Dissolution of silica component of glass network at early stage of corrosion in initially silica-saturated solution

Corrosion of glass in silica-saturated solution has been performed with the assumption that dissolution of silicate species from the glass network would not occur. Using surface-sensitive analytical techniques, we report experimental evidence suggesting the dissolution of silicate network species from a model nuclear waste glass, called international simple glass (ISG), in an aqueous solution initially saturated with soluble silica species. Results from low energy ion scattering and X-ray photoelectron spectroscopy reveal a complete depletion of mobile element species (B, Na) from the ISG surface and an enrichment of Zr on the outmost surface. In support of spectroscopic analyses, results from topographic imaging with atomic force microscopy show a stochastic dissolution of glass surface resulting in a higher surface roughness with increasing corrosion time in aqueous solution. This study shows that a true equilibrium between soluble silica species in the solution phase and silicate species in the glass network could not be warranted by performing corrosion experiments in the solution where dissolved silica species are initially equilibrated with amorphous silica in the presence of KOH. The leaching of mobile species (B, Na) could affect the saturation level of aqueous solution and induce further dissolution of the glass surface.     

Water Corrosion Behavior of Fluoroaluminate Glass

The water corrosion behavior of a typical fluoroaluminate glass was studied in both plate and powder form. The infrared spectra of the corroded plate sample exhibited an overall transmittance decrease but little absorption at 3450 cm^–1. Transmittance in the visible region was substantially reduced with an increase of haziness. Microscopic observation revealed a uniform corrosion of the glass surface without any detectable cracks. Dissolution of the glass powder in water resulted in a slight pH increase of the solution. The leaching was congruent and the dynamic dissolution process monitored by solution conductivity followed a logarithmic correlation with time. The corrosion behavior was greatly different from that of fluoride glasses in other systems.     

Chemical Durability of Na2O-K2O-CaO-SiO2 Glasses

The mixed-alkali effect on chemical durability was evaluated for bulk and powdered Na₂O-K₂O-CaO-SiO₂ glasses. Two contributions to this effect are discussed, i.e. those of glass structure and corrosion solution. The magnitude of the mixed-alkali effect depends on the pH of the corroding solution. For pH<9, alkali-ion leaching is the major mode of glass corrosion, and the mixed-alkali effect is large. However, for pH≥9 congruent glass dissolution is the rate-controlling corrosion mechanism, and the effect is diminished. When ion exchange is the rate-controlling corrosion mechanism the magnitude of the mixed-alkali effect depends on the extent of reaction. Since the kinetics of corrosion are influenced by the ratio of glass surface area to solution volume ( SA/V ), results obtained when this ratio is large may be misleading if extrapolated to smaller ratios. Physical evidence of the mixed-alkali effect in bulk glasses was obtained using scanning electron microscopy and ir reflection spectroscopy.     

Solution‐Controlled Dissolution of Supplementary Cementitious Material Glasses at pH 13: The Effect of Solution Composition on Glass Dissolution Rates

Supplementary cementitious materials (SCMs) are widely used to partially replace portland clinker in blended cements. Reducing clinker contents further without compromising the development of early strength necessitates a better assessment and enhancement of the reactivity of the available SCMs. To this purpose, the reactivity of synthesized calcium aluminosilicate glasses covering a compositional range from blast‐furnace slags (BFS) over fly ashes to silica fume was analyzed by dissolution experiments. Initial glass dissolution rates were measured at 20°C and pH 13, and with varying initial concentrations of aqueous Al, Ca, and Si. At pH 13, glass dissolution rates were observed to scale linearly with the glass Ca/(Al + Si) molar ratio. Ca‐rich blast‐furnace type glass dissolution was shown to be up to one order of magnitude faster than tectosilicate fly ash and silica fume type glass dissolution, supporting different pathways to dissolution. In solutions that are strongly undersaturated with respect to hydrous glass and hydration products, glass dissolution rates are independent of changes in solution undersaturation and aqueous Si activity. In contrast, dissolution rates decrease with aqueous Ca concentration for all glasses and with aqueous Al concentration for tectosilicate‐type glasses. The insights gained are instrumental in finding ways to enhance SCM reactivity.     

