Registration Dossier

Administrative data

Hazard for aquatic organisms

Freshwater

Hazard assessment conclusion:
PNEC aqua (freshwater)

Marine water

Hazard assessment conclusion:
PNEC aqua (marine water)

STP

Hazard assessment conclusion:
PNEC STP

Sediment (freshwater)

Hazard assessment conclusion:
PNEC sediment (freshwater)
PNEC value:
2 mg/kg sediment dw
Extrapolation method:
equilibrium partitioning method

Sediment (marine water)

Hazard assessment conclusion:
PNEC sediment (marine water)
PNEC value:
0.2 mg/kg sediment dw
Extrapolation method:
equilibrium partitioning method

Hazard for air

Air

Hazard assessment conclusion:
no hazard identified

Hazard for terrestrial organisms

Soil

Hazard assessment conclusion:
PNEC soil
PNEC value:
0.25 mg/kg soil dw
Extrapolation method:
equilibrium partitioning method

Hazard for predators

Secondary poisoning

Hazard assessment conclusion:
no potential for bioaccumulation

Additional information

The hydrolysis half-life of 1,1,1,3,3,3-hexamethyldisilozane (HMDZ) is <<1 min at pH 7, 25°C. The registered substance will hydrolyse in contact with water and atmospheric moisture to trimethylsilanol (TMS) (CAS 1066-40-6) and ammonia. REACH guidance (ECHA 2010A, R.16) states that “for substances where hydrolytic DT50 is less than 12 hours, environmental effects are likely to be attributed to the hydrolysis product rather than to the parent itself”. TGD and ECHA guidance, (EC 2003, ECHA 2010A) also suggest that when the hydrolysis half-life is less than 12 hours, the breakdown products, rather than the parent substance, should be evaluated for aquatic toxicity. Therefore, the environmental hazard assessment, including sediment and soil compartments due to water and moisture being present, is based on the properties of the silanol hydrolysis product, in accordance with REACH guidance. As described in Section 1.3, the silanol hydrolysis products may be susceptible to condensation reactions.

READ-ACROSS JUSTIFICATION

In order to reduce animal testing read-across is proposed to fulfil up to 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, because after hydrolysis occurs the resulting product has different structural features, physicochemical properties and behaviour.

Reliable short-term data are available for the registration substance. These data are used for concluding on the classification and labelling. However, given the rapid hydrolysis of hexamethyldisilazane, it is appropriate to carry out risk characterisation on the basis of the properties of the two hydrolysis products. Therefore, data for trimethylsilanol and ammonia are also considered in the assessment and PNECs are based on these data.

In the following paragraphs the approach taken to the assessment of hydrolysis products of 1,1,1,3,3,3-hexamethyldisilazane is outlined, taking into account structure, hydrolysis rate and physicochemical properties.

 

Data for the hydrolysis product trimethylsilanol

1,1,1,3,3,3 -Hexamethyldisilazane (CAS 999-97-3) is a monosilazane that hydrolyses rapidly in contact with water to form hydroxytrimethylsilane (known as trimethylsilanol CAS 1066-40-6) and ammonia (<<1 minute at pH 7, 25°C). The surrogate substance is trimethylsilanol (or hydroxytrimethylsilane, CAS 1066-40-6), the hydrolysis product of the registration substance. Trimethylsilanol is a monosilanol containing three methyl groups and is not subject to further hydrolysis. The aquatic data available for this substance indicates that it is of low short-term toxicity to aquatic organisms, with L(E)C50 values in the range 124 to >750 mg/l.

 

Considerations on the non-silanol hydrolysis products:

Ammonia is a known toxicant for aquatic organisms, in particular to fish and invertebrates. This compound drives the aquatic toxicity of the substance and is assessed separately to the silanol hydrolysis product. Further details on the assessment of the substance are presented in the toxicity test results section that follows.

