https://doi.org/10.1130/G48495.1

Manuscript received 22 July 2020 Revised manuscript received 12 October 2020 Manuscript accepted 6 February 2021

© 2021 The Authors. Gold Open Access: This paper is published under the terms of the CC-BY license. Published online 15 April 2021

Ferric iron triggers greenalite formation in simulated Archean seawater Isaac L. Hinz1, Christine Nims1, Samantha Theuer1, Alexis S. Templeton2 and Jena E. Johnson1 1Department of Earth and Environmental Sciences, University of Michigan, 1100 North University Avenue, Ann Arbor, Michigan 48109, USA 2Department of Geological Sciences, University of Colorado–Boulder, 2200 Colorado Avenue, Boulder, Colorado 80309, USA

ABSTRACT Estimates of dissolved iron concentrations in Sedimentary rock deposits provide the best records of (bio)geochemical cycles in the the pre-oxygenated ocean vary between 20 µM ­ancient ocean. Studies of these sedimentary archives show that greenalite, an Fe(II) silicate and 2 mM Fe(II) depending on constraints set with low levels of Fe(III), was an early chemical precipitate from the Archean ocean. To bet- by siderite (Holland, 1984; Canfield, 2005) or ter understand the formation of greenalite, we explored controls on iron silicate precipitation ferrous greenalite (Jiang and Tosca, 2019) satu- through experiments in simulated Archean seawater under exclusively ferrous conditions ration. These geochemical differences prevent or supplemented with low Fe(III). Our results confirm a pH-driven process promoting the direct translations between the high-concentra- precipitation of iron-rich silicate phases, and they also reveal an important mechanism in tion, high-pH synthesis experiments and ancient which minor concentrations of Fe(III) promote the precipitation of well-ordered greenalite processes in the Archean ocean. among other phases. This discovery of an Fe(III)-triggering iron silicate formation process A few previous studies explored iron silicate suggests that Archean greenalite could represent signals of iron oxidation reactions, poten- mineralization under reducing and geochemi- tially mediated by life, in circumneutral ancient seawater. cally relevant conditions plausible for the early ocean. Under reducing conditions with very low INTRODUCTION inclusions of early iron silicates (Rasmussen concentrations of iron and silica, Harder (1978) The early ocean shaped the evolution of mi- et al., 2015; Muhling and Rasmussen, 2020). produced Fe[II,III]-silica coprecipitates at pH croscopic life and supported the development of The best-preserved inclusions were character- 7–9 but did not include structural data to confirm

