Dihydrogen bis(acetato-O)difluoroborate(1-)

Short-term toxicity to fish

Description of key information

With high probability acutely not harmful to fish.

Additional information

In water, the acetic acid - boron trifluoride complex rapidly decomposes to form acetic acid and boron trifluoride dihydrate. The latter reacts further to boric acid and fluoroboric acid, which finally hydrolyses to yield hydrofluoric acid/ fluoride ions (BUA Report 261, 2005).

Studies on fish toxicity are available for all main hydrolysis products of the acetic acid – boron trifluoride complex.

The studies with acetic acid by Beslagic (2005) and Zok (1995) are GLP compliant and follow standard guidelines. The test substance is only 50% potassium acetate.The remainder was predominantly water with no components likely to prejudice the study outcome in relation to assessing the activity of potassium acetate.The reported LC50 > 300 mg/L is based on the concentration of the acetate ion. The NOEC and LC50 were adjusted for the molecular weight of the acetate compared to the parent molecule and also the fact that the test substance is only 50% potassium acetate.

Boron trifluoride dihydrate was tested in a static system, following the guideline DIN 38412 (L15) and using the golden orfe (Leucidus idus L.) as test organism. The mortality observed during the test was due to the low pH of the test solution at higher test substance concentrations. In the neutralised test solution, ran in parallel with 100 mg/L test substance, no mortality occurred (BASF AG, 1988). Of the data presented in the boric acid EU Risk Assessment Report, the most reliable acute tests on fish (4 day duration) show LC50 values in the range of 125 to 600 mg-B/L. The species tested include salmonids (Oncorhynchus kisutch, O. tshawtscha) and several endangered species such as Gila elegans, Ptychocheilus lucia, Xyrauchen texanus, and Catostomas latipinnis. Juveniles and fry appear to be the most sensitive fish life-stage (Hamilton, 1995; Hamilton and Buhl, 1990).

A few studies reported endpoints in the 5 to 15 mg-B/L range, but these were judged not reliable, or did not have sufficient information to permit data quality review. For example, Turnbull et al. (1954) reported a 24-hour TLm to bluegill of 4.6 mg-B/L in response to the test substance sodium tetraborate decahydrate. However, they also reported a 24-hour TLm of 2389 mg-B/L in response to the test substance boron trifluoride. Their procedure used relatively large fish (ca. 5 g, 7 cm). No information was provided on replication, intervals between test concentrations, or similar operational details. Guhl (1992a) reported 96-hour LC50 for zebrafish of 14.2 mg-B/L, but cited an unpublished study from Henkel KGaA. Terhaar et al. (1976) reported median lethal times for boric acid of 10 hours, exposed to 1750 mg-B/L which was extrapolated to an acute toxicity estimate of 17.5 to 175 mg-B/L. These studies cannot be adequately reviewed or compared to standard protocols, thus they cannot be judged reliable.