reaction mass of 2,2,3,3,5,5,6,6-octafluoro-4-(1,1,1,2,3,3,3-heptafluoropropan-2-yl)morpholine and 2,2,3,3,5,5,6,6-octafluoro-4-(heptafluoropropyl)morpholine

Bioaccumulation: aquatic / sediment

Additional information

It is not expected that FC-770 will have a significant presence in the aquatic compartment, and testing conducted under artificial conditions that force FC-770 to remain in solution would not provide a realistic assessment of bioaccumulation potential. A pilot BCF study demonstrated that testing under artificial conditions of closed test vessels with a high rate of flow-through was not adequate to maintain consistent test concentrations that were close to the nominal concentration, yet at the same time these test conditions caused extensive mortality and morbidity in fish. The test methodology was developed in response to Substance Evaluation decision SEV-D-2114434716-46-01/F, which mandated a BCF test under flow-through conditions that would maintain stable test concentrations of FC-770 in water. For both the uptake and the depuration data there were large variabilities noted in the fish concentrations from the same test chamber (i.e., the same time point), as well as swings in the tissue concentrations (e.g, day 21 to day 27 or day 27 to day 28). Both of these aspects of the fish concentration data contributed to the large confidence interval associated with the calculated provisional kinetic BCF. In addition, numerous issues were identified with this pilot study that, when combined, could lead to a substantially lower BCF.  The issues are documented as follows:

a.       The sampling method was open to the atmosphere.  In this preliminary test, no attempt was made to establish recovery from samples.  Volatilization losses of FC-770 during sampling cannot be excluded.  Such losses might explain the low observed test substance concentrations relative to the nominal (see below) in closed-bottle trials. We anticipate that the actual concentrations in the tank could have been much closer to nominal than the measured values, and therefore the true BCF could be much lower than the provisional kinetic BCF.

b.       The measured concentrations in this pilot study as well as the previous diluter trials was significantly lower than the nominal concentration by a factor of 7. FC-770 loss could have occurred via leaks in the test chambers, during daily feeding and cleaning operations while the test chambers were temporarily open, or through volatilization loss during sampling. Despite increased nominal concentrations in the pilot study relative to the preliminary trial (100 µg/L vs 60 µg/L), the measured concentration was actually lower in the pilot study relative to the preliminary trial (mean 9.3 µg/L in the pre-test equilibration period vs 19.9 µg/L in the preliminary trial). This difference demonstrates the difficulty of establishing exposure concentration for a substance that is both poorly soluble as well as volatile. It is worth noting also that the observed variability in concentration would have likely presented an issue if the test material had been radio labeled.

c.       Measured exposure concentration was quite variable in some cases even within the same test chamber (e.g., %SD between top middle and bottom samples ranged from 4.94% on day 1 to 70.7% on day 28), and across all the bottles (%SD of 58.94% across all measurements taken). Concentrations appeared to decrease over time with the exception of day 21. The reason for the apparent buildup and the observed variability are not known. It is possible that differences in organic carbon accumulation in each test system could have contributed to variability in measured concentration across the test chambers. Unfortunately, DOC/TOC values were not measured in the chamber during the duration of this pilot study, but only in the influent.

d.       Abnormalities (e.g., popeye) and fish mortality were present in both the solvent control and the test chambers.  Mortality was 33% and 83% in the control chambers and ranged from 0 to 67 % in the test chambers. The earliest mortality was observed at day 3. Mortality did not associate with exposure concentrations or observed morphological stress. In addition, mortality was not observed by the contract lab in a standard BCF study of an unrelated test material that was run with the same batch of fish at the same time as the pilot study.  The underlying mechanism for the observed mortality in both the solvent control and treatment replicates cannot be definitively identified based on the measurements and observations collected during the study but possible reasons include: (A) fish crowding during feeding-contact and competition for food stress at the small opening of the aspirator bottle test chambers (as a result, feeding procedure was changed on day 18 – the water level was lowered to increase the feeding area for the fish), (B) accumulation of fish waste and excess food in systems resulting in reduced water quality, (C) changes in daily water volumes during feeding after day 18, (D) changes in water quality concentrations and/or (E) high flow rates.

e.       High microbial growth in test chambers was noted in the test chambers. This prompted the laboratory to perform daily cleaning of the test chambers. Flow was stopped and medium drawn down via the stopcock.  Sides and bottom of the bottle were scraped and siphoned to dislodge biomass and remove uneaten food and debris, after which the flow was restarted, and medium level restored.  The process required ca. 30 minutes each day during which the test chambers were open to the atmosphere. The reason for the high microbial growth could not be precisely identified but possible explanations include 1) the use of solvent which could have acted as a carbon source for microbial activity 2) the glass equipment used could have been more conducive to microbial growth and more difficult to clean compared to other Teflon-lined equipment used regularly in the lab.

f.       The lack of a premixing chamber was a limitation of this study. The large variability in concentrations between the test chambers suggest that the fish may have been exposed as separate microcosms as opposed to one consistent exposure group. The use of a premixing chamber was avoided as a potential source of volatilization losses.

g.       High uncertainty in the measured BCF. The uncertainty in the water concentration measurements, the observed variability in water concentration, as well as the noted signs of fish stress all affect the reliability of the derived BCF value.

In the Substance Evaluation decision, a bioaccumulation feeding study was mandated if it were concluded that the flow-through exposure study was determined to be not feasible. For similar reasons to the flow-through test, however, a feeding study is not likely to be feasible. The pilot BCF result is used as such in the PBT Analysis.