In both low temperature materials synthesis (sol gel processing, chimie douce) and low temperature geochemistry (weathering, geothermal and hydrothermal reactions, diagenesis) metastable and/or poorly crystalline materials are encountered. Despite the lack of equilibrium, energetic considerations play a major role in the formation and transformation of such phases. Recent calorimetric studies on heats of formation of zeolites and related metastable frameworks, on nanophase alpha and gamma alumina, and on manganese oxides and hydroxides are summarized. These studies illustrate the effects of framework topology, cation content, hydration and surface energy on the formation, persistence, and eventual transformation to more stable phase assemblages of these environmentally important materials.

In this review paper, results on the A12O3-water interactions will be critically examined. The synthesis of alumina precursors rich in structural defects can be operated in an aqueous solution containing ligands with high affinity for A13+aq or by restricted hydrolysis of Al-alkoxides. Calcination of the gels yields transition aluminas containing four (IV), five (V) and six (VI) fold coordinated aluminum. The pore-size distribution, the Alv content and the degree of crystallinity are controlled by the degree of condensation of the oligomer species in the gel. In a dehydrated alumina two kinds of AlIV with isotropic chemical shifts at ∼73 and 58 ppm, respectively, are present. The distorted AlIV (chemical shift at 58 ppm) and Alv are surface species. Upon water chemisorption the distorted AlIV line disappears while the Alvline is reinforced. As bulk rehydration progresses the Alv line intensity decreases. It is only partially restored upon recalcination. Thus, hydration-rehydration cycles cure the solid from its defects and increase crystallinity. The onset of rehydration is the chemisorption of water on the surface Lewis sites just as Lewis sites result from a thorough dehydration.

The range of mixed-oxide materials used in ion exchange and catalytic materials is rapidly expanding from aluminosilicates such as zeolites and clays into materials whose structure and chemical properties are more difficult to predict and understand. A simple model based on local charge has been developed to predict distributions of local bonding configurations of both cations and anions in mixed oxides. The model, illustrated for the case of silicotitanates, provides qualitative predictions regarding both the acid-base and ion exchange properties of such materials as a function of composition.

An improved structural bond strength (SBS) model has been developed to quantify the correlation between glass compositions and their chemical durabilities. The SBS model assumes that the strengths of the bonds between cations and oxygens and the structural roles of the individual elements in the glass are the predominant factors controlling the composition dependence of the chemical durability of glasses. The structural roles of oxides in glass are classified as network formers, network breakers, and intermediates. The structural roles of the oxides depend on glass composition and the redox state of oxides. A12O3, ZrO2, Fe2O3, and B2O2 are assigned as network formers only when there are sufficient alkalis to bind with these oxides. CaO can also improve durability by sharing non-bridging oxygen with alkalis, relieving Si0 2 from alkalis. The binding order to alkalis is AI2O3>ZrO2>Fe2O 2>B2O2>CaO>SiO2. The percolation phenomenon in glass is also taken into account. The concentration of network formers has to reach a critical value for a glass to become durable; durable glasses are sufficient in network formers and have a complete network structure; poor durability glasses are deficient in network formers and the network is incomplete and discontinuous. The SBS model is capable of correlating the 7-day product consistency test durability of 42 low-level waste glasses with their composition with an R2 of 0.87, which is better than 0.81 obtained with an eight-coefficient empirical first-order mixture model on the same data set.

Local structures of high surface area sodium titanate materials have been examined as a function of pH using solid state 17O nuclear magnetic resonance (NMR) spectroscopy. Using 17O NMR, it is possible to determine the relative populations of nonbridging oxygens as well as oxygens bridging between two, three, and four titanium atoms. Results show that protonation of ion exchange sites in titanate materials is not a simple ion exchange process but involves a repolymerization and restructuring of the titanate network. The observed structural changes are consistent with reported phase stabilities of titanates in water as well as with known hydrolysis and condensation reactions.

