Volatiles in melts and fluids hold the key to understanding Earth's inner workings. While major uncertainties remain, advances in multi-disciplinary approaches continue to reveal how these elements move through and shape our planet.
This session brings together scientists investigating the full spectrum of volatile and elemental cycles, with a focus on their principal carriers—melts and fluids. We welcome contributions from petrology, geochemistry, and related disciplines, drawing on natural samples, experiments, and modelling.
Topics of interest include:
i) deep volatile cycles of H₂O, CO₂, halogens and sulfur;
ii) volatile mobilization and transfer during subduction in COHNS fluids and silicate melts;
iii) roles of volatiles in metamorphic and metasomatic processes;
iv) physical and chemical properties of volatiles in melts and fluids;
v) volatile storage in the lithospheric mantle;
vi) emissions and reservoirs in volcanic systems.
Orals: Fri, 8 May, 08:30–12:30 | Room K1
The Earth’s cratonic mantle is a major reservoir for carbon stored as carbonates as well as graphite/diamond. Over the billions of years of the cratons’ existence, migrating melts/fluids from the convecting mantle or released from subducted slabs have deposited carbon and other volatiles in the Thermal Boundary Layer (TBL) in the cratonic roots. This geodynamic history has created a geochemical environment that is variable on a small scale, where oxygen fugacity and lithology are heterogeneous and carbon storage and release are spatially controlled.
This complex history of carbon storage in, and release from the cratonic mantle is recorded in some suites of diamonds, such as small-grained polycrystalline diamond aggregates (PCAs). Due to their rapid crystallization, PCAs and their silicate paragenesis preserve chemical and isotopic heterogeneity that are testament to their episodic formation by small-scale melts in the TBL and the cratonic roots. While economically unimportant, PCAs represent sizable proportions of up to 20% of the diamond production in some kimberlites, attesting to the prevalence and importance of their formation mechanism in the deep carbon cycle. We present here a large dataset on PCAs and their silicate paragenesis from South Africa (Jacob et al., 2025) and discuss insights into the deep cycles of carbon and nitrogen that these samples offer.
Jacob, D.E., Stern, R.A., Czas, J., Reutter, M., Piazolo, S., Stachel, T. (2025) Polycrystalline diamond aggregates and their role in Earth’s deep carbon cycle. Geochimica et Cosmochimica Acta, 389, 136-156.
How to cite: Jacob, D., Stern, R., Czas, J., Piazolo, S., Stachel, T., and Foley, S.: Deep volatile cycling and diamond formation in the cratonic lithosphere, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15646, 2026.
Fluids and melts are the principal carriers of volatiles during subduction and continental collision, yet their composition and mutual relationships at high- and ultrahigh-pressure (HP-UHP) conditions remain incompletely constrained. Here we synthesize published and new data on fluid and melt inclusions in diamond-bearing metasedimentary rocks of the Seve Nappe Complex (SNC) and equivalent units in the Scandinavian Caledonides.
In the southern part of the SNC, Tväråklumparna paragneisses contain primary multiphase fluid inclusions in garnet with microdiamond, carbonate, and CO₂, locally accompanied by graphitic carbon, indicating partial retrogression of diamond during exhumation (Majka et al. 2014, Geology). At Åreskutan, diamonds occur as single or as part of multiphase fluid inclusions (MFI) in garnet coexisting with abundant single inclusions of graphite (Klonowska et al. 2017, J. Metamorph. Geol.) and crystallized former melt inclusions (MI) (Slupski 2023, PhD thesis). Multiphase fluid inclusions contain carbonates, hydrous phyllosilicates, rutile, quartz, diamond and graphite, together with a residual fluid phase dominated by CO₂ with minor CH₄ and N₂. Nanogranitoids occurring in the same microstructural domains contain a mineral assemblage consistent with the trapping of felsic melts and preserve measurable H₂O and CO₂ contents. Together, these inclusions provide direct constraints on the nature of deep fluids, indicating that significant amounts of carbon and minor nitrogen were mobilized during deep subduction.
Farther north, diamond-bearing gneisses from Saxnäs (Petrík et al. 2019, J. Petrol.) preserve MFI containing diamonds, carbonates, rutile and hydrous phyllosilicates. New FIB-SEM data show that microdiamonds occur as multiple grains within the MFI located inside the host garnet, commonly attached to the walls of the cavity. Residual fluid is CO2-rich; the presence of carbonates and phyllosilicates as step-daughter minerals suggests that the original fluid was C-O-H. Melt inclusions, possible nanogranitoids, are associated with diamond-bearing MFI. Graphite is widespread in these rocks, occurring both as inclusions and along grain boundaries, documenting extensive carbon re-equilibration during decompression and high-temperature overprint.
In the northernmost locality, the Heia (Nordmannvik) Nappe of northern Norway, garnet hosts diamond-bearing multiphase fluid inclusions coexisting with primary melt inclusions (Janák et al. 2024, J. Petrol.). Fluid inclusions contain CO₂, carbonates, hydrous phyllosilicates, and locally CH₄, whereas crystalized melt inclusions show a mineral assemblage consistent with a granitic composition. Their spatial association provides direct evidence for fluid–melt immiscibility at UHP conditions.
Taken together, these occurrences demonstrate that multiphase, diamond-bearing fluid inclusions and granitoid melts trapped by garnet in metasedimentary gneisses of the SNC consistently record carbon-dominated volatile systems, with carbon preserved as diamond, graphite, carbonate, CO₂, and reduced species such as CH₄, accompanied locally by N₂. Fluid and melt inclusions thus represent key archives for reconstructing volatile speciation, redox conditions, and mass transfer during deep subduction and UHP metamorphism of the orogen.
How to cite: Klonowska, I., Borghini, A., Janák, M., and Yoshida, K.: Fluid and melt inclusions as archives of volatile speciation in high- to ultrahigh-pressure metasedimentary rocks, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14216, 2026.
Carbon is mainly present as a CO2-rich fluid in lithospheric upper mantle, while upon further upward transport, fluid-involved reactions can be expected in the lower crust, like promoting granulite facies metamorphism. Isotope composition of fluids, like CO2-rich fluid inclusions in lower crustal xenoliths, serves as a direct tool to trace lithosphere-scale fluid processes and its potential effects on global carbon cycle. This study explores the significance and the fate of mantle-derived fluids, represented by primary and secondary fluid inclusion assemblages in metasedimentary granulites from the lower crust of the Pannonian Basin.
The studied xenoliths are made up mostly by garnet and sillimanite together with plagioclase, quartz, graphite, rutile and zircon in minor quantities. Garnet is mostly surrounded by fine-grained symplectitic aggregates of orthopyroxene, spinel ± plagioclase. The width of symplectitic corona and the extent of garnet breakdown vary at a thin section scale showing few micron sized rims and also the total replacement of garnet. Intact garnet, however hosts abundant primary graphite and fluid inclusions within its core, which were commonly co-entrapped with quartz, rutile and zircon. The smallest (2-3 µm) primary negative crystal-shaped fluid inclusions in garnet dominantly contain high-density (1.05-1.10 g/cm3) CO2-rich fluid. He-Ne isotope analyses on primary garnet-hosted primary fluid inclusions showed Rc/Ra ratios of 6.3 ± 0.2, thus suggesting a subcontinental lithospheric mantle origin. Results on combined quartz-in-garnet and zircon-in-garnet elastic thermobarometry indicate entrapment at UHT conditions. The intersection of the entrapment isomekes is at a P-T of 1.3 ± 0.4 GPa and 1100 ± 70 ºC. Such P-T conditions are far not compatible with the recent MOHO depth in the Pannonian Basin and clearly indicate that entrapment took place in a crust much thicker than in present days. In the light of previous studies from the Pannonian Basin, garnet is unstable in the present-day lower crust, due to pronounced lithospheric thinning during the Miocene. We provide a calculation on the quantity of CO2, which has been potential released by decompressional garnet breakdown. Such process serves as a newly discovered source of a delayed, secondary mantle degassing due to the residence of abundant mantle-derived CO2 in fluid inclusions in garnet, – the most common mineral in the lower crust –, for millions of years. Potential release of such CO2 may occur much later than leaving the stability field of garnet due to its metastable behavior, as evidenced by xenoliths.
