H. Wayne Nesbitt

Unraveling the effects of potassium metasomatism in sedimentary rocks and paleosols, with implications for paleoweathering conditions and provenance

Lutites are commonly metasomatized during diagenesis, but the analysis presented here accounts for most postdepositional change. Potassium metasomatism is particularly common, and typically involves the conversion of kaolin (residual weathering product) to illite by reaction with K + -bearing pore waters. Sandstones also undergo K metasomatism, which involves the replacement of plagioclase by potassium feldspar. These changes can be identified petrographically and are quantitatively accounted for by techniques discussed herein. Bulk chemical analyses and ternary diagrams are used to determine the amount of K addition, premetasomatized sediment composition, and composition of provenance areas. The premetasomatized mineralogy of paleosols can be compared with the mineralogy of recent soil profiles and thus, climate and topographic conditions determined for past weathering events. Some weathering indices lead to erroneous conclusions because, by excluding K 2 O from consideration, correction cannot be made for metasomatic effects.

Chemical processes affecting alkalis and alkaline earths during continental weathering

Abstract The alkali and alkaline earth concentrations in the Toorongo Granodiorite weathering profile are controlled by two competing processes; leaching of cations from primary minerals during their degradation to clays; fixation, by exchange and adsorption, of the same cations onto the secondary clay minerals. Degradation and leaching dominate the early weathering stages whereas during the advanced stages, exchange and adsorption onto clays are of most influence. The alkali and alkaline earth compositional changes in the Toorongo Granodiorite weathering profile are typical of changes occurring during weathering of the continents, consequently the following generalizations apply to continental weathering. Ca, Sr and Na are most rapidly and most strongly removed (as dissolved species) during weathering of fresh continental rocks. Although large quantities of Mg are transported to the marine environment as dissolved species, appreciable amounts remain (fixed in secondary clay minerals) at the weathering site to be removed during mass wasting of continental weathering profiles. Large quantities of Rb, Cs and Ba, fixed in continental weathering profiles by exchange and adsorption onto secondary clays, are transported from the continents only during mass wasting of the continents.

Weathering of granodioritic crust, long-term storage of elements in weathering profiles, and petrogenesis of siliciclastic sediments

Abstract The bulk composition and mineralogy of the Toorongo Granodiorite, Australia, are similar to average upper continental crust (AUCC). Weathering characteristics of the Toorongo profile consequently provide insight into large-scale chemical weathering of the upper crust. In situ weathered materials of the profile do not reflect parent granodiorite composition in quartz-plagioclase-K-feldspar (Q-P-K) or in quartz-feldspar-rock fragment (Q-F-L) compositional space. Intensive in situ weathering precludes sands, derived from mature weathering profiles through erosion, from reflecting their provenance. Where intensive chemical weathering has occurred, clay minerals and oxyhydroxides of the profile, and by inference muds derived therefrom, contain much more chemical information about provenance than do associated sands. Actinides, rare earth elements (REEs), many transition metals, and metalloids have accumulated in deep parts of the weathering profile at concentrations much greater than observed in the fresh granodiorite. Mass balance considerations require the bulk of these elements to have been derived from previously weathered, and now eroded, granodiorite. These elements were, and are, continually cycled from the intensely weathered uppermost soil zone to deeper, less weathered, zones of the profile where they accumulate. The profile therefore represents a large, continental elemental storage reservoir, the storage capacity of which has increased over time. Wherever erosion is sufficiently slow and chemical weathering sufficiently rapid, mature weathering profiles may become large, long-term storage reservoirs for actinides, REEs, and many other elements. The total REE contents of extremely weathered soil material are somewhat less than in the parent granodiorite, but they are enriched twofold to threefold in the zone of intermediate weathering relative to parent. Similar variations in total REEs are observed in some muds when normalized to their source (AUCC). These differences are attributed to a combination of chemical weathering and selective mass wasting of profiles. Homogenization of detritus in large sedimentary basins, however, produces muds with REE patterns and total REE contents similar to source (AUCC). Nd/Sm ratios are not influenced by chemical weathering, although both elements are mobilized by weathering and become enriched by over 200% relative to parent rock. Constancy of Nd/Sm in the profile indicates that Nd-Sm model ages derived from soils and sediments are not affected by chemical weathering. The least mobile trace elements of the profile are Sc, Cu, Nb, and Ta, but others are more mobile. Thorium, for example, is mobilized during weathering of the Toorongo Granodiorite and displays a twofold increase in the profile, as does the Th/Sc ratio. The ratio, however, varies by more than a hundredfold in major rock types so that Th/Sc (and other ratios) provides valuable information about provenance, although sensitivity is diminished somewhat by the effects of chemical weathering.

