Sedimentology and Geochemistry of Archean Silica Granules Sedimentology and Geochemistry of Archean Silica Granules

Sedimentology and geochemistry of Archean silica granules Sedimentology and geochemistry of Archean silica granules

Elizabeth J.T. Stefurak1,†, Donald R. Lowe1, Danielle Zentner1, and Woodward W. Fischer2 1Department of Geological and Environmental Sciences, Stanford University, Stanford, California 94305, USA 2Division of Geological and Planetary Sciences, California Institute of Technology, Pasadena, California 91126, USA

ABSTRACT via multiple stages of aggregation of silica Many pre–3.0 Ga cherty sequences include nanospheres and microspheres. Consistent <10-cm-thick layers of white-weathering or

The production of biogenic silica has with this hypothesis, Archean ocean chemis- translucent chert composed of nearly pure SiO2 dominated the marine silica cycle since early try would have favored particle aggregation (commonly >99 wt%), now microcrystalline Paleozoic time, drawing down the concentra- over gelling. Granule formation would have quartz (Lowe, 1999a). These white chert beds tion of dissolved silica in modern seawater been most favorable under conditions pro- are interbedded with compositionally contrast- to a few parts per million (ppm). Prior to moting rapid silica polymerization, including beds containing primary organic sedimen- the biological innovation of the fi rst silica ing high salinity and/or high concentrations tary grains (Lowe, 1999a; Walsh and Lowe, biomineralizing organisms in late Protero- of dissolved silica. Our observations suggest 1999), mixtures of siderite and fi ne aluminosili- zoic time, inputs of silica into Precambrian that granule sedimentation was often epi- cates, or hematite. These alternating lithologies seawater were balanced by strictly chemi- sodic, suggesting that granule formation may are known as banded black-and-white chert, cal silica and silicate precipitation processes, have also been episodic, perhaps linked to banded ferruginous (iron-bearing) chert, and although the mechanics of this abiotic ma- variations in these key parameters. banded iron formation, respectively (Fig. 1). rine silica cycle remain poorly understood. Stefurak et al. (2014) reported that many of the Cherty sedimentary rocks are abundant in INTRODUCTION white chert bands in these units are distinct sedi- Archean sequences, and many previous au- mentary beds composed entirely of sand-sized thors have suggested that primary precipita- The abundance and sedimentary style of chert subspherical primary silica granules (Fig. 1) and tion of amorphous silica could have occurred in early Archean rocks highlight a fundamental that silica granules were a common component in in Archean seawater. The recent discovery distinction between the modern and Archean other sedimentary layers as well, suggesting that that many pure chert layers in early Archean silica cycles. The early Archean silica cycle silica granules were a widespread grain type and rocks formed as sedimentary beds of sand- lacked the key components that dominate the represented a signifi cant mode of silica deposi- sized, subspherical silica granules has pro- modern silica cycle: continental weathering as tion in early Archean time. This observation aug- vided direct evidence for primary silica the dominant source and silica biomineraliza- mented the emerging view that the deposition of deposition. Here, we provide further sedi- tion as the dominant sink for dissolved silica in mud-, silt-, and sand-sized chemical sand grains mentological and geochemical analyses of marine waters (Maliva et al., 1989, 2005; Siever, composed of silica (Stefurak et al., 2014), iron early Archean silica granules in order to gain 1957, 1992; Treguer et al., 1995). Although the silicates (Rasmussen et al., 2013), and siderite a better understanding of the mechanisms of volume of continental crust during Archean time (Köhler et al., 2013) was involved in the deposi- granule formation. Silica granules are com- was much less than on modern Earth (Cawood tion of early Precambrian cherts and iron forma- mon components of sedimentary cherts from et al., 2012; Dhuime et al., 2012), Archean tions. This study presents additional petrographic a variety of depositional settings and water oceans had abundant silica sources in the form and geochemical analyses of silica granules with depths. The abundance and widespread dis- of weathering and alteration of mafi c and the goal of providing further insight into controls tribution of silica granules in Archean rocks ultramafi c rocks (Siever, 1992); mass balance on their formation and deposition. suggest that they represented a signifi cant requires that abundant sinks must also have been primary silica depositional mode and that present. Previous authors have suggested that GEOLOGIC BACKGROUND most formed by precipitation in the upper banded iron formations, diagenetic silicifi cation, part of the water column. The regular oc- and authigenic clay precipitation (Maliva, 2001; This study focuses on chert samples collected currence of silica granules as centimeter- Maliva et al., 1989, 2005; Siever, 1992; Stefurak in early Archean strata from both the Pilbara scale layers within banded chert alternating et al., 2015) were important sinks in the Archean block of Western Australia and the Barberton with layers of black or ferruginous chert silica cycle. It has also long been suggested that greenstone belt of South Africa. Samples from containing few granules indicates episodic primary chemical precipitation of amorphous the Pilbara block are from the Antarctic Creek granule sedimentation. Contrasting silicon silica phases also played a major role as a silica Member of the 3470 ± 1 Ma Apex Basalt (Byerly iso topic compositions of granules from dif- sink during Precambrian time (Lowe, 1999a; et al., 2002), which forms the middle part of the ferent depositional environments indicate Maliva et al., 2005; Siever, 1992), which is sup- Warrawoona Group (Fig. 2; Van Kranendonk that isotopic signatures were modifi ed during ported by the abundance of chert, especially in et al., 2002). The Antarctic Creek Member is a early diagenesis. Looking to modern siliceous pre–3.0 Ga Archean sequences. However, many 4–14-m-thick unit composed largely of silicifi ed sinters for insight into silica precipitation, we cherts also formed by diagenetic replacement felsic volcaniclastic sediments that also include suggest that silica granules may have formed and meta somatism of nonsiliceous primary an event bed of impact spherules and current- material (Lowe, 1999a), making primary silica deposited layers of silica granules (Stefurak †E-mail: [email protected] precipitates diffi cult to identify unambiguously. et al., 2014). The spherule bed has been corre-

GSA Bulletin; Month/Month 2015; v. 1xx; no. X/X; p. 1–18; doi: 10.1130/B31181.1; 10 ﬁ gures; 1 table; Data Repository item 2015105.; published online XX Month 2014.

ABC

2 cm 1 cm 500 μm D E F

1 cm 1 mm 1 mm GH I

2 cm 1 cm 500 μm

Figure 1. Layers of silica granules at different scales. (A–C) Black-and-white banded chert: (A) outcrop photo, Buck Reef Chert (locality Buck Reef Chert), (B) polished slab of white chert band showing faintly granular texture, upper Mendon Formation (locality BH-03), and (C) thin section of silica granules from a white chert band, upper Mendon Formation (locality SAF 521). (D–F) Lenticular granular layers within ferruginous shale, lower Mapepe Formation (locality SAF 183): (D) polished slab with lenses of two distinct granule types (arrows), both shown in thin section—a lower lens of mini- mally compacted, slightly ferruginous granules (E) and an upper lens of more compacted non-ferruginous granules (F). (G–I) Banded iron formation, lower Mapepe Formation: (G) outcrop photo showing chert bands (arrows) within jasper , (H) core photo highlighting ~1-cm-thick granular layer within banded iron formation (dashed outline) (sample SAF 649– 14), and (I) thin-section image of pure silica granules within slightly ferruginous matrix from the layer shown in H. See supplementary Table DR1 for additional stratigraphic information for samples shown and supplementary Figure DR1 for map of sample localities (see text footnote 1). lated with a similar 3470 ± 2 Ma unit, spherule Mendon Formations are part of the Onverwacht ferruginous chert at the base of the Kromberg bed S1, in the Barberton greenstone belt (Byerly Group, an 8–10-km-thick sequence of maﬁ c Formation (Lowe and Byerly, 1999). The age of et al., 2002; Glikson et al., 2004; Hickman and and ultramaﬁ c rocks with thin, interbedded the Buck Reef Chert is bracketed by two sets of Van Kranendonk, 2008; Lowe and Byerly, 1986). sedimentary chert units representing sediments U/Pb zircon ages of 3416 ± 5 Ma from a con- Samples from the Barberton greenstone belt deposited during periods of volcanic quiescence glomeratic unit in the lowest part of the Buck were collected from the Buck Reef Chert Mem- (Lowe and Byerly, 1999). The overlying Fig Reef Chert (Krüner et al., 1991) and 3334 ± ber of the Kromberg Formation, the upper Men- Tree Group is an ~180-m-thick sequence of 3 Ma from a felsic tuff from the uppermost don Formation, and the basal Mapepe Forma- immature terrigenous clastic rocks and felsic Kromberg Formation (Byerly et al., 1996). The tion (Fig. 2; Fig. DR11). Both the Kromberg and volcaniclastic units (Condie et al., 1970; Lowe basal unit of the Buck Reef Chert was deposited and Byerly, 1999). The Mapepe Formation is in a shallow, restricted setting, indicated by the the basal unit of the Fig Tree Group south of the common occurrence of wave ripples and evapo- 1GSA Data Repository item 2015105, Figures DR1-DR8 and Tables DR1 and DR2, is available at Inyoka fault (Lowe and Byerly, 1999). rite pseudomorphs after nahcolite (Lowe and http:// www .geosociety .org /pubs /ft2015 .htm or by re- The Buck Reef Chert is a 200–400-m-thick Fisher Worrell, 1999). The middle Buck Reef quest to editing@ geosociety .org. unit of banded black-and-white and banded Chert, composed of banded black-and-white

2 Geological Society of America Bulletin, v. 1XX, no. XX/XX Sedimentology and geochemistry of Archean silica granules

120° E 135° E Soanesville Group 3 Ma from the underlying Kromberg Formation MG (Byerly et al., 1996); intermediate ages mea- 10 Mapepe Formation ~3.26 Ga Mendon Formation ~3.33 Ga 15° S sured within the Mendon Formation of 3298 ± FTG 3 Ma (Byerly et al., 1996), 3287 ± 3 Ma (Decker,

M Pilbara 2013), and 3280 ± 9 Ma (Decker, 2013); and SSG Block Australia several ages from the base of the overlying Mapepe Formation ca. 3259 ± 3 Ma (Byerly Buck Reef Chert ~3.42 Ga 30° S et al., 1996; Decker, 2013; Krüner et al., 1991). 600 km Throughout the Barberton greenstone belt, Kelly Group the Mendon Formation is capped by a Kromberg Fm. Kromberg 5–70-m-thick unit of black chert, banded black- and-white chert, and banded ferruginous chert 25° S (Lowe and Byerly, 1999). The upper Mendon Formation chert unit is overlain with apparent conformity by the sediments of the Mapepe For- Barberton mation of the Fig Tree Group, which includes Greenstone Belt ferruginous shale, locally coarser clastic units, 30° S banded ferruginous chert, and hematitic banded South Africa iron formation (Lowe and Byerly, 1999). The Swaziland 200 km Mendon-Mapepe contact reﬂ ects a change from a relatively quiet, deep-water regime marked by 5 Hooggenoeg Fm. chemical, biogenic, and volcaniclastic sedimen- 20° E 30° E tation to a regime dominated by terrigenous and volcaniclastic sediments deposited in a range of Warrawoona Group Onverwacht Group depositional environments reﬂ ecting the initial stages of uplift and deformation in and around Antarctic Chert Member the greenstone belt (Lowe and Byerly, 1999; of the Apex Basalt ~3.47 Ga Lowe and Nocita, 1999).

METHODS

Sandstone Mafic volcaniclastic rocks Outcrops, polished hand samples, and pol- Conglomerate Felsic volcaniclastic rocks ished petrographic thin sections were used to Chert Basalt examine silica granule properties and occur- Komati Fm. rence at a variety of scales. Some lower Mapepe Various fine-grained Komatiite terrigenous rocks Formation samples were collected from the Felsic volcanic rocks BARB4 core of the 2011 International Conti- Granite nental Scientifi c Drilling Program (ICDP) in the Barberton greenstone belt. Granule sizes were measured using thin sec- 0 km tions of granules that were only slightly to mod- Figure 2. Location maps and general stratigraphic sections of the pre–3.0 Ga sequences erately compacted. For aggregates of granules, included in this study: the Barberton greenstone belt of South Africa (left; modifi ed after individual granules were measured rather than Lowe and Byerly, 1999) and the Pilbara block of Western Australia (right; modifi ed after the dimension of aggregates. Apparent grain Van Kranendonk, 2011). Abbreviations: M—Mendon Formation; FTG—Fig Tree Group; sizes were corrected using a Matlab script based MG—Moodies Group; SSG—Sulfur Springs Group. on the work of Schäfer and Teyssen (1987) to estimate the true grain sizes from observations of spherical segment diameters in thin section chert, was deposited in a relatively open shelf Previous studies have identifi ed in aggregate cuts through grains. Grains were approximated environment (Tice and Lowe, 2004). The upper at least fi ve episodes of volcanism within the as oblate spheres. Aspect ratios were measured Buck Reef Chert represents a deeper, quieter set- Mendon Formation, punctuated by the depo- and averaged in order to compare diameters of ting marked by fi nely laminated, banded ferru- sition of now-silicifi ed sedimentary layers. volume-equivalent spheres. ginous chert (Lowe and Byerly, 1999; Tice and However, most sections do not contain the full Gray-scale cathodoluminescence (CL) Lowe, 2004). The Buck Reef Chert is exposed sequence, potentially due to diachronous depo- images of areas of interest were mapped using a in the southernmost structural belt of the Bar- sition and/or local topographic highs imped- JEOL JSM-5600 scanning electron microscope berton greenstone belt and outcrops for over ing lateral continuity of volcanic units (Byerly, (SEM) fi tted with a Hamamatsu photo multi- 15 km along strike, indicating that deposition of 1999; Decker, 2013; Lowe and Byerly, 1999; plier tube (PMT), at the Stanford–U.S. Geo- banded black-and-white chert and banded ferru- Thompson Stiegler et al., 2010). The overall logical Survey Microanalysis Center at Stanford ginous chert was occurring continuously across age of the Mendon Formation is constrained by University (Stanford, California). Samples were relatively large areas on the Archean seafl oor. U/Pb zircon dating: a maximum age of 3334 ± coated with 15 nm of Au. Analysis settings were

Geological Society of America Bulletin, v. 1XX, no. XX/XX 3 Stefurak et al.

