N-cyclohexyl-N-methylcyclohexylamine

Auto flammability

Link to relevant study record(s)

Reference

Endpoint:

auto-ignition temperature (liquids)

Type of information:

read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

weight of evidence

Reliability:

2 (reliable with restrictions)

Justification for type of information:

1. HYPOTHESIS FOR THE ANALOGUE APPROACH
An analogue approach can be established for several cyclohexyl amines, based on common functional groups and similar physico-‐chemical properties. This method of data gap filling is in accordance with Regulation EC No. 1907/2006, Annex XI, Section 1.5, Grouping of substances and read-‐across, and uses concepts from ECHA’s Read-‐Across Assessment Framework (ECHA, 2015). This approach allows existing data from analogues to be used to fill data gaps in a testing program for a target substance (mDCHA), in place of undertaking actual toxicological testing.
The target substance is N-‐cyclohexyl-‐N-‐methylcyclohexylamine (mDCHA, CAS 7560-‐83-‐0, also known as N-‐methyldicyclohexylamine). The proposed analogues include dicyclohexylamine (DCHA, CAS 101-‐83-‐7) and N-‐dimethyl-‐N-‐cyclohexylamine (dmCHA, CAS 98-‐94-‐2, also known as N-‐cyclohexyl-‐N-‐dimethylamine).
2. SOURCE AND TARGET CHEMICAL(S) (INCLUDING INFORMATION ON PURITY AND IMPURITIES)
The target substance is N-cyclohexyl-N-methylcyclohexylamine (mDCHA, CAS 7560-83-0, also known as N-methyldicyclohexylamine). The source substances have been identified as DCHA (CAS 101-83-7) and dmCHA (98-94-2). The impurities of the target and each source substance have also been identified. If the impurity profile of the substance used in the source read-across study
3. ANALOGUE APPROACH JUSTIFICATION
The scenario for this read-‐across process is the analogue approach, used when read-‐across is employed between a small number of structurally similar substances (ECHA, 2015; RAAF, 2015). Each of the analogues possesses an amine group to which is bound at least one cyclohexane ring. In biological and environmental settings, each of the proposed analogues is expected to elicit the same type of effects, based on a common functional group of toxicological interest (i.e., the amine group with a reactive unshared pair of electrons).
The analogue approach is established for the target substance mDCHA and the source substances DCHA and dmCHA, on the following basis (REACH Annex XI, Section 1.5): common functional groups resulting in similar environmental and biological effects.
All the substances are structural analogues as amines, characterised by an unshared pair of electrons which, in water, becomes hydrated and alkaline. The lone pair of electrons also allows the amine to participate in hydrogen bonding and nucleophilic substitution reactions. Amines display characteristic toxicity of aquatic toxicity, point-‐of-‐contact mammalian irritation (skin, eye and mucosa), and nucleophilic/protein binding potentially leading to mutagenicity.
The target substance, mDCHA, is comprised of an amine with two cyclohexyl rings and a methyl group. This is a tertiary amine where the substituent groups result in low solubility in water, low vapour pressure and a high boiling point.
The first source substance, DCHA, possesses two cyclohexyl rings and one hydrogen substituent on the amine, qualifying it as a secondary amine. This substitution is thought to have no significant effect on toxicity, as neither the methyl (of mDCHA) nor the hydrogen displays significant chemical reactivity. However, there appears to be slightly higher acute mammalian toxicity with DCHA, perhaps due to the possibility of ionisation in DCHA or effects of steric hindrance of the methyl in mDCHA. DCHA may represent the “worst case”, so that any read-‐across from DCHA to mDCHA overestimates the toxicity, which may result in a more conservative classification. This is desirable in a “precautionary principal” approach.
The second proposed source substance, dmCHA, is a tertiary amine with one cyclohexyl group and two methyl groups. This is a smaller compound with a lower molecular weight than mDCHA or DCHA. This source substance displays some physico-‐chemical properties which differ from the other analogues, such as lower boiling point, higher vapour pressure and lower octanol-‐water partition coefficient (Kow). However, its behaviour in toxicity testing is strikingly similar to that of DCHA (See Table 3.) This consistency in toxicity test results lends support for the read-‐across approach and increases our confidence in the hypothesis that it is the amine which drives the toxicity.
The OECD QSAR Toolbox, Version 3.3.0, was used to identify analogues for the target substance (mDCHA). See Appendix 2. The profiling exercise indicates that mDCHA falls into an established category of aliphatic amines (as per U.S. Environmental Protection Agency (EPA)). This assignment supports the validity of the read-‐across approach. The source substances chosen are those amines which bear the closest structural similarity to the target substance, and which have sufficiently robust datasets available which are informative for data gap filling.
Specific chemical categories have previously been established around the dominant activity of the amine group. Two examples are from the OECD HPV Chemicals Programme: a large category of C1-‐13 Primary Amines (US/ICCA, 2011), which included both linear and cyclic structures, and a category of Tertiary Amines, one of which was dmCHA (US/ICCA, 2012).
The basis for both categories was the common chemical structure and reactivity of the amine group, specifically the lone pair of electrons. Chemicals in these categories displayed high alkalinity associated with skin and eye irritation, respiratory irritation, and histological evidence of irritation of the forestomach in rats to whom the amines were orally administered in repeated doses. Toxicity to aquatic organisms was also commonly observed.