In the light of our results, recent mantle degassing detected by surface-subsurface noble gas isotopic measurements does not always require a direct connection to the mantle or to a cooling magma chamber, but can be derived by destabilization of the lower crust. Accordingly, this mechanism should be taken into account when estimating geological carbon fluxes, as its contribution may be substantial and could potentially influence the overall carbon budget within the Earth system. Incorporating this factor into flux calculations is therefore essential, particularly in post-rift basins, characterized by significant crustal thinning for achieving a more accurate and comprehensive understanding of long-term carbon cycling and its geodynamic controls.
How to cite: Spránitz, T., Hencz, M., Lange, T. P., Keresztes, T., Fehér, K., Aradi, L. E., Molnár, K., Kovács, D., Szabó, Á., Török, K., Gilio, M., Alvaro, M., Szabó, C., Csicsek, Á., Fodor, L., and Berkesi, M.: Mantle-derived CO2-rich fluid entrapment in the lower crust and release via decompression: fluid inclusion systematics from metasedimentary granulite xenoliths in the Pannonian Basin, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17712, 2026.
The North Atlantic Igneous Province (NAIP) coincided in time with the Paleocene–Eocene Thermal Maximum (PETM), which is the most recent natural analogue for anthropogenic greenhouse gas emissions [1]. The temporal association between Large Igneous Provinces and climate perturbations throughout the Phanerozoic points to a potential causality via volatile emissions, especially of carbon species. Since East Greenland represents the closest location to the original centre of the mantle plume [2], we studied melt and fluid inclusions within intrusive and effusive rock samples from the (300 km3 sized) Skaergaard intrusion and its overlying (6-8 km thick) Plateau Basalt lava pile to constrain greenhouse gas emissions. Our Raman microspectroscopy data on melt and fluid inclusions hosted in magmatic minerals within these rock samples unveiled the occurrence of different carbon species. In the effusive rock samples, olivine and clinopyroxene phenocrysts host primary melt inclusions containing CO2 or elemental C within gas bubbles. In the intrusive rock samples, olivine crystals sometimes host CH4-bearing fluid inclusions, and quartz crystals usually host abundant multiphase (i.e., gaseous ± liquid ± solid phases) fluid inclusions, containing CH4 and sometimes CO2 along with H2O. Volatile species preserved by melt and fluid inclusions within magmatic minerals of Large Igneous Province rocks offer an exceptional window on the magmatic and thermogenic emissions released into the past surface system [3; 4]. Here, we attempt to reconstruct the evolution of carbon speciation throughout the emplacement of intrusive and effusive components of the NAIP across the PETM.
[1] Jones et al. (2019), Nat. Commun. 10, 5547.
[2] Larsen & Tegner (2006), Lithos 92, 181–197.
[3] Capriolo et al. (2020), Nat. Commun. 11, 1670.
[4] Capriolo et al. (2021), Nat. Commun. 12, 5534.
How to cite: Tegner, C., Capriolo, M., Muirhead, D., and Jones, S.: Carbon emissions from the North Atlantic Igneous Province: Insights from melt and fluid inclusions in East Greenland, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13731, 2026.
Metamorphic degassing has turned out to be an often-overlooked flux when considering the global carbon and sulfur cycles. Sediments containing organic matter and sulfidic minerals like pyrite undergo the very slow and diffuse process of metamorphic decarbonation and desulfidation. This makes it very difficult to quantify the amount of carbon and sulfur being released into the atmosphere via surface measurements, yet both of these fluxes can have profound effects on planetary habitability over time. The release of carbon and sulfur into the atmosphere also have opposite effects on the global temperature and operate on different timescales. While CO2 with a residence time of ~ 4 years in the atmosphere (Harde, 2017) has an overall warming effect, any SO2 released has a net cooling effect and stays in the troposphere for anywhere between a few hours to 14 days. Depending on the rate of devolatilization, the net effect on habitability will shift.
This study focuses on the Flinton Group, which is a part of the Mazinaw Domain in Ontario, Canada and was deposited between 1180 – 1150 Ma (Kinsman and Parrish, 1990; Sager-Kinsman and Parrish, 1993). Special emphasis is given to the Myer Cave Formation for this study, the lithology of which is defined by sulfidic and graphitic schists and pelites, calcitic and dolomitic marbles and marble clast breccias (Easton, 2006). The mineralogies of the schists can be characterized by quartz, dolomite, calcite, feldspars, muscovite, biotite, hornblende, sillimanite, graphite, pyrite and pyrrhotite along with traces of retrograde chlorite. Unlike the rest of the region that has undergone two phases of metamorphism, the Flinton Group seems to have been subjected to a single medium to high grade event at 976 ± 4 Ma (McCarron et. al., 2014).
The oxidation of graphitic carbon and its release into the atmosphere in the form of CO2 is recorded as a negative shift in the δ13C values of reduced carbon (RC)/graphite. We find that the δ13CRC values in the Flinton Group remain fairly consistent within the range of –19‰ to –25‰ on the Vienna Pee Dee Belemnite (VPDB) scale. Assuming a protolith with δ13CRC ~ –25‰, this suggests minimal mobilization of graphite across greenschist and amphibolite facies. Thermodynamic models show much earlier equilibration in temperature and activity of CO2 (aCO2) space. The sulfur isotope to be analyzed is the stable isotope 34S, which on the Vienna Canyon Diablo Troilite (VCDT) scale may show enriched values of ~10‰ to 40‰ or even more depending on the mass-based fractionation as well as the source of the deposits. Thermodynamic models in temperature vs sulfur fugacity (fS2) space will also bring to light the equilibration conditions and mobility of sulfur.
How to cite: Saha, S. and Stewart, E.: Thermodynamic and geochemical investigation into the fate of graphitic carbon and sulfides during orogenesis, from a field-based approach of the Proterozoic Ottawan orogeny, ON, Canada, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14549, 2026.
Lithospheric carbon fluxes are an essential piece of the geologic carbon cycle. Though more attention is given to volcanic emissions, previous studies suggest that collisional orogenic settings release a significant amount of metamorphic CO2 (Kerrick & Caldeira 1998, Groppo et al., 2017). Quantifying the CO2 released from decarbonation of mixed calcsilicates during mountain building events is important for understanding planetary climate and habitability. Crustal decarbonation is strongly controlled by the availability of mixed carbonate–silicate and dolomite-rich protoliths, which undergo decarbonation at substantially lower temperatures than pure carbonates. These mixed sediments promote efficient carbon release by reacting to form amphiboles, pyroxenes, and other ferro-magnesian silicate phases during metamorphism. Global datasets show that mixed carbonate–silicate and dolomitic rocks are especially abundant in the Paleoproterozoic and Mesoproterozoic (Cantine et. al, 2020). Consequently, modelling studies predict elevated metamorphic decarbonation fluxes in the Proterozoic, driven by both suitable protoliths and geothermal conditions (Stewart and Penman, 2024); however, there have been no field-based studies to test this hypothesis. Here we present preliminary results from a field-based test studying decarbonation from sediments buried and metamorphosed during the Proterozoic Grenville Orogeny.
The Grenville orogenic belt is a large-scale stack of crustal blocks (> 600 km wide) thrust over the older Archean crust as a result of convergence leading to the formation of the supercontinent Rodinia. The Grenville Orogen is thought to be a large hot long duration orogen (Rivers, 2008; Indares, 2020) and has been considered a Proterozoic analogue of present day orogens like the Himalayas. Regionally, the metamorphic grade increases from South to North from greenschist to upper amphibolite facies conditions.
We present a comparative study of decarbonation of two carbon bearing lithologic units: the Grenville Supergroup and the Flinton Group. The Grenville Supergroup consists primarily of metamorphosed marine carbonates which have undergone multiple generations of metamorphism corresponding to multiple orogenies. The younger Flinton Group was deposited < 1155 Ma under local fluvial to shallow marine conditions and has only undergone a single metamorphic event corresponding to the collision of Amazonia with Laurentia (i.e., the Ottawan Orogeny). Within the Flinton Group, we focus on the Fernleigh Formation consisting of laminated calcareous pelites to schists as a representative mixed calcareous – siliciclastic unit.