the earliest (bio)geochemical cycles. Chemically ized as greenalite [Fe(II)3Si2O5(OH)4], the iron iron silicate formation. Later studies similarly precipitated sediments record the (bio)geochem- end member of the two-layer serpentine group, showed the anoxic formation of poorly crystal- istry of their aqueous environments (Derry and but with ∼10%–25% Fe(III) (Johnson et al., line iron-silica coprecipitates but were unable Jacobsen, 1990), and deeper marine strata from 2018). This small proportion of Fe(III) in Ar- to robustly characterize these phases (Farmer, the Archean Eon (4.0–2.5 Ga) are dominated chean greenalite suggests that partial iron oxida- 1991; Konhauser et al., 2007). Recently, Tosca by iron formations (silica-rich chemical sedi- tion could have favored iron-silica precipitation, et al. (2016) examined Fe(II)-silicate formation ments with >15% iron) (Simonson and Hassler, and indeed, metastable Fe[II,III] hydroxy salts in Archean seawater-like solutions and proposed 1996). The long-term, widespread deposition (green rust phases) were proposed as potential that poorly ordered experimental precipitates of iron formations suggests anoxic conditions precursors to iron formation silicates and oxides detected at pH ≥7.5 were precursors to green- and high iron concentrations in seawater, yet (Halevy et al., 2017). alite. In contrast, at pH 7, Halevy et al. (2017) the accumulation of iron-rich sediments was Understanding the formation process of iron observed that Fe[II,III] green rust formed upon likely triggered by a chemical change or (bio) silicates under ancient marine conditions should partial Fe oxidation under low concentrations of geochemical process that induced mineral pre- yield insights into Archean (bio)geochemistry. silica (100–250 μM), transforming into three- cipitation (Holland, 1984; Bekker et al., 2014; Existing iron silicate synthesis protocols have layer iron silicates in solutions upon 22 °C and Johnson and Molnar, 2019). relied on oversaturated iron and silica chemi- 50 °C aging for days to months. The identification of the original iron cal conditions set at high pH (8 to >12) (e.g., While this research demonstrates that iron- precipitate(s) is integral to unraveling the an- Decarreau and Bonnin, 1986; Baldermann et al., silica precipitates can form in simulated ancient cient process(es) that induced the deposition 2014; Chemtob et al., 2015; Dzene et al., 2018). seawater, no study has conclusively shown of Archean marine iron formations. Though However, two independent models of early Earth greenalite formation in predicted Archean ocean the iron oxides prevalent in iron formations marine geochemistry estimated Archean oce- chemistry. It is also unclear whether the Fe(III) are largely accepted as primary components anic pH between ∼6.5 and 7.5 (Halevy and in the ancient iron silicate inclusions signals a of these deeper marine sediments (Beukes and Bachan, 2017; Krissansen-Totton et al., 2018). role for iron oxidation in the precipitation of Gutzmer, 2008; Bekker et al., 2014; Sun et al., Geological observations suggest that dissolved these iron-silica phases. Therefore, we explored 2015), rigorous characterization of well-pre- silica was near or at saturation with amorphous the precipitation and crystallization of iron and served chert associated with ca. 3.5 to ca. 2.4 Ga silica (∼1.9 mM) in the Archean ocean (Siever, silica precipitates in artificial ancient seawater at iron formations documented nanometer-scale 1992; Maliva et al., 2005; Stefurak et al., 2015). the predicted Archean pH range of 6.5–7.5 under

CITATION: Hinz, I.L., et al., 2021, Ferric iron triggers greenalite formation in simulated Archean seawater: Geology, v. 49, p. 905–910, https://doi.org/10.1130/G48495.1

Geological Society of America | GEOLOGY | Volume 49 | Number 8 | www.gsapubs.org 905

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/49/8/905/5362185/g48495.1.pdf by guest on 27 September 2021 both strictly ferrous conditions and in systems pH CONTROL ON IRON-RICH els. The quantity of solids at pH 7.5 in- with low levels of Fe(III). SILICATE FORMATION creased approximately fivefold with Fe(III) In measuring the mass of precipitated phas- (Table 1), producing deep-green precipitates EXPERIMENTAL METHODOLOGY es and the associated changes in solution chem- (Fig. 2A; Fig. S1). The 25 °C bulk solid phase Experiments were performed in 500 mL bo- istry (Table 1), we did not detect iron-silica was XRD-amorphous (Fig. S4) and >80% rosilicate glass bottles at 25 °C under anoxic precipitation, but observed possible colloidal Fe(II) (Table 1). SEM (Fig. S5A) and TEM conditions. Solutions were first supplemented silica at pH 6.5 and 7 under anoxic and ferrous (Figs. 2B–2E) imaging revealed abundant na- with 1 mM sodium orthosilicate and stirred for conditions (see the Supplemental Material). noscale tubular structures and poorly layered 24 h at their respective experimental pH (6.5, Low quantities of visible iron-silica coprecipi- curling structures. These were morphologi- 7, or 7.5), ensuring full silica depolymerization tate did form at the higher pH of 7.5 (Fig. S1). cally similar to the ferrous layered precipitate (Dietzel, 2000). We added seawater salts, then The mass of the observable precipitate at pH (Figs. S3D and S3E). HR-TEM also indicated ferrous or ferrous-ferric chloride and bicar- 7.5 was still not measurable, but changes in so- iron-oxide precipitates at 25 °C, including bonate buffer, and set the final pH of solutions lution chemistry after 5 days at 25 °C indicated euhedral crystals with magnetite lattice spac- ­(Table S1 in the Supplemental Material1). Given a decrease in Fe(II) from 1 mM to 0.78 mM ing (Fig. S5D). Even after 150 °C aging, XRD our focus on iron silicate formation, we omitted and a decrease in Si from 1 mM to 0.73 mM of the Fe[II,III] pH 7.5 experiment did not de- calcium in the media to preclude the precipita- (Table 1). We were unable to subsample the tect ∼7 Å silicates (Fig. S4), and TEM imag- tion of calcium carbonate. negligible pH 7.5 precipitate at 25 °C, so the ing showed curling layered structures instead After 5 days of low-temperature (25 °C) re- entire pH 7.5 precipitate was aged at 150 °C of well-ordered stacks (Figs. 2G–2I; Figs. action in the dark, the experiments were sub- and developed into a similar quantity of darker S6B–S6D). In HR-TEM, a lack of repeating sampled for aqueous and solid analyses and gray material (Fig. 1A). layers precluded definitive measurements, but secondary hydrothermal aging. For aging, sub- A nanoscale examination of this 150 °C– we estimated a basal layer spacing of ∼7.9 Å samples of 25 °C precipitates were sealed in aged ferrous pH 7.5 sample illuminated three (Fig. 2 K). While amorphous by selected area a gas-tight Parr vessel (22 mL) under anoxic distinct morphologies (Fig. 1). We observed ag- electron diffraction (SAED) (Fig. S3), the lay- conditions and heated at 150 °C for 5 days to gregates of layered structures, webs of amor- ered phases had elemental chemistry consis- enhance crystallization, simulating ∼100 m.y. phous gel, and rare hexagonal platelets (Fig. 1C; tent with a serpentine-group silicate similar of low-temperature diagenesis (Siever, 1983). Fig. S2C). The few hexagonal platelets were ten- to greenalite (Fig. S3) and lacked diagnostic