The OECD (Q)SAR Toolbox also assigned the target substance to the category of “SN1 iminium DNA binding” category. This behaviour reflects the reactivity of the unshared pair of electrons of nitrogen to engage in nucleophilic behaviour, including toward sites in DNA. This mechanism of action further supports the hypothesis that it is the chemical activity of nitrogen which characterises the toxicity of this category of amines rather than the substituents on the amine. Among the experimental results of mutagenicity testing for the target and source compounds, none was definitely demonstrative of a mutagen, and none is currently classified as such (including an EU Harmonized classification for DCHA). The weight of evidence for in vivo mutagenicity, where metabolism and DNA repair mechanism are operative, is that the substances are non-‐mutagenic.
The OECD (Q)SAR Toolbox assignment of the target substance to the category of “BfR and HESS skin and eye irritants” again supports the hypothesis of the amine structure as the functional group of toxicological interest. Point-‐of-‐contact irritation is observed with all the analogues as a result of its strong alkaline character. The measured pKa of both DCHA and dmCHA are over 10 (See Data Matrix, Appendix 2). These data suggest that attention be paid to the pH of solutions of amines, as used in toxicity testing, especially aquatic toxicity testing.
The propensity of the amine to ionise in solution is a critical factor impacting the decisions to fill data gaps by non-‐test alternatives. One common option is to predict the results of aquatic toxicity tests using ECOSAR (Q)SAR models from the EPA, which are based on structure-‐activity relationships for both chemical class and neutral organic (baseline) toxicity. ECOSAR has assigned the target substance and the source substances to the chemical class of aliphatic amines. The differential toxicity between the three aliphatic amines is therefore based on the respective Kow of each substance. The Kow is a less reliable predictor of relative toxicity for substances that are ionised than for those substances that are charge-‐neutral and non-‐ionised at environmentally relevant pH values (i.e., pH range of 4 to 9 for fresh surface waters). Since each of the three aliphatic amines is expected to exist in the ionised form at environmentally relevant pH, the identification of a common class for predicted toxicity (i.e., aliphatic amines) is more important than the estimation of relative toxicities between the three amines based upon Kow. The ECOSAR Reports for mDCHA, DCHA and dmCHA are found In Appendix 3.
Both DCHA and dmCHA have been shown to be readily biodegradable in ready biodegradability tests. The degradability of both source substances indicates that ready biodegradability of mDCHA may also be reasonably expected. However, the higher molecular weight and Kow value of mDCHA when compared to both DCHA and dmCHA indicate some caution that the target substance may be less biodegradable than either of its source substances. As indicated in Section 4 (above), the BIOWIN model reports an overall prediction that mDCHA is not readily biodegradable; DCHA and dmCHA are, similarly, predicted by BIOWIN to not be readily biodegradable. Therefore, BIOWIN results cannot be used to either fill a data gap nor to support the read-‐across from DCHA and dmCHA to mDCHA for this endpoint. The use of the analogue approach thus appears to be a more valid option to meet the registration requirement for this endpoint. However, it may be advisable to conduct a separate ready biodegradability test on mDCHA, in consultation with the testing laboratory to maximise conditions for a favourable test result (perhaps including an option for an extended or enhanced biodegradability test, in the event test results fall just below the degradation threshold at 28 days).
Read-‐across from DCHA or dmCHA to MDCHA for the adsorption coefficient (Koc) endpoint is not recommended. The preferred options include substance-‐specific laboratory testing; or estimation by QSAR, with a modeling software that predicts adsorption of the substance in the ionised state (mDCHA is ionised at environmentally-‐relevant pH values).
The relative aquatic toxicity of DCHA and dmCHA may be compared based on the values in Table 3 for acute fish, acute daphnid and algal toxicity. The table exhibits no clear pattern between the relative toxicity of the two source substances. The inability to discern a clear pattern of relative toxicity is confounded by the uncertainty between nominal and measured test concentration values. Perhaps more importantly, there is additional uncertainty as to whether the observed effects are pH-‐related effects in lieu of chemical toxicity; only the acute fish toxicity of dmCHA to Leuciscus idus is clearly reported with LC50 values for both the non-‐neutralised and pH-‐adjusted test solutions. Despite these two sources of uncertainty, the reported L(E)C50 values for all endpoints from both substances are clustered within an order of magnitude (with the exception of the LC50 from the pH-‐adjusted toxicity test for dmCHA on L. idus). Therefore, it is recommended that read-‐across information from both DCHA and dmCHA be used to represent the toxicity of mDCHA, using both the key and supporting studies from each registration file.
4. DATA MATRIX
See AARF document attached in Background material section.

Reason / purpose for cross-reference:

read-across source

Reason / purpose for cross-reference:

read-across: supporting information

Key result

Auto-ignition temperature:

255 °C

Atm. press.:

1 013 hPa

Conclusions:

the read-across is valid for classification and labelling, and for risk assessment

Key value for chemical safety assessment

Autoflammability / Self-ignition temperature at 101 325 Pa:

255 °C