Quantitation of PFAS in food with parts per trillion levels of sensitivity


Xanthippe Theurillat1, Claudia Mujahid1, Bjorn Eriksen1, Ashley Griffin2, Andrew Savage2, Thierry Delatour1, Pascal Mottier1, Jack Steed3, Michael Scherer4, Jianru Stahl-Zeng5

1Société des Produits Nestlé, Nestlé Research; 2Société des Produits Nestlé, Nestlé Quality Assurance Center, 3SCIEX, UK; 4SCIEX, Switzerland; 5SCIEX, Germany

Abstract


This technical note demonstrates a validated method for the analysis of 57 per- and polyfluoroalkyl substances (PFAS) compounds in a diverse range of food matrices. Using the SCIEX 7500 system and a simplified extraction method, the limit of quantitation (LOQ) levels met the Commission Regulation EU 2022/2388 for the 4 regulated PFAS compounds, including: PFOS, PFOA, PFNA and PFHxS. Specifically, LOQs were as low as 0.01 µg/kg for some PFAS in the food matrices (Table 1).1 The low LOQ levels were due to the sensitivity of the SCIEX 7500 system and the extensive contamination reduction steps that were implemented to reduce the PFAS background signal. The 7 food matrices tested included, baby food puree (beef), milk-based infant formula (as sold), full cream milk powder, fish, whole egg, soluble coffee and fish oil. The method showed good precision at the LOQ level for the 4 regulated compounds. The %CV was <20% for most matrices and <25% for all matrices. 

For more information refer to the publication on which this technical note is based.2

Table 1. The EU recommended (Rec.) and achieved LOQ in-sample concentrations for the 4 regulated PFAS compounds. All units are in µg/kg. (For more details, please see Tables 2 to 5). a: LOQ; b: indicative level. Indicative levels specify that further investigation needs to be performed however, this does not stop the products from being put on the market.

Key features of PFAS analysis in multiple foods when using the SCIEX 7500 system
 

  • Broad PFAS coverage. A method was developed and validated for 57 PFAS compounds following the EURL POPs guidelines.

  • Diverse food matrices. Multiple food matrices were tested including baby food puree (beef), milk-based infant formula (as sold), full cream milk powder, fish, whole egg, soluble coffee and fish oil.

  • Parts-per-trillion (ppt, µg/kg) sensitivity. LOQs in food matrices analysed were as low as 0.01 µg/kg

Introduction


PFAS compounds have now been established to be a major concern because of their inherent accumulation potential. These compounds do not effectively breakdown in the environment, due to the extremely strong carbon-fluorine bond that is prevalent in these compounds. This concern has become apparent to food manufacturers and producers, resulting in regulators increasingly requiring testing for these compounds. Therefore, this method was developed and validated for 57 PFAS compounds in 7 different food matrices.

The scope of analytes quantified includes compounds listed in both the Commission Recommendation EU 2022/1431 and, the U.S. Food and Drug Administration method for PFAS analysis in processed food. Several compounds were also included that were surveyed by service laboratories in Europe and mentioned in literature.3-4 The Commission Regulation EU 2022/2388 was followed to ensure that enforced limits were met for the relevant PFAS compounds mentioned.

The LOQs were established by spiking PFAS-free food items at levels which showed reliable identification and quantitation according to the EURL POPs guidance document criteria.5 Applicability of the method was further verified in 27 commodity check matrices with different moisture content: 6 baby food purees and baby cereals (as sold), 2 milk-based products (whey protein and skimmed milk powder), 4 fish/meat matrices, 4 egg items (egg yolk powder, white egg powder, whole egg powder and processed liquid egg yolk), 2 roasted and ground coffee samples and finally 9 fish oil, vegetable oil and fat matrices. The quantitation of linear PFOS (L-PFOS), PFOA, PFNA and, linear PFHxS (L-PFHxS) was carried out using the isotope dilution approach, to increase analytical rigour. All other compounds were quantified by isotope dilution apart from capstone A and B which were quantified by standard addition.