Dissolved iron, silica, and the Fe(III)/FeT tatively identified as green rust based on mor- green rust XRD peaks (Fig. S4). We also ob- (T—total) content of precipitates were mea- phological similarities to those identified by served magnesite and magnetite by TEM and sured using colorimetric assays. We used scan- ­Halevy et al. (2017). The layered structures were XRD after 150 °C aging (Fig. 2J; Figs. S4 and ning electron microscopy (SEM) paired with <10 nm thick and poorly ordered with curling S7). Though the introduction of Fe(III) at pH energy dispersive X-ray spectroscopy (EDS) features typical of frequent edge dislocations 7.5 did not substantially alter the morphology to image and collect micron-scale elemental (Figs. 1D–1F). One layered structure was suf- or identity of the poorly ordered silicate-like maps of the experimental precipitates. Trans- ficiently crystalline for HR-TEM and had 7.3 Å phase, it did greatly increase the quantity of mission electron microscopy (TEM), selected basal spacing consistent with a serpentine-group layered structures and eliminated the presence area electron diffraction (SAED), and scan- clay (Fig. 1F). This lattice fringe spacing and of the amorphous gel phase. ning and high-resolution TEM (STEM and the elemental chemistry (Fig. 1F; Fig. S3) sug- At pH lower than 7.5, precipitation occurred HR-TEM) enabled the compilation of na- gest the formation of two-layer Fe-Mg silicates. only upon the introduction of a mixture of fer- noscale structural and elemental data on the We therefore conclude that in ferrous iron ric and ferrous iron. Completely ferrous experi- mineral phases. Additionally, we used a cobalt- solutions, higher-pH aqueous conditions pro- ments did not yield extractable precipitates at pH sourced X-ray diffractometer (XRD) on anoxi- mote iron-rich silicate precipitation under an- 6.5 and 7 (Table 1; Fig. S1). However, with 5% cally sealed precipitate slurries to determine oxic, environmentally relevant conditions. This of the total iron added as Fe(III) to the lower- mineral d-spacings. A full description of the pH-driven precipitation mechanism initiates at pH experiments, orange solids formed after 5 ­experimental and analytical methods is avail- a pH of ∼7.5 with amorphous material, similar days of 25 °C precipitation and equilibration able in the Supplemental Material. to the findings of Tosca et al. (2016). These ex- (Fig. 3A; Fig. S1). The 25 °C precipitate largely perimental results are congruous with previous consisted of iron-silica coprecipitates in amor- studies that found high-pH aqueous conditions phous aggregates as determined by SEM and 1Supplemental Material. A complete description of the experimental methods, analytical techniques, and conducive to authigenic iron silicate clay forma- TEM (Fig. 3B; Fig. S8A). Although the 25 °C supporting data and information regarding mineral tion (Dzene et al., 2018). solids were dominated by amorphous material, identifications. Please visithttps://doi.org/10.1130/ we detected a thin wavy structure in the pH 7 GEOL.S.14320340 to access the supplemental FERRIC IRON TRIGGERS SILICATE precipitate that could correspond to early silicate material, and contact [email protected] with FORMATION layers (Figs. 3E–3G; Fig. S9A). In pH 6.5 exper- any questions. Underlying data can be found in the University of Michigan Deep Blue online repository: Adding 5% of the total iron as Fe(III) iments, we observed two examples of iron-rich https://doi.org/10.7302/k2hs-nh21. stimulated precipitation at all tested pH lev- ∼7 Å layered structures at 25 °C (Figs. 3B and

