Redox dynamics of subduction revealed by arsenic in serpentinite
Affiliations | Corresponding Author | Cite as | Funding information- Share this article
-
Article views:501Cumulative count of HTML views and PDF downloads.
- Download Citation
- Rights & Permissions
top
Abstract
Figures
Figure 1 Geology and mineralogy of the Tso Morari serpentinite samples. (a) Geological map showing sampling locations. (b) Typical serpentinite sample (polarised light) showing dominant antigorite (Ant), metamorphic olivine (Ol) and subordinate magnetite (Mt). (c) and (d) Magnetite grains embedded in antigorite matrix. (e) and (f) Antigorite matrix with slight variations in Mg-Al-Si-Fe content and typical magnetite grain displaying Cr-rich core and resorption features (backscattered electron mode). See Supplementary Information for samples identity and characteristics. | Figure 2 Representative As K-edge XANES spectra acquired on whole rock serpentinite (Serp) and its antigorite (Ant) and magnetite (Mt) enriched fractions, showing contributions of different arsenic formal redox states derived by linear combination fits using different sets of reference solid and aqueous (aq) compounds. Vertical dashed lines indicate the energy positions of the identified arsenic redox states. The best match of all serpentinite spectra was provided by a combination of the spectra of nickeline and aqueous arsenious and arsenic acid species or AsIII and AsV adsorbed on Al/Fe-bearing mineral surfaces. In contrast, our spectra are poorly matched by that of arsenolite, displaying characteristic features arising from AsIII-O-AsIII bonds (at 11876 and 11884 eV), and that of AsV partially incorporated in magnetite and coordinated by multiple Fe atoms via AsV-O-Fe bonds (e.g., feature at 11882 eV). See Table S-2 for full dataset and reference compounds, Figure S-1 for EXAFS spectra, and Supplementary Information text for more detailed comparisons. | Figure 3 Thermodynamic simulations of serpentinisation in the system harzburgite-sediment-aqueous fluid (starting mass ratios 10:1:10, respectively), along the T-P subduction path from Guillot et al. (2008). (a) Equilibrium relative abundances of key minerals. (b) Predicted AsV/AsIII species ratio in the aqueous fluid (dotted blue) and fO2 (dashed red) assuming bulk system equilibrium, as compared to the range of fO2 in the fluid (lower and higher bounds, solid red) corresponding to the range of AsV/AsIII ratios measured in serpentinite (solid green). | Figure 4 Arsenic redox transformation reactions along the T vs. depth subduction path of the Tso Morari metamorphic rocks. Arrows show the directions of the subduction and subsequent exhumation. Ant = antigorite, Mt = magnetite, Ol = olivine. |
Figure 1 | Figure 2 | Figure 3 | Figure 4 |
top
Introduction
Serpentinite formation and breakdown are major phenomena occurring in subduction zones. Knowledge of the redox conditions (fO2) in these processes is necessary to interpret the transfers of many major and trace elements existing in multiple oxidation states. Most natural, experimental, and modelling studies have tackled redox evolution during subduction using major redox sensitive elements such as C, S, and Fe, but little attention has been devoted to trace elements. Among them, arsenic may be a promising redox indicator because it exhibits a wide range of formal oxidation states, from −III to +V, yielding a variety of minerals from (sulf)arsenides to arsenates, and oxyhydroxide AsIII and AsV species in fluids that may be scavenged or released by major minerals and silicate melts depending on fO2 (e.g., Noll et al., 1996
Noll, P.D. Jr., Newsom, H.E., Leeman, W.P., Ryan, J.R. (1996) The role of hydrothermal fluids in the production of subduction zone magmas: Evidence from siderophile and chalcophile trace elements and boron. Geochimica et Cosmochimica Acta 60, 587–611. https://doi.org/10.1016/0016-7037(95)00405-X
; O’Day, 2006O’Day, P.A. (2006) Chemistry and mineralogy of arsenic. Elements 2, 77–83. https://doi.org/10.2113/gselements.2.2.77
; Perfetti et al., 2008Perfetti, E., Pokrovski, G.S., Ballerat-Busserolles, K., Majer, V., Gibert, F. (2008) Densities and heat capacities of aqueous arsenious and arsenic acid solutions to 350 °C and 300 bar, and revised thermodynamic properties of As(OH)30(aq), AsO(OH)30(aq) and iron sulfarsenide minerals. Geochimica et Cosmochimica Acta 72, 713–731. https://doi.org/10.1016/j.gca.2007.11.017
; Borisova et al., 2010Borisova, A.Y., Pokrovski, G.S., Pichavant, M., Fredier, R., Candaudap, F. (2010) Arsenic enrichment in hydrous peraluminous melts: Insights from femtosecond laser ablation-inductively coupled plasma-quadrupole mass spectrometry, and in situ X-ray absorption fine structure spectroscopy. American Mineralogist 95, 1095–1104. https://doi.org/10.2138/am.2010.3424
; Testemale et al., 2011Testemale, D., Pokrovski, G.S., Hazemann, J.-L. (2011) Speciation of AsIII and AsV in hydrothermal fluids by in situ X-ray absorption spectroscopy. European Journal of Mineralogy 23, 379–390. https://doi.org/10.1127/0935-1221/2011/0023-2104
; Scambelluri et al., 2019Scambelluri, M., Cannaò, E., Gilio, M. (2019) The water and fluid-mobile element cycles during serpentinite subduction: A review. European Journal of Mineralogy 31, 405–428. https://doi.org/10.1127/ejm/2019/0031-2842
).The present study thus examines the potential of arsenic for tracing subduction zone redox conditions through measurement of arsenic oxidation state and speciation in the Tso Morari serpentinites (NW Himalaya). These rocks were formed by hydration of forearc mantle peridotites by slab-derived fluids and subducted to depth of ∼100 km and temperatures of ∼650 °C before having been exhumed during the Himalayan orogenesis (Hattori and Guillot, 2007
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, 1–27. https://doi.org/10.1029/2007GC001594
and references therein). The serpentinites, constituted mostly of antigorite and magnetite (Fig. 1), are highly enriched in AsV and AsIII (up to ∼100 ppm As, which is ∼1000 times more than in mantle-derived rocks; Hattori et al., 2005Hattori, K., Takahashi, Y., Guillot, S., Johanson, B. (2005) Occurrence of arsenic (V) in forearc mantle serpentinites based on X-ray absorption spectroscopy study. Geochimica et Cosmochimica Acta 69, 5585–5596. https://doi.org/10.1016/j.gca.2005.07.009
; Witt-Eickschen et al., 2009Witt-Eickschen, G., Palme, H., O’Neill, H.St.C., Allen, C.M. (2009) The geochemistry of the volatile trace elements As, Cd, Ge, In and Sn in the Earth’s mantle: New evidence from in situ analyses of mantle xenoliths. Geochimica et Cosmochimica Acta 73, 1755–1778. https://doi.org/10.1016/j.gca.2008.12.013
). Combined with the well constrained geodynamic history of the Himalayan subduction (Supplementary Information), these serpentinites represent an excellent natural case to examine the redox cycle of arsenic in a palaeo-subduction zone across a wide range of temperatures (T) and pressures (P).We used synchrotron X-ray absorption spectroscopy (XAS), which is the most direct method to probe a trace element redox state, chemical bonding and coordination at the atomic scale. Arsenic K-edge X-ray absorption near-edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) spectra were acquired on a thoroughly characterised set of serpentinite rock samples as well as their antigorite- and magnetite-enriched mineral fractions (Supplementary Information; Table S-1). Combined with simulations of fluid-rock interactions using robust thermodynamic data for As-bearing minerals and aqueous species, the results allow us to propose a novel redox model of arsenic behaviour enabling it to unveil fO2 dynamics in subduction processes.