Methods



Sample preparation: Matrices were homogenized prior to the addition of a mixture of 33 internal standards (ISs). Multiple concentrations were used, depending on the compound (see Table S5 of Reference 5). Then, 5 mL of 80:20 (v/v), acetonitrile/water and a ceramic homogenizer were added to the sample. The sample was shaken at 1500 rpm for 3 min and then centrifuged for 10 min at 4000 g at room temperature. The resulting whole supernatant was transferred into a 50 mL polypropylene centrifuge tube.

Five mL of water was added and mixed and the pH was then adjusted to 7 with 1 mL of 1M phosphate buffer. Solid phase extraction (SPE) was then performed using SPE Strata™ PFAS WAX/GCB cartridges (P/N.: CS0-9207). The eluate was centrifuged at 13000 g for 10 minutes, supernatant transferred to a clean vial for analysis.

Note: Depending on the matrix, a concentration step was performed before the final centrifugation. The full sample preparation is detailed in reference 2.

Chromatography: Separation was performed using an Exion LC AD system. A delay column (Waters XBridge BEH C18, 2.1 x 50 mm, 3.5 µm) was used alongside the analytical column (Waters Acquity UPLC BEH C18 column (2.1 x 100 mm, 1.7 µm)), and guard column (Waters Acquity UPLC BEH C18 guard column, 2.1 x 5 mm, 1.7 µm). 

Mobile phase A was 2mM ammonium acetate solution buffered with ammonium hydroxide (pH adjusted to 10.4). Mobile phase B was 5mM 1-methylpiperidine in 50:50 (v/v) methanol/ acetonitrile. The gradient started with 1% mobile phase B and was held for 3.4 min. At 3.5 min the B% increased to 55%, then to 90% at 9 min, followed by an increase to 100% at 9.5 min, which was held until 11.5 min. Finally, at 11.6 min the B% was decreased to 1% and held for 3.4 min for a total run time of 15 minutes. The column temperature was set at 50°C and the flow rate was 0.4 mL/min. 

Mass spectrometry: The SCIEX 7500 system was used for the analysis.6 The system was operated in negative electrospray ionization mode, with scheduled MRM acquisition performed. See reference 2 for full mass spectrometry details. 

Data processing: All data were processed using SCIEX OS software, version 3.1.
 

Method performance compared to Commission Regulation EU 2022/2388 and Recommendation 2022/1431


In August 2022, the European Union (EU) released the Commission Recommendation EU 2022/1431, which suggested LOQs for PFOS, PFOA, PFNA and, PFHxS in several food matrices, including fruits and vegetables, fish and meat, eggs, fish oil and food for infants and young children. The lowest LOQs recommended for fruits, vegetables and, foods for infants and young children ranged from 0.001 µg/kg to 0.004 µg/kg depending on the compound. In the EURL POPs guidance, the scope of analytes without required LOQs included 32 additional compounds. In December 2022, the Commission Regulation EU 2022/2388 established maximum levels (MLs) for PFOS, PFOA, PFNA and, PFHxS, in eggs, fish and meat foodstuffs. The lowest MLs for PFOA and PFHxS were set at 0.3 µg/kg in eggs and for PFOA, PFNA and PFHxS were set at 0.2 µg/kg in fish and meat (beef, pork, poultry and sheep) intended to produce food for infants and young children. The ML applies to the sum of linear and branched isomers, whether they are chromatographically separated or not. Tables 2-5 below present the LOQs achieved for the 4 regulated compounds, compared with the enforced and and recommended limits. These results show that for the 4 regulated compounds, the enforced limits were easily met with the established method. The LOQs recommended by the EU were achieved for eggs and fish. The limit recommended for infant food, however, was not achieved, as the lowest level achieved was 0.010 µg/kg.

Table 2. The enforced EU limit, recommended LOQ and achieved LOQ in sample for PFOS. All units are in µg/kg. 

Table 3. The enforced EU limit, recommended LOQ and achieved LOQ in sample for PFOA. All units are in µg/kg.

Table 4. The enforced EU limit, recommended LOQ and achieved LOQ in sample for PFNA. All units are in µg/kg.

Table 5. The enforced EU limit, recommended LOQ and achieved LOQ in sample for PFHxS. All units are in µg/kg. 

