Paraffin waxes and Hydrocarbon waxes, chloro

Long-term toxicity to aquatic invertebrates

Additional information

Chronic Daphnia Toxicity

There are several 21-day reproduction studies inDaphnia magnaon liquid LCCPs. The most recent and best documented of these is the study by Sharpe and Penwell (2007), which tested a¹⁴C-labelled, C₂₅, 43% wt Cl chlorinated alkane (a single chemical not a commercial LCCP mixture) according to the OECD 211 method using a semi-static test procedure. In this study, a stock solution of the test substance was prepared in dimethylformamide and aliquots of this were added to Elendt’s M4 medium for use in the test (the amount of dimethylformamide in the test solution was 0.1 ml/L). The test medium had a hardness of 234 mg/L as CaCO₃. Test solutions were renewed every 48 hours during the test. The exposures were carried out using borosilicate glass beakers each containing 1 organism and 80 ml of test solution. A total of ten replicates per treatment group were carried out. The nominal concentrations tested were 2.0, 4.0, 8.0, 16, 32 and 64 μg/L and a dilution water control and solvent control group were also run. The actual concentration of chlorinated paraffin present was determined in “fresh” and “spent” solution at days 0-2, 4‑6, 10-12 and 18-20. The analytical method used was radiochemical analysis (direct scintillation counting of the test solution). The measured concentrations in “fresh” solutions ranged from 85% to 105% of the nominal values and the measured concentrations in the “spent” solutions ranged from 50% to 84% of the nominal values. The mean measured concentrations found in the six treatment groups were 1.6, 3.0, 5.4, 13, 27 and 55 μg/L respectively. It should be noted that this analytical method would not necessarily distinguish between truly dissolved chlorinated paraffin and non-dissolved chlorinated paraffin.  The endpoints determined in the study were adult length, reproduction (number of live offspring produced per parent animal surviving to the end of the study) and adult mortality. No statistically significant differences (at the p=0.05 level) were seen between the treatment groups and the control groups for either adult length or reproduction, as such the NOEC = 55 μg/L.

 

Chronic Toxicity to Aquatic Invertebrates

Substance	Species	Exposure	Value	Rel.	Reference
C18-20(liquid)
C18-20, 52% wt. Cl	Daphnia magna	21 days	No statistically significant effects observed at 2 μg/L (NOEC) (measured; based on re-analysis by Thompson)	2	Hooftman and Henzen/TNO 1993 Thompson 2007.
C18-20; 52% wt. Cl	Daphnia magna	21 days	~29 – 32 ug/L (NOEC) (measured)	2	Frank (1993); Frank and Steinhäuser (1994); (Thompson, 2001)
C20-30(liquid)
C25; 43% Cl	Daphnia magna	21 days	NOEC = 55 ug/L (measured) LOEC > 55ug/L (measured) EC50 > 55 ug/L (measured)	1	Sharpe & Penwell (2007)
C22-26; 43% Cl	Mytilus edulis	60 days	NOEC: 2.18 mg/L (measured) LOEC: 2. 18mg/L (measured)	2	Madeley & Thompson (1983b)
C20-30(solid)
C20-30; 70% Cl	Mytilus edulis	60 days	NOEC: 1.33 mg/L (measured) LOEC: 1.33mg/L (measured)	2	Madeley & Thompson (1983a)

 

 

A 21-day study ofC_18-20; 52-56% ClinDaphnia magnawas carried out by TNO (Hooftman and Henzen 1993). The substance tested was a commercial product containing 1.5% epoxidised soybean oil. The test was carried out according to the OECD 202 methodology using a semi-static test procedure (test solution renewal was carried out every 48-72 hours). The dilution water used was a synthetic medium (DSWL) prepared by addition of various salts to ground water. The hardness of the medium was 214 mg/L as CaCO₃. The study tested a saturated solution, a blank control (DSWL medium only) and a column control (DSWL medium pumped through a column containing unspiked chromosorb 60/80 mesh).  Four replicates (10 daphnids in 400 ml of test solution) were carried out for each treatment. The parent survival in both of the control groups was 97.5%. In the C_18-20, 52% wt. Cl treatment group the parent survival was 92.5%. This was not statistically significantly different (at the p=0.05 level) from the control groups. Therefore it was concluded that no treatment-related effects on parent morality were seen in the study. The reproduction rate (expressed as the cumulative number of young per living female) in the study was 113.6 in the blank control and 127.3 in the column control. The response of the two controls was not statistically significantly different (at the p=0.05 level). The reproduction rate in the C_18-20, 52% wt. Cl treatment group was 100.8. This was 88.7% of the blank control response and 79.1% of the column control response. These responses were analysed statistically in TNO (Hooftman and Henzen 1993) using the two-tailed Dunnett test with a 95% significance level. No statistically significant differences were found between the C_18-20, 52% wt. Cl treatment group and the blank control but it is not entirely clear from the report whether the reduction in the reproduction rate in the treatment group compared to the column control was statistically significant or not. In various parts of the report (including the overall summary) it is stated that this difference was statistically significant, but in other places there are statements that no statistically significant differences were found.