15 kV accelerating voltage, 38 µm spot size, and surements. The 28Si and 30Si ions were detected zones have lenticular geometries, thinning 39 mm working distance. using fast peak switching with one Faraday cup laterally into fi nely laminated black or gray Elements of interest (Ca, Mg, Fe, Al, and P (resistor value 10 × 10–11 ohms). The primary chert on a scale of ~10 m. The bases of granule or Ti) were mapped in carbon-coated (~14-nm- beam aperture size was ~400 μm in diameter. zones are often characterized by ferruginous thick) polished thin sections using a JEOL JXA- Critical illumination was used, with the diame- granules, while the tops are characterized by 8200 advanced electron probe micro-analyzer ter of the focused beam of ~15 μm. The focused iron-free granules. Dune-scale cross-stratifi ca- at the Division of Geological and Planetary beam was rastered across an area for the mea- tion (~20–30 cm scale) is often observed, but, Sciences Analytical Facility at the California surements, producing effective analyzed areas in contrast with the associated volcaniclastic Institute of Technology (Pasadena, California) of ~30 μm in width. The energy bandwidth sandstones, grading is absent from granule and the JEOL JXA-8230 SuperProbe electron was 45 eV. Count rates were typically between layers. Some granule layers have erosional probe micro-analyzer at the School of Earth 5 and 7 × 107 counts/s. Total analysis time for caps with evidence of scour and detrital lag Sciences Mineral Analysis Facility at Stanford each spot was 8 min, including 60 s presput- deposits containing translucent chert rip-up University (Stanford, California). Qualitative tering, fi eld and beam centering, and analyses clasts. Although silica granules occur within intensity maps without background correc- (20 cycles). Count times were 0.96 s for 28Si some volcaniclastic sandstones, volcaniclastic tions were collected, operating the electron and 8.00 s for 30Si. Each set of 12–16 sample grains are not observed in granule layers. probe in wavelength-dispersive X-ray spec- measurements was bracketed with 3–4 standard Overall, the Antarctic Creek Member where trometer (WDS) mode at 15 kV accelerating measurements. Each grain of NBS-28 quartz studied is characterized by current-deposited voltage, 100 nA (Caltech) or 10 nA (Stanford) sand accommodated 4–10 separate spot analy- sediments composed of volcaniclastic sand or beam current, and 100 ms (Caltech) or 10 ms ses, while the larger grains of Caltech Rose silica granules interbedded with thin beds of (Stanford) dwell time. Color CL images were Quartz typically accommodated >20 separate fi nely laminated chert (Fig. DR2 [see footnote mapped simultaneously with WDS elemental spot analyses. Isotope ratios are reported as 1]). The large-scale cross-stratifi cation and maps using the JEOL JXA-8230 at Stanford. per mil deviations from the NBS-28 quartz lenticularity of granule beds suggest that they The sensitivity of the CL of samples to beam standard (defi ned as 0‰) using delta notation: formed in a shallow-water setting characterized δ30 damage made collecting high-quality CL maps Si = 1000 × ([Rsample/RNBS-28] – 1). Typical by periods of moderate- to high-energy wave or challenging. 1σ external (standard-corrected) errors were current activity. The volcaniclastic sandstones For Si isotope analysis, select samples were ±0.30‰ determined by the variation of multi- represent an adjacent current-deposited facies; made into 1 in. (2.54 cm) circular polished ple spot measurements on known standards. the fi ning-upward character of some volcani- thin sections with standard grains embedded clastic units suggests that they could be chan- in epoxy as close to the analytical domains of RESULTS nel deposits. The fi nely laminated, fi ne-grained interest as possible, and then carefully polished chert represents lower-energy depositional envi- to obtain a fl at analytical surface for light and Depositional Environments ronments adjacent to these more current-active electron microscopy, electron microprobe, settings. The overall sequence is consistent with and secondary ion mass spectrometry (SIMS) Antarctic Creek Member deposition in a shallow, current-active setting analyses. The sample set of silica granule Silica granules in the Antarctic Creek Mem- indicating that granules were forming and being layers includes three samples of banded black- ber occur within a sequence of volcaniclas- deposited in very shallow water. and-white chert from the Mendon Formation, tic sandstone, laminated gray chert, granular one sample of ferruginous chert with lenses of chert, and translucent chert. The volcaniclastic Buck Reef Chert silica granules from the Mapepe Formation, sandstone is current-deposited, moderately- The sedimentology, lithofacies, and deposi- and one sample of banded iron formation from to well-sorted, and well-rounded sandstone, tional environments of the Buck Reef Chert have the Mapepe Formation. We used Caltech Rose often including accretionary lapilli and less been extensively studied by previous investiga- Quartz (δ30Si = –0.02‰ ± 0.10‰; Georg et al., often admixed silica granules. Units of vol- tors (Lowe and Byerly, 1999; Lowe and Fisher 2007) and NBS-28 pure quartz sand (δ30Si = caniclastic sandstone are tens of centimeters Worrell, 1999; Tice and Lowe, 2006). The 0‰) as standards. High-resolution silicon iso- to about a meter thick, with individual beds samples in this study are from the lower banded tope analyses were performed with the Cameca of 5–10 cm in thickness. Cross-laminations black-and-white chert facies of Tice and Lowe IMS-7f-GEO ion microprobe at the Caltech and large-scale cross-stratifi cation can be (2006; cf. fi g. 4 of Tice and Lowe, 2006). Black Microanalysis Center (California Institute of observed locally. Some beds fi ne upward and chert bands are composed of coarse sand-sized Technology, Pasadena, California) over three grade up into centimeter-scale units of fi ne- carbonaceous grains. Brecciation and/or plastic sessions. Samples were sputter-coated with grained laminated black, gray, or translucent deformation of white chert bands are common, 30 nm of Au for SIMS analysis. Samples were chert, which sometimes contain thin lenses of and coarse quartz cavity-fi lling cements often sputtered with an O– primary beam of 3–4 nA fi ne- to medium-grained siliciclastic sediment. occur beneath white chert bands or plates. The with 9 kV acceleration. Analysis spot diam- Similar fi ne-grained, fi nely laminated trans- lower banded black-and-white chert facies of eters were 30–40 µm, varying slightly between lucent, gray, and/or black chert also occurs the Buck Reef Chert overlies the basal evapo- sessions. Sample images in transmitted and without closely associated underlying vol- ritic unit, which includes pseudomorphs after refl ected light, in addition to elemental maps cani clastic sandstones and these are typically nahcolite and wave ripples (Lowe and Fisher acquired with the electron microprobe, were <10 cm thick. Silica granules occur in cur- Worrell, 1999). The lower banded black-and- used to avoid placing analytical spots in areas rent-deposited, well-sorted, typically massive white chert facies is in turn overlain by an with impurities (particularly iron oxides). A layers of coarse- to very coarse-grained gran- upper banded black-and-white chert facies (as mass resolving power of ~2400 (R = M/ΔM) ules. Sequences of granule beds are commonly defi ned in Tice and Lowe, 2006; this unit is was achieved, suffi cient for excluding contribu- ~1 m thick, but they can be as thin as 10 cm included in the lower Buck Reef Chert facies of tions from the 29Si1H hydride peak in 30Si mea- and up to several meters thick. Some granule Lowe and Byerly, 1999) and is characterized by

4 Geological Society of America Bulletin, v. 1XX, no. XX/XX Sedimentology and geochemistry of Archean silica granules

fi ne-grained, fi nely laminated, and slightly fer- Brecciation of white bands in banded black- Petrography and Sedimentology of ruginous banded chert with less common brec- and-white chert is less common than in the lower Silica Granules ciation of the white bands. The succeeding unit black-and-white banded chert facies of the Buck consists of the banded ferruginous chert facies, Reef Chert, likely refl ecting a lower-energy Despite some variability, several key proper- which is composed of intergrown microquartz paleoenvironment. Banded ferruginous chert of ties characterize granules from all depositional and siderite in unweathered subsurface samples the upper Mendon Formation is defi ned by alter- settings. Granules are everywhere sand-sized (Tice and Lowe, 2006). Locally, the Buck Reef nating white chert bands, and fi nely laminated subspherical grains with shapes ranging from Chert is capped by another unit of banded black- ferruginous chert bands. Ferruginous chert is nearly spherical to oblate spheroidal. The mean and-white chert up to 60 m thick (Lowe and often slightly rust-colored on the surface, prob- diameters for volume-equivalent spheres from Byerly, 1999). The Buck Reef Chert samples ably representing a sideritic component at depth. representative samples from a variety of deposi- included in this study are from the lower black- The upper Mendon Formation represents a tional settings range from 216 to 623 μm (fi ne- and-white chert unit. Our observations support relatively quiet depositional setting below storm to coarse-grained sand; Table 1). In addition, all the interpretation of Tice and Lowe (2006) that wave base and therefore a slightly deeper water granules appear to have been relatively soft and this facies represents a shelfal environment environment (>~200 m water depth) than the easily compacted. Granules are now composed above storm wave base at water depths that were lower black-and-white banded chert facies of of relatively pure microquartz, and the major- probably between 15 and 200 m. the Buck Reef Chert. ity of granules lack internal structures such as concentric laminations, irregular distributions Upper Mendon Formation Lower Mapepe Formation of trace impurities, relict textures refl ecting a Across the Barberton greenstone belt, the Within the study area in the central part of different primary mineralogy, or cracks formed uppermost Mendon Formation is composed of the Barberton greenstone belt, the Mapepe during dehydration and shrinkage of an initially ~20–75 m of black, banded black-and-white, Formation has an apparently conformable hydrated gel phase. Some iron-associated gran- and banded ferruginous chert (Lowe and Byerly, contact with cherts of the upper Mendon For- ules do display a crude structuring, with interi- 1999). The upper Mendon Formation black mation (Lowe and Byerly, 1999). This contact ors containing trace iron oxides and rims with chert and the black layers in banded black-and- is widely marked by a bed of spherules formed more abundant iron phases. white chert are dominated by fi nely plane-lami- through a large meteorite impact (Lowe and Some white bands within banded black- nated, silica-cemented deposits of fi ne-grained Byerly, 1986). The Mapepe Formation itself and-white chert of the Buck Reef Chert and carbonaceous matter and micaceous grains, shows a wide variety of facies deposited in the upper Mendon Formation are composed perhaps representing clays or altered volcani- a range of environments from shallow to exclusively of granules. These bands are gen- clastic particles. Many black cherts show a fi ne- deep water that vary from one structural belt erally 1–10 cm thick. Both matrix-supported grained granular texture on weathered surfaces to another within the central and southern (Fig. 1C) and grain-supported (Fig. 3B; Fig. and probably represent mixtures of fi ne ash and domains of the Barberton greenstone belt. DR1 [see footnote 1]) textures are observed carbonaceous matter. Many sections contain These complex facies and depositional set- within granule bands. Although structures like <2-cm-thick graded beds of fi ne ash, but these tings refl ect the initiation of uplift and defor- cross-laminations and grading are not com- generally lack current structures and prob- mation within the belt, providing sources of mon, they do occur in rare examples (Fig. DR1 ably represent air-fall material. Graded beds of clastic sediment and adjacent basins for its [see footnote 1]). While many white bands are accretionary lapilli occur in some sections. Cur- deposition and the accumulation of associated tabular and laterally continuous, others lens rent structures are represented mainly by small, deep-water, generally iron-rich chemical sedi- out laterally, and local brecciation is common widely spaced cross-laminations. Scour features ments. Samples from the Mapepe Formation (Fig. 4A). The bands of silica granules have include small, 2–5-cm-deep and 1–15-cm-wide included in this study are from ferruginous sharp contacts with the adjacent black chert scour pits, but large-scale cross-stratifi cation, lithofacies representing relatively quiet, deep- bands, which are composed primarily of silica scour, and evidence for string current activity water sedimentation—ferruginous cherts and containing detrital organic grains (Fig. 3B). and erosion are absent. banded iron formation. Compaction of granules is common within the

TABLE 1. RAW AND CORRECTED GRANULE DIAMETERS Equivalent μ,† σ,§ μ, σ, sphere Depositional Aspect raw# raw Φ,** corrected corrected diameter Sample name* Lithology environment n ratio (µm) (µm) raw (µm) (µm) (µm) Φ** Grain size AUS 36-1 Granular chert Subtidal 110 1.7 498 206 0.69 622 202 517 0.95 Coarse sand AUS 36-5 Granular chert Subtidal 50 1.9 639 268 0.37 772 288 623 0.68 Coarse sand BHR-02-13 Black chert Shelfal 300 1.7 204 118 1.97 256 137 216 2.21 Fine sand BHR-02-16 Black chert Shelfal 102 1.6 503 325 0.65 638 372 549 0.87 Coarse sand SAF 521-16 Banded black-and-white chert Shelfal 52 2.0 236 84 1.79 289 81 230 2.12 Fine sand SAF 183-2 Ferruginous chert Basinal 150 2.2 294 159 1.45 365 181 281 1.83 Medium sand SAF 649-14 Banded iron formation Basinal 300 2.2 302 166 1.37 386 187 299 1.74 Medium sand *See supplementary Table DR1 for more detailed stratigraphic and location information for these samples (text footnote 1). †μ—mean granule diameter. §σ—standard deviation of granule diameters. #“Raw” values refer to the grain diameters measured from thin sections prior to being corrected for the effects of the thin section providing random cuts through grains, as described in the methods section. Φ Φ ** = –log2 (grain diameter in mm); is the Krumbein phi scale, a standard logarithmic expression of grain size.

