Geoscience Reference
In-Depth Information
modified formulation of the Langmuir isotherm to simulate pH-dependent sorption behavior of
both As(III) and As(V). The formulation was incorporated into the variable saturated reactive
transport solver UNSATCHEM (Simunek and Suarez, 1994; Simunek et al. , 1996), to predict the
potential for As impacts on groundwater from gold mining-related activities. More recently Jeppu
and Clement (2012) introduced a modified sorption isotherm to simulate pH-dependent adsorp-
tion. The modified Langmuir-Freundlich (MLF) isotherm was used to successfully describe the
sorption behavior of arsenic on two sorbents, i.e., goethite and goethite-coated sand.
While in the above-mentioned examples the empirical formulations were able to successfully
describe the observed arsenic sorption behavior, it is important to note that their application is
strictly only warranted under the geochemical conditions for which they were originally developed.
In natural systems, the number of factors influencing the sorption efficiency of arsenic may
be large. Besides pH, arsenic concentration and the aquifer redox status, the concentration of
competing solutes such as phosphate, silicate or bicarbonate or feedback mechanisms between
ambient chemistry and the changes induced by the adsorption reaction itself may affect sorption
and it remains almost impossible to capture those impacts by relatively simple model approaches
(Goldberg, 2007). Thus, for field sites which exhibit spatial or temporal variability in geochemical
conditions, surface complexation models (SCMs) are typically used to describe the interactions
of arsenic with sorbing surfaces and the factors which impact these interactions. The SCMs allow
arsenic adsorption behavior to change as a function of varying geochemical conditions in space and
time. Of course the corresponding increase in model complexity comes at the price of an increased
computational demand and, perhaps more significantly, additional input data requirements.
The most influential study in this context was probably the work of Dzombak and Morel
(1990). They reviewed the available laboratory data for the adsorption of arsenic (and other ions)
by Fe-oxides and fitted the most reliable data with a diffuse double-layer SCM. This model,
while developed for adsorption to pure minerals and therefore ignoring the heterogeneity that
is present in natural groundwater systems, has been applied within many purely geochemical,
i.e., batch-type, as well as reactive transport modeling studies. One of the prominent reasons for
its widespread use is that the SCM and the accompanying thermodynamic database are readily
available within most of the earlier discussed geochemical models. However, other complexation
models have also been proposed, including the constant capacitance model (Stumm et al. , 1980),
the triple layer model (Davis et al. , 1978) and the CD-MUSIC model (Hiemstra and van Riemsdijk,
1999). Detailed appraisals of the various SCM models can be found in the review of Goldberg
et al. (2007) and Appelo and Postma (2005).
Surface complexation models are generally regarded as a progression from empirical formu-
lations, as they incorporate a more fundamental process-based representation of sorption and
desorption and allow for direct coupling of sorption and geochemical equilibria via a thermody-
namic database (Payne et al. , 2013). However, the most appropriate methods to simulate surface
complexation, especially in natural systems, and the methods to determine any associated model
parameters such as numbers and types of surface sites, surface area, relevant sorption reac-
tions and their associated equilibrium constants are still debated. Generalized SCMs, such as the
Dzombak and Morel surface complexation model, are developed on the basis of laboratory data
on sorption of different ions to pure mineral phases. In natural systems, however, sorption is often
controlled by an assemblage of multiple mineral phases. Two major alternative model approaches
have emerged to account for the resulting, more complex adsorption behavior exhibited by nat-
ural sediments, i.e., the component additivity (CA) and generalized composite (GC) approaches
(Arnold et al. , 2001; Davis et al. , 1998; 2005).
The CA approach assumes that the sorbent phase assemblage is composed of a mixture of
identifiable and quantifiable sorbing minerals. In this case, surface chemical reactions are known
for individual mineral components, e.g., from laboratory sorption experiments on goethite or
amorphous Fe-oxides. Thus, depending on the estimate or measurement of the relative amounts
of different sorbing minerals within the sorbent phase assemblage, sorption on each pure phase
mineral is combined in an additive fashion according to the relative concentrations or surface
areas of each mineral present in the natural substrate (Davis et al. , 1998; 2005). Adsorption by
Search WWH ::




Custom Search