top
Arsenic Redox and Structural State in Serpentinites
XANES spectra reveal a large range of arsenic redox forms coexisting in serpentinite samples (Fig. 2). Arsenate (AsVO4 tetrahedral coordination) is generally the dominant form both in serpentinite and its antigorite fraction (>95 % Ant) with an average of 55 ± 20 mol % (1 s.d.) of total As. The magnetite-enriched fraction (∼90 % Mt + 10 % Ant) contains even more AsV (from 85 to >95 mol %; Table S-2); however, it accounts for a minor part of AsV in the total As budget in serpentinite (Supplementary Information). Arsenate in both minerals coexists with arsenite (AsIIIO3 trigonal pyramidal coordination, from <5 to 55 mol % total As), as well as with arsenide (formal redox state As–III, <5 to 47 mol % total As) in some antigorite samples. EXAFS analyses confirm these findings by showing a mixture of AsIIIO3 and AsVO4 first shell coordinations for arsenide-free antigorite and magnetite (e.g., TSL16), and the simultaneous presence of Ni and O in the first atomic shell of As in antigorite samples containing both arsenide and arsenate, showing a mixture of As–IIINi6 and AsVO4 coordination environments (e.g., TLS19; Table S-3, Fig. S-1). Nickel arsenide nanoparticles were directly identified by transmission electron microscopy (TEM) in such samples whereas neither AsIII nor AsV individual solid phases were detected (Fig. S-2). The very weak (if any) arsenic second shell EXAFS signals, along with the lack of correlations of As redox state and total contents with those of Fe or other subordinate elements (e.g., Na, Al, V, Co, Ni) that commonly enter magnetite or antigorite structural sites, do not support substitution of AsIII and AsV in (Si,FeIII)O4 tetrahedral or (Mg,Al,FeII,III)O6 octahedral sites of both minerals (Supplementary Information). Indeed, the [AsIIIO3] geometry (Fig. S-1) is structurally incompatible with those sites. The [AsVO4]3– tetrahedron is also incompatible with octahedral sites in Mg/Fe layers and would induce strain and charge imbalance if substituted for smaller and higher charged [SiIVO4]4– in Si layers of antigorite. Therefore, our data collectively point to site unspecific intake of both AsIII and AsV, likely in structural imperfections common in between Mg-Si layers of antigorite, as oxyhydroxide anions of arsenious and arsenic acid similar to those dominant in aqueous solution (e.g., Perfetti et al., 2008
Perfetti, E., Pokrovski, G.S., Ballerat-Busserolles, K., Majer, V., Gibert, F. (2008) Densities and heat capacities of aqueous arsenious and arsenic acid solutions to 350 °C and 300 bar, and revised thermodynamic properties of As(OH)30(aq), AsO(OH)30(aq) and iron sulfarsenide minerals. Geochimica et Cosmochimica Acta 72, 713–731. https://doi.org/10.1016/j.gca.2007.11.017
; Testemale et al., 2011Testemale, D., Pokrovski, G.S., Hazemann, J.-L. (2011) Speciation of AsIII and AsV in hydrothermal fluids by in situ X-ray absorption spectroscopy. European Journal of Mineralogy 23, 379–390. https://doi.org/10.1127/0935-1221/2011/0023-2104
).top
Arsenic Redox Cycle during Subduction
The presence of contrasting arsenic oxidation states revealed in this study strongly suggests that the serpentinites have undergone large redox changes in T-P-time space during subduction. Using equilibrium thermodynamics, we simulated arsenic speciation and solubility during interactions of hydrous sediments (pelite ± seawater) with the mantle wedge (harzburgite) in variable proportions, followed by subduction of the produced serpentinite either in closed or open (i.e. partial fluid loss) systems (Fig. 3, Tables S-5, S-6), along the geothermal path established for the Tso Morari metamorphic rocks (Guillot et al., 2008
Guillot, S., Mahéo, G., de Sigoyer, J., Hattori, K.H., Pêcher, A. (2008) Tethyan and Indian subduction viewed from the Himalayan high- to ultrahigh-pressure metamorphic rocks. Tectonophysics 451, 225–241. https://doi.org/10.1016/j.tecto.2007.11.059
).We found that, independently of the initial redox state of arsenic, introduced with sediments or water either as sulfarsenide or arsenate, poorly soluble nickel arsenides such as orcelite (Ni5As2) and maucherite (Ni8As11) are the stable phases across a wide range of water/rock and sediment/harzburgite ratios (0.1–5 and 0.1–0.5), from early stages of serpentinisation (<200 °C) to at least 550 °C (Fig. 3). This T range corresponds to fO2 of –7 to –3 log units relative to the conventional fayalite-magnetite-quartz buffer (FMQ–7 to –3), as imposed by reactions of FeII-bearing olivine and pyroxene with water producing FeII/FeIII-bearing antigorite and magnetite while consuming oxygen (Fig. 4). The predicted aqueous arsenic concentrations are <1 ppb below 500 °C (Fig. S-5a), demonstrating that arsenic was not a fluid mobile element and was hosted by Ni arsenide phases, in full agreement with the XAS and TEM analyses (Figs. 2, S-2). Highly reducing conditions inherent to this subduction step are also supported by findings of methane-bearing inclusions in partly serpentinised olivine of the Nidar ophiolite complex adjacent to Tso Morari (Fig. 1a; Sachan et al., 2007
Sachan, H.K., Mukherjee, B.K., Bodnar, R.J. (2007) Preservation of methane generated during serpentinization of upper mantle rocks: Evidence from fluid inclusions in the Nidar ophiolite, Indus Suture Zone, Ladakh (India). Earth and Planetary Science Letters 257, 47–59. https://doi.org/10.1016/j.epsl.2007.02.023
). Reducing environments at the serpentinisation step are equally common in other subduction settings, for instance as evidenced by Fe-Ni arsenides in Kamchatka peridotite xenoliths altered by slab-derived fluids (Ishimaru and Arai, 2008Ishimaru, S., Arai, S. (2008) Arsenide in a metasomatized peridotite xenolith as a constraint on arsenic behavior in the mantle wedge. American Mineralogist 93, 1061–1065. https://doi.org/10.2138/am.2008.2746
