Entropy and structure of oxidized and reduced iron-bearing silicate glasses

The influence of ferrous and ferric iron on the low-temperature heat capacity and vibrational entropy of silicate glasses has been determined by adiabatic calorimetry. Two pairs of samples based on sodium disilicate and calcium Tschermak molecule compositions have been studied. Along with previous data for another Fe-bearing glass, these results have been used to complement the available set of composition independent partial molar relative entropies of oxides in silicate glasses with S₂₉₈ − S₀ values of 56.7 and 116 J/mol for FeO and Fe₂O₃, respectively. The calorimetric data indicate that the fraction of fivefold coordinated Al is significant in the CaO-“FeO”-Al₂O₃-SiO₂ system and that association of Ca²⁺ and Na⁺ with Fe³⁺ in tetrahedral coordination for charge compensation does not entail significant changes in coordination for these two cations. At very low temperatures, however, the heat capacity is no longer an additive function of composition because of unexpectedly high positive deviations from Debye laws. These anomalies are stronger for the reduced than the oxidized glasses and considerably larger than for iron-free glasses, but their origin cannot be established from the present measurements.     

The effect of titanium and phosphorus on ferric/ferrous ratio in silicate melts: an experimental study

The effect of TiO₂ and P₂O₅ on the ferric/ferrous ratio in silicate melts was investigated in model silicate melts at air conditions in the temperature range 1,400–1,550 °C at 1-atm total pressure. The base composition of the anorthite–diopside eutectic composition was modified with 10 wt % Fe₂O₃ and variable amounts of TiO₂ (up to 30 wt %) or P₂O₅ (up to 20 wt %). Some compositions also contained higher SiO₂ concentrations to compare the role of SiO₂, TiO₂, and P₂O₅ on the Fe³⁺/Fe²⁺ ratio. The ferric/ferrous ratio in experimental glasses was analyzed using a wet chemical technique with colorimetric detection of ferrous iron. It is shown that at constant temperature, an increase in SiO₂, TiO₂, and P₂O₅ content results in a decrease in the ferric/ferrous ratio. The effects of TiO₂ and SiO₂ on the Fe³⁺/Fe²⁺ ratio was found to be almost identical. In contrast, adding P₂O₅ was found to decrease ferric/ferrous ratio much more effectively than adding silica. The results were compared with the predictions from the published empirical equations forecasting Fe³⁺/Fe²⁺ ratio. It was demonstrated that the effects of TiO₂ are minor but that the effects of P₂O₅ should be included in models to better describe ferric/ferrous ratio in phosphorus-bearing silicate melts. Based on our observations, the determination of the prevailing fO₂ in magmas from the Fe³⁺/Fe²⁺ ratio in natural glasses using empirical equations published so far is discussed critically.     

Water and the density of silicate glasses

A review of published and newly measured densities for 40 hydrous silicate glasses indicates that the room-temperature partial molar volume of water is 12.0 ± 0.5 cm³/mol. This value holds for simple or mineral compositions as well as for complex natural glasses, from rhyolite to tephrite compositions, prepared up to 10–20 kbar pressures and containing up to 7 wt% H₂O. This volume does not vary either with the molar volume of the water-free silicate phase, with its degree of polymerization or with water speciation. Over a wide range of compositions, this constant value implies that the volume change for the reaction between hydroxyl ions and molecular water is zero and that, at least in glasses, speciation does not depend on pressure. Consistent with data from Ochs and Lange (1997, 1999), systematics in volume expansion for SiO₂–M₂O systems (M=H, Li, Na, K) suggests that the partial molar thermal expansion coefficient of H₂O is about 4 × 10⁻⁵ K⁻¹ in silicate glasses. Received: 30 June 1999 / Accepted: 5 November 1999     

A new approach to the equation of state of silicate melts: An application of the theory of hard sphere mixtures

