The adsorption-desorption distribution coefficient (Kd) is an important parameter for understanding the mobility of a compound in the environment (partitioning) and its distribution between water, sludge, soil and sediment compartments (OECD, 2000b). The Kd value is used in various environmental models for estimating the extent of removal on sludge during wastewater treatment, of leaching through a soil column, and of runoff from agricultural land into adjacent waters, as examples (Wauchope et al, 2002). The distribution of a chemical between water and soil, sediment or sludge compartments is dependent upon the characteristics of the chemical as well as the matrix; it may be additionally influenced by external factors such as temperature and rainfall (OECD, 2000b; Wauchope et al, 2002). Characterisation of Kd in the laboratory and extrapolation to the field can be difficult, largely because of the complexity of adsorption mechanisms that may be relevant for a given chemical and matrix; the mechanisms may be known and unknown in nature (Boxall et al, 2004; ter Laak et al, 2006).
Kd is often normalised to the organic content of the matrix, in order to obtain the organic carbon-water partition coefficient, KOC (Section 2.3.1); the latter is typically found in many of the current protocols today. While this approach was initially developed for hydrophobic compounds, it is not clear whether such normalisation is appropriate for ionisable substances (Chapter 4). Considerations of pH of the soil/sediment/sludge matrix may be more relevant endpoints for such normalisation, particularly when considering the potential effects of the pKa of the chemical on its potential ionisation and subsequent sorption (Nicholls and Evans, 1991; Franco and Trapp, 2008).
For direct measurement of Kd values, a batch or slurry mixing experiment is usually used: a mass of soil ms (g) is mixed with a volume V (ml) of water or another medium such as aqueous 0.1M CaCl2 (to minimise disruption of the soil mineral balance). A mass mp (g) of chemical is added to the slurry (or added to one phase before mixing) to give an initial concentration Ci =mp/V of chemical in the liquid phase. The slurry is then mixed gently in order to disturb the soil structure as little as possible for a period typically from 2 to 48 hours (24 h being a standard). Many studies have shown that the distribution of the chemical in the system approaches a pseudo-equilibrium within a few hours. The liquid phase is then analysed for the equilibrium concentration Ce of the chemical in aqueous solution. Ideally, the soil phase should also be analysed to confirm the mass balance, but this is usually more difficult and neglected.
The soil sorption constant Kd is then calculated as follows: assuming that all chemical removed from the solution is adsorbed by the soil, the mass of this, historically symbolised by x, is calculated as x= V(Ci-Ce). Then x/ms (Cs) is the concentration of chemical in the solid phase (g/g) and Kd is defined by
Thus, Kd (a partition coefficient) is a ratio of solid phase to solute concentrations. High values of Kd (of the order of 100 or more) indicate that, at any given time, the majority of the chemical is adsorbed to the soil surface and hence is less likely to move in soil, but it does not indicate the strength (reversibility) of that sorption. It should be noted that this is a macroscopic measurement and provides only indirect information on the type of surface interactions taking place. Often Kd values are determined over a range of concentrations at a constant temperature. The resultant plot is termed an adsorption “isotherm”, which can take a number of shapes as illustrated below (Figure 5).
It has been hypothesised by Giles et al. (1960) that the shape of the isotherm obtained is indicative of the adsorption processes occurring. When considering the shape of the isotherm, it is important to note the concentration range over which the isotherm has been produced, as different shapes may be obtained over different concentration ranges.
The C-type isotherm is proposed to indicate that the affinity of the sorbate for the adsorbent remains constant and that the number of adsorption sites is unlimited.
L-type isotherms indicate that the proportion of sorbate adsorbed increases more slowly as the amount of material adsorbed increases and are the most common isotherms obtained for crop protection products.
A more complex situation is represented by the S-type isotherm. This was suggested by Giles et al. (1960) to arise as a result of previously sorbed molecules interacting with the surface to facilitate greater adsorption of subsequent molecules up to a point at which the effect declines. An alternative explanation was given by Sposito (1984) who suggested that other species in solution compete for the sorbate until these species are fully reacted leaving subsequent sorbate to adsorb unhindered. Further possibilities are that that the sorbate has a greater affinity for itself than the surface or that at low concentrations the sorbate is competing against other species in solution (including water) for adsorption sites.
In the next sections, the Kd, KOC and isotherms are defined, and standard methods for the measurement of Kd (OECD 106, OPPTS 835.1110 and OECD 121) are evaluated in terms of their applicability to the ERA of ionisable compounds and recommendations given as to which method(s) are most appropriate.