Quartz and Feldspar Stability, Steady and Non‐Steady‐State Weathering, and Petrogenesis of Siliciclastic Sands and Muds

The kinetic and thermodynamic properties of quartz, plagioclase, and K‐feldspars, which constitute 70 to 80% of the upper crust, provide a framework for prediction of mineralogical and chemical changes involved in the production of siliciclastic sediments. Chemical weathering of bedrock may produce weathering profiles with distinct mineralogical zones, compositionally much different from the parent rock. Mass wasting of such profiles produces sediments that reflect the mineralogy of the zones exposed to mechanical erosion, rather than the composition of fresh bedrock. The relative rates of chemical weathering and mechanical erosion determine which mineralogical zones are exposed to mass wasting, and therefore control the compositions of siliciclastic sediments. Stable rates of chemical weathering and erosion result in steady‐state weathering, so that thickness and the mineralogical composition of eroded soil zones, and therefore the mineralogy of derived sediments, remain unchanged while steady‐state weathering prevails. Non‐steady‐state weathering occurs where climate and tectonism vary. Changing conditions alter rates of chemical weathering and physical erosion, resulting in exposure of different, and sometimes all, weathering zones of profiles, or exposure of bedrock. Such conditions result in production of sediments with diverse mineralogy, reflecting incipiently‐to‐highly weathered zones of profiles. Steady‐ and non‐steady‐state weathering of granitic rocks therefore can be assessed by study of variations in quartz and feldspar contents of sands and variations in bulk compositions of muds, as shown for sediments derived from the Sierra Nevada and Bega batholiths, and the Appalachian Piedmont. Compositional variation of sediments, or its absence, are controlled by the relative rates of chemical weathering and erosion and provides insight into climatic and tectonic conditions in source lands, as well as information about provenance composition.

Reactivity of surface chemical states on fractured pyrite

Abstract Synchrotron-radiation-excited photoelectron spectroscopy was used to monitor sulfur chemical states on fractured pyrite surfaces reacted with atmospheric gases. The results demonstrate that there are at least three distinct states at the pyrite surface and each is oxidised at a very different rate in air. The two surface chemical states are more reactive than bulk sulfur, the most reactive surface sulfur component being S 2− . The second chemical state is identified as the surface atom of the first disulfide layer (S 2− 2 ), and the least reactive species are sulfur atoms of disulfide groups beneath the surface layer (i.e., all sulfur atoms having bulk coordination). A model combining the interpretation of sulfur surface species after Nesbitt et al. (Am. Mineral. in press) and the proposed oxidation mechanism of Eggleston et al. (Am. Mineral. 81 (1996) 1036) was developed to explain the initial oxidation processes on pyrite surfaces, where air oxidation of pyrite commences with the oxidation of S 2− sites at the surface.

