Registration Dossier

Diss Factsheets

Environmental fate & pathways

Phototransformation in air

Currently viewing:

Administrative data

Link to relevant study record(s)

Description of key information

Phototransformation in air: Rate constant for reaction with OH radicals: Parent substance:
Rate constant of
1.49 x 10-12cm3molecule-1second-1 (half-life 10.8 days) (at tropospheric concentration of OH radicals 5E+05 molecule/cm3over 24 hour period (ECHA, 2016).

However, for the purposes of assessment of properties such as long-range transport potential, the measured concentrations are given priority, suggesting a half life in the range 1.7 to 2.7 days.

Key value for chemical safety assessment

Half-life in air:
10.8 d
Degradation rate constant with OH radicals:
0 cm³ molecule-1 s-1

Additional information

D4 is a member of the Reconsile Siloxane Category; siloxanes within the Category do not contain chromophores that would absorb visible or UV radiation, so direct photolysis is not likely to be significant. Indirect photolysis resulting from gas-phase reaction with photochemically-produced hydroxyl radicals occurs.

The Category hypothesis is that the rate of photo-oxidation by hydroxyl radicals of a compound is dependent on the constituent functional groups. A reliable experimental relative rates study with D4 and D5 (Atkinson, R., 1991) found that the NO3radical and O3reactions are of no importance as tropospheric removal processes for these substances. The dominant gas-phase chemical loss process is by reaction with the OH radical. Ameasured rate constant of 1.01 E-12 cm3molecule-1second-1at 24°C (calculated half-life 15.8 days, using tropospheric concentration of OH radicals 5E+05 molecule/cm3over a 24 hour period (ECHA, 2016)) was obtained by the same author.

Further measured data are available for D4. In anexperimental relative rates study, Sommerlade, R.et al.,(1993) found a reaction with the OH radical rate constant of 1.26 E-12 cm3molecule-1second-1at 24°C (calculated half-life 12.7 days, using tropospheric concentration of OH radicals 5E+05 molecule/cm3over 24 hour period (ECHA, 2016).

A measured OH radical rate constant (kOH­) of 0.95 x 10-12cm3/ molecule.sec was determined by Kim and Xu (2017), based on a relative rate method. The value obtained by Sommerlade et al.(2013) of 1.26 x 10-12is considered to be in good agreement. Further measurements by Safron et al. (2015) of1.90 x 10-12cm3/ molecule.sec and Xiao et al.(2015) of 2.34 x 10-12are slightly higher.

The EPIWIN AOPWIN programme has also been used unadapted to obtain values of the rate constant kOHfor reaction with hydroxyl radicals. The overall half-life in air under default conditions of hydroxyl radical concentration is calculated using the following expressions:

kdegair(d-1) = kOH(cm3/molecule.sec) x OH Concair(molecules/cm3) x 24 x 3600

DT50(d) = ln 2/ kdegair(d-1)

Where:

kdegair= total rate constant for degradation in air

kOH= rate constant for reaction with hydroxyl radicals

OH Concair= concentration of hydroxyl radicals in air = 5 E+05 OH molecules/ cm3

DT50= half-life

The concentration of hydroxyl radicals in air of 5 E+05 OH molecules/cm3, and the 24 hour photo period, are the values specified in ECHA Guidance on Information requirements and chemical safety assessment, Part R.16 Environmental exposure estimation (R.16.5.4.3. Photochemical reactions in the atmosphere) (ECHA, 2016).

The results are given in Table 4.1.3.

Table4.1.3Results of AOPWIN photodegradation in air calculations

Parameter

Result, D4

kOH(cm3/ molecule.sec)

1.20 x 10-12

kdegair(d-1)

5.2 x 10-02

DT50(days)

13.3

Kim and Xu (2017) also determined rate constants for three other cyclic siloxanes and four linear siloxanes, and compared the measured mean values of OH radical rate constants with the literature values for each substance. For D4, the measured values from the Kim and Xu (2017) study were close to those measured by Atkinson (1991) and those predicted by AOPWIN, but apparently smaller than those obtained by Safron et al. (2015) and Xiao et al. (2015). The measured values obtained by Kim and Xu for other volatile methyl siloxanes (VMS) also showed variabilities with available literature values (see Table 4.1.4). The arithmetic mean of the measured rate constants is therefore considered to give the best representation of the OH radical rate constant for the relevant VMS species. For D4, the mean measured value of 1.49 x 10-12cm3/ molecule.sec is used as the key value for exposure assessment, equivalent to a half-life in air of 10.8 days.

