James O. Leckie

Surface ionization and complexation at the oxide/water interface: I. Computation of electrical double layer properties in simple electrolytes

A new method for determination of intrinsic ionization and complexation constants of oxide surface sites from potentiometric titration data is reported. Using these experimental properties and the stoichiometry of surface reactions, surface charge, σo, adsorption density, Γi, and diffuse layer potentials, ψd, at the oxide/water interface are calculated. The numerical method permits simultaneous calculation of all surface and solution equilibrium states in both simple and complex electrolyte/colloid systems. This allows generalization to dilute ion adsorption in systems with complex solution chemistry. In electrolyte solutions of moderate or high concentration, the surface charge is dominated by surface complex formation of ionizable surface sites and electrolyte ions. In very dilute simple salt solutions the surface ionization reactions are more important. H+ and OH− are the principal potential determining ions (p.d.i.), but the electrolyte ions have a minor effect on the calculated surface potential, ψ0. The electrolyte ions and p.d.i. have a joint role in determining the magnitude of the surface charge via their reactions with surface sites. The important experimental parameters which characterize these contributions are ΔpKcumptex = p*Kcationint − p*Kanionint and ΔpKa = pKa2int − pKa1int.

MULTIPLE-SITE ADSORPTION OF CD, CU, ZN, AND PB ON AMORPHOUS IRON OXYHYDROXIDE

Abstract Adsorption of Cd, Zn, Cu, and Pb onto amorphous iron oxyhydroxide was measured as a function of pH, metal ion concentration, and adsorbent concentration. For each metal, there is a narrow pH band where fractional adsorption increases from near nil to near 100%. For fixed adsorbent concentration, the pH region of the pH-adsorption edge is independent of total adsorbate concentration when adsorption density is less than 10 −5.0 , 10 −3.7 , and 10 −2.3 moles adsorbate per mole Fe for adsorption of Cd, Cu, and Zn, respectively. At larger adsorption densities for these three metals and over the entire range of adsorption densities studied for Pb, the pH region of the adsorption edge becomes more alkaline as total adsorbate concentration increases. In no case did adsorption density attain a maximum, limiting value. The results suggest that the surface is composed of many groups of binding sites. The strength of binding between a given metal and the surface may vary by an order of magnitude or more from one site to another. At small adsorption densities all types of sites are available in excess, and adsorption can be described by the Langmuir isotherm. However, at higher adsorption densities, availability of the strongest binding sites decreases, leading to a decrease in the apparent adsorption equilibrium constant. This phenomenon occurs under conditions where only a few percent of all surface sites are occupied, and is inconsistent with available single-site models.

Surface ionization and complexation at the oxide/water interface II. Surface properties of amorphous iron oxyhydroxide and adsorption of metal ions

Abstract The site-binding model for the electrical double layer of hydrous oxides reported in a previous paper is applied to the adsorption of metal ions from dilute solution and to complex heterogeneous systems, i.e., amorphous iron oxyhydroxide. More than one stoichiometric surface reaction is usually needed to describe the adsorption behavior of dilute heavy metal ions. If mass law equations for surface reactions of metal ions are corrected for effects of the electrostatic field at the interface, the calculated adsorption density depends upon the type of surface species formed. It is shown that calculations with surface reactions involving hydrolytic complexes of metal ions, e.g., Pb(II), Cd(II), Cu(II), Ag(I), are more consistent with experimental adsorption data than complexation by bidentate surface sites. A table of intrinsic surface complexation constants for various metal ions and oxide substrates is presented. Similar to results reported earlier for major electrolyte ions, the stability constants of surface complexes of heavy metal ions with the silica surface are significantly less than for other oxide surfaces. Empirical surface parameters for model calculations with iron oxyhydroxide are derived and the results are compared with experimental adsorption data for Cu(II) and Ag(I).

