Surface speciation data

Simple ligands can adsorb on iron oxides to form a variety of surface species including mononuclear monodentate, mononuclear bidentate and binuclear mono or bi-dentate complexes (Fig. 11.2) these complexes may also be protonated. How adsorbed ligands (and cations) are coordinated to the oxide surface can be deduced from adsorption data, particularly from the area/adsorbed species and from coadsorption of protons. Spectroscopic techniques such as FTIR and EXAFS can provide further (often direct) information about the nature of the surfaces species and their mode of coordination. In another approach, the surface species which permit satisfactory modelling of the adsorption data are often assumed to predominate. As, however, the species chosen can depend upon the model being used, this method cannot provide an unequivocal indication of surface speciation confirmation by an experimental (preferably spectroscopic) technique is necessary. 

The association of pollutants such as trace metals, nutrients, and toxic organic molecules to colloids is intimately connected to the health of natural waters. Colloids, with their large specific surface area, play a dominant role in the transportation and eventual deposition of these pollutants. Of particular interest is the size speciation data. It is important to know not only the total amount of pollutant present but also where it is distributed. It has been inherently difficult to study pollutant-colloid interactions because of the lack of methods for particle size determination and fractionation as well as the low concentrations of pollutants present in many systems. This entry outlines a new approach using field-flow fractionation (FFF). 

Our present information on the effect of surface speciation on the reactivity of the surface (i.e., its tendency to dissolve) is summarized in Figure 4. Evidence for the formation of binuclear surface complexes is often circumstantial. Most researchers who modeled surface complex formation with oxy-anions could fit the adsorption data only by assuming the formation of binuclear complexes, usually in addition to mononuclear ones. 

Case Examples. The effects of various oxoanions on EDTA-pro-moted dissolution of lepidocrocite (y-FeOOH) have been studied by Bondietti et al. (33). EDTA was chosen as a reference system because it is dissolution-active over a relatively wide pH range. Phosphate, arsenate, and selenite markedly inhibit the dissolution at near-neutral pH values. At pH <5 phosphate, arsenate, and selenite accelerate the dissolution. It is presumed that the bi-nuclear surface complexes formed at near-neutral pH values by these oxoanions (Table II) inhibit the dissolution. Figure 8a displays data on the effect of selenite on EDTA-promoted dissolution, and Figure 8b shows that calculations on surface speciation by Sposito et al. (35) support the preponderance of binuclear selenite surface complexes in the neutral-pH range. Mononuclear surface species prevail at lower pH values. 

Figure 8. The effect of selenite on the EDTA-promoted dissolution of y-FeOOH 0.5 gIL). Part a At low pH the dissolution rate is increased by selenite at pH 7 it is strongly inhibited. Concentration of the ligands is given in inol/L. Part b Surface speciation on lepidocrocite as a function of pH according to Sposito et al. (35). These data suggest that binuclear selenite surface complexes are formed in the neutral pH range (from reference 33).

Nature of the Surface Complexes. The constant capacitance model assumes an inner-sphere molecular structure for surface complexes formed in reactions like equation 5a or 7. But this structure does not manifest itself explicitly in the composition dependence of Kc everything molecular is buried in which is an adjustable parameter. This encapsulating characteristic of the model was revealed dramatically by Westall and Hohl (13), who showed that five different surface speciation models, ranging from the Gouy-Chapman theory to the surface complex approach, could fit proton adsorption data on AL O., equally well, despite their mutually contradictory underlying molecular hypotheses [see also Hayes et al. (19)]. They concluded that "... no model will yield an unambiguous description of adsorption. .. . To this conclusion one may add that no model should provide such a description,... 

Johnston and Sposito (26) arrived at similar conclusions in a review of approaches to soil surface speciation and then went on to suggest an obvious alternative to the endless archiving of adsorption data ... 

It is inevitable that methodologies not equipped to explore molecular structure will produce ambiguous results in the study of surface speciation. The method of choice for investigating molecular structures is spectroscopy. Surface spectroscopy, both optical and magnetic, is the way to investigate surface species, and thus to verify directly the molecular assumptions in surface speciation models. When the surface species are detected they need not be divined from adsorption data, and the choice of a surface speciation model from the buffet of available software becomes a matter unrelated to goodness-of-fit. 

This chapter is organized under three topical areas. First, by analyzing the dissolution and speciation data of a basalt glass and various oxides we show that the surface characteristics and the dissolution behavior of complex oxides can be modeled from the properties of their constituent oxide components. Then we examine the main features of the steady-state dissolution of multiple oxides and their implications for the application of the surface coordination theory. Finally, we discuss the problem of the nature of the active dissolution sites by analyzing recent data on the dissolution of strained minerals. 

The pll dependence of (-potential, surface speciation and colloid stability ratios is presented in Figure 4. In panel (a) of Figure 4, the solid line represents the surface potential from the model, and the data points correspond to ( potential from electrokinetic measurements. The dashed line represents a reduced potential calculated at a distance of 4.0 nm from the surface. In panel (b). the surface species distribution is calculated by the SCF/DLM model. Under these conditions, the observed zero iiiterfacial potential is in the range pH 7.2 to 7.5, which suggests... 

