Registration Dossier

Environmental fate & pathways

Endpoint summary

Administrative data

Description of key information

Additional information

As per the introductory paragraphs to Annex XIII of REACH, on PBT Analysis, information used purposes of assessment of PBT/vPvB properties (including bioaccumulative properties) shall be based on data collected under relevant conditions. FC-770 is of very low water solubility (66 µg/L at 23 °C) and is very volatile (vapor pressure, 6.74 kPa at 20 °C). The extreme hydrophobicity of this molecule results in an experimental Henry's Law constant (expressed as the ratio of vapor phase partial pressure of FC-770 over aqueous phase concentration) of 1030 atm.m³/mol at 22 °C. A value for log Kow of 5.7 has been calculated based on water solubility and an experimentally-determined octanol solubility (32.3 g/L at 23 °C). However, it is not expected that FC-770 will have a significant presence in the aquatic compartment. For example, in exposure modeling expressing steady-state concentrations of FC-770 (i.e., as time approaches infinity), aquatic concentrations of FC-770 were in the range of 10 femtograms per liter, which is several orders of magnitude below the limit quantitation. A test conducted using a relevant environmental condition would therefore be impossible to conduct. The current registration update limits FC-770 uses to closed systems and articles. None of the registered uses of FC-770 have any inherent releases to water. All emissions are expected to be into the atmospheric compartment. FC-770 is not expected to partition from the atmosphere to surface waters. This conclusion is confirmed by Level III EQC modeling, in which transfer half-lives from the atmospheric compartment to water were several orders of magnitude longer than from other compartments to the atmosphere. Upon direct release of FC-770 to the aquatic compartment, the chemical would be expected to volatilize rapidly to the atmosphere. However, as noted FC-770 release under registered uses is entirely to the atmosphere. Therefore no exposure to aquatic organisms is expected.

Nevertheless, an aquatic bioaccumulation study was ordered under Substance Evaluation decision SEV-D-2114434716-46-01/F. In this decision , ECHA concluded that mass distribution may occur to all environmental compartments, especially if releases were to the aquatic compartment (although, as noted above, such release is not expected).  Further, ECHA expected that FC-770 would occur in the condensed state, and that distribution seems to be highly dependent on the environmental release pattern.  However model results depend crucially on physical/chemical input parameters.  By default, EpiSuite uses modeled distribution parameters (Henry’s Law constant, vapor pressure, log Kow, etc) to parameterize the Level III Mackay fugacity model.  The model results used in the SEV do not correct these defaults and are therefore not representative of the actual case. When the experimentally-determined parameters are used instead of modeled parameters, the distribution to soil, water and sediment predicted upon release to air becomes negligible. Distribution due to release to air was 90.3% to air, 0.2% to water, 8.9% to soil, and 0.6% to sediment when using modeled distribution parameters.  When experimentally-derived values are used to parameterize the fugacity model, distribution is 100.0 % to air, 5.1 x 10-6 % to water, 3.8 x 10-3 % to soil, and 1.8 x 10-5 % to sediment.

Persuant to the SEV, a pilot BCF study was conducted. A provisional BCF of 9585 was measured. The result is being carried forward to the PBT Analysis under REACH despite its shortcomings.

A log Koc value of 4.71 was calculated based on the laboratory-derived Kow value. However, there are no releases to soil expected from the uses of FC-770. Based on the predicted atmospheric fate, it is not expected that there will be significant exposure to sediment or soil organisms from atmospheric concentrations of FC-770. It should be noted that Koc is not the relevant quantity for partitioning from air to surfaces, and that Koa and/or Kp provide a better prediction of distribution of volatile substances to surfaces (where Koa is the octanol-air distribution coefficient and Kp is the general distribution coefficient of a substance from air to surfaces). FC-770 will not partition to soil or sediment from the atmosphere.

