Di-tert-butyl 1,1,4,4-tetramethyltetramethylene diperoxide

Hydrolysis

Link to relevant study record(s)

Reference

Reference 1

Endpoint:

hydrolysis

Type of information:

experimental study

Adequacy of study:

key study

Study period:

Experimental starting date: 26 July 2018 Experimental completion date: 02 October 2018

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

OECD Guideline 111 (Hydrolysis as a Function of pH)

Version / remarks:

At the request of the Sponsor, the procedure was modified such that analysis of the sample solutions was completed by solvent extraction performed in the original vessels used for incubation of the sample solutions. This was since the Sponsor believed that adsorption onto the surfaces of the vessels may be a significant influence on the observed sample solution concentrations; perhaps artificially skewing previous studies and resulting in a systematic overestimation of the rate of hydrolysis.

Deviations:

no

Qualifier:

according to guideline

Guideline:

EU Method C.7 (Degradation: Abiotic Degradation: Hydrolysis as a Function of pH)

Version / remarks:

At the request of the Sponsor, the procedure was modified such that analysis of the sample solutions was completed by solvent extraction performed in the original vessels used for incubation of the sample solutions. This was since the Sponsor believed that adsorption onto the surfaces of the vessels may be a significant influence on the observed sample solution concentrations; perhaps artificially skewing previous studies and resulting in a systematic overestimation of the rate of hydrolysis.

Deviations:

no

GLP compliance:

yes (incl. QA statement)

Specific details on test material used for the study:

Identification: Di-tert-butyl 1,1,4,4-tetramethyltetramethylene diperoxide
CAS no.: 78-63-7
Water solubility: 152 µg/L
Appearance/physical state: clear colorless liquid
Lot No.: 1002516115
Purity: see certificate of analysis
Expiry date: 01 March 2020
Storage conditions: room temperature, in the dark

Radiolabelling:

no

Details on sampling:

For analysis, 4 mL of internal standard solution was added directly to duplicate 25 mL aliquots of sample solution in the original vessels used for preparation and incubation (if relevant) of the test item solutions. The samples were then shaken horizontally on a flatbed shaker for a period of 15 minutes. After shaking, the phases were allowed to separate and the upper organic phase (i.e. the internal standard solution containing the extracted test item) then vialled for analysis.

Buffers:

Buffer solution

pH 4
Citric acid 0.06 mol dm-3
Sodium chloride 0.04 mol dm-3
Sodium hydroxide 0.07 mol dm-3

pH 7
Disodium hydrogen orthophosphate (anhydrous) 0.03 mol dm-3
Potassium dihydrogen orthophosphate 0.02 mol dm-3
Sodium chloride 0.02 mol dm-3

pH 9
Disodium tetraborate 0.01 mol dm-3
Sodium chloride 0.02 mol dm-3

Prior to use, these solutions were subjected to vacuum filtration, ultrasonication and degassing with nitrogen, to minimize their dissolved oxygen content.

Details on test conditions:

Test System
The determination was initiated using a procedure designed to be compatible with Method C.7 Abiotic Degradation, Hydrolysis as a Function of pH of Commission Regulation (EC) No 440/2008 of 30 May 2008 and Method 111 of the OECD Guidelines for Testing of Chemicals, 13 April 2004. At the request of the Sponsor, the procedure was modified such that analysis of the sample solutions was completed by solvent extraction performed in the original vessels used for incubation of the sample solutions. This was since the Sponsor believed that adsorption onto the surfaces of the vessels may be a significant influence on the observed sample solution concentrations; perhaps artificially skewing previous studies and resulting in a systematic overestimation of the rate of hydrolysis.

The test system consisted of sterile buffer solutions at pH’s 4, 7 and 9.