Geological Society of America Bulletin, v. 1XX, no. XX/XX 5 Stefurak et al.

Figure 3. Outcrop (A) and thin-section (B–G) images of representative occurrences AABB of silica granules associated with black- and-white banded chert and black chert. (A) Outcrop photo of white chert plate breccia, Buck Reef Chert (locality Buck Reef Chert). (B) Sharp contact between silica granule layer (white chert band) and adjacent detrital organic layer (black chert band), upper Mendon Formation (locality 2 cm 1 mm SAF 521). (C) Silica granules mixed with fi ner-grained detrital organic grains, lower Mapepe Formation (locality BHR-02). (D) Large silica granules associated with a C D variety of organic and volcanic grain types, upper Mendon Formation (locality SAF 521). (E) Silica granule intraclast within an intraclast breccia, upper Mendon Forma- tion (locality SAF 186). (F) Silica granules, intraclasts, detrital organic grains, and volcanic grains within an intraclast breccia bed from the same sample as E, upper 500 μm 500 μm Mendon Formation (locality SAF 186). (G) Organic grains and silica granules from a black chert, upper Mendon Formation EEFF (locality CQ-01). Silica granules (arrows) occur within the bed and at the base of the overlying graded bed. These granules are similar in size to and in some cases slightly coarser than the associated detrital organic grains. See supplementary Table DR1 for additional stratigraphic information for samples shown and supplementary Figure DR1 for map of sample localities (see text 500 μm 500 μm footnote 1). G white bands (maximum aspect ratios of ~9:1; Fig. 4), often obscuring primary textures. The surrounding matrix is microquartz, visually distinguishable due to trace organic matter or sider ite inclusions. In contrast with the common compaction of silica granules in white bands, carbonaceous grains in adjacent black chert bands typically show little to no compaction. This is seemingly in confl ict with the sequence of cementation implied by chert plate breccias, where white chert bands are often deformed as relatively rigid and impermeable plates while the surrounding carbonaceous sediment was still unlithifi ed. These observations suggest that, unlike silica granules, many organic grains were at least partially cemented very early, prior to deposition, making them more resistant to compaction. At the same time, in situ silica cementation occurred much earlier in pure layers of silica granules compared to 1 mm layers of carbonaceous sediment.

6 Geological Society of America Bulletin, v. 1XX, no. XX/XX Sedimentology and geochemistry of Archean silica granules

10 In the Antarctic Creek Member, coarse sand- mapped for either P or Ti, two poorly soluble sized, slightly compacted silica granules (aspect elements that might reveal possible precursor ratios in the range of 1.2:1–5:1, mean aspect mineralogy, but there was essentially no P or ratio 1.9:1) occur in 10-cm- to 1-m-thick, grain- Ti in any of the samples analyzed. Occurrence 8 supported and often cross-stratifi ed beds. Some of Al was limited to matrix material with trace granules from the Antarctic Creek Member are amounts of clay minerals, but these phases do composed purely of microquartz surrounded not occur in granule layers. by drusy quartz cements (Figs. 5A–5B). Other granules contain iron minerals closely associ- Cathodoluminescence 6 ated with the granules as rims of fi ne iron-oxide Silica granules and the adjacent matrix or grains (Figs. 5C–5D) and/or dispersed evenly cement are not distinguishable in gray-scale CL throughout granule interiors (Figs. 5E–5F). (Fig. DR6 [see footnote 1]), but some granules Lenticular centimeter-scale layers of granules are distinguishable using color CL (Fig. 6C). In 4 occur within ferruginous chert of the Mapepe the example shown in Figure 6, granules con- Formation, which consist of fi nely laminated taining fi ne trace inclusions of hematite are dis-

Granule aspect ratio sideritic chert (Figs. 1D–1F). Some of these tinguishable from the surrounding cement by granules are relatively pure silica (Figs. 1F and their brighter luminescence. Petrographically, 5G–5H), while others have trace inclusions of the granules appear to be compositionally vari- 2 micron-scale iron oxides (Fig. 1E). Most gran- able; similarly, CL brightness varies between ules are slightly to moderately compacted with granules. CL maps do not reveal any internal mean aspect ratios of ~2:1 (Fig. 4). The sur- textures or structuring (e.g., concentric layer- rounding matrix or cement is primarily micro- ing) apart from mirroring petrographic textures 0 quartz, although often containing ﬁ nely dis- reﬂ ecting slight variations in the abundance of

FC BIF persed hematite. The lenticular granule layers hematite inclusions. The peak emission wave-

BBWC could represent cross sections of starved ripples, length is ~640 nm (red). Apart from intensity, shallow although the examples examined in this study there were no distinct or systematic differences black chert do not display any cross-lamination. between CL spectra within the granules versus Silica granules also occur as centimeter-thick within the surrounding cement. intraclast breccia lenticular layers within banded iron formation in the lower Mapepe Formation (Figs. 1G–1I and Si Isotope Geochemistry Figure 4. Measurements of aspect ratios 5I). Excluding the granule lenses, the Mapepe Si isotopes are fractionated by many low- of silica granules associated with different banded iron formation is defi ned by fi nely temperature precipitation processes and can be litholo gies. Mean (circle), standard deviation laminated hematitic, sideritic, and cherty layers useful in differentiating different generations (line), minimum (diamond), and maximum with some interbedded layers of claystone or of silica and in tracking the cycling of silica (square) are given for each data set. Ellipses mudstone. The banded iron formation–associ- (Chakrabarti et al., 2012; Heck et al., 2011; next to the y-axis indicate dimensions for as- ated granules are composed of relatively pure Marin-Carbonne et al., 2012; Stefurak et al., pect ratios of 2, 4, 6, and 8. Abbreviations: silica, while the surrounding matrix/cement 2015; van den Boorn et al., 2007, 2010). The BBWC—banded black-and-white chert, contains trace hematite and siderite. Granules Si isotopic compositions of silica granules mea- FC—ferruginous chert, and BIF—banded are slightly to moderately compacted with mean sured in this study had a mean δ30Si of –0.28‰ iron formation. aspect ratios of ~2:1 (Fig. 4). As with granules and standard deviation of 1.42‰ (n = 338), in the ferruginous chert, the lenticular layers although analysis of variance (ANOVA) and in the banded iron formation may represent multiple comparison tests indicated that sam- Within nonbanded chert of the upper Men- starved ripple forms, but we have not observed ple means of granules associated with banded don Formation, granules also commonly occur cross-laminations that would confi rm this inter- black-and-white chert, banded iron formation, mixed in detrital beds with other sedimentary pretation, and some granule layers appear to be and ferruginous chert were signifi cantly dif- grain types, including: carbonaceous grains and matrix-supported (e.g., Fig. 1I). ferent (Fig. 7; Tables DR2 and DR3 [see foot- intraclasts of carbonaceous chert, volcanic grains note 1]). Banded black-and-white chert–associ- and intraclasts of volcaniclastic sediments, and Geochemistry ated silica granules had mean δ30Si of 0.77‰ and chert band intraclasts (Figs. 3C–3G). In occur- standard deviation of 0.45‰ (n = 189), while rences with organic grains, the mean aspect Electron Microprobe Maps banded iron formation–associated granules had ratios of silica granules are similar to other types Electron probe elemental maps confirm mean δ30Si of –0.70‰ and standard deviation of occurrences (2:1; Fig. 4), although in some petrographic observations showing that gran- of 0.67‰ (n = 68), and slightly ferruginous cases, particularly when the organic material ules are essentially pure SiO2 (Fig. 6; Figs. granules occurring in a lens within ferruginous occurs as dispersed, very fi ne-grained inclusions DR4 and DR5 [see footnote 1]). Some gran- shale had a mean δ30Si of –2.36‰ and standard rather than as clearly detrital grains, the silica ules associated with other iron-bearing facies deviation of 0.83‰ (n = 81). While neither the granules can be quite compacted (~10:1). Within contain trace inclusions of hematite (Fig. 6). banded iron formation–associated nor the ferru- intraclast breccia units, maximum compaction Iron-bearing carbonates, primarily siderite with ginous chert–associated silica granules had sig- observed (~2.5:1) is much less severe, likely minor ankerite and ankeritic dolomite, occur nifi cantly different mean δ30Si than the adjacent refl ecting differences in timing of silica cemen- within the matrix to the granules, but not within matrix or cement (Fig. 7), ANOVA and multiple tation between these lithofacies. the granules themselves. Granule samples were comparison tests indicated a small, but statisti-

Geological Society of America Bulletin, v. 1XX, no. XX/XX 7 Stefurak et al.

Figure 5. Thin-section images of representative occurrences of silica granules associ- A B ated with ferruginous chert, banded iron formation, and chert closely associated with D D ferruginous lithofacies. (A–B) Pure silica granules from the Antarctic Creek Member D D (locality AUS 36) in plane-polarized light Q Q (PPL; A) and cross-polarized light (XPL; B). 500 μm Q 500 μm Q Note drusy quartz cement (red D) in some PPL XPL pore spaces and the replacement of some granule interiors with coarse quartz (yellow Q). (C–D) Iron-oxide rims on silica granules C D similar to those shown in (A–B), in plane light (C) and cross-polarized light (D), from the Antarctic Creek Member (locality AUS 36). (E–F) Varied, somewhat ferruginous granules from the Antarctic Creek Member (locality AUS 36); image in F shows a close- up of the iron-rimmed grain from the lower- 200 μm PPL 200 μm XPL left corner of E. (G–H) Lens of compacted silica granules within ferruginous shale of the Mapepe Formation (locality SAF 183), E F in plane light (G) and cross-polarized light (H). (I) Thin lens of silica granules within Mapepe Formation banded iron formation (locality SAF 649). See supplementary Table DR1 for additional stratigraphic information for samples shown and supplementary Figure DR1 for map of sample localities (see text footnote 1). 500 μm 220000 μμmm GGH cally signifi cant difference in the sample means of banded black-and-white chert–associated granules (mean δ30Si = 0.77‰) and the associated microquartz matrix (mean δ30Si = 1.00‰; Table DR2 [see footnote 1]). The three banded black-and-white chert samples and the ferruginous chert sample were all collected from 1 mm 1 mm the same continuous stratigraphic section (Fig. PPL XPL DR7; Table DR1 [see footnote 1]; the banded black-and-white chert samples were collected within the upper 3 m of the Mendon Formation, I and the ferruginous chert sample was collected from 54 m into the lower Mapepe Formation. The granules analyzed in the three banded black-and-white chert samples have similar 1 mm δ30Si distributions and nonstatistically signifi - cant means (Fig. 7B), but they contrast strongly with the ferruginous chert sample, indicating in this case there is some relationship between associated silica granules in banded iron for- DISCUSSION observed granule Si isotopic composition and mation and ferruginous chert are broadly con- lithofacies and/or stratigraphic position. Silica sistent with many previous analyses of various Petrography and Environmental granules from banded black-and-white chert Archean banded iron formations fi nding them to Distribution have similar Si isotopic compositions to bulk have distinctly low δ30Si compositions (Fig. 7C; analyses of samples from van den Boorn et al. Andre et al., 2006; Chakrabarti et al., 2012; Del- The unstructured nature of iron-free gran- (2007, 2010) interpreted to have formed via vigne et al., 2012; Ding et al., 1996; Heck et al., ules and lack of relict textures resulting from silicifi cation by seawater silica, while the iron- 2011; Steinhoefel et al., 2009, 2010). diagenetic transformations differentiate silica