The interaction of both liquid and vapor water on nearly defect-free and defective TiO2(100) surfaces has been investigated using X-ray photoelectron spectroscopy (XPS) and ultraviolet photoemission spectroscopy (UPS). The study has focused on examining electronic or chemical defects as created in vacuum and after exposure to both liquid and vapor water. Defective surfaces were prepared by electron-beam exposure and Ar+ bombardment. For a nearly defect-free (100)1×1 surface, water coverage was ∼0.02 ML (1ML = 2.9×1015/cm2) at 104 L exposure to low vapor pressure water, ∼0.08 ML at 108 L exposure to higher vapor pressure water, and ∼0.12 ML with liquid water exposure, respectively. Defect intensities were greatly reduced after exposing defective surfaces to ∼102 L low vapor pressure water. More significantly, electron-beam induced defects were completely removed upon higher-pressure vapor exposure (>104 L) or liquid water exposure, while defects created by Ar+ bombardment were only partially removed. The adsorption behavior and surface defect reactivity for TiO2(100) 1×1 surfaces were compared with those for TiO2(110) surfaces. The adsorption rates for nearly defect-free (100) 1×1 and (110) surfaces are comparable. However, the rate of defect “healing” for a defective (100) surface is much faster than that for a defective (110) surface.

Metal carbonate surfaces are a common component of soils, sediments, terrestrial organisms, and many engineered structures. The surface-controlled growth and dissolution of metal carbonate minerals (e.g. calcite) often affects fluid flow in pipes (through scaling) as well as the nanoscale assembly of complex biologic structures (similar reactions control Ca levels in the oceans, and ultimately, long-term CO2 levels). Engineering metal carbonate surface chemistry to enhance or inhibit growth, dissolution, or crystal nucleation is therefore an area with obvious economic impact. Metal carbonate superstructures are rapidly altered by dissolution/re-precipitation reactions, necessitating the measurement of surface chemistry near equilibrium. Measurements of metal sorption using limited residence time reactors point to the importance of adsorbed/exchanged metal ions in controlling surface charge and electrokinetic behavior. The same reactions appear to control growth and dissolution rates as well.

Mixed hydroxides of calcium, zinc, cadmium and lead have been recently identified in the high pH environments of hydrating cement. FTIR, XRD, SEM, and SPM have been used to characterize these systems. A chemical equilibrium model of the early hydration of a zinc-doped cement/water system, Na-K-Ca-Zn-H-SO4-OH-Zn(OH)2-Zn(OH)3-Zn(OH)4-H2O, has been developed to better understand the mechanism of the surface formation of calcium hydroxyzincate (CHZ). The model is based on Pitzer's semi-empirical method for calculation of ion-activity coefficients at high ionic strength. The Pitzer parameters for Na+-Zn2+,Na+-Zn2+- and have been evaluated, and the results successfully predict the equilibria (solubilities) of Na2SO4-ZnSO4-H2O, NaOH-Zn(OH)2-H2O and KOH-ZnO-H2O systems. The chemical model clearly demonstrates that the formation of CHZ on the calcium-silicahydrate (C-S-H) surface is critically controlled by the Ca2+ ion concentration as well as pH of the pore water system. The results of this study suggest that the growth of CHZ is preceded by surface complex formation. Sequential charge control and sequential structure development have been used to discuss the surface selectivity of these compounds and their control of cement hydration.

The hydrolytic dissolution of some Zr-bearing minerals has been measured in aqueous KCl and KF solutions. Depending on the mineral being investigated, Zr concentrations varied from about 9 ppb to nearly 40 ppm in equilibrated solutions; this validates the notion of Zr mobility in some crustal fluids. At 50 °C, ΔfG were determined (using the measured solubility products) for Na catapleiite, elpidite, vlasovite and weloganite to be -4654.2±5.6, -7391.0±11.6, -5181.0±7.3 and -7130.5±1.0 kJ/mol, respectively. Calculated phase relationships amongst the zirconosilicates found at the Strange Lake peralkaline complex indicate that elpidite and gittinsite cannot coexist stably implying a two-stage hydrothermal mineralization process.