It should be noted that the true hydroxyl radical concentration is considered by some authorities to be higher than the default used in EU. Therefore, degradation rates may be higher than the values reported here. However, the Annex to this document indicates the half-lives in air necessary to model long-range transport potential.

Table4.1.4Reconsile Siloxane Category: Measured data and AOPWIN predictions for reaction with hydroxyl radicals in air

Substance

Rate constant for reaction with hydroxyl radicals (kOH(cm3/ molecule. sec))

Half-life (days)

Hexamethyldisiloxane (HMDS)

1.19 x 10-12(Sommerladeet al., 1993)

0.90 x 10-12(AOPWIN)

1.38 x 10-12(Atkinson, 1991)

1.32 x 10-12(Markgraf and Wells, 1997)

1.58 x 10-12(Kim and Xu, 2017)

13.5

17.8

11.6

12.2

10.2

Octamethyltrisiloxane (L3)

1.83 x 10-12(Markgraf and Wells, 1997)

2.15 x 10-12(Kim and Xu, 2017)

1.20 x 10-12(AOPWIN)

8.8

7.5

13.4

Decamethyltetrasiloxane (L4)

2.66 x 10-12(Markgraf and Wells, 1997)

3.37 x 10-12(Kim and Xu, 2017)

1.50 x 10-12(AOPWIN)

6.0

4.8

10.7

Dodecamethylpentasiloxane (L5)

4.03 x 10-12(Kim and Xu, 2017)

1.80 x 10-12(AOPWIN)

4.0

8.9

Hexamethylcyclotrisiloxane (D3)

0.90 x 10-12(AOPWIN)

0.52 x 10-12(Atkinson, 1991)

1.84 x 10-12(Xiaoet al. 2015)

0.91 x 10-12(Kim and Xu, 2017)

17.8

30.9

8.7

17.6

Octamethylcyclotetrasiloxane (D4)

1.26 x 10-12 (Sommerlade et al., 1993)

1.20 x 10-12 (AOPWIN)

1.01 x 10-12 (Atkinson, 1991)

1.90 x 10-12 (Safron et al. 2015)

2.34 x 10-12 (Xiao et al. 2015)

0.95 x 10-12 (Kim and Xu, 2017)

12.7

13.4

15.9

8.4

6.9

16.9

Decamethylcyclopentasiloxane (D5)

1.50 x 10-12(AOPWIN)

1.55 x 10-12(Atkinson, 1991)

2.60 x 10-12(Safronet al. 2015)

2.46 x 10-12(Xiaoet al. 2015)

1.46 x 10-12(Kim and Xu, 2017)

10.7

10.4

6.2

6.5

11.0

Dodecamethylcyclohexasiloxane (D6)

2.44 x 10-12(Kim and Xu, 2017)

2.80 x 10-12(Safronet al. 2015)

1.8 x 10-12(AOPWIN)

6.6

5.7

8.9

 

Sommerlade et al.(1993) identified the major products from the reaction of D4 with hydroxyl radicals as heptamethylhydroxycyclotetrasiloxane, along with smaller amounts of heptamethyl(hydroperoxymethyl) cyclotetrasiloxane and 1,2- bis (heptamethylcyclotetrasiloxanyl) ethane, and trace amounts of heptamethyl(hydroxymethyl) cyclotetrasiloxane and bis (heptamethylcyclotetrasiloxanyl) ether.

These degradation products are expected to be more soluble in water than D4, and to have a lower vapour pressure, and so are likely to be removed from the atmosphere by wet and dry deposition (Chandra, 1997). Chandramouli and Kamens (2001) confirmed this deposition process for a related substance (decamethylcyclopentasiloxane, D5). In this study nonamethylhydroxycyclopentasiloxane was identified as the main degradation product from D5 using an outdoor smog chamber that contained fine road dust. More than 99% of the hydroxy derivative formed partitioned onto the dust particles.

