Natural hydrogen is widely distributed, and its generation from rock and water is poorly known on the molecular scale. Here, we provide direct experimental evidence of hydrogen production from rock and offer an atomistic insight into the natural hydrogen origin based on first-principles molecular dynamics simulation using quartz as the model silicate mineral. The hypothetical reaction pathway of mechanoradical hydrogen generation is clarified, and the reductive environment and deprotonation process are shown to be essential for hydrogen generation. Hydrogen was generated from Si radicals and H2O in two steps involving charge transfer and inverse charge transfer. Oxygen generation was also observed in experiments and simulations, it was generated on an oxygen-rich surface with SiOO as an intermediate product. Simulations using water and ice VII (a high-pressure form of water) at different temperatures link our results to potential natural hydrogen production conditions.
Article
Identifying General Reaction Conditions for Mechanoradical Natural Hydrogen Production
https://doi.org/10.21203/rs.3.rs-4001833/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
Natural hydrogen is increasingly discovered under various geological conditions1, and the investigation of its genesis is important for the Earth's biosphere2-5, lithosphere4,6,7, energy cycles8,9, earthquake prediction10, and outer space exploration11. It can be a promising energy source portrayed as gold hydrogen in addition to grey, blue, and green hydrogens1,8,9. Hydrogen can also be generated by injecting water into reactive formations, known as orange hydrogen9,12. Minerals occurring in the crust and shallow mantle have the potential to be one of the sources of natural hydrogen by the reaction of their fresh surface with water during fracturing3-6,13–15. This phenomenon was first identified in the active Yamazaki fault in Japan10, and hydrogen is one of the products of complex physicochemical reactions that occur in fault-lubricated areas during earthquakes6,7. In subsequent experimental studies, it was confirmed that mechanoradical hydrogen generation is widespread in comminution silicate rocks, not only in the tectonic fault region1,3,6,16. H2 generation is not only influenced by the type and surface area of the rocks but also correlated with the localized mineral surfaces3,16, which indicates that the direction and fracture location may affect the amount of hydrogen production. Serpentinization16,17, radiolysis of water18, and degassing of magmas19 are also considered possible sources of natural hydrogen. The diverse sources and high diffusivity of abiotic hydrogen make it more complicated to identify the origin1,9,20.
Most natural hydrogen genesis is associated with water, and mantle conditions show favourable potential for hydrogen formation20,21. Ice VII is one of the states that may exist in this high-pressure mantle environment22,23, and under X-ray irradiation, ice VII can easily decompose into hydrogen and oxygen24,25. Quartz is often regarded as a model material for silicate minerals. Three types of surface species were found in quartz-aqueous solution (SiOH2+, SiOH0, SiO-)26. This indicates that hydrogen can be generated by deprotonation if mechanical radicals occur during comminution. When quartz is crushed, various cleavage planes are exposed27,28. It is generally assumed that silanol groups will be the end of the products. The cleavage in quartz determines the characteristics of the fresh surface, and the radicals on the surface vary by the fracture location under the same Miller index29.
In molecular-scale studies, hydrogen gas generation from rock comminution can be considered a reaction of free radicals on the Si-terminated surface with water. Most of the hydrogen radicals generated after the reaction are rapidly hydroxylated, and the Grotthuss mechanism also accelerates this process30,31. In previous molecular simulation studies, the oxygen exposed on the cleavage plane made the system not reductive, which could be the reason for the absence of hydrogen molecules30–32. Extensive research has been devoted to characterizing the surface chemistry of quartz, the reactivity of H2O or other geofluids, and the surface properties of quartz after hydroxylation33–36. However, to our knowledge, there has not yet been a successful molecular-scale simulation in generating hydrogen from quartz and water. The reaction pathway of rock-based hydrogen generation remains hypothetical.