Synergy between surface mechanochemistry and subsurface dissolution on wear of soda lime silica glass in basic solution

The polishing of oxide glass in aqueous solution is sensitive to not only the mechanical conditions applied by abrasives but also the chemistry of solution. This study elucidates the synergistic interactions of mechanical and chemical effects—especially, the synergetic effects of surface mechanochemical wear and subsurface dissolution are studied by measuring the material removal rate of soda lime silica (SLS) glass upon rubbing with a Pyrex glass ball in noncorrosive (neutral pH) in corrosive solutions (pH 10 and 13 NaOH) as a function of sliding speed. Based on the synergetic model of surface wear and subsurface dissolution, it is found that the mechanochemical surface reaction dominates the wear behavior of SLS glass in neutral and pH 10 solution conditions; the wear of SLS glass in pH 10 is enhanced, compared to the neutral pH case, due to the presence of OH^- ions at the sliding interface. In the case of pH 13, the dissolution of the densified subsurface region, which is formed due to interfacial friction during the surface wear, becomes significant, further enhancing the material removal yield. The finding provides an insight for designing an efficient polishing process in manufacturing of oxide glass materials with a good surface finish.     

Corrosion of Borosilicate Sealing Glasses for Molten Carbonate Fuel Cells

The development of a new sealant for molten carbonate fuel cells (MCFC) requires a study of the attack of molten carbonates on selected materials. Silica and Pyrex® glasses have better corrosion resistance against molten carbonates than other glasses, but they have unsuitable thermal expansion coefficients. Comparisons have also been made between borosilicate glasses of suitable thermal expansion for sealants for MCFC and silica and Pyrex® glasses. The corrosion kinetics in molten carbonates follows two limiting relations and involves two corrosion mechanisms. The weight loss varied linearly with time, indicating a dissolution of the glass network at short times. Longer times show corrosion, depending on the square root of time, typical of a diffusion mechanism and indicating formation of a protective layer on the surface of the glass. The main crystalline corrosion product is lithium methasilicate. The glass-corrosion rate follows the well-known Arrhenius law. These studies used scanning electron microscopy, X-ray diffraction, X-ray photoelectron spectroscopy, and chemical analysis. A general corrosion mechanism of borosilicate glasses in molten carbonates is proposed.     

Dissolution Rates of Commercial Soda–Lime and Pyrex Borosilicate Glasses: Influence of Solution pH

Dissolution rates of two commercial soda—lime silicate glasses and Pyrex borosilicate glass (Corning 7740) have been measured as a function of pH. The dissolution rate was constant with time and increased with increase in pH from 4 to 9. The activation energy for dissolution was nearly constant with pH for the Pyrex borosilicate glass, and decreased somewhat as pH increased for the soda—lime glasses. The openness of the glass surface influences strongly the rate of dissolution. The solubility of silica as a function of pH increases at pH above 8 but does not seem to be quantitatively related to dissolution rate. Other possible influences on this rate are briefly considered.     

Chemical Durability of Fluoride Glasses: III, The Effect of Solution pH

The aqueous corrosion behavior of glasses from two heavy-metal fluoride families (ZrF₄-based and BaF₂-ThF₄-based) was studied as a function of controlled solution pH and of glass composition. Leach test results demonstrate a profound pH dependence of corrosion over the range pH 2 to 13. Whereas the glasses readily undergo congruent matrix dissolution in acidic media, they exhibit a marked increase in chemical durability in neutral and basic solutions, with the optimum aqueous environment occurring near pH 7. Results of this study are discussed and compared with results reported earlier for leach tests of these two compositional families in unbuffered de-ionized water. In addition, several testing methods, both static and dynamic, are described and point to the importance of appropriate test selection in predicting the corrosion resistance of these glasses over prolonged periods.     