), and by graphite in the Alpine blueschist-to-eclogite facies (Malvoisin et al., 2012Malvoisin, B., Chopin, C., Brunet, F., Galvez, M.E. (2012) Low-temperature wollastonite formed by carbonate reduction: a marker of serpentinite redox conditions. Journal of Petrology 53, 159–176. https://doi.org/10.1093/petrology/egr060
). It is only above 550 °C, with onset of partial breakdown of antigorite and magnetite to olivine, that the solubility of the arsenide phases in the fluid as dominantly AsIII oxyhydroxide species does significantly increase (Fig. S-5a). This simulated scenario is consistent with the textures and compositional variations of magnetite observed in our samples likely reflecting recrystallisation/replacement phenomena (Fig. 1f) as well as the lack of arsenide phases in some of our samples (Figs. 2, S-2).Our modelling does not support the hypothesis of early intake of AsV by serpentinite (Hattori et al., 2005
Hattori, K., Takahashi, Y., Guillot, S., Johanson, B. (2005) Occurrence of arsenic (V) in forearc mantle serpentinites based on X-ray absorption spectroscopy study. Geochimica et Cosmochimica Acta 69, 5585–5596. https://doi.org/10.1016/j.gca.2005.07.009
), which is inconsistent with low fO2 values at early serpentinisation steps (<FMQ–3; Figs. 3, S-5). Such highly reducing conditions are also supported by ubiquitous experimental evidence of H2 production upon ultramafic rock hydration as well as the occurrence of reduced accessory minerals (sulfarsenides, Fe-Ni alloys) in serpentinisation reactions both in nature and laboratory (e.g., Frost, 1985Frost, B.R. (1985) On the stability of sulfides, oxides and native metals in serpentinite. Journal of Petrology 26, 31–63. https://doi.org/10.1093/petrology/26.1.31
; McCollom and Bach, 2009McCollom, T.M., Bach, W. (2009) Thermodynamic constraints on hydrogen generation during serpentinization of ultramafic rocks. Geochimica et Cosmochimica Acta 73, 856–875. https://doi.org/10.1016/j.gca.2008.10.032
; Marcaillou et al., 2011Marcaillou, C., Muñoz, M., Vidal, O., Parra, T., Harfouche, M. (2011) Mineralogical evidence for H2 degassing during serpentinization at 300 °C/300 bar. Earth and Planetary Science Letters 303, 281–290. https://doi.org/10.1016/j.epsl.2011.01.006
; González-Jiménez et al., 2021González-Jiménez, J.M., Piña, R., Saunders, J.E., Plissart, G., Marchesi, C., Padrón-Navarta, J.A., Ramón-Fernandez, M., Garrido, L.N.F., Gervilla, F. (2021) Trace element fingerprints of Ni–Fe–S–As minerals in subduction channel serpentinites. Lithos 400–401, 106432. https://doi.org/10.1016/j.lithos.2021.106432
). Oxidation to AsV during exhumation is equally unlikely because it would have required either influx of extremely oxidising fluids (i.e. fO2 ≈ FMQ+10) not evidenced during the inferred hydration events (Palin et al., 2014Palin, R.M., St-Onge, M.R., Waters, D.J., Searle, M.P., Dyck, B. (2014) Phase equilibria modelling of retrograde amphibole and clinozoisite in mafic eclogite from the Tso Morari massif, northwest India: constraining the P–T–M (H2O) conditions of exhumation. Journal of Metamorphic Geology 32, 675–693. https://doi.org/10.1111/jmg.12085
), or direct near surface oxidation and weathering. Both phenomena would have produced hematite/goethite and clays not observed in the Tso Morari rocks (Hattori and Guillot, 2007Hattori, 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, 1–27. https://doi.org/10.1029/2007GC001594
; Deschamps et al., 2010Deschamps, F., Guillot, S., Godard, M., Chauvel, C., Andreani, M., Hattori, K.H. (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. https://doi.org/10.1016/j.chemgeo.2009.10.002
). Therefore, arsenic redox captured by serpentinite likely reflects subduction phenomena at depth.top
Release of Highly Oxidised Fluids in Subduction Zones Revealed by Arsenic Redox
In light of our results demonstrating the absence of selective site specific AsV and AsIII incorporation in serpentine, no significant AsV vs. AsIII fractionation would be induced upon arsenic intake, thereby making the AsV/AsIII ratio in the mineral to be representative of that in the coexisting aqueous fluid. The thermodynamically predicted AsV/AsIII ratios (10–6 to 10–3) in the fluid phase in the 550–650 °C range of the deserpentinisation step under redox conditions of bulk system equilibrium are, however, much lower than the ratios measured in our antigorite and magnetite-enriched samples (from 0.4 to >11; Fig. 3b, Table S-1). Assuming that all dissolved species including AsV and AsIII rapidly re-equilibrate in aqueous solution at such elevated T, the measured AsV/AsIII ratios in antigorite would correspond to fO2 values of FMQ+8 to +12 in the fluid, which are ∼10 log units higher than those calculated in the bulk fluid-mineral system at equilibrium (∼FMQ; Fig. 4). This difference is beyond reasonable error margins of our fO2 estimates (±2 log units; Supplementary Information), and therefore reflects out-of-equilibrium oxidised fluid release during partial serpentinite decomposition, according to the formal reactions shown in Figure 4. Note that even though our fO2 estimates are close to mbar levels of partial PO2, the equivalent concentration of the aqueous molecular O20 species at these T-P conditions is only ∼10–10 molal. Therefore, the oxidation potential of the released fluid is mostly carried by more concentrated dissolved elements in their highest oxidation states, such as AsV, FeIII and SVI and, possibly, by reactive oxygen species like O2•− or H2O2. However, given the large predominance of both FeII and FeIII in serpentinite rock, the amount of all dissolved redox sensitive elements (Fe, As, S, O) in the fluid (Fig. S-5) would be insufficient to significantly alter the Fe redox ratio of the major solid phases and, consequently, to leave clearly detectable Fe redox imprint in the rock. Locally focused oxidised fluid release in the proximity of resorbing magnetite grains (Fig. 1) would also explain the more elevated AsV/AsIII ratios (∼10) found in magnetite-enriched fractions than in antigorite fractions (∼1).
