# Fileset

[2025_Noda_et_al_JGRPlanets.pdf](https://mdr.nims.go.jp/filesets/47794f7a-7a93-4329-8459-60b4fc58a2d3/download)

## Creator

Natsumi Noda, Yasuhito Sekine, Yoshio Takahashi, Keisuke Fukushi, [Hiroshi Sakuma](https://orcid.org/0000-0002-6522-0704), Takahiro Kawai, Mayuko Nakagawa, Norio KItadai, Kristin Johnson-Finn, Shawn Erin McGlynn

## Rights

[Creative Commons BY-NC Attribution-NonCommercial 4.0 International](https://creativecommons.org/licenses/by-nc/4.0/)

## Other metadata

[Hydrogen generation from ferrous saponite in reaction with H₂S-containing fluid: relevance to early Martian habitability](https://mdr.nims.go.jp/datasets/fec05e50-dda1-4dfd-bed0-1da85d2bcdff)

## Fulltext

Hydrogen Generation From Ferrous Saponite in Reaction With H2S‐Containing Fluid: Relevance to Early Martian HabitabilityHydrogen Generation From Ferrous Saponite in ReactionWith H2S‐Containing Fluid: Relevance to Early MartianHabitabilityNatsumi Noda1,2 , Yasuhito Sekine1,3,4 , Yoshio Takahashi2 , Keisuke Fukushi3 ,Hiroshi Sakuma5 , Takahiro Kawai2, Mayuko Nakagawa1,6, Norio Kitadai7,Kristin Johnson‐Finn1,8 , and Shawn Erin McGlynn1,9,101Earth‐Life Science Institute (ELSI), Institute of Science Tokyo, Tokyo, Japan, 2Department of Earth and PlanetaryScience, The University of Tokyo, Tokyo, Japan, 3Institute of Nature and Environmental Technology, KanazawaUniversity Kakuma, Kanazawa, Japan, 4Planetary Plasma and Atmospheric Research Center, Tohoku University Aoba,Sendai, Japan, 5National Institute for Materials Science, Ibaraki, Japan, 6Department of Earth and Planetary Science,Institute of Science Tokyo, Tokyo, Japan, 7Institute for Extra‐cutting‐edge Science and Technology Avant‐garde Research(X‐star), Japan Agency for Marine‐Earth Science and Technology (JAMSTEC), Yokosuka, Japan, 8RensselaerAstrobiology Research and Education (RARE) Center, Rensselaer Polytechnic Institute, Troy, NY, USA, 9Center forSustainable Resource Science, RIKEN, Wako, Japan, 10Blue Marble Space Institute of Science Seattle, Washington,DC, USAAbstract Molecular hydrogen is an important gas species for understanding the early Martian climate andredox chemistry. Through ancient aqueous alterations of crustal rocks, ferrous (Fe(II)) saponite formedabundantly on Mars. Subsequent intrusions of hydrothermal fluids may have resulted in a chemical reactionbetween the dissolved volatiles and the nearby rocks. Here we propose a new H2 generating reaction betweenferrous saponite and H2S‐containing fluids, which is possible on early Mars. A series of hydrothermalexperiments at a relatively low temperature of 90°C were performed under anoxic conditions using synthesizedferrous saponite to compare the resulting H2 concentration among various gas and fluid compositions. Based onthe relationship with the existence of H2S, reaction time, fluid pH, dissolved iron concentration, and amount ofminerals, we found that high levels of H2 (∼0.1 mmol/g ferrous saponite) were generated in the presence of H2Smost rapidly in moderate pH conditions. Our microscopic chemical analysis of mineral phases suggested thatferrous saponite served as both the iron source of pyrite precipitation and the electron source to form H2. Ourresults suggest that intrusions of H2S‐containing fluids into the saponite‐containing crust of Mars wouldgenerate H2, which could potentially provide locally concentrated chemical energy for chemoautotrophic life.Plain Language Summary How did early Mars maintain liquid water on its surface? How muchchemical energy was available in the aqueous environments? Molecular hydrogen (H2) is a key molecule thatcan address these questions concerning Martian habitability because it would have been an effective greenhousegas and could also provide chemical energy for primitive microbial life. While previous studies have suggestedthat bolide impacts, volcanoes, and iron oxidation in surface water could have generated H2 on Mars, wepropose a new mechanism that generates H2 through the interactions of clay minerals with hydrothermal fluids.Clay minerals are widely distributed in the Martian crust we observe today. Our experiments suggest that ifhydrothermal fluids containing volcanic hydrogen sulfide (H2S) had intruded into the clay‐bearing crust, H2would have been generated.1. IntroductionMars has been a unique extraterrestrial target for exploring aqueous chemistry and planetary habitability sincegeological evidence for the prolonged presence of liquid water on its surface dating back to ∼4 billion years ago(Ga) was reported (e.g., Howard et al., 2005; Fassett & Head, 2008). Despite geological evidence, climate modelshave indicated that achieving a continuously warm climate on early Mars is challenging without a significantsupply of greenhouse gas into the atmosphere considering the faint young Sun and the condensation of dense CO2atmosphere (Forget et al., 2013; Kasting, 1991; Wordsworth et al., 2013). In addition to the greenhouse effects ofatmospheric CH4, NH3, and SO2 (Haberle, 1998; Johnson et al., 2008), recent studies have proposed that thepresence of H2 at high levels of more than several % could effectively warm the surface via collision inducedRESEARCH ARTICLE10.1029/2024JE008538Key Points:• Chemical reactions between ferroussaponite and volatile‐containing hy-drothermal fluids are investigated• In our experiments, molecularhydrogen was generated by thereaction of ferrous saponite andhydrogen sulfide, forming iron sulfide• Interactions of crustal ferrous saponitewith hydrothermal fluids may haveprovided H2 on early Mars and otherwet rocky planetsCorrespondence to:N. Noda,natsumi@elsi.jpCitation:Noda, N., Sekine, Y., Takahashi, Y.,Fukushi, K., Sakuma, H., Kawai, T., et al.(2025). Hydrogen generation from ferroussaponite in reaction with H2S‐containingfluid: Relevance to early Martianhabitability. Journal of GeophysicalResearch: Planets, 130, e2024JE008538.https://doi.org/10.1029/2024JE008538Received 25 JUN 2024Accepted 28 NOV 2024Author Contributions:Conceptualization: Natsumi Noda,Yasuhito SekineData curation: Natsumi NodaFormal analysis: Natsumi Noda,Takahiro Kawai, Mayuko NakagawaFunding acquisition: Natsumi Noda,Yasuhito SekineInvestigation: Natsumi NodaMethodology: Natsumi Noda,Yoshio Takahashi, Keisuke Fukushi,Hiroshi Sakuma, Takahiro Kawai,Mayuko Nakagawa, Norio Kitadai,Kristin Johnson‐FinnProject administration: Yasuhito SekineResources: Yoshio Takahashi,Norio KitadaiSoftware: Yoshio TakahashiSupervision: Yasuhito SekineVisualization: Natsumi NodaWriting – original draft: Natsumi Noda,Yasuhito Sekine© 2024 The Author(s).This is an open access article under theterms of the Creative CommonsAttribution‐NonCommercial License,which permits use, distribution andreproduction in any medium, provided theoriginal work is properly cited and is notused for commercial purposes.NODA ET AL. 1 of 17https://orcid.org/0000-0002-2922-8207https://orcid.org/0000-0002-3544-3632https://orcid.org/0000-0001-7860-2341https://orcid.org/0000-0003-0398-8950https://orcid.org/0000-0002-6522-0704https://orcid.org/0000-0001-6015-8617https://orcid.org/0000-0002-8199-7011mailto:natsumi@elsi.jphttps://doi.org/10.1029/2024JE008538http://creativecommons.org/licenses/by-nc/4.0/http://creativecommons.org/licenses/by-nc/4.0/http://crossmark.crossref.org/dialog/?doi=10.1029%2F2024JE008538&domain=pdf&date_stamp=2024-12-25absorptions (CIAs) with CO2 atmosphere (e.g., Ramirez et al., 2014; Batalha et al., 2015; Turbet et al., 2017,2020; Wordsworth et al., 2017, 2021). A continuous H2 supply into the Martian atmosphere is required tomaintain a commensurate level against hydrogen escape into space (Batalha et al., 2015; Ramirez et al., 2014).One promising H2 source on early Mars is volcanic outgassing from the reduced mantle (Gaillard et al., 2013;Righter et al., 2008). Asteroidal impacts (Haberle et al., 2019), serpentinization of mafic igneous rocks (Chas-sefière et al., 2013, 2016), magnetite authigenesis within lakes (Tosca et al., 2018), and radiolysis of H2O ice(Tarnas et al., 2018) have also been discussed as sources of H2. Investigating additional H2 formation mecha-nisms, which could have existed in parallel, is important for reconstructing the climate on early Mars.In addition to the effect on climate, H2 could have also played an important role in habitability on Mars byproviding chemical energy for chemoautotrophic life. Redox reactions between H2 and CO2 provide metabolicenergy for methanogens and acetogens (e.g., McCollom & Shock, 1997; Fuchs, 2011). For example, the growthof methanogens in laboratory experiments is supported by H2 slightly lower than 100 ppm with a reaction energyof∼ − 20 kJ/mol (Cord‐Ruwisch et al., 1988; Zinder, 1993). Given the observations that metabolisms of terrestriallife can be supported by − 10 kJ/mol, the potential biological threshold is as low as ∼10 ppm H2, and can beincreased depending on other environmental factors such as pH and CO2 levels (Hoehler et al., 2001;Zinder, 1993). In addition to CO2, a variety of oxidants, such as oxyhalogens (Hecht et al., 2009; Quinnet al., 2013), O2 (Koyama et al., 2021; Lanza et al., 2016; Noda et al., 2019), O3 (Chaffin et al., 2017), SO3, andNO2 (Zolotov & Mironenko, 2007), might have been generated on early Mars via photochemical reactions.Reactions of H2 with these oxidants could have provided chemical energy for chemoautotrophic life in a similarway as occurs in some terrestrial hot spring environments, where H2 oxidation is coupled with species includingO2, NO3− , FeIII, AsV, S°, SO42− , and CO2 (Inskeep et al., 2005).Here we propose a new process of H2 generation on early Mars, focusing on ferrous (Fe(II)) saponite, a smectiteclay mineral, in the crustal rocks as a reductant for the formation of H2. The global distribution and chemicalcomposition of Mg/Fe smectite within the Martian crust have been determined using orbital observations (Carteret al., 2013; Ehlmann et al., 2008, 2009, 2011; Fukushi et al., 2019, 2022; Michalski et al., 2015; Mustardet al., 2008; Noda et al., 2022; Poulet et al., 2005; Rampe et al., 2020). A comparison of laboratory analog analysisand remote sensing data indicated that there would be two types of predominant Mg/Fe smectite on Mars: Fe(II)‐bearing trioctahedral smectite (ferrous saponite) and Fe(III)‐bearing trioctahedral smectite (ferrian saponite),which can be produced by oxidation of ferrous saponite (Chemtob et al., 2017; Michalski et al., 2015). TheCuriosity rover has also detected ferrian saponite in the sedimentary rocks of Gale Crater using an X‐raydiffraction (XRD) spectrometer, CheMin (Bristow et al., 2015; Treiman et al., 2014; Vaniman et al., 2014).These Mg/Fe smectites are interpreted to have mostly formed ferrous saponite through early aqueous alterationsof the Fe‐rich mafic composition of the Martian crust and then partially experienced oxidation to form ferriansaponite in the later geological history of Mars (e.g., Catalano, 2013; Ehlmann et al., 2011; Michalski et al., 2013).The