Performance of the Test
Preparation of the Test Solutions
Multiple sample solutions were prepared in capped glass vessels at a nominal concentration of 7.0 x 10-5 g/L (70 µg/L) for each of the three buffer solutions. This was achieved by dosing individual 25 mL aliquots of buffer solution with 200 µL of a stock solution of test item prepared at a nominal concentration of 9.0 mg/L in acetonitrile. This also achieved an acetonitrile co-solvent of 0.8% v/v to aid dissolution.
For each pH, at each time point, duplicate, independently incubated vessels of test item sample solution were taken for analysis. With the exception of testing detailed in Section 4.2 of this report, extraction of sample solutions for analysis was performed in the original vessels used for the preparation and incubation (if relevant) of the sample solutions.

The solutions were shielded from light whilst maintained at the test temperature.

Preliminary Test
No preliminary test was required for the test item, as from existing studies and the chemical structure, significant hydrolysis of the test item, i.e. a half-life of less than one year at 25 °C, could be readily predicted.

Tier 2
Testing was initiated at a temperature of 30 °C, to maximize the theoretical aqueous solubility of the test item, with the intention that testing was then to be performed at 20 °C and 10 °C, in order to maximize accuracy across environmentally relevant temperatures.

Sample solutions at all three pH’s were maintained at a temperature of 30.0 ± 0.5 °C for a period of at least 193 hours.

Duration:

193 h

pH:

4

Temp.:

30 °C

Initial conc. measured:

>= 0 - <= 0 g/L

Duration:

193 h

pH:

7

Temp.:

30 °C

Initial conc. measured:

>= 0 - <= 0 g/L

Duration:

193 h

pH:

9

Temp.:

30 °C

Initial conc. measured:

>= 0 - <= 0 g/L

Number of replicates:

2

Positive controls:

no

Negative controls:

no

Preliminary study:

Although the plots of the logarithm of concentration versus time approximated to second order kinetics (such a plot would be linear for pseudo-first order kinetics), analytical investigations into the influence of adsorption onto the surfaces of the test vessels invalidated the analytical data for consideration of the hydrolysis kinetics of the test item.

With reference to testing detailed previously, the test item was demonstrated to show a high affinity for adsorption to glassware. This testing indicated that test item dissolved in the aqueous buffer solutions may actually make no contribution to the concentration of test item detected on analysis. As such, it was concluded not to be possible to accurately quantify the hydrolysis kinetics of the test item whilst remaining within the sample solution concentration and co-solvent content constraints of the regulatory methods.

Transformation products:

no

Key result

pH:

4

Temp.:

30 °C

Remarks on result:

other: Due to this proven adsorption of the test item to glassware, it was not possible to identify and quantify the decreases in sample solution concentrations due to hydrolysis alone.

Key result

pH:

7

Temp.:

30 °C

Remarks on result:

other: Due to this proven adsorption of the test item to glassware, it was not possible to identify and quantify the decreases in sample solution concentrations due to hydrolysis alone.

Key result

pH:

9

Temp.:

30 °C

Remarks on result:

other: Due to this proven adsorption of the test item to glassware, it was not possible to identify and quantify the decreases in sample solution concentrations due to hydrolysis alone.

Details on results:

Discussion
On analysis of the aqueous buffer solution and glass vessel components of incubated test systems individually; it was demonstrated that the aqueous buffer solution made no contribution to the concentration of test item detected on analysis of the test system as a whole.
This finding invalidates the analytical procedure of extracting sample solutions in the original incubation vessel as a suitable method for quantification of the hydrolysis rate kinetics for the test item. This was since such a procedure risked underestimating the hydrolytic rate, as analyzed concentrations included a contribution from test item which was not available for hydrolysis.
The alternative procedure of analysing the aqueous solution only was also concluded as being unreliable. This was since analytical monitoring of the sample solution concentrations only would systematically overestimate the rate of hydrolysis, as analytical losses would also include a contribution from adsorption.

Evaluation of the distribution of the test item in the test system indicated that test item dissolved in the aqueous buffer solutions may actually make no contribution to the concentration of test item detected on analysis. 

Due to this proven adsorption of the test item to glassware, it was not possible to identify and quantify the decreases in sample solution concentrations due to hydrolysis alone. Analytical monitoring of the sample solution concentrations only would systematically overestimate the rate of hydrolysis, as analytical losses would also include a contribution from adsorption. However, inclusion of the adsorbed fraction in the analytical procedure, i.e. extraction of the incubated sample solutions within the original test vessels, risked underestimating the hydrolytic rate, as analyzed concentrations included a contribution from test item which was not available for hydrolysis.