The fO2 values derived from arsenic redox are significantly higher than those estimated based on Fe oxy-thermobarometry of metamorphic minerals associated with deep subduction (e.g., garnet, spinel, pyroxene, amphibole; Cannaò and Malaspina, 2018
Cannaò, E., Malaspina, N. (2018) From oceanic to continental subduction: Implications for the geochemical and redox evolution of the supra-subduction mantle. Geosphere 14, 2311–2336. https://doi.org/10.1130/GES01597.1
; Gerrits et al., 2019Gerrits, A.R., Inglis, E.C., Dragovic, B., Starr, P.G., Baxter, E.F., Burton, K.W. (2019) Release of oxidizing fluids in subduction zones recorded by iron isotope zonation in garnet. Nature Geoscience 12, 1029–1033. https://doi.org/10.1038/s41561-019-0471-y
), equilibrium thermodynamic simulations of fluid-rock interactions in this and previous work (e.g., Debret and Sverjensky, 2017Debret, B., Sverjensky, D.A. (2017) Highly oxidising fluids generated during serpentinite breakdown in subduction zones. Scientific Reports 7, 10351. https://doi.org/10.1038/s41598-017-09626-y
; Piccoli et al., 2019Piccoli, F., Hermann, J., Pettke, T., Connoly, J.A.D., Kempf, E.D., Vieira Duarte, J.F. (2019) Subducting serpentinites release reduced, not oxidized, aqueous fluids. Scientific Reports 9, 19573. https://doi.org/10.1038/s41598-019-55944-8
; Evans and Frost, 2021Evans, K.A., Frost, B.R. (2021) Deserpentinization in subduction zones as a source of oxidation in arcs: a reality check. Journal of Petrology 62, egab016. https://doi.org/10.1093/petrology/egab016
), and antigorite dehydration experiments (Maurice et al., 2020Maurice, J., Bolfan-Casanova, N., Demouchy, S., Chauvigne, P., Schiavi, F., Debret, B. (2020) The intrinsic nature of antigorite breakdown at 3 GPa: Experimental constraints on redox conditions of serpentinite dehydration in subduction zones. Contributions to Mineralogy and Petrology 175, 94. https://doi.org/10.1007/s00410-020-01731-y
). Their fO2 estimations commonly range from FMQ to FMQ+5, with an upper limit being close to the hematite-magnetite equilibrium. Our arsenic-derived fO2 values of FMQ+8 to +12 in the absence of hematite thus clearly indicate out-of-equilibrium release of highly oxidised fluids that is not captured by iron as the overwhelmingly dominant redox sensitive element in serpentinite. Interestingly, comparably high fO2 values were inferred from manganese-bearing metacherts in subduction mélanges (Tumiati et al., 2015Tumiati, 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. https://doi.org/10.1016/j.lithos.2014.12.008
), pointing to potentially wide occurrence of phenomena of oxidised fluid generation.Our findings thus provide novel insight into the redox dynamics and geochemical cycles in subduction zones. Like for arsenic, the transfer of other metalloids (e.g., P, Sb, Se, Te) in subduction zones may be fundamentally controlled by fO2, spanning over >15 orders of magnitude relative to FMQ across subduction, from early reducing conditions stabilising poorly soluble Fe/Ni phases of these elements, to highly oxidising conditions promoting their soluble oxyanions. More generally, our results highlight an important divide between the redox potential of a bulk serpentine rock (Tumiati et al., 2015
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. https://doi.org/10.1016/j.lithos.2014.12.008
; Evans and Frost, 2021Evans, K.A., Frost, B.R. (2021) Deserpentinization in subduction zones as a source of oxidation in arcs: a reality check. Journal of Petrology 62, egab016. https://doi.org/10.1093/petrology/egab016
; Galvez and Jaccard, 2021Galvez, M.E., Jaccard, S.L. (2021) Redox capacity of rocks and sediments by high temperature chalcometric titration. Chemical Geology 564, 120016. https://doi.org/10.1016/j.chemgeo.2020.120016
) and that of the out-of-equilibrium aqueous fluid phase generated therefrom. Such highly oxidised fluids may selectively carry major redox sensitive elements in their highest valence states (FeIII, MnIV, sulfate, carbonate), in contrast to equilibrium thermodynamic predictions in rock buffered systems that suggest reduced valence states of these elements to be equally (or more) abundant in the fluid. Our findings may thus provide new constrains on the speciation and transfer of metals and volatiles and their associated stable isotope signatures (e.g., Debret et al., 2016Debret, B., Millet, M.-A., Pons, M.-L., Bouilhol, P., Inglis, E., Williams, H. (2016) Isotopic evidence of iron mobility during subduction. Geology 44, 215–218. https://doi.org/10.1130/G37565.1
; Walters et al., 2019Walters, J.B, Cruz-Uribe, A.M., Marschall, H.R. (2019) Isotopic compositions of sulfides in exhumed high-pressure terranes: Implications for sulfur cycling in subduction zones. Geochemistry, Geophysics, Geosystems 20, 3347–3374. https://doi.org/10.1029/2019GC008374
) and, more globally, on the dynamics of redox reactions at depth.top
Acknowledgements
This work was funded by the Institut des Sciences de l’Univers of the Centre National de la Recherche Scientifique (INSU ASEDIT), the Partenariat Hubert Curien (PHC Germaine de Staël), the Agence Nationale de la Recherche (RadicalS ANR-16-CE31-0017), and the Institut Carnot (ISIFoR OrPet and AsCOCrit). The European Synchrotron Radiation Facility (ESRF) provided access to beam time and infrastructure. The FAME facility is supported by the CEA-CNRS-CRG consortium and the INSU. We thank A.-M. Cousin for help with figure layout, F. Maube and L. Menjot for SEM and XRD analyses, G. Morin, J. Rose, A. Foster, and C. van Genuchten for sharing XAS references, M. Galvez, M. Blanchard, M. Brounce and an anonymous reviewer for comments, and S. Myneni for editorial handling.