global formation of ferrous saponite under a steam atmosphere during magma ocean cooling is also suggestedby experiments at H2O‐CO2 super critical condition (i.e., temperature and pressure condition of 425°C and300 bar; Cannon et al., 2017).Although aqueous reactions involving various Fe(II)‐bearing minerals expected on early Mars, such as green rust,have been discussed (Tosca et al., 2018), it is unknown whether ferrous saponite can act as a reductant for protonreduction to form molecular hydrogen. Despite the relevance and occurrence of ferrous saponite on Earth (Badautet al., 1985; Kohyama et al., 1973), investigations of its ability to act as a reductant have been limited because ofits sensitivity to oxidation by O2 in the atmosphere (Noda et al., 2021; O’Loughlin et al., 2020). Several previousstudies have examined the formation and alteration processes of ferrous saponite on early Mars using synthesizedferrous saponite (e.g., Chemtob et al., 2017; Noda et al., 2021; Rivera Banuchi et al., 2022; Sakuma et al., 2022);however, H2 production from ferrous saponite has not been discussed. The global occurrence of deep hydro-thermal circulation reactive with the surrounding crust of early Mars has been proposed according to geomor-phological (e.g., fluvial features at deep craters) and mineralogical (e.g., clay units exposed at heavily erodedterrain) records and supported by hydrological calculations (Andrews‐Hanna et al., 2007; Ehlmann et al., 2011;Kite & Melwani Daswani, 2019; Michalski et al., 2013; Ojha et al., 2020; Salese et al., 2019). Thus, the crustalferrous saponite could have interacted with an intruded anoxic hydrothermal fluid that dissolves degassingcomponents such as carbon and sulfur volatiles.Writing – review & editing:Yasuhito Sekine, Yoshio Takahashi,Keisuke Fukushi, Hiroshi Sakuma,Mayuko Nakagawa, Norio Kitadai,Kristin Johnson‐Finn, ShawnErin McGlynnJournal of Geophysical Research: Planets 10.1029/2024JE008538NODA ET AL. 2 of 17 21699100, 2025, 1, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JE008538 by National Institute For, Wiley Online Library on [26/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons LicenseIn this study, we experimentally investigated the interactions between ferrous saponite and volatile (i.e., CO2 andH2S) containing fluid. Hydrothermal experiments were conducted under anoxic conditions with various fluidcompositions and pH values to constrain mechanisms of H2 generating reactions.2. Methods2.1. Synthetic Ferrous SaponiteThe ferrous saponite samples used in this study were synthesized using a sol‐gel method within decompositionvessels (HU‐100, San‐Ai Kagaku Co., Ltd., Aichi, Japan) in an acrylic glovebox connected to a deoxygenationsystem (SiOC‐2000 GB; STLab Co., Ibaraki, Japan), as described elsewhere (Noda et al., 2021). Anoxic con-ditions with an oxygen level (PO2) of <0.01 ppb and humidity ∼50% were maintained throughout all samplepreparations and collection procedures. Ultrapure water (Milli‐Q; Millipore) was bubbled with Ar gas (>99.999vol.%, Suzuki Shokan, Kanagawa, Japan) at 2.8 L min− 1 for 3 hr in the glovebox prior to use so as to reduce itsdissolved oxygen (DO) below 0.01 ppm (Noda et al., 2021). Throughout the synthesis procedure, oxidation ofFe2+ was effectively prohibited: the characterized saponite was only ∼4% Fe3+/ΣFe (Noda et al., 2021).The ideal half‐cell formula described as Na0.5(Mg1.5Fe2+1.5)(Si3.5Al0.5)O10(OH)2 corresponds well to orbiters'and rovers' analyses of Martian clay minerals (Michalski et al., 2015). We aimed at achieving this composition bymixing the reagents (Na4SiO4, FeSO4 · 7H2O, MgCl2 · 6H2O, and AlCl3 · 6H2O; Noda et al., 2021) in a propermolar ratio. The intended Fe/Mg M ratio (∼1.0) was confirmed using scanning electron microscopy and energy‐dispersive X‐ray spectrometry (SEM‐EDS; S‐3400N; Hitachi High‐Tech Corp., Japan). The synthesized ferroussaponite was dried and powdered in the glovebox using an agate mortar so that the typical grain size ranges1–100 μm, followed by sealing within high‐barrier nylon/polyethylene storage bags with oxygen absorbers(Oxygen‐Free, Dried Storage System, A‐500‐50S; I.S.O. Inc., Kanagawa, Japan).2.2. Anoxic Gas–Fluid–Mineral‐Phase ReactionCrimp‐top glass vials (effective volume of approximately 70 mL; Vial Bottle No. 7; Maruemu Corp., Osaka,Japan) sealed with Al‐capped Teflon‐laminated butyl stoppers (5‐112‐07; AS ONE Corp., Osaka, Japan) wereused as reaction vessels to encapsulate various mineral, fluid, and gas phases as initial conditions (Table 1). Boththe glass vials and stoppers were pre‐cleaned by soaking in ∼4 M HNO3 at room temperature (∼25°C) for 70 hrand then in ultrapure water for 60 hr. The glass vials were then heated in an electric furnace (FO410; YamatoScience, Tokyo, Japan) at 450°C for 12 hr.Solutions of sodium bicarbonate were prepared in the glove box from the deoxidized ultrapure water (DO < 0.01ppm) for the fluid phases. Sodium bicarbonate (NaHCO3, Guaranteed Reagent; Wako Pure Chemical Industries)was used because dissolved bicarbonates and CO2 gas can act as pH buffers (i.e., bicarbonate‐carbonate buffersystem). After the crimp‐sealing of the vials with typically 100± 3 mg of mineral powders and ∼10 mL of fluids,the headspace was purged with either pure Ar (>99.999 vol.%; Suzuki Shokan) or CO2 gas (>99.999 vol.%;Suzuki Shokan) at 100 mL min− 1 for 5 min. The headspace gas phase accounted for ∼60 ml in a ∼70 ml vial.In some of the experiments, H2S was introduced into glass vials containing minerals and fluids (see Section 2.3).H2S was prepared based on the methodology shown in the previous work (Kitadai et al., 2021) with the followingreaction:Na2S + 2HCl → 2NaCl + H2S. (1)To make H2S, precleaned vials containing 1.0 g of hydrated sodium sulfide (NaS2 · 9H2O; Kishida Chemical Co.Ltd., Osaka, Japan) were evacuated to <50 Pa and sealed in advance using a drying chamber (FDU‐1200, BSC‐2L; Tokyo Rikaki Co., Ltd., Tokyo, Japan). Then, ∼1.2 mL of 6 M HCl (Wako Pure Chemical Industries) wasadded so that the vial volume of ∼70 mL would be filled with generated H2S gas at the ambient temperature andpressure. Using a gas‐tight syringe (SGE Analytical 10MDR‐VLLMA‐GT; Trajan Scientific Japan, Kanagawa,Japan), 10 mL of the generated H2S gas was injected into each experimental vial after removing the same volumeof purged CO2 or Ar gas.All the crimped vials containing mineral, fluid, purged gas, and H2S were moved from the glovebox into a forced‐convection drying oven (ETTAS OFW‐450V; AS ONE) kept at a constant temperature of 90 ± 1°C. GeothermalJournal of Geophysical Research: Planets 10.1029/2024JE008538NODA ET AL. 3 of 17 21699100, 2025, 1, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JE008538 by National Institute For, Wiley Online Library on [26/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licenseestimates on the Noachian Martian crust have suggested that hydrothermal alteration at 90°C is presumable at afew to ∼5 km depth from the near‐freezing surface (Michalski et al., 2013; Ojha et al., 2020). By referencingprevious hydrothermal experiments with iron‐bearing minerals using similar experimental setups (Bargeet al., 2019; Heinen & Lauwers, 1996), we considered 90°C to represent the reaction procedure in the experi-mental timescale of the present study. For comparison, two vials with ferrous saponite and H2S injection wereplaced at the ambient temperature (∼23°C). To minimize gas leaks and air contaminations during the incubation,the vials were placed in the oven with the top side facing downward to ensure that the fluid covered the rubberstoppers.2.3. Experimental ConditionsFirst, we conducted comparative hydrothermal experiments on various mineral types. In these experiments, wefound H2 production specific for the co‐existence of ferrous saponite and gaseous H2S (see Section 3.1 below).Table 1Initial Conditions for Respective Groups of ExperimentsNumber of runs Duration (h)Mineral Fluid Headspace gasFluid pHeMass (mg)a Mass (g)b NaHCO3 (mM)c HCl (mM)c Purged gas H2S injectiondControl 1 3 863–975 − 10.2 − − Ar + 5.7–6.2Control 2 3 420–421 − 10.1 1.00 × 102 − CO2 + 7.0–7.21 94Mineral speciesFerrous saponite 3 429–438 100 10.4 1.00 × 102 − CO2 + 7.2–7.3Magnetite 3 422–438 102 10.4 1.00 × 102 − CO2 + 7.0–7.71 1,533Troilite 3 420–422 101 10.0 0.99 × 102 − CO2 + 7.1–7.6Iron‐nickel alloy 3 303–305 100 10.2 0.99 × 102 − CO2 + 7.1–7.2Ferrous saponite 3 420–443 96 10.0 1.00 × 102 − CO2 − 7.0–7.2Magnetite 3 438–443 98 10.0 1.00 × 102 − CO2 − 7.0–7.51 1,500Troilite 3 420–422 102 10.0 0.98 × 102 − CO2 − 7.0–7.1Iron‐nickel alloy 3 303–305 101 10.1 0.98 × 102 − CO2 − 7.1–7.2GroupI 5 34–1,791 100 10.0 − − Ar + 8.4–9.4i 7 29–1,774 27 10.0 − − Ar + 6.8–7.5II 5 34–1,791 101 10.1 1.00 × 102 − CO2 + 7.0–7.2ii 7 29–1,780 27 10.1 1.00 × 102 − CO2 + 7.0–7.2III 5 35–1,740 101 10.0 − − CO2 + 6.3–6.8iii 7 29–1,738 27 10.1 − − CO2 + 5.7–6.1IV 5 29–1,774 100 10.0 0.98 × 102 0.98 × 102 CO2 + 6.2–6.8iv 5 29–1,749 26 10.1 0.99 × 102 0.50 × 102 CO2 + 6.7–6.8V 6 29–1,780 100 10.0 − − CO2 − 6.3–6.6v 4 29–1,355 27 10.1 1.00 × 102 − CO2 − 7.0–7.3Note. The experiments for comparison among different mineral species (above) and for comparison of time evolution among various initial conditions (below). As forthe latter, the resulted fluid pH measured after the experiment are also included. aThe weighing accuracy was ±3 mg. bThe accuracy was within ±0.3 g. cThe accuracywas estimated to be ±2% of the value. The symbol “− ” indicates that the species was not included. dThe symbol “+” indicates that 10 mL in ∼60 mL of purged gas wasexchanged to H2S gas, while the symbol “− ” indicates that the whole headspace remained filled with purged gas. eThe lowest and highest values among the independentruns in the group with different reaction times are shown. The accuracy of each pH measurement was ±0.1.Journal of Geophysical Research: Planets 10.1029/2024JE008538NODA ET AL. 4 of 17 21699100, 2025, 1, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JE008538 by National Institute For, Wiley Online Library on [26/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons LicenseThen, we explored the time evolution of the H2 production via interactions with ferrous saponite with H2S‐containing fluids with different gas and fluid compositions to investigate the reaction mechanisms.In the comparative hydrothermal experiments, 4 types of Fe‐bearing minerals, that is, synthesized ferroussaponite, magnetite (Fe3O4, 95.0+%; Wako Pure Chemical Industries, Osaka, Japan), troilite (FeS, PracticalGrade; Wako Pure Chemical Industries) and iron‐nickel alloy (Fe:Ni = 90:10 wt.