Some potential limitations of the present study are as

Geochemical models, such as WHAM-VII (Tipping, 1998 and Lofts and Tipping, 2011) and the NICA-Donnan Model (Kinniburgh et al., 1999), include the assumption that trace metals sorbed on humic substances are entirely labile – i.e. they participate in a reversible dynamic equilibrium with free metal ions. However, there is increasing evidence that a non-labile fraction may exist in sub-micron sized colloidal phases and bound to dissolved species. For example, in a study of 18 soils, Ma et al. (2006) found non-isotopically exchangeable Cu in the solution phase filtered through 0.2 μm KW-6002 acetate filters; they suggested the Cu was likely to be associated with dissolved organic matter. Similar suggestions have been made by the same research group (Lombi et al., 2003 and Nolan et al., 2009) for Ni, Cu and Zn. Jackson et al. (2005) in a study of dissolved organic matter (DOM) in river water, measured Ni and U distribution in a colloidal phase separated by size exclusion chromatography (SEC) or flow field-flow fractionation (FIFFF) and concluded that Ni was ‘more labile’ than U because Ni2+ appeared to interact with the SEC column gel phase and the FIFFF membrane to a greater degree. By contrast Laborda et al. (2011), used asymmetrical flow field flow fractionation (AsFIFFF) to separate dispersed colloidal phases of isotopically spiked compost material and, from measurements of the isotopic ratio within different size fractions including the free inorganic phase, they concluded that all Cu and Pb present was isotopically exchangeable .