Info

River

PCBs, PAHs, PCDDs, PCDFs and substituted benzenes

Biomimetic extraction for toxicity assessment of aqueous contaminants

Effluents of wastewater treatment plant

Organochlorine pesticides, PCBs, PAHs

be considered, i.e. whether they will display linear (integrative), curvilinear or equilibrium uptake kinetics, which depends on sampling conditions and the compounds concerned.

In order to estimate the number of SPMDs required at a specific site Huckins et al. [2] suggested that the following estimated relationship between forecast sampling parameters and the analytical outcome should be applied:

where Rs is the sampling rate (L day-1), t is days of exposure, n the number of SPMDs, Cc the lowest environmental concentration of concern, Pr the method recovery for the analyte as a fraction of one, Et the fraction of the total sample injected into the analytical instrument, MQL the method quantitation limit and Vi the volume of the standard injected.

For example, to design a program for monitoring anthracene in a river with a flow rate of 50 cm s-1 and average temperature of 25°C the following considerations would apply. Our concentration of concern Cc is 10 ng L-1, since this is the detection limit for anthracene required in the EU quality criteria for the aquatic environment. Using the above relationship we can forecast that a single standard SPMD exposed to 10 ng L-1 of aqueous anthracene would sample around 987 ng of the compound during 21 days. The anthracene uptake rate by an SPMD is 4.7 L day-1 at 25°C and a flow rate 50 cm s-1 [5]. Assuming a Pr of 0.7, and an Et of 0.001, the analytical result (i.e. left side of the above equation) would be 690 pg. Clearly this would be much more than the MQL (690 times more, assuming an MQL of 1 pg mL-1 and a V/ of 1 mL), so one SPMD would be enough to obtain sufficient anthracene to analyze the compound at the level of concern stipulated by the EU quality criteria.

Practical applications have proved that one standard SPMD is enough for the concentration and subsequent analysis of a range of PAHs, PCBs and pesticides during 21 days. However, separate standard SPMDs should sometimes be used when determining the concentrations of dioxins, depending on the target detection limits.

Assessments of the study site conditions are also important when interpreting SPMD sampling and analysis results. Data on temperature, water body flow and turbidity should be available for the period of exposure when standard SPMDs are used without PRCs. All the above-mentioned factors, together with the potential for biofouling, will influence the uptake rate by the membrane. Uptake rates for many classes of compounds at various temperatures and flow rates have been established under both field and laboratory conditions and are presented in Huckins et al. [2].

Since environmental variables affect the SPMD uptake of all types of chemicals, it is very important to collect as much data as possible regarding each deployment site and field conditions during deployment and retrieval. When standard SPMDs are being deployed at multiple sites for comparative purposes, investigators should select sites with similar flow regimes or use uptake rates for different flows when calculating ambient concentrations at the respective sites. Uptake rates for low flow and different temperature regimes have been published by Huckins et al. [2]. Another approach would be to use SPMD PRCs to compensate for differences in the environmental conditions at each deployment site. Temperatures (at very least at the beginning and end of the exposure period), visual assessments of the extent of biofouling (e.g. light, medium, heavy and none) and estimates of flow rates should all be noted and recorded [6].

Surface water bodies are often stratified, i.e. often have layers with differing temperatures, densities and chemical compositions. Waters (even oceans) also have high degrees of patchiness that can be manifested (inter alia) in highly localized variations in the density of algal blooms. In various water bodies, the stratum from which the sample is collected is a very important consideration in chemical monitoring programs. Stratification can often be neglected in lakes shallower than 5 m [7]. However, in larger enclosed water bodies, such as lakes, lagoons and epicontinental seas such as the Baltic Sea, boundary layers like the thermocline and halocline in seawater profiles act as ''glide surfaces'' for water and pollutants carried by it. As shown in Fig. 14.2, there can