Editor: Satish Myneni
top
References
Borisova, A.Y., Pokrovski, G.S., Pichavant, M., Fredier, R., Candaudap, F. (2010) Arsenic enrichment in hydrous peraluminous melts: Insights from femtosecond laser ablation-inductively coupled plasma-quadrupole mass spectrometry, and in situ X-ray absorption fine structure spectroscopy. American Mineralogist 95, 1095–1104. https://doi.org/10.2138/am.2010.3424
Show in context
Among them, arsenic may be a promising redox indicator because it exhibits a wide range of formal oxidation states, from −III to +V, yielding a variety of minerals from (sulf)arsenides to arsenates, and oxyhydroxide AsIII and AsV species in fluids that may be scavenged or released by major minerals and silicate melts depending on fO2 (e.g., Noll et al., 1996; O’Day, 2006; Perfetti et al., 2008; Borisova et al., 2010; Testemale et al., 2011; Scambelluri et al., 2019).
View in article
Cannaò, E., Malaspina, N. (2018) From oceanic to continental subduction: Implications for the geochemical and redox evolution of the supra-subduction mantle. Geosphere 14, 2311–2336. https://doi.org/10.1130/GES01597.1
Show in context
The fO2 values derived from arsenic redox are significantly higher than those estimated based on Fe oxy-thermobarometry of metamorphic minerals associated with deep subduction (e.g., garnet, spinel, pyroxene, amphibole; Cannaò and Malaspina, 2018; Gerrits et al., 2019), equilibrium thermodynamic simulations of fluid-rock interactions in this and previous work (e.g., Debret and Sverjensky, 2017; Piccoli et al., 2019; Evans and Frost, 2021), and antigorite dehydration experiments (Maurice et al., 2020).
View in article
Debret, B., Sverjensky, D.A. (2017) Highly oxidising fluids generated during serpentinite breakdown in subduction zones. Scientific Reports 7, 10351. https://doi.org/10.1038/s41598-017-09626-y
Show in context
The fO2 values derived from arsenic redox are significantly higher than those estimated based on Fe oxy-thermobarometry of metamorphic minerals associated with deep subduction (e.g., garnet, spinel, pyroxene, amphibole; Cannaò and Malaspina, 2018; Gerrits et al., 2019), equilibrium thermodynamic simulations of fluid-rock interactions in this and previous work (e.g., Debret and Sverjensky, 2017; Piccoli et al., 2019; Evans and Frost, 2021), and antigorite dehydration experiments (Maurice et al., 2020).
View in article
Debret, B., Millet, M.-A., Pons, M.-L., Bouilhol, P., Inglis, E., Williams, H. (2016) Isotopic evidence of iron mobility during subduction. Geology 44, 215–218. https://doi.org/10.1130/G37565.1
Show in context
Our findings may thus provide new constrains on the speciation and transfer of metals and volatiles and their associated stable isotope signatures (e.g., Debret et al., 2016; Walters et al., 2019) and, more globally, on the dynamics of redox reactions at depth.
View in article
Deschamps, F., Guillot, S., Godard, M., Chauvel, C., Andreani, M., Hattori, K.H. (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. https://doi.org/10.1016/j.chemgeo.2009.10.002
Show in context
Oxidation to AsV during exhumation is equally unlikely because it would have required either influx of extremely oxidising fluids (i.e. fO2 ≈ FMQ+10) not evidenced during the inferred hydration events (Palin et al., 2014), or direct near surface oxidation and weathering. Both phenomena would have produced hematite/goethite and clays not observed in the Tso Morari rocks (Hattori and Guillot, 2007; Deschamps et al., 2010).
View in article
Evans, K.A., Frost, B.R. (2021) Deserpentinization in subduction zones as a source of oxidation in arcs: a reality check. Journal of Petrology 62, egab016. https://doi.org/10.1093/petrology/egab016
Show in context
The fO2 values derived from arsenic redox are significantly higher than those estimated based on Fe oxy-thermobarometry of metamorphic minerals associated with deep subduction (e.g., garnet, spinel, pyroxene, amphibole; Cannaò and Malaspina, 2018; Gerrits et al., 2019), equilibrium thermodynamic simulations of fluid-rock interactions in this and previous work (e.g., Debret and Sverjensky, 2017; Piccoli et al., 2019; Evans and Frost, 2021), and antigorite dehydration experiments (Maurice et al., 2020).