A comparison of compressional properties of silicate solids, glasses, and liquids reveals the following fundamental differences: (1) Liquids have much smaller bulk moduli than solids and glasses and the bulk moduli of various silicate melts have a narrow range of values; (2) Liquids do not follow the Birch’s law of corresponding state as opposed to solids and glasses; (3) The Grüneisen parameter increases with increasing pressure for liquids but decreases for solids; (4) The radial distribution functions of liquids show that the interatomic distances in liquids do not change upon compression as much as solids do. The last observation indicates that the compression of silicate melts occurs mostly through the geometrical arrangement of various units whose sizes do not change much with compression, i.e., the entropic mechanism of compression plays a dominant role over the internal energy contribution. All of the other three observations listed above can be explained by this point of view. In order to account for the role of the entropic contribution, we propose a new equation of state for multi-component silicate melts based on the hard sphere mixture model of a liquid. We assign a hard sphere for each cation species that moves in the liquid freely except for the volume occupied by other spheres. The geometrical arrangement of these spheres gives the entropic contribution to compression, while the Columbic attraction between all ions provides the internal energy contribution to compression. We calibrate the equation of state using the experimental data on room-pressure density and room-pressure bulk modulus of liquids. The effective size of a hard sphere for each component in silicate melts is determined. The temperature and volume dependencies of sphere diameters are also included in the model in order to explain the experimental data especially the melt density data at high pressures. All compressional properties of a silicate melt can be calculated using the calibrated sphere diameters. This equation of state provides a unified explanation for most of compressional behaviors of silicate melts and the experimental observations cited above including the uniformly small bulk moduli of silicate melts as well as the pressure dependence of Grüneisen parameters. With additional data to better constrain the key parameters, this equation of state will serve as a first step toward the unified equation of state for silicate melts.     

X-ray absorption study of Ti-bearing silicate glasses

Ti K-edge XANES spectra have been collected on a series of Ti-bearing silicate glasses with metasilicate and tetrasilicate compositions. The intensity of the preedge feature in these spectra has been found to change with glass composition and varies from 29 to 58% (normalized intensity) suggesting a variation in structural environent around the absorbing atom. The pre-edge peak intensity increases for the alkali titanium tetrasilicate glasses from 35% to 58% in the order Li < Na < K < Rb, Cs whereas for the metasilicate compositions there is a maximum for the K-bearing glass. The pre-edge peak intensity remains constant for the alkaline earth titanium metasilicate glasses, Ca and Sr (34%) but increases slightly for Ba (41%). As the intensity of this feature is inversely correlated with coordination number, a comparison of the pre-edge intensity data for the investigated glasses with those of materials of known coordination number leads us to establish a regression equation and to infer that the average coordination number of Ti in these glasses ranges from 4.8 to 5.8. Large alkali cations appear to stabilize a relatively low average coordination number for Ti in silicate melts. The Ti structural environment results appear also to vary as a function of SiO₂ content within the K₂O-TiO₂-SiO₂ system. A number of physical properties of the melts from which these glasses were quenched and of other Ti-bearing silicate melts, have been determined in recent years. Clear evidence of a variable coordination number of Ti, consistent with the interpretation of the present XANES data is available from density measurements. These and other property determinations are compared with the present spectroscopic observations in an attempt to relate structure and properties in these melts which contain a major component with variable coordination number.     

A study of Wilson's determination of ferrous iron in silicates

The chemical factors affecting the accuracy of Wilson's procedure for determination of ferrous iron in silicates have been investigated. This procedure utilizes pentavalent vanadium to oxidize ferrous iron as it is set free from the silicate by hydrofluoric acid. The quadrivalent vanadium produced is more resistant to oxidation by oxygen than ferrous iron, and conserves the reducing power of the silicate. The procedure is done at room temperature.Recovery of ferrous iron under various conditions has been studied to elucidate the chemical mechanisms of loss in the procedure. Some improvements have been made. The studies indicate that the procedure is of high accuracy.An effect resulting in loss of ferrous iron titre has been found, involving the chemical attack of solids by solutions containing oxygen and pentavalent vanadium. This loss does not appear to take place if all ion species are in solution.     

23Na NMR chemical shifts and local Na coordination environments in silicate crystals,melts and glasses

In order to decipher information about the local coordination environments of Na in anhydrous silicates from ²³Na nuclear magnetic resonance spectroscopy (NMR), we have collected ²³Na magic angle spinning (MAS) NMR spectra on several sodium-bearing silicate and aluminosilicate crystals with known structures. These data, together with those from the literature, suggest that the ²³Na isotropic chemical shift correlates well with both the Na coordination and the degree of polymerization (characterized by NBO/T) of the material. The presence of a dissimilar network modifier also affects the ²³Na isotropic chemical shift. From these relations, we found that the average Na coordinations in sodium silicate and aluminosilicate liquids of a range of compositions at 1 bar are nearly constant at around 6–7. The average Na coordinations in glasses of similar compositions also vary little with Na content (degree of polymerization). However, limited data on ternary alkali silicate and aluminosilicate glasses seem to suggest that the introduction of another network-modifier, such as K or Cs, does cause variations in the average local Na coordination. Thus it appears that the average Na coordination environments in silicate glasses are more sensitive to the presence of other network-modifiers than to the variations in the topology of the silicate tetrahedral network. Further studies on silicate glasses containing mixed cations are necessary to confirm this conclusion.     