 

Table 1 Summary of physico-chemical and ecotoxicological properties of the registered and surrogate substances.

CAS Number

999-97-3

1066-40-6

7664-41-7

Chemical Name

1,1,1,3,3,3-hexamethyldisilazane

Trimethylsilanol

Ammonia

Si hydrolysis product

Trimethylsilanol

n/a

n/a

Molecular weight (parent)

161.4

90.2

*17.03

Molecular weight (hydrolysis product)

90.2

n/a

 

log Kow(parent)

n/a

1.2

** 0.23

log Kow(silanol hydrolysis product)

1.2

n/a

n/a

Water sol (parent)

n/a

20000 mg/l but concentration dissolved in water may be limited above approximately 1000 mg/l by condensation reactions

**531000 mg/L

Water sol (silanol hydrolysis product))

20000 mg/l but concentration dissolved in water may be limited above approximately 1000 mg/l by condensation reactions

n/a

n/a

Vapour pressure (parent)

10.1 Pa

1900 Pa

*861100 Pa or 786.7 kPa

Vapour pressure (hydrolysis product)

1900 Pa

n/a

 

Hydrolysis t1/2at pH 7 and 25°C

<<1 min at 25°C:

n/a

 

Hydrolysis t1/2at pH 4 and 25°C

very rapid

n/a

 

Hydrolysis t1/2at pH 9 and 25°C

very rapid

n/a

 

Short-term toxicity to fish (LC50)

88 mg/l

270 mg/l

LC50 0.083 mg NH3/l

Short-term toxicity to aquatic invertebrates (EC50)

80 mg/l

124 mg/l

**LC50 0.16 mg NH3/l

Algal inhibition (ErC50and NOEC)

EC50: 50 mg/l and NOEC: 7.5 mg/l

EC50: >750 and NOEC:70 mg/l

**EC50 2.0 mg NH3/l

LOEC 3.0 mg NH3/l

Long-term toxicity to fish (NOEC)

n/a

n/a

NOEC 0.014 mg NH3/l

Long-term toxicity to aquatic invertebrates (NOEC)

n/a

n/a

**NOEC 0.066 mg NH3/l

Long-term sediment toxicity (NOEC)

n/a

n/a

n/a

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

n/a

n/a

* Plants LOEC Range: 3-40 ppm

Long-term terrestrial toxicity (NOEC)

n/a

n/a

n/a

**Source: Environment Agency Proposed EQS for Water Framework Directive Annex VIII Substances: Ammonia (unionised) 2007.

* Source : (D0966 SIDS Dossier Ammonia 7664417) Ammonia data sourced from SIAR for the Ammonia category (OECD, 2007; sponsored by USA, shared partnership with European Fertilizers Manufacturing Association (EFMA-Europe) and The Fertilizer Institute (TFI-US)).

Table 2: Comparison of short-term test results (L(E)C50's) for the registered substance and it's hydrolysis products, taking into account molecular weight adjustments and pH values

 Test Organism  Test pH  (in test with HMDZ)  HMDZ mg/L  Trimethylsilanol (TMS) mg/L as HMDZ (derived from the short-term toxicity (L(E)C50) results from the tests with TMS)  Unionised Ammonia (NH3) mg/L as HMDZ (derived from the short-term toxicity (L(E)C50) results from the tests with ammonia)
 Fish     Range  7.8 - 8.7 88     242     8.6
 Mean  8.25
 Aquatic Invertebrates     Range  6.6 - 8.5 80  111      77   
 Mean  7.55
 Alga        Range     Start of test: 8.5 - 9.3 50     7671         46      
 End of test: 8.7 - 9.8
 Mean  9.1