Experiments of hydrogen and oxygen generation from rocks
We measured the amount of hydrogen gas produced by different types of dry and wet rocks during high-speed friction (Fig. 1). The gas atmosphere was categorized into air and argon environments, and the production of hydrogen was measured under both conditions. Hydrogen production has been detected in a wide variety of rocks. Some rocks have been heated and dried in the oven which are called dry rocks. Wet rocks produced relatively more hydrogen than dry rocks, which shows water has a profound influence on the process of hydrogen generation. Dry rocks also produced hydrogen, because not all water was removed, such as moisture and fluid inclusions in the rock. The amount of hydrogen produced by high-speed friction is usually considered to be positively correlated with friction work6. The production of oxygen was detected in argon environments. In the air atmosphere, there were a lot of sparks from the high-speed friction of the rocks, and the amount of oxygen after the experiment was less than the initial amount. CO2 and CH4 were also produced during the experiment. Hydrogen generation has been demonstrated in both using a ball mill to crush rocks3,5,14–16 and high-velocity friction experiments6. However, the molecular-scale simulation successfully produces hydrogen to explain this process has not been shown yet.
Simulations of reaction between silicon-terminated quartz and water: hydrogen generation
We began the simulation with a newly cleaved quartz surface employed by Ledyastuti32, with a notable difference: we incorporated spin polarization due to the presence of the active radical surface. Despite this modification, hydrogen generation was not observed. To explore the influence of metallic elements, we extended our investigation by introducing Fe(OH)2 into the water system and substituting Si atoms with Fe atoms on various surface sites (fig. S1). However, these modifications did not result in hydrogen generation, as the H radicals generated in the process rapidly reacted with SiO. Building upon the known reductive properties of Fe(II) in serpentinization17 and considering the diverse surfaces formed during quartz crushing29,30, our subsequent approach aimed to construct a reductive closed system (fig. S2). According to the surface energy ordering29, the Q3 (100) surface of quartz30 is the most likely to produce a ≡ Si•-only surface. By employing this surface, we found a reaction path to generate molecular hydrogen, as shown in Fig. 2a. Due to the randomness of rock fracture, the comminution process produces mineral pieces without oxygen radicals, which behave reductively. This reductive closed system is important for the generation of hydrogen, especially in such a small molecular system. Otherwise, the generated hydrogen atom will immediately combine with oxygen, lowering the system's total energy.
Charge transfer is also necessary for the generation of hydrogen in this case. In the first step, positive charges were added to the system according to the number of missing oxygen atoms (as shown in the Methods section). After the addition of the positive charge, H+ was also observed, and the pH of the system was low. The previous study claims that H+ is also an accompanying product during the experiment15,16. For a strongly reducing environment, the fracture surface only retains ≡ Si• radicals and no ≡ SiO• radicals. We assume that there will be a probability that this reducing environment will occur, and this process was compensated by adding a positive charge. Without the initial charge addition, the simulation indicates the presence of SiOH, SiH, OH−, and H+ as the final components. However, with the inclusion of the charge, the freshly fractured surface's ≡ Si• radicals form a ≡ SiOH2+ structure in conjunction with water. This bonding was described as partially covalent and partially ionic, which was supported by the experiments38. These structures are crucial for the next step of hydrogen generation. In the second step, the coordinates of the last frame in the first step were used as the starting configuration. The positive charges that were previously added were removed. Notably, a rapid generation of hydrogen gas molecules was observed during this phase. These two steps to generate hydrogen can be seen as deprotonation processes, such as the 2-pK process described for ionization at oxide-water interfaces27. The surface eventually hydroxylates to SiOH with the lowest energy, and the extra hydrogen atoms have a probability of meeting each other to form hydrogen molecules.