The speciation of water in silicate melts

Previous models of water solubility in silicate melts generally assume essentially complete reaction of water molecules to hydroxyl groups. In this paper a new model is proposed that is based on the hypothesis that the observed concentrations of molecular water and hydroxyl groups in hydrous silicate glasses reflect those of the melts from which they were quenched. The new model relates the proportions of molecular water and hydroxyl groups in melts via the following reaction describing the homogeneous equilibrium between melt species: H₂O_molecular (melt) + oxygen (melt) = 2OH (melt). An equilibrium constant has been formulated for this reaction and species are assumed to mix ideally. Given an equilibrium constant for this reaction of 0.1–0.3, the proposed model can account for variations in the concentrations of molecular water and hydroxyl groups in melts as functions of the total dissolved water content that are similar to those observed in glasses. The solubility of molecular water in melt is described by the following reaction: H₂O (vapor) = H₂O_molecular (melt).These reactions describing the homogeneous and heterogeneous equilibria of hydrous silicate melts can account for the following observations: the linearity between f_H2O and the square of the mole fraction of dissolved water at low total water contents and deviations from linearity at high total water contents; the difference between the partial molar volume of water in melts at low total water contents and at high total water contents; the similarity between water contents of vapor-saturated melts of significantly different compositions at high pressures versus the dependence on melt composition of water solubility in silicate melts at low pressures; and the variations of viscosity, electrical conductivity, the diffusivity of “water,” the diffusivity of cesium, and phase relationships with the total dissolved water contents of melts.This model is thus consistent with available observations on hydrous melt systems and available data on the species concentrations of hydrous glasses and is easily tested, since measurements of the concentrations of molecular water and hydroxyl groups in silicate glasses quenched from melts equilibrated over a range of conditions and total dissolved water contents are readily obtainable.     

Quantification of Water Content and Speciation in Synthetic Rhyolitic Glasses: Optimising the Analytical Method of Confocal Raman Spectrometry

Quantification of water content in silicate glasses is of vital significance in understanding magma evolution and metamorphic anataxis. Here we provide a method for the determination of total dissolved water content and water speciation in silicate melts by confocal laser Raman spectrometry based on a set of hydrous rhyolitic glasses. A series of alumino-silicate glasses with water contents from 0.33 to 9.05% m/m were synthesised in a piston cylinder apparatus. Synchrotron-FTIR mapping shows that these glasses have relatively homogeneous distributions of dissolved water. Total water contents of the glasses were precisely measured by TC/EA-MS and FTIR. Both external and internal calibration were established for the quantitative analysis of water content and water speciation in the silicate glasses based on excellent linear correlation between total dissolved water content and integrated area of the water Raman band. Furthermore, by decomposing the total water Raman bands into four Gaussians components, the relative concentration of water speciation (OH groups and molecules H₂O_m) dissolved in the glasses was determined with a similar trend to water speciation data derived from FTIR. We suggest that the relative concentration of water speciation can be estimated in rhyolitic glasses with 4–8% m/m H₂O. Our work provides an accurate method to determine total water content and a potential tool to limit the relative concentration of water speciation dissolved in silicic glasses.     

Mössbauer spectroscopic studies of iron in organic material from natural sedimentary environments

⁵⁷Fe Mössbauer spectra have been obtained from samples of humic acid, fulvic acid and kerogen and from the organic material extracted from bituminous chalk with benzene-methanol. The spectra indicate that iron occurs in a trivalent form in the silicate residue of the humic acid fraction, as hydrated ferrous ions associated with the fulvic acid fraction, as pyrite in kerogen and in a form not detectable by Mössbauer spectroscopy in the benzene-methanol extract.     