The results of the studies for TMS and ammonia (NH3) have been converted into the equivalent amount of HMDZ that would generate that amount of TMS or NH3 upon full hydrolysis. For ammonia, the L(E)C50 results from the short-term tests were 0.083, 0.16 and 2.0 mg unionised NH3/l for fish, invertebrates and algae respectively. Taking into account the mean pH experienced in the short-term tests with HMDZ (to give the fraction of the total ammonia ionised) and the MW of HMDZ and ammonia, the amount of HMDZ that would produce an equivalent amount of unionised NH3 to the L(E)C50 values seen in the ammonia short-term tests was calculated. The calculated numbers for NH3 can therefore be interpreted as being the L(E)C50s that would be expected for HMDZ based on the data for ammonia. The fact that the actual L(E)C50s are equivalent to or lower than these calculated numbers indicates that HMDZ is not more toxic than would be expected based on its capacity to generate NH3.

Additional information about the aquatic ecotoxicity of ammonia

Ammonia is known to be toxic to the aquatic environment and has been reviewed by a number of different regulatory programmes including the SIAR for the Ammonia category (OECD, 2007; sponsored by USA, shared partnership with European Fertilizers Manufacturing Association (EFMA-Europe) and The Fertilizer Institute (TFI-US)), the Canadian Environmental Protection Act, 1999, Priority Substance List Assessment Report, Ammonia in the Aquatic Environment, and the UK Environment Agency Proposed EQS for Water Framework Directive Annex VIII Substances: Ammonia (unionised) 2007. The following information is taken from these sources; further review of the data has not been considered necessary. For the purpose of this chemical safety report, data for PNEC derivation has been taken from the UK Environment Agency (EA) EQS, where the EA have taken a thorough look at data which the authors of the present report believe is adequate for risk characterisation for REACH.

Ammonium salts dissociate in water to give ammonia/ammonium ion and, therefore, data for these salts may be used to assess the toxicity of ammonia.

The ammonia/ammonium ion in aqueous solution exists in equilibrium between NH3 and NH4+, depending on the pH. The unionised species is by far the most toxic; the ammonium ion is assumed to be essentially non-toxic. Therefore, toxicity due to ammonia increases with increasing pH (Clement Associates, 1990; USEPA, 1999), because the fraction of unionised ammonia increases according to the following equation (Emerson, 1975):

Fraction unionised = 1/(10pKa-pH+1)

At pH 8.5, the proportion of unionised ammonia is approximately 10 times that at pH 7.5.The concentration of unionised ammonia will be lower at higher ionic strengths of very hard fresh water or salt water environments. This effect can be significant in estuarine and marine waters. (Environment Canada, 2001b). Moreover, the pKa is reciprocally related to temperature. For every 9°C increase in temperature, the proportion of unionised ammonia approximately doubles (EA, 2007).

 

PNECaquatic-freshwater for hexamethyldisilazane may be derived in two ways:

  • From acute data for the silanol hydrolysis product, trimethylsilanol.
  • From the chronic data for ammonium compounds.

Acute data for trimethylsilanol

Fish: LC50 (96 h): 271 mg/l (equivalent to 242 mg/l hexamethyldisilazane)

Daphnia: EC50 (48 h): 124 mg/l (equivalent to 111 mg/l hexamethyldisilazane)

Algae: EC50 (96 h): >750 mg/l (equivalent to >671 mg/l hexamethyldisilazane)

The standard approach to PNEC calculation when acute data are available for three trophic levels is to apply an assessment factor of 1000 to the lowest lethal or effect concentration (E(L)C50). However, this assessment factor may be reduced by up to a factor of 10 if this is justified by the available data, for example:

  • Evidence from structurally similar compounds which may demonstrate that a higher or lower factor may be appropriate.
  • Knowledge of the mode of action as some substances by virtue of their structure may be known to act in a non-specific manner. A lower factor may therefore be considered. Equally a known specific mode of action may lead to a higher factor (ECHA 2009).

