Methods

Sample preparation: Sample preparation followed EPA method 533.⁶  Briefly, 250 mL of water was fortified with analytical and isotopic standards (Wellington Laboratories).  Water samples were passed through a conditioned anion exchange SPE column (Phenomenex Strata-X-AW Weak Anion 500 mg/6 mL). After rinsing with ammonium acetate the column bound residues were eluted with methanol containing 2% ammonium hydroxide, dried down under nitrogen gas, and reconstituted in 1 mL of 80:20 methanol/water containing isotopic performance standards.

Four real-world samples were collected from New England tap water and natural rivers, lakes, and ponds and then extracted following the EPA method 533 procedures. The isotopic dilution standards and isotopic performance standards were spiked as prescribed.

Care was taken to clean all sample preparation components with LC-MS grade methanol followed by MilliQ water to minimize contamination. This included sample containers, SPE connections, mobile phase bottles, and metal gas lines used for the dry down process.

Chromatography: HPLC separation was performed with the ExionLC AC system using a 12-minute gradient. The Phenomenex Gemini C18 (100 x 3.0 mm, 3.0 µm, 00D-4439-Y0) was used as the analytical column. The delay column used was a Phenomenex Luna C18 (30 x 3.0 mm, 00A-4252-Y0). The delay column was plumbed before the injection port to allow for separation of PFAS residues originating from the LC system and/or mobile phases from the analytical peak (Figure 2). Very low systemic contamination was observed for all analytes (1/3 of the MRL). The injection volume was 5 µL with a column temperature of 40°C using a flow rate of 0.6 mL/min. Mobile phases consisted of 10 mM ammonium acetate and methanol with 10 mM ammonium acetate. Figure 3 displays representative chromatographic separations that were achieved with the system. 

Mass spectrometry: A Turbo V ion source was used with electrospray ionization in negative ion mode on the SCIEX Triple Quad 4500 system. Acquisition was performed using Analyst software 1.7.2. The Scheduled MRM algorithm was used to optimize duty cycle to give greater than 12 scans across the chromatographic peaks.

Source conditions were optimized to give the highest ionization efficiency with respect to flow rate and mobile phase composition. Compound specific DP, CE, and CXP values were used to give maximum sensitivity and selectivity for all transitions and values were similar to previous PFAS SCIEX tech notes⁷.

Data processing: Data were processed using SCIEX OS software 2.1. The processing method was built to assign the appropriate isotopic dilution standard to the appropriate quantification transition for each PFAS analyte. This is accomplished in the software during construction of the processing method and allows for straightforward, internal standard corrected quantification.

Calibration curves were constructed between 0.5 – 15 ng/mL (corresponding to 2 – 60 ng/L in sample) and were forced through zero per method requirements.  Linearity was > 0.98 for all analytes. The statistics pane in SCIEX OS software allows for easy evaluation of the accuracy and precision of each calibration curve. This feature facilitated rapid assessments of method performance criteria specific to MRL calculations.

Method chromatography, precision, and accuracy

Excellent peak shape and analyte separation were achieved using the 12 min gradient (Figure 3). Target MRLs for IDC requirements included accuracy and precision characteristics. Requirements were for seven replicate spiked samples to display accuracies of 70%-130% and precision ≤ 20% CV. All compounds in the study were in acceptable ranges for their verified MRL values (Table 1). Most compounds displayed accuracy in the 80%-110% range or better. Precision values are listed for each compound alongside the compound MRL value in Table 1.

MRL determinations

For an MRL to be considered verified, the seven replicate extracts at the target MRL fortification must pass statistical rigor as described in EPA method 533. The standard deviation from the seven replicates is multiplied by t-value (3.963) to give the Half Range for the Prediction Interval of Results (HRPIR). Then the Upper and Lower Prediction Interval of Results (UPIR and LPIR) are calculated as follows:

Representative sample data

Several tap water and natural water sources were collected from New England and subjected to the extraction method. All tap water samples showed low level ng/L PFOS and PFOA (Table 2). The river water sample had higher concentrations compared to treated tap water.  However, PFAS findings were, in all cases, below the EPA drinking water recommendation for summed PFOA and PFOS (70 ng/L).⁸ Chromatograms of a representative water sample finding vs. the analytical standard are shown in Figure 5.

Conclusions

• The SCIEX Triple Quad 4500 system coupled to an ExionLC AC system can achieve, and in many cases surpass, target MRLs for PFAS residues specified in EPA method 533 and UCMR5
• Simple upgrades to the LC system can limit contamination and separate background contaminants from residues in the analytical injection
• The method was able to successfully test for PFAS contamination in New England water samples.

References

1. Gluge J et al. (2020) An overview of the uses of per- and polyfluoroalkyl substances (PFAS). Environmental science: processes and impacts. 22, 2345-2373
   
2. PFAS: Listed in OECD Global Database. EPA site.
3. Per- and Polyfluorinated substances (PFAS) biomonitoring fact sheet. CDC site.
4. Sunderland EM, et al. (2019) A Review of the Pathways of Human Exposure to Poly- and Perfluoroalkyl Substances (PFASs) and Present Understanding of Health Effects, J Expo Sci Environ Epidemiol. 29(2): 131–147.
5. Fifth Unregulated Contaminant Monitoring Rule. EPA site.
6. METHOD 533: DETERMINATION OF PER- AND POLYFLUOROALKYL SUBSTANCES IN DRINKING WATER BY ISOTOPE DILUTION ANION EXCHANGE SOLID PHASE EXTRACTION AND LIQUID CHROMATOGRAPHY/TANDEM MASS SPECTROMETRY. EPA method.
7. EPA method 533 for PFAS analysis in drinking water at low parts-per-trillion levels using the SCIEX Triple Quad 5500+ system. SCIEX technical note, RUO-MKT-02-12843-A.
8. Drinking Water Health Advisories for PFOA and PFOS. EPA site.