D. & Norry, M. J. (eds), Magmatism in the Ocean Basins . Geological Society Special Publication 42 , 313 – 345 . Google Scholar Crossref Search ADS Tahirkheli R. A. K. , Mattauer M. , Proust F. , Tapponnier P. ( 1979 ). The India Eurasia Suture Zone in Northern Pakistan: synthesis and interpretation of recent data at plate scale. In: Farah A. , De Jong K. A. (eds) Geodynamics of Pakistan . Geological Survey of Pakistan , 125 – 130 . Telus M. , Dauphas N. , Moynier F. , Tissot F. L. , Teng F. Z. , Nabelek P. I. , Craddock P. R. , Groat L. A. ( 2012 ). Iron, zinc, magnesium and uranium isotopic fractionation during continental crust differentiation: the tale from migmatites, granitoids, and pegmatites . Geochimica et Cosmochimica Acta 97 , 247 – 265 . Google Scholar Crossref Search ADS Teng F. Z. , Dauphas N. , Helz R. T. ( 2008 ). Iron isotope fractionation during magmatic differentiation in Kilauea Iki lava lake . Science (New York, N.Y.) 320 , 1620 – 1622 . Google Scholar Crossref Search ADS PubMed Teng F. Z. , Dauphas N. , Huang S. , Marty B. ( 2013 ). Iron isotopic systematics of oceanic basalts . Geochimica et Cosmochimica Acta 107 , 12 – 26 . Google Scholar Crossref Search ADS Tricart P. ( 1984 ). From passive margin to continental collision: a tectonic scenario for the western Alps . American Journal of Science 284 , 97 – 120 . Google Scholar Crossref Search ADS Tricart P. , Schwartz S. ( 2006 ). A north–south section across the Queyras Schistes lustrés (Piedmont zone, Western Alps): syncollision refolding of a subduction wedge . Eclogae Geologicae Helvetiae 99 , 429 – 442 . Google Scholar Crossref Search ADS Trommsdorff V. , Sánchez-Vizcaíno V. L. , Gómez-Pugnaire M. T. , Müntener O. ( 1998 ). High pressure breakdown of antigorite to spinifex-textured olivine and orthopyroxene, SE Spain . Contribution to Mineralogy and Petrology 132 , 139 – 148 . Google Scholar Crossref Search ADS Tumiati S. , Godard G. , Martin S. , Malaspina N. , Poli S. ( 2015 ). Ultra-oxidized rocks in subduction mélanges? Decoupling between oxygen fugacity and oxygen availability in a Mn-rich metasomatic environment . Lithos 226 , 116 – 130 . Google Scholar Crossref Search ADS Ulmer P. , Trommsdorff V. ( 1995 ). Serpentine stability to mantle depths and subduction-related magmatism . Science (New York, N.Y.) 268 , 858 – 861 . Google Scholar Crossref Search ADS PubMed Urey H. C. ( 1947 ). The thermodynamic properties of isotopic substances . Journal of the Chemical Society (Resumed) , 562 – 581 . Vils F. , Pelletier L. , Kalt A. , Müntener O. , Ludwig T. ( 2008 ). The lithium, boron and beryllium content of serpentinized peridotites from ODP Leg 209 (Sites 1272A and 1274A): implications for lithium and boron budgets of oceanic lithosphere . Geochimica et Cosmochimica Acta 72 , 5475 – 5504 . Google Scholar Crossref Search ADS Vitale-Brovarone A. , Martinez I. , Elmaleh A. , Compagnoni R. , Chaduteau C. , Ferraris C. , Esteve I. ( 2017 ). Massive production of abiotic methane during subduction evidenced in metamorphosed ophicarbonates from the Italian Alps . Nature Communications 8 , doi: 10.1038/ncomms14134. Weyer S. , Ionov D. A. ( 2007 ). Partial melting and melt percolation in the mantle: the message from Fe isotopes . Earth and Planetary Science Letters 259 , 119 – 133 . Google Scholar Crossref Search ADS Weyer S. , Anbar A. D. , Brey G. P. , Munker C. , Mezger K. , Woodland A. B. ( 2005 ). Iron isotope fractionation during planetary differentiation . Earth and Planetary Science Letters 240 , 251 – 264 . Google Scholar Crossref Search ADS Williams H. M. , Bizimis M. ( 2014 ). Iron isotope tracing of mantle heterogeneity within the source regions of oceanic basalts . Earth and Planetary Science Letters 404 , 396 – 407 . Google Scholar Crossref Search ADS Williams H. M. , McCammon C. , Peslier A. H. , Halliday A. N. , Teutsch N. , Levasseur S. , Burg J.-P. ( 2004 ). Iron isotope fractionation and the oxygen fugacity of the mantle . Science (New York, N.Y.) 304 , 1656 – 1659 . Google Scholar Crossref Search ADS PubMed Williams H. M. , Nielsen S. G. , Renac C. , Griffin W. L. , O'Reilly S. Y. , McCammon C. A. , Pearson N. , Viljoen F. , Alt J. C. , Halliday A. N. ( 2009 ). Fractionation of oxygen and iron isotopes by partial melting processes: implications for the interpretation of stable isotope signatures in mafic rocks . Earth and Planetary Science Letters 283 , 156 – 166 . Google Scholar Crossref Search ADS Williams H. M. , Peslier A. H. , McCammon C. , Halliday A. N. , Levasseur S. , Teutsch N. , Burg J. P. ( 2005 ). Systematic iron isotope variations in mantle rocks and minerals: the effects of partial melting and oxygen fugacity . Earth and Planetary Science Letters 235 , 435 – 452 . Google Scholar Crossref Search ADS Wykes J. L. , Newton R. C. , Manning C. E. ( 2008 ). Solubility of andradite, Ca3Fe2Si3O12, in a 10 mol% NaCl solution at 800°C and 10 kbar: implications for the metasomatic origin of grandite garnet in calc-silicate granulites . American Mineralogist 93 , 886 – 892 . Google Scholar Crossref Search ADS © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
TI - Carbonate Transfer during the Onset of Slab Devolatilization: New Insights from Fe and Zn Stable Isotopes
JF - Journal of Petrology
DO - 10.1093/petrology/egy057
DA - 2018-06-01
UR - https://www.deepdyve.com/lp/oxford-university-press/carbonate-transfer-during-the-onset-of-slab-devolatilization-new-dPZA2VAGNj
SP - 1145
VL - 59
IS - 6
DP - DeepDyve
ER -