TABLE 1. SUMMARY OF EXPERIMENTAL RESULTS 0% Fe(III) 0% Fe(III) 0% Fe(III) 5% Fe(III) 5% Fe(III) 5% Fe(III) Experimental condition pH 6.5 pH 7.0 pH 7.5 pH 6.5 pH 7.0 pH 7.5 Dried mass after 12 h (mg) ± 0.01 3.18 4.68 3.31 6.54 7.73 15.86 Silica in solution after 5 days, 25 °C (mM) ± standard deviation 0.57 ± 0.2 0.77 ± 0.2 0.73 ± 0.2 0.34 ± 0.2 0.39 ± 0.2 0.38 ± 0.2 Fe(II) in solution after 5 days 25 °C (mM) ± standard deviation 0.99 ± 0.2 0.86 ± 0.2 0.78 ± 0.2 0.58 ± 0.2 0.40 ± 0.2 0.31 ± 0.2 Fe(II) in solids after 5 days, 25°C – – – 0.96 0.89 0.18 FeT Fe(III) in solids after 5 days, 150°C – – – 0.60 0.54 0.30 FeT

906 www.gsapubs.org | Volume 49 | Number 8 | GEOLOGY | Geological Society of America

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/49/8/905/5362185/g48495.1.pdf by guest on 27 September 2021 A B C D E F

Figure 1. Transmission electron microscopy (TEM), scanning TEM (STEM), and high-resolution TEM (HR-TEM) images of experimental pre- cipitates at pH 7.5. Minimal gray-colored precipitates formed in ferrous conditions after 150 °C aging, shown in an epitube (A), with TEM images revealing amorphous films (B), hexagonal platelets (C, enlarged in inset), and silicate-like curled clusters (B, D–E) with representative elemental chemistry of layered iron silica phases (“layered”) (D inset) and ∼7 Å lattice spacing (E,F). HAADF—High-angle annular dark-field imaging; Rel.—relative.

3C; Figs. S8C–S8E). This uncommon but well- ordered 7 Å silicates. TEM imaging revealed clay and/or carbonate green rust (Fig. S4). ordered stacked phase had elemental chemistry considerable amorphous precipitates and mi- SEM and TEM images showed a common consistent with a Fe-Mg silicate (Fig. S3); thus, nor crystalline phases identified as magnetite, phase with hexagonal morphology, low-Si in tandem with its basal spacing, we interpret it hematite, and magnesite (Figs. 3H–3S; Figs. elemental chemistry, and 3.1 Å a-axis spac- to be high-Mg greenalite. S3, S7, S10, and S11). XRD confirmed these ing characteristic of a carbonate green rust Aging of the pH 6.5 and 7 low-Fe(III) minerals and also indicated the presence of (Fig. 3S; Figs. S3 and S10A) (Christiansen precipitates at 150 °C yielded a diversity of one or more phases with ∼7 Å spacing, con- et al., 2009). We also observed substantial phases, including extensive formation of well- sistent with the presence of a serpentine-group formation of a layered phase with 7.3–7.4 Å