8. Aug 18. Aug 28. Aug 7. Sept 17. Sept 27. Sept 7. Oct 17. Oct 27. Oct 31. Oct

8. Aug 18. Aug 28. Aug 7. Sept 17. Sept 27. Sept 7. Oct 17. Oct 27. Oct 31. Oct

be substantial temporal variations in the salinity profiles of seawater bodies. These variations should also be taken into account when investigating phenomena such as the transport of pollutants by floodwaters, which usually consist of freshwater. For this reason, oceanographic data on the floodwater distribution were used in a temporal monitoring program of organochlorine compounds close to the Swedish coast following a flooding episode in Western Europe [8]. Oceanographic investigations confirmed that floodwater moved toward the Swedish coast from the North Sea and began to reach the coast in the second half of March 1995. The freshwater current in the North Sea was mixed with the seawater mass during a storm and descended below the halocline (at approximately 20-m depth) just before reaching the Swedish coast. Strong winds at the end of March caused substantial turbulence in the water and temporary movement of the floodwater masses away from the coast. After considering this information, SPMDs were deployed at two stations at 24-m depth a few miles off the coast where floodwater masses were expected to be found. SPMDs were changed every 12-16 days during the 12-week study period. Time trends were detected in which, inter alia contents of DDTs and (to lesser degrees) PCBs, dieldrin, chlordane and HCHs were high during the first two sampling periods when the floodwater arrived, and subsequently declined [8].

When sampling river water, we must also consider the heterogeneity of the system, because the common assumption that running water environments are generally homogeneous in composition, due to mixing by the current, may be erroneous. For example, sum PCBs obtained from two SPMD sampling points (St5 and St5a) situated 15 m apart across the River Umea were found to differ 60-fold, as shown in Fig. 14.3. The average flow of the River Umea is 340 m3 s_1 [10]. Later, a source was found approximately 500-m upstream from sampling sites St5 and St5A, causing the near-shore (10 m) sampling site St5 to be much more polluted than the other.

Lounch et al. have investigated the influence of the spatial variability of the target chemical within the water column on the interpretation of SPMD results [11]. This is particularly important when considering the impact of point sources. They placed three sets of SPMD along a transect across the small river (25 m wide at the deployment site) to investigate if one membrane is sufficient to characterize the cross-section of the river, or a specific river mile or water column at a specific point along the river by measuring PAHs with log Kow values ranging from 4.08 to 5.61. The results showed that we should not assume even a

St3 St4 St5 St5A

Fig. 14.3. Sampling results of sum PCB from River Umea; St3 was located upstream of the city, St4 in the city center and St5 downstream of an old industrial area. St5A was located at the same point along the river as St5, but 15 m further toward the middle of the river [10].

St3 St4 St5 St5A

Fig. 14.3. Sampling results of sum PCB from River Umea; St3 was located upstream of the city, St4 in the city center and St5 downstream of an old industrial area. St5A was located at the same point along the river as St5, but 15 m further toward the middle of the river [10].

west

Fig. 14.4. Influence of spatial variability on sampling by SPMDs during sampling across a cross-section of the river. Three sampling sites across the river (east, centre and west) were chosen [11].

west

Fig. 14.4. Influence of spatial variability on sampling by SPMDs during sampling across a cross-section of the river. Three sampling sites across the river (east, centre and west) were chosen [11].

small canalized river to be homogeneous or ''well-mixed'', even if the river sampling point is relatively distant from point sources of contaminants (in the cited case the nearest point source was 10.4 km away) (see Fig. 14.4). Thus, the spatial variability in water column concentrations needs to be taken into account when interpreting the results from SPMD sampling. Gradients of increasing SPMD residues were found for most of the analyzed PAHs from east to west across the river cross-section [11].

Samples must be taken sufficiently far downstream of effluent discharges and other potential sources of contamination to be sure that the waters have been thoroughly mixed; otherwise samples should be taken from the same side of the stream as the effluent/contamination source.

By choosing SPMDs with PRCs that do not interfere with the analysis, such as perdeuterated PAHs, with low-to-moderate SPMD fuga-city, that have been added to the SPMD lipid prior to deployment, most of the field interactions discussed above can be corrected for. Currently, commercially available SPMDs (ExposMeter AB, Sweden) contain a PRC mixture of four deuterated PAH compounds, four C13-labeled PCB compounds and one chlorinated naphthalene.

SPMDs should be sheltered from direct sunlight. Orazio et al. [12] showed that PAHs spiked into SPMDs deployed in a protective cage at 1-m depth (clear pond) were not influenced by photolysis. However, PAHs in naked SPMDs (with no protective cage) deployed under the same conditions did undergo photolysis.

