Reaction mass of bis(oxiran-2-ylmethyl) terephthalate and tris(oxiran-2-ylmethyl) benzene-1,2,4-tricarboxylate

Exposure related observations in humans: other data

Administrative data

Endpoint:

exposure-related observations in humans: other data

Type of information:

experimental study

Adequacy of study:

key study

Study period:

13 February 2019

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

test procedure in accordance with national standard methods

Justification for type of information:

The powder coating manufacturing process involves a melt mixing step upon which a minor amount of the registered substance, typically 5 ± 2%, is incorporated into a solid polyester resin via extrusion at 100-140°C under industrial conditions. The mixture optionally contains fillers, pigments and additives.

A powder coating formulation is solid, amorphous, partly crystalline.

The polyester resins are functionalized with terminal end carboxylic groups, which react with the terminal epoxide groups of PT910/912 upon cure at 160-200°C. Thereby, encapsulating and subsequently reducing or even removing the potential for the registered substance to be bioavailable when incorporated into a powder coating matrix. In an attempt, to prove that the registered substance is not bioaccessible following accidental exposure to representative powder coatings containing <7% of the registered substance in-vitro testing was commissioned. The testing was designed to determine and quantify if the main constituents of the registered substance can leach from representative powder coating matrices following exposure to simulated perspiration.

Thereby, simulating an accidental dermal exposure scenario following removal of protective gloves.

Cross-reference

Reason / purpose for cross-reference:

exposure-related information

Data source

Reference

Reference Type:

study report

Title:

Unnamed

Year:

2019

Report date:

2019

Materials and methods

Type of study / information:

Preliminary proof of concept study which will be developed further to better understand the bioaccessibility of two key components within powder coating formulations following the incorporation of the the registered susbtance

Endpoint addressed:

not applicable

Test guideline

Qualifier:

according to guideline

Guideline:

other:

Version / remarks:

Concept is based on a study (EUR 29282 EN) conducted in 2018 by the European Commission’s Joint Research Centre. The registrant considers this study to set a precedent for investigation of bio-accessibility

Deviations:

not applicable

Principles of method if other than guideline:

To determine the potential for bis(2,3-epoxypropyl) terephthalate (CAS 7195-44-0) and tris(oxiranylmethyl) benzene- 1,2,4-tricarboxylate (CAS 7237-83-4) to leach out of 3, 5 and 7% powder coating formulations at various time points (0, 2, 8, 24 hours, 5 and 10 days) and into a commercially available simulated perspiration (BS EN 1811:2011, Stabilized, Pickering Laboratories) an analytical method using liquid chromatography mass spectrometry (LCMS) was developed. The LCMS method was validated for linearity, limit of quantification (LOQ), limit of detection (LOD), accuracy, and reproducibility.

Aliquots of the powder coating formulations were prepared at a concentration of approximately 10 g/L in the synthetic perspiration and extracted at the requested time intervals. Time intervals included: T0 = start, T1 = 2 hours, T2 = 8 hours, T3 = 24 hours, T4 = 5 days, and T5 = 10 days. After the appropriate
time interval, aliquots of the extractions were diluted with LCMS grade acetonitrile, and then sonicated for 20 minutes at 30°C, except for T0 extracts. T0 extraction was performed by putting the powder coating formulations into the synthetic perspiration for 30 seconds and then extracting the perspiration
with further dilution using LCMS grade acetonitrile. After sonication all samples were filtered prior to LCMS analysis. Two powder coating formulation samples were randomly chosen to be extracted in duplicate for each extraction time interval, in order to test the extraction reproducibility. Sample pH was
measured before and after each extraction and was consistently 6.5.

The LC separation was achieved using mobile phase consisting of LCMS Optima™ grade acetonitrile and 0.1% formic acid with a Waters C18 UPLC column. A Waters UPLC and Xevo TQD mass spectrometer were utilized in positive mode with electrospray ionization. The bis(2,3-epoxypropyl) terephthalate and tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate analytes eluted at 4.16 and 2.27 minutes, respectively. The analytes were detected using multiple reaction monitoring (MRM). A common plastic additive, dibutyl phthalate, was found to interfere with the detection of bis(2,3- epoxypropyl) terephthalate. The interfering contaminant is thought to originate from the plastic bottle used to ship the synthetic sweat. Upon introduction of the contaminate it was impossible to remove or separate the dibutyl phthalate signal from that of bis(2,3-epoxypropyl) terephthalate. Therefore, further method validation and quantitation was performed only for tris(oxiranylmethyl) benzene- 1,2,4-tricarboxylate.

The analyte identification was based on the peak retention time and the mass of ions correlated to bis(2,3-epoxypropyl) terephthalate and tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate in the mass spectra. Bis(2,3- epoxypropyl) terephthalate elutes at approximately 4.2 minutes, and tris(oxiranylmethyl) benzene-1,2,4- tricarboxylate elutes at approximately 2.3 minutes. Chromatograms of standard solutions of bis(2,3-epoxypropyl) terephthalate and tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate are displayed in Figures 2-3 (see study report). Note a smaller peak is detected at 2.3 minutes in chromatograms for tris(oxiranylmethyl) benzene-1,2,4- tricarboxylate, which correlate to similar ions present from bis(2,3-epoxypropyl) terephthalate, which is also present in the Araldite PT 910 standard. Chromatograms of bis(2,3-epoxypropyl) terephthalate and tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate detected in extracted powder coating formulation VPK81262-0-03 at T2 (8 hours) are shown in Figures 4-5 (see study report).