A B C D E

F G H IKJ

Figure 2. Transmission electron microscopy (TEM), scanning TEM (STEM), and high-resolution TEM (HR-TEM) images of 5% Fe(III)/FeT (T—total) precipitates at pH 7.5, producing green precipitates at 25 °C (A) dominated by poorly ordered curling structures (B–E). The layered phase (“layered”) had iron silicate-like chemistry by energy dispersive X-ray spectroscopy (D inset). TEM imaging of this Fe[II,III] precipi- tate aged at 150 °C (F) shows clusters of layered structures alongside amorphous material (G–I) and nanomagnetite (J). The layered phases (“layered”) had chemistry consistent with an iron-magnesium silicate by EDS (I inset). Layered structures in H had 7.9 Å lattice spacing (K). HAADF—High-angle annular dark-field imaging; Rel.—relative.

Geological Society of America | GEOLOGY | Volume 49 | Number 8 | www.gsapubs.org 907

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/49/8/905/5362185/g48495.1.pdf by guest on 27 September 2021 A B C D E F

G

H I J K L

M

NO P Q R

S

Figure 3. Transmission electron microscopy (TEM), scanning TEM (STEM), and high-resolution TEM (HR-TEM) images of pH 6.5 and 7

experimental precipitates. (A) Orange precipitates formed at 25 °C and pH 6.5 with 5% Fe(III)/FeT (T—total). (B) TEM imaging of amorphous material, with one inset showing representative energy dispersive X-ray spectroscopy (EDS) (“Amorphous”) and another inset of selected area electron diffraction (SAED) of amorphous material, along with 7 Å Fe-Mg silicates (C, with inset showing representative EDS (“layered”).

(D) Orange-green precipitate at 25 °C and pH 7 with 5% Fe(III)/FeT is amorphous (E, with one inset showing representative EDS (“Amorphous”) and another inset of SAED of amorphous material), along with rare acicular structures (F,G). (H) Orange-brown precipitates from the 150 °C aging of 5% ferric iron pH 6.5 samples (I), including ∼7 Å layered phases (J,K) with a diffraction pattern and Fe-Mg silicate chemistry by EDS consistent with serpentine group minerals (J insets), iron-oxide spherules (L), and magnetite in triangular platelets (M). (N) Deep-green precipitates form from the 150 °C aging of pH 7 solids containing 5% Fe(III), and indicate multiple phases (O) including both 7 Å layered and amorphous phases (P,Q), and green rust as hexagonal platelets (O), iron oxide nano-inclusions (R), and hexagonal green rust (S). HAADF— High-angle annular dark-field imaging; Rel.—relative.

spacing (Figs. 3 K and 3Q) and serpentine IMPLICATIONS mineral inclusions in Archean marine chert. Our group–like elemental chemistry (Fig. S3). Our aggregated data confirm that high-Mg experimental products contained coexisting ox- Additionally, SAED on multiple orientations­ greenalite was a major mineral product in the idized and reduced phases in thermodynamic of this phase displayed a polynanocrystalline crystallized Fe[II,III] pH 6.5 and 7 experiments. disequilibrium, while well-preserved Archean diffraction pattern consistent with serpentine- However, we observe two discrepancies between chert hosts primary greenalite inclusions with- group minerals (Fig. 3J). this experimental assemblage and the early iron- out magnetite and hematite (Rasmussen et al.,

908 www.gsapubs.org | Volume 49 | Number 8 | GEOLOGY | Geological Society of America