Disorder and the extent of polymerization in calcium silicate and aluminosilicate glasses: O-17 NMR results and quantum chemical molecular orbital calculations

Estimation of the framework connectivity and the atomic structure of depolymerized silicate melts and glasses (NBO/T > 0) remains a difficult question in high-temperature geochemistry relevant to magmatic processes and glass science. Here, we explore the extent of disorder and the nature of polymerization in binary Ca-silicate and ternary Ca-aluminosilicate glasses with varying NBO/T (from 0 to 2.67) using O-17 NMR at two different magnetic fields of 9.4 and 14.1 T in conjunction with quantum chemical calculations. Non-random distributions among framework cations (Si and Al) are demonstrated in the variation of relative populations of oxygen sites with NBO/T. The proportion of non-bridging oxygen (NBO, Ca-O-Si) in the binary and ternary aluminosilicate glasses increases with NBO/T. While the trend is consistent with predictions from composition, the detailed fractions apparently deviate from the predicted values, suggesting further complications in the nature of polymerization. The proportion of each bridging oxygen in the glasses also varies with NBO/T. The fractions of Al-O-Si and Al-O-Al increase with increasing polymerization as CaO is replaced with Al₂O₃, while that of Si-O-Si seems to decrease, implying that activity of silica may decrease from calcium silicate to polymerized aluminosilicates (X_SiO2=constant). Quantum chemical molecular orbital calculations based on density functional theory show that a silicate chain with Al-NBO (Ca-O-Al) has an energy penalty (calculated cluster energy difference) of about 108 kJ/mol compared with the cluster with Ca-O-Si, consistent with preferential depolymerization of Si-networks, reported in an earlier O-17 NMR study [Allwardt, J., Lee, S.K., Stebbins, J.F., 2003. Bonding preferences of non-bridging oxygens in calcium aluminosilicate glass: Evidence from O-17 MAS and 3QMAS NMR on calcium aluminate glass. Am. Mineral.88, 949-954]. These prominent types of non-randomness in the distributions suggest significant chemical order in silicate glasses that leads to a decrease in silica activity coefficient and will be useful in modeling transport properties of melts.     

Sulfur speciation and network structural changes in sodium silicate glasses: Constraints from NMR and Raman spectroscopy

Information about the state of sulfur in silicate melts and glasses is important in both earth sciences and materials sciences. Because of its variety of valence states from S²⁻ (sulfide) to S⁶⁺ (sulfate), the speciation of sulfur dissolved in silicate melts and glasses is expected to be highly dependent on the oxygen fugacity. To place new constraint on this issue, we have synthesized sulfur-bearing sodium silicate glasses (quenched melts) from starting materials containing sulfur of different valence states (Na₂SO₄, Na₂SO₃, Na₂S₂O₃ and native S) using an internally heated gas pressure vessel, and have applied electron-induced SKα X-ray fluorescence, micro-Raman and NMR spectroscopic techniques to probe their structure. The wavelength shift of SKα X-rays revealed that the differences in the valence state of sulfur in the starting compounds are largely retained in the synthesized sulfur-bearing glasses, with a small reduction for more oxidized samples. The ²⁹Si MAS NMR spectra of all the glasses contain no peaks attributable to the SiO_4-nS_n (with n > 0) linkages. The Raman spectra are consistent with the coexistence of sodium sulfate (Na₂SO₄) species and one or more types of more reduced sulfur species containing S-S linkages in all the sulfur-bearing silicate glasses, with the former dominant in glasses produced from Na₂SO₄-doped starting materials, and the latter more abundant in more reduced glasses. The ²⁹Si MAS NMR and Raman spectra also revealed changes in the silicate network structure of the sulfur-bearing glasses, which can be interpreted in terms of changes in the chemical composition and sulfur speciation.     

玻璃和熔体包裹体中水含量的拉曼光谱定量测定方面的进展评述

This paper is focused on the progress in the determination of water in glasses and melt inclusions with Raman spectroscopy.Using the presented"Comparator Technique"the water content of a sample is determined by simple comparison with a known standard.A calibration curve is not necessary.Furthermore,with this technique the water concentration in silicate melt inclusions can be determined without exposing the inclusions for measurements.This is very important for extremely water-rich melt inclusions,which would loose H_2O on exposure.     

