The potential of diffusive gradient in thin film (DGT) as a long-term monitoring tool to assess trace level mercury (Hg) in surface waters was evaluated. A piston type DGT sampler and a plate-type device that could hold 15 DGTs were designed. The device contained piston type DGT samplers with varying diffusive gel thicknesses, that is, 0.5, 0.75, and 1.0 mm, respectively. Three DGT devices were deployed in a lake for 5 weeks, and two were deployed in a stream for 3 weeks. In the lake, the total Hg (THg) mass accumulated in the DGT varied between 0.05 and 0.15 ng, which increased with an increase in deployment time and decreased with an increase in agarose diffusion gel thickness. The DGT concentration in the lake water for a 2 week period was estimated to be about 0.8-1.0 ng/L, which was close to the measured value of 1.1 (± 0.13) ng/L, using the grab sampling technique. However, the DGT estimated at 4 and 6 weeks showed a concentration of about 0.5-0.7 ng/L, which is about twice as small as that measured by grab sampling. This underestimation of the THg levels in water appear to be caused by additional thicknesses of the physical diffusive boundary layer (0.15, 0.5, 1.29 mm) and biofilm, outside the DGT filter. The predicted DGT concentration in the upper stream of the Nakdong River was estimated to be about 0.8-1.4 ng/L, which is similar to the value of 1.22 (± 0.29) ng/L measured in the field by grab sampling. The concentration of THg was estimated to be about 1.0-1.2 ng/L, which is similar to the values measured by grab sampling. The additional diffusion thickness formed outside the DGT filter was 0.018 mm and 0.093 mm at 1 and 3 weeks, respectively, which is not larger than the diffusion gel thickness (0.5-1.0 mm). This was because DGT was installed in a region where the flow velocity is high, and the thickness of the diffusion boundary layer outside the filter is negligible.
Keywords: Diffusive gradient in thin film technique; Lakes; Long-term monitoring; Mercury; Rivers.