The potential for vandalism and theft should also be taken into account when selecting the deployment site. In shallow waters, the devices can be deployed under various shelters that may be present, e.g. bridges or trees, which can protect membranes from direct sunlight and make them less readily visible to people who may interfere with them. In deeper waters, SPMDs are usually attached to an anchor line and float, which can be kept just below the water surface. To illustrate the possible risks, in a monitoring project in Nicaragua POCIS were deployed at four locations and local people were hired for 3 weeks to guard them. During the deployment period, one sampler disappeared and the woman hired to guard this sampler went upstream and retrieved another sampler, with which she replaced the lost one. Luckily each device was numbered, so the replacement was discovered, the woman was not paid, but her actions resulted in the loss of data from two, instead of just one, sampling sites.

Due to the integrative nature of the sampling process, SPMDs can be deployed for sampling intervals ranging from days to months depending on the expected levels of contaminants and their properties. Generally, we have found deployments of 14-30 (routinely 21) days to be sufficient to sequester quantifiable levels of most environmentally relevant hydrophobic contaminants. During the 21 days in normal surface conditions, the sequestered amounts of only a few of the normally targeted lipophilic organic compounds exceed the linear sampling range, and thus the possibility for TWA concentration calculations. However, several factors should be taken into account when selecting an appropriate interval for integrative sampling with SPMDs, including the linear uptake time, the types of analytes targeted and the analytical sensitivity (i.e. method detection and method quantification limits) required, the time resolution needed for defining changes in waterborne chemical concentrations and environmental variables (e.g. flow-rate, temperature, expected level of biofouling, potential for vandalism or other damage to SPMDs).

In order to predict the time that SPMDs can be deployed while retaining the ability to calculate TWA values, the following equations can be used:

ln 0.5KswVs^ 0.693Kow Vs

ln 0.05KswVs 2.99Kow Vs n .

Rs Rs where t1/2 and t95 are the times required to reach 50% and 95% of the equilibrium concentrations, respectively, and Vs is SPMD volume. Returning to the previous example of a program for monitoring anthracene in a river with a flow rate of 50 cm s—1 and average temperature of 25°C, the following considerations would apply. The anthracene uptake rate by a standard SPMD is 4.7 L day-1 at 25°C and a flow rate of 50 cm s—1 [5], while the log Ksw, log Kow and Vs values are 4.67, 4.54 and 4.9 cm3, respectively. Using the above relationship [3], we can forecast that the t1/2 for anthracene would be 25 days, or in other words, uptake of anthracene by SPMDs should remain integrative for 25 days.

14.2.2 SPMD storage considerations

SPMDs exposed to air will concentrate/sample vapor-phase chemicals, therefore care must be taken to prevent their contamination during storage prior to, during and after the deployment. SPMDs must be stored in the vapor-tight, solvent-rinsed sealed metal cans provided by the supplier, and ideally should be kept frozen (< —15°C) until deployment. If PRCs are used in any of the SPMDs, the PRC-containing SPMDs must be kept separate from the others. Furthermore, although it may not be completely essential to transport SPMDs to the field at low temperatures (the SPMDs are in a clean atmosphere until the seal on the can is broken), it is always preferable to maintain them at freezing or near-freezing temperatures; especially SPMDs with PRCs, during transport to and from the sampling sites to minimize losses of the PRC compounds. A variety of coolants can be used for shipping SPMDs, including ice, blue ice and dry ice. However, some commercial cooling blocks may contain bactericides, e.g. triclosan, which may contaminate the samples during transport.

14.2.3 Precautions/procedures during deployment and retrieval of SPMDs

Sampling will start once the SPMD has been removed from its airtight can. Therefore, the deployment area should be examined for potential sources of contamination. Since there are many sources of vapor phase contaminants, including inter alia engine exhaust gases, gasoline, diesel fuel, oils, wheel dust, tars, paints, solvents and cigarette smoke. It is essential to avoid exposing the samplers to the atmosphere for longer than necessary before they are deployed in the water. Samplers containing PRCs should be protected from UV light (sunlight) during the handling procedure since just 1-2 min exposure can alter the PRC concentration. Hand lotions, cologne, perfume, powered gloves, etc. must not be used because these materials can contain target chemicals. After the SPMD has been deployed, the lids are resealed on the shipping cans and the empty cans are stored refrigerated until the SPMDs are retrieved, when the same cans should be re-used.

