## Cdgt MagelDGel Af Df dDw11

AGel is the thickness of the gel, Af is the thickness of the pre-filter, d is the thickness of the DBL and Df and Dw are the diffusion coefficients in the pre-filter and DBL, respectively. Since DGel — Df, for a typical filter and APA2 gel, the Af term can be incorporated into Ag. On rearrangement, this gives

If simultaneous measurements are made with DGT devices with different gel layer thicknesses, a plot of 1/M versus Dg is a straight line with a slope (m) of 1/(CDGTDGelAt) and a y-intercept of d/CDGTDwAt. The thickness of the DBL, d, and the concentration in the solution, CDGT, can therefore be calculated:

By measuring Ca and Mg using DGT devices loaded with three different gel layer thicknesses, Alfaro-De la Torre et al. [19] calculated a DBL thickness in an acidic (pH 5.3-5.6) and oligotrophic, Precambrian Shield lake. A mean value of d — 0.0317 0.022 cm was obtained for the whole water column, since no trend was observed with depth. Zhang et al. [7] used a combination of nine DGT devices (three diffusive gel layer thicknesses in triplicate) to measure phosphorus in a shallow quiescent eutrophic pond and estimate a DBL thickness of 0.039 cm. Similarly, Zhang [16] estimated a DBL thickness of 0.0387 0.005 cm for a humic-rich stream, from measurements of Ni, Cu and Zn.

The calculated d might not only represent the thickness of the DBL, but could incorporate a term due to biofouling as mucilaginous bacterial biofilms can form in <2 days [7,60]. The assumption that biofouling affects all DGT devices equally, irrespective of their gel layer thicknesses, is likely valid as they all have the same geometry and pre-filter. In this case, the effect of biofouling on DGT can be calculated using the same approach as for estimating d (Eqs. (11.9) and (11.10)), but d now becomes the mean thickness of the biofilm during the deployment time plus the thickness of the DBL. Where only two gel layer thicknesses with accumulated masses M1 and M2 are available, following equation can be used:

Warnken et al. [21] found when making very precise measurements in the laboratory under well-controlled stirring regimes at different gel layer thicknesses that the above procedure for estimating DBL had problems. They concluded that the effective area of resin-gel that accumulates metal is larger than the geometric area of the window of the DGT device. This effective area of 3.8 cm2 for a standard solution device should be used when measurements are made at a range of gel layer

## Post a comment