Gobas et al. (2009) defined a BCF as the ratio of the steady-state chemical concentrations in an aquatic water-respiring organism (CB, g chemical/kg wet weight) and the water (CW, g chemical/L) determined in a controlled laboratory experiment in which the test organisms are exposed to a chemical in the water (but not in the diet) (units: L/kg wet weight).
It is now accepted that to standardize measurements of bioaccumulation for easy comparison, metrics such as BCF, BAF, and BMF should be normalized to a known lipid content. However, definitions given by Gobas et al. (2009) for BCF, BAF, and BMF seem to indicate (via the units presented as wet weight) that data should be used as whole-body concentrations in the framework. The majority of data discussed in this section and the Biomagnification factors section is based on lipid normalized values reported in the literature but for comparative reasons, where possible, the data has also been transformed into whole-body concentrations.
Before 2000, a number of studies had reported BCFs in fish for MCCPs, all of which were below 1000 (EU 2005). However, these studies were criticized on the grounds that the exposure (water) concentrations used were in excess of the solubility of the substance and therefore may have overestimated exposure and thus underestimated the BCF. Furthermore, most of the studies preceded, and therefore were not consistent with, standardized procedures (e.g., OECD Guideline 305). In response to these criticisms, and as part of the European Risk Assessment Regulation an OECD-compliant study was carried out (Thompson et al. 2000) to determine the BCF in rainbow trout (Oncorhynchus mykiss), exposed to a 51% chlorinated MCCP, monitored using 14C-radiolabeled chlorinated n-pentadecane (≈C15H26Cl6). The exposure period was 35 days followed by 42 days depuration.
The maximum BCF reported was 1100 at a nominal exposure concentration of 1.0 μg/L (mean measured 0.93 μg/L), calculated by the kinetic method (whole-body, extrapolated to steady-state). Because this was based on radiolabeled residues and may therefore have included metabolites of the parent substance, this value can be considered as a worst-case estimate. The study was carried out to the then current OECD test guideline and no growth correction or lipid normalization was carried out (as now recommended; OECD 2012). Further analysis of the original data for growth correction and lipid normalization to the now recommended 5% (OECD 2012) gives a BCF of 1000. (Lipid normalization was calculated using a representative 10% fish lipid content [estimated from in-house data] because no lipid measurements were determined in the original study.) The resulting BCF of 1000 is lower than the BCF criterion specified in the current regulatory schemes for the EU (BCF 2000), Canada, and the United Nations Environment Program (BCF 5000) and lower than the indicator for “B status possible” (BCF 5000) recommended by Gobas et al. (2009).
A further study has been conducted to determine the BCF in rainbow trout (O. mykiss) of a lower single-chain length material with a relatively low chlorination level (45% chlorinated [14C]-n-tetradecane; ≈C14H25.5Cl4.5) (Vaughan and Hurd 2010). The exposure period was 35 days followed by 42 days depuration. A whole-body BCF of 6600 was reported and a maximum BCF obtained of 9100 at a nominal exposure concentration of 0.50 μg/L (mean measured 0.34 μg/L), calculated by the kinetic method. Further manipulation of the data to conform with the 2012 OECD 305 test guideline recommendations gives a lipid normalized growth corrected kinetic BCF of 15 000, well in excess of the 5000 B status possible recommended by Gobas et al. (2009). As for the previous study, this study was based on the analysis of radiolabeled residues and may therefore have included metabolites of the parent substance that may have been incorporated into lipid or protein tissue in the fish. At the end of the study, analysis of fish taken from the end of the depuration phase for parent compound indicated significant metabolism, with a mean metabolite percentage of 21% reported (range 17%–30%, nonextractable residues considered as bound metabolites in fish tissue) (Leonards and van Beuzekom 2010), although the mean percentage of metabolites observed may also have been underestimated. Fish samples were only available for analysis at the end of the depuration period, therefore metabolites and parent compound may have been excreted. Because the potential metabolites are likely to be more polar and may have different depuration rates to the parent compound, the ratio of parent–metabolite may have been different at the end of depuration compared to the end of uptake. The level of metabolites may therefore have been underestimated if they had depurated at a faster rate than the parent compound, see “Metabolism of chlorinated paraffins” for further discussion.