%, 99.9% purity; KojundoChemical Laboratory, Saitama, Japan), were compared with and without H2S (Table 1). Two types of controlexperiments without minerals were also conducted: one with Ar gas, pure water, and H2S injection, and the otherwith the bicarbonate‐carbonate buffer system and H2S injection (Table 1). The headspace gas composition wasanalyzed typically after∼400 hr of incubation. In addition to three time runs each, one with different reaction timewere analyzed in some conditions. The Fe‐bearing materials other than ferrous saponite (magnetite, troilite, andFe‐Ni alloy) correspond to (hydr)oxide, sulfide, and metal possibly available on Mars, whose aqueous reactionshave been discussed in simulating paleoenvironments of Mars and Earth (e.g., Haberle et al., 2019; Heinen &Lauwers, 1996; Tosca et al., 2018). The Curiosity rover detected magnetite and a low level of pyrite within thelake deposition (Vaniman et al., 2014). The reaction between pyrite and H2S has been demonstrated and proposedas a plausible H2 source in the reducing paleoenvironment on Earth (Heinen & Lauwers, 1996). Iron contained inmeteorites could have been oxidized by water, providing significant H2 into the early Martian atmospheretemporarily after impacts (Haberle et al., 2019; Wordsworth et al., 2021).In terms of the time evolution of H2 production from ferrous saponite, 10 groups of experiments (groups I–V andi–v) with different gas and fluid compositions were compared (Table 1). Groups i–v contained lower amounts offerrous saponite (27 mg) than groups I–V (100 mg) (Table 1). In group I and i experiments, Ar gas was introducedinstead of CO2. In the group II–IV and ii–iv experiments, CO2, NaHCO3 and/or HCl were introduced in differentamounts so that different pH and ionic strengths were tested (Table 1). Group V and v experiments contained noH2S to compare against samples containing sulfide (Table 1). Each group comprised four to seven individual runswith reaction times ranging from 30 to 1,800 hr. Analysis of each run (i.e., gas measurement and collection offluid and mineral samples) was done only once after the given reaction time with a few exceptions in respectivegroups for gas measurement. Blank experiments without minerals, CO2, and H2S were performed in duplicatewith long reaction times (∼900 hr) to check the background detection levels of gas species from the reaction ofwater with the glass vials.2.4. Gas and Liquid AnalysesThe concentrations of H2 in the headspace gas of the glass vials were determined using Gas Chromatography withBarrier‐discharge Ionization Detector (GC‐BID; GC‐2010 Plus, BID‐2010 Plus; Shimadzu, Kyoto, Japan) andMicropacked‐ST column (2.0 m × 1.0 mm I.D.; Shinwa chemical Industries), with an initial temperature of 35°C(held 9 min) ramped to 100°C at 16°C min− 1 (held 4 min), then to 150°C at 22°C min− 1 (held 10 min). Headspacegas was collected and injected manually using a gas‐tight syringe (A‐2; Valco Instruments, TX, USA). Theconcentration of H2 was quantified based on the chromatogram peak area and the injection volume ((40–100) ± 0.5 μL). The error and the lower limit of quantitation were derived from uncertainties in peak detection,calibration line, and injection volume.After incubation and gas analysis, the fluid and mineral samples for subsequent analysis were collected in theglovebox. The mixture of fluid and mineral phases collected with disposable PP syringes (SS‐10LZ; TerumoCorp.; Tokyo, Japan) was separated using a membrane filter (0.2 μm PTFE Membrane JGWP02500 or 0.025 μmMCE Membrane VSWP02500; Merck Millipore Ltd., Darmstadt, Germany) placed in a filter holder (SwinnexFilter Holder Φ25 mm SX0002500; Merck Millipore Ltd., Darmstadt, Germany). Mineral samples gathered onthe filter were vacuum‐dried, placed in microtubes, and sealed in the oxygen‐free storage bags. The pH of thefiltered fluid was measured at room temperature using a pH meter (LAQUA twin pH‐22B; Horiba Co., Ltd.)calibrated prior to use. In the experiments, minerals, including ferrous saponite, partly dissolved, releasing Na, Al,Si, Mg, and Fe into fluids. The concentrations of dissolved Fe and Mg in the fluid were measured usinginductively coupled plasma–atomic emission spectrometry (ICP–AES; SPS5510; SII NanoTech, Tokyo, Japan).The collected fluid samples were diluted ∼8‐fold in precleaned polypropylene vials with 0.1 M HNO3 preparedby diluting 68 wt.% HNO3 ultrapure analytical reagent (TAMAPURE‐AA‐100; Tama Chemicals Co., Ltd.) withultrapure water. The diluted standard solutions (XSTC‐622; SPEX CertiPrep, LLC.) were prepared and analyzedfor calibration.Journal of Geophysical Research: Planets 10.1029/2024JE008538NODA ET AL. 5 of 17 21699100, 2025, 1, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JE008538 by National Institute For, Wiley Online Library on [26/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License2.5. Chemical Analysis of Mineral PhasesThe collected mineral samples were subjected to synchrotron‐based chemicalanalyses in order to visualize changes in the chemical state. In scanningtransmission X‐ray microscopy (STXM), X‐ray absorption near‐edge struc-tural (XANES) mapping with a high spatial resolution of about 50 nm wasperformed. The preparation and analytical procedures for STXM measure-ments were based on our previous study (Noda et al., 2021). Fine grains of thedried mineral samples were mounted on a 400 mesh Cu grid with a supportingfilm (No. 1608; JEOL Ltd., Tokyo, Japan), attached to the sample holder, andstored in a dry oxygen‐free storage system in the glovebox without any airexposure until STXM analysis (Noda et al., 2021). The stacked images ob-tained around the Fe L‐edge (700–732 eV) were accumulated and convertedto extract the XANES spectra at an optical density using the analyticalsoftware aXis2000 (Hitchcock, 2009). The XANES spectra of the bulkpowdered samples were obtained using Beamline 19B at the Photon Factoryfor reference. Both the extracted and bulk XANES spectra were analyzedusing Athena software v. 0.9.26 (Ravel & Newville, 2005).3. Results3.1. Comparison of H2 Formation for Different MineralsFigure 1 compares the measured H2 concentrations with different iron‐bearing minerals and without minerals (controls). Among the tested min-erals, ferrous saponite showed the most significant difference in H2 con-centration between the presence and absence of H2S (Figure 1). The highestlevels of H2 (∼4 × 104 ppm) were produced in the experiment with Fe–Nialloy regardless of H2S injection (Figure 1). This is attributed to the reac-tion of water with metallic Fe (Haberle et al., 2019),Fe0 + H2O → FeO + H2(g) . (2)Low levels of H2 (<102 ppm) were measured in the experiments usingmagnetite after 400 and 1,500 hr of the experimental time (Figure 1). The H2concentrations with H2S was ∼40 ppm higher than those without H2S, whichcan be attributed to H2 contained in the injected H2S: small amounts of H2(<40 ppm in the reaction vial) was found in the prepared H2S gas as shown inthe control experiments (Figure 1), possibly due to the corrosion of the needle used for the injection of HCl. Thereaction between water and magnetite could form H2 and iron hydroxide (e.g., ∼0.1 μmol H2/g mineral at 100°C:Mayhew et al., 2013); however, the fact that the H2 concentration is comparable to those of the control experi-ments supports the idea that the contribution of this H2 generation (∼10 nmol of H2 for 100 mg of magnetite) issmall compared to the total H2 production.H2 concentration in the experiment with troilite was about twofold increased by H2S injection. This can beattributed to the formation of pyrite (FeS2), as observed under terrestrial aqueous conditions (Heinen & Lauw-ers, 1996; Rickard, 1997; Wächtershäuser, 1992; Wan et al., 2017),FeS + H2S → FeS2 + H2(g) . (3)Here, H2S acts as an electron acceptor due to the fact that FeS has the highest occupied molecular orbital (HOMO;+0.5 eV) less stable than the lowest unoccupied molecular orbital (LUMO) of H2S (− 1.1 eV) (Rickard &Luther, 1997). Relatively high H2 concentration (∼1 × 104 ppm) without H2S injection indicates that water‐troilite interaction also forms H2.Figure 1. Comparison of H2 concentrations in the headspace gas among thereactions with various minerals. Control experiments without minerals arealso shown. Colored bars and errors represent the averages and standarddeviations of three repeated experiments, respectively. Their mean reactiontimes (hours (h)) and mineral names are provided at the bottom of x‐axis.The color of the bars corresponds to the presence (pink) or absence (green) ofH2S in the initial headspace gas compositions. Open diamonds indicate theresults from three individual experiments under the same reaction conditionswith quantitation errors. Solid squares show the results with differentreaction times from those shown at the bottom of x‐axis. The initialconditions of the respective experiments are provided in Table 1.Journal of Geophysical Research: Planets 10.1029/2024JE008538NODA ET AL. 6 of 17 21699100, 2025, 1, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JE008538 by National Institute For, Wiley Online Library on [26/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons LicenseExperimental triplicates with synthesized ferrous saponite showed similar H2 concentrations of ∼1 × 103 ppmwith H2S, which is two orders of magnitude higher than those without H2S (Figure 1). In contrast to those withtroilite, H2 concentration was limited (<20 ppm) when no H2S was injected (Figure 1). This significantdependence on the existence of H2S suggests ferrous saponite is involved directly or catalytically in H2 generationin the interaction with H2S. H2 generating reaction was also confirmed in the experiments performed at ambienttemperature (∼23°C) instead of 90°C with the same initial composition. The low H2 concentrations (∼2 × 102ppm after ∼450 hr reaction; Supplementary Table S1) were consistent with the slow reaction rate at the lowtemperature than that at 90°C.3.2. Time Evolution of H2 Concentration in Reaction With Ferrous SaponiteTo further explore the reaction mechanism and the role of ferrous saponite in generating H2, the different reactionconditions, including fluid pH by varying the gas and fluid compositions and mass of ferrous saponite, werecompared (Table 1). Independent 4–7 runs with various reaction times were conducted for each group to measurethe resulting concentrations of gaseous H2 and dissolved Fe and Mg.H2 concentration reached ∼(4− 7) × 103 ppm after the reaction time of 1,750 hr with 100 mg of ferrous saponiteand H2S injection (groups I–IV in Figure 2a). No significant H2 (<50 ppm) was generated over the reaction timeFigure 2. H2 concentrations in the headspace gas as a function of reaction time for (a) the group I‐V experiments (with 100 mgof ferrous saponite) and for (b) the group i‐v experiments (27 mg of ferrous saponite). Respective dot lines connect the datafrom single experiments measured at different reaction times. Except for these, each data point corresponds to a singleexperiment prepared individually and ended at the indicated reaction time. Error bars show the quantitation uncertainties inpeak detection, calibration line, and injection volume. The experimental conditions of the respective groups are provided inTable 1.Journal of Geophysical Research: Planets 