View in article
More generally, our results highlight an important divide between the redox potential of a bulk serpentine rock (Tumiati et al., 2015; Evans and Frost, 2021; Galvez and Jaccard, 2021) and that of the out-of-equilibrium aqueous fluid phase generated therefrom.
View in article
Frost, B.R. (1985) On the stability of sulfides, oxides and native metals in serpentinite. Journal of Petrology 26, 31–63. https://doi.org/10.1093/petrology/26.1.31
Show in context
Such highly reducing conditions are also supported by ubiquitous experimental evidence of H2 production upon ultramafic rock hydration as well as the occurrence of reduced accessory minerals (sulfarsenides, Fe-Ni alloys) in serpentinisation reactions both in nature and laboratory (e.g., Frost, 1985; McCollom and Bach, 2009; Marcaillou et al., 2011; González-Jiménez et al., 2021).
View in article
Galvez, M.E., Jaccard, S.L. (2021) Redox capacity of rocks and sediments by high temperature chalcometric titration. Chemical Geology 564, 120016. https://doi.org/10.1016/j.chemgeo.2020.120016
Show in context
More generally, our results highlight an important divide between the redox potential of a bulk serpentine rock (Tumiati et al., 2015; Evans and Frost, 2021; Galvez and Jaccard, 2021) and that of the out-of-equilibrium aqueous fluid phase generated therefrom.
View in article
Gerrits, A.R., Inglis, E.C., Dragovic, B., Starr, P.G., Baxter, E.F., Burton, K.W. (2019) Release of oxidizing fluids in subduction zones recorded by iron isotope zonation in garnet. Nature Geoscience 12, 1029–1033. https://doi.org/10.1038/s41561-019-0471-y
Show in context
The fO2 values derived from arsenic redox are significantly higher than those estimated based on Fe oxy-thermobarometry of metamorphic minerals associated with deep subduction (e.g., garnet, spinel, pyroxene, amphibole; Cannaò and Malaspina, 2018; Gerrits et al., 2019), equilibrium thermodynamic simulations of fluid-rock interactions in this and previous work (e.g., Debret and Sverjensky, 2017; Piccoli et al., 2019; Evans and Frost, 2021), and antigorite dehydration experiments (Maurice et al., 2020).
View in article
González-Jiménez, J.M., Piña, R., Saunders, J.E., Plissart, G., Marchesi, C., Padrón-Navarta, J.A., Ramón-Fernandez, M., Garrido, L.N.F., Gervilla, F. (2021) Trace element fingerprints of Ni–Fe–S–As minerals in subduction channel serpentinites. Lithos 400–401, 106432. https://doi.org/10.1016/j.lithos.2021.106432
Show in context
Such highly reducing conditions are also supported by ubiquitous experimental evidence of H2 production upon ultramafic rock hydration as well as the occurrence of reduced accessory minerals (sulfarsenides, Fe-Ni alloys) in serpentinisation reactions both in nature and laboratory (e.g., Frost, 1985; McCollom and Bach, 2009; Marcaillou et al., 2011; González-Jiménez et al., 2021).
View in article
Guillot, S., Mahéo, G., de Sigoyer, J., Hattori, K.H., Pêcher, A. (2008) Tethyan and Indian subduction viewed from the Himalayan high- to ultrahigh-pressure metamorphic rocks. Tectonophysics 451, 225–241. https://doi.org/10.1016/j.tecto.2007.11.059
Show in context
Using equilibrium thermodynamics, we simulated arsenic speciation and solubility during interactions of hydrous sediments (pelite ± seawater) with the mantle wedge (harzburgite) in variable proportions, followed by subduction of the produced serpentinite either in closed or open (i.e. partial fluid loss) systems (Fig. 3, Tables S-5, S-6), along the geothermal path established for the Tso Morari metamorphic rocks (Guillot et al., 2008).
View in article
Thermodynamic simulations of serpentinisation in the system harzburgite-sediment-aqueous fluid (starting mass ratios 10:1:10, respectively), along the T-P subduction path from Guillot et al. (2008).
View in article
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, 1–27. https://doi.org/10.1029/2007GC001594
Show in context
These rocks were formed by hydration of forearc mantle peridotites by slab-derived fluids and subducted to depth of ∼100 km and temperatures of ∼650 °C before having been exhumed during the Himalayan orogenesis (Hattori and Guillot, 2007 and references therein).
View in article
Oxidation to AsV during exhumation is equally unlikely because it would have required either influx of extremely oxidising fluids (i.e. fO2 ≈ FMQ+10) not evidenced during the inferred hydration events (Palin et al., 2014), or direct near surface oxidation and weathering. Both phenomena would have produced hematite/goethite and clays not observed in the Tso Morari rocks (Hattori and Guillot, 2007; Deschamps et al., 2010).
View in article
Hattori, K., Takahashi, Y., Guillot, S., Johanson, B. (2005) Occurrence of arsenic (V) in forearc mantle serpentinites based on X-ray absorption spectroscopy study. Geochimica et Cosmochimica Acta 69, 5585–5596. https://doi.org/10.1016/j.gca.2005.07.009
Show in context
Our modelling does not support the hypothesis of early intake of AsV by serpentinite (Hattori et al., 2005), which is inconsistent with low fO2 values at early serpentinisation steps (<FMQ–3; Figs. 3, S-5).
View in article
The serpentinites, constituted mostly of antigorite and magnetite (Fig. 1), are highly enriched in AsV and AsIII (up to ∼100 ppm As, which is ∼1000 times more than in mantle-derived rocks; Hattori et al., 2005; Witt-Eickschen et al., 2009).
View in article
Ishimaru, S., Arai, S. (2008) Arsenide in a metasomatized peridotite xenolith as a constraint on arsenic behavior in the mantle wedge. American Mineralogist 93, 1061–1065. https://doi.org/10.2138/am.2008.2746
Show in context
Reducing environments at the serpentinisation step are equally common in other subduction settings, for instance as evidenced by Fe-Ni arsenides in Kamchatka peridotite xenoliths altered by slab-derived fluids (Ishimaru and Arai, 2008), and by graphite in the Alpine blueschist-to-eclogite facies (Malvoisin et al., 2012).
