5.4.1 Laboratory exposure experiments

The performance of the MESCOs described here (Figs. 5.1 and 5.2) for the long-term monitoring of SOCs was investigated by conducting exposure experiments with the flow-through exposure chambers described above. The concentrations in the exposure chambers were between 16 ngm-3 (FLU) and 561 ngm-3 (g-HCH) (see Table 5.2). The concentrations were constant throughout the whole exposure period. The reproducibility, given as the RSD values, was found in both exposure chambers ^12.1% except FLU (17.7%), the compound with the lowest concentration.

In Fig. 5.4, the characteristic update curves of the analyte PCB 28 for the MESCOs A1, B1 and B2 are given. During the whole expose time of 15 days (sampler A1) and 16 days (samplers B1 and B2), respectively, the uptake of the pollutant was linear. The results of the linear regression are given in Table 5.2. The correlation coefficients (r2) of the plots for the MESCOs A1 and A2 (use of the conditioned chamber) were in the range 0.96-1.00 and that of the plots for the MESCOs B1 and B2 (exposure chamber at room temperature) were in the range 0.94-0.99.

The sampling rates RS were calculated using Eq. (5.2). The RS (see Table 5.2) values are in the range 70-320 mL h-1 (sampler A1), 630-4300 mL h-1 (sampler A2), 90-1220 mL h-1 (sampler B1) and 270-540 mL h-1 (sampler B2). Compared with other common passive samplers, such as standard SPMDs [10] the RS values of the samplers studied here are relatively low. For example, RS values of standard SPMDs (450 cm2 surface area) of 190 and 160 L h-1 were calculated for PCB 28 and PCB 52, respectively. This means that the sampling

Fig. 5.4. Uptake profiles for PCB 28 by exposure of the MESCOs A1, B1 and B2 in the calibration atmosphere.

Fig. 5.4. Uptake profiles for PCB 28 by exposure of the MESCOs A1, B1 and B2 in the calibration atmosphere.

efficiency of the SPMDs is two or three orders of magnitude higher than that of the MESCOs described here, mainly due to the higher surface area of the SPMDs. However, the overall sensitivity of the four types of the MESCOs is sufficient because, in the case of these samplers, the total amount of analyte accumulated in the receiving medium is transferred to the GC-MS, whereas, in the case of SPMDs, only a small part of the extract is injected (commonly 1-2 mL, i.e. 1-2% of the analytes sampled) to avoid the potential contamination of the chromatographic system by impurities with higher molecular weights, such as lipids, still present in small quantities in the extracts following the clean-up.

5.4.2 Comparison of the different MESCO types

The MESCOs investigated in this study differ with regard to the membrane thickness (sampler type A: 250 mm, sampler type B: 50 mm), the area of the membrane (type A1: 2.8 cm2, type A2: 12.0 cm2, type B1: 8.1cm2, type B2: 10.2 cm2) and the volume of the enclosed receiving medium (type A1: 24 mL, type A2: 250 mL, type B1: 47 mL, type B2: 125 mL).

The comparison of the two MESCOs of the type A shows that the enlargement of the surface area of the membrane by a factor of about four results in an increase in the RS values by a factor of 7-15 (except FLU). The significantly increased RS values of MESCO A2 are to be expected by the relationship between the surface area of the membrane and the uptake rate RS. Accordingly when MESCO A2 is used, the average amounts of the analytes collected are approximately one order of magnitude higher than when using MESCO A1, resulting in an improved sensitivity. Additionally, the silicone tube-filled sampler is less expensive than the stir bar filled MESCO A1.

The effective membrane surfaces of the MESCOs of the type B differ insignificantly and the uptake rates of HCB, PCB 101, a-HCH and g-HCH are comparable. The RS values for ¿-HCH are greater for the spiral-rod sampler, in contrast the uptake rate of PCB 28 for the stir-bar sampler is considerably larger.

The comparison of sampler types A1 and B1 (volume of the receiving phase of B1 is twice as large as that of A1 and the membrane surface of type B1 is also larger (factor 3.6)) shows that the RS values of B1 for a-HCH, g-HCH, HCB and PCB 28 measured with both sampler types are much higher.

As described by Vrana et al. [1] a negative intercept of the linear uptake curve can be interpreted as a lag-phase between the start of exposure and the penetration of analyte through the diffusion-limiting membrane, i.e. the time required for the analyte to pass the membrane. This lag time for the MESCOs of the type B (50 mm membrane thickness) should be much shorter than for the MESCOs of the type A (250 mm membrane). The calculated lag times for the MESCOs A1 and A2 are in a range between 16 and 65 h. The lag-times for the MESCOs B1 and B2 calculated from the results of the exposure experiments differ much more so that for these samplers no reliable values exist. Further investigations are necessary.

5.4.3 On-site exposure experiments

The field exposure of the MESCO samplers A1 and A2 was performed according to the experimental conditions described above. To screen the site-relevant pollutants, one of the loaded samplers was analysed by TDSA/GC-MS with the MS operated in the scanning mode. Subsequently, the other loaded samplers were analysed in SIM mode (see Section 5.3), for selected relevant pollutants.

Results of this field study are given in Table 5.3. The accumulated amounts of analytes are listed for both types of sampler at an exposure

0 0

Post a comment