
 From:
Ginger Moser/RTP/USEPA/US
 To:
Tim Benner/DC/USEPA/US@EPA
 Date:
03/29/2010 08:26 AM
 Subject:
Re: Possible impact on wetlands - comments needed


Hi Tim

Below, I have copied Glen Thursby's review of the TFA information.  In my naive reading, it appears that he agrees that the risk of TFA is minimal, which I believe was the original conclusion.  Hopefully this will address the concerns that OAR had?

From Glen:
I looked over the information you sent relative to TFA -- with particular emphasis on the aquatic plant toxicity data (my area of expertise). The data I evaluated were those summarize in the document entitled, Revised Draft Assessment of the Potential Impacts of HFO-1234yf and the Associated Production of TFA on Aquatic Communities and Local Air Quality, prepared by ICF International. As I mentioned earlier my comments are restricted to the aquatic toxicity test data. I do not have the expertise to evaluate the environmental fate information presented. I briefly reviewed the following papers that are the source of the aquatic plant data in that report:

Benesch, et al. 2002. Investigation of Effects of Trifluoroacetate on Vernal Pool Ecosystems. Environmental Toxicology and Chemistry. 21(3):640-647. 
Berends et al. 1999. Toxicity of Trifluoroacetate to Aquatic Organisms. Environmental Toxicology and Chemistry. 18(5):1053-1059. 
Boutonnet, et al 1999. Environmental Risk Assessment of Trifluoroacetic Acid. Human and Ecological Risk Assessment: An International Journal. 5(1):59 -124.
Hanson, ML and Solomon, KR. 2004a. Haloacetic acids in the aquatic environment. Part I: Macrophyte Toxicity. Environmental Pollution 130:371-383.
Hanson, ML and Solomon, KR. 2004b. Haloacetic acids in the aquatic environment. Part II: Ecological Risk Assessment. Environmental Pollution 130:385-401.
One general comment. The data from Berends et al. (1999) are reported in mg NaTFA/L and those from Hanson and Solomon (2004a) and Benesh et al. (2002) in mg TFA/L. The ICF report authors might want to consider normalizing all of the data to TFA. TFA is approximately 83% of the molecular weight of NaTFA.

The number of aquatic plant species for which data are available in the Boutonnet et al. (1999) report and the one by Hanson and Solomon (2004a) is greater than the number of values typically available to assess the effects of toxic compounds on this community. In brief, the methods used were standard methods and generally appropriate for the type of testing performed. The data reported in the risk assessment by Boutonnet et al. (1999) were published in more detail in Berends et al. (1999). It is this latter paper that contains the data for the most sensitive alga tested, Selenastrum capricornutum (currently known as Pseudokirchneriella subcapitata (Korshikov) F. Hindák -- the use of Raphidocelis subcapitata did not last long). The EC50 for this species was reported as 4.8 mg.L NaTFA (ca. 4 mg/L TFA) and the NOEC was 0.12 mg/L NaTFA (0.1 mg/L TFA). Although these values are substantially lower that those reported for the other 10 algal species, this range in sensitivity among algae is not uncommon. In a risk characterization for pesticides registration the initial evaluation relative to aquatic plants would consist of taking the lowest EC50 (4 mg/L in this case) and dividing that into the expected environmental concentration. The level of concern for this ratio is 1. Any ratio greater than 1 would be cause for greater scrutiny. Based on some of the summary information presented I would expect that the ratio in this case would be several orders of magnitude less than 1. I've overly simplified this, but you get the idea.
The data presented by Hanson and Solomon (2004a) are for three species of aquatic vascular plants, Lemna gibba, Myriophyllum spicatum and Myriophyllum sibiricum. The authors present effects concentrations for several different endpoints for each species. The lowest EC50s for each were 618, 222 and 341 mg/L TFA, respectively. It interesting to note that the EC50 for L. gibba using frond number from Berends et al. (913 mg/L) was very similar to that calculated by Hanson and Solomon (884 mg/L) for this same endpoint (the EC50 from Berends et al. was corrected for TFA only). In the European Union initial risks to aquatic vascular plants are evaluated by dividing the EC50 from a standard test using Lemna (usually L. gibba) by 10. This is generally considered to result in a threshold concentration protective of aquatic vascular plants. In the case of TFA this would be ca. 90 mg/L. 
Benesh, et al. (2002) evaluated the potential toxic effect of TFA to species common in vernal pools (dry in summer). They studied the effects on microbial communities as well as on germination and growth or several wetland plant species. The following summary is from their abstract.
      Microbial respiration for three vernal pool soils and an agricultural soil was not affected by TFA exposures (0, 10, 100, 1,000, and 10,000 mg/L), and degradation of TFA by microbial communities was not observed in soils incubated for three months. Trifluoroacetate accumulated in foliar tissue of wetland plant species as a function of root exposure concentration (100 and 1,000 mg/L TFA), and accumulation was found to stabilize or decrease after the second or third month of exposure. Seeds accumulated TFA as a function of root exposure concentration; however, germination success was not affected. No adverse physiological responses, including general plant health and photosynthetic and conductance rates, were observed for root exposures at the TFA concentrations used in this study. Based on the soils and plant species used in this study, predicted TFA concentrations will not adversely affect the development of soil microbial communities and vernal pool plant species.
Finally, Hanson and Solomon (2004b) summarized the risks to freshwater aquatic plants of several different haloacetic acids (HAA) of which TFA was one. Their main conclusions are summarized below from their paper.
      Haloacetic acids at current environmental concentrations do not pose a threat to these aquatic plants in European, Canadian or African freshwater systems as concluded through the use of both deterministic and probabilistic methods of ecological risk assessment. Several orders of magnitude separate effects concentrations from exposure concentrations for these HAAs. Even mixtures of HAAs do not appear to pose a significant risk to these plants. Myriophyllum spp. and L. gibba are not exceptionally sensitive to HAA toxicity. Most effect measures are in the mg/l range and some endpoints, especially pigments, do not exhibit effects until the g/l range. Current concentrations of HAAs in the environment rarely exceed the ng/l range. The degradation products of HAAs are carbon dioxide and halide ions and in themselves do not pose a risk to these plants.

I hope this brief summary helps. I am sorry that I could not help relative to the environmental fate questions. 

Ginger
_____________________________
Virginia C. Moser, Ph.D., D.A.B.T.
Acting Regulatory Support Coordinator
Research Planning and Coordination Staff

Toxicity Assessment Division (MD B105-04)
US Environmental Protection Agency
Research Triangle Park, NC   27711
(919) 541-5075
