Low-level quantification of PFAS in water samples with high sensitivity and precision 

Methods

Sample preparation: A surface water sample and a low-level standard (50 ppt) were analyzed in this study. The standard consisted of a comprehensive suite of PFAS compounds, including perfluorinated carboxylic acids (PFCAs), fluorotelomer sulfonates (FTSs), sulfonamides and fluoroether acids (Wellington Laboratories, Guelph, ON). Table 1 contains a list of all compounds studied. The surface water sample was prepared by extraction of 500 mL of water through weak anion exchange (WAX) SPE cartridges. The eluent was reduced in volume under N₂ gas and reconstituted in an 80:20 methanol and water mixture. Five replicates of the standard and 3 replicates of the sample were analyzed.

Chromatography: An ExionLC system was plumbed with a Phenomenex Gemini C18 (5 μm, 3 x 50 mm) column between the LC pumps and the autosampler. This column temporarily retains contaminants from the mobile phases and LC system and elutes them later in the gradient.⁵ A Phenomenex Gemini C18 (3 µm, 3 x 100 mm) column was used as the analytical column. An LC flow rate of 500 µL/min was used along with a 1 µL injection volume. Mobile phases were water and methanol, both amended with 10 mM ammonium acetate. The LC gradient is shown in Figure 2.

Mass spectrometry: A SCIEX 7500 system was used with an OptiFlow Pro ion source and an electrospray ionization (ESI) probe in negative ion mode. The source and MS parameters were as follows: TEM = 400°C, GS1 = 40, GS2 = 70, ISV = -2000, CUR = 40, CAD = 14. For comparison purposes, a SCIEX Triple Quad 6500+ system was used with the IonDrive Turbo V ion source. Primary (and, in most cases, secondary) multiple reaction monitoring (MRM) transitions were monitored for each PFAS compound based upon previous studies and were identical for both instruments. Each system was optimized for maximum sensitivity.   

Data processing: Data acquisition and automatic data processing were performed using SCIEX OS software and analytics in SCIEX OS software. Flagging rules were specified within the analytics module to quickly flag good and bad data.

Key innovations that improve sensitivity

The SCIEX 7500 system was designed to push sensitivity boundaries beyond those previously established with the high-end SCIEX Triple Quad 6500+ system. These sensitivity improvements are realized through several areas.⁴ For example, the E Lens probe of the OptiFlow Pro ion source more efficiently breaks up droplets, releases ions and actively drives ions toward the orifice. The 4.3x increase in the area of the entrance orifice allows more ions to enter the instrument, improving ion sampling. The D Jet ion guide more efficiently captures and focuses ions from the ESI spray plume, allowing more ions to traverse the instrument to be analyzed and detected. All of these enhancements improve the total number of ions available for analysis in the system. Since the noise does not increase proportionally, the S/N ratio increases, quite dramatically for some compounds, for SCIEX 7500 system data.

Because of these changes in source and entrance optics, PFAS compounds were first evaluated using a range of source and entrance parameters to find optimum conditions. Although there were some dependencies on source temperature and ionization voltage as a function of PFAS class and chain length, conditions were found that produced good signal for all compounds. Interestingly, lower voltages were generally required using the OptiFlow Pro ion source than previous source designs due to the function of the E Lens probe. 

Signal-to-noise comparison

The sensitivities of the SCIEX Triple Quad 6500+ system and the SCIEX 7500 system were evaluated by injecting the low-level standard onto each system. Table 1 lists all of the PFAS compounds contained within the low-level standard. The S/N was automatically calculated within SCIEX OS software using the “peak-to-peak” algorithm and is listed for each MRM transition. The table also shows the improvement factor in S/N for the SCIEX 7500 system versus the SCIEX Triple Quad 6500+ system. Improvements in S/N for the SCIEX 7500 system were on average around 5.5x but ranged anywhere from ~1x to more than 50x. 

Figure 1, Figure 3 and Figure 4 show MRM peaks obtained for 3 representative PFAS compounds on the SCIEX 7500 system compared with the SCIEX Triple Quad 6500+ system. Integrated peak areas are shaded in purple while the gray highlighted areas represent the noise used by the S/N algorithm.

For the SCIEX 7500 system data, peak areas and total peak intensities are greater while the noise is maintained at relatively low levels, resulting in increased S/N in all cases. This ultimately results in improved sensitivity and lower limits of quantification (LOQ). For example, as shown in Figure 4, the increased S/N using the SCIEX 7500 system enables quantification of important, lower-level isomers for N-ethyl perfluorooctane sulfonamide (EtFOSA).

Precision comparison 

Five replicate injections were made for the low-level standard and %CVs were calculated. As shown in Figure 5, precision was excellent with CVs of less than 10% for the SCIEX 7500 system data.

For the SCIEX Triple Quad 6500+ system data, CVs were generally less than 20%, with some outliers ranging as high as 30%. The lower %CV of the SCIEX 7500 system represents improved precision and therefore improved data quality near the lower LOQ.

Water sample evaluation

The surface water sample was analyzed in triplicate on both the SCIEX Triple Quad 6500+ system and the SCIEX 7500 system. As shown in Figure 6, Figure 7 and Figure 8, the higher sensitivity of the SCIEX 7500 system resulted in higher signals and greater S/N for all compounds, transitions and replicates.

For example, this increased sensitivity enabled the detection of low-level isomers of N-methylperfluorooctane sulfonamidoacetic acid, or MeFOSAA (Figure 7), and it increased confidence in the detection of perfluorotridecanoic acid, or PFTrA (Figure 8), a hydrophobic compound that normally would not be expected to be present in surface water.

Conclusions

The increased sensitivity of the SCIEX 7500 system allows a new level of performance for the analysis of PFAS compounds.

• Lower-level compounds and important lower-level isomers can be detected and quantified.
• Improved S/N leads to lower limits of detection.
• Unexpected compounds that are detected can be quantified with higher certainty.
• Higher precision is achieved for replicate analyses, leading to more confidence in the quantified amounts.
• Less sample can be utilized for the analysis.
• Sample preparation strategies can be simplified.
• Intuitive and friendly SCIEX OS software handles all data acquisition and processing, streamlining the entire workflow.

References

1. Buck RC et al. Perfluoroalkyl and polyfluoroalkyl substances in the environment: terminology, classification, and origins. Integr Environ Assess Manag. 2011 Oct;7(4):513–541.
2. De Silva AO et al. PFAS Exposure Pathways for Humans and Wildlife: A Synthesis of Current Knowledge and Key Gaps in Understanding. Environ Toxicol Chem. 2021 Mar;40(3):631-657.
3. European Chemicals Agency (ECHA): Perfluoroalkyl chemicals (PFAS).           
4. Enabling new levels of quantification. SCIEX technical note, RUO-MKT-02-11886-A.
5. Quantitation of PFASs in water samples using LC-MS/MS large-volume direct injection and solid phase extraction. SCIEX technical note, RUO-MKT-02-4707-A.