Registration Dossier

Diss Factsheets

Ecotoxicological information

Endpoint summary

Administrative data

Description of key information

Additional information

One experimental study of reliability 1 (based on international guideline and GLP compliance) and two QSAR results (iSafeRat® Holistic HA-QSAR v1.5) are available to assess the aquatic toxicity of the registered substance, on aquatic invertebrates and algae. The results are presented in the table below.

Table 6.1/1: Summary of aquatic toxicity data available



Measured values

Predicted values

(iSafeRat® Holistic HA-QSAR v1.5)

Short-term toxicity to aquatic invertebrates

48h-EL50 = 7.77 mg/L (nominal loading values)

48h-EL50 = 4.4 mg/L

Toxicity to aquatic algae


72h-ErL50 = 11 mg/L

The experimental short-term toxicity study on the aquatic invertebrate, Daphnia magna, was performed on the registered substance. In this study, the 48h-EL50 value, based on nominal loading values, was determined at 7.77 mg/L (95%-Confidence Limit: 6.85 – 8.90 mg/L) where the major constituent was measured and found to be stable over the study period.

This study was supported by a conservative QSAR result on aquatic invertebrates performed on the registered substance also.


The QSAR used a calculation method to predict the ecotoxicological values which would be expected when testing the substance under experimental conditions in a laboratory following OECD Guidelines adapted for testing as a mixture using the WAF method. This method has previously been validated in an internal publication (Bicheral and Thomas, 2014; available in the Endpoint Study Record). This algorithm is based on a QSAR model (iSafeRat® Holistic HA-QSAR v1.5) which has been validated to be compliant with the OECD recommendations for QSAR modelling (OECD, 2004). The QSAR is based on validated data for a training set of chemicals derived from experimental studies, for which the concentrations of the test item had been determined by chemical analyses over the test period. Based on the chemical structure of ketones, the three major constituents of the test substance are expected to fall on the MOA 1 regression line and they can attributed to the class of non-polar narcotic compounds (MOA 1). Further to this, the effect loading rate of the mixture is determined by using a series of calculation steps using phase equilibrium thermodynamics and excluding the non-bioavailable fraction. The 48h-EL50 value based on mobility of daphnids was determined to be 4.4 mg/L.

Based on this data, the QSAR result obtained on daphnids supports the experimental study and can be considered slightly conservative compared to the in vivo result but strongly supports the validity of the QSAR and its usefulness as a predictor of ecotoxicity for this substance. For the algae endpoint, the same calculation method was used (iSafeRat Holistic HA-QSAR v1.5). Thus the prediction method can be considered sufficient to assess the toxicity of the registered substance to algae. The 72h-ErL50 based on growth rate of algae was determined to be 11 mg/L. An in vivo algae study is therefore not considered necessary.