Test item concentration in analysed samples

 

pH	Sample Identity	Replicate	Analyzed Concentration (g/L)
4	“Complete” Sample	A	2.67 x 10^-5
		B	2.98 x 10^-5
	“Solution Only” Sample	A	None detected
		B	None detected
	“Glass Only” Sample	A	2.92 x 10^-5
		B	2.45 x 10^-5
7	“Complete” Sample	A	2.57 x 10^-5
		B	2.71 x 10^-5
	“Solution Only” Sample	A	None detected
		B	None detected
	“Glass Only” Sample	A	1.92 x 10^-5
		B	1.57 x 10^-5
9	“Complete” Sample	A	1.55 x 10^-5
		B	2.36 x 10^-5
	“Solution Only” Sample	A	None detected
		B	None detected
	“Glass Only” Sample	A	1.95 x 10^-5
		B	1.98 x 10^-5

 

PLEASE REFER TO THE ATTACHED DOCUMENTS FOR:

1) Examples of typical chromatography

2) Instrument responses

Validity criteria fulfilled:

yes

Conclusions:

Evaluation of the distribution of the test item in the test system indicated that test item dissolved in the aqueous buffer solutions may actually make no contribution to the concentration of test item detected on analysis.
Due to this proven adsorption of the test item to glassware, it was not possible to identify and quantify the decreases in sample solution concentrations due to hydrolysis alone. Analytical monitoring of the sample solution concentrations only would systematically overestimate the rate of hydrolysis, as analytical losses would also include a contribution from adsorption. However, inclusion of the adsorbed fraction in the analytical procedure, i.e. extraction of the incubated sample solutions within the original test vessels, risked underestimating the hydrolytic rate, as analyzed concentrations included a contribution from test item which was not available for hydrolysis.

Executive summary:

The hydrolysis behavior of di-tert-butyl 1,1,4,4-tetramethyl tetramethylene diperoxide (CAS 78 -63 -7), as a function of pH, has been assessed using a procedure designed to be compatible with Method C.7 Abiotic Degradation, Hydrolysis as a Function of pH of Commission Regulation (EC) No 440/2008 of 30 May 2008and Method 111 of the OECD Guidelines for Testing of Chemicals, 13 April 2004. At the request of the Sponsor, the procedure was modified such that analysis of the sample solutions was completed by solvent extraction performed in the original vessels used for incubation of the sample solutions. This was since the Sponsor believed that adsorption onto the surfaces of the vessels may be a significant influence on the observed sample solution concentrations; perhaps artificially skewing previous studies and resulting in a systematic overestimation of the rate of hydrolysis.

The test item was demonstrated to show a high affinity for adsorption to glassware, such that it was concluded not to be possible to accurately quantify the hydrolysis kinetics of the test item whilst remaining within the sample solution concentration and co-solvent content constraints of the regulatory methods.  Additional investigative testing indicated that test item dissolved in the aqueous buffer solutions may actually make no contribution to the concentration of test item detected on analysis. 

Due to this proven adsorption of the test item to glassware, it was not possible to identify and quantify the decreases in sample solution concentrations due to hydrolysis alone. Analytical monitoring of the sample solution concentrations only would systematically overestimate the rate of hydrolysis, as analytical losses would also include a contribution from adsorption. However, inclusion of the adsorbed fraction in the analytical procedure, i.e. extraction of the incubated sample solutions within the original test vessels, risked underestimating the hydrolytic rate, as analyzed concentrations included a contribution from test item which was not available for hydrolysis. 

Such adsorption may be reduced through the use of a higher initial fortified test item concentration, combined with an increased organic solvent co-solvent content in the aqueous buffer solutions to achieve dissolution; however the regulatory methods limit the maximum permitted parameters to an initial fortified test item of less than half the water solubility of the test item and a maximum co-solvent content of 1% v/v.