10.1029/2024JE008538NODA ET AL. 7 of 17 21699100, 2025, 1, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JE008538 by National Institute For, Wiley Online Library on [26/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licenseof 1,750 hr in the group V experiments without H2S (Figure 2a), confirming that H2S is essential in H2 productionfrom ferrous saponite. The experimental duration required for the H2 production was different among the groupI− IV experiments. Group II showed the most rapid increase in the H2 concentration in the experiment with a fewdays (∼100–200 hr) and reached a plateau at (6–7) × 103 ppm after 600 hr of reaction (Figure 2a). H2 concen-trations also started to increase for a relatively short duration in the group I experiments and reached 8 × 103 ppmafter 1,350 hr (Figure 2a). Neutral pH values (∼7.1) were maintained throughout the reaction time in the group IIowing to the bicarbonate‐carbonate buffer system; meanwhile, the pH of the group I experiments was alkaline (pH∼8–9) without CO2 or NaHCO3 (Table 1). Much longer reaction times were required for the H2 generation ingroup III and IV experiments with weakly acidic pH (pH∼6.2–6.8) (Table 1). In group III and IV experiments, H2concentrations remained low (<7 × 102 ppm) in reaction times shorter than 700 hr and then increased to∼ 4 × 103ppm after 1,750 hr reaction (Figure 2a). These results indicate that the fluid pH affected the kinetics of the H2production reaction with ferrous saponite.In the group i‐v experiments with lower starting amounts (∼27 mg) of ferrous saponite than the group I–V ex-periments (∼100 mg), the H2 concentration of the group i–v experiments became generally lower than that of thegroup I–V experiments (Figure 2b). Groups i, ii, and iv had a moderate pH of∼7, despite differences in gas and fluidcomposition (Table 1). The H2 concentrations in these groups generally increased to∼(1–2)× 103 ppm in the shortreaction time within 250 hr and then decreased to∼7× 102 ppm after a longer duration of >1,000 hr (Figure 2b). Onthe contrary, the group iii experiments at pH∼6 showed a relatively low H2 concentration of (5–7)× 102 ppm in therelatively short experiments of 100–600 hr and increased to >1× 103 ppm after reaction times longer than 1,350 hr(Figure 2b). The observed trend of slower H2 production under acidic conditions compared to neutral conditions isconsistent with the results of larger amounts of ferrous saponite (i.e., the difference between group III and II ex-periments). Little H2 was found in group v experiments for all reaction times tested as no H2S was introduced in theexperiments (Figure 2b).Figure 3 shows the changes in H2 concentration and fluid pH as a function of Mg and Fe concentrations dissolvedin the filtered fluid samples. Low Mg concentrations (<0.1 mM) were measured with an alkaline pH ofFigure 3. Results of (a, b) H2 concentrations and (c, d) fluid pH as a function of the concentrations of (a, c) Mg and (b, d) Fe dissolved in the filtered fluid samples. Thetop panels show data from group I–IV experiments with 100 mg of ferrous saponite, while the bottom panels show group i–iv experiments with 27 mg of ferroussaponite. Each data point represents a single separated experiment that ended at various reaction times (29–1,791 hr). Stars indicate the data from the experiments withthe shortest reaction time (29–34 hr) among each experimental group. Error bars show the quantitation uncertainty of each experiment.Journal of Geophysical Research: Planets 10.1029/2024JE008538NODA ET AL. 8 of 17 21699100, 2025, 1, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JE008538 by National Institute For, Wiley Online Library on [26/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licenseapproximately 9 in group I, while high Mg concentrations (∼2–3 mM) were observed at pH ∼ 6.5 in groups IIIand IV (Figure 3c, top). Among groups i‐iv, group iii with the lowest pH (∼5.8) showed the highest Mg con-centrations (∼1–2 mM) (Figure 3c, bottom). These results show that the Mg concentration is controlled by thefluid pH and is independent of H2 generation (Figures 3a and 3c). The varied Mg concentration in group I was dueto the variation in fluid pH. Since no pH buffer (carbonate‐bicarbonate buffer) was used, the fluid pH of group Ican be varied depending on small differences in the initial materials among the runs.In contrast to Mg, Fe concentrations varied with reaction time and were less correlated with pH (Figure 3d)despite the 1:1 molar ratio of Mg and Fe in our synthesized ferrous saponite. Figure 3b indicates that the fluids ofgroups I–IV shared similar Fe concentrations of ∼10− 2 mM at the beginning of the reaction. Then, in the groups Iand II experiments, Fe concentrations decreased to ∼10− 3 mM in response to increasing H2 concentrations(Figure 3b, top), suggesting that dissolved Fe could have been consumed by H2 generating reaction in theseexperiments. Fe concentrations increased to >0.1 mM in the group III and IV experiments without increases in H2concentrations (Figure 3b, top), suggesting that the dissolution of Fe in ferrous saponite did not trigger rapid H2generation. Significant H2 detection with low dissolved Fe concentration (<∼10− 3 mM) at neutral pH conditionwas also confirmed starting with the smaller amount of saponite (Figure 3b, bottom).Figure 4. Result of STXM analysis. (a–c) Normalized XANES spectra are calibrated at the absorption edge energy, E₀, of 707.4 eV. (a) Spectra extracted from the areashown in panel (d). Vertical dashed lines correspond to 708 and 710 eV. (b) Bulk XANES spectra of synthesized ferrous saponite stored under anoxic conditions, ferroussaponite oxidized by ambient air, and various Fe‐bearing reference minerals. (c) XANES spectra of the endmember components used for SVD mapping of STXMstacked images. (d–f) Results of STXM measurements performed for the mineral sample collected after ∼600 hr of reaction for the group ii experiment. (d) Opticaldensity map at a single X‐ray energy of 707.4 eV. (e) Residuals of SVD mapping of the stacked image shown in optical density. (f) Results of SVD mapping visualizedin RGB color. The relative contributions of the three components in panel (c) are shown, with ferrous saponite in green, air‐introduced Fe(III)‐bearing saponite in red,and pyrite in blue.Journal of Geophysical Research: Planets 10.1029/2024JE008538NODA ET AL. 9 of 17 21699100, 2025, 1, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JE008538 by National Institute For, Wiley Online Library on [26/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License3.2.1. Characterization of Reactions in the Solid PhaseSTXM analysis was performed for the dried mineral samples retrieved from group ii (Figure 4) and II(Figures 5a–5c) experiments, where neutral pH was maintained during rapid H2 generation. For comparison, theSTXM analysis was also performed for the samples of the group V experiment (Figure 5d), where the absence ofH2S resulted in no H2 detection. The grains with the proper optical thickness were searched and chosen to obtainstacked images.The strong absorptions in the extracted XANES spectra at around 708 and 710 eV are generally attributed toferrous and ferric iron, respectively. Most regions exhibited XANES spectra similar to the bulk XANES spectra ofsynthesized ferrous saponite obtained without any air exposure (area 1 and 2 in Figures 4a and 4d, in comparisonwith the line labeled “Ferrous saponite” in Figure 4b), while some areas showed relatively high absorption ataround 710 eV (area 3 in Figures 4a and 4d). The absorption at 710 eV was significant in the bulk XANES spectraof ferrous saponite exposed to O2 by introducing ambient air in the reaction vial instead of Ar or CO2 togetherwith ∼10 mL of pure water and kept at 90°C for ∼900 hr (the line labeled with “Air‐introduced” in Figure 4b).Given the effective aerial oxidation, the absorption at 710 eV should be attributed to Fe(III)‐bearing (hydro‐)oxide(i.e., ferrihydrite and magnetite; Figure 4b) or Fe(III)‐bearing saponite (i.e., ferrian saponite; Chemtobet al., 2015, 2017). Moreover, a broad absorption band at around 714 eV was also found (area 4 in Figures 4a and4d). This spectral feature only appeared in pyrite (FeS2) among the various Fe bearing standards we tested,including iron (hydro‐)oxide, sulfide, and carbonate (Figure 4b).The detailed spatial distribution of the chemical states of Fe was visualized using single‐value decomposition(SVD) mapping with four endmember components. In addition to the extracted XANES spectra of the opticallythick area, the bulk XANES spectra of the synthesized ferrous saponite, aerially oxidized ferrous saponitecontaining Fe(III), and commercially purchased pyrite were used as endmembers (Figure 4c). The three bulkXANES spectra were normalized to have similar optical densities at the pre‐ and post‐edges to create appropriatecolor maps (Figure 4c). Using these endmembers, the residuals of the fitting were suppressed to an optical densitybelow 0.5 (Figure 4e), which was minimized using other available references in Figure 4b. The result of SVDmapping shows that sub‐micrometer‐sized pyrite (blue particles) occurred widely within and beside the ferrousFigure 5. Result of SVD mapping of STXM measurements performed for the mineral sample collected after (a) 37, (b) 227,and (c) 1,350 hr of reaction in the group II experiment, and (d) 1,759 hr of reaction in the group V experiment. Thevisualization of the stacked image is done in the same condition as Figure 4f. The color represents the relative concentrationsof the three endmembers, with ferrous saponite in green, air‐introduced Fe(III)‐bearing saponite in red, and pyrite in blue.Journal of Geophysical Research: Planets 10.1029/2024JE008538NODA ET AL. 10 of 17 21699100, 2025, 1, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JE008538 by National Institute For, Wiley Online Library on [26/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensesaponite through the reaction of the group ii experiments (Figure 4f). The presence of Fe(III)‐bearing materials(reddish features) surrounding pyrite indicates that ferrous iron oxidation to ferric iron may occur in conjunctionwith pyrite formation (Figure 4f).Figures 5a–5c compares the STXM results in the SVD mapping of the group II experiment with different reactiontimes. After a reaction time of 37 hr, sub‐micrometer‐sized pyrite particles were first generated in ferroussaponite, while no Fe(III)‐bearing areas were found. This indicates that oxidation by air during the experiment andanalysis was prevented (Figure 5a). After 227 hr of reaction, Fe(III)‐bearing areas started to appear in the samplein addition to pyrite particles (Figure 5b). Fe(III)‐bearing areas appeared in the vicinity of pyrite after 1,350 hr ofreaction (Figure 5c). Fe(III)‐bearing features and pyrite was less significant in the sample from the experimentwithout H2S (Figure 5d). The minor Fe(III)‐bearing area and pyrite