Assessment of Nanosecond Laser Ablation Multi‐Collector Inductively Coupled Plasma‐Mass Spectrometry for Pb and Sr Isotopic Determination in Archaeological Glass: Mass Bias Correction Strategies and Results for Corning Glass Reference Materials

This work presents an evaluation of various methods for in situ high‐precision Sr and Pb isotopic determination in archaeological glass (containing 100–500 μg g^?1 target element) by nanosecond laser ablation multi‐collector‐inductively coupled plasma‐mass spectrometry (ns‐LA‐MC‐ICP‐MS). A set of four soda‐lime silicate glasses, Corning A–D, mimicking the composition of archaeological glass and produced by the Corning Museum of Glass (Corning, New York, USA), were investigated as candidates for matrix‐matched reference materials for use in the analysis of archaeological glass. Common geological reference materials with known isotopic compositions (USGS basalt glasses BHVO‐2G, GSE‐1G and NKT‐1G, soda‐lime silicate glass NIST SRM 610 and several archaeological glass samples with known Sr isotopic composition) were used to evaluate the ns‐LA‐MC‐ICP‐MS analytical procedures. When available, ns‐LA‐MC‐ICP‐MS results for the Corning glasses are reported. These were found to be in good agreement with results obtained via pneumatic nebulisation (pn) MC‐ICP‐MS after digestion of the glass matrix and target element isolation. The presence of potential spectral interference from doubly charged rare earth element (REE) ions affecting Sr isotopic determination was investigated by admixing Er and Yb aerosols by means of pneumatic nebulisation into the gas flow from the laser ablation system. It was shown that doubly charged REE ions affect the Sr isotope ratios, but that this could be circumvented by operating the instrument at higher mass resolution. Multiple strategies to correct for instrumental mass discrimination in ns‐LA‐MC‐ICP‐MS and the effects of relevant interferences were evaluated. Application of common glass reference materials with basaltic matrices for correction of ns‐LA‐MC‐ICP‐MS isotope data of archaeological glasses results in inaccurate Pb isotope ratios, rendering application of matrix‐matched reference materials indispensable. Correction for instrumental mass discrimination using the exponential law, with the application of Tl as an internal isotopic standard element introduced by pneumatic nebulisation and Corning D as bracketing isotopic calibrator, provided the most accurate results for Pb isotope ratio measurements in archaeological glass. Mass bias correction relying on the power law, combined with intra‐element internal correction, assuming a constant ⁸⁸Sr/⁸⁶Sr ratio, yielded the most accurate results for ⁸⁷Sr/⁸⁶Sr determination in archaeological glasses     

Structure of Cl-containing silicate and aluminosilicate glasses: A Cl MAS-NMR study

Chlorine-35 magic angle spinning (MAS) nuclear magnetic resonance (NMR) spectra were collected at 14.1 and 18.8 Tesla fields to determine the atomic scale structural environments of the chloride ions in anhydrous and hydrous silicate and aluminosilicate glasses containing 0.2 to 0.7 wt% Cl. NMR peaks are broad and featureless, but are much narrower than the total chemical shift range for the nuclide in inorganic chlorides. Peak widths are primarily due to quadrupole interactions and to a lesser extent to chemical shift distributions. Peak positions are quite different for the Na- and Ca-containing glasses, suggesting that most Cl⁻ coordination environments contain network modifier cations. Comparison of peak positions and shapes for silicate and aluminosilicate glasses containing either Na or Ca suggests that there is no obvious contribution from Cl⁻ bonded to Al, and relative quantitation of peak areas indicates that there is no systematic undercounting of ³⁵Cl spins in the aluminous vs. the Al-free samples. In Ca-Na silicate glasses with varying Ca/(Ca + Na), the mixed-cation glasses have intermediate chemical shifts between those of the end members, implying that there is not a strong preference of either Ca²⁺ or of Na⁺ around Cl⁻. Hydrous Na-aluminosilicate glasses with H₂O contents up to 5.9 wt% show a shift to higher frequency NMR signal with increasing H₂O content, while the quadrupole coupling constant (C_Q) remains constant at ∼3.3 MHz. However, the change in frequency is much smaller than that expected if H₂O systematically replaced Na⁺ in the first-neighbor coordination shell around Cl⁻. A series of hydrous Ca-aluminosilicate glasses with H₂O contents up to 5.5 wt% show no shift in NMR signal with increasing H₂O content. The C_Q remains constant at ∼4.4 MHz, again suggesting no direct interaction between Cl⁻ and H₂O in these samples.     

