2,2'-(cyclohexylimino)bisethanol

Endpoint summary

Description of key information

The registration substance has a measured log Kow of 0.87 by using OECD107 Shake-flask method.

It is agreed that this value cannot be considered as valid indicator of bioaccumulation potential alone as the partitioning behaviour will be impacted by the surfactant behaviour. Therefore, a weight-of evidence approach has been applied for the bioaccumulation endpoint. The following information has been included in addition.

 

OECD 107 method suggests that “when the shake flask method is not applicable, e.g. for surface active materials, a calculated value or an estimate based on the individual n-octanol and water solubilities should be provided”. 

 

For surface active substances, which can form micelles and show a complex solubility behaviour, due to aggregation, the monomer concentration, and hence the thermodynamic activity, reaches a limiting value at the critical micelle concentration (cmc). It remains approximately constant when the total concentration is further increased. For ecotoxicological models requiring a solubility value, the critical micelle concentration is therefore the appropriate parameter. This is also suggested in ECHA Guidance R.7a.

The measured cmc value of the registration substance is 28.2g/L at 25°C. (as. IUCLID Ch. 4.8)

As the substance is fully miscible in every ratio with n-octanol the density can be used as a quantitative descriptor of the solubility in n-octanol. This certainly constitutes a worst case assumption, since the maximum solubility is taken as 100 % test substance.

The measured density of the substance is 1.0382 g/cm³ at 20°C. (s. IUCLID Ch. 4.4 and 4.9)

With the cmc value and the density a valid log Kow can be calculated to be 1.57. This value can be seen as descriptor of the worst-case bioaccumulation potential of the substance by passive diffusion driven by hydrophobicity. It is well below 4.5 indicating low bioaccumulation potential.

 

The BCF value for the registration substance has been calculated by the BCFBAF v.3.01 model implemented in the US EPA EPIWIN Suite v4.11. Using SMILES OCCN(CCO)C1CCCCC1 and the log Kow of 1.57 all calculated BCF (with and without biotransformation, all trophic levels, regression based) are <12 L/kg ww. In addition, a calculation has been made with BIONIC V2, using the log KMW discussed later for C8-tertiary amine and resulting in BCF values <2L/kg ww independent of including or excluding growth dilution, biotransformation or fecal egestion.

 

The registration substance is a tertiary amine, which is mainly protonated under physiological or environmental conditions (pH 5-9). The degree of protonation is exclusively determined by the pKa of 9.1 and the ambient pH. At pH5 99.99%, at pH7 99.42% and at pH9 63.17% are protonated respectively. Consequently, under environmental conditions, the substance is mainly protonated and a cationic surfactant.

 

Ionic substances cannot pass the lipid bilayer of a membrane by simple passive diffusion. They will be transported by active transport with the help of membrane proteins (carrier). In doing so the transport proteins involved consume metabolic energy, usually ATP. It is very unlikely that exogenous, ionic substances will be transported through the semi permeable lipid bilayer by using metabolic energy.

 

Phospholipid-water partitioning coefficients (kplip-w) can be used to investigate more realistic partitioning behavior of cationic surfactants, like the registration substance, as phospholipid bilayers or membranes have zwitterionic headgroups at the interface. Cations can interact favorably with the phosphate group. [1]  Log kplip-w values are published for several amine substances. These values decrease with lower C-chain length of the alkyl chain. Published values for tertiary alkyl-dimethyl-amines are: C8N(CH3)3+: 2.35; C10N(CH3)3+: 3.65; C12N(CH3)3+: 5.30. [2]

 

The registration substance is a tertiary alkyl amine with one cyclohexyl-alkyl-group and two ethoxy groups attached to the amino function. The differences to linear alkyl-dimethyl-amines are the bulkier alkyl chain leading to increased steric hinderance, which is caused by the ring-structure, and two ethoxy side groups, which are more polar than methyl groups as they bear hydroxy functions. Nevertheless, these differences are of low importance as for partitioning processes the main driver is the mostly protonated ammonium function and considering the EAWAG-modelled pathway the metabolic clearance can be assumed. For the next arguments the relevant structural differences and their impact are discussed.

Within the ECO37 project of CEFIC Long range initiative (LRI) BCF values are predicted from phospholipid-water partitioning coefficients for cationic surfactants based on 5% lipid content in wet fish and a phospholipid fraction of 1.25%. [3]

By using the presented equation (ionic fish BCFconservative (ww) = 0.0125 x kplip-w or log BCFconservative = log kplip-w – 1.9) the obtained BCF values can be compared with the thresholds defined in REACH legislation. With this, substances with a log kplip-w ≥ 5.2 are B and log kplip-w ≥ 5.6 are vB.