Preliminary results indicate metamorphic decarbonation in the rocks of the Grenville Orogeny was controlled by the mixed silicate – carbonate bulk composition of protoliths. Rocks of the carbonate-dominated Grenville Supergroup show hindered decarbonation due to limitation of reactant silicate minerals. Upper amphibolite grade, pyroxene bearing rocks of the Grenville Supergroup show ~ 20% decarbonation at ~ 700°C. In contrast, rocks of the Fernleigh formation show enhanced decarbonation (~80-90 %) even at lower amphibolite grades. Decarbonation reactions like amphibole-in and pyroxene-in also occur at comparatively lower temperatures in these carbonate-limited rocks.
We will present detailed results of carbon mobilization using stable isotope geochemistry and thermodynamic modelling. Decarbonation estimates for different metamorphic facies will offer field-based insights into the solid Earth metamorphic flux associated with the Grenville Orogeny during the Proterozoic eon.
How to cite: Sengupta, K. and Stewart, E.: Protolith chemistry controls decarbonation in a Proterozoic Orogen : A field-based test from the Grenville Orogen, Ontario, Canada, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-877, https://doi.org/10.5194/egusphere-egu26-877, 2026.
Carbon recycling in continental subduction zones is a fundamental component of the global deep carbon cycle, yet the efficiency of carbon transfer in these systems remains poorly constrained compared to oceanic subduction zones. Here we report the first zinc isotope data for ultrahigh-pressure (UHP) marbles and carbonated eclogites from the Dabie–Sulu orogenic belt to evaluate the behavior of carbonate minerals under conditions of continental subduction zones. Zinc isotope systematics of UHP marbles reveal contrasting behaviors of different carbonate species. Coupled declines of zinc concentrations and isotope ratios in calcite marbles compared with their sedimentary protoliths provide direct evidence for dissolution, but the dissolution fraction of calcite carbonates, as a whole, is almost negligible. For dolomite marbles, the elevated zinc concentrations and decreased zinc isotope ratios indicate that dissolved carbonates underwent strong refixation within the slab. These observations imply that carbonate dissolution occurs during continental subduction but is spatially restricted, despite some similarities to carbonate behavior in oceanic subduction zones. Carbonated eclogites exhibit zinc isotope compositions comparable to or heavier than those of MORB, reflecting variable extents of carbonate–silicate interaction within the subducted continental slab. These carbonate–silicate interactions further facilitate the stabilization and retention of carbonate minerals during deep continental subduction. As a whole, carbonates in subducted continental slabs remain relatively conservative. This can be attributed to (i) the low density of subducted continental crust, (ii) the limited availability of fluids, and (iii) the thick layers of carbonates in the continent. Consequently, most continental carbonates are retained within the slab and may be exhumed back to the crust, undergo diapiric ascent, or be underplated beneath thick continental lithosphere, rather than being extensively dissolved at sub-arc depths.
How to cite: Sun, W.-Y. and Liu, S.-A.: Carbon conservation in continental subduction zones revealed by zinc isotopes in ultrahigh-pressure marbles and carbonated eclogites, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13313, 2026.
Hydrothermal circulation and associated alteration of the oceanic lithosphere are the first order control on Earth’s volatile cycles and have been proposed as a potential driver of the emergence of life on our planet. Constraining the extent of oceanic lithosphere’s alteration, and its consequences on lithospheric composition, carbon budget including abiotic organic compound formation is thus key.
While these processes have been investigated at mid-ocean ridges (MOR), oceanic transform faults (OTFs), which regularly segment MOR, have received comparatively little attention. Recent studies, however, suggest that these plate boundaries can be the locus of deep mantle hydration by downward percolation of seawater-derived fluids (to depths of ~ 25-30 km on (ultra)slow spreading ridges; Prigent et al., 2020; Wang et al., 2022), as well as mantle carbonation by upward percolation of magmatic-derived carbon-rich fluids within the fault zone (Klein et al., 2024). Such fluid circulation is key in establishing chemical, particularly redox, disequilibria that influence carbon speciation. In addition, subduction of fracture zones, the fossilized portion of OTFs, is associated with higher slab seismicity and enriched geochemical signatures in overlying arc lavas (e.g. Paulatto et al., 2017). Together, these observations identify OTF as an important yet poorly constrained component of the Earth’s volatile cycle, potentially influenced by both hydrothermal and magmatic processes.
This study focuses on constraining the deep volatile cycle on OTFs, with a particular emphasis on carbon. Using deformed and hydrated peridotites from two OTFs of the Southwest Indian Ridge, we characterized water bearing-components (e.g. amphibole, fluid inclusions) that formed during high temperature deformation (700-900°C).
Hydrated silicate phases (e.g. amphibole) serve as indicators of fluid-rock reactions. Trace element concentrations and enrichments in chlorine, boron and lithium suggest a hydrothermal origin for the fluids interacting with the studied mantle rocks, even at great depths.
Fluid inclusions (FIs), mainly hosted in olivine, occur as trails formed near the brittle-ductile transition of the host mineral. Some trails are associated with the formation of the high-temperature shear bands, suggesting syn-deformational fluid trapping. Raman spectroscopy and FIB-SEM analyses of FIs in olivine reveal various crystalline (including serpentine, brucite, magnetite) and gaseous phases (CH4 and H2) in FIs, suggesting intense fluid-olivine reactions during rock cooling. Carbon-bearing phases, including methane and carbonaceous compounds, also formed together with molecular hydrogen, which was likely produced during olivine serpentinization. Methane concentrations (25-136 ppm) and δ13C-CH4 values (-3.8 to -21.5‰), measured for the first time at OTFs, overlap those reported from SWIR gabbros and sediment-starved hydrothermal systems; more work is needed before making robust constraints on the carbon source.
Overall, our results highlight OTF as active sites of deep hydrogen and carbon cycling and emphasize their role in controlling volatile speciation during high-temperature deformation of the upper mantle.
Klein et al. (2024). Proc. Natl. Acad. Sci. U.S.A. 121, e2315662121.
Paulatto et al. (2017). Nat Commun 8, 15980.
Prigent et al. (2020). Earth and Planetary Science Letters 532, 115988.
Wang et al. (2022). Nat Geosci 15, 741–746.
How to cite: Joanno, S., Prigent, C., Bickert, M., Andreani, M., Vitale Brovarone, A., Montagnac, G., Herviou, C., and Menez, B.: High-temperature peridotite mylonites reveal deep organic carbon cycle at Oceanic Transform Faults, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17344, 2026.
Magmas generated by partial melting of mantle rocks are the main carriers of volatile species (e.g., CO2, H2O, SO2) and trace elements (rare earth elements, Hg, Co, etc.) to the Earth’s surface. Mercury (Hg) is of particular interest because it has been widely used over the last decade as a marker of large-scale volcanic eruptions in sedimentary records, owing to its relatively long atmospheric residence time (0.5–2 years; Bagnato et al. 2007), and its association with mass extinction events (Percival et al. 2018). Although Hg is present at low abundance in the silicate Earth (10 ppb; McDonough and Sun 1995), isotopic studies on sediments point to a predominantly volcanic origin (Grasby et al. 2019). However, no experimental work has yet constrained the mechanisms by which Hg is mobilized from mantle sources to the atmosphere, and only a few geochemical studies on meteorites and peridotite xenoliths suggest that sulfide minerals are the main Hg host at depth (Canil et al. 2015).
In this study, Hg solubility in Fe–Ni–S alloy was investigated at 6 GPa and 700–1400 °C using a rotating multi-anvil apparatus (MavoPress LPT 500-400/50 with a Walker-type module) at the Department of Earth Sciences, Sapienza University of Rome. The starting materials consisted of a mixture of pure Fe and Ni powders doped with 5 wt.% natural cinnabar (HgS) as the Hg source, allowing quantitative analysis by electron microprobe. In addition, Hg solubility in synthetic melts was examined at 3-6 GPa, 1300-1550 °C, and oxygen fugacity buffered near the graphite–CO2 redox equilibrium, using six-anvils cubic presses at the Center for High Pressure Science & Technology Advanced Research (HPSTAR), Beijing (Wu et al. 2024; Xu et al. 2025). Two starting compositions were employed, a synthetic picritic glass and a carbonate–silicate glass, each mixed with ~5 wt.% natural HgS.