8 Geological Society of America Bulletin, v. 1XX, no. XX/XX Sedimentology and geochemistry of Archean silica granules

Figure 6. Petrographic textures in slightly additional iron phases could collect on the gran- ferruginous granules (A–B) correspond with A ule surfaces, perhaps as they were reworked by relative intensity maps of Si and Fe (using currents. wavelength-dispersive X-ray spectroscopy Silica granules occur in many depositional [WDS]) and color cathodoluminescence settings and water depths, from shallow subtidal (CL) (C). Thin-section images in plane-po- (Antarctic Creek Member) to shelfal in the Buck larized light (PPL; A) and cross-polarized Reef Chert to deep basinal (Mapepe banded iron light (XPL; B) show differences in composi- formation). They also occur in different types of tion between the granules and a contrast in deposits, from lenticular or tabular layers com- crystallinity among granules and between posed purely of granules (Fig. 1), to intraclast granules and cement. The CL data (C, red) breccia units with a variety of grain types (Figs. show that the granules have slightly higher 500 μm 3D–3F), to black chert composed mostly of CL intensity than the adjacent cement. See PPL organic grains with a minor component of silica supplementary Table DR1 for additional granules (Figs. 3C and 3G). The similarities of stratigraphic information for sample shown granule properties, including size, shape, sus- (see text footnote 1). B ceptibility to compaction, lack of internal structure, and composition, across a wide range of depositional environments also suggest that the granules share a common origin. The uniformly granules from other grain types like accretionary sand-sized, subspherical, well-rounded granules lapilli (Lowe, 1999b), impact spherules (Lowe are inconsistent with formation as rip-up clasts, and Byerly, 1986), volcanic grains (DiMarco which would be expected to display a range and Lowe, 1989), ooids, or other primary car- of grain sizes, shapes, and internal textures. bonate grains (Maliva et al., 2005). Although The occurrence of granules in shallow-water silica diagenesis, which involves several stages depositional environments indicates that gran- of dissolution and reprecipitation (Williams 500 μm ules probably formed within the shallow water et al., 1985), could theoretically have obscured XPL column. Within the deepest-water lithofacies some primary textures, it is unlikely that textures considered here, the banded iron formation of defi ned by insoluble impurities like organic mat- C the lower Mapepe Formation, granule layers ter, iron oxides, or clays, and elements like P, Ti, are relatively thin and irregularly spaced (Fig. and Al would have been completely obscured in 1H). This pattern suggests either that there were every granule occurrence. At the same time, we some processes preventing the delivery or pres- cannot completely rule out the possibility that ervation of granules in deep-water environments the granules initially contained some primary (e.g., obliteration during diagenesis) or that con- heterogeneities that have been homogenized ditions for granule formation were more favor- during diagenesis, such as structuring of the able in the shallow water column above shelfal original amorphous silica phase or inclusions environments, such that current transport was of more soluble minerals like carbonates. In required for granules to be delivered to more the latter case, the strongest evidence that the basinal environments. granules did not originate as carbonate grains is 500 μm One notable aspect of silica granules is their the lack of carbonate inclusions, ghosts, or even relationship to the phenomenon of banding: rare examples of only partially silicifi ed gran- many relatively pure, whitish to translucent ules such as those commonly observed in later chert bands or lenses within banded black-and- 14 12 424 Precambrian silicifi ed carbonates (e.g., Maliva white chert and banded ferruginous chert show et al., 2005). The ubiquitous homogeneity of granular textures and were therefore deposited iron-free silica granules throughout a range of as beds of silica granules. In theory, hydraulic

depositional and diagenetic settings suggests Si (counts) sorting could explain this juxtaposition of pure 0 (counts) Fe 0 37

that they originated as compositionally homo- CL, 553-717 nm granule layers with layers of contrasting com- geneous particles. position if both layers were current-deposited Some iron-associated granules do show a and there were density contrasts between gran- rudimentary structuring deﬁ ned by rims of 1972; Simonson, 1987, 2003). This indicates the ules and other grain types. These conditions iron oxides (Figs. 1E and 5C–5F). Some of coprecipitation of silica and iron phases, rather are inconsistent with a number of observations, these granules also include trace iron phases than alternating deposition of silica and iron which suggest that the two band types represent within the granule interiors (Figs. 1E and 5E). layers. The iron-rich rims, on the other hand, separate depositional events. The compositional Although the distribution of iron within gran- appear to have accumulated after silica deposi- contrasts between white chert bands composed ules is slightly heterogeneous, it does not tion ceased. These distribution patterns of iron of silica granules and the adjacent black chert deﬁ ne concentric laminations as in oolitic iron in and around silica granules are both consistent bands (Fig. 3B) or laminated ferruginous sedi- granules that occur in some granular iron for- with the hypothesis that granules formed rap- ment (Figs. 5G–5I) are sharp rather than grada- mations (Beukes and Klein, 1990; Dimroth idly, in some cases incorporating iron, likely as tional, and silica granules commonly occur as and Chauvel, 1973; Goode et al., 1983; Gross, iron oxyhydroxides or carbonates. Afterwards, a minor component within black chert bands,

Geological Society of America Bulletin, v. 1XX, no. XX/XX 9 Stefurak et al.

neither of which is consistent with hydraulic BBWC carbonaceous 1.6 A matrix (N=1; n=17) sorting of compositionally distinct layers as parts of the same depositional event. Evidence BBWC cht matrix of deposition by currents does not occur consis- (N=2; n=110) 1.2 tently throughout banded chert facies: Although BBWC granules the carbonaceous grains in some black chert (N=3; n=189) bands (e.g., within the lower facies of the Buck 0.8

Density BIF Fe matrix (N=1; n=92) Reef Chert; see Tice and Lowe, 2006) appear to have been deposited and/or reworked by cur- BIF granules (N=1; n=68) 0.4 rents, granular white chert bands can also be jux- FC cht cement (N=1; n=24) taposed with more fi ne-grained carbonaceous or FC Fe granules (N=1; n=81) ferruginous chert facies representing quiet sus- 0 pension settling. Hydraulic sorting alone cannot BBWC granules by sample: explain the occurrence of relatively pure granule 1.6 B layers juxtaposed with different primary sedi- SAF 521−14 (n=18); ment types and in different depositional envi- μ=0.6‰, σ=0.4‰ ronments. Instead, deposition of silica granules 1.2 SAF 521−15 (n=110); must have occurred only episodically. μ=0.7‰, σ=0.4‰ SAF 521−16 (n=60); 0.8 Composition of Granules and Associated Density μ=0.9‰, σ=0.5‰ Grains, Matrix, and Cements FC granules from same 0.4 stratigraphic section: The composition of other grains and matrix FC Fe granules (N=1; n=81) materials associated with silica granules depends μ=-2.4‰, σ=0.8‰ on the depositional setting. When cements are 0 present, they are always silica cements, support- Data from previous studies: ing the inference that the concentration of dis- 1.6 C solved silica in pore fl uids was high, either due to BIF, previous studies the occurrence of primary amorphous silica (e.g., (n=359) the granules themselves) or as a consequence of 1.2 van den Boorn et al. a high concentration of silica in seawater. Black S−cherts (n=22) cherts are commonly dominated by silicifi ed 0.8 Data from this study: organic grains and a microquartz matrix often Density BBWC granules containing trace carbonaceous material, while (N=3; n=189) deeper-water iron-rich sediments juxtapose len- 0.4 BIF granules (N=1; n=68) ticular granule layers with much fi ner-grained laminated ferruginous sediment composed FC Fe granules (N=1; n=81) of mixtures of hematite, siderite, phyllosili- 0 −4 −2 0 2 4 cates, and silica. Similarly, the matrix to silica δ SiNBS-28 (‰) granules within black chert is microquartz that commonly contains trace organic matter, while Figure 7. Kernel-smoothing probability density estimates of δ30Si data from silica granules the matrix within ferruginous settings contains and associated matrix and/or cement associated with different lithologies. (A) All data from fi ne hematite, siderite, and silica. The Antarctic this study, divided by lithofacies and texture. The samples included are from banded black- Creek granules are notable in the co-occurrence and-white chert (BBWC; black), banded iron formations (BIF), and ferruginous chert (FC). of iron-bearing minerals as trace constituents Spot analyses from within granules (solid lines) are differentiated from the associated matrix within granules. Most ferruginous chert–asso- or cement (dotted or dashed lines). The number of samples (N) and the total number of spot ciated and banded iron formation–associated analyses (n) are distinguished for each category. The three samples of the banded black-and- granules in Barberton samples contain little to white chert are from within the same section of the upper Mendon Formation. The ferru- no iron (Figs. 1F, 1I, and 5G–5I), except for ginous chert sample is from the lower Mapepe Formation higher in the same stratigraphic one isolated example, a lens of loosely packed, section as the upper Mendon samples. The banded iron formation sample is from the BARB4 grain-supported, well-sorted, subangular iron- core drilled in the Manzimnyama syncline. (B) Spot analyses from within silica granules of bearing granules (Fig. 1E) occurring within each of the banded black-and-white chert samples and the ferruginous chert sample collected laminated ferruginous chert (Fig. 1D) within from ~55 m higher in the same stratigraphic section. The means of the banded black-and- 2 cm of a more typical silica granule layer (Fig. white chert samples (listed in key to the right) are not statistically distinct, but the ferruginous 1F). These granules appear to be current-depos- chert sample is clearly different. (C) Comparison of granule data from this study (solid lines) ited and may represent a turbidite composed of with Si isotope data from previous studies (dashed lines) of cherts interpreted to record a sea- grains transported from shallower water, which water δ30Si signature (van den Boorn et al., 2007, 2010) and many measurements of Archean could explain the contrast in composition, sort- banded iron formations (Andre et al., 2006; Chakrabarti et al., 2012; Delvigne et al., 2012; ing, rounding, packing, and compaction with the Ding et al., 1996; Heck et al., 2011; Steinhoefel et al., 2009, 2010). adjacent granule lens.

10 Geological Society of America Bulletin, v. 1XX, no. XX/XX Sedimentology and geochemistry of Archean silica granules

The overall CL intensity of microquartz Formation (Fig. 1E) include some oblong grain son, 1985, 2003). The iron-free and iron-bearing from the cherts in this study is low, but color shapes, but these grains are loosely packed, silica granules detailed in this study may there- CL allows the recognition of some contrasts in and the axes of grain elongations are often not fore represent points on a spectrum of silica CL intensity that mirror petrographic textures aligned, indicating that many of these grains granule compositions, with shallow-water gran- (Fig. 6). The peak CL wavelength observed were irregular in shape when they were depos- ules apparently containing more iron than those using color CL (Fig. 6) is ~640 nm, an intrin- ited, rather than being deformed during burial. deposited in deeper-water environments. The sic wavelength observed in most quartz due to In contrast with the uniformly rounded nature main distinction, therefore, between the shal- nonbridging oxygen hole centers (Götze et al., of iron-free silica granules, some iron-bearing low-water ferruginous chert–associated silica 2001). Red emissions strongly dominate, and silica granules are angular to subangular. These granules in this study and granular iron forma- overall CL intensity is relatively low, both of observations both suggest that coprecipitation tions is the absence of associated iron-oxide or which are common features in low-grade meta- of silica and iron resulted in earlier lithifi cation iron-silicate granules. Since most granular iron morphic rocks (Boggs and Krinsley, 2006; of iron-bearing silica granules, making them formations are latest Archean or Proterozoic in Boggs et al., 2002). The samples do not display more likely to be broken or abraded by grain age (Bekker et al., 2014; Simonson, 2003), this spectral peaks indicative of structural substitu- collisions but less susceptible to postdeposi- distinction could be related to the rise of oxygen. tions of other elements like Fe3+ (705 nm) or tional compaction. Al (several possible peaks in the green or blue Si Isotope Geochemistry range; Götze et al., 2001). Relationship to Granular Iron Formation The contrasts between the δ30Si composi- Granule Compaction The silica granules described here bear some tions of silica granules and associated matrix resemblance to granules in granular iron forma- and/or cement from banded black-and-white Most iron-free granules appear to have been tion in that they are well-rounded, uniquely Pre- chert (three samples, 316 spot analyses), banded soft when deposited, based on the common cambrian chemical sand grains (Bekker et al., iron formation (one sample, 160 spot analy- occurrence of slight to severe postdepositional 2010, 2014; Klein, 2005; Simonson, 2003; ses), and ferruginous chert (one sample, 105 compaction. The degree of compaction varies Trendall, 2002). Some granular iron formations spot analyses) are striking (Fig. 7), although it between samples (Fig. 4). Silica granules within contain oolitic grains with concentric layering, is important to point out that this is a relatively white bands of banded black-and-white chert although non-oolitic iron granules are much small data set and therefore may not refl ect the include the most severely compacted examples, more common (Bekker et al., 2010; Beukes full range of δ30Si variability within each of even though independent textural evidence indi- and Klein, 1990; Dimroth and Chauvel, 1973; these lithofacies. At the same time, the data pre- cates that the white bands lithifi ed early (Knauth Goode et al., 1983; Gross, 1972; Simonson, sented here are broadly consistent with analyses and Lowe, 1978, 2003; Lowe, 1999a). The 1987, 2003). Although granular iron formations from previous Si isotopic studies. The granules lesser degree of compaction of granules in other are rich in SiO2 (~50 wt%), they also contain a from banded black-and-white chert are similar lithologies might indicate either that granules signifi cant amount of iron (>15 wt%, but com- in Si isotopic composition to previous analyses had already begun to lithify before being depos- monly ~30 wt%) in the form of iron oxides or of chert not associated with iron, interpreted to ited or that some local aspect of depositional iron silicates (Bekker et al., 2010, 2014; Klein, refl ect a seawater δ30Si signature (Fig. 7C; van conditions prevented the granules from being 2005; Simonson, 2003; Trendall, 2002). While den Boorn et al., 2007, 2010; Stefurak et al., as severely compacted. Alternately, organic-rich the silica granules described here co-occur with 2015). Similarly, the banded iron formation sediments may have been more fl uid-like during iron oxides or carbonates in some settings, they and ferruginous chert samples are isotopically compaction, with the result that granules mixed also commonly occur without associated iron depleted relative to the mean value of bulk sili- with black chert precursor sediments would minerals. Even when iron is associated with cate earth (δ30Si = –0.4‰), consistent with pre- have experienced lower compaction stresses the silica granules, it is only a trace constitu- vious analyses of Archean banded iron forma- than in grainstone-like granule layers. In the ent within the chert. The silica granules here tions (Fig. 7C; Andre et al., 2006; Chakrabarti case of intraclast breccias, the abundant lithifi ed are clearly distinct from the iron-oxide and et al., 2012; Delvigne et al., 2012; Ding et al., or partially lithifi ed intraclasts (Figs. 3E–3F), iron-silicate granules that occur in granular iron 1996; Heck et al., 2011; Steinhoefel et al., often including angular chert plate fragments, formations. 2009, 2010). There are two main explanations could have helped protect the relatively minor At the same time, many previous authors have for these distinctions: Either (1) the mecha- component of silica granules from the effects reported the occurrence of silica-rich granules nisms forming these different granule types had of compaction. In the case of deeper-water in addition to iron-rich granules within granu- drastically different Si isotopic fractionation ε δ30 δ30 ferruginous chert–associated or banded iron lar iron formations (Bekker et al., 2010; Beukes factors ( = Siprecip – Sifl uid) between fl uid formation–associated granule lenses, the less- and Klein, 1990; Dimroth and Chauvel, 1973; and precipitate, or (2) a substantial amount of extreme compaction likely resulted from some Goode et al., 1983; Gross, 1972; Simonson, silica from isotopically contrasting fl uids was combination of aging and strengthening of the 1987, 2003). These silica-rich granules appear contributed to each granule association during amorphous silica during transport and/or set- to commonly have trace inclusions of iron initial granule formation and/or diagenesis (e.g., tling, earlier cementation due to differences in phases, much like the iron-bearing granules we the granules associated with ferruginous chert diagenetic conditions, and a slower background observe in the Antarctic Creek Member (Figs. incorporated silica from fl uids of much lower depositional rate than that of banded black-and- 1E–1F) and the isolated lens in the Mapepe δ30Si than the banded black-and-white chert white chert, resulting in smaller forces exerted Formation (Fig. 1E). Furthermore, most granu- granules). by overlying sediment prior to cementation. lar iron formations are also current-reworked Previous authors have suggested that Si iso- Silica granules containing iron from the Ant- deposits of sand-sized chemical grains repre- topic fractionation associated with coprecipita- arctic Creek Member (Figs. 5E–5F) and an iso- senting relatively shallow shelfal environments tion of silica and iron oxyhydroxides (hematite lated lens in ferruginous chert of the Mapepe (Beukes and Klein, 1990; Gross, 1972; Simon- precursor) was responsible for the low δ30Si