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/49/8/905/5362185/g48495.1.pdf by guest on 27 September 2021 2016). Another inconsistency arises in our Mg- REFERENCES CITED Halevy, I., Alesker, M., Schuster, E.M., Popovitz- enriched (∼10 at% [atomic percent]) experi- Baldermann, A., Dohrmann, R., Kaufhold, S., Nick- Biro, R., and Feldman, Y., 2017, A key role mental greenalite compared to the Mg-depleted el, C., Letofsky-Papst, I., and Dietzel, M., 2014, for green rust in the Precambrian oceans and The Fe-Mg-saponite solid solution series—A the genesis of iron formations: Nature Geosci- (<0.5 at%) ancient greenalite (Johnson et al., hydrothermal synthesis study: Clay Minerals, ence, v. 10, p. 135–139, https://doi.org/10.1038/ 2018). Future experimental work should explore v. 49, p. 391–415, https://doi.org/10.1180/clay- ngeo2878. the impact of lower aqueous Mg concentration min.2014.049.3.04. Harder, H., 1978, Synthesis of iron layer silicate min- on silicate chemistry, the effect of simulated dia- Bekker, A., Planavsky, N.J., Rasmussen, B., Krapež, erals under natural conditions: Clays and Clay B., Hofmann, A., Slack, J.F., Rouxel, O.J., and Minerals, v. 26, p. 65–72, https://doi.org/10.1346/ genesis on Mg content in greenalite, and the role Konhauser, K.O., 2014, Iron formations: Their CCMN.1978.0260108. of concurrent carbonate-mineral precipitation as origins and implications for ancient seawa- Holland, H.D., 1984, The Chemical Evolution of the an alternative Mg sink. ter chemistry, in Mackenzie, F.T., ed., Treatise Atmosphere and Oceans: Princeton, New Jer- While not an exact replica of the Archean on Geochemistry (second edition), Volume 9: sey, Princeton University Press, 598 p., https:// ocean, our experiments demonstrate that another Sediments, Diagenesis and Sedimentary Rocks: doi​.org/10.1515/9780691220239. Amsterdam, Elsevier, p. 561–628, https://doi​ Jiang, C.Z., and Tosca, N.J., 2019, Fe(II)-carbonate formation mechanism of iron-rich silicates ex- .org/10.1016/B978-0-08-095975-7.00719-1. precipitation kinetics and the chemistry of an- ists wherein ferric iron induces the precipita- Beukes, N.J., and Gutzmer, J., 2008, Origin and oxic ferruginous seawater: Earth and Planetary tion of iron silicate phases (analogous to the ­paleoenvironmental significance of major iron Science Letters, v. 506, p. 231–242, https://doi​ Fe[II,III]-phosphate system; Cosmidis et al., formations at the Archean-Paleoproterozoic .org/10.1016/j.epsl.2018.11.010. boundary, in Hagemann, S., et al., eds., Band- Johnson, J.E., and Molnar, P.H., 2019, Widespread 2014). We observed the formation of well-or- ed Iron Formation-Related High-Grade Iron and persistent deposition of iron formations dered 7 Å Fe-Mg silicates at lower pH values Ore: Society of Economic Geologists Reviews for two billion years: Geophysical Research­ than those previously reported for Fe(II) sili- in Economic Geology 15, p. 5–47, https://doi​ Letters, v. 46, p. 3327–3339, https://doi​ cate formation (≥pH 7.5; Tosca et al., 2016; .org/10.5382/Rev.15.01. .org/10.1029/2019GL081970. this study) or Mg silicate formation ( pH 8.6; Canfield, D.E., 2005, The early history of atmospher- Johnson, J.E., Muhling, J.R., Cosmidis, J., Ras- > ic oxygen: Homage to Robert M. Garrels: An- mussen, B., and Templeton, A.S., 2018, Low- Tosca et al., 2011) experiments. nual Review of Earth and Planetary Sciences, Fe(III) greenalite was a primary mineral Low amounts of ferric iron with 150 °C v. 33, p. 1–36, https://doi.org/10.1146/annurev​ from Neoarchean oceans: Geophysical Re- aging at the lower pH values of 6.5 and 7 ap- .earth.33.092203.122711. search Letters, v. 45, p. 3182–3192, https://doi​ peared to stimulate the precipitation of iron- Chemtob, S.M., Nickerson, R.D., Morris, R.V., Agres- .org/10.1002/2017GL076311. ti, D.G., and Catalano, J.G., 2015, Synthesis and Konhauser, K.O., Amskold, L., Lalonde, S.V., Posth, rich silicates, with initial layered phases start- structural characterization of ferrous trioctahedral N.R., Kappler, A., and Anbar, A., 2007, Decou- ing to form at 25 °C (e.g., Fig. 3C). Unlike smectites: Implications for clay mineral genesis pling photochemical Fe(II) oxidation from shal- previous work that produced green rust at 25 and detectability on Mars: Journal of Geophysical low-water BIF deposition: Earth and Planetary °C from partial iron oxidation with lower, less- Research: Planets, v. 120, p. 1119–1140, https:// Science Letters, v. 258, p. 87–100, https://doi​ relevant Si concentrations (Halevy et al., 2017), doi.org/10.1002/2014JE004763. .org/10.1016/j.epsl.2007.03.026. Christiansen, B.C., Balic-Zunic, T., Dideriksen, K., Krissansen-Totton, J., Arney, G.N., and Catling, D.C., our higher-Si (1 mM) experiments instead gen- and Stipp, S.L.S., 2009, Identification of green 2018, Constraining the climate and ocean pH of erated amorphous Fe-Si precipitates at 25 °C. rust in groundwater: Environmental Science & the early Earth with a geological carbon cycle While green rust did form after 150 °C aging, Technology, v. 43, p. 3436–3441, https://doi​ model: Proceedings of the National Academy our observations suggest that the mineraliza- .org/10.1021/es8011047. of Sciences of the United States of America, Cosmidis, J., Benzerara, K., Morin, G., Busigny, V., v. 115, p. 4105–4110, https://doi.org/10.1073/ tion pathway to the greenalite-like silicates Lebeau, O., Jézéquel, D., Noël, V., Dublet, G., pnas.1721296115. did not proceed through a metastable green- and Othmane, G., 2014, Biomineralization of Maliva, R.G., Knoll, A.H., and Simonson, B.M., 2005, rust phase but rather through an amorphous iron-phosphates in the water column of Lake Secular change in the Precambrian silica cycle: precursor. Pavin (Massif Central, France): Geochimica et Insights from chert petrology: Geological Society In the context of the ferruginous Archean Cosmochimica Acta, v. 126, p. 78–96, https:// of America Bulletin, v. 117, p. 835–845, https:// doi.org/10.1016/j.gca.2013.10.037. doi.org/10.1130/B25555.1. oceans (Poulton and Canfield, 2011), our re- Decarreau, A., and Bonnin, D., 1986, Synthesis and Muhling, J.R., and Rasmussen, B., 2020, Widespread sults suggest that the oxidation of a small portion crystallogenesis at low temperature of Fe(III)- deposition of greenalite to form Banded Iron For- of ferrous iron would trigger iron-rich silicate smectites by evolution of coprecipitated gels: mations before the Great Oxidation Event: Pre- precipitation. This oxidative step, potentially Experiments in partially reducing conditions: cambrian Research, v. 339, 105619, https://doi​ Clay Minerals, v. 21, p. 861–877, https://doi​ .org/10.1016/j.precamres.2020.105619. mediated by iron-oxidizing photosynthesis or .org/10.1180/claymin.1986.021.5.02. Poulton, S.W., and Canfield, D.E., 2011, Ferrugi- through reactions between ferrous iron and oxy- Derry, L.A., and Jacobsen, S.B., 1990, The chemi- nous conditions: A dominant feature of the gen, may have been essential for the precipita- cal evolution of Precambrian seawater: Evidence ocean through Earth’s history: Elements, v. tion of Archean iron silicates if marine pH was from REEs in banded iron formations: Geochi- 7, p. 107–112, https://doi.org/10.2113/gsele- 6.5–7 as estimated (Halevy and Bachan, 2017). mica et Cosmochimica Acta, v. 54, p. 2965–2977, ments.7.2.107. https://doi.org/10.1016/0016-7037(90)90114-Z. Rasmussen, B., Krapež, B., Muhling, J.R., and Su- Consequently, if Archean oceanic pH can be Dietzel, M., 2000, Dissolution of silicates and the vorova, A., 2015, Precipitation of iron silicate constrained below 7.5, the extensive and pro- stability of polysilicic acid: Geochimica et Cos- nanoparticles in early Precambrian oceans marks longed deposition of low-Fe(III) iron silicates mochimica Acta, v. 64, p. 3275–3281, https://doi​ Earth’s first iron age: Geology, v. 43, p. 303–306, (Johnson et al., 2018; Muhling and Rasmussen, .org/10.1016/S0016-7037(00)00426-9. https://doi.org/10.1130/G36309.1. Dzene, L., Brendlé, J., Limousy, L., Dutournié, P., Rasmussen, B., Muhling, J.R., Suvorova, A., and 2020) could signal the presence of early mi- Martin, C., and Michau, N., 2018, Synthesis of Krapež, B., 2016, Dust to dust: Evidence for crobial life. iron-rich tri-octahedral clay minerals: A review: the formation of “primary” hematite dust in Applied Clay Science, v. 166, p. 276–287, https:// banded iron formations via oxidation of iron ACKNOWLEDGEMENTS doi.org/10.1016/j.clay.2018.09.030. silicate nanoparticles: Precambrian Research, We thank Itay Halevy, Jens Gutzmer, and an anonymous Farmer, V.C., 1991, Synthetic allophane and layer- v. 284, p. 49–63, https://doi.org/10.1016/j.pre-