The results show that Hg increasingly partitions into the Fe-Ni alloy with rising temperature. In the presence of silicate melts, Hg concentrations of up to ~1700 ppm under sulfur-saturated conditions are observed, with similar contents in both carbonate-silicate and picritic melts. Additionally, Hg abundance is primarily controlled by the concentration of dissolved sulfur. These experimental constraints are finally compared with the limited available data on Hg concentrations in natural volcanic rocks to better quantify the deep Hg cycle.
References
Bagnato E. et al. (2007). Atmos. Environ., 41(35), 7377-7388.
Canil et al. (2015). Chem. Geol., 396, 134-142
Grasby S.E. et al. (2019). ESR, 196, 102880.
McDonough W.F., Sun S.S. (1995). Chem. Geol., 120 (3-4), 223-253
Percival L. M. et al. (2018). AJS, 318(8), 799-860
Wu P. et al. (2024). Matter Radiat. Extremes 9, 027402
Xu Y. et al. (2025). Matter Radiat. Extremes 11, 017803
How to cite: Benedetti, F., Marras, G., Lei, S., Morelli, T., Lin, Y., and Stagno, V.: Experimental constraints on mercury solubility both in Fe-Ni(-S) metal and volatile-bearing silicate melts at high pressure and temperature, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20923, 2026.
Fluid–rock interaction and microbial life are intimately connected. One process that is recognized to feed microbial communities is serpentinization, during which mantle minerals (olivine and pyroxenes) react with fluids to form serpentine and magnetite. This process also produces H2 and, in the presence of C-rich units, CH4. These serpentinization-derived, reduced C-H forms represent energy sources for microbial activities, as documented at multiple sites on present-day and ancient seafloor (i.e., mid-ocean ridges and ophiolites). Notably, however, serpentinization does not only occur at crustal levels that are shallow enough to overlap with T conditions permitting microbial life (the “biotic fringe”; ≤ 135 °C). Instead, several cases (e.g., the Monte Maggiore ultramafic massif, France) have been documented where oceanic lithosphere was majorly serpentinized only during subduction. This implies that major amounts of reduced energy sources may not be released until as deep as the plate interface, where microbial life is precluded.
We show that serpentinization occurring at P–T conditions that are prohibitive for microbial life can still play a fundamental role in nurturing microbial communities at shallow levels in the mantle wedge. At least from the Phanerozoic, slab-derived serpentinizing fluids have induced the formation of plate-interface metasomatic rocks (PIMRs) in the mantle wedge worldwide. Importantly, these PIMRs were recognized as fluid pathways by previous studies, and their exhumation path overlaps with the biotic fringe. Using micro-Raman spectroscopy, we identified CH4, H2 and N2 in fluid inclusions in high-P/primary (e.g., jadeite) and lower-P/secondary (e.g., albite, analcime) minerals constituting Phanerozoic PIMRs all over the world, confirming their role in transferring deeply tapped, reduced energy sources from the plate interface to the biotic fringe. U-Pb geochronology of primary (i.e., zircon) and secondary (i.e., titanite) minerals in Phanerozoic-exclusive PIMRs confirms that such fluxes were protracted for tens–hundreds of millions of years, thus being able to sustain subsurface microbial communities in the mantle wedge. Our thermodynamic modelling confirms that, as subduction regimes became progressively cooler across geological time, reduced C-H forms like the detected CH4 and H2 became dominant over oxidized ones (e.g., CO2). High-P serpentinization-derived fluids thus became optimal for sustaining microbial life in the Phanerozoic. However, serpentinization-derived energy sources may have never reached the biosphere in the mantle wedge without the emergence of fluid pathways like PIMRs, which emerged worldwide only in the Phanerozoic. Our study indicates that the emergence of these Phanerozoic-exclusive PIMRs, combined with cooler subduction regimes, may have played a pivotal role in promoting the proliferation and diversification of microbial life in this eon by boosting the supply of energy sources towards the biotic fringe.
While it is true that the release of slab-derived fluids induces potentially catastrophic geological processes that can disrupt life as we know it (e.g., high-magnitude earthquakes and highly explosive volcanic eruptions), if properly channelized and under the right geodynamic conditions, such fluids can also play a key role in sustaining hidden life on Earth.
How to cite: Peverelli, V., Olivieri, O. S., Tsujimori, T., Giovannelli, D., Shi, G., Cannaò, E., Piccoli, F., and Vitale Brovarone, A.: The role of plate-interface metasomatic rocks in nurturing subsurface microbial life, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7159, 2026.
Convergent margins are key engines of mass transfer between Earth’s surface and interior, governing long-term fluxes of volatiles, redox-sensitive elements, and carbon. Subduction transports large quantities of water and carbon into the mantle, while fluid release and metasomatic reactions move these components into the mantle wedge and, in places, toward the surface. Central to this exchange is mantle wedge serpentinization, the slab-derived hydration of mantle wedge peridotite that alters redox conditions and generates reduced compounds such as H2 and CH4, directly linking deep Earth processes to carbon cycling and energy availability for subseafloor biosphere.
In the modern Mariana forearc, serpentinite mud volcanoes provide a rare natural laboratory to directly interrogate the products and consequences of mantle wedge serpentinization. Our recent geochemical, isotopic, and lipid biomarker findings demonstrate that the availability of abiotic geofuels produced during serpentinization exerts a first-order control on the subsurface chemosynthetic microbial communities on a temporal scale (Kumawat et al., 2025). We present in situ stable oxygen isotope measurements of serpentine from several Mariana mud volcanoes, combined with published pore fluid δ18O compositions. They define systematic spatial trends in serpentinization temperature, from cold, trench-proximal settings to progressively hotter conditions deeper in the mantle wedge. Using our newly developed serpentine–water calibration, our data imply that these thermal gradients regulate redox evolution and the production of reduced volatiles and organic components, establishing dynamic energy landscapes that sustain life under high pH, nutrient limitation, and episodic substrate delivery.
While mantle wedge serpentinization, serpentinite mud volcanism, and associated biospheres are increasingly well-constrained in the modern Mariana forearc, their occurrence and significance in the geological record remains largely unconstrained. We also present geochemical evidence for Early Cretaceous serpentinite mud volcanism preserved within the paleo-forearc basin within the Coast Ranges of California. Elevated fluid-mobile element inventories, systematic oxygen isotope record of serpentine, and textural evidence for mud-supported serpentinite transport are complemented by an extensive methane-seep fossil record and lipid biomarker signatures indicative of chemosynthesis-based ecosystems. Together, these observations suggest that mantle wedge serpentinization and focused fluid discharge have been major volatile and energy providers in Mesozoic convergent margins.
By integrating modern forearc observations with ancient geological archives, this work highlights serpentinization as a persistent and efficient mechanism for mass transfer, redox modulation, and volatile cycling at convergent margins. These processes not only shape mantle metasomatism and arc evolution but also link deep Earth volatile pathways to the limits of habitability in the deep biosphere through Earth history.
Kumawat, P., Albers, E., Bach, W. et al. Biomarker evidence of a serpentinite chemosynthetic biosphere at the Mariana forearc. Commun Earth Environ 6, 659 (2025). https://doi.org/10.1038/s43247-025-02667-6
How to cite: Kumawat, P., Albers, E., Shapiro, R. S., Peckmann, J., Scicchitano, M. R., Menapace, W., Klein, F., Frederichs, T., Roud, S. C., Hansen, C., Monien, P., Klügel, A., Vogt, C., Toro, M., Taubner, H., Shervais, J. W., Jean, M. M., Wheat, C. G., Schubotz, F., and Bach, W.: Serpentinization and Subduction Mass Transfer Processes: Implications for Chemosynthetic Life in Modern and Ancient Forearcs, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21342, 2026.