View in article
Malvoisin, B., Chopin, C., Brunet, F., Galvez, M.E. (2012) Low-temperature wollastonite formed by carbonate reduction: a marker of serpentinite redox conditions. Journal of Petrology 53, 159–176. https://doi.org/10.1093/petrology/egr060
Show in context
Reducing environments at the serpentinisation step are equally common in other subduction settings, for instance as evidenced by Fe-Ni arsenides in Kamchatka peridotite xenoliths altered by slab-derived fluids (Ishimaru and Arai, 2008), and by graphite in the Alpine blueschist-to-eclogite facies (Malvoisin et al., 2012).
View in article
Marcaillou, C., Muñoz, M., Vidal, O., Parra, T., Harfouche, M. (2011) Mineralogical evidence for H2 degassing during serpentinization at 300 °C/300 bar. Earth and Planetary Science Letters 303, 281–290. https://doi.org/10.1016/j.epsl.2011.01.006
Show in context
Such highly reducing conditions are also supported by ubiquitous experimental evidence of H2 production upon ultramafic rock hydration as well as the occurrence of reduced accessory minerals (sulfarsenides, Fe-Ni alloys) in serpentinisation reactions both in nature and laboratory (e.g., Frost, 1985; McCollom and Bach, 2009; Marcaillou et al., 2011; González-Jiménez et al., 2021).
View in article
Maurice, J., Bolfan-Casanova, N., Demouchy, S., Chauvigne, P., Schiavi, F., Debret, B. (2020) The intrinsic nature of antigorite breakdown at 3 GPa: Experimental constraints on redox conditions of serpentinite dehydration in subduction zones. Contributions to Mineralogy and Petrology 175, 94. https://doi.org/10.1007/s00410-020-01731-y
Show in context
The fO2 values derived from arsenic redox are significantly higher than those estimated based on Fe oxy-thermobarometry of metamorphic minerals associated with deep subduction (e.g., garnet, spinel, pyroxene, amphibole; Cannaò and Malaspina, 2018; Gerrits et al., 2019), equilibrium thermodynamic simulations of fluid-rock interactions in this and previous work (e.g., Debret and Sverjensky, 2017; Piccoli et al., 2019; Evans and Frost, 2021), and antigorite dehydration experiments (Maurice et al., 2020).
View in article
McCollom, T.M., Bach, W. (2009) Thermodynamic constraints on hydrogen generation during serpentinization of ultramafic rocks. Geochimica et Cosmochimica Acta 73, 856–875. https://doi.org/10.1016/j.gca.2008.10.032
Show in context
Such highly reducing conditions are also supported by ubiquitous experimental evidence of H2 production upon ultramafic rock hydration as well as the occurrence of reduced accessory minerals (sulfarsenides, Fe-Ni alloys) in serpentinisation reactions both in nature and laboratory (e.g., Frost, 1985; McCollom and Bach, 2009; Marcaillou et al., 2011; González-Jiménez et al., 2021).
View in article
Noll, P.D. Jr., Newsom, H.E., Leeman, W.P., Ryan, J.R. (1996) The role of hydrothermal fluids in the production of subduction zone magmas: Evidence from siderophile and chalcophile trace elements and boron. Geochimica et Cosmochimica Acta 60, 587–611. https://doi.org/10.1016/0016-7037(95)00405-X
Show in context
Among them, arsenic may be a promising redox indicator because it exhibits a wide range of formal oxidation states, from −III to +V, yielding a variety of minerals from (sulf)arsenides to arsenates, and oxyhydroxide AsIII and AsV species in fluids that may be scavenged or released by major minerals and silicate melts depending on fO2 (e.g., Noll et al., 1996; O’Day, 2006; Perfetti et al., 2008; Borisova et al., 2010; Testemale et al., 2011; Scambelluri et al., 2019).
View in article
O’Day, P.A. (2006) Chemistry and mineralogy of arsenic. Elements 2, 77–83. https://doi.org/10.2113/gselements.2.2.77
Show in context
Among them, arsenic may be a promising redox indicator because it exhibits a wide range of formal oxidation states, from −III to +V, yielding a variety of minerals from (sulf)arsenides to arsenates, and oxyhydroxide AsIII and AsV species in fluids that may be scavenged or released by major minerals and silicate melts depending on fO2 (e.g., Noll et al., 1996; O’Day, 2006; Perfetti et al., 2008; Borisova et al., 2010; Testemale et al., 2011; Scambelluri et al., 2019).
View in article
Palin, R.M., St-Onge, M.R., Waters, D.J., Searle, M.P., Dyck, B. (2014) Phase equilibria modelling of retrograde amphibole and clinozoisite in mafic eclogite from the Tso Morari massif, northwest India: constraining the P–T–M (H2O) conditions of exhumation. Journal of Metamorphic Geology 32, 675–693. https://doi.org/10.1111/jmg.12085
Show in context
Oxidation to AsV during exhumation is equally unlikely because it would have required either influx of extremely oxidising fluids (i.e. fO2 ≈ FMQ+10) not evidenced during the inferred hydration events (Palin et al., 2014), or direct near surface oxidation and weathering. Both phenomena would have produced hematite/goethite and clays not observed in the Tso Morari rocks (Hattori and Guillot, 2007; Deschamps et al., 2010).
View in article
Perfetti, E., Pokrovski, G.S., Ballerat-Busserolles, K., Majer, V., Gibert, F. (2008) Densities and heat capacities of aqueous arsenious and arsenic acid solutions to 350 °C and 300 bar, and revised thermodynamic properties of As(OH)30(aq), AsO(OH)30(aq) and iron sulfarsenide minerals. Geochimica et Cosmochimica Acta 72, 713–731. https://doi.org/10.1016/j.gca.2007.11.017
Show in context
Among them, arsenic may be a promising redox indicator because it exhibits a wide range of formal oxidation states, from −III to +V, yielding a variety of minerals from (sulf)arsenides to arsenates, and oxyhydroxide AsIII and AsV species in fluids that may be scavenged or released by major minerals and silicate melts depending on fO2 (e.g., Noll et al., 1996; O’Day, 2006; Perfetti et al., 2008; Borisova et al., 2010; Testemale et al., 2011; Scambelluri et al., 2019).