As a conservative approach the published value for C8-alkyl-dimethyl amine can be used to estimate the kplip-w value of the registration substance without taking into account that the attached C-chain is only C6 and the resulting kplip-w should be lower. This effect is further intensified as with increasing “bulkiness” or steric hinderance, the ionic interaction is further reduced. [4]  This demonstrates that even by using a more certain system, introducing ionic driven partitioning, the registration substance has a low bioaccumulation potential.

 

Within the same ECO37 project the enzymatic clearance of several amine substances have been investigated by using rainbow trout S9 liver extract in an in-vitro biotransformation assay. For C9- and C10-alkyl-dimethyl amine the CLS9-values are slightly above 300 ml/h/g liver. As the optimum of clearance has been found for C12-tertiary amine with lower rates for higher and lower C-chain-numbers the value for the registration substance, a C6-tertiary amine, is expected to be slightly lower, but the values show that the possible low intake of the substance is opposed by a significant clearance caused by fast biotransformation.

 

In addition to the investigation of phospholipid-water partitioning and in-vitro biotransformation the ECO37 project includes in-vivo bioaccumulation studies using 3 juvenile rainbow trout per sampling point over 14 days exposure phase (8 sampling points) and 28 days of depuration phase (10-12 sampling points). 4 tertiary alkyl-dimethyl amines have been investigated with an alkyl-C-chain from 9 to 14. As result the obtained BCF values increase with increasing C-chain number, indicating that the result for the registration substance, a C6-tertiary amine, will be lower than the C9-substance. For C9-alkyl-dimethyl amine the following kinetic parameter and BCF have been found for pH7.8: Uptake rate (K1) = 2.2 L/kg.h; Elimination rate (K2) = 0.057 1/h; BCFkinetic = 40 L/kg.

 

The results from the Cefic LRI ECO37 project, are not officially published until now, but papers are in preparation and the results have been circulated and presented within the community of ecotoxicologists, e.g. at SETAC 2019 in Helsinki. Preliminary findings have been presented to ECHA on 24th of May 2019 during the “Surfactant training day” by Dr. Steven Droge. Consequently, the results of the ECO37 project are high quality, published research results which must be used as available data. The project has been performed, by institutions with high reputation, in cooperation of the university of Stockholm, the university of Amsterdam, Armitage Environmental Science and Arnot Research and Consulting. Supporting regulatory scientists from authorities of Germany (Umweltbundesamt), Canada and USA (EPA) were involved in the project.

 

Overall, it has been shown that the registration substance has a low potential for uptake into the organism independent if the driving mechanism is hydrophobicity (log Kow=1.57, calculated BCF < 12 L/kg) or includes an ionic mechanism (log kplip-w < 2.35, conservatively derived from C8-alkyl-dimethyl amine).

It is very unlikely that the ionic registration substance would pass biological membranes.

For similar substances (C9-alkyl-dimethyl amine and C10-alkyl-dimethyl amine), within the ECO37 project fast biotransformation has been shown with measured in-vitro clearance rates (CLS9 of around 300 ml/h/g liver) and an in-vivo bioaccumulation experiment indicated low potential to bioaccumulate (Uptake rate (K1) = 2.2 L/kg.h; Elimination rate (K2) = 0.057 1/h; BCFkinetic = 40 L/kg). All similar substances have longer alkyl-chains than the registration substance, and in all cases the values increase with increasing alkyl-chain-length. Therefore, all values can be seen as worst-case conservative approach. Considering this overall picture within the weight-of-evidence approach, especially in combination with animal welfare considerations, an additional BCF study would not generate any new information and should be avoided, as the presented data show a low potential for bioaccumulation. This is supported by the result of the P-assessment which has been shown that the registration substance is not P and not vP.

   

[1]  Armitage, J.M.; Erickson, R.J.; Luckenbach, T.; Ng, C.A.; Prosser, R.S.; Arnot, J.A.; Schirmer, K.; Nichols, J.W. Assessing the bioaccumulation potential of ionizable organic compounds: Current knowledge and research priorities. Environ. Toxicol. Chem. 2017, 36 (4), 882-897; 10.1002/etc.3680.

[2]  Timmer, N.; Droge, S.T.J. Sorption of cationic surfactants to artificial cell membranes: comparing phospholipid bilayers with monolayer coatings and molecular simulations. Environ. Sci. Technol. 2017, 51, 2890-2898; 10.1021/acs.est.6b05662.

[3]  Paper in preparation; Steven T.J. Droge, Peter Scherpenisse, Jon A. Arnot, James M. Armitage, Michael McLachlan, John Nichols, Peter von der Ohe, Mark Bonne, Geoff Hodges; surfactant bioconcentration assessment should focus on phospholipid binding instead of inadequate octanol-water partition ratio estimates.

[4]  Armitage, J.M.; Arnot, J.A.; Wania, F.; Mackay, D. Development and evaluation of a mechanistic bioconcentration model for ionogenic organic chemicals in fish. Environ. Toxicol. Chem. 2013, 32 (1), 115-128