 

In order to investigate these results further, a re-analysis of the data from this study was developed by Thompson (2007). The statistical analysis was carried out on the same dataset as summarised in the TNO (Hooftman and Henzen 1993) study and incorporated the dataset for a medium-chain chlorinated paraffin (C_14-1752% wt Cl) that was also tested and reported in the same publication as the C_18-20, 52% wt. Cl substance. The data were found to be normally distributed and the variances were homogenous, satisfying the requirements for the use of ANOVA (with Dunnett’s procedure to maintain a family error rate of 0.05) to determine significant differences from the column control. Using this procedure, no significant difference was seen between the reproduction rate in the C_18-20, 52% wt. Cl treatment group (or the medium-chain chlorinated paraffin treatment group) and the column control. In addition, a further analysis was carried out comparing the treatment groups with the pooled control group (the combined blank control and column control) and this again showed no statistically significant differences for the C_18-20, 52% wt. Cl treatment group.

 

Overall, the results of the TNO (1993) study indicate that although a reduction in reproduction rate occurred in the study compared with the control group, the statistical significance of this reduction is marginal, and depends on the exact method used to analyse the data. (If the data for the MCCP are included in the analysis then there is no significance difference. If, however, the MCCP data are omitted from the analysis a statistically significant difference is observed between the LCCP treatment group and the column control, and combined controls, but not with the blank control alone). Either way the study can be interpreted as showing a NOEC (or a response close to a NOEC) at a concentration of 2 μg/L; the limit dose of the study. As this is a single value, dose response could not be determined from this study. 

 

Frank (1993) conducted a 21‑day reproduction study using dilutions of the water soluble fractions of two stock solutions (100mg/L and 10 g/L loading rates) of a C_18-2052% Cl LCCP. The 10 g/L stock solution equated to an LCCP concentration of 462 – 519 ug/L, while no LCCP was detected in the 100 mg/L stock solution (ie concentration <20 ug/L). The dilutions used were 1:2, 1:4, 1:8 and 1:16 for the 100 mg/L loading and 1:4, 1:8, 1:16, 1:32 and 1:64 for the 10 g/L loading. In these experiments, the test medium was changed 3 times/week and 10 animals per concentration were used. The tests were carried out at 20°C and a pH of 7.79‑8.44. Two end‑points were determined in the study: effects on parent mortality and effects on reproduction (number of offspring/adult). Parent mortality in the controls was 0% in the test carried out with the 10 g/L nominal stock solution and 10% in the test carried out with the 100 mg/L nominal stock solution. Elevated mortality was seen in the exposed populations: for the 10 g/L stock solution the LOEC was determined as the 1:8 dilution (approximately 58‑65 µg/L) and the NOEC was determined as the 1:16 dilution (approximately 29‑32 µg/L); for the 100 mg/L nominal stock solution the LOEC was determined as the 1:4 dilution and the NOEC was determined as the 1:8 dilution (concentrations were measured by AOX (adsorbable organic halogen) analysis; based on the detection limit for the analysis of the stock solution, these dilutions equate to LOEC and NOEC of <5 and <2.5 µg/L respectively). From the dose response curves it appears that 100% parent mortality occurred at concentrations of around <10 µg/L in the 100 mg/L nominal stock solution experiments and around 125 µg/L in the 10 g/L nominal stock solution experiments.

 

For the reproduction end point, the average number of young per adult in controls was 72.3 (variability 17.8%) in the 100 mg/L nominal stock solution series of experiments and 73.5 (variability 6.2%) in the 10 g/L nominal stock solution experiments. A significant reduction in the number of young per adult was seen in some of the exposed organisms. For the 100 mg/L nominal stock solution, this effect on reproduction was significantly different from the control groups at the lowest concentration tested (a 1:16 dilution which is equivalent to a chlorinated paraffin concentration of <1.2 µg/L, based on the detection limit of the analytical method used). Thus the NOEC/LOEC for this series of experiments was <1.2 µg/L. Similarly, for the 10 g/L nominal stock solution effects were again seen at the lowest concentration tested (a 1:64 dilution, which is equivalent to a chlorinated paraffin concentration of 7.3‑8.1 µg/L). This value was treated as the LOEC for this series of experiments.  The report also indicates that the NOEC is very close to this value since using a different statistical method, the Dunnett Test rather than the Williams Test, the effects seen at this concentration were not statistically significantly different from controls.