Synchrotron x-ray absorption fine structure (XAFS) spectroscopy provides quantitative information on the speciation of Cr in complex oxide samples, including the Cr(VI):Cr(III) ratio, without any sample pretreatment. We discuss several XAFS methods that can be used for this purpose, including: (1) measurement of the height of the pre-edge peak of Cr K-edge spectra which is caused by ls→3d electronic transitions occurring predominately in the Cr(VI)O4 tetrahedron; (2) fitting observed XANES spectra using combinations of experimental spectra from Cr(III)- and Cr(VI)-containing model compounds; (3) fitting observed EXAFS spectra using combinations of experimental spectra from Cr(III)- and Cr(VI)-containing model compounds; and (4) ab-initio modeling of observed Cr K-EXAFS spectra using theoretical phase-shift and amplitude parameters for Cr(III)O6 and Cr(VI)O4 models. The first method is shown to be the simplest and most accurate. The samples examined using these methods include highly pure model compounds, Cr3+2O3 and Na2Cr6+O4·4H42O, as well as more complex soil samples taken from Cr-contaminated industrial sites and Cr waste disposal pits. We present XAFS evidence for the reduction of Cr(VI) to Cr(III) after reaction with the Fe(Il)-bearing oxide magnetite, Fe3O4, and show that Cr associated with magnetite in soils is reduced, whereas that associated with non-Fe(II)-bearing silicate minerals is not. Also presented are ab-initio multiplescattering calculations of XANES spectra of an Cr(VI)O4 tetrahedron, which help explain the XANES features of this Cr species in solids and aqueous solutions.

Because of its importance in the fields of environmental and mineral engineering, the adsorption mechanism of metal ions on oxides has been studied rather extensively using such techniques as titration, adsorption isotherm determination, flotation and spectroscopy. However, limited work has been done to compare the behavior of hydrophilic silicates with those that are naturally hydrophobic. Our investigation was undertaken to provide a better understanding of the adsorption mechanism of selected metal ions at the solid/liquid interface through simultaneous measurements of zeta potentials, adsorption densities and turbidity on colloidal suspensions of two silicate minerals, quartz and talc, as a function of pH, to delineate the effect of hydrophobicity and surface heterogeniety on these processes.

The surface charge components of two Georgia kaolinites of differing degree of crystallinity (KGa-1 and KGa-2) were measured using procedures based on charge balance concepts. Permanent structural charge density (Σ0) was determined by measuring the surface excess of Cs, which is highly selective to permanent charge sites. The values of Σ0 determined were -6.3 ± 0.1 and -13.6 ± 0.5 mmol kg-1 for kaolinites KGa-l and KGa-2, respectively. The net proton surface charge density (σH) was determined as a function of pH by potentiometric titration in 0.01 mol dm-3 LiC1. Correction from apparent to absolute values of ΣH was made by accounting for Al release during dissolution, background ion adsorption, and charge balance. Lithium and Cl adsorption accounted for the remainder of the surface charge components. Changes in surface charge properties with time were measured after mixing times of 1, 3, and 15 hours (the latter representing “equilibrium”). Time-dependent behavior is believed to be caused by mineral dissolution followed by readsorption or precipitation of Al on the mineral surface. Both the point of zero net charge (p.z.n.c.) and the point of zero net proton charge (p.z.n.p.c.) changed with mixing time, generally increasing. The “equilibrium” p.z.n.c. values were approximately 3.6 and 3.3 for KGa-1 and KGa-2, respectively, while the corresponding p.z.n.p.c. values were about 5.0 and 5.4. The p.z.n.c. results were in good agreement with previous studies, but the values of p.z.n.p.c. were higher than other values reported for specimen kaolinite.

For applications requiring colloidal particles, it is desirable that they be monosized to better control the structure and the properties. In a number of systems, the monosized particles come together to form aggregates that are also monosized. A model is presented here to explain the formation of these monosized aggregates.

The aqueous chemistry of the Zr-H2O system is reviewed. It is shown that the thermodynamically stable solid phase for the Zr-H2O system at 25°C is the monoclinic phase. After a review of the chemistry of the pure material in water, a summary of the solution equilibria of the Zr-H2O system is presented followed by a discussion of the aqueous phase stability of the monoclinic and tetragonal phases of the pure ZrO2 system. Last, a discussion of the phase stability of the complex metal oxides, particularly the Y2O3-containing TZP in aqueous solution will be reviewed. It will be shown that the phase stability for the complex metal oxide-zirconia systems may be best understood as corrosion with the alloying agent (e.g., Y2O3) subject to leaching from a zirconia-rich interface.

