Peak, D., E.J. Elzinga and D. L. Sparks. 2001. Understanding sulfate adsorption mechanisms on iron(iii) oxides and hydroxides: Results from ATR-FTIR spectroscopy. In Heavy Metals Release in Soils, H. M. Selim and D. L. Sparks. eds. Lewis Publishers, Boca Raton.
Heavy Metals Release in Soils Selim; H. Magdi, Louisiana State University, Baton Rouge, Louisiana, USA
Understanding Sulfate Adsorption Mechanisms on Iron (Ill) Oxides and Hydroxides: Results from ATR-FTIR Spectroscopy
Derek Peak, Evert J. Elzinga, and Donald L. Sparks
Sulfate (SO2-4aq) is a weakly basic Group VI oxyanion with a metal center that has a charge of +6. In aqueous solution it exists as either the fully-deprotonated form, or as the singly protonated bisulfate (HSO-4aq) ion (Stumm and Morgan, 1991). The pKa for the protonation reaction is ~1.9, making the fully deprotonated form the dominant ion under normal soil conditions. Sulfate ions have a hydrated radius of about 4 Å. At the present time, the chemistry of sulfate in the soil environment is still poorly understood. In fact, the mechanisms of sulfate sorption have often been the subject of debate, both historically and in the current scientific literature. Sulfate is of interest to soil chemists for environmental and agronomic reasons. It is an essential micronutrient for plant growth. Neither deficiency nor toxicity symptoms are commonly seen in cultivated soils, but sulfate can occur in extremely high levels near sites of mine waste deposition as a result of hydrogen sulfide oxidation (Persson and Lovgren, 1996). Sulfate is a product in the geochemical cycling of pyrite and therefore plays an important role in marine sediment chemistry.
Macroscopic studies of sulfate sorption have suggested that sulfate adsorbs via an outer-sphere (electrostatic) adsorption mechanism on both soils and reference minerals (Charlet et al., 1993). This conclusion is supported primarily by two observations: (1) ionic strength has a great effect on the amount of sulfate that is adsorbed, with increasing adsorption as ionic strength decreases, and (2) no adsorption of sulfate is usually seen above the point of zero charge of the mineral. This fact potentially makes iron and aluminum oxides important sites for sulfate adsorption in soils, since these components have high points of zero charge and are commonly found in soils.
He and colleagues (1996) studied the stoichiometry of hydroxyl-sulfate exchange on gamma aluminum oxide and kaolinite using a back-titration technique. Combining a thermodynamic approach with their data, they suggested that the observed hydroxyl release upon sulfate adsorption need not be associated with a ligand-exchange mechanism. Instead, they suggested a mechanism consisting initially of surface site protonation and generation of a hydroxyl in solution via a reaction such as Al-OH0 + H20 -> Al-OH+2 + OH-. They proposed that this protonation was then followed by the formation of an outer-sphere surface complex: Al-OH+2 SO2-4. This mechanism accounts for the observed proton consumption to maintain pH (neutralizing the hydroxyl generated) without requiring an inner-sphere surface complex.
It has also been shown that the rate of gibbsite dissolution in the presence of sulfate is more rapid than in the presence of chloride (Ridley et al., 1997). While an enhancement of dissolution in the presence of ligands is often attributed to inner-sphere surface complex formation, the explanation for this observed effect was the formation of aluminum-sulfate complexes in solution that enhance mineral solubility. These aluminum-sulfate complexes result in more aluminum being released from the surface because they keep the free aluminum concentration in solution much lower than when only chloride is present.
Sposito (1984) suggested that sulfate adsorption might be of an intermediate nature, sometimes sorbing as an outer-sphere complex and sometimes as an inner-sphere complex via a ligand exchange mechanism. This concept was supported by the observations of Yates and Healy (1975), who investigated sulfate adsorption on both α-FeOOH and α-Cr2O3. Although the rates of hydroxyl exchange for the two sorbents are markedly different, the rate and extent of sulfate adsorption was very similar, implying an outer-sphere complexation mechanism. However, sulfate adsorption also shifts the point of zero charge to higher values on both α-FeOOH and α-Cr2O3, which is consistent with inner-sphere complexation.