The total energy shows that the system with the generated hydrogen is lower regardless of the temperature and pressure conditions (see Figs. 2b,c, and S3), which indicates that the hydrogen molecule as a reaction product is reasonable, and the reaction path discovered here should be noted. We found that sufficient SiOH2 needs to be formed in the first step of the simulation, which requires lower temperature conditions, especially in a low-pressure relatively sparse system. When comparing high-pressure and low-pressure conditions, if the density of water is sufficient to form SiOH2 in a saturated state (i.e., ice VII system), the temperature will have a lower impact on the formation of SiOH2. However, under low-pressure conditions where water is not saturated, the temperature has a strong effect on the formation of SiOH2 and consequently affects the generation of hydrogen. For instance, in a 1,500 K water system, with a limited number of water molecules (e.g., 32 ≡ Si• radicals and 32 water molecules), the structure of SiOH2 decreases during the initial step. Furthermore, after the second step of deprotonation, Si–H structures are formed, leading to a reduction in the production of hydrogen molecules.
Influence of temperature and pressure on hydrogen generation
For both water and ice VII systems, the temperature was set at 500, 1,000, and 1,500 K, respectively. The pressure range in the simulation is from 0.01 GPa to 5.39 GPa (Table 1, fig. S4). During the first step of the simulation, the predominant species generated were ≡ SiOH2+ and ≡ SiOH, with some H3O+ formation, but no hydrogen molecules appeared, as shown in Fig. 3a. In the second step of the simulation, upon removal of the positive charge, nearly all SiOH2 species transformed into SiOH, leading to the production of hydrogen molecules, as shown in Fig. 3b. Therefore, SiOH2 can be regarded as an intermediate product that connects hydrogen generation and quartz surface hydroxylation. We have thus analyzed the products SiOH2, SiOH, and hydrogen molecules for the low-pressure water system (Fig. 3c,d,e) and high-pressure ice VII system (Fig. 3f,g,h), respectively. Interestingly, for the low-pressure water system, the amount of hydrogen produced at three different temperatures in order from most to least is 1,000 K, 500 K, and 1,500 K (Fig. 3e, Table 1). This can be explained by SiOH2 in Fig. 3c and SiOH in Fig. 3d. When SiOH2 is produced in the first step, the quantities produced are 1,000 K, 500 K, and 1,500 K in descending order (Table 1). The amount of SiOH2 directly affects the amount of H radicals that can be generated subsequently. SiOH2 is not a stable structure, and its lifetime is further reduced at a temperature of 1,500 K. When the positive charge is removed in the second step, the amount of SiOH2 decreases rapidly, while the amount of SiOH starts to increase. Ultimately, the degree of hydroxylation is close for each system at different temperatures (Fig. 3d,g). Under this simulated condition, the formation amount of SiOH2 is correlated with the density and pressure of water. The system of ice VII generates more hydrogen molecules than those with low-pressure water, as expected. In the ice VII system, there is no significant difference in the amount of hydrogen generation at different temperatures (Table 1). Ice VII exhibits higher density compared to the low-pressure water phase, limiting the available space for water molecules to escape. This is indirectly evidenced by the consistent amount of SiOH2 in the three systems. At the conclusion of the initial simulation step (with positive charges) spanning 3 ps, the quantity of SiOH2 remains nearly identical across the three temperatures. The hydroxylation state of ice VII is observed to be consistent for all three systems. Consequently, the quantity of hydrogen gas produced appears to be less influenced by temperature. This consistency is also reflected in similar simulations conducted on different quartz surfaces (Q2), as depicted in figs. S5 and S6. In the state of water at ordinary pressure, the amount of hydrogen generated may be highest at a suitable intermediate temperature based on the lifetime of SiOH2 and the degree of hydroxylation. As the pressure increases, the impact of temperature differences on hydrogen production diminishes.
In Japan's Nankai area, substantial quantities of biogenic CH4 are stored in the form of methane hydrate, with geogenic hydrogen being considered one of their potential sources4,39. The hydrogen concentration in the deep sample was reported to be three orders of magnitude higher than that in the shallow sample4, suggesting that the deeper environment may favour hydrogen generation. Considering the geological context of Nankai subduction zone where seismic faulting and fluid circulation constantly occur, the elevated concentration of H2 may stem from fault friction and/or water-rock interactions. Indeed, seismic imaging results in the area showed a large fault branched from the deep plate interface, indicating that the hydrogen gas may be from a deeper plate boundary interface40. Greater hydrogen production at high pressures could predict that a deeper subsurface environment is more favourable for hydrogen production. However, this topic is still under debate, and additional evidence needs to be presented in the future4,39.
