Comparison of PFAS sample preparation between WAX + GCB dispersive and WAX/GCB cartridges from water and soil extracts according to EPA draft 1633


Recent draft EPA Method 1633 specifies the use of a weak anion exchange (WAX) polymeric SPE sorbent in combination with a graphitized carbon black (GCB) sorbent for cleaning solid samples, soil, biota, sediment or non-potable water samples. . This procedure adds time to the cleaning step and presents the possibility of loss of analytes and introduction of inaccuracy. This article describes a significant improvement to guidelines in which both sorbents are contained in a single tube, offering the potential to reduce sample processing time and achieve equivalent accuracy and precision.

Polyfluoroalkyl substances (PFAS) are a group of manufactured chemicals that have been used in industry and consumer products since the 1940s. It is estimated that there are over 6,000 possible forms of PFAS compounds, but most n were not made. The most studied and characterized are perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) (1), which have been replaced by other PFAS in the United States in recent years. The EPA has developed Draft Method 1633, “Analysis of Per- and Polyfluoroalkyl Substances (PFAS) in Aqueous, Solid, Biosolid, and Tissue Samples by LC-MS/MS” (2) to facilitate the analysis of PFAS ( 3). This method was designed as an EPA 1600 series method. The DOD then published QSM 5.4, which had specific QA criteria for 1633. This method involves a two-step solid phase extraction (SPE) approach using weak anion exchange (WAX) plus separate graphitized carbon black (GCB) cleaning using a GCB powder, known as Dispersive GCB (dGCB). For water samples, dGCB powder is added after extraction. The purpose of the additional GCB cleaning step is to remove matrix that may be causing interference and to reduce bias. The limitations of dGCB use are well known, as stated in EPA Draft Method 1633, “it is important to minimize the time the sample extract is in contact with carbon”. Besides these practical limitations, adding dGCB is labor intensive and therefore not ideal due to the extra time required to add, mix, and centrifuge the dGCB for each sample, especially in high throughput labs. The cartridges have therefore been developed as a single stacked cartridge with WAX and GCB sorbents which performs like a traditional SPE cartridge with an integrated polishing step to meet method guidelines. These cartridges reduce the need more extraction tubes compared to current draw method. The utility of WAX and GCB in a stacked SPE format for EPA 533 and 537.1 for a variety of water matrices has previously been demonstrated (4,5). The objective of this study was to validate the performance of the method for the broader list of compounds in EPA 1633 and to demonstrate the same utility for water and soil extracts. Figure 1 shows the SPE cartridges used for this comparative study.

As a method guide, EPA 1633 makes provisions for demonstrating equivalence, “…a laboratory is permitted certain options to improve separations or reduce measurement costs. These options include alternative procedures for extraction, concentration and clean-up, as well as changes to sample volumes, columns and detectors” (2). In order to demonstrate equivalence, initial precision and recovery (IPR) as described in the section 9.2.1 of EPA 1633 were used for comparison studies for water and soil matrices.For aqueous extract comparisons, 500 mL of reagent water for each sample was used for four samples For the comparative soil study, a 5g batch of Ottawa Sand was spiked and extracted.The extraction panel was expanded by adding three additional analytes to reflect the California Water Dashboard for PFAS compounds in wastewater discharges (6).

Comparison of aqueous extracts

For the water samples, the precision of the extractions was determined by comparing the percent relative standard deviation (%RSD) of the spiked native PFAS analytes between cartridges A, B, and C shown in Table 1. water were extracted by two different procedures. Samples were either extracted with WAX followed by dGCB powder as described in EPA 1633, or extracted using WAX/GCB double stacking alone. When using the stacked cartridge, the dGCB was not used.

Figure 2 shows the PFAS analyte extraction precision of the four water samples in % RSD. These results were compared with those obtained by the single laboratory project method. Overall, the precision was lower in this study compared to the published method. Although there were minor variations between phases, the difference in precision between phases was not statistically significant in the sample size. Perfluorooctadecanoic acid (PFODA) was one of the standards tested according to California Water Board recommendations. This analyte had the lowest precision but was still well below 15%.

Comparison of soil extracts

Table 2 shows the results of the comparative study between the dGCB+WAX cartridge and the single GCB/WAX cartridge of PFAS compounds recovered from spiked Ottawa sand extracts.

