Registration Dossier

Categories

Categories

Category name:
Polyol substances: Group 1 Ether-linked substances (Illing & Barrat 2010)

Justifications and discussions

Category definition:
The target substances are short chain oligomers formed from core molecules containing multiple hydroxyl or amino functional groups or a combination of the two. These functional groups are alkoxylated with propylene oxide or ethylene oxide. The alkoxylation of the core molecules results in multiple free terminal hydroxyl groups, and are therefore these substances are termed as “polyols.” These oligomers have been defined by the European Union as ‘No Longer Polymers’ (NLP) because they do not meet the OECD definition of polymers.
As a group, these NLPs have similar physical, chemical and toxicological properties. The distinguishing feature among the NLPs within this category is the ether linkages between the core molecules and the ethoxy or propoxy repeating units.
Category rationale:
3 CATEGORY JUSTIFICATION
The NLPs in this category are formed primarily from propoxylation or ethoxylation of the hydroxyl functionalities of in the core molecules. All target substances are UVCB substances.
Since the members of this category have a variety of structures with various anticipated breakdown products and metabolites, Scenario 6 (different substances with qualitatively similar properties) was chosen for as the basis of justification for this category.

3.1 STRUCTURAL CHARACTERIZATION
Most of the NLPs in this category contain polyhydroxy core substances, primarily sugars and small aliphatic polyols with the exception of the last two members of the category, which are discussed below. Since the reactive sites of all the molecules in this group are either ethoxylated or propoxylated, the formed oligomers are expected to have similar properties that are primarily determined by the nature of these alkoxyl repeating units.

It is noted that there are two substances present in this category with an amine as the core molecule, 2,2'-iminodiethanol, propoxylated (CAS 35176-06-8) and 2,2',2''-nitrilotriethanol, propoxylated (CAS 37208-53-0). These two amines are included in Category 1 because they form ether linkages.
For CAS 37208-53-0, there are no replaceable protons on the N-atom in the core triethanolamine (TEA) molecule. Therefore, propoxylation of TEA occurs on the terminal hydroxyl groups of the ethanolic moieties via ether linkages. In the case of CAS 35176-06-8, the core molecule, diethanolamine (DEA) has one replaceable proton on the N-atom. As a result, the first propoxyl group will be added by the replacement of that proton. This will convert the DEA molecule to a close analog of TEA. All further propoxylation will then occur on the terminal hydroxyl groups by the creation of ether linkages. Thus, for both of these amine-containing core molecules, ether is the primary linkage used to form the resulting oligomers.

3.2 PHYSICAL AND CHEMICAL CHARACTERIZATION
The physico-chemical properties of the members of this category are similar. All members are liquids at ambient temperatures. Given that these materials are UVCB substances, there is a range of molecular weights that is influenced by both the core molecule and the degree of alkoxylation. The molecular weights for this category of NLPs range from 337 to 1974 Daltons across the category with relative densities between 1.04 and 1.2. Vapor pressures are consistently low with values ranging from 1.2 x 10-7 to 8.4 x 10-4 hPa. across the category.

All oligomers in Category 1 are highly water-soluble (range ≥ 240,000 - < 1,000,000 mg/L) which would indicate low partition coefficient values. The oligomers in Category 1 can exhibit varying degrees of surface activity which makes determination of partition coefficients experimentally unreliable. Therefore, the range of log Pow values were determined for the lowest and highest molecular weight homologues using the QSAR model ClogP (BioByte Corp). The estimated range of log Pow for Category 1 is -3.6 to 1.8.

Likewise, the assessment of pKa is complicated due to the difficulty of isolating the ionisable components of the oligomers. The dissociation constant was estimated using the QSAR model ACD Labs pKa DB software. The estimated pKa range of Category 1 is 12.8 to 15.7, indicating that ionisation is not likely in aqueous environments encountered in most mammalian and ecosystems.

3.3 ABSORPTION AND ELIMINATION KINETICS
Due to the low vapor pressure of the Category 1 substances, significant exposure by inhalation is unlikely. Based on the water solubility and partition coefficients absorption by oral and dermal routes of exposure are expected to be limited.
There is no in vivo experimental information available to characterize the absorption, distribution and elimination kinetics of the NLPs. Instead, a modelling approach has been applied as described by Illing and Barratt (2007 and 2010).
Based on these calculations, the rate of dermal penetration is limited for all the NLPs. The most bioavailable NLP is predicted to be Propylidynetrimethanol, propoxylated (CAS 25723-16-4) with a log Kp of -4.7 cm/hr and the least bioavailable is predicted to be Sucrose, ethoxylated and propoxylated (CAS 26301-10-0) with a log Kp of -13.4.
Molecules with molecular weights of less than 500 g/mole are small enough to be candidates for absorption by passive diffusion from the GI tract. The molecular weight of the NLPs varies based on the nature of the core molecule and the number of sites that are alkoxylated but generally exceed, or are close to, the molecular weight limit. The estimated molecular weight of the NLPs in Category 1 range from 337 to 1974 g/mole. As a result, absorption from the GI tract is also likely to be limited.