There is sustained interest in the potential coupling between geodynamic processes and the deep volatile cycle, including oxygen and other redox-sensitive elements. Most current global mass-balance estimates are based on the redox budget [1] or redox capacity [2] of subducted inputs with highly variable compositions and hydration states (sediments, mafic and ultramafic rocks), and on corresponding global outputs. Past efforts have largely focused on the deep-water cycle, particularly in subduction zone settings. However, the evolution of the redox state and redox budget of these diverse inputs during subduction has only recently begun to be addressed, and many fundamental questions remain unresolved. A key issue is the extent to which oxidised species are transferred from the slab to the mantle wedge. This problem is difficult to evaluate when subducting lithologies are assumed to behave independently and as closed systems during dehydration reactions. Increasing evidence instead points to significant exchange of aqueous fluids among contrasting lithologies, with critical and non-linear effects on the redox capacity of fluids ultimately transferred to the mantle wedge [3]. In addition, high-pressure hydrodynamics, driven by dynamic permeability changes in compacting rheologies, remain poorly constrained. The role of the mantle wedge as a potential oxygen reservoir is therefore an emerging topic of interest.
In this contribution, we present a series of natural, theoretical, and experimental case studies based on analyses of COHS components in different lithologies. These observations are complemented by bulk and in situ stable-isotope data, which further support mixing of aqueous fluids from different sources under variable pressure–temperature conditions. Most mafic and ultramafic input lithologies show prograde evolution under highly oxidising conditions and possess a high redox budget. However, interaction with lithologies containing minor amounts of reduced phases, such as graphite-bearing metapelites, produces distinctive petrological and geochemical signatures and substantially reduces the oxidising capacity of the interacting lithologies. In particular, sulphur and carbon efficiently track these interactions and represent the most effective vectors for redox-budget transfer from the slab to the mantle wedge. New data on the role of the cold mantle wedge as an oxygen reservoir are also presented.
Overall, these observations highlight the need to integrate lithological interactions and fluid exchange into models of subduction-zone processes, accounting for secular and global variations in input lithologies. Future constraints on ferric iron in key high-pressure hydrous phases and on the stability of sulphur-bearing phases will enable the development of improved thermodynamic models, leading to more robust predictions of the redox capacity of high-pressure aqueous fluids.
[1] Evans (2012) Earth-Science Reviews, 113, [2] Galvez, M. E., Müntener, O., & Jaccard, S. L. (2025). Geophysical Research Letters, 52 [3] Padrón-Navarta et al. (2023) Nature Geosciences, 16,
This project has been funded through the ERC project OZ (DOI: 10.3030/101088573).
How to cite: Padrón-Navarta, J. A., Bukała, M., Menzel, M., Ramón-Fernández, M., Cristóbal, L. S., Garduño, I., Rosenthal, A., and Lopez-Sanchez, M.: New perspectives on the oxygen and deep water cycle, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20943, 2026.
Seafloor mantle exhumation along detachment faults is well known from slow to ultra-slow spreading ridges, i.e. oceanic core complex, and from magma-poor margins at ocean-continent transitions. It is also thought to occur in supra-subduction zone environment (SSZ) although the detachment architecture and associated magmatism, deformation and metasomatism are still poorly understood. The Western Limassol Forest Complex (WLFC), in the Troodos ophiolite, Cyprus, is characterized by extensive mantle rock exhumation associated with complex magmatism, extensive metasomatism and ultramafic-hosted volcanogenic massive sulfide mineralizations. The exhumation of mantle rocks in the WLFC has been interpreted to be the result of a poly-tectonic evolution, including transform fault-related deformation along the Arakapas transform fault during the Cretaceous, overprinted by Miocene thrusting associated with the Yesavara thrust belt. However, new field observations, together with structural and geochronological data from the WLFC indicate that the initial exhumation of the mantle sequence occurred along an earlier Cretaceous detachment fault.
Ultramafic rock metasomatism in the WLFC is dominated by serpentinization but other seafloor-like metasomatisms, such as sulfide and magnetite mineralizations as well as carbonation are also present. Preliminary in-situ calcite and magnetite U-Pb dating by LA-ICP-MS give ages ranging between 85.6 ±12 Ma and 92±2 Ma for seafloor metasomatism in the WLFC. In this contribution we present an overview of seafloor metasomatism preserved in the WLFC using field, petrographic and geochemical evidences, and discuss how it relates to mantle exhumation during the early stages of the Troodos ophiolite evolution. The WLFC detachment, although strongly overprinted by transform and thrust-related deformations appears to be one of the few well-preserved oceanic detachments in an SSZ environment worldwide.
How to cite: Patten, C. G. C., Peillod, A., Junge, M., Rogowitz, A., Hector, S., Coltat, R., Beranoaguirre, A., Bilau, A., and Kontny, A.: Seafloor metasomatism associated with oceanic detachment in supra-subduction zone ophiolite: evidence from the Western Limassol Forest Complex, Troodos, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17283, 2026.
Subduction zone fluids control mass transfer and crust-mantle evolution, yet their detailed sources and fluid-rock interaction processes remain debated. Jadeitite, formed either by direct precipitation from Na–Al–Si-rich fluids (P-type) or by metasomatic replacement of magmatic protoliths (R-type), serves as a unique archive of subduction zone fluids. We present high-precision Mg–Fe isotopic data for jadeitites and jadeite-rich rocks from the Rio San Juan Complex (RSJC), Dominican Republic. RSJC jadeitites exhibit low δ26Mg values (−0.92‰ to −0.16‰) that lack correlations with carbonate indicators (e.g., CaO/Al2O3, CaO/TiO2 and Sr/Nd), precluding a significant contribution from sedimentary carbonates. Instead, the coupling of light Mg isotopes with MgO–Ni–Cr enrichment indicates a substantial contribution from serpentinizing fluids. In contrast to the light Fe isotope signatures of Myanmar jadeitites, RSJC jadeitites display relatively high δ56Fe values (−0.08‰ to 0.29‰). Systematic covariations between δ56Fe values and redox-sensitive proxies (V/Sc, U/Th, Ce anomalies and Sb/As) suggest that Fe isotope heterogeneity is primarily controlled by fluid redox conditions. By integrating petrological and geochemical constraints, we propose that forearc serpentinization acts as a critical redox filter that governs the coupled Mg–Fe isotope heterogeneity of jadeitites. Olivine-dominated serpentinization generates reducing conditions that promote light Fe isotope fractionation, as recorded by Myanmar jadeitites, whereas orthopyroxene-involved serpentinization buffers the system under relatively oxidizing conditions, preserving heavier Fe isotopic signatures in RSJC jadeitites. Jadeitite-forming fluids are best explained as mixtures of altered oceanic crust-derived and serpentinizing fluid components. Consequently, forearc serpentinization exerts critical control on the redox state and chemical heterogeneity of fluids transferred to the mantle wedge and arc magmas.
How to cite: Chen, K., Chen, Y.-X., Tsujimori, T., Schertl, H.-P., Takahashi, N., Huang, F., and Maresch, W. V.: The impact of forearc serpentinization on the composition of subduction-zone fluids revealed by Mg–Fe isotopes in jadeitites, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4635, 2026.
The migration of fluids released during slab-dehydration in deep subduction environments is strongly controlled by deformation and lithological discontinuities such as serpentinized shear zones. However, the geological meaning of geochemical fingerprints and how they relate to transport mechanisms and spatial scales of fluid flux in deformed serpentinite at depth remain poorly understood. We herein focus on three subduction-related mantle sections: an intra-slab serpentinized shear zone (Monviso, Italian Alps), an underplated ultramafic sliver (Zagros suture zone, Iran) and the former base of a mantle wedge (Polar Urals, Russia). Most major and trace element signatures of serpentinites appear rather homogeneous along the transects. In contrast, boron isotopic signatures (δ11B) show systematic variations at several hundred meters scale approaching the main structural boundaries for each locality. A decrease is observed in the Monviso and Urals localities (from c. 25‰ to c. 7‰, and from c. 16‰ to c. 0‰, respectively), while the Zagros section shows an increase from c. 1‰, up to c. 9‰ in the most sheared and serpentinized samples. These variations reflect complex fluid-rock interactions processes including B loss or B addition associated with protolith and fluid variability. This demonstrates that major shear zones exert a first-order control on the serpentinite boron isotopic signature. We combine boron isotopic data with Darcy-based flux models to quantify the volume of rock influenced by paleo-fluid fluxes in deep ultramafic settings. These combined petrostructural and isotopic constraints highlight the importance of fracture-controlled fluid flow in slab-top serpentinites, and yield a time-integrated permeability of the (partly) serpentinized base of the mantle wedge in the range of 10-19 to 10-18 m².