When represents a metal cation not in the background electrolyte, the intrinsic constants are determined by fitting the triple layer model to adsorption edge data. This fitting entails a surface speciation calculation with previously measured values of the intrinsic constants in Eq. 5.61, the capacitance parameters Ci and C2, and the parameter Af. The computation includes Eqs. 5.58, 5.59, and 5.69, as well as surface charge and mole balance equations imposed as constraints. " ... 

Identification of the specific species of the adsorbed oxyanion as well as mode of bonding to the oxide surface is often possible using a combination of Fourier Transform Infrared (FTIR) spectroscopy, electrophoretic mobility (EM) and sorption-proton balance data. This information is required for selection of realistic surface species when using surface complexation models and prediction of oxyanion transport. Earlier, limited IR research on surface speciation was conducted under dry conditions, thus results may not correspond to those for natural systems where surface species may be hydrated. In this study we review adsorbed phosphate, carbonate, borate, selenate, selenite, and molybdate species on aluminum and iron oxides using FTIR spectroscopy in both Attenuated Total Reflectance (ATR) and Diffuse Reflectance Infrared Fourier Transform (DRIFT) modes. We present new FTIR, EM, and titration information on adsorbed arsenate and arsenite. Using these techniques we... 

Recent developments in spectroscopic techniques offer the opportunity to increase our understanding of oxyanion surface speciation and binding. This understanding is essential to properly use mechanistic sorption models, such as the constant capacitance model and the triple layer model. A recent criticisms of these models is that selection of the surface species and reaction from the sorption data alone results in an empirical model which could be replaced with the traditional Langmuir model (7). However determination of the surface species and reactions will constrain the parameterization and allow for mechanistic evaluation of the sorption models. Knowledge of the actual species and reaction should also enable more generalized prediction of sorption behavior outside the range of the actual experiment, which is not possible at present. 

The combination of shifts in EM with surface charge measurements and titration data can be interpreted to provide important information on the surface speciation, which is complimentary to spectroscopic data. Titration data provide information on the net proton or hydroxyl release from the surface. Among the techniques that can be utilized is suspension of the oxide in the background electrolyte, adjustment of the pH to the value of interest, adjustment of the background electrolyte with the adsorbing ion, then mixing the two and measuring the proton or hydroxyl mass necessary to return to the specified pH, in conjunction with determination of the concentration of anion adsorbed. These data are then reported in mol H or OH per mmol anion adsorbed. 

FIG. 8 Surface speciation of phosphate-doped silica gel (S sp 187m /g, loading of phosphate 1.2 mmol/g), retrieved from titration data by applying the charge regulation model based on the reaction set... 

FIG. 5 Alumina surface speciation> assuming one positively and one negatively charged species each, (a) Relative amounts of species, data from Ref. 25. (b) Absolute coverages, with numbers of centers obtained by model calculation [18]. B, from Ref. 18 C, from Ref 19 D, from Ref. 18. 

Sverjensky, D. A. and K. Fukushi. 2006. A predictive model (ETLM) for As(III) adsorption and surface speciation on oxides consistent with spectroscopic data. Geochimica et Cosmochimica Acta 70, no. 15 3778-3802. doi 10.1016/j.gca.2006.05.012. 

In this section we illustrate the application of GC modeling to the schist materials and discuss the role of data range and surface speciation in the parameterization of GC models. Previously, Waite et al. (2000) developed a two-site GC model for the U(VI) adsorption data for systems equilibrated with air however, the GC model calculations of Waite et al. (2000) assumed that only the surface species, SO2UO2 and SO2UO2COI-, were present (where S denotes a generic surface site, strong or weak), based on the earlier ferrihydrite DDL model of Waite et al. (1994). Because the GC model calculations presented here are performed with an NEM model, it is possible that different surface species may yield a better fit to the experimental data than those found in a DDL model, because of the effect of the EDL terms on adsorption calculations (Dzombak Morel, 1990). 

The surface speciation in Model 3 simulations for systems equilibrated with air and 1% CO2 is shown in Fig. 4-11 and 4-12. The ternary carbonate complexes in Model 3 were of greater importance in comparison to Model 2. This emphasizes again the significance of collecting adsorption data as a function of the partial pressure of CO2, in order to accurately represent the effect of carbonate complexation on both aqueous and surface speciation. Model 3 speciation better represented the experimental data not only as a function of CO2, but also as a function of pH in the pH range from 5 to 7. 

The CA and GC modeling approaches can differ in the selection of surface species to fit experimental data. Surface speciation in the CA modeling approach is normally fixed by a previous fit of an SCM to a reference mineral phase, such as ferrihydiite (Waite et al., 1994 Dzombak Morel, 1990). Ideally, spectroscopic data (XAS, FTIR, and others) are used to constrain the selection of surface species in CA models for reference minerals, so that molecular scale details of the bonding are included within the model as well as EDL terms however, as is illustrated in the CA model calculations presented here and elsewhere (Waite et al., 2000 Davis et al., 1998), application of CA models to soils and sediments is not straightforward. Without very detailed characterization of the physical and chemical characteristics of the surface, several assumptions need to be made to apply a CA model. 

---