View in article
Therefore, our data collectively point to site unspecific intake of both AsIII and AsV, likely in structural imperfections common in between Mg-Si layers of antigorite, as oxyhydroxide anions of arsenious and arsenic acid similar to those dominant in aqueous solution (e.g., Perfetti et al., 2008; Testemale et al., 2011).
View in article
Piccoli, F., Hermann, J., Pettke, T., Connoly, J.A.D., Kempf, E.D., Vieira Duarte, J.F. (2019) Subducting serpentinites release reduced, not oxidized, aqueous fluids. Scientific Reports 9, 19573. https://doi.org/10.1038/s41598-019-55944-8
Show in context
The fO2 values derived from arsenic redox are significantly higher than those estimated based on Fe oxy-thermobarometry of metamorphic minerals associated with deep subduction (e.g., garnet, spinel, pyroxene, amphibole; Cannaò and Malaspina, 2018; Gerrits et al., 2019), equilibrium thermodynamic simulations of fluid-rock interactions in this and previous work (e.g., Debret and Sverjensky, 2017; Piccoli et al., 2019; Evans and Frost, 2021), and antigorite dehydration experiments (Maurice et al., 2020).
View in article
Sachan, H.K., Mukherjee, B.K., Bodnar, R.J. (2007) Preservation of methane generated during serpentinization of upper mantle rocks: Evidence from fluid inclusions in the Nidar ophiolite, Indus Suture Zone, Ladakh (India). Earth and Planetary Science Letters 257, 47–59. https://doi.org/10.1016/j.epsl.2007.02.023
Show in context
Highly reducing conditions inherent to this subduction step are also supported by findings of methane-bearing inclusions in partly serpentinised olivine of the Nidar ophiolite complex adjacent to Tso Morari (Fig. 1a; Sachan et al., 2007).
View in article
Scambelluri, M., Cannaò, E., Gilio, M. (2019) The water and fluid-mobile element cycles during serpentinite subduction: A review. European Journal of Mineralogy 31, 405–428. https://doi.org/10.1127/ejm/2019/0031-2842
Show in context
Among them, arsenic may be a promising redox indicator because it exhibits a wide range of formal oxidation states, from −III to +V, yielding a variety of minerals from (sulf)arsenides to arsenates, and oxyhydroxide AsIII and AsV species in fluids that may be scavenged or released by major minerals and silicate melts depending on fO2 (e.g., Noll et al., 1996; O’Day, 2006; Perfetti et al., 2008; Borisova et al., 2010; Testemale et al., 2011; Scambelluri et al., 2019).
View in article
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. https://doi.org/10.1016/j.lithos.2014.12.008
Show in context
Their fO2 estimations commonly range from FMQ to FMQ+5, with an upper limit being close to the hematite-magnetite equilibrium. Our arsenic-derived fO2 values of FMQ+8 to +12 in the absence of hematite thus clearly indicate out-of-equilibrium release of highly oxidised fluids that is not captured by iron as the overwhelmingly dominant redox sensitive element in serpentinite. Interestingly, comparably high fO2 values were inferred from manganese-bearing metacherts in subduction mélanges (Tumiati et al., 2015), pointing to potentially wide occurrence of phenomena of oxidised fluid generation.
View in article
More generally, our results highlight an important divide between the redox potential of a bulk serpentine rock (Tumiati et al., 2015; Evans and Frost, 2021; Galvez and Jaccard, 2021) and that of the out-of-equilibrium aqueous fluid phase generated therefrom.
View in article
Testemale, D., Pokrovski, G.S., Hazemann, J.-L. (2011) Speciation of AsIII and AsV in hydrothermal fluids by in situ X-ray absorption spectroscopy. European Journal of Mineralogy 23, 379–390. https://doi.org/10.1127/0935-1221/2011/0023-2104
Show in context
Among them, arsenic may be a promising redox indicator because it exhibits a wide range of formal oxidation states, from −III to +V, yielding a variety of minerals from (sulf)arsenides to arsenates, and oxyhydroxide AsIII and AsV species in fluids that may be scavenged or released by major minerals and silicate melts depending on fO2 (e.g., Noll et al., 1996; O’Day, 2006; Perfetti et al., 2008; Borisova et al., 2010; Testemale et al., 2011; Scambelluri et al., 2019).
View in article
Therefore, our data collectively point to site unspecific intake of both AsIII and AsV, likely in structural imperfections common in between Mg-Si layers of antigorite, as oxyhydroxide anions of arsenious and arsenic acid similar to those dominant in aqueous solution (e.g., Perfetti et al., 2008; Testemale et al., 2011).
View in article
Walters, J.B, Cruz-Uribe, A.M., Marschall, H.R. (2019) Isotopic compositions of sulfides in exhumed high-pressure terranes: Implications for sulfur cycling in subduction zones. Geochemistry, Geophysics, Geosystems 20, 3347–3374. https://doi.org/10.1029/2019GC008374
Show in context
Our findings may thus provide new constrains on the speciation and transfer of metals and volatiles and their associated stable isotope signatures (e.g., Debret et al., 2016; Walters et al., 2019) and, more globally, on the dynamics of redox reactions at depth.
View in article
Witt-Eickschen, G., Palme, H., O’Neill, H.St.C., Allen, C.M. (2009) The geochemistry of the volatile trace elements As, Cd, Ge, In and Sn in the Earth’s mantle: New evidence from in situ analyses of mantle xenoliths. Geochimica et Cosmochimica Acta 73, 1755–1778. https://doi.org/10.1016/j.gca.2008.12.013
Show in context
The serpentinites, constituted mostly of antigorite and magnetite (Fig. 1), are highly enriched in AsV and AsIII (up to ∼100 ppm As, which is ∼1000 times more than in mantle-derived rocks; Hattori et al., 2005; Witt-Eickschen et al., 2009).
View in article
top
Supplementary Information
The Supplementary Information includes:
Download the Supplementary Information (PDF).