Whelan et al.(2004) assessed the atmospheric fate of volatile methyl siloxanes (VMS) and their degradation products. The assessment used a simple equilibrium-partitioning model to investigate the relative rates of removal of two representative VMSs (the linear siloxane decamethyltetrasiloxane, L4, and the cyclic siloxane, D4) and their hydroxyl-substituted degradation products by reaction and atmospheric deposition. The modelling is based on the work of Atkinson, R. (1991) and Sommerlade, R.et al.(1993), which demonstrates that siloxanes break down in the atmosphere to form hydroxyl-substituted degradation products by reaction with OH radicals. As substitution proceeds the silanols become increasingly water-soluble and less volatile, and so tend to be washed out of the atmosphere by wet deposition. The silanols are also assumed to undergo hydrolysis reactions when dissolved in liquid water droplets. Removal of the silanols from the atmosphere by dry deposition is also accounted for. The model indicated that L4 and D4 and the monohydroxy degradation products occur mainly in the vapour phase, whereas the further degradation products occur mainly in the dissolved and particulate phases. Overall, it is concluded that >99% of L4 and D4 are removed from the atmosphere as silanols in wet deposition and <1% are removed in dry deposition.

All the aforementioned studies are focused on homogenous reactions. Although they definitely represent the major characteristics of D4 degradation in the atmospheric environment, the real atmosphere is much more complex. For example, the atmosphere contains a combination of multiple oxidants such as O3, OH and other free radicals, as well as UV radiation and aerosols. In addition, cVMS release also follows distinguishable spatial and temporary patterns: they are released mostly to the urban and suburban atmosphere where the O3, OH radical and aerosol concentrations may be much higher than the rural or remote regions. When cVMS are transported from the source region through the air, they move along with those oxidants, which may increase their exposure to the intensified atmospheric degradation processes. The overall half-lives of cVMS therefore may largely depend on the resident time of cVMS in urban and suburban atmosphere.

In order to better understand the environmental fate of cVMS under the more realistic atmospheric conditions, several projects were initiated both at the University of Iowa and Dow Corning Corporation. The specific objectives of those studies are twofold: To determine the removal of gas-phase cVMS by combination of multiple oxidants in the presence of UV radiation and aerosols; to determine the effects of the unique release pattern of cVMS and any additional removal mechanism on the overall half-lives of cVMS in the atmosphere.

In University of Iowa studies, the uptake of D4 and D5 vapours (up to saturated vapour pressure) by carbon black and several types of reactive mineral dust aerosols in the absence and presence of O3and hydroxyl radicals was simulated in an atmospheric chamber at room temperature and monitored by FT-IR spectroscopy (Navea et al., 2009a; b; c). It was found that the heterogeneous uptake (removal from gas phase) of D4 and D5 by mineral aerosols such as kaolinite, hematite and quartz was rapid and significant (Navea et al., 2009a). Under dry conditions (<1% RH) in the absence of O3and hydroxyl radicals, removal of both D4 and D5 by hematite and kaolinite were characterised by two processes. The initial fast removal process completed within one minute accounts for 30 to 50% total removal for D4, but 50~70% for D5. The subsequent slow removal process had a pseudo-first order kinetics with rate constant k at room temperature varying in the range of 0.5~0.9 x 10-3s-1(corresponding to half-lives from 24~13 minutes) for D4, and in a range of 0.2~4 x 10-3s-1(corresponding to half-lives from 53 minutes to 12 minutes) for D5 for hematite and kaolinite, respectively. The reactivity of various aerosols was in the order, kaolinite >hematite >quartz, for both D4 and D5 after the reaction rates were normalised to surface area of the aerosols.

In the presence of O3but no solar radiation, no detectable change was observed in the concentration of the gas phase D4 or D5 after 50 minutes’ exposure (Naveaet al., 2009b), consistent with the low rate constant obtained for O3/D4 reactions in the previous study (Atkinson, 1991). Under dry (<1% RH) conditions, the introduction of aerosols such as hematite and kaolinite triggered immediate removal of both cVMS and O3(Naveaet al., 2009b). The kinetics of the gas phase removal apparently implied multiple heterogeneous processes, significantly different than that obtained with aerosols in the absence of O3. The major differences were in two aspects: first, addition of O3slowed the uptake of the both D4 and D5 by the aerosols, while existence of D4 and D5 in the gas phase also slowed the decomposition of gas phase O3relative to the aerosol-containing control with no cVMS, suggesting the competition of cVMS with O3for surface sites of the aerosols (Naveaet al., 2009b). In addition, the constant concentration profile observed at reaction times greater than 200 minutes in the cVMS/aerosol systems without O3was replaced with a linear decline in D4 and D5 concentrations with increase of the reaction time when O3was present. The disappearance of the surface saturation characteristics (constant concentration profile as time increases) increased the total removal over a longer time period (Naveaet al., 2009b).