How to cite: Angiboust, S., Minnaert, C., Sanhueza, J., Romer, R., Skoblenko, A., Sobolev, I., and Muñoz-Montecinos, J.: Boron isotopes unravel cryptic fluid-rock interactions in sheared subduction zone serpentinites , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11617, 2026.
Boron is a key volatile tracer in subduction systems. It is concentrated in the pore waters of subducting sediments prior to diagenesis, partitions between aqueous and solid phases, is highly fluid-mobile, and is progressively released during devolitization. Exchangeable (aqueous + adsorbed) boron is primarily released by desorption at low temperatures (≤150 °C) and lattice bound boron is released by breakdown of hydrous phases at higher temperatures (<350 °C). During compaction, diagenesis, and dehydration of sediment in the forearc, released boron migrates with fluids and volatiles to seafloor seeps and mud volcanoes, or is retained and entrained within the subducting slab, where it can appear in arc lavas. Its widespread distribution makes boron an effective tracer throughout subduction. However, previous studies have not examined which subduction zone characteristics control the concentrations of exchangeable boron in subducting sediments, nor how these characteristics - and the resulting boron fluxes - vary among margins, limiting its use as a quantitative tracer of volatile recycling.
To address this, we investigated the mechanisms governing boron adsorption on compositionally representative Circum-Pacific trench sediments and quantified adsorbed boron input to subduction zones. We conducted boron adsorption experiments on sediments from Costa Rica, Barbados, Cascadia, Nankai, and the Hikurangi trench obtained from IODP drill cores. Sediment surface areas (SA) and aluminum-oxide (Al-O) contents were characterized using multi-point Brunauer–Emmett–Teller (BET) analysis and X-ray photoelectron spectroscopy (XPS), respectively.
Results show that boron adsorption onto marine sediments is controlled by SA and lithology, specifically the abundance of surface aluminum-oxide sites. SA varies by up to a factor of 4.6 among margins (e.g., 130 m2/g at Hikurangi versus 28 m2/g at Nankai), while aluminum-oxide surface concentration varies by a factor of 1.3 (e.g., 16 Al-O At% at Nankai versus 22 Al-O At% at Japan). Adsorbed boron flux varies by up to a factor of 9 between sampled subduction zones, being greatest at Hikurangi (5 kg yr⁻¹ m⁻¹) and similar at the Japan and Nankai trenches (0.6 kg yr⁻¹ m⁻¹) - higher SA and Al–O content at the Japan Trench is offset by reduced sediment input compared to Nankai - producing similar adsorbed boron fluxes. Adsorbed boron dominates the total exchangeable boron entering the trench, accounting for 97%, 84%, and 82% of the flux at Hikurangi, Nankai, and Japan, respectively, and contributes the largest variability among aqueous (up to 1.6×), adsorbed, and lattice-bound (~3.5×) boron. Total boron flux is highest at Hikurangi (18 kg yr⁻¹ m⁻¹), lowest at northern Japan (5 kg yr⁻¹ m⁻¹), and intermediate at Nankai (8 kg yr⁻¹ m⁻¹).
These findings indicate that boron fluxes into subduction zones vary substantially among margins, primarily due to differences in adsorbed boron at the trench, which largely reflect variations in sediment SA. Adsorbed boron dominates early boron release through desorption during initial burial and devolatilization, making it the most important boron reservoir for volatile tracing in the shallow forearc. The zone with the highest adsorbed boron flux, Hikurangi, corresponds to the most elevated boron concentrations observed at seeps. This work highlights that adsorbed boron - a previously overlooked component- is critical to volatile transport and geochemical cycling throughout the subduction complex.
How to cite: Ferrie, N., Saffer, D., Breecker, D., and Emslie, S.: Comparative Boron Budgets of Subduction Zones and Implications for Volatile Cycling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15940, 2026.
Mantle wedge serpentinization is driven by fluids released from subducting slab and represents a key mechanism of crust-mantle chemical exchange in subduction zones. However, the relative contributions of different slab lithologies to forearc serpentinization remain poorly constrained, despite the consequences for mineralogy, redox state, and the volatile budget in the mantle wedge. Here, we present radiogenic (87Sr/86Sr) and stable strontium isotope (d88Sr) data for pore fluids, mantle wedge serpentinites, and subducted lithologies (metasediments, metabasalts), recovered from the Mariana Forearc during IODP Expedition 366.
Across all sample types, 87Sr/86Sr values cluster narrowly between 0.705 and 0.706, indicating strong buffering by a dominant Sr reservoir. These values overlap those of altered oceanic crust, with only a minor sedimentary contribution, implying extensive fluid-rock interaction and redistribution of Sr along the decollement zone. In contrast, d88Sr values vary substantially and reflect both source and process-dependent modification of Sr during fluid transport. Compared to incoming altered oceanic crust (ODP801), metabasalts show increased Sr concentrations accompanied by lower d88Sr, consistent with carbonate-controlled Sr addition during interaction with deeper slab-derived fluids along the plate interface. Metasediments largely retain their incoming d88Sr composition, with one sample recording anomalously heavy values (0.63‰).
Pole fluids and serpentinites from the shallowest sites exhibit the heaviest d88Sr values (∼0.6‰), which decrease systematically at greater depths (∼0.3‰ and ∼0.2‰). We interpret these trends as reflecting a depth-dependent transition in Sr-hosting phases and fluid sources, with sulphate precipitation in subducted sediments producing isotopically heavy fluids at shallower levels, and carbonate dissolution-precipitation reactions dominating Sr budgets and d88Sr signatures at greater depths. Together, these results demonstrate that while radiogenic Sr isotopes are homogenized by forearc fluid-rock interaction, stable Sr isotopes provide a sensitive tracer of slab-fluid provenance and reaction pathways during mantle wedge serpentinization.
How to cite: Rajič, K., Robinson, L., Menzies, C., Nowell, G., Debret, B., Evans, A. D., and Burton, K.: Isotopic constraints on slab-derived fluids and mantle wedge serpentinization, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18227, 2026.
Since the fictional journey by Jules Verne in 1864, more than 100,000 non-fictional articles have been published to follow the journey to the center of the Earth. Instead of deciphering the Runic manuscript to set the passage for the journey, seismologists have constructed their own maps of the Earth’s interior, on which geochemists color resident minerals. While the team of professor Liedenbrock embarked their epic journey through a volcano in Iceland, modern mineral physicists subject minerals to the conditions expected in the Earth’s interior and observe the deep world in situ. While it is unclear whether the protagonists in the novel could complete their journey to the very center of the Earth, the scientific journey deep into our planet still awaits much more to be discovered.
In this talk, I will showcase what has been added through our own scientific journey to the Earth’s interior. We set water, one of the most important volatile species on the Earth’s surface, as the supporting actor in our journey. We used diamond anvil cells to let the water meet minerals under the conditions expected deep inside the Earth and irradiated X-rays to watch their interactions. As the fictional team found a subterranean river during their journey, we observed how much more water can be added into mineral carriers during subduction processes [1,2], how deep water can be delivered by transferring the carriers [3], and what would happen when water reaches the destination at the core-mantle boundary, as depicted in the novel by lightening clouds over the subterranean ocean [4]. Deep in time, natural force could have created conditions for life by the action of heat and moisture. By simulating the reactions in the early magma ocean by the brightest X-ray pulses, we observed how water could have nurtured the conditions required for the origin of life on Earth [5].
References
[1] Hwang, et. al., Nature Geoscience, 10 (2017), 947-953
[2] Bang, et. al., Nature Communications, 16 (2025), 2279
[3] Bang, et. al., Nature Communications, 15 (2024), 4428
[4] Kim, et. al., Nature Geoscience, 16 (2023) 1208-1214.
[5] Choi, et. al., Science Advances, 9 (2023), eadi6096
How to cite: Lee, Y.: “Journey to the center of the Earth”: a mineral physicist’s revisit, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15534, 2026.