Modeling ionic strength effects on cation adsorption at hydrous oxide/solution interfaces

Abstract The effects of changes in ionic strength on the adsorption behavior of selenite (SeO32−) and selenate (SeO42−) on goethite and hydrous ferric oxide have been modeled using a generalized version of the triple-layer surface complexation model. Selenite adsorption, which is relatively unaffected by changes in ionic strength, is best modeled assuming that selenite forms an inner-sphere (surface coordination) complex; selenate adsorption, which is markedly reduced by increasing ionic strength, is best modeled assuming that selenate forms an outer-sphere (ion-pair) surface complex. The modeling results suggest that it is possible to distinguish between inner-sphere and outer-sphere anion surface complexes by studying the effects of ionic strength on anion partitioning.

Surface ionization and complexation at the oxide/water interface. 3. Adsorption of anions

Abstract The site-binding model for the electrical double layer of hydrous oxides reported in a previous paper is applied to adsorption of anions from dilute solution. Generally, more than one stoichiometric surface reaction is needed to describe the adsorption behavior of divalent weak acid anions. If mass law equations for surface reactions are corrected for effects of the electrostatic field at the interface, the calculated adsorption density depends upon the type of surface species formed. It is shown that calculations which consider formation of surface complexes by protonated anionic forms, e.g., HCrO 4 − , HSeO 4 − , HSO 4 − , are more consistent with experimental adsorption data than complexation by bidentate surface sites. Modeling results predict that adsorbed anions are more easily protonated than those in bulk water, and a qualitative explanation for this phenomenon is presented. The model applies over a wide range of solute concentrations and accounts for effects of changes in composition of the supporting electrolyte. In addition calculated results for a shift in pH PZC upon specific adsorption of sulfate are in reasonable agreement with other experimental studies.

In Situ X-ray Absorption Study of Surface Complexes: Selenium Oxyanions on α-FeOOH

A novel application of x-ray absorption spectroscopy has provided structural information for ions sorbed at oxide-water interfaces. As an example, in situ extended x-ray absorption fine structure (EXAFS) measurements of adsorbed selenate and selenite ions at ah α-FeOOH(goethite)—water interface have been performed; these measurements show that selenate forms a weakly bonded, outer-sphere complex and that selenite forms a strongly bonded, inner-sphere complex. The selenite ion is bonded directly to the goethite surface in a bidentate fashion with two iron atoms 3.38 angstroms from the selenium atom. Adsorbed selenate has no iron atom in the second coordination shell of selenium, which indicates retention of its hydration sphere upon sorption. This method provides direct structural information for adsorbed species at solid-liquid interfaces.

Surface Complexation Models: An Evaluation of Model Parameter Estimation Using FITEQL and Oxide Mineral Titration Data

Abstract The ability of surface complexation models (SCMs) to fit sets of titration data as a function of changes in model parameters was evaluated using FITEQL and acid-base titration data of α-FeOOH, α-Al2O3, and TiO2. Three SCMs were evaluated: the triple-layer model (TLM), the constant capacitance model (CCM), and the diffuse-layer model (DLM). For all models evaluated, increasing the model input value for the total number of surface sites caused a decrease in the best-fit Log K values of the surface protolysis constants. In the case of the CCM, the best-fit surface protolysis constants were relatively insensitive to changes in the value of the capacitance fitting parameter, C1, particularly for values of C1 greater than 1.2 F/m2. Similarly, the best-fit values of TLM surface electrolyte binding constants were less influenced by changes in the value of C1 when C1 was greater than 1.2 F/m2. For a given C1 value, the best-fit TLM values of the electrolyte binding constants were sensitive to changes in ΔpKa up to ΔpKa values of 3. For ΔpKa values above 3, no changes in the best-fit electrolyte binding constants were observed. Effects of the quality and extent of titration data on the best-fit values for surface constants are discussed for each model. A method is suggested for choosing a unique set of parameter values for each of the models.