Many research groups have modeled sulfate adsorption on both soils and pure mineral components, with differing results. He and co-workers (1997) utilized the triple layer model with outer-sphere surface complex formation to describe sulfate adsorption on y-alumina and kaolinite. Chalet and co-workers (1993) modeled the adsorption of sulfate on an aluminum-coated TiO2, δ-A1203, and an acidic forest soil. They found that outer-sphere complexation described the data well. Sulfate adsorption on goethite has been modeled using several different approaches. Zhang and Sparks (1990) modeled sulfate adsorption on goethite using the triple-layer model and an outer-sphere surface complex. Davis and Leckie (1980) employed a modified triple-layer model to represent sulfate adsorption on goethite. They determined that a mixture of outer-sphere and inner-sphere sulfate surface complexes best described their data. The charge-distribution multisite complexation approach (CD-Music) has been used by several researchers to model sulfate adsorption on goethite. Geelboed and colleagues (1997) described sulfate complexation using an inner-sphere bidentate binuclear surface complex, while Rietra and co-workers (1999) used an inner-sphere monodentate surface complex. Persson and Lovgren (1996) utilized an extended double-layer model to describe their experimental data and concluded that an outer-sphere surface complex was most likely. Ali and Dzombak (1996) modeled adsorption of sulfate on goethite in both the absence and presence of simple organic acids using a generalized two-layer model. Three different inner-sphere surface complexes of sulfate were required to describe the experimental data.
There are a few possible reasons that modelers have not completely agreed about the nature of sulfate-goethite complexation mechanisms. First of all, different syn thesis methods and pretreatment techniques can result in goethite with quite different surface properties. If the morphology and crystallinity of the sorbent varies, then the surface chemistry can sometimes also be different. For example, Sujimoto and Wang (1998) found that hematite morphology had a significant effect on the mechanism of sulfate sorption. Another possible source of error is the fact that the design of some surface complexation models excludes some potential surface complexes. This could result in an incorrect assignment of sulfate surface-complexation mechanisms. For example, the original triple-layer model only considered outer-sphere complex formation, and the CD-MUSIC model as currently implemented only describes inner-sphere complex formation. A third possibility is that none of the surface complexation models used are robust enough to converge only on one unique solution when several surface complexes are occurring simultaneously. Definitive mechanistic information from spectroscopy can constrain models to physically relevant complexes and can therefore improve model refinement.
Although somewhat contradictory to the macroscopic laboratory studies, there is microscopic and spectroscopic evidence of sulfate inner-sphere surface complex ation. Transmission infrared spectroscopic studies of sulfate adsorption on goethite and hematite (Parfitt and Smart, 1978; Tumer and Kramer, 1991) revealed the formation of sulfate bidentate binuclear surface complexes on both solids. XPS studies (Martin and Smart, 1987) also validated this sorption mechanism. More recently, Persson and Lovgren (1996) concluded that outer-sphere adsorption of sulfate on goethite was occurring based on results from diffuse reflectance infrared (DRIFT) spectroscopy. However, these spectroscopic experiments all involved potential sample alteration via either drying, the application of heat and pressure, or dilution in a salt, which could have modified the structure of the original sorption complex.
Due to the potential artifacts in ex situ spectroscopy, it is greatly preferable to conduct in situ experimentation to elucidate interactions that occur in aqueous suspensions. In situ experiments using scanning tunneling microscopy (STM) (Eggleston et al., 1998) and ATR-FTIR spectroscopy (Eggleston et al., 1998; Hug, 1997) to determine the adsorption mechanism of sulfate on hematite have more recently shown that inner-sphere monodentate surface complexes form at the hematite surface under aqueous conditions. Degenhardt and McQuillan (1999) found that sulfate forms primarily outer-sphere surface complexes on chromium (III) oxide hydroxide, with some splitting of infrared bands being observed and attributed to electrostatic forces. Peak et al. (1999) utilized ATR-FTIR to better understand the adsorption of sulfate on goethite. We determined that sulfate forms only outer-sphere surface complexes above pH 6.0, and that it forms a mixture of outer- and inner-sphere surface complexes at pH less than 6.0. This is quite significant, as it demonstrates a continuum between different adsorption mechanisms that can potentially explain the discrepancies in earlier macroscopic and surface complexation modeling research.