in Figure 5d can be attributed to the un-certainty in SVD fitting. The chemical state of ferrous iron was largely unchanged in the experiments without H2S(Figure 5d), supporting that the occurrence of Fe(III)‐bearing areas in the experiments with H2S were associatedwith the H2 generating reaction.4. Discussion4.1. Mechanism for H2 Generation From H2S‐Containing Fluids With Ferrous SaponiteThe results of our experiments, including both comparative hydrothermal experiments (Section 3.1) and timevariation experiments (Section 3.2) on the hydrothermal interactions between ferrous saponite and volatile‐containing fluids, can be summarized as follows:1. Ferrous saponite was involved in H2 generation when it interacted with H2S containing fluid (Figure 1). LittleH2 detection without H2S in long incubation experiments strongly suggests that H2S played an essential role inH2 production (Figure 2).2. As for the time variation, H2 generation occurred most rapidly under circumneutral pH conditions (Figure 2).H2 production proceeds more slowly under weakly acidic (pH ∼6.5) conditions than under alkaline (pH ∼9)conditions (Figure 2).3. As for the fluid composition, a decrease in Fe concentration with an increase in H2 concentration was observedin neutral‐alkaline pH, suggesting that dissolved Fe is involved H2 generation (Figure 3). Relatively high Feconcentrations (>0.1 mM), likely due to the effective dissolution of ferrous saponite under weakly acidicconditions, did not always result in rapid H2 generation (Figure 3).4. As for the mineral phases, only ferrous saponite among all the tested minerals demonstrated distinct H2generation triggered by the presence of H2S (Figure 1). The lower initial abundances of ferrous saponite tend toresult in lower H2 concentrations (Figure 2). Sub‐micrometer‐sized pyrite is formed with the H2 generation(Figure 4). Oxidation to Fe(III)‐bearing materials also occurred around the pyrite particles (Figure 5).Concerning points (1) and (2) above, one possible interpretation for the pH dependence on the reaction kinetics isthat H2 production may be controlled by the availability of HS− rather than H2S. HS− is thermodynamically morestable at pH > 7 than H2S while H2S is more stable than HS− at pH < 7. The point of zero charge (ZPC) of Na‐smectite has been reported as 7.8–8.2 at an ionic strength of 0.001–0.1 (Arfaoui et al., 2012; Avena & DePauli, 1998; Manohar et al., 2006). Assuming a similar ZPC value for ferrous saponite, the negative charge of itssurface at high pH (>∼8) could inhibit the interaction of HS⁻ with ferrous saponite. Accordingly, H2 generationcould be optimal at circumneutral pH (∼7). This finding is contrasting to H2 generation with troilite because thereaction of HS⁻ with troilite has been excluded by the large HOMO‐LUMO gap (>+6 eV) (Rickard &Luther, 1997).Concerning the point (3), one possible pathway that ferrous ion involved to form H2 is the oxidization reaction asfollows:3Fe2+ + 4H2O → Fe3O4(magnetite) + 6H+ + H2(g) . (4)However, the fact that H2 generation requires the presence of H2S (point 1) suggests that Fe2+ would havecontributed to H2 production in a different pathway from the reaction (Equation 4). In addition, the generation of6H+ suggests that the reaction (Equation 4) can proceed efficiently under high pH conditions (e.g., SeyfriedJournal of Geophysical Research: Planets 10.1029/2024JE008538NODA ET AL. 11 of 17 21699100, 2025, 1, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JE008538 by National Institute For, Wiley Online Library on [26/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licenseet al., 2007), while H2 concentration increased more rapidly in the neutral pH condition in group II experimentsthan in the alkaline condition in group I experiments, despite the similar range of Fe concentration.Concerning point (4), the fact that the initial amount of the mineral controlled the final H2 concentration dem-onstrates that ferrous saponite is involved in the reaction. The low H2 production in the experiments withmagnetite also shows that ferrous saponite would play a role more than a source of dissolved Fe2+. Given theformation of pyrite, the alternative reaction (Equation 4) to form H2 is as follows:Fe2+ + 2HS− → FeS2 + H2(g), (5)in which both Fe2+ and HS− react to form H2 and pyrite. Given that ferrous saponite can be oxidized by thesubstitution of structural O2− by OH− (Sakuma et al., 2022), we propose that the following bulk H2 generatingreaction is consistent with our experimental observation in which the reaction (Equation 5) is involved:2Na0.5 (Mg1.5 Fe2+1.5) (Si3.5Al0.5)O10(OH)2 + Fe2+ + 2HS−→ 2Na0.5 (Mg1.5 Fe2+0.5Fe3+1.0) (Si3.5Al0.5)O11(OH) + FeS2 + 2H2(g).(6)In this reaction, ferrous saponite acts as (i) an electron donor to reduce HS− and (ii) a source of dissolved iron toform pyrite. The stoichiometry of the reaction (Equation 6) indicates that 10 μmol of Fe2+ is required for aheadspace of 60 mL to contain 8 × 103 ppm of H2. This Fe supply corresponds to ∼1 mM of Fe2+ dissolution intoa 10 mL fluid. This level of Fe2+ dissolution approximately corresponds to the measured Mg concentration in thefluid. Given the synthesized ferrous saponite composition of Fe:Mg= 1:1, Fe2+ supply via dissolution of saponiteis consistent with the measured H2 production. Reaction (Equation 6) might be described by excluding the re-action (Equation 5) as follows:2Na0.5 (Mg1.5 Fe2+1.5) (Si3.5Al0.5)O10(OH)2→ 2Na0.5 (Mg1.5 Fe2+0.5Fe3+1.0) (Si3.5Al0.5)O11(OH) + H2(g).(7)However, the absence of H2 generation in the experiments with ferrous saponite and without H2S indicate that thereaction (Equation 7) does not proceed independently from pyrite formation. Although there is another possibilitythat ferrous saponite catalyzes reaction (Equation 5), our STXM results show that the Fe(III)‐bearing feature wasless significant in the solid sample retrieved after the reaction without H2S. This observation supports the hy-pothesis that reaction (Equation 5) may stimulate the reaction (Equation 7) in the experiments with H2S. Wecannot rule out the intermediate formation of FeS mackinawite instead of the direct formation of pyrite in re-actions (Equations 5 and 6), which may have happened on timescales shorter than the time interval of our ex-periments. Further work will be needed to assess if Fe2+ and HS− form such a reaction intermediate in amaturation process to form pyrite.4.2. Potential H2 Yield on MarsOur experimental results suggest that pre‐existing ferrous saponite within the crust of early Mars may haveformed H2 if subsurface fluids that contain H2S have intruded and interacted in the later stage after the saponiteformation. H2S in the fluids is likely supplied as a degassing species (Gaillard et al., 2013; Gaillard & Scail-let, 2009). In this section, we estimate the potential impact of this H2 production process on early Mars.We first assumed that H2 production would be in a proportional relationship to the reactive amount of ferroussaponite based on our experimental results on the comparison of H2 production for different initial amounts offerrous saponite (Section 3.1). Figure 6 shows that H2 production per 1 g of mineral reaches (0.5–2) × 10− 1 mmolin most of the runs with H2S, generally independent of fluid pH after a long reaction time, while it reaches below 3× 10− 3 mmol when magnetite or no H2S is contained.Concerning the amount of ferrous saponite within the Martian crust, previous remote sensing studies haveinterpreted that ferrous saponite has been generated globally within the Martian crust by Noachian, the era inwhich the active hydrothermal circulation at the subsurface is expected (Ehlmann et al., 2011; Michalskiet al., 2013). Here, we defined the Global Equivalent Layer of Clay minerals (GEL‐C) as the hypotheticalJournal of Geophysical Research: Planets 10.1029/2024JE008538NODA ET AL. 12 of 17 21699100, 2025, 1, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JE008538 by National Institute For, Wiley Online Library on [26/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensethickness of clay minerals within the crust averaged throughout the planetarysurface. Based on remote‐sensing observations of canyons and craters onMars, GEL‐C can reach 5–10 km (Scheller et al., 2021). This range isconsistent with the model results for clay mineral formation via interactionsbetween a steam atmosphere and primitive crust (Cannon et al., 2017). Theamount of reactive ferrous saponite in the mass MFeSap, can be described asfollows:MFeSap(x) = ρcr × SMars × GEL‐C × x , (8)where ρcr is the density of the crust (assumed as 2.8 g cm− 3: Knapmeyer‐Endrun et al., 2021) and SMars is the surface area of Mars (1.44 × 108 km2).To deal with the uncertainty in the reaction completeness, the variable x wasgiven as the mass fraction of ferrous saponite that interacted with H2S‐con-taining fluids relative to the total clay minerals represented as GEL‐C. Inaddition to the uncertainty in the mineral composition of crustal clay minerals,they may have been buried 20–30 km below the surface (Cannon et al., 2017),while groundwater could have intruded shallower into the crust up to a depthof ∼10 km and not active well at a deeper layer due to the increasedcompaction (Kite & Melwani Daswani, 2019; Michalski et al., 2013),implying a certain portion of embedded clay would have not involved theinteraction.Given the H2 production per mass of ferrous saponite, RH2 :FeSap, as aparameter, the global H2 yield on Mars by the proposed reaction, YH2, can becalculated as follows:YH2(x) = RH2 :FeSap ×MFeSap(x). (9)Figure 7 shows YH2(x) calculated using GEL‐C of 5 km and RH2 :FeSap of 3× 10− 3 to 2 × 10− 1 mmol g− 1 (Figure 6).This figure compares with global H2 yields from volcanic outgassing estimated by the previous studies (Grottet al., 2011; Scheller et al., 2021). With a nominal RH2:FeSap of 0.1 mmol g− 1 (Figure 6), YH2(x) could reach acomparable level with volcanic outgassing when x is larger than ∼10% (Figure 7). Although the value of x isdifficult to constrain, this requirement (i.e., x > 10%) might be unlikely because of the lack of evidence for global‐scale hydrothermal alterations in the Martian crust. Our estimate suggests that the reaction with ferrous saponitemay not have been an effective H2 source to cause a strong greenhouse effect compared with volcanism. Rather,interactions between ferrous saponite and H2S‐containing fluids may be more significant as a source of locallyconcentrated H2 on early Mars as they could provide chemical energy for chemoautotrophic life, such asmethanogens.There are many uncertainties in our estimate of H2 production, particularly the assumption of RH2:FeSap. Forsimplicity, no dependence on temperature, pressure, or other reaction conditions (e.g., water‐to‐rock ratios) wasconsidered here. A supply of H2S by fluids would be one of the most important factors that determines the H2production efficiency from saponites on early Mars. The outgassing rate of sulfur