Reference 2

Endpoint:

hydrolysis

Type of information:

experimental study

Adequacy of study:

supporting study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

guideline study with acceptable restrictions

Qualifier:

according to guideline

Guideline:

OECD Guideline 111 (Hydrolysis as a Function of pH)

Version / remarks:

only tier 3 was conducted.

Deviations:

yes

GLP compliance:

no

Specific details on test material used for the study:

Chemical name: 2,5-Dimethyl-2,5-di(tet-butylperoxy)hexane
EC name: Di-tert-butyl 1,1,4,4-tetramethyltetramethylene diperoxide
CAS number: 78-63-7
Batch/lot no.: 1002516115
Purity: 96.2%
Expiration date: 03-01-2020
Storage : Refrigerated ± 5°C

Radiolabelling:

no

Analytical monitoring:

yes

Details on sampling:

A stock solution of the test substance was prepared in dichloromethane, this stock solution was used to spike the test substance to the M4 test media, An amount of 100 μL of the test substance stock solution was spiked to the M4 medium, in separate 20 mL glass vials (with PTFE septa) containing 9.9 mL of sterilized M4 medium.

Details on test conditions:

The identification of the hydrolysis products was carried out according to OECD guideline 111, Tier 3 (OECD, 2004) with some modifications. Sterilization of the M4 test medium, prepared according OECD guideline 201 (OECD, 2012) was performed by autoclaving at 121°C for 20 minutes. Sterilization of glassware was performed by heating at 180°C for at least 30 minutes. The spiked buffer solutions, prepared in the sterilized sample vials, were placed in a thermostatically controlled water bath in the dark at a temperature of 12 ± 0.5°C. Samples were taken on different time intervals and analyzed by GC-MS and NMR to determine which hydrolysis products were formed.

Modifications to the guideline:
· No Thymol was added to the buffer solutions. The addition of Thymol as a disinfectant was not necessary because the buffer solutions were sterilized.
· Only Tier 3 of OECD guideline 111 was conducted in this study, as tier 1 and 2 were already determined (Powley, 2008).
· Initial concentration of the test substance was above maximum water solubility. This was necessary to generate enough potential hydrolysis product(s) to make identification possible.

Positive controls:

no

Negative controls:

no

Transformation products:

yes

Remarks:

The intension was to analyze them, but no transformation products were formed.

Details on hydrolysis and appearance of transformation product(s):

No transformation products were formed during the course of the study.

% Recovery:

94.1

Temp.:

12 °C

Duration:

4 d

% Recovery:

96.1

Temp.:

12 °C

Duration:

44 d

% Recovery:

91.7

Temp.:

12 °C

Duration:

61 d

Remarks on result:

other: including amount found adsorbed to glassware

Details on results:

Concentrations were calculated form a single calibration standard, as it was not the intention to quantify the test substance but to identify the hydrolysis products. However as no hydrolysis products were formed the test substance concentrations are reported instead, to demonstrate the unexpected stability of the test substance.

One of the expected hydrolysis products, tert-Butanol, could not be determined with the described GCMS method. Our presumption/explanation is that tert-Butanol and DCM co-elute, making it impossible to determine the tert-Butanol. However as tert-Butanol could be accurately determined with the NMR, it was not necessary to optimize the GC-MS method. The other expected hydrolysis product 2,5-Dimethyl-2,5-hexanediol could be detected with the GC-MS and eluted at approximately 7.7 minutes. The test substance elutes at approximately 16 minutes.

Validity criteria fulfilled:

not applicable

Conclusions:

No hydrolysis products were found during the course of this study. The test substance did not hydrolyze, or not to the extent that should be expected based on the hydrolysis study performed by DuPont (Powley, 2008).

Executive summary:

Objective

Determine the hydrolysis products of 2,5-Dimethyl-2,5-di(tet-butylperoxy)hexane.

Approach

The samples were analyzed using two techniques, NMR and GC-MS to increase the possibility of identifying unexpected hydrolysis products, besides the two expected hydrolysis products tert-butanol and 2,5-Di-methyl-2,5-hexanediol.