Hodges and Johnson (1987) used a miscible displacement technique to follow the kinetics of sulfate adsorptionldesorption on goethite. They found that the reaction kinetics could best be described using diffusion-limited kinetics models. This is not surprising considering the nature of the miscible displacement technique. A miscible displacement setup is a continuous flow system where a small amount of sorbent is injected into a thin disk (usually a filter holder with filter paper). A dilute solution of the sorbate in a constant ionic strength background is flowed through this disk until equilibrium is reached. At this point the solution is changed to pure background electrolyte, and the desorption reaction is monitored. The main problems with this experiment are that there is little mixing inside the thin disk and it is difficult to ensure that the sorbent is distributed evenly inside the thin disk without preferential flow pathways. Both these drawbacks tend to make diffusional forces extremely important in modeling the experimental results. Hodges and Johnson (1987) were unable to clearly determine whether sulfate adsorption was due to ligand exchange or electrostatic attraction in these studies. Pressure-jump chemical relaxation studies investigating sulfate adsorption on goethite (Zhang and Sparks, 1990) suggested an outer-sphere complexation mechanism. This seems somewhat at odds with the recent observation that sulfate fonns both outer-sphere and inner sphere complexes on goethite at pH below 6. However, pressure-jump studies cannot conclusively determine the mechanism by which a reaction proceeds. Instead they determine the number of reaction steps and the rate constants for those individual steps. This information must then be coupled with an equilibrium model andlor spectroscopic studies to accurately assess the identity of these individual reaction products. Furtherinore, if reactions are occurring simultaneously rather than sequentially, then a pressure-jump experiment may not distinguish them effectively if the time scales of the reactions are similar.
The relationship between the symmetry of sulfate complexes and their infrared spectra is well established (Nakamoto, 1986), and it is possible to assign molecular symmetry based on the number and position of peaks that appear in the mid-infrared region. The relationship between the symmetry of surface complexes and the resulting infrared spectrum is summarized in Figure 8.1. With the Attenuated Total Reflectance (ATR) technique under aqueous conditions, there are two infrared sulfate vibrations that are accessible to spectroscopic investigation. They are the non-degenerate symmetric stretching ν1 and the triply degenerate asymmetric stretching ν3 bands (Persson and Lovgren, 1996). As a free anion in solution, sulfate has tetrahedral symmetry and belongs to the point group Td. For this symmetry, only one broad peak at approximately 1100 cm-1, due to the triply degenerate ν3 band, is usually observed. In some cases the ν1 band is also weakly active and appears at around 975 cm-1. Since outer-sphere complexes retain their waters of hydration and form no chemical bonds, it is expected that the symmetry of outer-sphere sulfate complexes is similar to aqueous sulfate. However, distortion due to electrostatic effects could shift the ν3 to higher wave number and cause the ν1 band to become IR active. If sulfate is present as an inner-sphere complex, the symmetry is lowered. As a result, the ν1 band becomes infrared active, and theν3 band splits into more than one peak. In the case of a monodentate inner-sphere surface complex, such as observed for sulfate adsorption on hematite, C3ν symmetry results. The ν3 band splits into two peaks, one at higher wave number and one at lower wave number, while the ν1 band becomes fully active at about 975 cm-1 (Hug, 1997; Nakamoto, 1986). If sulfate forms a bidentate binuclear (bridging) surface complex, the symmetry is further lowered to C2ν, and the ν3 band splits into three bands between 1050 and 1250 cmd5, while the v, band is shifted to around 1000 cm-1 (Nakamoto, 1986). Figure 8.2 contains a table (adapted from Hug, 1997) that details the positions of ν1 and ν3 sulfate bands reported in the literature for different molecular configurations.
In this chapter, we present results from investigation of the mechanism of sulfate adsorption on goethite, hematite, and ferrihydrite as a function of pH. Additionally, the effects of surface loading and ionic strength on sulfate adsorption on goethite were studied. Understanding how adsorption mechanisms are affected by reaction conditions is of considerable interest to developers of surface complexation models as well as soil scientists, but there are few studies that use in situ spectroscopy to study soil chemical reactions over a wide range of conditions.