The chemistry of dissolved Fe(III) was studied in the Scheldt estuary (The Netherlands). Two discrete size fractions of the dissolved bulk (< 0.2 µm and < 1 kDa) were considered at three salinities (S = 26, 10 and 0.3).Within the upper estuary, where fresh river water meets seawater, the dissolved Fe concentration decreases steeply with increasing salinity, for the fraction < 0.2 µm from 536 nM at S = 0.3 to 104 nM at S = 10 and for the fraction < 1 kDa from 102 nM to 36 nM Fe. Further downstream, in the middle and lower estuary, this decrease in the Fe concentration continues, but is far less pronounced. For all samples, the traditionally recognised dissolved strong organic Fe-binding ligand concentrations are lower than the dissolved Fe concentrations.
Characteristics of dissolved Fe-binding ligands were determined by observing kinetic interactions with adsorptive cathodic stripping voltammetry. From these kinetic experiments we concluded that apart from the well-known strong Fe-binding organic ligands (L, logK' = 19–22) also weak Fe-binding ligands (P) existed with an a value (binding potential = K' · [P]) varying between 1011.1 and 1011.9. The presence of this relatively weak ligand explained the high concentrations of labile Fe present in both size fractions in the estuary. This weak ligand can retard or prevent a direct precipitation after an extra input of Fe.
The dissociation rate constants of the weak ligand varied between 0.5 × 10- 4 and 4.3 × 10- 4 s- 1. The rate constants of the strong organic ligand varied between kd = 1.5 × 10- 3–17 × 10- 2 s- 1 and kf = 2.2 × 108–2.7 × 109 M- 1 s- 1. The dissociation rate constant of freshly amorphous Fe-hydroxide was found to be between 4.3 × 10- 4 and 3.7 × 10- 3 s- 1, more labile or equal to the values found by Rose and Waite [Rose, A.L., Waite, T.D., 2003a. Kinetics of hydrolysis and precipitation of ferric iron in seawater. Environ. Sci. Technol., 37, 3897–3903.] for freshly precipitated Fe in seawater.
Kinetic rate constants of Fe with the ligand TAC (2-(2-Thiazolylazo)-p-cresol) were also determined. The formation rate constant of Fe(TAC)2 varied between 0.1 × 108 and 3.6 × 108 M- 1 s- 1, the dissociation rate constant between 0.2 × 10- 5 and 17 × 10- 5 s- 1 for both S = 26 and S = 10. The conditional stability constant of Fe(TAC)2 (ßFe(TAC)2') varied between 22 and 23.4 for S = 10 and S = 26 more or less equal to that known from the literature (logßFe(TAC)2' = 22.4; [Croot, P.L., Johansson, M., 2000. Determination of iron speciation by cathodic stripping voltammetry in seawater using the competing ligand 2-(2-Thiazolylazo)-p-cresol (TAC). Electroanalysis, 12, 565–576.]). However, at S = 0.3 the logßFe(TAC)2' was 25.3, three orders of magnitude higher. Apparently the application of TAC to samples of low salinity can only be done when the correct ßFe(TAC)2' is known.