Effect of calcium on dissolution and precipitation reactions of amorphous silica at high alkalinity

A better understanding of silica dissolution–precipitation reactions at high pH aqueous solutions allows for promotion of favorable (e.g., pozzolanic) reactions and mitigation of deleterious (e.g., alkali-silica) reactions in concrete. In this paper, the kinetics and products of silica glass dissolution are studied as a function of solution pH, temperature, and availability of calcium. It was observed that dissolution rate versus time increases linearly with pH and reaches a maximum at pH = 14, with slower dissolution at higher alkalinities. In solutions with similarly high pH, but saturated with portlandite, glass dissolution is significantly slower. This is due to formation of a dense, low porosity, and strongly bonded C–S–H layer on the surface of glass, which serves as a barrier against diffusion of OH⁻ and alkali ions towards the substrate glass. This protective layer forms only when Ca is abundant and portlandite saturation can be maintained on a local scale.     

Nuclear Waste Glass Durability: I, Predicting Environmental Response from Thermodynamic (Pourbaix) Diagrams

Pourbaix diagrams are shown to conceptually explain dissolution of glasses in terms of their thermodynamic stability and surface layer formation. To predict the long-term effects of ground-water contact on the durability of nuclear waste glass in a geologic repository, the nature of the glass dissolution process has been extensively examined. Hydration thermodynamics has been used to quantify the role of glass composition and the effect of solution pH. The glass compositions examined vary widely in composition and include natural, lunar, medieval, and nuclear waste glasses, as well as some glass-ceramics. The known effects of solution pH and oxidation potential (Eh) on glass dissolution are empirically described by thermodynamically calculated Pourbaix diagrams. Statistical analysis of over 300 glass durability tests demonstrates that the Pourbaix diagram can be quantified because of (1) the colinearity of the hydration free energy with solution pH defined by the Nernst equation and (2) the colinearity of the solution pH with the concentration of dissolved silicon and boron in the solution defined by the pH dependence of the ion activities. Construction of Pourbaix diagrams to describe glass dissolution serves to unify and categorize various existing experimental glass dissolution data.     

Interface Reactions Between Metals and Ceramics: Refractory Metals–Fused SiO2 System

In a study of the interfacial reactions between SiO₂ and the refractory metals Ta, Nb, V, and Mo, a wide range of reactions was observed. Investigation of normal and tapered sections at the interface showed considerable reaction and solution in the case of V, reactions in the cases of Ta and Nb, and very little reaction in the case of Mo. The major interfacial reaction was oxygen corrosion followed by (most possibly) reaction with silica glass to form silicates. X-ray data showed that the oxides were the principal components at the interface. Evidence was obtained for the formation of silicides in the case of the system Ta-SiO₂. A very low melting silicate was formed in the case of V and silica.     

Corrosion and Infrared Study of Some γ-Irradiated Lead-Phosphate Glasses Doped with MoO3

In this study we compared the effect of three different concentrations of MoO₃ on the corrosion of sodium phosphate glass containing lead and aluminum oxides to develop its durability. The corrosion was studied in deionized water and acidic solution of 0.1 N HCl. The dissolution mechanism is interpreted depending on the glass weight loss versus time and change in the glass surface. MoO₃ had a stabilizing effect on the glass, causing a decrease in the glass dissolution rate. Acidic medium was found to strongly enhance the glass dissolution. The pH decrease in water with increase in acidic solution gives evidence of the release of some phosphorous ions to form phosphoric acid. The consequent enhancement of the glass durability was influenced by the presence of MoO₃ and showed a remarkable advance by doping with 3 wt% of MoO₃. FT-infrared absorption spectra of the studied glasses show IR vibrational bands due to phosphate groups mainly of the metaphosphate and pyrophosphate units together with the sharing and interference of IR vibrations due to Pb–O bonds. Gamma irradiation produces a minor effect on the IR spectra, and enhances the base glass durability at the beginning of the corrosion process, which can be related to the shielding behavior of the high lead glass. However, glass containing 3% MoO₃ has the best durability and is the least affected by irradiation. The glass durability corresponding to its structure is verified by IR spectroscopy.