For each compound, the precision was measured to ensure that the method was repeatable for the 4 regulated compounds in all matrices analyzed. Figure 1 shows whether a %CV ≤20% or <25% was achieved for the LOQ (see Tables 1-5 for the LOQ values). These two specific criteria were based on the EURL POPs guideline that states that a %CV of ≤20% can be used for compliance testing whereas a %CV ≤25% is suitable for monitoring purposes only. 

Figure 1. The %CV at the LOQ for the 4 regulated compounds (PFOS, PFOA, PFNA and PFHxS) in all matrices. Apart from full cream milk powder and yellow pollack fish all the precision data provided a %CV value <20%. All precision data had a %CV <25% indicating that values >20% were within 20–25% and within an acceptable range for monitoring purposes.2

In addition to the 4 regulated compounds, all other analytes were assessed based on criteria defined in the EURL POPs guideline.5 Figure 2 shows the method performance of each compound in matrix at the LOQ and 5x LOQ based on the criteria established in the EURL POPs guideline.

Figure 2. Method performance for the 53 compounds (PFOS, PFOA, PFNA and PFHxS covered in Figure 1) analyzed at both the LOQ and at 5x LOQ. For many of the compounds analyzed, the criteria defined in the EURL POPs guidelines were achieved. It is of note however, that some compounds fell outside of the criteria. The yellow pollack fish was a more challenging matrix to analyze and had the most compounds that did not meet the criteria. In addition, PFBA, PFPeA, HPFHpA, PF4OPeA, NMeFOSE-M and N-EtFOSE-M used only one transition and capstone A and B were only included in full cream powder, baby food and milk-based infant formula.2

Reducing background PFAS contamination


It has been well documented that it can be challenging to achieve a clean blank injection when analyzing PFAS compounds. These challenges arise because these compounds are inherent in the environment and there are therefore many potential sources of contamination. Table 6 highlights the changes made to reduce background contamination. Simple changes in lab equipment and clothing can help to limit PFAS contamination. 

Table 6. Recommended practices to reduce PFAS contamination in the lab. Following the guidelines outlined in this table can help to reduce inherent contamination when analyzing PFAS compounds.

Conclusion
 

  • A method was validated for 57 relevant PFAS compounds in multiple food matrices, using 33 isotopically labelled standards to ensure accurate quantitation

  • Quantitation was achieved in food matrices at ng/kg levels of sensitivity as required by EURL POP guidelines

  • Contamination mitigation steps were used to reduce the background PFAS observed.

  • The Commission Regulation EU 2022/2388 limits were met for the 4 regulated PFAS compounds. The EURL POPs criteria were achieved for most of the compounds in all matrices analyzed

  • The Commission Recommendation EU 2022/1431 indicative limits were met for the four regulated PFAS compounds in baby food

References
 

  1. COMMISSION REGULATION (EU) 2022/2388 amending Regulation (EC) No 1881/2006 as regards maximum levels of perfluoroalkyl substances in certain foodstuff 7 th December 2022.

  2. Theurillat, X.; Mujahid, C.; Eriksen, B.; Griffin, A.; Savage, A.; Delatour, T.; Mottier, P. An LC-MS/MS method for the quantitative determination of 57 per- and polyfluoroalkyl substances at ng/kg levels in different food matrices. Food Additives & Contaminants: Part A. 2023, 40 (7), 862-877.

  3. COMMISSION RECOMMENDATION (EU) 2022/1431 on the monitoring of perfluoroalkyl substances in food, 24 August 2022.

  4. US FDA. C-010.02: Determination of 16 per and polyfluoroalkyl substances (PFAS) in processed food using liquid chromatography-tandem mass spectrometry (LCMS/MS), 19th December 2021.

  5. EURL POPs, Guidance document on analytical parameters for the determination of per- and polyfluoroalkyl substances (PFAS) in food and feed, 11th May 2022.

  6. Enabling new levels of quantification. SCIEX technical note, RUO-MKT-02-11886-A.

  7. AGS guide to environmental sampling, 14th Jan 2019.
Request information