Under simulated solar radiation and in the presence of O3, hematite and kaolinite aerosols remove up to 50 to 70 % gas phase D4 and 60 to 90% gas phase D5 within 400 minutes under dry condition (<1% RH) (Navea, J. G. et al., 2009c). An increase in humidity under those conditions actually accelerates the removal presumably due to formation of hydroxyl radicals through photolysis of O3in the presence of water.

The University of Iowa’s studies were conducted under relatively high concentrations of cVMS (>10 mg L-1) due to limitation of the non-destructive analytical technique employed for cVMS analysis. In follow-up studies conducted at Dow Corning Corporation (Kimet al., 2009; Kim and Xu, 2009a; b), the mechanism for D4 sorption in the high concentration range is verified as polymerisation of the sorbed D4 catalysed by clay surface. Modeling assessment results suggested that aerosol effect on the overall D4 degradation in natural environment should be relatively small by this polymerisation mechanism (Navea, J.et al., 2011). However, D4 concentration in atmosphere is in nanogram to micrograms per cubic meter, or 3 to 6 orders of magnitude lower than those tested in Navea’s studies.

At low D4 concentration range, reactive adsorption via depolymerisation was observed on aerosol surface (Kim and Xu, 2009b). Under those conditions (initial CD4 <0.3 mg/l, RH 10~80%), 60% to 97% of sorbed D4 was not desorbable (Kim and Xu, 2009a; b). Surface speciation analysis of the sorbed D4 revealed that almost all sorbed D4 in this low concentration range was transformed to silanols within 2 hrs at 28% RH, due to surface-facilitated hydrolysis (or depolymerisation) (Kim, J. and Xu, S., 2009b), similar to that found in dry soil (Xu, S. 1999; Xu, S. and Chandra, G., 1999).

Similar studies have been carried out with D5 (Kim and Xu, 2010 and 2011), and showed similar results, i.e. the mechanism for adsorption at high concentration range is polymerisation of the sorbed D5 by the clay surface, whereas at the low D5 concentration range, reactive adsorption via depolymerisation was observed on the aerosol surface (sorbed D5 being transformed to silanols and smaller cyclics (D3 and D4) as transient intermediates).

A recent study (Kim and Xu, 2016) investigated the sorption and desorption behaviours of airborne VMSs (including D4) on nine major primary and secondary atmospheric aerosols (RH 30%). It was found that sorption and desorption of VMS took place via a two-phase process, which included an initial rapid step, followed by slower subsequent step. The initial rapid step was favoured especially at low concentrations. Some aerosols such as carbon black and sea salts interacted reversibly with D4 whereas other aerosols such as kaolinite and sulfates showed highly irreversible sorption for the VMS, especially at low concentrations. Values of apparent aerosol-air partition coefficients ranged 0.09-50.4 L/m(2) for D4, with carbon black having the largest values.

These results suggest that the heterogeneous interaction of D4 with mineral aerosols, therefore, can be an important mechanism in reducing the concentrations and transport of this volatile siloxane compound in the environment. The exact effects from this depolymerisation by aerosol on the half-life of airborne D4 could not be estimated at this juncture. Nevertheless, the actual half-life of D4 in air should be shorter than that calculated based solely on the homogenous reaction rate.

More recent work using actual field monitoring data has tested this hypothesis (Xu et al., 2017), which has been drawn from extensively in the Annex to this document. Evidence is presented there which shows that on the basis of measurements of the concentration of D4 in the northern hemisphere, the degradation rate of D4 in air could be around six times faster than the rate based on the rate constant (Atkinson 1991) reported above,indicating that there may be additional removal processes for airborne D4 currently not accounted for in the determination of half-life in air based on photodegradation alone.