The flux of carbon, sulfur, and other volatiles between Earth’s mantle and surface plays a fundamental role in shaping the planet’s long-term geochemical cycles. However, modeling the fate of these volatiles at elevated pressures and tracking their oxidation state remains challenging. These difficulties are reflected in the large uncertainties that persist in global carbon and sulfur budget estimates.
Recent advances in thermodynamic modeling have incorporated electrolytic aqueous fluid speciation and open-system frameworks, significantly improving our understanding of slab devolatilization. These developments have clarified the roles of carbon and sulfur dissolution, mass transfer, and associated redox conditions during subduction. At the same time, the volume and complexity of model outputs have increased substantially, creating a need for modern tools for efficient data handling, processing, and visualization.
To investigate the redox budget and COHS fluid speciation across a global suite of subduction thermal models, we developed ThermoPathX, a software framework for automated thermodynamic modeling. ThermoPathX uses PerpleX [1] as its computational engine and enables streamlined preparation and execution of one- and two-dimensional (X–Y) models, followed by the construction of semi-3D (X–Y–Z) models through an iterative workflow based on extracted and processed results. This multi-dimensional approach allows simultaneous analysis of fluid-release pulses, fluid composition and speciation, and oxygen fugacity during prograde metamorphism along a wide range of subduction-zone P–T paths.
We tested ThermoPathX's capacity to explore the potential effects of thermal regimes and initial redox budget on fluid redox capacity by examining the evolution of metapelites during subduction metamorphism. Continental metapelites show a systematic decrease in Fe3+/ΣFe with increasing metamorphic grade during regional metamorphism [2]. Here, we examine whether a similar trend occurs in subduction-zone metapelites and whether such behavior can be explained by intrinsic (closed-system) devolatilization, or instead requires open-system interaction with externally derived reduced fluids. Our modeling indicates that intrinsic devolatilization alone is sufficient to reduce the bulk Fe3+/ ΣFe ratio and the overall redox budget, driven by the loss of oxidized volatile components in aqueous fluids due to the oxidation of graphite and the reduction of ferric iron in silicates in the rock. This reduction is more pronounced along warm subduction geotherms.
[1] Connolly, 2009 (doi: 10.1029/2009GC002540); [2] Forshaw & Pattison, 2023 (doi: 10.1130/G50542.1)
This research work was funded by the European Commission – NextGenerationEU, through Momentum CSIC Programme: Develop Your Digital Talent (MMT24-IACT-01; M.Bukała) and the ERC CdG, OZ: Deep Earth’s Oxygen recycling at subduction Zones Grant Agreement 101088573.
How to cite: Bukała, M., Padrón-Navarta, J. A., Menzel, M. D., and Garrido, C. J.: Redox evolution of metamorphic COHS fluids in subduction zones: Insights from automated thermodynamic modeling with ThermoPathX, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11288, 2026.
Major and trace element contents, including H2O, of garnet and clinopyroxene were determined in eclogite and pyroxenite xenoliths from the Diavik diamond mine, Slave Craton, Canada. Three eclogite (A, B, C) and two pyroxenite types (B, C) are distinguished on the basis of garnet composition (A: high Ca/low Mg; B: high Ca/high Mg; C: low Ca/high Mg).
The 20 xenoliths were derived from a 100 km depth range (c. 120–220 km; one sample: 230-240 km). The data show that pyroxenite and eclogite of C-type invariably come from lower depth (<175 km) than B-type pyroxenite and eclogite (>200 km). Type-A eclogite is present in both intervals. The equilibrium conditions of orthopyroxene-bearing samples (exclusively C-types) point to a geotherm equivalent to 37 mW/m2 surface heat flow.
The contents of structural H2O in clinopyroxene are variable in the sample set (123–1509 wt.ppm) with an average of 440 wt.ppm. Excluding three samples with exceptionally high contents, the range is reduced to 123–522 ppm (17 samples). Clinopyroxene in B-type eclogite and pyroxenite (i.e., 360–1149 and 225–1509 ppm) has considerably higher and much more heterogeneous H2O contents than C-type samples (eclogite: 123–165 ppm; pyroxenite 321–393 ppm). Structural H2O of clinopyroxene is positively correlated to some trace (Cu, Ni, and K) but not to major elements and also depends on the rock type. Eclogitic clinopyroxene has lower H2O contents relative to pyroxenite. While the contents of B-type eclogite and pyroxenite overlap, there is a distinct gap between C-type eclogite and pyroxenite.
In garnet, the contents of structural H2O are low (0– 41 ppm) and correlate neither with rock type nor with mineral composition. Most garnet grains additionally contain molecular H2O (in contrast to pyroxene), which is correlated to compositional parameters of both garnet (positive: Mg; negative: Ca, Sr, Be, Na) and clinopyroxene (positive: MREE, Ca, Th; negative: Al, K, Na, Li).
The observation that structural H2O in both minerals is unrelated to major elements, the highly variable contents of structural H2O in clinopyroxene, and the lack of an H2O-Ca correlation in garnet are unusual and point to disequilibrium. This, and the correlation of the mineral composition with molecular H2O – being of secondary origin – indicate that structural H2O does not reflect equilibrium at PT peak conditions. All these characteristics imply that structural H2O was affected by secondary processes related to metasomatism due to reaction with a hydrous fluid or melt. Metasomatic changes that led to lower clinopyroxene Na and Zn contents along with higher contents of Mg#, Cr, Sr, REE, Pb, Th, U, and Cu. The data indicate that Diavik pyroxenite formed from an eclogitic precursor: pyroxenite B from eclogite B at greater (200-240 km) and pyroxenite C from eclogite C at lower depth (120-175 km).
How to cite: Schmädicke, E., Gose, J., and Stachel, T.: Pyroxenite from Diavik, Canada: metasomatic origin from eclogite, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3367, 2026.
Volatiles play a fundamental role in arc magmatism and ore-forming processes (Wilkinson, 2013; Zellmer et al., 2015). However, their origin and distribution within the lower crust of arcs remain poorly constrained, despite the potential of the lower crust to act as a significant volatile reservoir. Well-preserved fossil arc sections such as the Talkeetna arc (Alaska) provide rare opportunities to investigate volatile behavior across near-complete profiles of lower arc crust (Clift et al., 2005; Hacker et al., 2008).
In this study, we aim to combine petrography, thermobarometry, volatile concentrations, major and trace element geochemistry to constrain the distribution and origin of volatiles within the lower crust of the Talkeetna arc. We selected a sample suite that consists of 14 mafic to ultramafic lower crustal cumulates, including pyroxenites and gabbroic lithologies, representative of different levels of the arc crust.
Volatile content (H2O, C, Cl, S, F), together with major and trace elements concentrations are measured in Nominally Anhydrous Minerals (NAMs), including clinopyroxene (cpx) and orthopyroxene (opx) using SIMS, EPMA and LA-ICP-MS, respectively. Volatile concentrations show significant variability among mineral phases and lithologies. Clinopyroxene systematically contains at least 200 ppm H2O across all lithologies with an average concentration of 650 ppm H2O. Orthopyroxene generally contains lower H2O contents, below 400 ppm H2O, with an average of 300 ppm H2O.
These non-negligible H2O concentrations in NAMs have important implications for volatile recycling and storage in the lower crust (Wallace, 2005; Bekaert et al., 2021). Although NAMs contain lower H2O concentrations than hydrous phases, their volumetric dominance may make them a major volatile reservoir on Earth. By integrating volatile concentrations with estimates of crustal thicknesses and lithological proportions, we provide constraints on the volatile budget of the Talkeetna lower crust, and assess its volatile storage capacity. This study provides a point of comparison of volatile budgets to other fossil arcs, such as Kohistan (Urann et al., 2022), and investigates the role of the lower crust as a long-term volatile reservoir.
How to cite: Sanchez, M., Le Roux, V., Piani, L., Barry, P. H., and Hudak, M. R.: The lower crust as a volatile reservoir: new constraints from the Talkeetna arc (Alaska), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11694, 2026.
