Registration Dossier

Administrative data

Hazard for aquatic organisms


Hazard assessment conclusion:
PNEC aqua (freshwater)
PNEC value:
0.002 mg/L
Assessment factor:
Extrapolation method:
assessment factor
PNEC freshwater (intermittent releases):
0.003 mg/L

Marine water

Hazard assessment conclusion:
PNEC aqua (marine water)
PNEC value:
0 mg/L
Assessment factor:
Extrapolation method:
assessment factor


Hazard assessment conclusion:
PNEC value:
10 mg/L
Assessment factor:
Extrapolation method:
assessment factor

Sediment (freshwater)

Hazard assessment conclusion:
PNEC sediment (freshwater)
PNEC value:
8.9 mg/kg sediment dw
Assessment factor:
Extrapolation method:
assessment factor

Sediment (marine water)

Hazard assessment conclusion:
PNEC sediment (marine water)
PNEC value:
0.89 mg/kg sediment dw
Assessment factor:
Extrapolation method:
assessment factor

Hazard for air


Hazard assessment conclusion:
no hazard identified

Hazard for terrestrial organisms


Hazard assessment conclusion:
PNEC soil
PNEC value:
0.083 mg/kg soil dw

Hazard for predators

Secondary poisoning

Hazard assessment conclusion:
PNEC oral
PNEC value:
5.3 mg/kg food
Assessment factor:

Additional information

The hydrolysis half-life of hexamethyldisiloxane (HMDS, L2, CAS 107-46-0) is 116 hours (5 days) at pH7 and 25˚C. The water solubility of the substance is low (0.93 mg/l) and the log Kow is high (5.1). It is therefore likely that, under the flow-through exposure conditions of the fish study that the test organisms will have predominantly been exposed to the registration substance. In the semi-static long-term invertebrate study and the static algal test it is likely that exposure will have been predominantly to the registration substance and a small proportion of its hydrolysis products. The test substance is volatile but measures were taken to maintain exposure concentrations by conducting the key fish test under flow-through conditions and the invertebrate and algal tests under sealed conditions.



In order to reduce the need for testing, read-across is proposed to fulfil REACH Annex X requirements for the registered substance from substances that have similar structure and physicochemical properties. Ecotoxicological studies are conducted in aquatic medium or in moist environments; therefore the hydrolysis rate of the substance is particularly important since after hydrolysis occurs the resulting product has different physicochemical properties and structure.

In aqueous media, HMDS hydrolyses in water with a half-life of 116 hours at pH 7 and 25°C, therefore the chemical safety and hazard assessment will focus on the properties of the parent substance.

The registration substance hexamethyldisiloxane (HMDS, L2, CAS 107-46-0) and the substance used as a surrogate are members of the Reconsile Siloxane Category. Substances in this category tend to have slow hydrolysis rates, low water solubility, high log Kow, high adsorption coefficients and slow degradation in the sediment compartment. For substances with a log Kow of 8 and above, no long-term toxicity effects are seen with aquatic organisms due to the low water solubility limiting the amount of substance that can be taken up by organisms. In the environment, the substances will adsorb to particulate matter and will partition to soil and sediment compartments.

Additional information is given in a supporting report (PFA, 2017) attached in Section 13 of the IUCLID 6 dossier.

In the following paragraphs the category read-across approach for the registered substance is outlined, taking into account structure and physicochemical properties.

Read-across from octamethyltrisiloxane (L3, CAS 107-51-7) to hexamethyldisiloxane (HMDS, CAS 107-46-0):

The registration substance, hexamethyldisiloxane (HMDS, CAS 107-46-0), and the source substance, octamethyltrisiloxane (L3, CAS 107-51-7), are members of the Siloxane Category. HMDS and L3 are linear siloxanes with two silicon atoms and one oxygen atom, and three silicon and two oxygen atoms, respectively. Each silicon atom is fully substituted with methyl groups. Refer to Section 1 for more structural information on the registration substance including diagrams.

In the context of long-term toxicity to fish: HMDS and L3 have similar physicochemical properties in respect of behaviour in water under flow-through conditions: medium molecular weight (162.38 and 236.54 respectively), fairly low water solubility (0.93 and 0.034 mg/l at 23˚C respectively), high log Kow (5.1 and 6.6 respectively) and high log Koc (3.0 and 4.3 respectively). Both substances have negligible biodegradability and moderate hydrolysis rates (half-life of 116 h (5 d) and 329 h (13.7 d) respectively at pH 7 and 25˚C).