Conclusions

No hydrolysis products were found during the course of this study. The test substance did not hydrolyze, or not to the extent that should be expected based on the hydrolysis study performed by DuPont (Powley, 2008).

Recommendations

The results during this experiment were so contradicting to the results of the hydrolysis study performed by DuPont (Powley, 2008). That we recommend checking if the determined hydrolysis rate is correct, that hydrolysis was for instance not influenced by adsorption of the test substance to the test vessel wall. As a follow up on this study the adsorption behavior of the test substance was checked (van Dam, 2018).

Reference 3

Endpoint:

hydrolysis

Type of information:

experimental study

Adequacy of study:

supporting study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: screening study performed to check for adsorption

Qualifier:

no guideline followed

Principles of method if other than guideline:

A stock solution of test substance was prepared in acetonitrile, this stock solution was used to spike the test substance to the M4 test media, An amount of 100 μL of the test substance stock solution was spiked to the M4 medium in separate 20 mL glass vials (with PTFE septa) not exceeding 1% (v/v) solvent. Test was spiked at a concentration of 75 μg/L test substance, this equals approx. half the water solubility.

GLP compliance:

no

Specific details on test material used for the study:

Chemical name: 2,5-dimethyl-2,5-di(tert-butylperoxy)hexane
CAS: 78-63-7
Batch/Lot no.: 1002516115
Water solubility: 152 μg/L (DuPont 2008)
Storage: Refrigerated ± 5 °C

Radiolabelling:

no

Analytical monitoring:

yes

Details on sampling:

Sample preparation / extraction
The samples taken at the different time intervals were extracted before analysis. Extraction was performed by adding 10 mL of DCM containing internal standard to the test vessel and shaking vigorously for 1 minute, subsequently the vial was left to stand for 10 minutes to allow for good phase separation. A subsample of the DCM layer was analyzed, to quantify the amount of test substance.

Details on test conditions:

Sterilization of the M4 test medium, prepared according OECD guideline 211 (OECD, 2012) was performed by autoclaving at 121°C for 20 minutes. Sterilization of glassware was performed by heating at 180°C for at least 30 minutes. The prepared sample vials were placed in a thermostatically controlled water bath in the dark at a temperature of 20 ± 0.5°C. Samples were taken on different time intervals and extracted before analysis by GC-MS, to quantify the amount of test substance.

Positive controls:

no

Negative controls:

no

Transformation products:

no

Details on results:

From the data displayed in table 1 below it can be seen that after 3 days there is still on average 87 % of test substance available. This is a significant difference to the half live of 3.9 hours and 4.4 hours at 20°C in pH buffers 7 and 9, respectively determined in the previous study (DuPont 2008). So when performing a total extraction from the test vessel significantly more test substance is still present compared to measuring just the water phase. However the amount of test substance after 6 days is 67.8%, on average. Unfortunately there are not enough time points measured to make a solid conclusion. But this indicates there is perhaps indeed hydrolysis taking place, however at a much slower rate than determined by DuPont. This should be confirmed by repeating the experiment with more analysis at more points in time.

Table 1: Concentrations of the test substance in M4 medium over time

Sample	Conc. (µg/L) T=0d	Conc. (µg/L) T=3d	Recovery (%) T=3d	Conc. (µg/L) T=6d	Recovery (%) T=6d
M4 medium I	74.1	63.0	85.0	49.4	66.7
M4 medium II	70.8	63.7	90.0	48.0	67.8
M4 medium III	71.3	61.8	86.7	49.2	69.0
average	72.1	62.8	87.2	48.9	67.8

Validity criteria fulfilled:

not applicable

Conclusions:

When performing a total extraction from the test vessel significantly more test substance is still present compared to measuring just the water phase. However the amount of test substance after 6 days is 67.8%, on average. Unfortunately there are not enough time points measured to make a solid conclusion. But this indicates there is perhaps indeed hydrolysis taking place, however at a much slower rate than determined by DuPont. This should be confirmed by repeating the experiment with more analysis at more points in time.