According to EPA 1633, soil extraction requires an initial extraction of 0.3% methanolic ammonium hydroxide (see section 11.3.4–11.3.7) (2), cleaning with dGCB, then the extract is passed over the SPE cartridge. Thus, cleaning and SPE are in reverse order to water samples. To simplify cleaning and one-step SPE, a stacked cartridge containing dGCB on top of SPE WAX was used. Figure 3 shows the extraction precision of PFAS analytes from soil extracts in % RSD using Ottawa sand as reference material. These results were compared with those obtained by the single laboratory project method. The %RSD for native standards had excellent accuracy below 10% for all PFAS standards except PFODA. Both procedures produced equivalent results for the cartridge comparison tests.


Whether using single WAX ​​+ dGCB cartridges or dual WAX/GCB cartridges, these results demonstrated equivalence for an EPA 1633 PFAS panel from water and soil samples. Dual-layer cartridges with the elimination of dispersive SPE provided performance equivalent to WAX cartridges specified in Draft EPA 1633 method for all 40 EPA 1633 parameters plus PFHxDA, PFODA and 10:2FTS for soil and water, in accordance with the requirements of IDOC (Section 9.1.2) and DOD QSM 5.4 Table B-24. For soil samples, only PFODA – which is not included in the EPA 1633 PFAS panel – failed the initial test demonstration of capability (IDC) for both procedures. In our laboratory, the elimination of dispersive SPE reduced the labor for each analytical batch (20 samples) by approximately 30 min for manual cleaning of the SPE cartridge. Eliminating the filtration step would provide an additional reduction of 30 minutes of work. The incorporation of dual-layer cartridges into the workflow has enabled the automation of the entire cleaning procedure, with the potential for a significant reduction in labor and improved data reproducibility.


  2. US EPA Draft Method 1633, “Analysis of Per- and Polyfluoroalkyl Substances (PFAS) in Aqueous, Solid, Biosolids, and Tissue Samples by LC-MS/MS”:
  4. Phenomenex, Comparison of PFAS Recoveries Between WAX/GCB Cartridge Format and Dispersive GCB for DoD Compliance (TN-0145): you
  5. EPA Draft Method 1633 for PFAS Analysis: Evaluation of SPE Extraction Options. NEMC 2022 Environmental Symposium:

Richard F. Jack is the global market development manager for environmental and food markets at Phenomenex Corporation. He has over 18 years of experience in chromatography and mass spectrometry for the environmental, semiconductor, chemical and pharmaceutical industries. Richard collaborates with global regulatory agencies to develop validated methods through new applications, instruments, column chemistries and software. Richard is a former EPA Science Advisor for the EPA Panel on Hydraulic Fracturing, co-author of EPA Standards 557 and 557.1 as well as ASTM D8001 and method updates D4327 and D6919. He is currently the second vice-chairman of the ASTM D19 subcommittee on water analysis. He was previously responsible for vertical marketing at Thermo Fisher Scientific, Dionex and Hitachi High Technologies, where he designed analytical instruments including ion chromatography (IC) and high performance liquid chromatography (HPLC) systems, pumps, autosamplers and a variety of detectors. . Richard got his Ph.D. in Biochemistry and Anaerobic Microbiology from Virginia Tech University in Blacksburg, Virginia, USA. He received his MSc in Ecology from the University of Tennessee in Knoxville, Tennessee, USA.

Sam Lodge is Business Development Manager – Environmental at Phenomenex in Torrance, California, USA.

Adam Robinson is the LC–MS/MS Analytical Laboratory Supervisor at Bureau Veritas’ Mississauga, Ontario facility. In addition to maintaining operational excellence, he has contributed to the development of new analytical test methodologies in support of the environmental and food science business lines, including the new EPA Draft Method 1633. .

Thushara Johnson is a laboratory technician with Bureau Veritas, Mississauga, Ontario. In addition to performing routine organic processing of environmental samples, she is a member of the PFAS development team; help BV and its customers meet the unique analytical challenges and testing requirements of PFAS.

Adnan Khan is an analyst at Bureau Veritas, Mississauga, Ontario. Prior to joining BV, he was a research assistant at Western University in Canada where he published peer-reviewed articles. He now applies this experience in his current role within the BV research and development team. He looks forward to further contributing in the field of PFAS analytical chemistry R&D.

Heather Lord joined Bureau Veritas in 2012 and is currently Senior Manager of Bureau Veritas Advanced Organic Testing Services. She manages all aspects of ultra-trace testing and reporting on priority and emerging organic pollutants such as PFAS, PCBs, dioxins, furans and PAHs in food and environmental media.


About Author

Comments are closed.