Geological Society of America Bulletin, v. 1XX, no. XX/XX 11 Stefurak et al. values observed in banded iron formations have higher δ30Si (mean δ30Si = –2.0‰). This up to 100–200 ppm (the solubility of opal-A, (Chakrabarti et al., 2012; Fischer and Knoll, distinction indicates that, at least in this case, fl u- depending on temperature; Iler, 1979). In solu- 2009). In theory, this could explain why silica ids with a particularly light δ30Si value interacted tions supersaturated with respect to opal-A, sil- granules with ferruginous inclusions (Fig. 7A, with these granules during diagenesis. Given the ica monomers polymerize fi rst to dimers, then “FC Fe granules”) have lower δ30Si than the typical fl uid-precipitated Si fractionation factor cyclic polymers, and fi nally to compact spheri- iron-free silica granules (Fig. 7A, “BIF gran- of –1‰ to –2‰ (Basile-Doelsch et al., 2005; cal particles, which grow via Ostwald ripening ules”), i.e., that Si isotopic fractionation was Ding et al., 1996; Geilert et al., 2014), cements to stable colloidal particles 5–10 nm in diameter enhanced by coprecipitation with iron in the with δ30Si of –3.9‰ likely precipitated from a (above pH 7; Iler, 1979). former type but not the latter. However, the iron- fl uid with a δ30Si value in the range of –2‰ to Modern siliceous sinters provide a natural poor silica granules in the banded iron forma- –3‰, which is signifi cantly lighter than esti- system for studying silica precipitation that is tion sample still have signifi cantly lower δ30Si mates for Archean seawater, with δ30Si in the potentially analogous to Archean oceans. Some than the banded black-and-white chert granules, range of ~+2‰ to +3‰ (Abraham et al., 2011; modern and ancient sinter deposits contain sili- so differences in fractionation during primary van den Boorn et al., 2007, 2010), or analyses ceous oncoids 1–10 cm in diameter (Guidry and granule formation cannot explain the full range of modern hydrothermal fl uids, which yielded Chafetz, 2003; Jones and Renaut, 1997; Jones of variability observed in our data. a δ30Si of –0.3‰ (De La Rocha et al., 2000). et al., 1999, 2001; Lynne et al., 2008; Renaut The different granule types must therefore Si isotopic compositions this depleted require et al., 1996) or pisoids (Lowe and Braunstein, have incorporated silica from isotopically con- light δ30Si generated via one or more previous 2003; Walter, 1976; Walter et al., 1996). These trasting fl uids: higher δ30Si for the shallower- stages of precipitation/dissolution or adsorption/ large siliceous grains are not appropriate ana- water banded black-and-white chert–associated desorption. Both banded iron formation– and logues for silica granules: Apart from their size granules, intermediate δ30Si for the banded iron ferruginous chert–associated granules are adja- discrepancy, these grains are irregular in shape formation–associated granules, and surprisingly cent to layers that are rich in iron. Adsorption of and often contain relatively insoluble impuri- low δ30Si for the ferruginous chert–associated silica onto iron oxides is known to fractionate ties (e.g., organic matter, iron oxides, clays) that granules. Since banded iron formation repre- Si isotopes (Delstanche et al., 2009). Release distinguish the wrinkly or irregular concentric sents the deepest-water depositional environ- of this adsorbed silica with light δ30Si could internal laminations visually and geochemi- ment of these three lithofacies, there is no linear have contributed to the light isotopic signatures cally. Although some of these features could relationship between paleo–water depth and Si observed in our banded iron formation and fer- be overprinted during diagenesis, it is unlikely isotope composition of silica granules. The lack ruginous chert granule samples. The laminated that traces of a complex original structure, if of isotopic contrast between granules and the ferruginous chert adjacent to the lenses of gran- similarly defi ned by insoluble phases, would be adjacent matrix or cement in the iron-associated ules also contains phyllosilicates. Authigenic eliminated in all grains during lithifi cation. lithologies (Fig. 7A) furthermore indicates that clays can have fairly light δ30Si compositions Opal-A “microspheres” commonly occur in the cements precipitated from fl uids of similar (mean δ30Si = ~–1.4‰) due to preferential siliceous sinter deposits (Braunstein and Lowe, Si isotopic composition to the adjacent gran- incorporation of 28Si (De La Rocha et al., 2000; 2001; Jones and Renaut, 2007; Lowe and Braun- ules, perhaps due to exchange of silica during Ding et al., 1996; Douthitt , 1982; Georg et al., stein, 2003; Lynne and Campbell, 2004; Lynne the stages of dissolution and reprecipitation 2009; Opfergelt et al., 2010, 2012; Ziegler et al., et al., 2005, 2008; Rodgers et al., 2004; Smith characterizing phase transformations from opal- 2005), so any release of silica from these phases et al., 2001, 2003; Tobler et al., 2008). These A to opal-CT and fi nally to microquartz. These during diagenesis could also have contributed microspheres are generally <5 μm in diam- inferences fi t with the previous suggestion that to pore fl uids with low δ30Si compositions. The eter, with mean diameters of <2 μm (Jones there was a δ30Si chemocline in Precambrian behavior of Si isotope compositions of chert and Renaut, 2007; Rodgers et al., 2004). Some oceans (Chakrabarti et al., 2012), but they are during diagenesis has important implications microspheres show nanospheric substructure seemingly in confl ict with the conclusion that all for determining which silica phases record sea- (Rodgers et al., 2004; Smith et al., 2003), indi- granules originated in relatively shallow water water chemistry rather than diagenetic fl uids. cating that they formed via aggregation of nano- (Stefurak et al., 2014), which would suggest that The variability of δ30Si compositions of silica meter-scale colloidal particles. Once silica col- they should have shared similar initial isotopic granules observed here suggests that future loids reach a certain size (~5 nm), aggregation compositions. There are two main scenarios analyses of large silica granule data sets can pro- reduces the interfacial free energy more effec- that could have produced the observed patterns: vide important insights into seawater δ30Si and tively than individual particle growth, resulting (1) primary Si isotopic variability inherited the fractionation of Si isotopes under different in the formation of microspheres (Iler, 1979; from depth-related and/or lateral variations in diagenetic conditions. Smith et al., 2003). Smith et al. (2003, p. 577) seawater δ30Si, or (2) overprinting of a smaller suggested that the apparent maximum size limit range of primary granule δ30Si compositions by Processes of Granule Formation of microspheres in sinter deposits (~10 μm) is a distinct, locally derived δ30Si signatures during consequence of a similar principle: “Once silica diagenesis. Silicic acid molecules (H4SiO4) combine particles grow to a certain size, there is very little In reality, both primary and diagenetic pro- and grow into particles by polymerization. This change in energy content with surface area, and cesses probably contributed to the observed Si polymerization usually occurs via dehydration there is an optimum upper limit on the size to isotopic variability of silica granules, although reactions that convert two silanol groups into a which particles may grow.” a larger sample set is required to better evalu- siloxane bond (i.e., R-Si-O-H + H-O-Si-R ↔ This consistent upper bound on the diameters ate their relative contributions. In one example R-Si-O-Si-R + H2O). The aqueous chemistry of microspheres from modern siliceous sinters (Fig. DR8 [see footnote 1]), δ30Si does appear to of silica at low temperatures is dominated by is two orders of magnitude less than the aver- be sensitive to primary textural boundaries, with kinetics: Although a solution is technically satu- age size of the Archean silica granules described δ30 lower Si in cements and some granules (mean rated with respect to quartz at 6–10 ppm SiO2, here. Although there is some systematic size δ30Si = –3.9‰), while other adjacent granules monosilicic acid is relatively stable in solution variation observed in sinter microspheres,

12 Geological Society of America Bulletin, v. 1XX, no. XX/XX Sedimentology and geochemistry of Archean silica granules thought to be related to variations in environ- environmental factors, silica granules cannot be some pools (Fig. 8). Some granules clearly mental conditions (e.g., pH, salinity, tempera- easily explained as extraordinarily large opal-A formed around nuclei of β-quartz or other local ture, dissolved silica concentration; Rodgers microspheres. lithic material (Fig. 8B), but many do not have et al., 2004), these variations are still on the In the modern Akinomiya siliceous sinter nuclei identiﬁ able in thin section (Figs. 8A and order of a few microns. To the extent that the deposit (Japan), examined and sampled as part 8C), although this could be a result of random optimum microsphere size appears to be a func- of this study, sand-sized subspherical granules cuts through grain interiors rather than a true tion of surface free energy and a small range of of amorphous silica occur along the edges of absence of nuclei. The cortices of these granules

A B

1 mm 500 μm

C D

200 μm 100 μm

Figure 8. Thin-section images of subspherical, sand-sized amorphous silica granules from the modern Akinomiya siliceous sinter deposit in Japan. The sinter is located at 38°57′32″N, 140°33′26″E, and is 0–80 yr old (Nakanishi et al., 2003). (A) Loosely packed sinter granules. The blue coloration in the pore spaces between the samples is epoxy. (B) Sinter granule with a euhedral volcanic β-quartz grain as its nucleus and thin layers of isopachous silica cement along the grain edges. (C) Somewhat irregularly shaped granule. Silica microspheres of 30–50 μm in diameter can be distinguished along the edges of the granule, but the interior is structureless. (D) Distinguishable silica microspheres within a sinter granule. (E) Geopetal ﬁ ll composed of silt-sized silica microspheres, possibly distinguishable due to extremely thin coatings of impurities. (F) Close-up of layered microsphere ﬁ ll from E.

