Determination of 1, 4-dioxane using automated SPE, compliant with US EPA 522

Introduction

1, 4-dioxane is a compound that has become known as an emerging contaminant which may cause negative health effects in humans. The US Agency for Toxic Substances and Disease Registry (ATSDR) states that exposure to 1, 4-dioxane at high levels may cause liver and kidney damage.

1,4-dioxane is also reasonably anticipated to be a human carcinogen based on sufficient evidence of carcinogenicity in animals.¹ The US EPA has also classified 1,4-dioxane as “likely to be carcinogenic to humans” by all routes of exposure.² Recent research has evaluated exposure through drinking water and food, generating a comprehensive picture of possible carcinogenicity.

1,4-dioxane exposure occurs from a variety of sources, it’s used as a stabilizer in in certain chlorinated solvents, therefore it can be found in many products that are known to use chlorinated solvents such as; paint strippers, dyes, greases, anti-freeze and aircraft de-icing fluids. Dioxane was also used as a solvent to facilitate SN2 reactions in chemical synthesis because of its polar aprotic nature. Dioxane is also a by-product of
ethoxylation reactions, many of which are carried out on a regular basis is cosmetic products that contain sodium laureth sulfate.² This reagent is so common among cosmetic products that detectable amounts of 1,4-dioxane can be found in nearly 57% of baby shampoos and 97% of hair relaxers. The FDA and the EU Scientific Committee on Consumer Safety, working on the advice of the International Cooperation on Cosmetics Regulation (ICCR), recommended the limit for 1, 4-dioxane in finished cosmetic products be less than 10 ppm.³

Since the main source of 1,4-dioxane is currently cosmetic products, it is no surprise that it can be found in both drinking water and ground water tables. Japan has observed levels in surface water up to 42.8 µg/L and found up to 79 µg/L in groundwater samples. In this case, a high correlation was observed with the presence of 1,1,1-trichloroethane. 1,4-Dioxane was found at a concentration of 0.2–1.5 μg/L in tap water samples from six cities in Kanagawa, Japan, in 1995–1996.4 In the US, 1, 4-dioxane was included in the third Unregulated Contaminant Monitoring Rule (UCMR-3), a list of candidate contaminant compounds based on toxicity and occurrence. The list of 30 compounds is monitored in large public water supplies and selected small water supplies to better understand the occurrence and magnitude in drinking water to aid in deciding if regulation is warranted.⁵ The results of UCMR-3 have not resulted in a regulated maximum contaminant level of 1, 4-dioxane, but some states are beginning to set regulations. These regulations can be observed in Table 12 The EPA risk assessments indicate that the drinking water concentration representing a 1 x 10-6 cancer risk level for 1,4-dioxane is 0.35 μg/L.⁶

Several concerns have arisen about measurement of 1, 4- dioxane in water samples due to dioxane’s high affinity for water. The compound is completely miscible in water and although it is volatile, it is difficult to purge from water.
Evaluation of 1,4-dioxane can be done by a number of existing US EPA methods employing liquid-liquid extraction or purging to remove 1, 4-dioxane from water for GC/MS measurement but these methods have proved to have worse detection limits than desired. US EPA method 522 from the drinking water program specifies solid phase extraction (SPE) and GC/MS analysis using single ion monitoring (SIM) and is the most successful method to date.²
This application note will evaluate the performance of the Biotage® Horizon 5000 automated solid phase extraction system in conjunction with US EPA Method 522.

Experimental

The extraction was performed using the Biotage® Horizon 5000 automated solid phase extraction system, using the extraction program displayed in Table 2. A 500 mL water sample size was extracted at a neutral pH. To improve method performance, the consumable used for this application note was a 3-gram, 6 cc coconut charcoal cartridge. This change not only demonstrated optimal recovery rates but it also allowed the 5000 system to be operated at a sample loading speed of 3. This operational change allowed for the sample loading rate to be increased to approximately 25 mL/min from 10 mL/min and is method compliant due to the language in section 1.6 of EPA method 522. This saved up to approximately 20 minutes per sample. The analytical step was performed using GC/MS in the single ion mode (SIM) for the best sensitivity. The conditions for the Agilent 7890A GC coupled with the Agilent 5975C mass spectrometer are presented in Table 3.

Figure 1: Two Biotage® Horizon 5000 extractors equipped with carbon cartridges for extraction. Both extractors are controlled using the PC in the middle of the image.

Results and discussion

Table 6 in EPA Method 522 lists the initial demonstration of capability (IDC) requirements as well as the quality control requirements for the analysis of 1,4-dioxane. Table 7 in Method 522 lists the ongoing quality control requirements that must continually be met.
The method states that a low background of the system and the reagents must be determined by examining a lab reagent blank (LRB). A surrogate is added to the reagent blank to ensure that the extraction was performed to the standard of the method. The 1, 4-dioxane and background interferences must be less than or equal to 1/3 of the MRL in order to continue with the IDC requirements. The results for one LRB sample are presented in Table 4.

A set of four laboratory fortified blanks (LFBs) was extracted on the Biotage® Horizon 5000 to determine the initial demonstration of precision (IDP). The precision (relative standard deviation (RSD)) of all four samples must be ≤20%. The precision results are presented in Table 5.

The initial demonstration of accuracy (IDA), presented in Table 6, uses the same four LFBs that were used for determining the IDP. The method specifies that in order to demonstrate accuracy, the mean recovery of the LFBs must be +/- 20% of the true value. The true value for each of the four samples was 10 µg/L.
 