 

The same 21‑day results were reported by Frank and Steinhäuser (1994). In addition, this paper also reports the results of a similar test (and an acute study) carried out using the water soluble fraction from a 1 mg/L nominal stock solution, with no analytical measurement during these additional experiments. This stock solution appears to have been prepared in a different manner from those previously. A solution of the chlorinated paraffin in pentane was prepared and a portion of this was added to the water, followed by stirring for 1 hour at 40°C to evaporate the solvent, prior to filtration. This showed no effects on parent mortality or reproduction in the test. It was not possible to determine the concentration of AOX (and hence chlorinated paraffins) present in these solutions, and so the actual exposure concentration is unknown.

 

There are a number of significant deficiencies in the studies by Frank (1993) and Frank and Steinhäuser (1994) that make these data very difficult to interpret in terms of the effects of the C_18‑20liquid chlorinated paraffins ( EA 2009). However, the raw data from the 21-day Frank (1993) and the Frank and Steinhäuser (1994) studies have been obtained andreanalysed[D1] (Thompson, 2001) and discussed at length in the Environmental Risk Assessment of LCCPs (EA 2009). Overall, a 21-day NOECof approximately 29‑32 µg/L can be determined from the 1:16 dilution of the 10 g/L stock solution, based on parent mortality.The analysis of these data suggest that while there are deficiencies, they were useful in developing the predicted no effect concentration (PNEC) of 2.9 μg/L for C_18-20LCCPs ( EA 2009). 

 Chronic Toxicity to Mussel

 

The toxicity of two LCCPs to the common mussel (Mytilus edulis) has been determined over 60 days (Madeley and Thompson, 1983a and 1983b). The compounds tested were the same C_22‑26, 43% wt. Cl and a C_>20, 70% wt. Cl products (mixed with a small amount of radiolabelled n‑pentacosane that had been chlorinated to a similar degree) as used in the 60‑day rainbow trout study by Madeley and Thompson (1983b). The test used a flow‑through system with filtered natural seawater (salinity 35‰, pH 8.0‑8.3, dissolved oxygen 6.1‑8.25 mg/L) at 15°C. Acetone was present in the test solutions at a concentration of 0.5 ml/L (500 ppm), and acetone controls were also run using this concentration. Groups of 30 mussels were used for each exposure concentration; the mussels were fed during the test (algal cells were added to the in‑flowing water at a rate of 1.0‑1.1x10⁹cells/day).

 

For the C_22‑26, 43% wt. Cl substance two nominal concentrations of 0.32 mg/L and 3.2 mg/L were tested. The mean measured concentrations for these two exposures over the duration of the test were determined as 0.12 mg/L and 2.18 mg/L respectively by¹⁴C‑analysis. Measurements by parent compound analysis (using a TLC method) at various times during the test were in general agreement with these values. The 2.18 mg/L exposure solution was reported to be cloudy in appearance, with a fine white deposit being observed, indicating that the solubility of the substance in the test system had been exceeded. No significant mortality occurred in the test. Reduced filter feeding activity compared to control populations was consistently observed in the 2.18 mg/L exposure group from the 7th day onwards, and it was thought that this may reflect a response to the "particulate" nature of the test solution.

 

For the C_>20, 70% wt. Cl substance, the two nominal concentrations tested were 0.56 mg/L and 1.8 mg/L. The mean measured concentrations at these two exposures were determined as 0.46 mg/L and 1.33 mg/L by¹⁴C‑measurements. Occasional parent compound analyses were in general agreement with these values. Some deposition of the test substance was noted at the highest concentration, indicating that the solubility of the substance in the test system had been exceeded. No mortality as a result of exposure to the chlorinated paraffin was seen in the test. Again, reduced filtration activity was seen in the 1.33 mg/L exposure group when compared with the control populations, particularly during the second half of the exposure period. However, this observation was variable, and on a number of occasions the filter feeding activity was comparable to controls and so this reduction in activity was considered to be minimal and tentative.

 [D1]Need to get and add here. Somewhere in the archive