Executive summary:

The hydrolysis speed of the test substance, 2,5-dimethyl-2,5-di(tert-butylperoxy)hexane has already been determined (DuPont 2008). However the results from our experiment to determine the hydrolysis products (AkzoNobel 2018) conflicts with the previous study from DuPont. The aim of this study is to check if the test substance truly hydrolyses or if it is adsorption to the test vessel wall.

When performing a total extraction from the test vessel significantly more test substance is still present compared to measuring just the water phase. However the amount of test substance after 6 days is 67.8%, on average. Unfortunately there are not enough time points measured to make a solid conclusion. But this indicates there is perhaps indeed hydrolysis taking place, however at a much slower rate than determined by DuPont. This should be confirmed by repeating the experiment with more analysis at more points in time.

Reference 4

Endpoint:

hydrolysis

Type of information:

experimental study

Adequacy of study:

supporting study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: screening study to check for adsorption and the use of silanization

Qualifier:

no guideline followed

GLP compliance:

no

Specific details on test material used for the study:

EC name: Di-tert-butyl 1,1,4,4-tetramethyltetramethylene diperoxide
CAS: 78-63-7
Batch/Lot no.: 1002516115
Water solubility: 152 μg/L (DuPont 2008)
Composition: see CoA (Annex 1)
Storage: Refrigerated ± 5 °C

Radiolabelling:

no

Analytical monitoring:

yes

Details on sampling:

Sample preparation / extraction
At each time interval the water phase was decanted into a new vial, already containing 5 mL DCM with internal standard. Subsequently 5 mL DCM was added to the empty vial to extract the test substance from the test vial. Both vails were shaken vigorously for 1 minute. The vials were left to stand for 10 minutes to allow for good phase separation (for the one containing water). A subsample of the DCM layer was analyzed for quantification of the test substance.
At the last time interval, the emptied glass vials were extracted for a second time with DCM to examine if more test substance could be extracted from the test vessels. Subsequently other extraction solvents were used to examine if more could be extracted.

Details on test conditions:

Sterilization of buffer pH 7, prepared according OECD guideline 111 (OECD, 2004) was performed by autoclaving at 121°C for 20 minutes. Sterilization of glassware was
performed by heating at 180°C for at least 30 minutes. The prepared sample vials were placed in a thermostatically controlled water bath in the dark at a temperature of 20 ± 0.5°C. Samples were taken on different time intervals and extracted before quantitative analysis by GC-MS.

Silanizing test vessels
Silanization of the test vessels was achieved by the following procedure.
· Rinse 20 mL headspace vials with approximately 5 mL of methanol.
· Subsequently rinse the vial with toluene.
· Add 5 mL of a 5% (v/v) dimethyldichlorosilane in toluene solution to the vial and shake the vials for at
least 10 minutes and then discard the solution.
· Rinse the vial with approximately 5 mL toluene.
· Rinse the vial with approximately 5 mL methanol and allow to evaporate/dry before use.

Positive controls:

no

Negative controls:

no

Transformation products:

no

Details on hydrolysis and appearance of transformation product(s):

See attached tables (under illustration)

Details on results:

There is (except for day 7) a significant difference between the extracted amount of test substance from the untreated and the silanized glass vials (determined with T-test at 95% probability); from silanized glass vials significantly less could be extracted.
Until day 7 the measured concentrations in the water phase are higher for the untreated vials than the silanized vials, but the total recoveries of the test substance are better for the untreated vials. Since total recoveries for the silanized glass vials are lower, it is considered that adsorption to glassware is not different between untreated and silanized vials, but that the extraction method is less efficient for silanized glass vials. Attempts to improve the extraction efficiency by extracting the vials for a second time, or the use of different solvents (i.e. Hexane and Acetonitrile) had no effect, for both untreated and silanized vials. Therefore, we could not conclude that silanization of the test vials decreases adsorption.
Total recovery of the test substance at day 7 is well above 90% for the ‘normal’ glass vials, indicating that the test substance is stable over a period of at least 7 days. After day 7 of the test the total recovery slowly decreases. If it is assumed that the decreasing substance concentration is truly hydrolysis and not decreasing extraction efficiency, the rate of hydrolysis is much lower (half-life > 21d at 20 °C, pH 7) than determined in a previous study, where a half-life of 3.9 hours at 20°C in pH 7 was determined (Powley,
2008). This lower hydrolysis rate, and or adsorption of the test substance to the test vessel is confirmed by the results found in several other studies (van Dam, 2018; van Dam and ter Weele, 2018 and OConnor, 2019).