The log octanol-air partition coefficient (log Koa) of FC-770 was calculated and used to address potential exposure to terrestrial organisms from the atmosphere. As per Kelly and Gobas (1), chemicals with log Koa < 4 do not biomagnify regardless of the Kow because of efficient elimination via air exhalation. REACH Technical Guidance Chapter R.7c.10.3.4 pg 28 states that “…biomagnification (is) predicted to occur in many mammals at a log Koa above 5.” FC-770 has a log Koa of 1.47, and thus would not be expected to bioaccumulate in air-breathing organisms. Further evidence on terrestrial bioaccumulation is available from mammalian studies. According to REACH Technical Guidance Chapter R.11.4.1.2, p. 53:

“If chronic toxicity studies with mammals are available, the complete absence of effects in the long-term is an indication that the compound is either chronically non-toxic and/or that it is not taken up to a significant extent. Although this is only indirect information on the uptake of a substance, it may be used together with other indicators, e.g. referring to non-testing information, to conclude in a weight of evidence approach that a substance is likely to be not B or vB.”

Results from a 28-day oral study of FC-770 in the rat (no effects were seen up to 1000 mg/kg bw by oral gavage, the highest dose used in the study), a reproduction study of FC-770 in the rat (no effects seen up to 1000 mg/kg bw by oral gavage), and a 90-day repeat dose inhalation study (using perfluoromethyl morpholine, a highly similar structural analog that is being submitted as a read-across value) with the rat (NOAEL of 49,821 ppm) suggest that the bioaccumulation potential of this substance is negligible in mammals.

Furthermore, pharmacokinetic data obtained from two rat studies demonstrate an absence of FC-770 uptake into the blood which precludes FC-770 from entering systemic circulation and distribution to internal organs. There was no quantifiable FC-770 in serum, liver, or urine samples from these two studies. FC-770 was recovered only in the feces, providing direct evidence that administered FC-770 was not absorbed.

FC-770 has not been found to be metabolized in vivo. Analysis of urine samples after oral doses of up to 1000 mg/kg bw found no quantifiable parent FC-770, no quantifiable FC-770 isomers, no quantifiable FC-770 morpholino ring-related metabolites, and no quantifiable other fluorochemical metabolite species of any kind derived from the FC-770 mixture.

The pharmacokinetic data is direct evidence that supports the conclusion that FC-770 is not bioaccumulative. The argument that all perfluorinated organic chemicals bioaccumulate by the mechanism of binding to plasma proteins, rather than a lipid partitioning mechanism, is not supported by current pharmacokinetic principles. This is a fundamental misconception, as it has been widely demonstrated that plasma protein binding is based upon the polarity of the molecule in question, with acidic molecules binding to plasma albumin, alkaline molecules bind to plasma α1-acid glycoprotein, and neutral molecules demonstrating low protein binding (2).  Thus, it is not appropriate to draw an equivalence between the acidic molecule perfluorooctanesulfonic acid and a neutral, non-polar substance such as FC-770 with respect to plasma protein binding.  In reality, the binding of a molecule to plasma albumin is potentiated by the molecule’s polar or ionic head groups to specific polar or ionic amino acid residues in or near the plasma protein's hydrophobic pocket (3).  As FC-770 lacks any strongly polar functionalities, binding to plasma proteins would not be expected, would not be similar to that of perfluorooctanesulfonic acid, and thus is not expected to bioaccumulate by plasma protein binding.

Based on the mammalian pharmacokinetic and toxicity studies and physical property data such as Henry’s Law Constant and log Koa, FC-770 is not expected to be biomagnified in the top of the food chain. The provisional BCF can be interpreted as indicating that FC-770 has bioaccumulative properties in aquatic organisms under unrealistic conditions.

References

1) Kelly, B.C. and F.A.P.C. Gobas. 2003. An Arctic Terrestrial Food-Chain Bioaccumulation Model for Persistent Organic Pollutants. Environ. Sci. Technol Vol. 37, No. 13, pp 2966–2974.

2) Leslie Benet, Deanna Kroetz, and Lewis Sheiner.  Pharmocokinetics The Dynamics of Drug Absorption, Distribution, and Elimination in Goodman & Gilman’s The Pharmacological Basis of Therapeutics edited by Joel Harman and Lee Limbird, pages 3-27.  Ninth Edition, New York: McGraw-Hill, 1996.

3) Ferenc Zsila, Zsolt Bikadi, David Malik, Peter Hari, Imre Pechan, Attila Berces, Eszter Hazai.  2011. Evaluation of drug–human serum albumin binding interactions with support vector machine aided online automated docking. Bioinformatics Vol. 27, No. 13, pp. 1806–1813.