Paleoclimatic control on the composition of the Paleoproterozoic Serpent Formation, Huronian Supergroup, Canada: a greenhouse to icehouse transition

Unlike other feldspathic arenites in the ∼ 12 km thick Huronian Supergroup, strata of the 250–350 m thick Serpent Formation contain abundant plagioclase. Deposited in a distal alluvial setting, the Serpent Formation consists of arenites with minor shale and siltstone. The plagioclase abundance could reflect unroofing of a unique source relative to other Huronian units, or could be due to less intense chemical weathering. Major-, trace- and rare earth-element (REE) geochemistry of the Serpent Formation are used to evaluate these two possibilities. Serpent Formation source rocks form part of the Archean Superior Province, which mainly consists of tonalitic, granitic and supracrustal assemblages and their metamorphosed equivalents. In Al2O3CaO* + Na2OK2O (ACNK) space, shales define a linear array which indicates that the unweathered source contained plagioclase and K-feldspar in the ratio of 5:1. This particular array indicates that Serpent fluvial systems tapped a variably weathered carapace developed on the Superior Province, rather than purely fresh bedrock. A modelled mixture of 80% tonalite- and 20% granite-group rocks reproduces REE and trace element characteristics of the Serpent Formation sandstones and shales: average Eu/Eu*= 0.87, LaN/SmN= 4.4, GdNYbN= 1.8 and Th/Sc= 1.65. Mass balance considerations indicate that less than 5% supracrustal-group rocks can be added at the expense of the granite-group. A unique source for Serpent detritus is rejected because underlying and overlying units have approximately the same provenance composition. Plagioclase:K-feldspar ratios in Serpent sandstones range from 2:1 to 1:2, which achieves mass balance among source, mudstone and sandstone compositions when plotted in ACNK space. Achievement of mass balance implies that sandstones and mudstones were derived from the same weathering profiles. These findings indicate that plagioclase preservation in the Serpent Formation is related to less intense (compared with other Huronian arenites) paleoweathering conditions, probably a harbinger of the widespread continental glaciation recorded in the overlying Gowganda Formation.

Paleoclimatology and provenance of the glaciogenic Gowganda Formation (Paleoproterozoic), Ontario, Canada: A chemostratigraphic approach

The Gowganda Formation (dated as about 2.3 Ga) is widely (but not universally) interpreted as glaciogenic, and our chemostratigraphic study supports such an interpretation. The bulk compositions of diamictite matrix materials and associated argillites are considered to have resulted from the mixture of two detrital components; one is glacial flour, similar in composition to the unweathered Archean upper crust, whereas the other is chemically weathered detritus resembling average Proterozoic shale in composition. The first component has a chemical index of alteration (CIA) value of ∼50, reflecting the abundance of feldspar and dearth of aluminous clay minerals. The second component has a CIA value of ∼70 and was produced primarily in weathering profiles by chemical weathering, typical of temperate climatic regimes. Diamictite matrix materials (92 samples) have an average CIA value of 57. The argillites have higher values; the average is 62 (97 samples). The lower CIA values of the diamictites reflect a higher proportion of glacial flour, suggesting a stronger influence of frigid climatic conditions than that which prevailed during deposition of the argillites. Systematic trends in CIA values are developed across diamictite-argillite boundaries and within a thick argillite between two major diamictite units. These values reflect climatic variations, with maximum amelioration during deposition of the middle section of the argillite. Rare earth element data and Th/Sc and Ti/Al ratios from the diamictite matrix materials all suggest a provenance that included a large supracrustal component (∼45%), together with lesser amounts of tonalite-trondhjemite-granite and granite. The proportion of supracrustal materials such as volcanic rocks and shales may have been exaggerated by selective comminution during glacial transport. The matrix materials of the Gowganda diamictites are severely depleted in Ca and proportionally enriched in Na, relative to the source. The anorthitic component of plagioclase has been selectively replaced by an albitic component during diagenesis and metamorphism. A chemostratigraphic approach provides high-resolution data for reconstruction of paleoclimatic conditions and valuable information concerning provenance and metasomatic processes.