Validity criteria fulfilled:

not applicable

Conclusions:

Total recovery of the test substance at day 7 is well above 90% for the ‘normal’ glass vials, indicating that the test substance is stable over a period of at least 7 days. After day 7 of the test the total recovery slowly decreases. If it is assumed that the decreasing substance concentration is truly hydrolysis and not decreasing extraction efficiency, the rate of hydrolysis is much lower (half-life > 21d at 20 °C, pH 7) than determined in a previous study, where a half-life of 3.9 hours at 20°C in pH 7 was determined (DuPont,
2008). This lower hydrolysis rate, and or adsorption of the test substance to the test vessel is confirmed by the results found in several other studies (van Dam, 2018; van Dam and ter Weele, 2018 and OConnor, 2019).

Executive summary:

Objective

The objective of this study is to examine if silanizing the test vessel will reduce the adsorption of Di-tertbutyl 1,1,4,4-tetramethyltetramethylene diperoxide to the test vessel. As determined in other studies (van Dam, 2018; van Dam and ter Weele, 2018) the decreasing concentration of Di-tert-butyl 1,1,4,4 -tetramethyltetramethylene diperoxide is mainly caused by adsorption, rather than hydrolysis behavior.

Approach

A stock solution of test substance was prepared in acetonitrile, this stock solution was used to spike the test substance to a pH 7 buffer solution. The test substance stock solution was spiked to the buffer solution in separate 20 mL glass vials (with PTFE septa) not exceeding 1% (v/v) solvent. This was done in parallel to ‘normal’ glass vials and silanized vials. The test was spiked at a concentration of 75 μg/L test substance, this equals approximately half the water solubility.

Results

Total recovery of the test substance at day 7 is well above 90% for the ‘normal’ glass vials, indicating that the test substance is stable over a period of at least 7 days. After day 7 of the test the total recovery slowly decreases. If it is assumed that the decreasing substance concentration is truly hydrolysis and not decreasing extraction efficiency, the rate of hydrolysis is much lower (half-life > 21d at 20 °C, pH 7) than determined in a previous study, where a half-life of 3.9 hours at 20°C in pH 7 was determined (Powley, 2008). This lower hydrolysis rate, and or adsorption of the test substance to the test vessel is confirmed by the results found in several other studies (van Dam, 2018; van Dam and te Weele, 2018 and OConnor, 2019).

Description of key information

Due to proven adsorption of the test item to glassware, it was not possible to identify and quantify the decreases in sample solution concentrations due to hydrolysis alone.  Analytical monitoring of the sample solution concentrations only would systematically overestimate the rate of hydrolysis, as analytical losses would also include a contribution from adsorption.  However, inclusion of the adsorbed fraction in the analytical procedure, i.e. extraction of the incubated sample solutions within the original test vessels, risked underestimating the hydrolytic rate, as analyzed concentrations included a contribution from test item which was not available for hydrolysis.

Di-tert-butyl 1,1,4,4-tetramethyltetramethylene diperoxide adsorbs strongly to glassware and loss of the compound is solely or mainly due to adsorption and not to hydrolysis. Even if the compound would be susceptive to hydrolysis it would not be a significant route of degradation in the environment because the compound would adsorb to suspended solids and other organic material present in surface water.

Additional information

Three GLP studies evaluating the hydrolysis are available. The first two are screening studies; the third study is the definitive study.