Hydrogen generation and charge transfer
In the previous study, it was hypothesized that two types of bond cleavage may occur in quartz: homolytic cleavage without charge transfer, and heterolytic cleavage with charge transfer15,16. It was further hypothesized that the homolytic process (without charge transfer) leads to hydrogen generation, and H+ is a product of heterolytic cleavage. Charge analysis revealed a distinct contrast between scenarios involving and not involving charge transfer (Fig. 4a). Here, we found that hydrogen molecules were generated in two steps, with the first step, H+ was produced in the form of SiOH2+ or H3O+ with the help of charge transfer, in the second step, hydrogen molecules were produced via inverse charge transfer (Fig. 4b, c). In the absence of charge transfer, following the free radical reaction path results in an average valence state of + 3.6 for active silicon on the surface. This diminishes its reducing properties; the simulation along this pathway shows no hydrogen generation, and the main product is SiH. Upon introducing charge transfer, the average valence of Si becomes + 4, leading to the formation of SiOH2. The application of inverse charge transfer restores the average valence state of surface silicon to + 3.8 which is higher than its original level without charge transfer. This is our current understanding based on first principles simulation, and further research is needed on charge transfer and the effects of different surfaces or other elements such as Fe and Mg.
Friction experiments involving various types of rocks have substantiated the production of hydrogen6, correlating it with seismic activity11. In our simulations, quartz was employed as a unit to model rock and fresh surfaces devoid of reactive oxygen exposure, resulting in the production of hydrogen. Remarkably, we discovered that the direct generation of hydrogen through free radicals alone is not viable in the simulation owing to the inherent instability of these radicals. Regarding charge transfer, we note this phenomenon occurs in geological formations41. During the process of rock friction, the distribution of electrons will be uneven, which can be reflected as charge transfer in our simulation. This discovery offers a perspective on the reaction pathway of natural hydrogen generation.
Oxygen generation and peroxy bond
Geological sources of oxygen are essential for the early Earth’s anoxic atmosphere and for the evolution of early life5,34,42. Oxygen production was also observed in our experiments and simulations. The average distance between the two oxygen atoms in the generated oxygen is 1.26 Å, which agrees with the previous study26. The system with an oxygen generation is shown in fig. S7 and the charge of related atoms is shown in Fig. 4d. On the oxygen-enriched quartz fracture surface at 273 K, we observed the generation of oxygen without charge transfer (Fig. 4e). A low temperature was selected to maintain the water film; otherwise, the film would be vaporized (see our consideration on the system in the Simulation details). The generation of the two oxygen atoms is attributed to SiO and H2O, respectively. This is corroborated by the alterations in the electron states of Si and the two O atoms, confirming oxygen production (Fig. 4e). In the simulation, SiO loses the O atom and subsequently combines with OH to form Si-OH. The process of H2O losing oxygen is relatively complex. The two lost H atoms first form Si-OH with SiO on the fracture surface, and the other forms H3O+ through multiple transfers with water molecules. The valence state of Si in SiO, which supplies oxygen atoms, has not changed. The oxygen from H2O and the O from SiO both changed from − 2 valence to 0 valence. We traced the distance between the two oxygen atoms that generated oxygen, and the reaction pathway is shown in Fig. 4f and g. Initially, the two oxygen atoms belong to SiO and H2O. At 400 fs, one of the OH bonds in H2O is the first to separate, and at 565 fs, the other H atom in H2O separates, and the O atom forms O-O. The SiO distance starts to oscillate at 440 fs, and the oscillation intensifies at 565 fs when the SiOO structure is formed. The SiOO structure breaks up completely at 1,000 fs, the SiO distance rapidly increases, and O2 escapes. Due to the substantial presence of SiO on the oxygen-enriched surface, H is utilized in the production of SiOH, while OH persists and has the potential to combine with SiO, forming SiOOH. Eventually, the H of SiOOH combines with an OH to generate H2O, Si combines with an OH to generate SiOH, and OO is released to generate O2. Throughout the process of oxygen generation, we observed SiOO as an intermediate product, indicating the significance of peroxyl bonding in this phenomenon. Notably, the pathways of oxygen production are complex, and the representations in our simulations illustrate only one of the potential routes.