 

Trimethylsilanol is part of a category of organosilicon substances containing only alkyl, halogen, alkoxy or silanol groups attached to the Si atom. The category has a low hazard profile and data are consistent with a simple polar narcosis mechanism. It is considered that the weight of evidence for the number of substances (ca.40 substances have reliable measured data) justifies a reduction of the assessment factor from the normal value of 1000. A value of 500 is used, to reflect the increased confidence in the individual values due to the low variability across the category. This is discussed further in PFA, 2013 Y.

The lowest L(E)C50 value for trimethylsilanol is 124 mg/l for invertebrates. Therefore, PNECaquatic is 124/500 = 0.25 mg/l.

Acute data for ammonia

Short-term PNECaquatic for ammonia has been derived by the UK Environment Agency for the purpose of setting an Environmental Quality Standard (EQS) under Annex VIII of the Water Framework Directive (WFD). This PNEC is considered to also be adequate for the purpose of the risk characterisation of ammonia derived from hydrolysis of HMDZ.

Fish are the most sensitive taxon to acute exposure with the lowest valid acute LC50 value being 0.068 mg/l of nitrogen present as NH3 for Oncorhynchus gorbuscha (pink salmon) alevins exposed to ammonia for a 96-hour period. Similar sensitivity was evident in other studies with the same species, and Prosopium williamsoni (mountain whitefish) and the crustacean Hyalella were only slightly less sensitive. The recommended PNECfreshwater of 6.8 μg l-1 un-ionised ammonia is based on the 96-hour LC50 for pink salmon and application of a reduced assessment factor of 10 (instead of 100) since this datum is supported by an extensive body of toxicological data for a wide range of taxa. There is no existing short-term EQS for ammonia. The LC50 of 0.068 mg/l of nitrogen present as NH3 is equivalent to 0.083 mg NH3/L (unionised ammonia). The PNECfreshwater can then be derived for unionised ammonia by applying an assessment factor of 10, therefore PNECfreshwater = 0.0083 mg NH3/l.

Assuming a pH 7.0, the ratio of NH3 to NH4+ in solution is 1:189. This means that a PNEC of 0.0083 mg NH3/l is equivalent to 1.57 mg ammonia/ammonium/l. This is equivalent to 14.1 mg/l of hexamethyldisilazane. In reality, the PNEC in terms of total loading would vary significantly according to test conditions.

Chronic data for ammonia

Long-term PNECaquatic for ammonia has been derived by the UK Environment Agency for the purpose of setting an Environmental Quality Standard (EQS) under Annex VIII of the Water Framework Directive (WFD). This PNEC is considered to also be adequate for the purpose of the risk characterisation of ammonia derived from hydrolysis of HMDZ.

The lowest credible concentration of unionised ammonia at which long-term effects were found is 0.022 mg/l of nitrogen present as NH3 when a cumulative mortality of 71 per cent was observed for eggs, larvae and fry of rainbow trout (Oncorhynchus mykiss) over 73 days exposure. No NOEC value was derived in the study since effects were observed at the lowest exposure concentration. However, data on the concentration–response curve for similar effects in other fish have been evaluated, and these indicate 2–3 times difference in exposure concentration for a 50 per cent reduction in survival (from the controls) in such an early life stage test (EA 2007).

The PNECfreshwater from long-term data, can be derived based on the 73-day LOEC of 0.022 mg/l of nitrogen present as NH3 to which a factor of 2 is applied to estimate a NOEC and an assessment factor of 10 to extrapolate to a PNECfreshwater of 1.1 μg l-1 unionised ammonia. This factor is justified on the basis of a large body of data, including a multigenerational study with a sensitive species and a series of simulated ecosystem studies (EA 2007).

The LOEC of 0.022 mg/l of nitrogen present as NH3 is the equivalent to 0.027 mg NH3/L (unionised ammonia). This value can also be converted to a NOEC by applying a factor of 2. This gives a NOEC for unionised ammonia of 0.014 mg NH3/L. The PNECfreshwater can then be derived for unionised ammonia by applying an assessment factor of 10 therefore PNECfreshwater = 0.0014 mg NH3/L. Assuming that the contribution of NH4+ to toxicity may be considered negligible, the PNEC in terms of ammonia/ammonium total loading may be calculated.