SPMDs are deployed in the field in deployment devices that minimize abrasion of the membrane in the turbulent environment, buffer external flow, protect them from mechanical damage caused by sharp items in the water and/or living organisms and minimize their exposure to sunlight. The commercially available stainless steel deployment canisters, such as the one shown in Fig. 14.5, hold two to five standard SPMDs mounted on spider racks. The perforated surface of the device permits adequate water exchange rates and the rack design prevents SPMD both from coming into contact with the canister walls and from self-adherence, which would reduce the diffusion surface area. The robust construction of the device allows membranes to be deployed in such highly turbulent environments as oceans, industrial pipes and rivers.

In highly polluted environments, when chemicals are visible on the surface of the water (for example oil films), it is recommended to minimize their exposure to the surface when the sampler is submerged.

Fig. 14.5. A commercially available stainless steel deployment canister that holds a maximum of two SPMD racks and an airtight tin can for storage of membranes.

When the SPMDs are retrieved their exposure to sunlight must also be minimized, since some target analytes may be rapidly photodegraded.

Experienced or trained technicians should be assigned the tasks of deploying and retrieving SPMDs in order to avoid questions such as the one I was asked by a consultant, ''I just received your membranes and took them out from the cans to examine them in my office. How do I mount them on the rack?'' Personnel with accreditation based on the BSI PAS 61:2006 standardization protocol should be used if possible [13].

As mentioned previously in this chapter, environmental conditions affect the sampling rates of SPMDs, so data on relevant variables such as temperature (ideally obtained using temperature loggers that can measure and record temperatures every 10 min during the deployment, down to 200-m depth) should be acquired; a minimum requirement is to record the temperature at the beginning and end of the exposure. Flow rates and the extent of biofouling on the membrane surface should also be recorded, even if PRCs are used in the membranes to provide compensatory factors for the influence of environmental variables. Sometimes it is useful to take notes on additional phenomena such as visible discoloration of the triolein, visible damage to the membrane and the ''feel'' of the surface of the membrane (e.g. if it feels ''fatty''). Johnson et al. [14] performed a study on hazard assessment of oil spills and they noticed that extracts from membranes retrieved during the first 2 weeks had distinctive chocolate colors. However, the colored residues were no longer visible in the extracts from SPMDs retrieved in weeks 4, 6, 8 and 12.

Global positioning systems (GPS) devices are recommended for identifying SPMD deployment sites. Following retrieval from the exposure medium, SPMDs should be immediately dried with tissue paper, sealed inside the same labeled metal cans and transported (frozen or near frozen) back to the analytical laboratory in a cooler. Biofouling or layers of dirt must be carefully removed from the SPMDs directly after retrieval of the membrane, using tissue paper and maybe gentle wetting with hexane-extracted water. Removing salty water is also important since it will cause corrosion of the inside of the tin can, which can cause increased numbers of particles in the can and thus reduce the measured amounts of target compounds inside the membranes. Marl deposits within the biofilm may lower the dialytical recovery unless an acid wash is included in the precleaning procedure. Harsh cleaning should always be avoided since it can harm the membrane, especially if calcitric biofouling has occurred, since the calcitric deposits may tear the membranes during handling. Small losses of triolein from a membrane during sampling or after cleaning may be detectable only if membranes with an appropriate PRC (e.g. octachloronaphthalene) have been used. However, large losses of triolein can also be measured gravimetrically.

If it is necessary to delay the shipping of exposed SPMDs more than a few hours they should be stored frozen at — 15°C in their sealed metal cans. Failure to maintain exposed SPMDs under freezing conditions can result in losses of analytes with relatively high fugacity (e.g. naphthalene). However, no measurable losses of 2,4,5-trichlorophenol (which has high fugacity from SPMDs at room temperature) were observed from SPMDs stored at -15°C for 6 months in sealed cans in a study by Huckins et al. [2].

0 0

Post a comment