Our experiments provide direct evidence of hydrogen and oxygen production during rock friction. In addition, the generation of hydrogen molecules occurs through a two-step process in our simulation. Given the instability of Si radicals, we employed charge transfer and inverse charge transfer processes to transform surface Si· into Si+ and subsequently facilitate the formation of SiOH2+, SiOH, and eventually H2. Compared to water, high-pressure ice VII simulations at different temperatures produced more hydrogen. Furthermore, O2 was generated on a reactive oxygen-rich ruptured surface in simulations, indicating that the peroxy bond is the key intermediate. This study reveals a potential reaction pathway of rock comminution generating hydrogen, which inspires additional attempts to use first-principles molecular dynamics to study water-rock interactions using different minerals, such as serpentinization, where charge transfer may play an important role as well. It may further be used for an optimal design to stimulate hydrogen generation by injecting water into reactive formations13.
Experimental details
In high-velocity friction experiments, we employed representative rocks that constitute Earth's crust, including basalt, dunite, granite, marble, and sandstone. Dry specimens underwent a two-week drying process in an oven at 110 °C, while wet specimens were saturated with distilled water in a vacuum desiccator for approximately two days before installation in the high-velocity friction apparatus. Experiments were conducted using a modified rotary-shear, high-speed friction testing apparatus6,41. High-speed slip on a simulated fault under normal stress was achieved by pressing together a pair of hollow/solid cylindrical specimens using an air actuator. Subsequently, one specimen was held stationary while the other was rotated at high speed by a servo motor. Seismic faulting was reproduced in a pressure vessel connected to a gas chromatograph (VARIAN CP4900), and then hydrogen concentration in the vessel was measured immediately after the experiment. The experiments were conducted with a consistent slip velocity of 1.6 m/s and a displacement of 10 m. The simulated fault underwent an acceleration from 0 to 1.6 m/s in 0.5 s, maintained a speed of 1.6 m/s for 6.3 s, and then decelerated to 0 m/s in 0.5 s. Some of the hydrogen production data we published in the previous work and oxygen production in argon atmosphere and some hydrogen data were added in this work.
Simulation details
Ab initio molecular dynamics simulations (AIMD) and spin-polarized density functional theory (DFT) calculations were performed using the CP2K/Quickstep package42. Grimme’s D3 dispersion corrections were employed43. Exchange and correlation were treated with the Perdew−Burke−Ernzerhof (PBE) generalized gradient approximation (GGA) functional44. The core electrons were addressed by norm-conserving pseudopotentials45. A double-ζ-with-polarization (DZVP) basis set was used, and the cutoff value of 280 Ry was chosen because it gives a good compromise between computational accuracy and efficiency. Systems of α-quartz and water/ice VII were prepared by cleaving a quartz model in the middle along the (100), which contains 40 Si atoms and 64 O atoms in a quartz slab. Lattice parameters of quartz−water/ice VII systems are a = 20.00000 Å; b = 11.86572 Å; c = 10.86324 Å, α = β = γ = 90°. In the water system, 32 water molecules were added, which corresponds to the presence of 32 Si radicals on the quartz surface. In the ice VII system, the quartz surface and void space were the same. A proton-disordered ice VII structure consisting of 16 water molecules with zero net dipole moment was used46. The unit cell was duplicated three times along the x and y directions and then cut in the y direction to fit the void space. Finally, there were 54 water molecules in the system. In all the simulations, the volume was fixed, and the temperature was controlled at 500, 1,000, and 1,500 K by a Nosé-Hoover thermostat47. The pressure was an output; for the water system, the pressures were 0.01, 0.86, and 1.30 GPa at 500, 1,000, and 1,500 K, respectively; for the ice VII system, the pressures were 2.77, 4.25, and 5.39 GPa at 500, 1,000, and 1,500 K, respectively (Table 1). An additional simulation system with Si-terminated and SiO terminated surfaces was prepared (Fig. 4b), which was sandwiched by two water films. A large vacuum was used to avoid proton transfer through the Grotthuss mechanism. The initial motivation was to observe whether the hydrogen can be generated on the Si-terminated surface without a charge transfer or not. However, an oxygen molecule was generated during the simulation. This simulation allows us to discuss the importance of charge transfer in the hydrogen generation process. Moreover, the observation of oxygen enabled us to discuss the mechanism of oxygen generation from water-rock interactions.