Using the controlled growth technique, nanocrystalline ZrO2 powders have een prepared from solution. By variation of the Y-content (0-8 mol %), redispersable monoclinic, tetragonal or cubic ZrO2 powders with particle sizes between 5 and 10 mn were obtained after crystallization at elevated temperature and pressure. Nanodisperse suspensions of the powders have been used for colloidal processing techniques such as tape casting, slip casting or extrusion. The resulting green bodies with densities of 55 % and average pore sizes of 5 nm could be sintered at temperatures below 1100 °C leading to monoclinic, tetragonal or cubic ZrO2 ceramics.

Vanadium oxide deposited in thin layers from aqueous colloidal dispersions exhibits electronic conductivity by a small polaron hopping conduction mechanism. Conductivity and static dissipative properties of coatings are unaffected by changes in humidity. Because vanadium oxide is highly colored, the deposition of effective antistatic coatings which are transparent and colorless requires that the percolative (networking forming) properties of the colloidal particles be maximized. The percolative properties of the colloid are strongly influenced by morphology of the dispersed particles and the extent to which they are well dispersed in the aquasol. These properties are determined by the synthetic route to the colloid. Vanadium oxide is the most potent antistatic agent known and has been found to provide antistatic properties even when as little as 1 milligram per square meter is used.

The transformation of transition aluminas (i.e. θ-A12O3) to α-A12O3 results in an interconnected, porous microstructure that is commonly described as ‘wormy’ or‘vermicular’. The effect of heat generated during the exothermic transformation at the θ-Al2O3/α-Al2O3 interface on the development of the vermicular α-A12O3 morphology was analyzed in boehmite-derived thin films. The vermicular structure may arise in an analogous manner to the thermal dendrites formed in metals during solidification. However, the analysis shows that heat generation and transfer at the interface does not play a significant role in the development of α-A12O3 dendrites.

We have investigated the structures of U(VI) oxides precipitated from room temperature aqueous solutions at low ionic strength as a function of pH. Using the uranium LIII - edge extended x-ray absorption fine structure (EXAFS) as a probe of the local structure around the uranium, a trend is observed whereby the axial oxygen bond lengths from the uranyl groups increase from 1.80 Å at pH=7 to 1.86 Å at pH=11. A concomitant decrease in the equatorial oxygen and nearest-neighbor uranium bond lengths also occurs with increasing pH. Expansion of the linear O=U=O group is seen directly at the L111 absorption edge where multiple scattering resonances systematically shift in energy. EXAFS curve-fitting analysis on these precipitates and a sample of synthetic schoepite indicate that the structure of the species formed at pH=7 is similar to the structure of schoepite. At pH=11, the precipitate structure is similar to that of a uranate.

The mean particle size of magnetite precipitated in aqueous solution can be adjusted and stabilized against ripening over a large nanometric range (1.5 to 12.5 nm) by controlling the pH and the ionic strength imposed by a non-complexing salt in the precipitation medium. The higher the pH and the ionic strength are, the smaller the particle size is. An explanation for this phenomenon is based on the lowering of the oxide-solution interracial tension due to the surface electrostatic charge increase. A critical pH value corresponding to the saturation of the interface is defined and calculated. When the precipitation is effected above this critical pH value, the spontaneous decrease in surface area by ripening is avoided and, for a given ionic strength, the particle size depends only on the acidity. This model correlates well with the experimental results. This provides the first experimental example of thermodynamic stabilization of oxide nanoparticles.

The first stages of the hydrolysis of Fe(III) in presence of PO4 were studied at the local (P k-edge and Fe k-edge EXAFS spectroscopy) and semi-local scale (SAXS). The presence of PO4 hinders the polymerization of iron at the edge sharing dimer step. Phosphate probably governs the local structure leading to the formation of small and dense clusters. The fitting of both k-edge EXAFS spectroscopies requiered the use of a new 3D model of each nucleation step.