Isooctyl acrylate

Toxicity to aquatic algae and cyanobacteria

Description of key information

A weighted average of 0.535 mg/L was calculated using QSAR results based on the contribution of each class of components (C7-C9) to the total composition.
Estimated ErC50, C7: 0.714 mg/L.
Estimated ErC50, C8: 0.542 mg/L.
Estimated ErC50, C9: 0.410 mg/L.

Additional information

The toxicity of isooctyl acrylate to aquatic algae and cynaobacteria was assessed in one laboratory study and by QSAR analysis. The laboratory study was conducted under a protocol based on OECD TG201. It established that the 48-hour EC50 for growth rate in Pseudokirchneriella subcapitata was <1.28 mg/L. Since the endpoint was exceeded at the lowest concentration tested and the control was no longer in exponential phase after 48 hours, the result is indicative of the toxicity that might be expected in a full test. Simulations were therefore undertaken to fill the data gap. Simulation was an internal QSAR based on toxicity of four simple, branched or linear aliphatic acrylate esters to Pseudokirchneriella subcapitata in 72-hour growth inhibition studies conducted according to OECD TG201. Data for the model was obtained from publicly-available reports with the exception of data from an internal study of 2-methylbutyl acrylate (EC# 256-170-3), the results of which had not been previously published. A robust summary providing details of the 2-methylbutyl acrylate study is included in this data set as read-across to justify the validity of the result and to improve and to allow incorporation into the model.

The sole molecular descriptor in the model was the number of carbon atoms in the esterified alcohol subunit. The model therefore is based on a common chemistry and mechanism of action, and has an independent variable that does not depend on precise structure or on a physicochemical property subject to experimental error.¹ Log- transformed data was entered into linear regression. The resulting equation had a high degree of linearity and robustness :

Log (1/EC50) = 0.1536 * Carbon number - 1.6978, R-squared = 0.987, Q2LOO = 0.9993 (explained variance in prediction)

Isooctyl acrylate is a UVCB material and is not fully characterized. However, the typical distribution of congeners in IOA may be described as follows:

Reaction products of propenoic acid with C7 alcohols (branched): >1.25%

Reaction products of propenoic acid with C8 alcohols (branched): >94.0%

Reaction products of propenoic acid with C9 alcohols (branched): >4.0%

Accordingly, the simulation was run for each class of components. The predicted ErC50 values of the C7, C8, and C9 components were 0.714 mg/L, 0.542 mg/L, and 0.410 mg/L, respectively. A weighted average of 0.535 mg/L was calculated based on the contribution of each class of components to the total composition. 

The weighted predictions provide a reasonable estimate of the toxicity of IOA. All toxicity values are based on reliable studies. The short-chain aliphatic acrylate esters, branched or linear, act by a common mode of toxicity, i. e., Michael-type addition to nucleophilic species such as proteins. The use of carbon number rather than an experimentally derived value reduces potential experimental error and bias due to inconsistent measurement methods of, e. g., log P. Each carbon number can therefore be judged as a class. That branching has little impact on toxicity is demonstrated by high linearity of a regression containing a mixture of branched and linear compounds. Long-chain acrylate esters, in contrast, operate as simple narcotics, exerting less overall toxicity than the short-chain compounds. The experimental study, while flawed and not considered reliable, demonstrates that IOA has a toxicity in the range suggested by the regression.  This indicates that the mechanism of isooctyl acrylate toxicity, like the shorter acrylates, is Michael addition rather than simple narcosis. Further, purified 2-ethylhexyl acrylate (2-EHA, CAS# 103-11-7) has a 72-hour ErC50 to Desmodesmus subspicatus of 1.71 mg/L.² 2-EHA is a significant component of IOA, and the reported value lends further support to the estimate. A QSAR for toxicity of esters including 2-EHA has been published, however parameterization of the model was conducted before downward revision of 2-EHA EC50 due to low solubility of this compound.¹ Therefore, the QSAR developed for this dossier provides the most reliable means of assessing toxicity of IOA.

References:

1, Papa E., Battaini F., and Gramatica P. 2005. Ranking of aquatic toxicity of esters modelled by QSAR. Chemosphere 58:559-570. 

2, European Chemicals Bureau. 2005. Risk Assessment Report for 2-Ethylhexyl Acrylate.