Geological Society of America Bulletin, v. 1XX, no. XX/XX 13 Stefurak et al. appear to have formed by aggregation of silica silica concentration are both important param- H spheres 10–50 μm in diameter; these spheres eters for the silica system and could well explain O O H Silicic acid are often clearly visible along granule edges the differences between modern siliceous sinters Si monomer (Fig. 8C) and, in limited cases, in incompletely and Archean marine environments. OOH cemented granule interiors (Fig. 8D). Silica Previous authors have suggested that the con- H microspheres also settled out of suspension as centration of silica in the Archean ocean was small geopetal fi lls (Fig. 8E–8F). Many of the at or above the saturation of amorphous silica microspheres observed here are larger than the (Maliva et al., 2005; Siever, 1992), meaning that apparent maximum microsphere size (~10 μm) silica would have existed in colloidal, rather H O observed in other sinter deposits. This could than monomeric form (Iler, 1979). Proposed H O be due to differences in environmental condi- models for Archean primary silica precipitation O Dimers and Si Si H tions or to an additional intermediate stage of must therefore take the polymerization behavior H O oligomers O O O aggregation that produced spherical particles of colloidal silica into consideration. Polymer- H H tens of microns in diameter. The Akinomiya ization is most rapid in the range of pH 4–8 sinter deposit is only ~80 yr old (Nakanishi (Fig. 10), and higher concentrations of silica et al., 2003), and most of the granules have translate to more rapid polymerization (Iler, Colloidal little discernible internal structure apart from 1979). Polymerization is inhibited from pH 8 to 5 nm the nucleus/cortex boundary—they lack con- 10 (Fig. 10) because of the increasing particle- nanospheres centric laminations analogous to ooids. Some particle repulsion caused by negative surface granules have fi nely layered rims of isopachous charges due to dissociation of H+ from surface silica cements that appear to have precipitated silanol groups (R-Si-O-H; Iler, 1979). When after the granules were deposited (Fig. 8B). The electrolyte concentrations are low, circum- 5 nm Aggregation Akinomiya sinter granules could potentially be neutral pH favors gelling, in which colloidal of nanospheres analogous to Archean silica granules, although particles link together to form a branched net- whether the modern sinter granules can form work of chains such that the liquid plus colloid without nuclei is still an open question. system does not change in silica concentration Polymerization and aggregation are com- but becomes increasingly viscous (Iler, 1979). mon processes in silica precipitation at multiple However, when cations are present (Na+, Ca2+, spatial scales, be it the initial polymerization etc.) in the pH range 6–10, the negative surface 2 μm Microspheres of nanometer-scale colloids, the aggregation charges on colloidal particles are mitigated, and of nanometer-scale colloidal particles to form coagulation is favored, in which colloidal parti- microspheres, or, as observed in the Akinomiya cles clump together into relatively dense aggre- sinter deposit, the aggregation of rather large gates (Baxter and Bryant, 1952; Iler, 1979). microspheres to form sinter granules. We there- Although the pH of Archean seawater is poorly fore hypothesize that Archean silica granules constrained, we can reasonably assume that it may have formed via multistage aggregation was not strongly acidic (pH < 4) or alkaline Aggregation analogous to processes observed in siliceous (pH > 10). In addition, the salinity of Archean of microspheres sinters, but including an additional stage or seawater was almost certainly higher than in the 2 μm stages of aggregation of microspheres to form modern alkali-chloride siliceous hot springs dis- macroscopic silica granules (Fig. 9). cussed here (Knauth, 2005). We can therefore These additional stages of aggregation are not expect that aggregation played a more signifi - ? common in modern siliceous hot springs—most cant role in Archean marine silica precipitation do not contain silica granules like those in the than in modern siliceous sinters, which perhaps Akinomiya sinter deposit. Several aspects of the explains why sand-sized silica granules are not 200 μm Granules depositional conditions make modern sinters common in these modern systems. imperfect analogues to Archean marine environments. For example, the maximum water depths Episodicity of Granule Formation Figure 9. Schematic outline of the hypoth- in siliceous hot springs are shallow, and currents and Deposition esis that silica granules formed via multiple are variable but generally unidirectional, both stages of aggregation. Silicic acid monomers of which could affect the length of time during One of the most striking characteristics of (H4SiO4) polymerize to form dimers and which particles of a particular size stay in sus- early Archean banded cherts is the regular alter- oligomers, which eventually grow to form pension. Many siliceous hot springs also have nation of layers of contrasting composition, one spherical nanospheres with characteristic brackish alkali-chloride waters with very high component of which is the relatively pure chert diameters on the order of 5 nm. Within mod- concentrations of silica (400–500 ppm; Jones layer of silica granules, indicating that deposi- ern siliceous sinters, colloidal nanospheres and Renaut, 2007; Rodgers et al., 2004); neither tion and probably formation of silica granules aggregate to form silica microspheres with of those conditions is a particularly good fi t for occurred episodically. This repetitive nature of characteristic diameters on the order of Archean oceans, which are thought to have been granule deposition and the local consistency in 2 μm. We propose that aggregation of micro- perhaps supersaline (Knauth, 2005) and only sat- thicknesses of silica granule layers and the inter- spheres could produce spherical granules urated to moderately supersaturated with respect vening black or ferruginous chert layers suggest with characteristic diameters similar to to amorphous silica (Siever, 1992). Salinity and that this cyclic deposition of granules and con- those observed in this study, ~200 μm.

14 Geological Society of America Bulletin, v. 1XX, no. XX/XX Sedimentology and geochemistry of Archean silica granules

coagulation brackish/saline scale would depend on the mixing time of shal- polymerization low granule-forming waters with deeper silica- mode gelling stable sols fresh water enriched waters and variations of the other critical environmental variables (e.g., seasonal/ annual variations in salinity). Granule layers in the deepest-water lithofacies examined here, the lower Mapepe Formation banded iron forma- silica sol no salt SiO2 tion, are thinner, less regular, and often farther stability 0.1 M NaCl dissolves apart stratigraphically than those observed in the shelfal Buck Reef Chert. As we have argued 0.2 M NaCl herein, this seems to suggest that granules either increasing negative surface charge formed exclusively over shelf environments or that granules formed in the water column over 46 81012 basinal environments did not make it to the pH seafl oor as effi ciently. Either way, it seems that not all granule formation episodes recorded in Figure 10. Schematic diagram illustrating the effects of pH and sa- shallower-water environments had equivalents linity on polymerization of colloidal silica. The y axis represents sta- in deeper-water facies. The depositional rates bility of the silica sol (the colloidal silica-water system), where higher in the deeper-water facies, banded ferruginous values correspond to silica being more stable in colloidal form, chert and banded iron formation, could there- while lower values correspond to systems favoring polymeri zation, fore still be relatively slower than shallower- resulting in coagulation of larger aggregate particles or gelling of water facies (cf. Tice and Lowe, 2004). branched networks. In the absence of electrolytes, particle repulsion due to negative particle surface inhibits polymerization above pH 7, Effects of Postdepositional Alteration and the colloidal system is relatively stable. Added cations, however, mitigate surface charges over a greater range in pH and promote The observations presented here suggest that + coagulation over gelling. For comparison, the concentration of Na silica granules have been affected physically and in the modern ocean is 0.47 M (DOE, 1994). Figure is modifi ed after chemically by their extended histories of dia- Iler (1979) and Williams and Crerar (1985). genesis and metamorphism. Some of the conse- quences of this postdepositional alteration relate to the appearance of the silica granules, most trasting sediments occurred with regular fre- variations are unlikely to have triggered granule notably including widespread compaction. A quency. Granule formation may therefore have formation. At slightly alkaline pH (8–10), how- more subtle but signifi cant change is the oblitera- been driven by periodic fl uctuations in key envi- ever, increases in salinity can strongly infl uence tion of any primary textural features with smaller ronmental parameters occurring over geologi- polymerization rates (Fig. 10). Finally, the rate length scales than the average size of microquartz cally short time scales. Temperature, salinity, of silica precipitation is proportional to the con- domains, ~30 μm. This makes it impossible to pH, and concentration of dissolved silica, the centration of dissolved silica, so higher silica directly test our granule formation hypothesis principal parameters infl uencing silica polymer- concentrations produce more rapid precipitation using traditional transmitted or refl ected light ization, could all have plausibly varied regularly (Rimstidt and Barnes, 1980) and polymerization petrography of Archean silica granules: Micro- on seasonal, annual, or multi-annual time scales. rates (Iler, 1979). Salinity and silica concentra- scopic substructures like those observed in the A rapid drop in temperature can cause a fl uid tion can therefore both affect polymerization modern granules have been destroyed. to become supersaturated with respect to silica, rates and could have fl uctuated over seasonal, Similarly, the data presented here suggest promoting silica precipitation (Williams and annual, and/or multi-annual time scales via vari- that the Si isotopic compositions of some silica Crerar, 1985). However, periodic fl uctuations in ations in evaporative concentration and/or fl uc- granules have likely been altered during dia- seawater temperature would not have been the tuations in inputs (cf. Bekker et al., 2014). We genesis. One likely explanation for the distinct most effective trigger for rapid granule forma- hypothesize that these two parameters are more Si isotopic compositions observed in silica tion because temperature changes would have likely to have triggered silica granule formation granules from different lithofacies is that they been gradual and small in magnitude, resulting than periodic changes in temperature or pH. resulted from the incorporation of contrasting in only small changes in the degree of saturation If the production and deposition of silica Si isotopic signatures from local pore fl uids, with respect to amorphous silica. granules were seasonal or annual, an impor- perhaps infl uenced by the mineralogical and As discussed already, there are complicated tant corollary is that the depositional rates of Si isotopic composition of adjacent sediments. relationships among pH, salinity, and stability banded chert must have been quite rapid—sev- In theory, Si isotopic compositions recorded by of the colloidal silica-water system (Fig. 10). In eral centimeters per year as refl ected by both primary silica precipitates should provide some general, polymerization of silica is most rapid the background sedimentation (black chert key insights into any spatial variations in the Si at moderate pH, and aggregation is favored bands or ferruginous chert bands) and the sil- isotopic composition of seawater. At this point, over gelling in the presence of cations, which ica granule event beds (white chert bands). We however, we cannot control for the magnitudes would have been abundant in Archean sea- could expect to observe a characteristic time and mechanisms of any changes in Si isotopic water (Knauth, 2005). Unless pH was fl uctuat- scale required to return silica concentration to composition; in other words, the initial Si iso- ing between circumneutral and either strongly its critical level after being diminished by the topic compositions of the granules analyzed acidic (pH < 4) or alkaline (pH > 10) values, pH previous granule-forming episode. This time here are unknown. Notably, if this interpreta-