3.4 METABOLISM
If absorbed, the NLPs target substances are expected to be dealkoxylated to yield the core substances and propane-1,2-diol (propylene glycol) and/or ethane-1,2-diol (ethylene glycol) depending on the monomer present in the oligomer NLP. The fragments generated from the dealkoxylation and the remaining core molecules are then further metabolized or excreted.

While no experimental data are available with these NLPs, the dealkyloxylation metabolic sequence has been demonstrated with alcohol ethoxylates (Drotman 1980, Talmadge 1994). In mammalian systems, the ether links between the fatty alcohol and the ethoxylate moiety is cleaved to release the free fatty alcohol and the ethoxy remnants of varying lengths, which are largely excreted in the urine. Formation of the corresponding glycols also occurs but direct excretion of the oligomers appears to predominate (ATSDR 1977, 2010, Drotman 1980, JECFA 2002).

In the environment, alkylphenol ethoxylates have been found to undergo a two-step degradation process. Initially, there is a step-wise removal of ethoxyl groups by aerobic and anaerobic processes. In addition, some depolymerization of the ethoxylated chains occurs through abiotic means (Jonkers et al 2001). The second stage degradation involves the ultimate conversion of the intermediate products to carbon dioxide, water, and inorganic salts (Ahel et al 1994).

3.5 BIOLOGICAL AND TOXICOLOGICAL DATA
The biological activity inherent in the members of this category is similar across all endpoints. The available data demonstrates a pattern of similar toxicological properties indicating that the members of the category possess low potential for toxicological hazard including anticipated breakdown products and metabolites.

3.5.1 Acute Toxicity
Acute oral and dermal toxicity data is available throughout the category, with each substance possessing LD50’s equal to or greater than 2000 mg/kg for each route of exposure. This finding is consistent with the limited dermal penetration predicted for these NLPs. Data relating to skin irritation in the form of either an in vitro or in vivo test is available for all members of the category; all substances are classed as non-irritating. Similarly all members of the category have been shown to be non-irritating to eyes in at least one study. Data shows that ten of the fifteen polyols possess unequivocally negative in vivo skin sensitisation properties.

3.5.2 Genotoxicity
With regards to genotoxicity all but one of the category substances have data demonstrating a lack of mutagenicity in the Ames test. Five of the fifteen substances have chromosome aberration test data, all of which are negative and three of the fifteen substances are negative in in vivo gene mutagenicity assays. Collectively, this information strongly indicates that the category as a whole can be considered to lack genotoxic activity.

3.5.3 Chronic Toxicity
In repeat dose toxicity studies, four of the category substances have NOAEL’s equal to greater than 1000 mg/kg/day in 28 day studies. There are two reproduction/developmental screening studies on category substances. In both studies there were no reproductive or developmental effects observed and parental NOAELs of greater than 1000 mg/kg/day. A prenatal development study gave similar results of greater than 1000 mg/kg/day for both maternal and foetal effects.

3.6 ENVIRONMENTAL DATA
3.6.1 Biodegradation
Based on ISOPA (2011), the inclusion of substances within this category in both abiotic and biologically-mediated reactions will be governed by their common structural features. These features include the terminal hydroxy groups and repeating ether linkages which are common to all members within this category.

3.6.2 Hydrolysis
There is one study available to address the hydrolysis endpoint of this category, Propylidynetrimethanol, ethoxylated (CAS 50586-59-9). This study demonstrates the ether linkages characterising this category are resistant to hydrolytic degradation at pH 4, 7, and 9.

3.7 ECOTOXICOLOGICAL DATA
3.7.1 Acute Aquatic Toxicity
Acute toxicity data is available for all members of the category (with 2 exceptions) elaborated further below, with each substance demonstrating:

• Fish toxicity LC50 of > 100 mg/L
• Daphnia toxicity EC50 of > 100 mg/L
• Algae toxicity EC50 (growth) > 100 mg/L.

This demonstrates low overall toxicity of the category to aquatic organisms.

3.7.2 Chronic Aquatic Toxicity
Based on the available acute aquatic toxicity data, ISOPA (2011) concluded that the most sensitive species from the 3 trophic levels is Daphnia; therefore it is appropriate to use existing chronic Daphnia read across to address chronic toxicity to aquatic species within the category.
The available data demonstrates a pattern of similar ecotoxicological properties across the category indicating that the members of the category possess low potential for ecotoxicological hazard. A full break down of the available toxicological data is presented in the Annex to this document.

3.8 PERSISTENCE, BIOACCUMULATION AND TOXICITY (PBT)
All substances within the category, according to their physico-chemical properties, are considered highly water soluble (>1000 mg/L), unlikely to bioaccumulate (log Pow <4.5) and unlikely to dissociate within the environment (low Pka). All substances are considered to be mobile within soil (log Koc <5).

4 CONCLUSION
This data confirms the category of UVCB NLP Polyols described within this document is suitable for data gap filling by read across according to the guidelines set out in the ECHA RAAF frame work.
The category read across hypothesis is appropriate due to significant structural similarity throughout the category members. This chemical similarity is further supported by adequate, reliable scientific data presented in the read across category data matrix Annex (Annex 2), this data confirms a pattern of similar physico-chemical, environmental, and toxicological properties amongst the category substances and satisfies the read across criteria as set out in the RAAF.