Data analysis and visualization
A script is used to analyze the atomic bonding in the system. Incorporating relaxation effects, the initial threshold values for determining the bonding of HH, SiO, and OH were set at 1.0, 2.1, and 1.3 Å, respectively. If the distance between two atoms is below this threshold under periodic boundary conditions, they are considered bonded. The upper and lower limits of the fluctuation range are determined by adding or subtracting 0.1 Å from the threshold value. The trajectory data were initially identified using visual molecular dynamics (VMD)48. To enhance the visualization of the generated structures, such as H2, SiOH2, SiOH, and SiH, ChimeraX49 was utilized. VESTA50 was employed to illustrate the hydrogen generation of the entire system and minimize the influence of the crystal skeleton in the diagram. The movie of hydrogen and oxygen generation process
Data availability
All data are available in the main text or the supplementary information. Requests for materials and codes should be addressed to [email protected] and [email protected].
Acknowledgments
We thank the Japan Society for the Promotion of Science (JSPS) for a Grant-in-Aid for Transformative Research Areas and Scientific Research (21H05202, 22H05108, and 23K04647).
Author contributions
Conceptualization: W.C., Y.L., Y.M., T.T. Methodology: Y.L., W.C., T.H. Investigation: W.C., Y.L., T.H., T.T. Funding acquisition: T.T., Y.M. Writing – original draft: W.C. Writing – review & editing: W.C., Y.L., Y.M., T.H., T.T.
Competing interests
Authors declare that they have no competing interests.
- Zgonnik, V. The occurrence and geoscience of natural hydrogen: a comprehensive review. Earth-Sci. Rev.203, 103140 (2020).
- Kelley, D.S. et al. A serpentinite-hosted ecosystem: the Lost City hydrothermal field. Science307, 1428–1434 (2005).
- Telling, J. et al. Rock comminution as a source of hydrogen for subglacial ecosystems. Nat. Geosci.8, 851–855 (2015).
- Ijiri, A. et al. Deep-biosphere methane production stimulated by geofluids in the Nankai accretionary complex. Sci. Adv.4, eaao4631 (2018).
- Stone, J., Edgar, J.O., Gould, J.A. & Telling, J. Tectonically-driven oxidant production in the hot biosphere. Nat. Commun.13, 4529 (2022).
- Hirose, T., Kawagucci, S. & Suzuki, K. Mechanoradical H2 generation during simulated faulting: implications for an earthquake‐driven subsurface biosphere. Geophys. Res. Lett.38, L17303 (2011).
- Di Toro G. et al. Fault lubrication during earthquakes. Nature471, 494–498 (2011).
- Prinzhofer, A., Cissé, C. S. T. & Diallo, A. B. Discovery of a large accumulation of natural hydrogen in Bourakebougou (Mali). Int. J. Hydrogen Energy43, 19315–19326 (2018).
- Rabesandratana, T. Hidden hydrogen. Science379, 630–636 (2023).