Geological Society of America Bulletin, v. 1XX, no. XX/XX 15 Stefurak et al. tion is correct, this behavior would contrast with REFERENCES CITED Decker, N.B., 2013, High Resolution Tephra and U/Pb Chronol ogy of the 3.33–3.26 Ga Mendon Formation, observations that Si isotope compositions of Abraham, K., Hofmann, A., Foley, S.F., Cardinal, D., Harris, Barberton Greenstone Belt, South Africa [M.S. thesis]: Archean cherts were largely rock-buffered and C., Barth, M.G., and Andre, L., 2011, Coupled silicon- Baton Rouge, Louisiana, Louisiana State University, 70 p. relatively resistant to resetting during diagen- oxygen isotope fractionation traces Archaean silici- De La Rocha, C.L., Brzezinski, M.A., and DeNiro, M.J., fi cation: Earth and Planetary Science Letters, v. 301, 2000, A fi rst look at the distribution of the stable iso- esis and metamorphism (Marin-Carbonne et al., p. 222–230, doi: 10 .1016 /j .epsl .2010 .11 .002 . topes of silicon in natural waters: Geochimica et Cos- 2012; Stefurak et al., 2015), perhaps signaling Andre, L., Cardinal, D., Alleman, L.Y., and Moorbath, S., mochimica Acta, v. 64, no. 14, p. 2467–2477, doi: 10 differences in the timing and/or mechanisms of 2006, Silicon isotopes in ~3.8 Ga West Greenland .1016 /S0016 -7037 (00)00373 -2 . rocks as clues to the Eoarchean supracrustal Si cycle: Delstanche, S., Opfergelt, S., Cardinal, D., Elsass, F., Andre, silica diagenesis experienced by silica granules Earth and Planetary Science Letters, v. 245, p. 162– L., and Delvaux, B., 2009, Silicon isotopic fractionation compared to silica cements and other early dia- 173, doi: 10 .1016 /j .epsl .2006 .02 .046 . during adsorption of aqueous monosilicic acid onto iron Basile-Doelsch, I., Meunier, J.D., and Parron, C., 2005, oxide: Geochimica et Cosmochimica Acta, v. 73, no. 4, genetic chert-precursor silica phases. The most Another continental pool in the terrestrial silicon cycle: p. 923–934, doi: 10 .1016 /j .gca .2008 .11 .014 . productive future research directions may there- Nature, v. 433, p. 399–402, doi: 10 .1038 /nature03217 . Delvigne, C., Cardinal, D., Hofmann, A., and Andre, L., fore be experimental: evaluating the silica gran- Baxter, S., and Bryant, K.C., 1952, 577. Silica sols. Part II. 2012, Stratigraphic changes of Ge/Si, REE + Y and sil- Conditions of stability: Journal of the Chemical Soci- icon isotopes as insights into the deposition of a Meso- ule formation hypothesis presented here and ety, p. 3021–3024, doi: 10 .1039 /jr9520003021 . archaean banded iron formation: Earth and Planetary documenting the physical and chemical changes Bekker, A., Slack, J.F., Planavsky, N., Krapez, B., Hofmann, Science Letters, v. 355–356, p. 109–118, doi: 10 .1016 associated with diagenesis and metamorphism. A., Konhauser, K.O., and Rouxel, O.J., 2010, Iron /j .epsl .2012 .07 .035 . formation: The sedimentary product of a complex in- Department of Energy (DOE), 1994, Physical and thermo- terplay among mantle, tectonic, ocean, and biospheric dynamic data, in Dickson, A.G., and Goyet, C., eds., CONCLUSIONS processes: Economic Geology and the Bulletin of the Handbook of Methods for the Analysis of the Vari- Society of Economic Geologists, v. 105, p. 467–508, ous Parameters of the Carbon Dioxide System in Sea Silica granules were a common grain type doi: 10 .2113 /gsecongeo.105 .3 .467 . Water, Version 2: Oak Ridge, Tennessee, Oak Ridge in early Archean cherty sediments and appear Bekker, A., Planavsky, N.J., Krapež, B., Rasmussen, B., National Laboratory, Carbon Dioxide Information to have represented a signifi cant mode of silica Hofmann, A., Slack, J.F., Rouxel, O.J., and Konhauser, Analysis Center, 22 p. K.O., 2014, Iron formations: Their origins and impli- Dhuime, B., Hawkesworth, C.J., Cawood, P.A., and Storey, deposition in the Archean silica cycle. Their cations for ancient seawater chemistry, in Turekian, C.D., 2012, A change in the geodynamics of conti- sand size, occurrence in non-ferruginous set- H.D.H.K., ed., Treatise on Geochemistry (2nd ed.): nental growth 3 billion years ago: Science, v. 335, Oxford, Elsevier, p. 561–628. no. 6074, p. 1334–1336, doi:10 .1126 /science .1216066 . tings, nearly pure SiO2 composition, and uni- Beukes, N.J., and Klein, C., 1990, Geochemistry and sedi- DiMarco, M.J., and Lowe, D.R., 1989, Petrography and formly subspherical, well-rounded shapes dif- mentology of a facies transition—From microbanded provenance of silicifi ed Early Archaean volcaniclastic ferentiate silica granules from other associated to granular iron-formation—in the early Proterozoic sandstones, eastern Pilbara block, Western Australia: Transvaal Supergroup, South Africa: Precambrian Sedimentology, v. 36, p. 821–836, doi:10 .1111 /j .1365 detrital grain types or from granular iron forma- Research, v. 47, p. 99–139, doi:10 .1016 /0301 -9268 -3091 .1989 .tb01748 .x . tion. The sharp contacts between pure granule (90)90033 -M . Dimroth, E., and Chauvel, J.-J., 1973, Petrography of the layers and the adjacent detrital organic layers in Boggs, S., Jr., and Krinsley, D., 2006, Application of Cath- Sokoman Iron Formation in part of the Central Lab- odoluminescence Imaging to the Study of Sedimentary rador Trough, Quebec, Canada: Geological Society of banded black-and-white chert suggest that the Rocks: Cambridge, UK, Cambridge University Press, America Bulletin, v. 84, no. 1, p. 111–134, doi:10 .1130 two types of layers were formed from distinct 176 p. /0016-7606 (1973)84 <111: POTSIF>2 .0 .CO;2 . Boggs, S., Jr., Kwon, Y.-I., Goles, G.G., Rusk, B., Krinsley, Ding, T., Jiang, S., Wan, D., Li, Y., Li, J., Song, H., Liu, Z., depositional events and that production of silica D., and Seyedolali, A., 2002, Is quartz cathodolumines- and Yao, Z., 1996, Silicon Isotope Geochemistry: Bei- granules was episodic. We suggest that the gran- cence color a reliable provenance tool? A quantitative jing, Geological Publishing House, 125 p. ules were formed via multistage aggregation examination: Journal of Sedimentary Research, v. 72, Douthitt, C.B., 1982, The geochemistry of the stable isotopes no. 3, p. 408–415, doi: 10 .1306 /102501720408 . of silicon: Geochimica et Cosmochimica Acta, v. 46, of silica, in a manner somewhat analogous to Braunstein, D., and Lowe, D.R., 2001, Relationship between p. 1449–1458, doi: 10 .1016 /0016 -7037 (82)90278 -2 . silica precipitation observed in modern siliceous spring and geyser activity and the deposition and mor- Fischer, W.W., and Knoll, A.H., 2009, An iron shuttle for sinter deposits. In systems with near-neutral to phology of high temperature (>73°C) siliceous sinter, deepwater silica in Late Archean and early Paleo protero- Yellowstone National Park, Wyoming, U.S.A.: Journal zoic iron formation: Geological Society of America moderately alkaline pH and high concentrations of Sedimentary Research, v. 71, no. 5, p. 747–763, doi: Bulletin, v. 121, p. 222–235. of dissolved silica (saturated to supersaturated 10.1306 /2DC40965 -0E47 -11D7 -8643000102C1865D . Geilert, S., Vroon, P.Z., Roerdink, D.L., Van Cappellen, P., Byerly, G.R., 1999, Komatiites of the Mendon Formation: and Van Bergen, M.J., 2014, Silicon isotope fraction- with respect to amorphous silica), increased Late-stage ultramafi c volcanism in the Barberton ation during abiotic silica precipitation at low tem- salinity raises the rate of silica polymerization greenstone belt, in Lowe, D.R., and Byerly, G.R., eds., peratures: Inferences from fl ow-through experiments: and favors aggregation of silica particles over Geologic Evolution of the Barberton Greenstone Belt, Geochimica et Cosmochimica Acta, v. 142, p. 95–114, South Africa: Geological Society of America Special doi: 10 .1016 /j .gca .2014 .07 .003 . gelling. Variations in silica concentration and Paper 329, p. 189–211. Georg, R.B., Halliday, A.N., Schauble, E.A., and Reynolds, salinity could therefore have triggered episodic Byerly, G.R., Kroner, A., Lowe, D.R., Todt, W., and Walsh, B.C., 2007, Silicon in the Earth’s core: Nature, v. 447, production of silica granules. M.M., 1996, Prolonged magmatism and time con- p. 1102–1106, doi: 10 .1038 /nature05927 . straints for sediment deposition in the early Archean Georg, R.B., Zhu, C., Reynolds, B.C., and Halliday, A.N., ACKNOWLEDGMENTS Barberton greenstone belt: Evidence from the Up- 2009, Stable silicon isotopes of groundwater, feldspars, per Onverwacht and Fig Tree Groups: Precambrian and clay coatings in the Navajo Sandstone aquifer, We thank the following for their assistance: Y. Guan, Research, v. 78, p. 125–138, doi:10 .1016 /0301 -9268 Black Mesa, Arizona, USA: Geochimica et Cosmo- secondary ion mass spectrometry; E. Gottlieb and M. (95)00073 -9 . chimica Acta, v. 73, p. 2229–2241, doi:10 .1016 /j .gca Coble, sample preparation, scanning electron micros- Byerly, G.R., Lowe, D.R., Wooden, J.L., and Xie, X., .2009 .02 .005 . copy, and cathodoluminescence; C. Ma, electron probe 2002, An Archean impact layer from the Pilbara and Glikson, A.Y., Allen, C., and Vickers, J., 2004, Multiple (Caltech); and R. Jones, electron probe (Stanford). Kaapvaal cratons: Science, v. 297, no. 5585, p. 1325– 3.47-Ga-old asteroid impact fallout units, Pilbara cra- 1327, doi: 10 .1126 /science.1073934 . ton, Western Australia: Earth and Planetary Science Stefurak was supported by a National Science Foun- Cawood, P.A., Hawkesworth, C.J., and Dhuime, B., 2012, Letters, v. 7040, p. 1–14. dation Graduate Fellowship. Fischer was supported The continental record and the generation of conti- Goode, A.D.T., Hall, W.D.M., and Bunting, J.A., 1983, The by National Aeronautics and Space Administration nental crust: Geological Society of America Bulletin, Nabberu Basin of Western Australia, in Trendall, A.F., (NASA) Exobiology program (NNX09AM91G) and v. 125, no. 1–2, p. 14–32. and Morris, R.C., eds., Iron-Formation: Facts and Prob- the David and Lucile Packard Foundation. This study Chakrabarti, R., Knoll, A.H., Jacobsen, S.B., and Fischer, lems: Amsterdam, Netherlands, Elsevier, p. 295–323. was carried out using funds provided by the School of W.W., 2012, Si isotope variability in Proterozoic Götze, J., Plötze, M., and Habermann, D., 2001, Origin, Earth Sciences, Stanford University, to Lowe. We are cherts: Geochimica et Cosmochimica Acta, v. 91, spectral characteristics and practical applications of grateful to Sappi Forest Products, the Mpumalanga p. 187–201, doi: 10 .1016 /j .gca .2012 .05 .025 . the cathodoluminescence (CL) of quartz—A review: Condie, K.C., Macke, J.E., and Reimer, T.O., 1970, Petrol- Mineralogy and Petrology, v. 71, p. 225–250, doi: 10 Parks Board (J. Eksteen and L. Loocks), and Taurus ogy and geochemistry of early Precambrian graywackes .1007 /s007100170040 . Estates (C. Wille) for access to private properties. We from the Fig Tree Group, South Africa: Geological So- Gross, G.A., 1972, Primary features in cherty iron-forma- thank Nic Beukes, Andrey Bekker, and Assoc. Editor ciety of America Bulletin, v. 81, p. 2759–2776, doi: 10 tions: Sedimentary Geology, v. 7, no. 4, p. 241–261, Henning Dypvik for their helpful reviews. .1130 /0016 -7606 (1970)81 [2759: PAGOEP]2 .0 .CO;2 . doi: 10 .1016 /0037 -0738 (72)90024 -3 .

