Surface complexation models adsorption experiments

In order to test the reversibility of metal-bacteria interactions, Fowle and Fein (2000) compared the extent of desorption estimated from surface complexation modeling with that obtained from sorption-desorption experiments. Using B. subtilis these workers found that both sorption and desorption of Cd occurred rapidly, and the desorption kinetics were independent of sorption contact time. Steady-state conditions were attained within 2 h for all sorption reactions, and within 1 h for all desorption reactions. The extent of sorption or desorption remained constant for at least 24 h and up to 80 h for Cd. The observed extent of desorption in the experimental systems was in accordance with the amount estimated from a surface complexation model based on independently conducted adsorption experiments. 

Batch adsorption experiments by Yee and Fein (2002) using aqueous Cd, B. subtilis, and quartz as a function of pH showed that the thermodynamic stability constants, determined from binary systems, could successfully describe the distribution of Cd between the aqueous phase and the bacterial and mineral surfaces. The constants could also be used to estimate the distribution of mass in systems, and construct a surface complexation model. 

The surface complexation models used are only qualitatively correct at the molecular level, even though good quantitative description of titration data and adsorption isotherms and surface charge can be obtained by curve fitting techniques. Titration and adsorption experiments are not sensitive to the detailed structure of the interfacial region (Sposito, 1984) but the equilibrium constants given reflect - in a mean field statistical sense - quantitatively the extent of interaction. 

The adsorption data is often fitted to an adsorption isotherm equation. Two of the most widely used are the Langmuir and the Freundlich equations. These are useful for summarizing adsorption data and for comparison purposes. They may enable limited predictions of adsorption behaviour under conditions other than those of the actual experiment to be made, but they provide no information about the mechanism of adsorption nor the speciation of the surface complexes. More information is available from the various surface complexation models that have been developed in recent years. These models represent adsorption in terms of interaction of the adsorbate with the surface OH groups of the adsorbent oxide (see Chap. 10) and can describe the location of the adsorbed species in the electrical double layer. 

The surface complexation approach discussed in this section is suitable for adsorption, which dominates at relatively short equilibration times and relatively low concentrations of the adsorbate (up to a few hours, and up to 10 mol dm respectively, for typical experimental conditions). In principle this model is not suitable for long equilibration times or high adsorbate concentrations. Successful applications of surface complexation model SCM to uptake data obtained by coprecipitation (e.g.. Ref. [80]) and other sorption experiments which can be hardly described as adsorption has been reported. In such instances, however, SCM should be rather regarded as a data fitting model than as a mechanistic model. 

During the early development of these surface complexation models, the selection of a particular reaction was based on its ability to lit adsorption data collected in macroscopic experiments in which the extent of sorption was expressed by quantifying the mass of solute lost from solution as a function of pH. Model refinements were based on attempts to more accurately describe surface charge be-... 

However, as we noted early in this chapter, numerous assumptions are employed in the field applications of surface complexation models. Davis et al. (1998) noted that surface complexation models are mainly developed from well-controlled laboratory experiments. It is unclear how the models can be applied to soil and sediments where the double layers of the heterogenous particles may interact and the competitive adsorption of many different ions can cause significant changes in the electrical properties of mineral-water interfaces. 

The model used to evaluate surface chemistry in these systems is the constant capacitance surface complexation model. This model has been used to describe the adsorption of cations (41) and anions (4,8) onto oxides similar to those used in our experiments. A significant difference between those studies and the present study is that we have adapted the model to simulate some of the interactions that might occur between particles in a binary suspension. 

However, models of one reaction were based on processes with sufficiently high rates, when the equilibrium was reached in short experiments. In these experiments the researchers dealt mostly with ion exchange whose rate was very high and controlled mostly by a mechanism of the film diffusion. With an increase in adsorbate concentration becomes noticeable the effect of surface complexation (specific adsorption) and of chemisorption. Their typical feature may be irreversibility. Such processes have a much more complex mechanism and different rates. In order to study their effect, initially is provoked adsorption in conditions when ion i concentration on the surface C is 0 and is drastically overestimated in water solution (C, ). 

One characteristic of the surface complexation modelling with PHREEQC2 is that the pH of the column outflow is also predicted while considering several reactions. As a result it must be stated that this modelling of the pH seems to be a problem. In all three column experiments a steady increase in pH was measured during the adsorption step and a steady decrease during the desorption step. The general effect of proton con-... 

However, the predicted uranium concentration at the edge of the plume is far higher than the observation. Observational data indicate that the maximum uranium concentration outside of the plume is less than 92 pCi/L, while the model predicts it to be in excess of 1000 pCi/L. This discrepancy between observed and predicted uranium concentrations is an indication that the model parameters used to predict U partitioning may not accurately represent conditions at the site. Examination of Dzombak and Morel s (1990) work reveals that they have used a predicted surface complexation constant rather than retrieve a value from experiments. Newer experimental data show that uranium(VI) adsorption onto ferrihydrite can be fitted by a two-site model with a bidentate complex (Waite et al 1994). However, this type of modeling ability is not included with minteqa2. Another explanation could be that co-precipitation is ignored, but apparently is a major attenuation mechanism at a similar site (Opitz et al. 1983). 

The constants for the surface complexation of calcium, sulphate, phosphate and arsenate are included in the file minteq.dat. This data was not sufficient for the modelling of the column experiments performed and had to be augmented. Surface complexation constants for magnesia and chromate for amorphous iron hydroxide were taken from Dzombak and Morel (1990) (see Table 12.3). Van Geen (1994) showed that also carbon dioxide has to be considered for the modelling of adsorption. Carbon dioxide is not mentioned in Dzombak and Morel (1990). The database used contains complexation constants derived from data of Van Geen et al. (1994) and were reoptimized by Dr. C. A. J. Appelo (Amsterdam) for use with PHREEQC2 and amorphous iron hydroxide. This data was transferred from the database file PHREEQC.dat to Minteq.dat. 

The limitations imposed on DDL theory as a molecular model by these four basic assumptions have been discussed frequently and remain the subject of current research.In Secs. 1.4 and 3.4 it is shown that DDL theory provides a useful framework in which to interpret negative adsorption and electrokinetic experiments on soil clay particles. This fact suggests that the several differences between DDL theory and an exact statistical mechanical description of the behavior of ion swarms near soil particle surfaces must compensate one another in some way, at least in certain applications. Evidence supporting this conclusion is considered at the end of the present section, whose principal objective is to trace out the broad implications of Eq. 5.1 as a theory of the interfacial region. The approach taken serves to develop an appreciation of the limitations of DDL theory that emerge from the mathematical structure of the Poisson-Boltzmann equation and from the requirement that its solutions be self-consistent in their physical interpretation. TTie limitations of DDL theory presented in this way lead naturally to the concept of surface complexation. 

---