- Wakita, H., Kita, I., Fujii, N. & Notsu, K. Hydrogen release: new indicator of fault activity. Science210, 188–190 (1980).
- Waite, J.H., et al. Cassini finds molecular hydrogen in the Enceladus plume: evidence for hydrothermal processes. Science356, 155–159 (2017).
- Osselin, F., et al. Orange hydrogen is the new green. Nat. Geosci.15, 765–769 (2022).
- Kita, I., Matsuo, S. & Wakita, H. H2 generation by reaction between H2O and crushed rock: an experimental study on H2 degassing from the active fault zone. J. Geophys. Res.87, 10789–10795 (1982).
- Saruwatari, K., Kameda, J. & Tanaka, H. Generation of hydrogen ions and hydrogen gas in quartz–water crushing experiments: an example of chemical processes in active faults. Phys. Chem. Miner.31, 176–182 (2004).
- Delogu, F. Hydrogen generation by mechanochemical reaction of quartz powders in water. Int. J. Hydrogen Energy36, 15145–15152 (2011).
- Mayhew, L. E., Ellison, E. T., McCollom, T. M., Trainor, T. P. & Templeton, A. S. Hydrogen generation from low-temperature water–rock reactions. Nat. Geosci.6, 478–484 (2013).
- Klein, F., Grozeva, N. G. & Seewald, J. S. Abiotic methane synthesis and serpentinization in olivine-hosted fluid inclusions. Proc. Natl Acad. Sci.U. S. A. 116, 17666–17672 (2019).
- Lin, L.H. et al. Radiolytic H2 in continental crust: nuclear power for deep subsurface microbial communities. Geochem. Geophys. Geosys.6, Q07003 (2005).
- Gaillard, F., Scaillet, B. & Arndt, N. T. Atmospheric oxygenation caused by a change in volcanic degassing pressure. Nature478, 229–232 (2011).
- Klein, F., Tarnas, D. J. & Bach, W. Abiotic sources of molecular hydrogen on earth. Elements16, 19–24 (2020).
- Bali, E., Audétat, A. & Keppler, H. Water and hydrogen are immiscible in Earth’s mantle. Nature495, 220–222 (2013).
- Tschauner, O. et al. Ice-VII inclusions in diamonds: evidence for aqueous fluid in Earth’s deep mantle. Science359, 1136-1139 (2018).
- Sanz, E., Vega, C., Abascal, J. L. F. & MacDowell, L. G. Phase diagram of water from computer simulation. Phys. Rev. Lett.92, 255701 (2004).
- Mao, W.L. et al. X-ray–Induced dissociation of H2O and formation of an O2–H2 alloy at high pressure. Science314, 636–638 (2006).
- Liang, Y. & Tse, J. S. Mechanisms for the formation of H2 and O2 from x-ray irradiated dense ice. Phys. Rev. B79, 104105 (2009).
- Duval, Y., Mielczarski, J. A., Pokrovsky, O. S., Mielczarski, E. & Ehrhardt, J. J. Evidence of the existence of three types of species at the quartz−aqueous solution interface at pH 0-10: XPS surface group quantification and surface complexation modeling. J. Phys. Chem. B106, 2937–2945 (2002).
- Bloss, F. D. & Gibbs, G. V. Cleavage in quartz. Am. Mineral.48, 821–838 (1963).
- Murashov, V. V. Reconstruction of pristine and hydrolyzed quartz surfaces. J. Phys. Chem. B109, 4144–4151 (2005).
- Deng, Y. et al. Crystal face dependent wettability of α-quartz: elucidation by time-of-flight secondary ion mass spectrometry techniques combined with molecular dynamics. J. Colloid Interface Sci.607, 1699–1708 (2022).
- Adeagbo, W. A., Doltsinis, N. L., Klevakina, K. & Renner, J. Transport processes at α-quartz-water interfaces: insights from first-principles molecular dynamics simulations. ChemPhysChem9, 994–1002 (2008).