A study of the bioconcentration in carp (Cyprinus carpio) of a 49% chlorinated tridecane (C13H23.2Cl4.8), separately monitoring Cl5 (50% Cl), Cl6 (54% Cl), and Cl7 (58% Cl) components in water and fish using high performance liquid chromatography mass spectrometry (LC/MS) has been reported. The fish were exposed to nominal concentrations of 1.0 and 10 μg/L of the test substance for 62 days. Preliminary data (UNEP 2009) provides whole-body steady-state BCFs of 1700, 2000, and 2800 for Cl5, Cl6, and Cl7, respectively at the higher exposure concentration (1500, 1600, and 2300, respectively, at the lower exposure concentration). The similarity of the BCFs at the different exposure levels indicates that the dosing system had achieved full dissolution of the components at both concentrations. In addition Cl4, Cl8, and Cl9 components were monitored, but the levels were considered too low to provide accurate BCFs; approximate values after 62 days were 300, 2600, and 1800, respectively for the higher exposure concentration. Depuration was also monitored but the data have not been reported. These data suggest that bioconcentration increases with increasing chlorination of the tridecane, up to Cl7 (58% Cl), with a possible slight decline thereafter. The increase cannot be attributed to increasing hydrophobicity, because the authors also reported that measured solubilities of the congeners increased with increasing chlorination −50, 70, and 90 μg/L for Cl5, Cl6, and Cl7, respectively. Therefore, the increasing BCF might suggest that metabolism of the components of the parent substance was decreasing with increasing chlorination.
All available whole-body fish BCFs obtained for representative single-chain length materials over the range C10 to C17 using procedures compatible with OECD Guideline 305, are plotted against chain length in Figure 1. The plot shows percentage chlorination level and distinguishes between analysis based on the radiolabel (14C) or parent compound. Results are shown for studies previously mentioned above and a further study on a C11 58% Cl material (Madeley and Maddock 1983b).
Figure 1. Plot showing all available whole-body fish bioconcentration factors obtained for representative single chain length materials over the range C10 to C17 plotted against C number. The plot shows the percentage chlorination level and distinguishes between analysis based on the radiolabel (14C) or parent compound. Results are shown for studies previously mentioned above and a further study on a C11 58% Cl material (Madeley and Maddock 1983b).
Download figure to PowerPoint
The recent BCF for 45% chlorinated tetradecane (C14) appears to be unusually high compared with those obtained for C13 and C15 substances, which might suggest that substantial metabolism occurs below a “critical” level of chlorination. In that case, determinations by radiochemistry would increasingly overestimate the BCF with decreasing chlorination level. The potential for metabolism of chlorinated paraffins is discussed further later.
Defined as the ratio of the steady-state chemical concentrations in an aquatic water-respiring organism (CB, g chemical/kg wet weight) and the water (CW, g chemical/L) determined from field data in which sampled organisms are exposed to a chemical in the water and in their diet (units: L/kg wet weight).
Field data relevant to a BAF assessment of C14–17 chlorinated alkanes, based on this definition, are provided by Houde et al. (2008).
Houde et al. (2008) report concentrations of C14–17 chlorinated alkanes in biota and water samples from Lake Ontario and Lake Michigan. Bioaccumulation factors are calculated between water and 4 species of fish and plankton from Lake Ontario, but are expressed as lipid normalized values. As described above, these values have also been corrected to whole-body (wet weight) units, as specified in the framework definition, using the mean lipid content of the samples as given in the article.
Mean whole-body (total C14–17 chlorinated alkanes) BAFs range from 35 000 L/kg wet weight (lipid normalized Log BAF = 6.5) for water-Diporeia to 958 000 L/kg wet weight (lipid normalized Log BAF = 7.3) for water-sculpin (Cottus cognatus). These values are considerably in excess of the framework criterion of 5000 indicative of B status possible (Gobas et al. 2009), and markedly different from the laboratory BCF data discussed above. However, it should be noted that these BAFs are based on water samples taken at different locations and at different times to the biota samples. Thus, it is possible that the biota concentrations reflect exposure to temporally or spatially varying ambient levels, in particular in relation to local sources of contamination. Indeed, Houde et al. (2008) (Supplemental Data) show that the measured water concentration of C14–17 chlorinated alkanes varied from 47 pg/L in 2002 to <0.5 pg/L in 2004, at the same sample location. Furthermore, the biota were sampled from the western, industrialized (Toronto, Hamilton, etc.) end of Lake Ontario, whereas the water samples were taken in the central and eastern areas of the lake. Each fish species value was based on only 2 samples (except for lake trout [Salvelinus namaycush], 7 samples) and the water concentration was based on only 4 samples of which 2 were below the detection limit. Therefore, the results quoted in Houde et al. (2008) do not fully meet acceptance criteria given for field studies. Parkerton et al. (2008) stated that guidance for the conduct of field bioaccumulation studies is lacking, and further efforts are needed to describe best practices for field investigations. Field samples should sufficiently represent the habitat and exposure regime of the selected organism, and samples should be taken at the same locations over a period of time (Weisbrod et al. 2009). Further acceptance criteria for field measurements are given in Burkhard et al. (2011).