on early Mars has been esti-mated by extrapolation of the suggested sulfur outgassing rates of ∼109 cm− 2 s− 1 on modern Earth (Halmeret al., 2002) and 1010–1011 cm− 2 s− 1 on early Earth (Claire et al., 2014; Ranjan et al., 2018). An order ofmagnitude higher sulfur content of the Martian mantle than that of Earth, and an approximately 1:1 for a H2S:SO2degassing ratio from Martian magma (oxygen fugacity of IW+1) could have resulted in H2S supply flux of∼1012 cm− 2 s− 1 (Gaillard & Scaillet, 2009, 2014). The provided H2S can be effectively dissolved into alkalinefluids. Given low water‐to‐rock ratios in the subsurface and the presence of basalt, alkaline pH (7.5–10) should beachieved for subsurface (hydrothermal) fluids on Mars (Kikuchi & Shibuya, 2021; Kite & Melwani Das-wani, 2019; Noda et al., 2022).One possibility for increasing the total H2 production is the dependence of RH2 :FeSap on H2S concentration influids. Since large fractions of unreacted ferrous saponite remained in our solid samples after the reactionsFigure 6. A summary of our results of H2 production in the experiments witha reaction time longer than 1,300 hr, normalized with the mass of the mineralagainst the fluid pH. The measured H2 concentrations are converted into themolar amounts using a headspace gas volume of 60 ± 1 mL. Each data pointis from a single experiment with errors due to the uncertainties inquantitation and normalization. Gray broken/dotted lines represent H2production of 0.2, 0.1, 0.05, and 0.003 mmol g− 1.Journal of Geophysical Research: Planets 10.1029/2024JE008538NODA ET AL. 13 of 17 21699100, 2025, 1, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JE008538 by National Institute For, Wiley Online Library on [26/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License(Figure 5), RH2 :FeSap could become higher with a more effective supply of H2S than those in our sealed experi-ments. The temperature dependence of RH2:FeSap would be also crucial in evaluating the global effects of the H2formation process found in this study. RH2:FeSap can vary largely in reality; however, the aim of our estimate is toprovide a first‐order approximation of H2 production from ferrous saponite. A detailed evaluation of these as-sumptions requires further experimentation and is beyond the scope of this study.5. ConclusionsThis study investigated the reaction driven by redox‐sensitive ferrous saponite to examine the role of ferroussaponite as an effective reductant in anoxic aqueous environments on early Mars. Based on a series of anaerobichydrothermal experiments and molecular geochemical analysis, we discovered that ferrous saponite generates H2via interaction with H2S‐containing hydrothermal fluid at a relatively low temperature of 90°C. A comparison ofthe H2 concentration in headspace gas, the concentration of Fe dissolved in the fluid, and the chemical species ofiron in the recovered mineral among various reaction times, initial fluid and gas compositions, and containedamount of ferrous saponite were performed. These series of chemical and mineralogical analyses enabled us topropose that ferrous saponite is involved in the reaction by providing iron to form pyrite in addition to acting as anelectron donor to generate H2.It is difficult to precisely quantify the impact of the H2 production process we proposed on early Mars at this stage.The real reaction conditions on early Mars are not well constrained, while our experiments were conducted at onlyone pressure and temperature. Despite such uncertainties, the new pathway to convert H2S into H2 through fluid‐Figure 7. Estimated global H2 yields occurs through the proposed reaction between ferrous saponite and H2S‐containingfluids (gray broken/dotted lines) and through volcanic outgassing during and after the Noachian period (after 4.1 Ga, coloredsolid lines). The former are calculated for RH2 :FeSap values of 0.2, 0.1, 0.05, and 0.003 mmol g− 1 as a function of thepercentage of reacted ferrous saponite in crustal clay mineral (i.e., x in reaction Equations 8 and 9). The latter was obtainedbased on the parameterized model (Equation 18 and Table 3 in Grott et al., 2011) using H2/H2O ratio in volcanic gas for theoxygen fugacity of IW (iron–wüstite)+ 1 (Gaillard & Scaillet, 2009; Hirschmann & Withers, 2008; Wordsworth et al., 2021) attypical outgassing conditions (1,450 K, 5 bar; Ramirez et al., 2014). The maximum and minimum estimates of initial mantleH2O contents, XH2O, for two cooling models were respectively applied, referring to a previous study (Scheller et al., 2021).Journal of Geophysical Research: Planets 10.1029/2024JE008538NODA ET AL. 14 of 17 21699100, 2025, 1, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JE008538 by National Institute For, Wiley Online Library on [26/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licenserock interaction under anoxic environments has a potential impact on planetary habitability. To evaluate whetherthe H2 production via the proposed process can be important as a greenhouse effect gas, further experiments toinvestigate the H2S and temperature dependence of this reaction would be crucial. Since ferrous saponite andvolatile‐containing fluids would have been available not only on early Mars but also on early Earth (Cata-lano, 2013; Gaillard & Scaillet, 2014), similar H2 production could have occurred on these terrestrial planets.Data Availability StatementThe analytical data generated for Figures 1− 3 and 6 and the source data for Figures 4 and 5 are available fordownload from the open access repository (Noda et al., 2024).ReferencesAndrews‐Hanna, J. C., Phillips, R. J., & Zuber, M. T. (2007). Meridiani Planum and the global hydrology of Mars. Nature, 446(7132), 163–166.https://doi.org/10.1038/nature05594Arfaoui, S., Hamdi, N., Frini‐Srasra, N., & Srasra, E. (2012). Determination of point of zero charge of PILCS with single and mixed oxide pillarsprepared from Tunisian‐Smectite. Geochemistry International, 50(5), 447–454. https://doi.org/10.1134/S0016702912030020Avena, M. J., & De Pauli, C. P. (1998). Proton Adsorption and Electrokinetics of an Argentinean Montmorillonite. Journal of Colloid andInterface Science, 202(1), 195–204. https://doi.org/10.1006/jcis.1998.5402Badaut, D., Besson, G., Decarreau, A., & Rautureau, R. (1985). Occurrence of a ferrous, trioctahedral smectite in recent sediments of Atlantis IIdeep, Red Sea. Clay Minerals, 20(3), 389–404. https://doi.org/10.1180/CLAYMIN.1985.020.3.09Barge, L. M., Flores, E., Baum, M. M., VanderVelde, D. G., & Russell, M. J. (2019). Redox and pH gradients drive amino acid synthesis in ironoxyhydroxide mineral systems. Proceedings of the National Academy of Sciences of the United States of America, 116(11), 4828–4833. https://doi.org/10.1073/pnas.1812098116Batalha, N., Domagal‐Goldman, S. D., Ramirez, R., & Kasting, J. F. (2015). Testing the early Mars H2‐CO2 greenhouse hypothesis with a 1‐Dphotochemical model. Icarus, 258, 337–349. https://doi.org/10.1016/j.icarus.2015.06.016Bristow, T. F., Bish, D. L., Vaniman, D. T., Morris, R. V., Blake, D. F., Grotzinger, J. P., et al. (2015). The origin and implications of clay mineralsfrom Yellowknife Bay, Gale crater, Mars. American Mineralogist, 100(4), 824–836. https://doi.org/10.2138/am‐2015‐5077CCBYNCNDCannon, K. M., Parman, S. W., & Mustard, J. F. (2017). Primordial clays on Mars formed beneath a steam or supercritical atmosphere. Nature,552(7683), 88–91. https://doi.org/10.1038/nature24657Carter, J., Poulet, F., Bibring, J.‐P., Mangold, N., & Murchie, S. (2013). Hydrous minerals on Mars as seen by the CRISM and OMEGA imagingspectrometers: Updated global view. Journal of Geophysical Research: Planets, 118(4), 831–858. https://doi.org/10.1029/2012je004145Catalano, J. G. (2013). Thermodynamic and mass balance constraints on iron‐bearing phyllosilicate formation and alteration pathways on earlyMars. Journal of Geophysical Research: Planets, 118(10), 2124–2136. https://doi.org/10.1002/jgre.20161Chaffin, M. S., Deighan, J., Schneider, N. M., & Stewart, A. I. F. (2017). Elevated atmospheric escape of atomic hydrogen from Mars induced byhigh‐altitude water. Nature Geoscience, 10(3), 174–178. https://doi.org/10.1038/ngeo2887Chassefière, E., Langlais, B., Quesnel, Y., & Leblanc, F. (2013). The fate of early Mars' lost water: The role of serpentinization. Journal ofGeophysical Research: Planets, 118(5), 1123–1134. https://doi.org/10.1002/jgre.20089Chassefière, E., Lasue, J., Langlais, B., & Quesnel, Y. (2016). Early Mars serpentinization‐derived CH4 reservoirs, H2‐induced warming andpaleopressure evolution. Meteoritics & Planetary Sciences, 51(11), 2234–2245. https://doi.org/10.1111/maps.12784Chemtob, S. M., Nickerson, R. D., Morris, R. V., Agresti, D. G., & Catalano, J. G. (2015). Synthesis and structural characterization of ferroustrioctahedral smectites: Implications for clay mineral genesis and detectability on Mars. Journal of Geophysical Research: Planets, 120(6),1119–1140. https://doi.org/10.1002/2014JE004763Chemtob, S. M., Nickerson, R. D., Morris, R. V., Agresti, D. G., & Catalano, J. G. (2017). Oxidative alteration of ferrous smectites and im-plications for the redox evolution of early Mars. Journal of Geophysical Research: Planets, 122(12), 2469–2488. https://doi.org/10.1002/2017JE005331Claire, M. W., Kasting, J. F., Domagal‐Goldman, S. D., Stüeken, E. E., Buick, R., & Meadows, V. S. (2014). Modeling the signature of sulfurmass‐independent fractionation produced in the Archean atmosphere. Geochimica et Cosmochimica Acta, 141, 365–380. https://doi.org/10.1016/j.gca.2014.06.032Cord‐Ruwisch, R., Seitz, H.‐J., & Conrad, R. (1988). The capacity of hydrogenotrophic anaerobic bacteria to compete for traces of hydrogendepends on the redox potential of the terminal electron acceptor. Archives of Microbiology, 149(4), 350–357. https://doi.org/10.1007/bf00411655Ehlmann, B. L., Mustard, J. F., Fassett, C. I., Schon, S. C., Head, J. W., III, Des Marais, D. J., et al. (2008). Clay minerals in delta deposits andorganic preservation potential on Mars. Nature Geoscience, 1(6), 355–358. https://doi.org/10.1038/ngeo207Ehlmann, B. L., Mustard, J. F., Murchie, S. L., Bibring, J.‐P. P., Meunier, A., Fraeman, A. A., & Langevin, Y. (2011). Subsurface water and claymineral formation during the early history of Mars. Nature, 479(7371), 53–60. https://doi.org/10.1038/nature10582Ehlmann, B. L., Mustard, J. F., Swayze, G. A., Clark, R. N., Bishop, J. L., Poulet, F., et al. (2009). Identification of hydrated silicate minerals onMars using MRO‐CRISM: Geologic context near Nili Fossae and implications for aqueous alteration. Journal of Geophysical Research,114(E2). E00D08. https://doi.org/10.1029/2009JE003339Fassett, C. I., & Head, J. W. (2008). The timing of martian valley network activity: Constraints from buffered crater counting. Icarus, 195(1), 61–89. https://doi.org/10.1016/j.icarus.2007.12.009Forget, F., Wordsworth, R., Millour, E., Madeleine, J.‐B., Kerber, L., Leconte, J., et al. (2013). 