Environmental toxicity data for siloxanes are consistent with a non-polar narcosis mechanism. Given the similar properties, structural similarities, and expected mode of action it is considered valid to read-across data from octamethyltrisiloxane (L3) to hexamethyldisiloxane (HMDS). Additional information is given in a supporting report (PFA, 2017) attached in Section 13 of the IUCLID 6 dossier.

In the context of sediment toxicity: A category approach is applied to this endpoint and is detailed in the Siloxane Category report (PFA, 2017). The hypothesis for read across of sediment ecotoxicity evidence within the Siloxanes Category is that no structure-based or property-based pattern is evident from the category data set of existing studies, although patterns are identifiable associated with extrinsic aspects of test design to which effects may be attributed. The approach will be revisited in the event that reliable new data become available. With this in mind, a single overall interpretation is made across the category. To fulfil the requirements of REACH, a conservative approach is made by reading across on a nearest-neighbour basis the reliable data within the category.

In the context of the RAAF, Scenario 6 is expected to apply to this endpoint. It is considered that effects observed in benthic organisms are associated primarily with extrinsic factors (such as pH, sediment type and organic carbon source and dosing concentration) associated with test design and not to structural similarities as such.

A total of twenty-four sediment toxicity studies for siloxanes and nineteen results from studies of standard duration in standard test species have been reviewed in detail. There is a general trend for studies using natural sediment, which all have pH <~8, to show no effects, or higher NOECs than those with artificial sediment. No significant toxicity (NOEC <100 mg/kg) in any organism is found at pH near 7 (pH 7±0.3) with natural sediment. The data suggest that it is possible to read across sediment toxicity data between different siloxane structures, especially where natural sediment data are available, given that the studies which are not suspected to be confounded by extrinsic factors show relatively minimal effects across the dataset.

Data from Lumbriculus variegatus are available with HMDS, and data from Chironomus riparius and Hyalella azteca are read-across from octamethyltrisiloxane (L3, CAS, 107-51-7). The selection of read-across substances is based on structural similarity and key physico-chemical properties (log Kow, log Koc, degradation).

HMDS and L3 are both linear siloxanes with two silicon atoms and one oxygen atom, and three silicon and two oxygen atoms, respectively. Each of the silicon atoms are fully substituted with methyl groups.

TheLumbriculus variegatusstudy with HMDS and theHyalella aztecastudy with L3 were conducted in natural sediment, and theChironomus ripariusstudy with L3 was conducted in artificial sediment. All are considered to be reliable and not confounded by extrinsic factors.

Given that HMDS and L3 have reasonably similar physicochemical properties and are structurally similar, it is considered valid to read across sediment toxicity data from L3 to HMDS in the context of the Siloxanes Category.


In the context of terrestrial toxicity: Terrestrial studies with siloxanes such as HMDS are considered to be technically difficult to conduct due to their high volatilisation potential (as indicated by high Henry’s Law Constant and low octanol-air partition coefficient) and the potential for degradation in soil. Soil testing according to guideline methods does not allow for a renewal of the substrate and hence re-application of test substance. Therefore, there is potential for the organisms to not be exposed to the test material for a sufficiently long period of time for effects to be expressed, as well as the difficulty of quantifying actual exposure concentrations. OECD TG 222 acknowledges that the test method may not be applicable to substances for which the air/soil partition coefficient is greater than one, or to substances with vapour pressure exceeding 300 Pa at 25°C. Terrestrial toxicity testing with HMDS and L3 has shown this to be the case (see discussion in IUCLID Section 6.3.1). HMDS meets both of these criteria (Kair-soil = 2.8, VP 5500 Pa); for L3, the vapour pressure criteria is met, and the air/soil partition coefficient is close to one (Kair-soil = 0.9, VP 530 Pa).

The physico-chemical properties of L3 are reasonably similar to those of HMDS, but the former should have greater stability in soil: HMDS has a higher tendency to volatilise from soil compared to L3, based on its higher vapour pressure (5500 Pa versus 530 Pa at 25°C) and lower tendency to partition to organic matter (log Koc 3.0 versus 4.3) than L3. This is expressed in the higher Kair-soil for HMDS (2.8) compared to L3 (0.9). HMDS also has a faster homogeneous hydrolysis rate (t½ 116 h for HMDS versus 329 h for L3, at pH 7 and 25°C), and so is expected to degrade faster than L3 in soil. It is also well established that siloxanes undergo clay-catalysed hydrolysis in soil (Xu et al., 1998 and Xu, 1998), with half-lives increasing with increasing molecular size of the siloxane (Xu and Chandra, 1999).