reviewer for input that improved the manuscript. This silicate formation in SiO2-Al2O3-FeO-Fe2O3- camres.2016.07.003. study was supported by NASA Exobiology funding MgO-H2O systems at 23°C and 89°C in a Siever, R., 1983, Burial history and diagenetic reac- (award #80NSSC18K1060 to Johnson and Templeton). ­calcareous environment: Clays and Clay Miner- tion kinetics: American Association of Petroleum We thank David Diercks (Colorado School of Mines, als, v. 39, p. 561–570, https://doi.org/10.1346/ Geologists Bulletin, v. 67, p. 684–691, https:// Golden, Colorado, USA) for assistance with TEM CCMN.1991.0390601. doi.org/10.1306/03B5B67D-16D1-11D7- imaging and analyses, and Victoria Jarvis (McMaster Halevy, I., and Bachan, A., 2017, The geologic history 8645000102C1865D. University, Hamilton, Ontario, Canada) for support in of seawater pH: Science, v. 355, p. 1069–1071, Siever, R., 1992, The silica cycle in the Precambri- developing anoxic Co-XRD methodology. https://doi.org/10.1126/science.aal4151. an: Geochimica et Cosmochimica Acta, v. 56,

Geological Society of America | GEOLOGY | Volume 49 | Number 8 | www.gsapubs.org 909

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/49/8/905/5362185/g48495.1.pdf by guest on 27 September 2021 p. 3265–3272, https://doi.org/10.1016/0016- America Bulletin, v. 127, p. 1090–1107, https:// and Planetary Science Letters, v. 306, p. 11–22, 7037(92)90303-Z. doi.org/10.1130/B31181.1. https://doi.org/10.1016/j.epsl.2011.03.041. Simonson, B.M., and Hassler, S.W., 1996, Was the Sun, S., Konhauser, K.O., Kappler, A., and Li, Y.-L., Tosca, N.J., Guggenheim, S., and Pufahl, P.K., 2016, deposition of large Precambrian iron formations 2015, Primary hematite in Neoarchean to Pa- An authigenic origin for Precambrian greenalite: linked to major marine transgressions?: The Jour- leoproterozoic oceans: Geological Society of Implications for iron formation and the chemis- nal of Geology, v. 104, p. 665–676, https://doi​ America Bulletin, v. 127, p. 850–861, https:// try of ancient seawater: Geological Society of .org/10.1086/629861. doi.org/10.1130/B31122.1. America Bulletin, v. 128, p. 511–530, https://doi​ Stefurak, E.J.T., Lowe, D.R., Zentner, D., and Fischer, Tosca, N.J., Macdonald, F.A., Strauss, J.V., Johnston, .org/10.1130/B31339.1. W.W., 2015, Sedimentology and geochemistry of D.T., and Knoll, A.H., 2011, Sedimentary talc Archean silica granules: Geological Society of in Neoproterozoic carbonate ­successions: Earth Printed in USA

910 www.gsapubs.org | Volume 49 | Number 8 | GEOLOGY | Geological Society of America

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/49/8/905/5362185/g48495.1.pdf by guest on 27 September 2021