3D modelling of the early Martian climate under adenser CO2 atmosphere: Temperatures and CO2 ice clouds. Icarus, 222(1), 81–99. https://doi.org/10.1016/j.icarus.2012.10.019Fuchs, G. (2011). Alternative pathways of carbon dioxide fixation: Insights into the early evolution of life? Annual Review of Microbiology, 65(1),631–658. https://doi.org/10.1146/annurev‐micro‐090110‐102801Fukushi, K., Sekine, Y., & Rampe, E. B. (2022). Reconstruction of pH, redox condition, and concentrations of major components in ancient liquidwater from the Karasburg member, Murray formation, Gale Crater, Mars. Geochimica et Cosmochimica Acta, 325, 129–151. https://doi.org/10.1016/j.gca.2022.02.005AcknowledgmentsThis study was financially supported byKAKENHI JSPS (Grant JP19J22396,JP20H00195, JP21H04515, JP22K21344,JP22KJ1296, and JP23H00144). TheSTXM analyses were performed with theapproval of the Photon Factory (No.2018S001).Journal of Geophysical Research: Planets 10.1029/2024JE008538NODA ET AL. 15 of 17 21699100, 2025, 1, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JE008538 by National Institute For, Wiley Online Library on [26/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttps://doi.org/10.1038/nature05594https://doi.org/10.1134/S0016702912030020https://doi.org/10.1006/jcis.1998.5402https://doi.org/10.1180/CLAYMIN.1985.020.3.09https://doi.org/10.1073/pnas.1812098116https://doi.org/10.1073/pnas.1812098116https://doi.org/10.1016/j.icarus.2015.06.016https://doi.org/10.2138/am-2015-5077CCBYNCNDhttps://doi.org/10.1038/nature24657https://doi.org/10.1029/2012je004145https://doi.org/10.1002/jgre.20161https://doi.org/10.1038/ngeo2887https://doi.org/10.1002/jgre.20089https://doi.org/10.1111/maps.12784https://doi.org/10.1002/2014JE004763https://doi.org/10.1002/2017JE005331https://doi.org/10.1002/2017JE005331https://doi.org/10.1016/j.gca.2014.06.032https://doi.org/10.1016/j.gca.2014.06.032https://doi.org/10.1007/bf00411655https://doi.org/10.1007/bf00411655https://doi.org/10.1038/ngeo207https://doi.org/10.1038/nature10582https://doi.org/10.1029/2009JE003339https://doi.org/10.1016/j.icarus.2007.12.009https://doi.org/10.1016/j.icarus.2012.10.019https://doi.org/10.1146/annurev-micro-090110-102801https://doi.org/10.1016/j.gca.2022.02.005https://doi.org/10.1016/j.gca.2022.02.005Fukushi, K., Sekine, Y., Sakuma, H., Morida, K., & Wordsworth, R. (2019). Semiarid climate and hyposaline lake on early Mars inferred fromreconstructed water chemistry at Gale. Nature Communications, 10(1), 1–11. https://doi.org/10.1038/s41467‐019‐12871‐6Gaillard, F., Michalski, J., Berger, G., McLennan, S. M., & Scaillet, B. (2013). Geochemical reservoirs and timing of sulfur cycling on Mars.Space Science Reviews, 174(1–4), 251–300. https://doi.org/10.1007/s11214‐012‐9947‐4Gaillard, F., & Scaillet, B. (2009). The sulfur content of volcanic gases on Mars. Earth and Planetary Science Letters, 279(1), 34–43. https://doi.org/10.1016/j.epsl.2008.12.028Gaillard, F., & Scaillet, B. (2014). A theoretical framework for volcanic degassing chemistry in a comparative planetology perspective andimplications for planetary atmospheres. Earth and Planetary Science Letters, 403, 307–316. https://doi.org/10.1016/j.epsl.2014.07.009Grott, M., Morschhauser, A., Breuer, D., & Hauber, E. (2011). Volcanic outgassing of CO2 and H2O on Mars. Earth and Planetary ScienceLetters, 308(3–4), 391–400. https://doi.org/10.1016/j.epsl.2011.06.014Haberle, R. M. (1998). Early mars climate models. Journal of Geophysical Research, 103(E12), 28467–28479. https://doi.org/10.1029/98je01396Haberle, R. M., Zahnle, K., Barlow, N. G., & Steakley, K. E. (2019). Impact degassing of H2 on early Mars and its effect on the climate system.Geophysical Research Letters, 46(22), 13355–13362. https://doi.org/10.1029/2019GL084733Halmer, M. M., Schmincke, H.‐U., & Graf, H.‐F. (2002). The annual volcanic gas input into the atmosphere, in particular into the stratosphere: Aglobal data set for the past 100 years. Journal of Volcanology and Geothermal Research, 115(3), 511–528. https://doi.org/10.1016/S0377‐0273(01)00318‐3Hecht, M. H., Kounaves, S. P., Quinn, R. C., West, S. J., Young, S. M. M., Ming, D. W., et al. (2009). Detection of perchlorate and the solublechemistry of Martian soil at the Phoenix lander site. Science, 325(5936), 64–67. https://doi.org/10.1126/science.1172466Heinen, W., & Lauwers, A. M. (1996). Organic sulfur compounds resulting from the interaction of iron sulfide, hydrogen sulfide and carbondioxide in an anaerobic aqueous environment. Origins of Life and Evolution of the Biosphere: The Journal of the International Society for theStudy of the Origin of Life, 26(2), 131–150. https://doi.org/10.1007/BF01809852Hirschmann, M. M., & Withers, A. C. (2008). Ventilation of CO2 from a reduced mantle and consequences for the early Martian greenhouse.Earth and Planetary Science Letters, 270(1), 147–155. https://doi.org/10.1016/j.epsl.2008.03.034Hitchcock, A. P. (2009). aXis 2000 – Analysis of X‐ray images and spectra. Retrieved from http://unicorn.mcmaster.ca/aXis2000.htmlHoehler, T. M., Alperin, M. J., Albert, D. B., & Martens, C. S. (2001). Apparent minimum free energy requirements for methanogenic Archaeaand sulfate‐reducing bacteria in an anoxic marine sediment. FEMS Microbiology Ecology, 38(1), 33–41. https://doi.org/10.1111/j.1574‐6941.2001.tb00879.xHoward, A. D., Moore, J. M., & Irwin, R. P. (2005). An intense terminal epoch of widespread fluvial activity on early Mars: 1. Valley networkincision and associated deposits. Journal of Geophysical Research, 110(E12). E12S14. https://doi.org/10.1029/2005JE002459Inskeep, W. P., Ackerman, G. G., Taylor, W. P., Kozubal, M., Korf, S., & Macur, R. E. (2005). On the energetics of chemolithotrophy innonequilibrium systems: Case studies of geothermal springs in Yellowstone National Park. Geobiology, 3(4), 297–317. https://doi.org/10.1111/j.1472‐4669.2006.00059.xJohnson, S. S., Mischna, M. A., Grove, T. L., & Zuber, M. T. (2008). Sulfur‐induced greenhouse warming on early Mars. Journal of GeophysicalResearch, 113(E8), 8005. https://doi.org/10.1029/2007je002962Kasting, J. F. (1991). CO2 condensation and the climate of early Mars. Icarus, 94(1), 1–13. https://doi.org/10.1016/0019‐1035(91)90137‐iKikuchi, S., & Shibuya, T. (2021). Thermodynamic constraints on Smectite and Iron oxide formation at Gale Crater, Mars: Insights into potentialfree energy from Aerobic Fe oxidation in lake water–groundwater mixing zone. Minerals, 11(4), 341. https://doi.org/10.3390/min11040341Kitadai, N., Nakamura, R., Yamamoto, M., Okada, S., Takahagi, W., Nakano, Y., et al. (2021). Thioester synthesis through geoelectrochemicalCO2 fixation on Ni sulfides. Communications Chemistry, 4(1), 1–9. https://doi.org/10.1038/s42004‐021‐00475‐5Kite, E. S., & Melwani Daswani, M. (2019). Geochemistry constrains global hydrology on Early Mars. Earth and Planetary Science Letters, 524,115718. https://doi.org/10.1016/j.epsl.2019.115718Knapmeyer‐Endrun, B., Panning, M. P., Bissig, F., Joshi, R., Khan, A., Kim, D., et al. (2021). Thickness and structure of the Martian crust fromInSight seismic data. Science, 373(6553), 438–443. https://doi.org/10.1126/science.abf8966Kohyama, N., Shimoda, S., & Sudo, T. (1973). Iron‐rich saponite (ferrous and ferric forms). Clays and Clay Minerals, 21(4), 229–237. https://doi.org/10.1346/ccmn.1973.0210405Koyama, S., Terada, N., Nakagawa, H., Kuroda, T., & Sekine, Y. (2021). Stability of atmospheric redox states of early mars inferred from timeresponse of the regulation of H and O Losses. The Astrophysical Journal, 912(2), 135. https://doi.org/10.3847/1538‐4357/ABF0ACLanza, N. L., Wiens, R. C., Arvidson, R. E., Clark, B. C., Fischer, W. W., Gellert, R., et al. (2016). Oxidation of manganese in an ancient aquifer,Kimberley formation, Gale crater, Mars. Geophysical Research Letters, 43(14), 7398–7407. https://doi.org/10.1002/2016GL069109Manohar, D. M., Noeline, B. F., & Anirudhan, T. S. (2006). Adsorption performance of Al‐pillared bentonite clay for the removal of cobalt(II)from aqueous phase. Applied Clay Science, 31(3), 194–206. https://doi.org/10.1016/j.clay.2005.08.008Mayhew, L. E., Ellison, E. T., McCollom, T. M., Trainor, T. P., & Templeton, A. S. (2013). Hydrogen generation from low‐temperature water‐rock reactions. Nature Geoscience, 6(6), 478–484. https://doi.org/10.1038/ngeo1825McCollom, T. M., & Shock, E. L. (1997). Geochemical constraints on chemolithoautotrophic metabolism by microorganisms in seafloor hy-drothermal systems. Geochimica et Cosmochimica Acta, 61(20), 4375–4391. https://doi.org/10.1016/S0016‐7037(97)00241‐XMichalski, J. R., Cuadros, J., Bishop, J. L., Darby Dyar, M., Dekov, V., & Fiore, S. (2015). Constraints on the crystal‐chemistry of Fe/Mg‐richsmectitic clays on Mars and links to global alteration trends. Earth and Planetary Science Letters, 427, 215–225. https://doi.org/10.1016/j.epsl.2015.06.020Michalski, J. R., Cuadros, J., Niles, P. B., Parnell, J., Deanne Rogers, A., & Wright, S. P. (2013). Groundwater activity on Mars and implicationsfor a deep biosphere. Nature Geoscience, 6(2), 133–138. https://doi.org/10.1038/ngeo1706Mustard, J. F., Murchie, S. L., Pelkey, S. M., Ehlmann, B. L., Milliken, R. E., Grant, J. A., et al. (2008). Hydrated silicate minerals on Marsobserved by the Mars reconnaissance orbiter CRISM instrument. Nature, 454(7202), 305–309. https://doi.org/10.1038/nature07097Noda, N., Imamura, S., Sekine, Y., Kurisu, M., Fukushi, K., Terada, N., et al. (2019). Highly oxidizing aqueous environments on early Marsinferred from scavenging pattern of trace metals on manganese oxides. Journal of Geophysical Research: Planets, 124(5), 1282–1295. https://doi.org/10.1029/2018je005892Noda, N., Sekine, Y., Takahashi, Y., Fukushi, K., Sakuma, H., Kawai, T., et al. (2024). Hydrogen generation from ferrous saponite in reaction withH2S‐containing fluid: Relevance to early Martian habitability [Dataset]. https://doi.org/10.6084/m9.figshare.26011219Noda, N., Sekine, Y., Tan, S., Kikuchi, S., Shibuya, T., Kurisu, M., et al. (2022). Characterization of groundwater chemistry beneath Gale Crateron early Mars by hydrothermal experiments. Icarus, 386, 115149. https://doi.org/10.1016/j.icarus.2022.115149Noda, N., Yamashita, S., Takahashi, Y., Matsumoto, M., Enokido, Y., Amano, K., et al. (2021). Anaerobic microscopic analysis of ferroussaponite and its sensitivity to oxidation by Earth's air: Lessons learned for analysis of returned samples from mars and carbonaceous asteroids.Minerals, 11(11), 1244. https://doi.org/10.3390/min11111244Journal of Geophysical Research: Planets 10.1029/2024JE008538NODA ET AL. 16 of 17 21699100, 2025, 1, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JE008538 by National Institute For, Wiley Online Library on [26/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttps://doi.org/10.1038/s41467-019-12871-6https://doi.org/10.1007/s11214-012-9947-4https://doi.org/10.1016/j.epsl.2008.12.028https://doi.org/10.1016/j.epsl.2008.12.028https://doi.org/10.1016/j.epsl.2014.07.009https://doi.org/10.1016/j.epsl.2011.06.014https://doi.org/10.1029/98je01396https://doi.org/10.1029/2019GL084733https://doi.org/10.1016/S0377-0273(01)00318-3https://doi.org/10.1016/S0377-0273(01)00318-3https://doi.org/10.1126/science.1172466https://doi.org/10.1007/BF01809852https://doi.org/10.1016/j.epsl.2008.03.034http://unicorn.mcmaster.ca/aXis2000.htmlhttps://doi.org/10.1111/j.1574-6941.2001.tb00879.xhttps://doi.org/10.1111/j.1574-6941.2001.tb00879.xhttps://doi.org/10.1029/2005JE002459https://doi.org/10.1111/j.1472-4669.2006.00059.xhttps://doi.org/10.1111/j.1472-4669.2006.00059.xhttps://doi.org/10.1029/2007je002962https://doi.org/10.1016/0019-1035(91)90137-ihttps://doi.org/10.3390/min11040341https://doi.org/10.1038/s42004-021-00475-5https://doi.org/10.1016/j.epsl.2019.115718https://doi.org/10.1126/science.abf8966https://doi.org/10.1346/ccmn.1973.0210405https://doi.org/10.1346/ccmn.1973.0210405https://doi.org/10.3847/1538-4357/ABF0AChttps://doi.org/10.1002/2016GL069109https://doi.org/10.1016/j.clay.2005.08.008https://doi.org/10.1038/ngeo1825https://doi.org/10.1016/S0016-7037(97)00241-Xhttps://doi.org/10.1016/j.epsl.2015.06.020https://doi.org/10.1016/j.epsl.2015.06.020https://doi.org/10.1038/ngeo1706https://doi.org/10.1038/nature07097https://doi.org/10.1029/2018je005892https://doi.org/10.1029/2018je005892https://doi.org/10.6084/m9.figshare.26011219https://doi.org/10.1016/j.icarus.2022.115149https://doi.org/10.3390/min11111244Ojha, L., Buffo, J., Karunatillake, S., & Siegler, M. (2020). Groundwater production from geothermal heating on early Mars and implication forearly Martian habitability. Science Advances, 6(49). eabb1669. https://doi.org/10.1126/sciadv.abb1669O’Loughlin, E. J., Boyanov, M. I., Kemner, K. M., & Thalhammer, K. O. (2020). Reduction of Hg(II) by Fe(II)‐Bearing Smectite Clay minerals.Minerals, 10(12), 1079. https://doi.org/10.3390/min10121079Poulet, F., Bibring, J.‐P., Mustard, J. F., Gendrin, A., Mangold, N., Langevin, Y., et al. (2005). Phyllosilicates on Mars and implications for earlyMartian climate. Nature, 438(7068), 623–627. https://doi.org/10.1038/nature04274Quinn, R. C., Martucci, H. F. H., Miller, S. R., Bryson, C. E., Grunthaner, F. J., & Grunthaner, P. J. (2013). Perchlorate radiolysis on Mars and theorigin of Martian soil reactivity. Astrobiology, 13(6), 515–520. https://doi.org/10.1089/ast.2013.0999Ramirez, R. M., Kopparapu, R., Zugger, M. E., Robinson, T. D., Freedman, R., & Kasting, J. F. (2014). Warming early Mars with CO2 and H2.Nature Geoscience, 7(1), 59–63. https://doi.org/10.1038/ngeo2000Rampe, E. B., Bristow, T. F., Morris, R. V., Morrison, S. M., Achilles, C. N., Ming, D. W., et al. (2020). Mineralogy of Vera Rubin ridge from theMars science laboratory CheMin instrument. Journal of Geophysical Research: Planets, 125(9). e2019JE006306. https://doi.org/10.1029/2019JE006306Ranjan, S., Todd, Z. R., Sutherland, J. D., & Sasselov, D. D. (2018). Sulfidic anion concentrations on early earth for surficial origins‐of‐lifechemistry. Astrobiology, 18(8), 1023–1040. https://doi.org/10.1089/ast.2017.1770Ravel, B., & Newville, M. (2005). ATHENA, artemis, hephaestus: Data analysis for X‐ray absorption spectroscopy using IFEFFIT. Journal ofSynchrotron Radiation, 12(4), 537–541. https://doi.org/10.1107/S0909049505012719Rickard, D. (1997). Kinetics of pyrite formation by the H2S oxidation of iron (II) monosulfide in aqueous solutions between 25 and 125°C: Therate equation. Geochimica et Cosmochimica Acta, 61(1), 115–134. https://doi.org/10.1016/s0016‐7037(96)00321‐3Rickard, D., & Luther, G. W. (1997). Kinetics of pyrite formation by the H2S oxidation of iron (II) monosulfide in aqueous solutions between 25and 125°C: The mechanism. Geochimica et Cosmochimica Acta, 61(1), 135–147. https://doi.org/10.1016/S0016‐7037(96)00322‐5Righter, K., Yang, H., Costin, G., & Downs, R. T. (2008). Oxygen fugacity in the Martian mantle controlled by carbon: New constraints from thenakhlite MIL 03346. Meteoritics & Planetary Sciences, 43(10), 1709–1723. https://doi.org/10.1111/j.1945‐5100.2008.tb00638.xRivera Banuchi, V. B., Liu, W., Yee, N., Legett, C., Glotch, T. D., & Chemtob, S. M. (2022). Ultraviolet photooxidation of smectite‐bound Fe(II)and implications for the origin of Martian nontronites. Journal of Geophysical Research: Planets, 127(5). https://doi.org/10.1029/2021je007150Sakuma, H., Morida, K., Takahashi, Y., Fukushi, K., Noda, N., Sekine, Y., & Tamura, K. (2022). Synthesis of ferrian and ferro‐saponites:Implications for the structure of (Fe,Mg)‐smectites formed under reduced conditions. American Mineralogist, 107(10), 1926–1935. https://doi.org/10.2138/am‐2022‐8231Salese, F., Pondrelli, M., Neeseman, A., Schmidt, G., & Ori, G. G. (2019). Geological evidence of planet‐wide groundwater system on Mars.Journal of Geophysical Research: Planets, 124(2), 374–395. https://doi.org/10.1029/2018JE005802Scheller, E. L., Ehlmann, B. L., Hu, R., Adams, D. J., & Yung, Y. L. (2021). Long‐term drying of Mars by sequestration of ocean‐scale volumes ofwater in the crust. Science, 372(6537), 56–62. https://doi.org/10.1126/SCIENCE.ABC7717Seyfried, W. E., Foustoukos, D. I., & Fu, Q. (2007). Redox evolution and mass transfer during serpentinization: An experimental and theoreticalstudy at 200°C, 500bar with implications for ultramafic‐hosted hydrothermal systems at Mid‐Ocean Ridges. Geochimica et CosmochimicaActa, 71(15), 3872–3886. https://doi.org/10.1016/j.gca.2007.05.015Tarnas, J. D., Mustard, J. F., Sherwood Lollar, B., Bramble, M. S., Cannon, K. M., Palumbo, A. M., & Plesa, A.‐C. (2018). Radiolytic H2production on Noachian Mars: Implications for habitability and atmospheric warming. Earth and Planetary Science Letters, 502, 133–145.https://doi.org/10.1016/j.epsl.2018.09.001Tosca, N. J., Ahmed, I. A. M. M., Tutolo, B. M., Ashpitel, A., & Hurowitz, J. A. (2018). Magnetite authigenesis and the warming of early Mars.Nature Geoscience, 11(9), 635–639. https://doi.org/10.1038/s41561‐018‐0203‐8Treiman, A. H., Morris, R. V., Agresti, D. G., Graff, T. G., Achilles, C. N., Rampe, E. B., et al. (2014). Ferrian saponite from the Santa MonicaMountains (California, U.S.A., Earth): Characterization as an analog for clay minerals on Mars with application to Yellowknife Bay in GaleCrater. American Mineralogist, 99(11–12), 2234–2250. https://doi.org/10.2138/am‐2014‐4763Turbet, M., Boulet, C., & Karman, T. (2020). Measurements and semi‐empirical calculations of CO2 + CH4 and CO2 + H2 collision‐inducedabsorption across a wide range of wavelengths and temperatures. Application for the prediction of early Mars surface temperature. Icarus, 346,113762. https://doi.org/10.1016/j.icarus.2020.113762Turbet, M., Forget, F., Head, J. W., & Wordsworth, R. (2017). 3D modelling of the climatic impact of outflow channel formation events on earlyMars. Icarus, 288, 10–36. https://doi.org/10.1016/J.ICARUS.2017.01.024Vaniman, D. T., Bish, D. L., Ming, D. W., Bristow, T. F., Morris, R. V., Blake, D. F., et al. (2014). Mineralogy of a mudstone at Yellowknife Bay,Gale crater, Mars. Science, 343(6169). 1243480. https://doi.org/10.1126/science.1243480Wächtershäuser, G. (1992). Groundworks for an evolutionary biochemistry: The iron‐sulphur world. Progress in Biophysics and MolecularBiology, 58(2), 85–201. https://doi.org/10.1016/0079‐6107(92)90022‐xWan, M., Schröder, C., & Peiffer, S. (2017). Fe(III):S(‐II) concentration ratio controls the pathway and the kinetics of pyrite formation duringsulfidation of ferric hydroxides. Geochimica et Cosmochimica Acta, 217, 334–348. https://doi.org/10.1016/j.gca.2017.08.036Wordsworth, R., Forget, F., Millour, E., Head, J. W., Madeleine, J.‐B., & Charnay, B. (2013). Global modelling of the early Martian climate undera denser CO2 atmosphere: Water cycle and ice evolution. Icarus, 222(1), 1–19. https://doi.org/10.1016/J.ICARUS.2012.09.036Wordsworth, R., Kalugina, Y., Lokshtanov, S., Vigasin, A., Ehlmann, B., Head, J., et al. (2017). Transient reducing greenhouse warming on earlyMars. Geophysical Research Letters, 44(2), 665–671. https://doi.org/10.1002/2016GL071766Wordsworth, R., Knoll, A. H., Hurowitz, J., Baum, M., Ehlmann, B. L., Head, J. W., & Steakley, K. (2021). A coupled model of episodic warming,oxidation and geochemical transitions on early Mars. Nature Geoscience, 14(3), 127–132. https://doi.org/10.1038/s41561‐021‐00701‐8Zinder, S. H. (1993). Physiological ecology of methanogens. In Methanogenesis (pp. 128–206). Springer US. https://doi.org/10.1007/978‐1‐4615‐2391‐8_4Zolotov, M. Y., & Mironenko, M. V. (2007). Timing of acid weathering on Mars: A kinetic‐thermodynamic assessment. Journal of GeophysicalResearch, 112(E7). https://doi.org/10.1029/2006je002882Journal of Geophysical Research: Planets 10.1029/2024JE008538NODA ET AL. 17 of 17 21699100, 2025, 1, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JE008538 by National Institute For, Wiley Online Library on [26/12/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttps://doi.org/10.1126/sciadv.abb1669https://doi.org/10.3390/min10121079https://doi.org/10.1038/nature04274https://doi.org/10.1089/ast.2013.0999https://doi.org/10.1038/ngeo2000https://doi.org/10.1029/2019JE006306https://doi.org/10.1029/2019JE006306https://doi.org/10.1089/ast.2017.1770https://doi.org/10.1107/S0909049505012719https://doi.org/10.1016/s0016-7037(96)00321-3https://doi.org/10.1016/S0016-7037(96)00322-5https://doi.org/10.1111/j.1945-5100.2008.tb00638.xhttps://doi.org/10.1029/2021je007150https://doi.org/10.1029/2021je007150https://doi.org/10.2138/am-2022-8231https://doi.org/10.2138/am-2022-8231https://doi.org/10.1029/2018JE005802https://doi.org/10.1126/SCIENCE.ABC7717https://doi.org/10.1016/j.gca.2007.05.015https://doi.org/10.1016/j.epsl.2018.09.001https://doi.org/10.1038/s41561-018-0203-8https://doi.org/10.2138/am-2014-4763https://doi.org/10.1016/j.icarus.2020.113762https://doi.org/10.1016/J.ICARUS.2017.01.024https://doi.org/10.1126/science.1243480https://doi.org/10.1016/0079-6107(92)90022-xhttps://doi.org/10.1016/j.gca.2017.08.036https://doi.org/10.1016/J.ICARUS.2012.09.036https://doi.org/10.1002/2016GL071766https://doi.org/10.1038/s41561-021-00701-8https://doi.org/10.1007/978-1-4615-2391-8_4https://doi.org/10.1007/978-1-4615-2391-8_4https://doi.org/10.1029/2006je002882 description Hydrogen Generation From Ferrous Saponite in Reaction With H2S‐Containing Fluid: Relevance to Early Martian Habitability 1. Introduction 2. Methods 2.1. Synthetic Ferrous Saponite 2.2. Anoxic Gas–Fluid–Mineral‐Phase Reaction 2.3. Experimental Conditions 2.4. Gas and Liquid Analyses 2.5. Chemical Analysis of Mineral Phases 3. Results 3.1. Comparison of H2 Formation for Different Minerals 3.2. Time Evolution of H2 Concentration in Reaction With Ferrous Saponite 3.2.1. Characterization of Reactions in the Solid Phase 4. Discussion 4.1. Mechanism for H2 Generation From H2S‐Containing Fluids With Ferrous Saponite 4.2. Potential H2 Yield on Mars 5. Conclusions Data Availability Statement