Self-Etching Reconstruction of Hierarchically Mesoporous F-TiO2 Hollow Microspherical Photocatalyst for Concurrent Membrane Water Purifications

We report a large-scale self-etching approach for the synthesis of monodispersed mesoporous F-TiO2 hollow microspheres. The self-etching derived from HF was elucidated by the morphology, chemical composition, and crystal size evolutions from solid to hollow microspheres with the increase in the concentration of H2SO4. The resulting TiO2 hollow microspheres exhibited ease for the concurrent membrane filtration and photocatalysis, providing high potential for engineering application in advanced water treatment, for not only increasing water production but also improving water quality.

Complexation of carbonate species at the goethite surface: Implications for adsorption of metal ions in natural waters

Headspace Pco, was measured with an infrared gas analyzer over an equilibrated goethite suspension to determine adsorption of carbonate species in the pH range 3 to 8. For a 2 g/L goethite suspension in 0.1 N NaC104 (-3 1O-4 M surface sites), the fraction of carbonate species adsorbed increased from 0.15 at pH 3 to a maximum of 0.56 at pH 6. In 0.0 1 N NaC104, the fraction of carbonate species adsorbed at pH 6 increased to 0.67. The total concentration of CO2 in the suspension increased from about 0.4 to 0.6 1O-4 M in the pH range of these experiments. The development of surface charge at the goethite surface was determined in the pH range 4 to 11 by potentiometric titration under controlled low CO2 conditions. No hysteresis was observed between the acid and base legs of titrations in 0.10,0.03, and 0.01 N NaC104 resulting in a pH,,, of 8.9. The carbonate species adsorption data were modelled using the least squares optimization program FITEQL for the diffuse double-layer model and the triple- layer model using stoichiometries of the type Fe-OCOOH and Fe-OCOO for surface bound carbonate species. The model results are consistent with separate experiments showing a significant reduction in chromate adsorption on goethite as the partial pressure of CO* was increased from <5 to 450 and 40,000 patm. Our data suggest that mineral oxide surface sites which control solid/solute partitioning of metal ions in natural systems may be largely bound to adsorbed carbonate species.

Cd2+ uptake by calcite, solid-state diffusion, and the formation of solid-solution: Interface processes observed with near-surface sensitive techniques (XPS, LEED, and AES)

Abstract Cadmium uptake by calcite from aqueous solution was studied using techniques sensitive to the near-surface: X-ray photoelectron spectroscopy (XPS), low-energy electron diffraction (LEED), and Auger electron spectroscopy (AES). These techniques allowed direct observations of structure and bonding environments at the calcite surface. The results indicate that the main processes involved in cadmium uptake by calcite are adsorption and solid-state diffusion into the crystal, which leads eventually to the formation of solid-solution. Pure calcite crystals were cleaved from precleaned Iceland spar and were briefly exposed to aqueous solutions containing various concentrations of Cd2+, CO32−, ClO4−, and/or Cl−. Some Cd2+ was radiolabelled. LEED results demonstrate that the calcite surface is atomically ordered, even after hydration and cadmium uptake. γ-scintillation data from crystals exposed briefly to solutions of 109Cd2+ indicate that surface uptake ranged from the equivalent of about 1 to 4 monolayers. XPS analyses in the first 2 hours after exposure detected Cd within the top 30 A, but crystals stored in air or in ultra-high vacuum showed a decrease in Cd surface concentration with time such that after two days, Cd was barely detectable in the near-surface region. In other experiments, LEED verified the crystallinity of otavite (CdCO3) grown epitaxially over the {101} cleavage faces of calcite, and XPS showed almost no Ca in the near-surface on scans taken immediately after precipitation; but after storage for a month in ultrahigh vacuum, binding energy shifts and the presence of a Ca peak strongly suggested the development of solid-solution by diffusion through the solid. No Cd enrichment was observed at sites of surface defects using AES, indicating that solid-state diffusion into the mineral surface was not accomplished simply by migration along microfractures alone. This work suggests that solid-state diffusion may play a role in the rate and extent of uptake of certain trace metals from solution and probably leads to the formation of solid-solution in calcite and other carbonate minerals. It also suggests that the process of diffusion into the solid mineral host should be considered in hydrogeochemical models that intend to simulate and predict trace-metal mobility in carbonate terrains.