- Ledyastuti, M., Liang, Y. & Matsuoka, T. The first‐principles molecular dynamics study of quartz‐water interface. Int. J. Quantum Chem.113, 401–412 (2013).
- Prigiobbe, V., Suarez Negreira, A. & Wilcox, J. Interaction between olivine and water based on density functional theory calculations. J. Phys. Chem. C117, 21203–21216 (2013).
- He, H. et al. A mineral-based origin of Earth’s initial hydrogen peroxide and molecular oxygen. Proc. Natl Acad. Sci.U. S. A.120, e2221984120 (2023).
- Futera, Z., Yong, X., Pan, Y., John, S. T. & English, N. J. Formation and properties of water from quartz and hydrogen at high pressure and temperature. Earth Planet. Sci. Lett.461, 54–60 (2017).
- Ding, Z. & Selloni, A. Electron trapping and ion leaching at the Li-modified quartz–water interface. J. Phys. Chem. C124, 26741–26747 (2020).
- Stolte, N., Hou, R. & Pan, D. Nanoconfinement facilitates reactions of carbon dioxide in supercritical water. Nat. Commun.13, 5932 (2022).
- Hochstrasser, G. & Antonini, J. F. Surface states of pristine silica surfaces: I. ESR studies of Es′ dangling bonds and of CO2− adsorbed radicals. Surf. Sci.32, 644–664 (1972).
- Suzuki, N., Koike, K., Kameda, J. et al. Thermogenic methane and hydrogen generation in subducted sediments of the Nankai Trough. Commun. Earth. Environ.5, 97 (2024).
- Tsuji, T., Ashi, J., Strasser, M. & Kimura, G. Identification of the static backstop and its influence on the evolution of the accretionary prism in the Nankai Trough. Earth and Planet. Sci. Lett.431, 15–25 (2015).
- Freund, F. T., Takeuchi, A. & Lau, B. W. Electric currents streaming out of stressed igneous rocks–A step towards understanding pre-earthquake low frequency EM emissions. Phys. Chem. Earth 31, 389–396 (2006).
- Shimamoto, T. & Tsutsumi, A., A new rotary-shear high-velocity frictional testing machine: Its basic design and scope of research. Struct. Geol.39, 65–78 (1994).
- VandeVondele, J., Krack, M., Mohamed, F., Parrinello, M., Chassaing, T. & Hutter, J. Quickstep: fast and accurate density functional calculations using a mixed Gaussian and plane waves approach. Comput. Phys. Commun.167, 103–128 (2005).
- Grimme, S., Antony, J., Ehrlich, S. & Krieg, H. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys.132, 154104 (2010).
- Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett.77, 3865-3868 (1996).
- Goedecker, S., Teter, M. & Hutter, J. Separable dual-space Gaussian pseudopotentials. Phys. Rev. B54, 1703−1710 (1996).
- Kuo, J. L. & Klein, M. L. Structure of Ice-VII and Ice-VIII: a quantum mechanical study. J. Phys. Chem. B108, 19634–19639 (2004).
- Nosé, S. A molecular dynamics method for simulations in the canonical ensemble. Mol. Phys. 52, 255–268 (1984).
- Humphrey, W., Dalke, A. & Schulten, K. VMD: visual molecular dynamics. J. Mol. Graphics14, 33–38 (1996).
- Goddard, T. D.et al. UCSF ChimeraX: meeting modern challenges in visualization and analysis. Protein Sci.27, 14–25 (2018).
- Momma, K. & Izumi, F. VESTA 3 for three-dimensional visualization of crystal, volumetric and morphology data. J. Appl. Crystallogr.44, 1272–1276 (2011).
Table 1. Number of hydrogen molecules, SiOH2 and SiOH produced during the two-step reactions.
There is NO Competing Interest.
- O2.mp4
Oxygen generation
- H2.mp4
Second step of hydrogen generation
- SiOH2.mp4
First step of hydrogen generation
- HydrogenSupplementaryInformation20240301.pdf
Supplementary materials