57Fe Mössbauer spectroscopic analysis of chlorite

Five chlorite samples have been studied by ⁵⁷Fe Mössbauer spectroscopy at different temperatures. The spectra are consistently described by a superposition of one ferric and three ferrous quadrupole doublets. All hyperfine data exhibit only minor variations from sample to sample, meaning that the structural and electronic properties of chlorite are rather insensitive to compositional changes and local cation disorder. Two of the ferrous doublets are ascribed to cis and trans co-ordinations in the T-O-T layers. The third ferrous doublet is interpreted as being due to ferrous ions in the hydroxide sheets. Its hyperfine parameters are close to those of the former doublets and the observed deviations from the parameter values found for pure brucite are discussed. Approximately 20 to 25 percent of the total iron is situated in the hydroxide sheets. All five chlorites contain about 10 percent Fe³⁺ which is most probably located on octahedral sites. The temperature dependence of the ferrous quadrupole splittings for two chlorites has been interpreted on the basis of the thermal populations of the two lowest electronic states of the Fe²⁺ ions. From this, the deformations of the different octahedral co-ordinations could be estimated.     

A composition-independent quantitative determination of the water content in silicate glasses and silicate melt inclusions by confocal Raman spectroscopy

A new approach was developed to measure the water content of silicate glasses using Raman spectroscopy, which is independent of the glass matrix composition and structure. Contrary to previous studies, the compositional range of our studied silicate glasses was not restricted to rhyolites, but included andesitic, basaltic and phonolitic glasses. We used 21 glasses with known water contents for calibration. To reduce the uncertainties caused by the baseline removal and correct for the influence of the glass composition on the spectra, we developed the following strategy: (1) application of a frequency-dependent intensity correction of the Raman spectra; (2) normalization of the water peak using the broad T–O and T–O–T vibration band at 850–1250 cm⁻¹ wavenumbers (instead of the low wavenumber T–O–T broad band, which appeared to be highly sensitive to the FeO content and the degree of polymerization of the melt); (3) normalization of the integrated Si-O band area by the total number of tetrahedral cations and the position of the band maximum. The calibration line shows a ±0.4 wt% uncertainty at one relative standard deviation in the range of 0.8–9.5 wt% water and a wide range of natural melt compositions. This method provides a simple, quick, broadly available and cost-effective way for a quantitative determination of the water content of silicate glasses. Application to silicate melt inclusions yielded data in good agreement with SIMS data.     

The effect of crystallinity on dissolution rates and CO2 consumption capacity of silicates

Comparison of measured far-from-equilibrium dissolution rates of natural glasses and silicate minerals at 25 °C and pH 4 reveals the systematic effects of crystallinity and elemental composition on these rates. Rates for both minerals and glasses decrease with increasing Si:O ratio, but glass dissolution rates are faster than corresponding mineral rates. The difference between glass and mineral dissolution rates increases with increasing Si:O ratio; ultra-mafic glasses (Si:O ? 0.28) dissolve at similar rates as correspondingly compositioned minerals, but Si-rich glasses such as rhyolite (Si:O ∼ 0.40) dissolve ?1.6 orders of magnitude faster than corresponding minerals. This behaviour is interpreted to stem from the effect of Si-O polymerisation on silicate dissolution rates. The rate controlling step of dissolution for silicate minerals and glasses for which Si:O > 0.28 is the breaking of Si-O bonds. Owing to rapid quenching, natural glasses will exhibit less polymerisation and less ordering of Si-O bonds than minerals, making them less resistant to dissolution. Dissolution rates summarized in this study are used to determine the Ca release rates of natural rocks at far-from-equilibrium conditions, which in turn are used to estimate their CO₂ consumption capacity. Results indicate that Ca release rates for glasses are faster than those of corresponding rocks. This difference is, however, significantly less than the corresponding difference between glass and mineral bulk dissolution rates. This is due to the presence of Ca in relatively reactive minerals. In both cases, Ca release rates increase by ∼two orders of magnitude from high to low Si:O ratios (e.g., from granite to gabbro or from rhyolitic to basaltic glass), illustrating the important role of Si-poor silicates in the long-term global CO₂ cycle.