Distinguishing Adsorption and Surface Precipitation of Phosphate and Arsenate on Hydrous Iron Oxides

The interaction between ions and solid surfaces in soils, natural waters, and wastewaters is an important control on the concentration and mobility of the ions. Many reactions in natural waters and wastewaters occur at the solid-solution interface. Ions (e.g., arsenate) that adsorb strongly on the surface of naturally occurring solids are generally far less mobile in the environment than ions that do not adsorb strongly (e.g., arsenite). Understanding these reactions is an important component of modelling the movement of the ions in the environment, and in designing treatment systems. 

For the past two decades, ion adsorption on hydrous metal oxide solids has been modelled using the surface complexation model (SCM), which was developed by Stumm and co-workers. The SCM assumes that the adsorbing ion forms a surface complex with the adsorbing site, similar to the formation of a dissolved complex. The modelling is complicated by the effect of the surface charge on the concentration of the ion at the reaction site, and there are several variations of the SCM that use different descriptions for the distribution of the ions around the charged surface.

Unfortunately, the surface complexation model does not work well for the adsorption of anions on hydrous metal oxides, such as iron and aluminium oxides that are common in many environmental systems. The reactions occurring between anions such as phosphate, arsenate, and arsenite at the iron or aluminium oxide surface appear to be far more complex than is envisioned in the SCM model. 

Both phosphate and the different arsenic species are of significant environmental interest. Phosphate is an important plant nutrient as well as a major stimulant of eutrophication in fresh waters, while arsenic is a toxic element found with an increasing frequency in drinking water aquifers in this region of the world. The arsenic contamination of drinking water supplies in the Bengal Basin of India and Bangladesh is affecting tens of millions of people. Since adsorption is an important control on the mobility of these ions in groundwater as well as a potential means of treating the waters, understanding the reactions occurring during adsorption is of more than just theoretical interest. 

One of the difficulties of applying the SCM to anion adsorption is in distinguishing between surface complexation and surface precipitation. Surface complexation is the formation of a single layer of anions attached to the surface of the oxide, whereas precipitation involves the formation of multiple layers of the anion, along with the metal ion from the adsorbing oxide. The distinction is important, since the ions complexed on the surface will behave differently from precipitated ones, and surface precipitation requires a different model than surface complexation. However, it is difficult to determine where the surface reaction changes from the monolayer surface complex formation to the multiple layer surface precipitate. 

The surfaces of the hydrous oxides generally have a positive charge in natural systems, due to protonation of surface sites. The formation of a surface complex between a negatively charged anion and a positively charged surface yields a more negatively charged surface. The addition of another layer of surface precipitate, on the other hand, should have almost no effect on the surface charge, since the nature of the surface has not been changed, merely the thickness of the surface precipitate has changed. One way to differentiate between surface complexation and surface precipitation is thus by the effect on the surface charge. Studies on the surface charge of goethite (a common form of iron oxide) particles with varying levels of phosphate coverage show that the incoming phosphates do not all have the same effect on surface charge (Figure 1). In the figure, particle mobility/zeta potential was measured as an indicator of surface charge. Initially, the incoming phosphate had a constant effect on the surface charge. Above a given surface coverage, however, the amount of negative charge induced on the surface by each phosphate ion greatly decreased, as shown by the break in the curves at each pH value. The second portion of the curve appears to be where surface precipitation is occurring, whereas the first portion is the surface complex formation. Surprisingly, the phosphate concentration, at which the transition occurs, is well below the saturation concentration for iron phosphate in the solution. The surface is clearly promoting the formation of the iron phosphate precipitate. 

The formation of a surface precipitate should result in the burial of the phosphate on the surface, making the phosphate unavailable. The surface complex, on the other hand, is assumed to be in equilibrium with phosphate in solution. However, studies on the competitive adsorption of arsenate and phosphate on the goethite surface have shown that the opposite is true. In these studies, one anion was added to the goethite first, followed by the competing anion at a later time. The first ion must desorb before the competing ion can replace it. Rather than an exchange of the ions at low surface coverage (where surface complexation should predominate) and much less exchange at high surface coverage, where surface precipitation occurs, we found that at low surface coverage, the initially added ion does not exchange at all, whereas at higher surface coverage (or longer times), an increasing amount of the ion added first will exchange with the competing ion. Figure 2 shows the results of arsenate competition with phosphate at the higher surface coverage, when phosphate was added first and the arsenate was added later. The graph shows the amount of phosphate on the surface with the log of time after the start of the experiment. The amount of phosphate on the surface increases linearly with the log time due to a slow reaction between phosphate and the solid surface. When arsenate is added, a portion of the phosphate exchanges with the arsenate. The amount of exchangeable phosphate increases from almost zero at the start of the experiment, to increasingly larger amounts with time. The amount of non-exchangeable phosphate remains relatively constant, at the amount that was adsorbed initially. These results suggest that the initial reaction (presumably the surface complexation) results in a species that is non-exchangeable, whereas the slow, precipitation reaction results in an exchangeable form of phosphate. 

These results have improved our understanding of the interaction between anions such as phosphate and arsenate and hydrous metal oxides such as goethite. The interaction involves both surface complexation and surface precipitation, even at relatively low concentrations of the anion. Surprisingly, it is the surface precipitate that is involved with exchange reactions with other anions, rather than the surface complex. The surface complex appears to be very strongly bound and will not desorb at all. These results may be helpful in designing treatment systems, suggesting that we should control conditions in a way that the surface complexation reaction predominates as compared to surface precipitation.