Table: Summary of ecotoxicological and physicochemical properties for the registered substance and the surrogate substance (aquatic and terrestrial toxicity)

CAS Number



Chemical Name

Hexamethyldisiloxane (L2, HMDS)

Octamethyltrisiloxane (L3)

Si hydrolysis product

Trimethylsilanol (2 moles)

 Dimethylsilanediol (1 mole) and Trimethylsilanol (2 moles)

Molecular weight (parent)



Molecular weight (hydrolysis product)


Dimethylsilanediol 92.2 Trimethylsilanol 90.2

log Kow (parent)



log Kow (silanol hydrolysis product)


Dimethylsilanediol -0.41 Trimethylsilanol 1.2

log Koc (parent)



Water solubility (parent)

0.93 mg/l at 23˚C

0.034 mg/l at 23˚C

Water solubility (silanol hydrolysis product from complete hydrolysis)

995 mg/l

Dimethylsilanediol 1.0E+06 mg/l (limited to ca. 1000 mg/l by condensation reactions) Trimethylsilanol 995 mg/l

Vapour pressure (parent)

5500 Pa at 25˚C

530 Pa at 25˚C

Vapour pressure (hydrolysis product)

1900 Pa

Dimethylsilanediol 7 Pa Trimethylsilanol 1900 Pa

Hydrolysis t1/2 at pH 7 and 25°C

116 hours (5 days)

329 hours (13.7 days)

Hydrolysis t1/2 at pH 5 and 25°C

1.4 hours

5.09 hours

Hydrolysis t1/2 at pH 9 and 25°C

12.4 hours

9.76 hours

Short-term toxicity to fish (LC50)

460 μg/l

>19.4 μg/l (LC50 > achievable limit of solubility)

Short-term toxicity to aquatic invertebrates (EC50)


>20 μg/l (EC50 > achievable limit of solubility)

Algal inhibition (ErC50 and NOEC)

ErC50 >550 μg/l;  EC10 90 μg/l

ErC50: >9.4 μg/l; NOEC: ≥9.4 μg/l (ErC50 > highest loading rate, reported as geometric mean measured)

Long-term toxicity to fish (NOEC)

30-d NOEC 0.02 mg/l QSAR prediction.

NB the result is very similar to that for the read-across substance L3

≥27 μg/l (NOEC ≥ achievable limit of solubility)

Long-term toxicity to aquatic invertebrates (NOEC)

80 μg/l

≥14.3 μg/l (NOEC ≥ highest loading rate, reported as geometric mean measured)

Sediment toxicity (NOEC)


88 mg/kg dwt Chironomus riparius; 97 mg/kg dwt Hyalella azteca

Short-term terrestrial toxicity (L(E)C50)



Long-term terrestrial toxicity (NOEC)




Conclusion on classification

The substance is classified as follows:

According to Regulation (EC) No 1272/2008:

Aquatic Acute 1 (Hazard statement: H400: Very toxic to aquatic life.). An M-factor of 1 applies.

Aquatic Chronic 2 (Hazard statement: H411: Toxic to aquatic life with long lasting effects)

According to Directive 67/548/EEC:

N; R50 Dangerous for the environment; Very toxic to aquatic organisms.

Reliable short-term toxicity tests results are available for the effects of the test substance on freshwater fish (Oncorhynchus mykiss), and algae (Selenastrum capricornutum (new name: Pseudokirchneriella subcapitata)). A 96-hour LC50 value of 0.46 mg/l (expressed as measured concentration) or 3.02 mg/l (nominal concentration) has been determined for effects on mortality of O. mykiss. A 70-hour ErC50 value of >0.55 mg/l and ErC10 of 0.09 mg/l have been determined for the effects of the substance on growth rate of S. capricornutum.

In longer-term tests a 90-day NOEC of ≥0.027 mg/l in fish is read across from a structural analogue and a chronic NOEC of ca. 0.02 mg/l has been estimated for the effects of the substance in fish. A 21-day EC50 value of 0.30 mg/l and NOEC of 0.08 mg/l have also been determined for the effects of the substance on reproduction of D. magna.

The substance is hydrolytically unstable with a half-life of 116 h at pH 7 and 25°C, meaning that for classification and labelling purposes, it is considered to be a rapidly degradable substance. The hydrolysis product, trimethylsilanol, has a log Kow of 1.19.