A SIMS and XPS study of dissolving plagioclase

Abstract In an earlier report, we showed that altered layers formed on the surface of dissolving labradorite feldspar grains, and that the thickness of these layers (up to hundreds of angstroms) is strongly dependent on the pH of the reactant solution. In this paper, we show that the thickness of these altered layers also depends strongly on the composition of the plagioclase feldspar. Secondary ion mass spectrometry (SIMS) and X-ray photoelectron spectroscopy (XPS) have been used to characterize these altered layers. During dissolution, Ca and Al are removed from the solid material to form an altered layer residually enriched in Si with very similar profiles for Ca and Al. In acidic solutions (pH 3.5) for 90 days, the altered layers increase in thickness from a few hundred angstroms to many hundred angstroms in the order: albite

Reactivity of surface sites on fractured arsenopyrite (FeAsS) toward oxygen

Abstract The reactivity of arsenopyrite (FeAsS) fractured surfaces toward oxygen was studied using synchrotron radiation excited photoelectron spectroscopy (SXPS). The spectra of the pristine surface provide evidence of different As and S surface sites. Signals for As3d with 0.33 eV lower and 0.37 eV higher binding energy than the bulk signal are attributed to arsenic surface sites with filled and unfilled dangling bonds, respectively, caused by the rupture of Fe-As and As-S bonds. Sulfur surface sites with filled dangling bonds, bonded to three iron as well as to two iron and one arsenic atom, give rise to a composite signal in the S2p spectra shifted by 0.79 eV to lower binding energy. Reaction of oxygen with FeAsS surfaces in ultra-high vacuum reveals fast oxidation of As surface sites with filled dangling bonds to As species of increasing oxidation state. The detection of oxidation states As0, As2+, As3+, and As5+ indicate a consecutive reaction scheme for arsenic oxidation involving elementary one-electron transfer steps. The Fe3p spectra have a corresponding intensity increase of a component with a binding energy 1.1 eV higher than the Fe3p signal emitted from the pristine surface. This signal is assigned to Fe bonded to oxidized arsenic. The very small changes in the S2p spectra together with their decreased intensity indicate the formation of an arsenic and iron containing overlayer of oxidation products on top of the FeAsS mineral surface where the S2p signal arises from. In air oxidation of arsenic continues with As5+ being the final oxidation product. An additional Fe3p signal at 56.1 eV binding energy is attributed to Fe bonded to O atoms formed during Fe oxidation. Sulfur oxidation leads to numerous intermediate oxidation products with sulfate being the final product. During air oxidation of up to 30 min, the sulfur signal at the low binding energy side of the S2p spectrum is broadened which is probably caused by S2- formed in layers underneath As and Fe oxidation products. These oxidation products reach the surface by diffusion from the bulk (reaction induced segregation). A model of homogeneous oxidized layers on arsenopyrite indicates that reaction with air has produced a layer containing iron bonded to oxygen on top of the increasingly oxidized arsenic and iron containing layer. The Fe-O overlayer is about 1.8 monolayers thick and is probably formed through interaction of water with iron surface sites

Characteristics of altered labradorite surfaces by SIMS and XPS

Abstract Altered layers, produced by the reaction of plagioclase (labradorite) with dilute solutions of pH 3.5 and 5.7, form at feldspar surfaces. These layers have been identified using secondary ion mass spectrometry (SIMS) and X-ray photoelectron spectroscopy (XPS). During reaction, Na, Ca and Al are extracted from, the solid in preference to Si, yielding an altered layer enriched in Si. The thickness and composition of the altered layers are dependent on the pH of the reactant solution. Using the difference in the observed secondary ion intensity (from the SIMS profiles) between the reacted and control samples, the thickness of the altered layers can be estimated. The estimated thickness of these layers ranges from tens of angstroms forpH = 5.7 conditions, up to hundreds of angstroms for pH = 3.5 conditions. Scanning electron microscope (SEM) studies do not show any evidence for the formation of secondary phase(s); however, extensive weathering features ( e.g. etch pits) are present on some of the reacted surfaces.