Assuming a pH 7.0, the ratio of NH3 to NH4+ in solution is 1:189. This means that a PNEC of 0.0014 mg NH3/l is equivalent to 0.26 mg ammonia/ammonium/l. This is equivalent to 2.4 mg/l of hexamethyldisilazane. In reality, the PNEC in terms of total loading would vary significantly according to test conditions.

Unionised ammonia is toxic to aquatic organisms at concentrations below 1 mg/l. Fish are more sensitive to ammonia toxicity, followed by Daphnia then algae. The toxicity of ammonia is highly dependent on the pH and temperature of the test system, since the fraction of unionised ammonia will decrease with lowering pH and temperatures.

The lowest reliable NOEC for long-term effects of ammonia on marine biota is 0.066 mg/l of nitrogen present as NH3 for growth of Dover sole (Solea solea). Short-term test data are available that indicate this may not be the most sensitive taxonomic group (molluscs and echinoderms are more sensitive following acute exposure) and so an assessment factor of 100 is recommended to protect these taxa, resulting in a PNECsaltwater of 0.66 μg l-1 unionised ammonia (EA 2007).

At pH 7.0, this is equivalent to 0.12 mg ammonia/ammonium/l.

The marine dataset is too small to draw firm conclusions on possible differences between the sensitivities of freshwater and saltwater organisms, particularly with regard to chronic effects. However, marine taxa do not appear to differ markedly from the range of sensitivities obtained for corresponding freshwater species. Consequently, freshwater toxicity data for ammonia have been considered in the derivation of PNEC values for the protection of marine life.

Using the PNECfreshwater result of 0.0014 mg NH3/l and applying an additional assessment factor of 10, this results in a PNECaqua (marine water) of 0.00014 mg NH3/L.

Assuming that the contribution of NH4+to toxicity may be considered negligible, the PNEC in terms of ammonia/ammonium total loading may be calculated.

Due to the increased pH of saltwater, a pH of 8 has been assumed and the ratio of NH3 to NH4+ in solution is 1:19.8. This means a PNECaqua (marine water) of 0.00014 mg NH3/L is equivalent to 0.0028 mg ammonia/ammonium/l.

Conclusion

Given the rapid hydrolysis of hexamethyldisilazane, it is appropriate to carry out risk characterisation on the basis of the properties of the two hydrolysis products. Therefore, the PNECs based on data for trimethylsilanol and ammonia are used.

No suitable data for the toxicity of the substance or its silanol hydrolysis product to marine organisms has been identified. Therefore, PNECaquatic-marine is derived from PNECaquatic-freshwater by application of an additional assessment factor of 10.

The documented PNECs are considered adequate for the present REACH risk characterisation. They have been calculated using the current ECHA guideline, including the most conservative assessment factors, and are used for the registration under the regulation 1907/2006 dated June 1st 2007 (REACH) only. They should not be used for other regulatory purposes (e.g., OELs) without further consideration and evaluation.

Conclusion on classification

Reliable acute toxicity tests results are available for freshwater fish (Brachydanio rerio), invertebrates (Daphnia magna) and algae (Scenedesmus subspicatus). LC50 or EC50 values for the three organisms were all between 10 and 100 mg/l. No chronic studies are available for fish or invertebrates. The substance hydrolyses rapidly, however, the hydrolysis product trimethylsilanol may persist in the environment. Therefore, the proposed classification and labelling is as follows:

According to Regulation (EC) No 1272/2008:

Aquatic Acute: Not classified.

Aquatic Chronic: Category 3

According to Directive 67/548/EEC:

R52-53 Harmful to aquatic organisms, may cause long-term adverse effects in the aquatic environment.