The screening studies at 50 °C (Vos, 2010; Powley, 2008) were performed under GLP according to OECD guideline 111. In both studies, significant loss of parent compound occurred within 24 hours at pH 4, 7 and 9. As the possible hydrolysis product TBA (tertiary- butyl alcohol) could not be detected no confirmation of the increase of this product in 24 hours could be given. In the definitive study (Habeck, 2011), performed under GLP and following OECD Guideline 111 the results also showed a decrease of parent compound over time. As again only concentrations of parent compound were followed over time, no confirmation of hydrolysis route could be given. Loss of parent compound was wrongly interpreted as hydrolysis and a half-life between 2.7 and 2.8 hours for all pHs (4, 7 and 9) at 25 °C was derived.

Although the studies were not performed in a wrong manner, the conclusion drawn was not correct, therefore these studies were invalidated, and additional studies were conducted.

 

A non-GLP screening study at 20 °C to check for adsorption to glassware was performed (van Dam, 2018). Total extraction of test vessel and water phase was performed using DCM. After 3 and 6 days there was still on average 87% and 67.8%, respectively of parent compound available showing that significant adsorption occurred. Another non-GLP screening study at 12 °C (Van Dam and ter Weele, 2018) was performed checking for hydrolysis products over a longer period of time (61 days). Total extraction of test vessels was performed in the same manner as by van Dam (2018) and analyses of hydrolysis products were done by NMR and GC-MSto increase the possibility of identifying unexpected hydrolysis products, besides the two expected hydrolysis products tert-butanol and 2,5-Di-methyl-2,5-hexanediol. No hydrolysis products could be detected during this study. Parent compound was also analyzed over time and 91.7% was still present after 61 days showing that this substance is stable at 12 °C.

To substantiate these findings a GLP study according to OECD 111 guideline (OConnor, 2019) was performed. The procedure was modified such that analysis of the sample solutions was completed by solvent extraction performed in the original vessels used for incubation of the sample solutions.

The test item was demonstrated to show a high affinity for adsorption to glassware. Additional investigative testing indicated that test item dissolved in the aqueous buffer solutions may actually make no contribution to the concentration of test item detected on analysis. 

Due to this proven adsorption of the test item to glassware, it was not possible to identify and quantify the decreases in sample solution concentrations due to hydrolysis alone.  Analytical monitoring of the sample solution concentrations only would systematically overestimate the rate of hydrolysis, as analytical losses would also include a contribution from adsorption.  However, inclusion of the adsorbed fraction in the analytical procedure, i.e. extraction of the incubated sample solutions within the original test vessels, risked underestimating the hydrolytic rate, as analyzed concentrations included a contribution from test item which was not available for hydrolysis.

In an attempt to reduce adsorption to glassware and determine true hydrolysis another non-GLP study at 20 °C (van Dam, 2019) was conducted. Silanized as well as non-silanized test vessels were filled with pH 7 buffer solutions spiked at a concentration of 75 μg/L test compound, this equals approximately half the water solubility.

The results of this study were not conclusive. From silanized test vessels significantly less compound could be extracted. Until day 7 of the test the measured concentrations in the water phase were higher for non-silanized test vessels, but the total recoveries of the test compound were better for the non-silanized vessels. Since total recoveries for the silanized test vessels were lower, it is considered that adsorption to glassware is not different between non-silanized and silanized vessels, but that the extraction method is less efficient for silanized vessels. Attempts to improve the extraction efficiency had no effect, for both non-silanized and silanized vessels. Therefore, it could not be concluded that silanization of the test vessels decreases adsorption.

Total recovery of the test substance at day 7 was well above 90% for non-silanized test vessels, indicating that the test compound is stable over a period of at least 7 days. After day 7 of the test the total recovery slowly decreased, but after 21 days still over 67.5% could be recovered.

 

All in all the results summarized above show that Di-tert-butyl 1,1,4,4-tetramethyltetramethylene diperoxide adsorbs strongly to glassware and loss of the compound is solely or mainly due to adsorption and not to hydrolysis. Even if the compound would be susceptive to hydrolysis it would not be a significant route of degradation in the environment because the compound would adsorb to suspended solids and other organic material present in surface water.