Chemcatcher With That Of Other Sampling Devices

The performances of passive samplers can be compared for a range of classes of pollutants, and evaluated alongside other methodologies. For example, calibration data for hydrophobic organic pollutants are available in the literature for SPMDs [22] and the MESCO sampling devices [23,24]. These devices differ in their design geometry and the materials used in their construction. However, the sampling rate is directly proportional to the sampler functional surface area. Consequently, the highest sampling rates will be achieved with passive samplers having the largest surface area, such as the standard size SPMDs (450 cm2 in comparison to 17.5 cm2 for the Chemcatcher). It is therefore necessary to compare the performances on a surface area specific basis, i.e. with sampling rates expressed as volume of water cleared for a chemical, per unit time and unit surface area (Lday^cm-2). In making this comparison it is necessary to take into account reported variations in sampling rates with exposure conditions. Although the most calibration studies reported in the literature were performed in flow-through systems, they were not all conducted under identical conditions (temperature and turbulence). However, if these limitations are taken into account an approximate comparison of sampling rates can be made. The surface-specific sampling rates of three passive sampling devices (MESCO, SPMD and non-polar Chemcatcher) are similar for PAHs compounds with three and four aromatic rings, and range from 5 to 13mLday~1cm~2. This indicates that the uptake of these compounds by the three different samplers is governed overall by a similar mass transfer process; this is most likely to be diffusion across the aqueous boundary layer.

A similar comparison can be made for the polar variant of Chem-catcher and the POCIS. The surface area of the standard configuration of POCIS is 41 cm2 (see Chapter 8), in comparison with 17.5 cm2 for the Chemcatcher. The two samplers are fitted with similar diffusion membrane materials, both are made of PES. The surface-specific sampling rates at room temperature for atrazine and simazine were approximately a factor 2 higher for the Chemcatcher than those reported by Alvarez (Table 8.4 in Chapter 8). This is a reasonable agreement, and the observed difference may be caused by differences in the calibration conditions for the two sets of samplers.

While for the metal version of Chemcatcher, uptake is limited by diffusion in water across the boundary layer and the CA membrane, for the DGT it is restricted by metal diffusion across the hydrogel and only minor effects of the boundary layer are reported [25]. For both samplers, free ions and organic/inorganic metal complexes are able to dissociate within the time required to cross the diffusion layers will accumulate and therefore the TWA concentration will be representative of these fractions. A major difference between these devices is the procedure for the calculation of TWA concentrations. While laboratory-based calibration data are used to calculate TWA concentrations with the Chemcatcher, concentrations for DGT are obtained using known metal diffusivities for the hydrogel layer measured in the laboratory.

In order to evaluate the performance of the Chemcatcher and the DGT when responding to simulated peaks of metal concentrations, a 5-day tank experiment was conducted using Meuse river water. TWA concentrations were measured and compared with the equivalent concentrations from unfiltered, filtered (0.45 mm) and ultra-filtered (5 kDa) spot samples. Figure 9.10 shows a comparison of TWA concentrations measured by the Chemcatcher and the DGT, relative to spot sampling concentrations. While for Cd and Ni, the Chemcatcher slightly underestimates TWA concentrations, the DGT is in better agreement with filtered fractions of these metals. Similar results are obtained for both samplers for Zn and closest agreement is with the filtered fraction. For Cu, both samplers underestimate the filtered concentration while clearly overestimating the ultra-filtered fraction. Generally, results appear in agreement with the speciation of these metals under those conditions. Overall, TWA concentrations obtained using the Chem-catcher appear to have a slight bias as most data points are below the 1:1 relationship. This may be related to the selection of laboratory

Fig. 9.10. Comparison of TWA Cd (O), Cu (A), Ni (V) and Zn (□) concentrations measured by Chemcatcher and DGT relative to TWA concentrations (unfiltered, filtered (0.45 mm) and ultra-filtered (5 kDa)) measured by spot sampling during a 5-day long tank experiment with spiked metals simulating fluctuating concentrations in natural Meuse river water.

Fig. 9.10. Comparison of TWA Cd (O), Cu (A), Ni (V) and Zn (□) concentrations measured by Chemcatcher and DGT relative to TWA concentrations (unfiltered, filtered (0.45 mm) and ultra-filtered (5 kDa)) measured by spot sampling during a 5-day long tank experiment with spiked metals simulating fluctuating concentrations in natural Meuse river water.

calibration data for set levels of temperature and turbulence that differ slightly from conditions observed during the experiment.

0 0

Post a comment