Seven LFBs were extracted to confirm the minimum reporting level (MRL) and determine the half range for the prediction interval of results (HRPIR). This data set provides an RL for the Biotage® Horizon 5000 automated solid phase extraction instrument. The MRL and HRPIR data is presented in Table 7.

The equation for calculating HRPIR is as follows:
HRP_IR = 3.963 S
Where S is the standard deviation and 3.963 is a constant value for seven replicates

Data from Table 7 was also used to confirm the upper and lower prediction interval of results (PIR). These two limits must be met in order to confirm that the MRL is valid. The upper PIR limit must be less than or equal to 150% while the lower PIR limit must be greater than or equal to 50%. The data for the upper and lower PIR limits is presented in Table 8. The equations for calculating Upper PIR and Lower PIR are as follows:

Upper: (Mean + HRPIR / Fortified Concentration) *100
Lower: (Mean - HRPIR / Fortified Concentration) *100

A method detection limit (MDL) (optional for an IDC) was calculated using the procedure in 40CFR, part 136 for an initial MDL. Eight LFBs were spiked at low concentration (0.15 µg/L) and extracted through the Biotage® Horizon 5000 over a period of one month. The standard deviation of the eight replicates was multiplied by the Student’s T value of 2.998 to calculate the MDL. The results for the MDL study are presented in Table 9.

A method detection limit (MDL) (optional for an IDC) was calculated using the procedure in 40CFR, part 136 for an initial MDL. Eight LFBs were spiked at low concentration (0.15 μg/L) and extracted through the Biotage® Horizon 5000 over a period of one month. The standard deviation of the eight replicates was multiplied by the Student’s T value of 2.998 to calculate the MDL. The results for the MDL study are presented in Table 9.

Figure 2: Recovery values for twenty samples spiked with surrogate.

The three-gram cartridges performed exceedingly well and passed all IDC and ongoing QC requirements. The three-gram cartridge allowed for faster sample processing because the larger sorbent volume allows the sample to be pulled through faster while preventing breakthrough. The measured recovery of twenty surrogates is presented in Figure 2.
A blind performance testing sample (PT) was analysed in order to make sure that the extraction process, as well as the analytical method, are capable of quantitation. A sample was received from an accredited provider and extracted using the Biotage® Horizon 5000. The results as well as the acceptance criteria are presented in Table 11.

Conclusion

This application note proves that EPA method 522 can be successfully implemented in a laboratory using the Biotage® Horizon 5000 automated solid phase extraction system.
Four LFB samples were analysed for precision and accuracy, yielding an average recovery value of 85.25% with an RSD of 2.93%. Both values meet the acceptance criteria of the method. The batch to batch quality control requirements set by the EPA method are easily met using this extraction method.
A blind performance test sample validated the accuracy of results obtained for drinking water. The automation of this method provides less analyst intervention which reduces any possible outside contamination. The 10 mL final extract volume eliminates any losses due to evaporation while the larger sorbent bed has allows for faster flow rate with better performance. All of these factors lead to an increase in productivity while easily meeting all of the quality control requirements for EPA Method 522.

References

1. Agency for Toxic Substances and Disease Registry, Toxicological Health Profile for 1, 4-Dioxane (2012), https:// www.atsdr.cdc.gov/toxprofiles/tp187-c2.pdf, accessed July 5, 2018.

2. US EPA Technical Fact Sheet –1,4-Dioxane, November 2017, https://www.epa.gov/sites/production/files/2014-03/ documents/ffrro_factsheet_contaminant_14-dioxane_ january2014_final.pdf, accessed July 5, 2018.

3. Scientific Committee on Consumer Safety (SCCS), The Report of the ICCR Working Group: Considerations on Acceptable Trace Level of 1,4-Dioxane in Cosmetic Products, December 2015. https://ec.europa.eu/health/scientific_committees/consumer_safety/docs/sccs_o_194.pdf, accessed July 5, 2018.

4. 1,4-Dioxane in Drinking-water, Background document for development of WHO Guidelines for Drinking-water Quality, World Health Organization 2005. https://www.who.int/ water_sanitation_health/dwq/chemicals/14dioxane0505. pdf, accessed July 8, 2018

5. Third Unregulated Contaminant Monitoring Rule Fact Sheet, https://www.epa.gov/sites/production/files/2015-10/ documents/ucmr3_factsheet_general.pdf, accessed July 5, 2018.

6. 2018 Edition of the Drinking Water Standards and Health Advisories Tables, https://www.epa.gov/sites/production/ files/2018-03/documents/dwtable2018.pdf, accessed July 5, 2018.

7. David T. Adamson, Elizabeth A. Piña, Abigail E. Cartwright, Sharon R. Rauch, R. Hunter Anderson, Thomas Mohr, John A. Connor, 1,4-Dioxane drinking water occurrence data from the third unregulated contaminant monitoring rule, Science of the Total Environment 596–597 (2017) 236–245.

8. US EPA Method 522, Method 522_Determination of 1,4-Dioxane in Drinking Water by Solid Phase Extraction (SPE) and Gas Chromatography/ Mass Spectrometry (GC/ MS) with Selected Ion Monitoring (SIM) Sep, 2008, https:// nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P100EQ8A.txt, accessed July 6, 2018.

Literature number: AN927