Representative overlaid quantitation results of tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate are displayed in Figure 6 (see study report). The method accuracy results are listed in Table 3 (see study report) and the reproducibility results are listed in Table 4 (see study report). The calibration curve was
acceptably linear over the range of 5 ppb (μg/L) to 1000 ppb (μg/L) tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate. The square of correlation coefficient value (R2) for the calibration curve displayed in Figure 1 is 0.9997 (see study report).

The detection and quantitation of bis(2,3-epoxypropyl) terephthalate was not possible due to the presence of dibutyl phthalate. As mentioned, the dibutyl phthalate is thought to originate from the plasticbottle containing the simulated sweat. Tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate was detected
in all sample extracts at all extraction times. Quantitation results are listed in Table 1 (see study report). The trend in extraction amount of tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate over time is shown in Figure 6 (see study report). Most samples show a consistent trend of increased extraction of
tris(oxiranylmethyl) benzene-1,2,4- tricarboxylate over time, up to 5 days. At 5 days most samples see a decrease in the amount of tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate detected. Similarly, at 10 days the amount of tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate decreases further. Most likely
other unknown components are also leaching, and at longer elapsed times the sample gets too complex for mass spectrometry detection.

LCMS is prone to matrix effects; if a significant amount of material is present around the same retention time as tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate, it can cause a loss of sensitivity for the analyte. A scan LCMS was performed to attempt to determine if an obvious component was present. No major
ions were detected at a similar retention time as tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate, but the overall background was elevated of mass ranges of 100-200 Da, 200-300 Da, and similar extracted mass ranges. If detection and quantification of tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate at
longer elapsed times is required a sample pre-treatment would be required.

GLP compliance:

yes (incl. QA statement)

Method

Ethical approval:

not applicable

Details on study design:

Test design was in-vitro

Exposure assessment:

measured

Details on exposure:

A total of nine representative powder coating formulations were supplied by downstream customers to Huntsman Advanced Materials and subsequently shipped to Intertek, Pickering Laboratories, USA. The representative formulations contained 3, 5 or 7% of the registered substance. During the leaching studies 10g of each powder coating was incubated with the artificial perspiration resulting in nominal concentrations of 0.3, 0.5 or 0.7 g of tegistered substance equivalent to nominal concentrations of 0.06, 0.1 or 0.14 g of tris(oxiranylmethyl) benzene-1,2,4 -tricarboxylate and 0.195, 0.325 and 0.455 g of bis(2,3-epoxypropyl) terephthalate.

Results and discussion

Results:

The results indicate that using a simulated scenario the maximum leachable amount of tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate from powder coating formulations containing 3, 5 or 7% registered substace was equivalent to 67.2 μg (0.05% of the nominal amount of tris(oxiranylmethyl) benzene-
1,2,4-tricarboxylate present in a 10g sample).

Due to the interfering peak of dibutyl phthalate that was identified in the synthetic sweat, thought to be a contaminant from the shipping container it was necessary during these trials to compare and quantify the amounts of bis(2,3-epoxypropyl) terephthalate by calculation using the measured amounts of
tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate which is not ideal but may allow some interpretation and assumptions to be made. Using internal company analytics the registered substance typically contains 20% tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate and 65% bis(2,3-epoxypropyl) terephthalate and therefore a multiplication factor of 3.25 was applied to the measured quantities of tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate to provide estimated amounts of bis(2,3-epoxypropyl) terephthalate in the3,5 and 7% powder coating formulations.

Applicant's summary and conclusion

Conclusions:

The results of this proof of concept study support the technology and scientific theory and indicate that once the registered substance is incorportated via extrusion into powder coating formulations at <7% the main chemical species components become encapsulated within a polyester resin which inherently significantly reduces the bioaccessibility, mitigates exposure and subsequently reduces the bioavailability of the substance if accidental exposure were to occur.

It was concluded based on this proof of concept study that within a powder coating formulation containing 7% of the registered substance only 0.0018 % of the initial registered substance incorporated into the polyester resin would be bioaccessible in the unlikely event of exposure occurring. As this study was a proof of concept further experiment work is in progress to further support and build a weight of evidence approach.

Executive summary:

In conclusion, due to effects on male fertility (sperm) which were identified in a 2-week preliminary test that was designed to act as a range finder for an OECD 443 extended one generation study it was considered necessary to reclassify the reaction mass of bis(2,3-epoxypropyl) terephthalate and tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate (PT-910) as a category 1B reprotoxin. To mitigate “hazard” based concerns in the powder coating sector (industrial users), develop a weight of evidence and for potential use when classification powder coatings which have a composition containing a maximum of 7% the registered substance and finally to provide additional data to support the” risk” characterisation, a bespoke test utilising a synthetic perspiration was conducted.

Considering the high molecular weight polyester resins used in the manufacture of powder coating materials, it was hypothesised that once the substance is mixed into a powder coating at 7% there is a matrix effect which effectively locks the registered substance within the formulation (a polyester resin) and inherently reduces the bioaccessibility of the components during handling or in the event of accidental exposure. To test the hypothesis, a leaching study using 10 g of powder coating formulations containing the registered substance in synthetic sweat followed by measurement of tris(oxiranylmethyl) benzene-1,2,4 -tricarboxylate and bis(2,3-epoxypropyl) terephthalate leaching after 2, 8, 24 hours, 5 and 10 days using a validated LC/MS analytical method. Using the results of this study conservative estimates of bioaccessibility and subsequently availability for the tris(oxiranylmethyl) benzene-1,2,4-tricarboxylate, bis(2,3-epoxypropyl) terephthalate and total (summation of the two components which represents 85% of the PT-910 composition) were determined. It was concluded that within a powder coating formulation containing 7% of the registered substance only 1.883 x 10-03 % would be available in the unlikely event of exposure occurring.

This data was produced as part of a proof of concept study and will be further developed to create a bioaccessibility weight of evidence case.