     

Stress Corrosion and Static Fatigue of Glass

Stress corrosion cracking of six glasses was studied using fracture mechanics techniques. Crack velocities in water were measured as a function of applied stress intensity factor and temperature, and apparent activation energies for crack motion were obtained. Data were consistent with the universal fatigue curve for static fatigue of glass, which depended on glass composition. Of the glasses tested, silica glass was most resistant to static fatigue, followed by the low-alkali aluminosilicate and borosilicate glasses. Sodium was detrimental to stress corrosion resistance. The crack velocity data could be explained by the Charles and Hillig theory of stress corrosion. It is probable that stress corrosion of glass is normally caused and controlled by a chemical reaction between the glass and water.     

Mesh-capped probe design for direct pH measurements at an actively corroding metal surface

Quantitative characterization of surface reaction species improves understanding of reaction mechanisms, especially for heterogeneous reactions such as in corrosion and electrolysis. A universal mesh-capped surface pH probe was developed in order to investigate surface proton concentrations for aqueous metal dissolution/deposition reactions. This technique can be generally applied to characterize the surface pH either for bare metal reactions or metals corroding beneath a corrosion product film. Measurements with the new probe design have shown a good degree of reproducibility. The surface pH measurements during corrosion of mild steel have shown that a higher surface pH is achieved at the steel surface as compared with that of the bulk solution, just as predicted from theory. Chemical buffering effects on surface pH were observed during corrosion by contrasting the measurements obtained in HCl deaerated solutions with those containing CO₂, i.e., carbonic acid (H₂CO₃) and acetic acid (CH₃COOH/HAc) solutions. It was found that the surface pH was mainly controlled by the water chemistry and chemical buffer capacity. The surface pH is affected by the deposited corrosion scale porosity. A higher deviation of surface pH compared to the bulk value was observed for more dense layers with lower porosity.     

The dissolution behavior in alkaline solutions of a borosilicate glass with and without P2O5

There are a variety of applications for glasses in alkaline environments, including glass fibers and glass‐coated steel to reinforce concrete structures. To understand how a simple glass reacts in such environments, the dissolution behavior of a 25Na₂O–25B₂O₃–50SiO₂ (mol%) glass, doped with and without 3 mol% P₂O₅, in pH 12 KOH and pH 12 KOH saturated with Ca²⁺ ions was studied. Ca²⁺ ions in the solution significantly reduce the glass dissolution rate by forming a passivating calcium silicate hydrate (C–S–H) gel layer on the glass surface. When these corroded glasses were then exposed to Ca‐free KOH, the C–S–H layer redissolves into the undersaturated solution and the glass dissolution rate increases. For phosphate‐doped borosilicate glass, PO₄^3? units released from the dissolving glass react with Ca²⁺ ions in saturated solutions to form crystalline hydroxylapatite on the glass surface, but this layer does not protect the glass from corrosion as well as the C–S–H does. The nature of the C–S–H layer was characterized by Raman spectroscopy, which reveals a gel layer constituted mainly of silicate anions.     

Bone bonding ability behavior of some ternary borate glasses by immersion in sodium phosphate solution

Some ternary borate glasses of varying compositions were prepared and corrosion behaviors of such ternary borate glasses after immersion in aqueous dilute phosphate solution were studied using different immersion times. Fourier transform infrared (FTIR) absorption spectral measurements were done before and after immersion in the mentioned solution for extended times up to 3 days to justify the appearance of the characteristic FT infrared bands due to calcium phosphate (hydroxyapatite (HA)) which is considered as the potential indication of bone bonding ability.Experimental IR data confirm the beginning of the appearance of FTIR bands at about 580 and 650 cm⁻¹ after 1 day and the complete resolution with its characteristic split form after 3 days and more.The corrosion behavior of the studied borate glasses is explained in relation to a suggested hydrolysis and direct dissolution mechanism. The ease of dissolution of all the borate glasses constituents explains the rapid conversion and formation of hydroxyapatite within the borate glass matrix as indicated and confirmed by X-ray diffraction.     