16 Geological Society of America Bulletin, v. 1XX, no. XX/XX Sedimentology and geochemistry of Archean silica granules

Guidry, S.A., and Chafetz, H.S., 2003, Anatomy of siliceous Park: The role of biological and abiological processes Renaut, R.W., Jones, B., and Rosen, M.R., 1996, Primary hot springs: Examples from Yellowstone National Park, in sedimentation: Canadian Journal of Earth Sciences, silica oncoids from Orakeikorako Hot Springs, North Wyoming, USA: Sedimentary Geology, v. 157, no. 1–2, v. 40, no. 11, p. 1611–1642, doi: 10 .1139 /e03 -066 . Island, New Zealand: Palaios, v. 11, no. 5, p. 446–458, p. 71–106, doi: 10 .1016 /S0037 -0738 (02)00195 -1 . Lowe, D.R., and Byerly, G.R., 1986, Early Archean silicate doi: 10 .2307 /3515212 . Heck, P.R., Huberty, J.M., Kita, N.T., Ushikubo, T., Kozdon, spherules of probable impact origin, South Africa and Rimstidt, J.D., and Barnes, H.L., 1980, The kinetics of R., and Valley, J.W., 2011, SIMS analyses of silicon Western Australia: Geology, v. 14, no. 1, p. 83–86, doi: silica-water reactions: Geochimica et Cosmochimica and oxygen isotope ratios for quartz from Archean and 10 .1130 /0091 -7613 (1986)14 <83: EASSOP>2.0 .CO;2 . Acta, v. 44, no. 11, p. 1683–1699, doi: 10 .1016 /0016 Paleoproterozoic banded iron formations: Geochimica Lowe, D.R., and Byerly, G.R., 1999, Stratigraphy of the -7037 (80)90220 -3 . et Cosmochimica Acta, v. 75, no. 20, p. 5879–5891, west-central part of the Barberton greenstone belt, Rodgers, K.A., Browne, P.R.L., Buddle, T.F., Cook, K.L., doi: 10 .1016 /j .gca .2011 .07 .023 . South Africa, in Lowe, D.R., and Byerly, G.R., eds., Greatrex, R.A., Hampton, W.A., Herdianita, N.R., Hickman, A., and Van Kranendonk, M. J., 2008, Marble Bar Geological Evolution of the Barberton Greenstone Holland, G.R., Lynne, B.Y., Martin, R., Newton, Z., Geological Sheet 2855: Geological Survey of Western Belt, South Africa: Geological Society of America Pastars, D., Sannazarro, K.L., and Teece, C.I.A., 2004, Australia, scale 1:100,000. Special Paper 329, p. 1–36. Silica phases in sinters and residues from geothermal Iler, R.K., 1979, The Chemistry of Silica: New York, John Lowe, D.R., and Fisher Worrell, G., 1999, Sedimentology, fi elds of New Zealand: Earth-Science Reviews, v. 66, Wiley & Sons, 866 p. mineralogy, and implications of silicifi ed evaporites in p. 1–61, doi: 10 .1016 /j .earscirev .2003 .10 .001 . Jones, B., and Renaut, R.W., 1997, Formation of silica the Kromberg Formation, Barberton greenstone belt, Schäfer, A., and Teyssen, T., 1987, Size, shape and orienta- oncoids around geysers and hot springs at El Tatio, South Africa, in Lowe, D.R., and Byerly, G.R., eds., tion of grains in sands and sandstones—Image analysis northern Chile: Sedimentology, v. 44, p. 287–304. Geologic Evolution of the Barberton Greenstone Belt, applied to rock thin-sections: Sedimentary Geology, Jones, B., and Renaut, R.W., 2007, Microstructural changes South Africa: Geological Society of America Special v. 52, p. 251–271, doi: 10 .1016 /0037 -0738 (87)90064 -9 . accompanying the opal-A to opal-CT transition: New Paper 329, p. 167–188. Siever, R., 1957, The silica budget in the sedimentary cycle: evidence from the siliceous sinters of Geysir, Haukada- Lowe, D.R., and Nocita, B.W., 1999, Foreland basin sedi- The American Mineralogist, v. 42, p. 821–841. lur, Iceland: Sedimentology, v. 54, p. 921–948, doi: 10 mentation in the Mapepe Formation, southern-facies Siever, R., 1992, The silica cycle in the Precambrian: Geo- .1111 /j .1365 -3091 .2007 .00866 .x . Fig Tree Group, in Lowe, D.R., and Byerly, G.R., eds., chimica et Cosmochimica Acta, v. 56, p. 3265–3272, Jones, B., Renaut, R.W., and Rosen, M.R., 1999, Actively Geologic Evolution of the Barberton Greenstone Belt, doi: 10 .1016 /0016 -7037 (92)90303 -Z . growing siliceous oncoids in the Waiotapu geothermal South Africa: Geological Society of America Special Simonson, B.M., 1985, Sedimentological constraints on the area, North Island, New Zealand: Journal of the Geo- Paper 329, p. 233–258. origins of Precambrian iron-formations: Geological logical Society of London, v. 156, no. 1, p. 89–103, Lynne, B.Y., and Campbell, K.A., 2004, Morphologic and Society of America Bulletin, v. 96, no. 2, p. 244–252, doi: 10 .1144 /gsjgs .156 .1 .0089 . mineralogic transitions from opal-A to opal-CT in low- doi: 10 .1130 /0016 -7606 (1985)96 <244: SCOTOO>2 .0 Jones, B., Rosen, M.R., and Renaut, R.W., 2001, “Geyser temperature siliceous sinter diagenesis, Taupo volcanic .CO;2 . Eggs” from Te Whakarewarewatangaoteopetauaawa- zone, New Zealand: Journal of Sedimentary Research, Simonson, B.M., 1987, Early silica cementation and subse- hiao, North Island, New Zealand: Journal of Sedimen- v. 74, no. 4, p. 561–579, doi: 10 .1306 /011704740561 . quent diagenesis in arenites from four early Protero- tary Research, v. 71, no. 1, p. 190–204, doi:10 .1306 Lynne, B.Y., Campbell, K.A., Moore, J.N., and Browne, zoic iron formations of North America: Journal of /060100710190 . P.R.L., 2005, Diagenesis of 1900-year-old siliceous Sedimentary Research, v. 57, no. 3, p. 494–511. Klein, C., 2005, Some Precambrian banded iron-formations sinter (opal-A to quartz) at Opal Mound, Roosevelt Simonson, B.M., 2003, Origin and evolution of large Pre- (BIFs) from around the world: Their age, geologic Hot Springs, Utah, U.S.A.: Sedimentary Geology, cambrian iron formations, in Chan, M.A., and Archer, setting, mineralogy, metamorphism, geochemistry, v. 179, no. 3–4, p. 249–278, doi:10 .1016 /j .sedgeo .2005 A.W., eds., Extreme Depositional Environments: Mega and origin: The American Mineralogist, v. 90, no. 10, .05 .012 . End Members in Geologic Time: Geological Society of p. 1473–1499, doi: 10 .2138 /am .2005 .1871 . Lynne, B.Y., Campbell, K.A., Moore, J., and Browne, America Special Paper 370, p. 231–244. Knauth, L.P., 2005, Temperature and salinity history of the P.R.L., 2008, Origin and evolution of the Steamboat Smith, B.Y., Rodgers, K.A., and Browne, P.R.L., 2001, Precambrian ocean: Implications for the course of mi- Springs siliceous sinter deposit, Nevada, U.S.A.: Sedi- Opal-A—The fi rst 100 days, in Proceedings of the 23rd crobial evolution: Palaeogeography, Palaeoclimatol- mentary Geology, v. 210, no. 3–4, p. 111–131, doi:10 New Zealand Geothermal Workshop: Auckland, New ogy, Palaeoecology, v. 219, p. 53–69, doi:10 .1016 /j .1016 /j .sedgeo .2008 .07 .006 . Zealand, Geothermal Institute, University of Auckland, .palaeo .2004 .10 .014 . Maliva, R.G., 2001, Silicifi cation in the Belt Supergroup p. 131–136. Knauth, L.P., and Lowe, D.R., 1978, Oxygen isotope geo- (Mesoproterozoic), Glacier National Park, Montana, Smith, B.Y., Turner, S.J., and Rodgers, K.A., 2003, Opal-A chemistry of cherts from the Onverwacht Group (3.4 USA: Sedimentology, v. 48, p. 887–896, doi:10 .1046 and associated microbes from Wairakei, New Zealand: billion years), Transvaal, South Africa, with implica- /j .1365 -3091 .2001 .00399 .x . The fi rst 300 days: Mineralogical Magazine, v. 67, tions for secular variations in the isotopic composition Maliva, R.G., Knoll, A.H., and Siever, R., 1989, Secular no. 3, p. 563–579, doi: 10 .1180 /0026461036730118 . of cherts: Earth and Planetary Science Letters, v. 41, change in chert distribution: A refl ection of evolving Stefurak, E.J.T., Lowe, D.R., Zentner, D., and Fischer, W.W., p. 209–222, doi: 10 .1016 /0012 -821X (78)90011 -0 . biological participation in the silica cycle: Palaios, v. 4, 2014, Primary silica granules—A new mode of early Knauth, L.P., and Lowe, D.R., 2003, High Archean climatic no. 6, p. 519–532, doi: 10 .2307 /3514743 . Archean sedimentation: Geology, v. 42, no. 4, p. 283– temperature inferred from oxygen isotope geochem- Maliva, R.G., Knoll, A.H., and Simonson, B.M., 2005, Sec- 286, doi: 10 .1130 /G35187 .1 . istry of cherts in the 3.5 Ga Swaziland Supergroup, ular change in the Precambrian silica cycle: Insights Stefurak, E.J.T., Fischer, W.W., and Lowe, D.R., 2015, Tex- South Africa: Geological Society of America Bulle- from chert petrology: Geological Society of America ture-specifi c Si isotope variations in Barberton green- tin, v. 115, no. 5, p. 566–580, doi:10 .1130 /0016 -7606 Bulletin, v. 117, no. 7–8, p. 835–845, doi:10 .1130 stone belt cherts record low temperature fractionations (2003)115 <0566: HACTIF>2 .0 .CO;2 . /B25555 .1 . in early Archean seawater: Geochimica et Cosmo- Köhler, I., Konhauser, K.O., Papineau, D., Bekker, A., and Marin-Carbonne, J., Chaussidon, M., and Robert, F., 2012, chimica Acta, v. 150, p. 26–52. Kappler, A., 2013, Biological carbon precursor to dia- Micrometer-scale chemical and isotopic criteria (O and Steinhoefel, G., Horn, I., and von Blanckenburg, F., 2009, genetic siderite with spherical structures in iron forma- Si) on the origin and history of Precambrian cherts: Im- Micro-scale tracing of Fe and Si isotope signatures in tions: Nature Communications, v. 4, p. 1741, doi:10 plications for paleo-temperature reconstructions: Geo- banded iron formation using femtosecond laser abla- .1038 /ncomms2770 . chimica et Cosmochimica Acta, v. 92, p. 129–147, doi: tion: Geochimica et Cosmochimica Acta, v. 73, no. 18, Krüner, A., Byerly, G.R., and Lowe, D.R., 1991, Chronol- 10 .1016 /j .gca .2012 .05 .040 . p. 5343–5360, doi: 10 .1016 /j .gca .2009 .05 .037 . ogy of Early Archaean granite-greenstone evolution Nakanishi, T., Izawa, E., and Watanabe, K., 2003, Miner- Steinhoefel, G., von Blanckenburg, F., Horn, I., Konhauser, in the Barberton Mountain Land, South Africa, based alogical and geochemical characteristics of siliceous K.O., Beukes, N.J., and Gutzmer, J., 2010, Decipher- on precise dating by single zircon evaporation: Earth sinter in Japan, in Proceedings of the 25th New Zea- ing formation processes of banded iron formations and Planetary Science Letters, v. 103, p. 41–54, doi:10 land Geothermal Workshop, University of Aukland, from the Transvaal and the Hamersley successions by .1016 /0012 -821X (91)90148 -B . p. 155–160. combined Si and Fe isotope analysis using UV femto- Lowe, D.R., 1999a, Petrology and sedimentology of cherts Opfergelt, S., Cardinal, D., Andre, L., Delvigne, C., second laser ablation: Geochimica et Cosmochimica and related silicifi ed sedimentary rocks in the Swa- Bremond, L., and Delvaux, B., 2010, Variations of δ30Si Acta, v. 74, p. 2677–2696, doi: 10 .1016 /j .gca .2010 .01 ziland Supergroup, in Lowe, D.R., and Byerly, G.R., and Ge/Si with weathering and biogenic input in tropi- .028 . eds., Geologic Evolution of the Barberton Greenstone cal basaltic ash soils under monoculture: Geochimica Thompson Stiegler, M., Lowe, D.R., and Byerly, G.R., Belt, South Africa: Geological Society of America et Cosmochimica Acta, v. 74, no. 1, p. 225–240, doi: 10 2010, The petrogenesis of volcaniclastic komatiites in Special Paper 329, p. 83–114. .1016/j .gca .2009 .09 .025 . the Barberton greenstone belt, South Africa: A textural Lowe, D.R., 1999b, Shallow-water sedimentation of ac- Opfergelt, S., Georg, R.B., Delvaux, B., Cabidoche, Y.-M., and geochemical study: Journal of Petrology, v. 51, cretionary lapilli–bearing strata of the Msauli Chert: Burton, K.W., and Halliday, A.N., 2012, Silicon iso- p. 947–972, doi: 10 .1093 /petrology /egq008 . Evidence of explosive hydromagmatic komatiitic topes and the tracing of desilication in volcanic soil Tice, M.M., and Lowe, D.R., 2004, Photosynthetic micro- volcanism, in Lowe, D.R., and Byerly, G.R., eds., weathering sequences, Guadeloupe: Chemical Geol- bial mats in the 3,416-Myr-old ocean: Nature, v. 431, Geologic Evolution of the Barberton Greenstone Belt, ogy, v. 326–327, p. 113–122, doi:10 .1016 /j .chemgeo p. 549–552, doi: 10 .1038 /nature02888 . South Africa: Geological Society of America Special .2012 .07 .032 . Tice, M.M., and Lowe, D.R., 2006, The origin of carbona- Paper 329, p. 213–232. Rasmussen, B., Meier, D.B., Krapez, B., and Muhling, J.R., ceous matter in pre–3.0 Ga greenstone terrains: A re- Lowe, D.R., and Braunstein, D., 2003, Microstructure of 2013, Iron silicate microgranules as precursor sedi- view and new evidence from the 3.42 Ga Buck Reef high-temperature (>73°C) siliceous sinter deposited ments to 2.5-billion-year-old banded iron formations: Chert: Earth-Science Reviews, v. 76, p. 259–300, doi: around hot springs and geysers, Yellowstone National Geology, v. 41, p. 435–438, doi: 10 .1130 /G33828 .1 . 10 .1016 /j .earscirev .2006 .03 .003 .

Geological Society of America Bulletin, v. 1XX, no. XX/XX 17 Stefurak et al.

Tobler, D.J., Stefansson, A., and Benning, L.G., 2008, In- ronments: Geochimica et Cosmochimica Acta, v. 74, Walter, M.R., Desmarais, D., Farmer, J.T., and Hinman, situ grown silica sinters in Icelandic geothermal areas: p. 1077–1103, doi: 10 .1016 /j .gca .2009 .09 .009 . N.W., 1996, Lithofacies and biofacies of mid-Paleo- Geobiology, v. 6, p. 481–502, doi: 10 .1111 /j .1472 -4669 Van Kranendonk, M.J., 2011, Pilbara craton, in Gargaud, M., zoic thermal spring deposits in the Drummond Basin, .2008 .00179 .x . Amils, R., Quintanilla, J.C., Cleaves, H.J., II, Irvine, Queensland, Australia: Palaios, v. 11, no. 6, p. 497– Treguer, P., Nelson, D.M., Van Bennekom, A.J., DeMaster, W.M., Pinti, D.L., and Viso, M., eds., Encyclopedia of 518, doi: 10 .2307 /3515187 . D.J., Leynaert, A., and Queguiner, B., 1995, The silica Astrobiology: Berlin, Springer-Verlag, p. 1257–1260. Williams, L.A., and Crerar, D.A., 1985, Silica diagenesis: II. balance in the world ocean: A reestimate: Science, Van Kranendonk, M.J., Hickman, A.H., Smithies, R.H., Nel- General mechanisms: Journal of Sedimentary Petrol- v. 268, p. 375–379, doi: 10 .1126 /science .268 .5209 .375 . son, D.M., and Pike, G., 2002, Geology and tectonic ogy, v. 55, no. 3, p. 312–321. Trendall, A.F., 2002, The signiﬁ cance of iron-formation in evolution of the Archean North Pilbara terrain, Pilbara Williams, L.A., Parks, G.A., and Crerar, D.A., 1985, Silica the Precambrian stratigraphic record, in Altermann, craton, Western Australia: Economic Geology and the diagenesis: I. Solubility controls: Journal of Sedimen- W., and Corcoran, P.L., eds., Precambrian Sedimentary Bulletin of the Society of Economic Geologists, v. 97, tary Petrology, v. 55, no. 3, p. 301–311. Environments: A Modern Approach to Ancient Depo- p. 695–732. Ziegler, K., Chadwick, O.A., White, A.F., and Brzezinski, sitional Systems: International Association of Sedi- Walsh, M.M., and Lowe, D.R., 1999, Modes of accumula- M.A., 2005, δ30Si systematics in a granitic saprolite, mentologists Special Publication 33, p. 33–66. tion of carbonaceous matter in the early Archean: A Puerto Rico: Geology, v. 33, no. 10, p. 817–820, doi: van den Boorn, S.H.J.M., van Bergen, M.J., Nijman, W., and petrographic and geochemical study of the carbona- 10 .1130 /G21707 .1 . Vroon, P.Z., 2007, Dual role of seawater and hydro- ceous cherts of the Swaziland Supergroup, in Lowe, thermal ﬂ uids in early Archean chert formation: Evi- D.R., and Byerly, G.R., eds., Geologic Evolution of the SCIENCE EDITOR: CHRISTIAN KOEBERL dence from silicon isotopes: Geology, v. 35, no. 10, Barberton Greenstone Belt, South Africa: Geological ASSOCIATE EDITOR: HENNING DYPVIK p. 939–942, doi: 10 .1130 /G24096A .1 . Society of America Special Paper 329, p. 115–132. MANUSCRIPT RECEIVED 1 AUGUST 2014 van den Boorn, S.H.J.M., van Bergen, M.J., Vroon, P.Z., Walter, M.R., 1976, Geyserites of Yellowstone National REVISED MANUSCRIPT RECEIVED 29 DECEMBER 2014 de Vries, S.T., and Nijman, W., 2010, Silicon isotopic Park: An example of abiogenic “stromatolites,” in MANUSCRIPT ACCEPTED 5 FEBRUARY 2015 and trace element constraints on the origin of ~3.5 Ga Walter, M.R., ed., Stromatolites: Amsterdam, Nether- cherts: Implications for Early Archaean marine envi- lands, Elsevier, p. 87–112. Printed in the USA

18 Geological Society of America Bulletin, v. 1XX, no. XX/XX