Tungsten isotopes of subducted materials have been extensively investigated at sub-arc depths, yet W recycling into the deeper mantle remains poorly constrained. We present the first stable W isotope data from Cenozoic intraplate basalts in eastern China. These basalts display light W isotopes (δ186W = 0.020–0.074‰) compared to MORB, complementing to most arc magmas. Coupled with low W/Th but high Nb/U, TiO2/Al2O3, and Ce/Pb ratios, these signatures indicate a deep mantle source modified by the stagnant Pacific oceanic crust in the mantle transition zone. Systematic trends of decreasing δ186W with increasing (87Sr/86Sr)i but decreasing εNd(t) and Ce/Pb, from basanites to tholeiitic basalts, reflect deep mantle heterogeneity and increasing contributions from subducted sediments. We suggest melts derived from the recycled residual slab metasomatize the big mantle wedge, generating the low-δ186W basalts. Therefore, W isotopes in continental basalts provide a powerful means to trace the recycled crustal components into deep mantle.
How to cite: Ma, L.: Tungsten isotopes of intraplate basalts and implications for deep W recycling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18383, 2026.
The determination of aqueous fluid-rock partitioning of trace elements has progressed from (1) simple ‘before and after” analysis of solids, assuming that the discrepancy is dissolved in the fluid, through (2) separation of fluid solute from rock in double capsules and (3) separation in diamond traps and consequent analysis of the solute, to (4) cryoablation and analysis of fluid and its solute in diamond traps in the frozen state [1,2]. The last of these is now accepted as the best method currently available, but very few results are available and most of these have concentrated on eclogite in subduction zones and on H2O and Cl fluids.
Appreciable amounts of CO2 and N2 may be present in fluids, particularly in the upper mantle beneath stable continents, but their effect on mobilising trace elements and acting as metasomatic agents has not been quantified. The host rocks through which fluids flow beneath continents include pyroxenites and hydrous assemblages that may differ from those in subduction zones. We present experimentally determined fluid/rock partition coefficients (Df/r) for peridotite and pyroxenite assemblages in equilibrium with a variety of fluid compositions (H2O, H2O+CO2, H2O+CO2+NaCl, H2O+NaCl, and H2O+NH3) for a large range of trace elements at 1.5 GPa pressure and 800 °C. Experimental fluids were separated from the rocks at high pressures using a glassy carbon trap, which has better ablation characteristics than diamond but remains similarly inert. The fluid and solutes in the trap were analysed by cryocell laser ablation ICP-MS in the frozen state.
We show that these mixed aqueous fluids have higher Df/r for pyroxenite than for peridotite. Df/r for the LILE are 10-100 times higher in pyroxenite than peridotite, especially in H2O+NH3 fluids. We confirm earlier conclusions under different conditions that saline fluids dissolve more LILE than pure H2O. Df/r for the HFSE are low and we do not see high Df/r or strong fractionation of LREE from HREE. Cu, Zn and Ni have the highest Df/r amongst the first-row transition elements. Pt and Re have higher Df/r than the HFSE and are most mobile in non-saline aqueous fluids. The fluid composition affects key geochemical ratios: in peridotite > while in pyroxenite ≈ . In pyroxenite assemblages with H2O + CO2 fluids > , whereas in pure H2O, < . The drastic lowering of Df/r for many elements with decreasing pressure probably leads to an optimal pressure-temperature window for fluid-induced metasomatism in the upper mantle. This will vary in depth depending on geodynamic setting (geothermal gradient) and the lithologies present.
[1] Kessel et al. (2005) Nature 437, 724-727.
[2] Rustioni et al. (2019) Geochemical Perspectives Letters 11, 49-54.
How to cite: Foley, S., Phillips, M., and Shcheka, S.: Experimental fluid/rock partition coefficients for H2O with CO2, NaCl and NH3 fluids with peridotite and pyroxenite by cryoablation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15599, 2026.
An increase in fluorine concentration significantly enhances rim growth rates between periclase and wollastonite, as previously reported by Franke and Joachim-Mrosko (2022). However, the mechanisms responsible for this acceleration, as well as the quantitative relationship between volatile content and rim growth dynamics, remain only partially understood. A more detailed understanding of these processes is critical for our understanding of the role of volatiles in metamorphic and metasomatic reactions.
To address this, we conducted high-p-T experiments at 1000 °C and 1.5 GPa for 20 minutes, with fluorine contents ranging from 1 to 2 wt%. For example, at 1 wt% F, rim growth reached an overall thickness of approximately 50 µm. Our results indicate a stagnating reaction rate between 1-3 wt% F. This leads to rim growth dynamics being divided into three distinct regimes:
In regime 1 (0-1wt% F) fluorine gradually replaces OH at grain boundaries within the reaction rim leading to increasing reaction rates. This also leads to fluorine accumulating at the rim-wollastonite boundary. In regime 2 (1-3wt% F) fluorine has almost completely replaced OH leading to further fluorine addition not affecting reaction rates. In regime 3 (>3wt% F) increasing rim growth rates can be explained by microstructural changes from a mosaic to a lamellar structure, as well as the formation of a pore network.
This implies that the quantitative effect of fluorine on elemental mobility along grain and phase boundaries in a reaction rim is a more complicated relationship, whose rate and influence can be dependent on multiple parameters such as chemical composition, temperature and pressure. Thus, specifically designed experiments are required to use reaction rims in natural systems as geofluidometers.
References:
Franke, M. G., Joachim-Mrosko, B. (2022). The effect of fluorine on reaction-rim growth dynamics in the ternary CaO-MgO-SiO2 system. American Mineralogist, 107(8), 1477–1486
How to cite: Nikolaizig, C., Weber, L., and Joachim-Mrosko, B.: The effect of fluorine on rim growth dynamics in the ternary CaO-MgO-SiO₂ system, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3866, 2026.
Water is the most abundant volatile dissolved in magmas and exerts a major influence on the physical and chemical behavior of silicate melts and glasses. Even small amounts can strongly modify viscosity, phase relations, and crystal nucleation, and affect properties such as heat capacity and phase separation. Because of its broad impact on geological processes – from partial melting to magma ascent and crystallization – accurate quantification of water in silicate melts is essential.
Fourier-transform infrared spectroscopy (FTIR) is a widely used technique for determining water content in both natural and synthetic glasses. Its high sensitivity, non-destructive nature, and capability to analyze microscopic regions make it suitable for a broad range of samples. FTIR also allows distinguishing between different hydrogen-bearing species, such as hydroxyl groups and molecular water, through characteristic absorption bands. Silicate glasses exhibit multiple infrared bands associated with hydrogen, three of which are commonly used to quantify water: combination bands of H2O and OH species at 5200 cm-¹ and 4500 cm-¹, respectively, and the fundamental O–H stretching band at 3530 cm-¹. Quantifying water from these bands requires knowledge of molar absorption coefficients, which depend on glass composition. Previous calibrations focused on relatively polymerized compositions with non-bridging oxygen per tetrahedral cation ratios (NBO/T) of 0–0.8, leaving highly depolymerized melts poorly constrained.
To address this gap, we measured molar absorption coefficients in highly depolymerized hydrous peridotitic glasses (NBO/T ≈ 2.5). These glasses were synthesized using a rapid-quench multi-anvil technique, which preserve water in compositions previously inaccessible. Absolute hydrogen contents (0.3–4.7 wt.% H2O) were determined independently using elastic recoil detection analysis, providing a robust basis for FTIR calibration. We determined both linear and integrated molar absorption coefficients of combination bands at 5200 cm-¹ and 4500 cm-¹, and the fundamental O–H stretching band.
We compiled and critically assessed over 350 published values of molar absorption coefficients across a wide range of glass compositions, creating the largest database of its kind. Combining these data with our new measurements for highly depolymerized peridotitic glasses, we identify systematic correlations between molar absorption coefficients and compositional parameters, such as SiO2 wt.% and mol.%, excess modifiers, and the Al+Si/cations ratio, with coefficients generally decreasing as melts become more depolymerized. These trends enable reliable prediction of absorption coefficients for both combination and fundamental bands, extending FTIR calibrations to ultramafic melts and providing a consistent framework for accurate water quantification across diverse silicate glasses.
How to cite: Bondar, D., C. Withers, A., Di Genova, D., Zandonà, A., Bureau, H., Khodja, H., Kurnosov, A., Fei, H., and Katsura, T.: Quantifying water in silicate glasses using FTIR: Extending calibrations to highly depolymerized compositions, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8807, 2026.