Molecular dynamics study on the co-doping effect of Al2O3 and fluorine to reduce Rayleigh scattering of silica glass

Further attenuation of Rayleigh scattering of silica glass is required to extend the capacity of long-distance optical fiber communication. To theoretically examine the effect of aluminum and fluorine co-doping on the density fluctuations of silica glass, which is related to Rayleigh scattering, a set of force-matching potentials (FMP) for simulating F-doped aluminosilicates was developed using Bayesian optimization based on density functional theory calculations. Molecular dynamics (MD) simulations with the new FMP could evaluate the densities of silica glasses to which a small amount of fluorine was doped and those of aluminosilicate glasses with a wide range of aluminum contents within reasonable accuracy. The FMP successfully represents the changing role of aluminum from a network former without a compensating cation (threefold coordination) to that with a compensating cation and a charge compensator in the aluminosilicates. Indeed, relative concentrations of four, five, and sixfold-coordinated aluminum observed by NMR measurements were reproduced better than the original Teter potential at a high aluminum content. At an aluminum content lower than 1 mol%, threefold-coordinated aluminum was observed, which is consistent with ESR measurements. After careful validations of the FMP, the effect of the co-doping of alumina and fluorine on the density fluctuations of silica glass was computationally examined. Consequently, it was expected that the co-doping might not sufficiently attenuate the Rayleigh scattering, even though 1 wt% fluorine would be able to reduce the density fluctuations of aluminosilicate glasses for some extent. This is because more alumina-doping increases density fluctuations of silica glass if the drawing temperature and procedure are the same with those for silica glass fiber. Thereby, a possible fabrication process to sufficiently attenuate the Rayleigh scattering of the F-doped aluminosilicate glass was proposed, according to the density fluctuation analysis at high temperature.     

A Mechanism of Corrosion-Induced Roughening of Glass Surfaces

Atomic force microscopy (AFM), X-ray photoelectron spectroscopy (XPS), and solution analysis by inductively coupled plasma mass spectrometry (ICP-MS) were used to investigate the molecular scale processes responsible for the roughening of glass surfaces due to aqueous corrosion. The study of atomically smooth fiber and melt surfaces allowed direct investigation of the atomic and molecular scale effects of dissolution on surface roughness. The combined use of these analytical techniques clearly showed that the change in RMS roughness with aqueous corrosion could be directly related to the concentration of silica released to solution from the glass; cation leaching alone did not generate detectable roughening. It is well known that nano-/microscale surface roughness can influence strength, optical response, adsorptivity, and other surface properties of glass. It is shown here that the roughening of silicate glass surfaces can be expected based on the amount of silica released from the glass and does not show a dependence on the extent of modifier ion leaching. It is also suggested that the glass composition dependence of this roughening may be a measure of the nanoscale heterogeneity of the glass network structure.     

Evolution of Interfacial Microstructure between Barium Titanate and Binary Glasses

Evolution of interfacial microstructure between BaTiO₃ and the binary glasses used as frits in thick-film technology has been studied. Possible reaction mechanisms for the interfacial compounds have been determined by comparing the results of the sintering study with the reaction of BaTiO₃, powder with glass powders. The interfacial microstructure can be divided into two types. The first type of interfacial microstructure is observed for glasses rich in PbO or Bi₂O₃ network modifiers and is characterized by the penetration of glass into the BaTiO₃ grain boundaries, accompanied by incorporation of Pb into the perovskite lattice. The second type of interfacial microstructure is observed for glasses rich in B₂O₃ or SiO₂ network formers and is characterized by the formation of interfacial compounds.     

Acid Durability of Sodium and Potassium Galliosilicate Glasses

The relative chemical durability in nitric acid of three series of glasses in the (Na,K)₂O-Ga₂O₃-SiO₂ system have been studied. In these compositions, gallium is proposed to behave as either a network modifier or former depending on the gallium to alkali ion ratio of the glass. The mechanisms of acid attack as found in the present work are postulated to be a combination of ion exchange and network dissolution dependent upon the glass composition/structure. This results in complex trends in the acid dissolution of these glasses as functions of the gallium to alkali ion ratio, and silica content.