Comparison of pressurized solvent and reflux extraction methods for the determination of perfluorooctanoic acid in polytetrafluoroethylene polymers using LC-MS-MS

Abstract

Both pressurized solvent extraction (PSE) and reflux extraction in various solvents were used to select the most efficient system for the determination of the quantity of perfluorooctanoic acid (PFOA) present in polytetrafluoroethylene polymers. After evaporating the solvent, PFOA was determined using liquid chromatography tandem mass spectrometry. Ethanol, water and methanol gave comparable results and were shown to be good solvents for this extraction. Acetonitrile was a reasonable solvent using the reflux extraction method, but not with PSE. Chloroform resulted in poor recovery for both extraction methods. PSE proved to be the more efficient extraction method.

---

Introduction

In 1999 Moody et al. reported the presence of perfluorinated compounds in groundwater impacted by fire-fighting activities, even though the activities had concluded many years before the analyses were conducted.¹ Several investigators have reported that human sera contained low concentrations (viz., ppb) of perfluorinated organic molecules [e.g., perfluorooctane sulfonate (PFOS), perfluorooctanoic acid (PFOA)] in the general population.^2–5 Other reports showed that low levels of these and similar compounds could be found in wildlife and in the environment.^6–8

Organofluorine compounds have unusual properties that make them difficult to measure.⁹ In addition, perfluorocarboxylates are used in the production of many fluoropolymers¹⁰ as a processing aid.^11,12 Fluoropolymers are important inert components commonly used in laboratory apparatus and analytical laboratory instrumentation. If residual perfluorocarboxylates are present in analytical systems, their levels and ease of extraction will be important in deciding the low levels of quantitation needed, especially for exposure, environmental, and toxicological studies.

In this study, we compare solvents and extraction methods used for the determination of perfluorooctanoic acid (PFOA), a common fluoropolymer processing aid, in two polytetrafluoroethylene (PTFE) polymers.

Based on the work of Vandenburg et al.,¹³ we selected pressurized solvent extraction and boiling under reflux to determine optimal conditions for these extraction measurements. Since PFOA is quite soluble in water¹⁴ and since water is a good solvent used to ascertain potential exposure to PFOA from perfluoropolymers, water was selected for this study. Ethanol has been used for studies involving food contact in FDA studies.¹⁵ Acetonitrile and methanol are common solvents. Chloroform was evaluated since it is a common halogenated solvent.

Experimental

Apparatus

Reflux extractions were done with a 500-mL round-bottom flask with glass magnetic stirring bar and a glass water condenser. The pressurized solvent extraction was done with a Dionex (Sunnyvale, CA) accelerated solvent extractor (ASE^® series 200).

Materials and reagents

Polytetrafluoroethylene fluoropolymer resins were obtained from a commercial lot and an intermediate of a commercial lot (average particle size approximately 500 µm). Liquid chromatographic grade or analytical grade water, methanol, acetonitrile, chloroform and standard Ottawa sand were obtained from EMD Chemicals, Inc. (Darmstadt, Germany). Ethanol (100%) was obtained from Sigma–Aldrich (Milwaukee, WI). A dual ¹³C enriched perfluorooctanoic acid was synthesized in house.

Standards for the seven-point calibration curve were prepared by dilution with methanol from a 1000 ppb (µg L⁻¹) standard solution to concentrations of 0.5, 1, 5, 10, 25, 50, and 100 ppb.

Reflux extraction procedure

Five reflux extraction systems were run concurrently. One hundred mL of solvent was added to each 500-mL round-bottom flask with a glass stirring bar and reflux water condenser. The solvent was refluxed for 20 min and the solvent discarded. Then, 250 mL of the solvent of interest was added to each flask. To one of the flasks was added 1.0 mL of the recovery check standard that was subsequently used to determine percent recovery of PFOA. Approximately 0.75 g of each polymer was weighed (± 0.1 mg) and added to three of the remaining flasks. The fifth flask had only solvent, which served as the solvent blank. The solvents in the prepared flasks were refluxed for two hours, cooled and filtered through a no. 40 filter paper (Whatman, Clifton, NJ). The solvents were then evaporated on a Buchi rotary evaporator (Brinkmann, Westbury, NY) to remove most of the solvent (approximately 10 mL remaining). The remaining solvent was then transferred to either a 30 mL or 60 mL disposable collection vial and evaporated to dryness under nitrogen on a temperature-controlled heating block (ambient for chloroform, 45 °C for methanol, 55 °C for ethanol and acetonitrile, and 85 °C for water).

Pressurized solvent extraction procedure

Five pressurized solvent extractions were performed concurrently. The instrument's solvent reservoir was filled with the appropriate solvent and the lines rinsed four times with approximately 5 mL of solvent. The cells were filled with Ottawa sand to approximately 3 mm from the top. Each cell was closed and loaded onto the instrument. The cells with sand were preconditioned. The preconditioning cycle was set to 1500 psi (10.3 kPa), 150 °C, 7 min heating time, 100% volume flush, 240 second purge for one cycle. The solvents were then discarded and new collection vessels set in place. Approximately 5/6 of the sand was removed from three of the vessels and a weighed quantity of polymer added to each vessel. Each cell was refilled to 3 mm from the top with the sand and the lid secured. To the fourth sand-filled cell 1.0 mL of recovery check standard was added on top of the sand and the lid secured. The fifth sand filled flask served as the solvent blank. The extraction conditions were set to 1600 psi (11.0 kPa), 150 °C, 7 min heating time, 100% volume flush, 240 s purge for 4 cycles. The collection vials were removed when the extraction was complete and placed in a heating block where they were evaporated to dryness under nitrogen (ambient for chloroform, 45 °C for methanol, 55 °C for ethanol and acetonitrile and 85 °C for water).

Analytical method

The concentration of PFOA in the reconstituted extracts was determined using high performance liquid chromatography coupled with negative ion electrospray tandem mass spectrometry (LC-MS-MS) (Micromass Quatro Ultima, Beverly, MA). The extracts were reconstituted by adding 1 mL of 100% methanol to the vial and shaking for 30 min on a wrist-action shaker. The methanol was then placed into a 5-mL volumetric flask, and brought to volume with the LC mobile phase A, as described below. The dual ¹³C-enriched standard (final concentration 50 ppb) was added as an internal standard to all the reconstituted samples.

The analyte was separated using an Agilent 1100 liquid chromatograph (Wilmington, DE) modified with low dead-volume internal tubing. A guard column, Hypersil C₁₈ 2 × 50 mm (Thermo Keystone, Bellefonte, PA), was installed between the mixer and the autoinjector. Twenty-five microliters of the reconstituted extract were injected onto a Hypersil ODS 2.1 × 200 mm (Thermo Keystone, Bellefonte, PA) at a flow rate of 0.3 mL min⁻¹ and maintained at 60 °C. Duplicate injections were made for all samples.

Initial gradient conditions were 15% mobile phase B, where mobile phase A is 2 mM ammonium acetate–1% methanol and B is 100% methanol. A linear gradient was used from 15–67% B over 16 min. The conditions were returned to 15% B for two additional minutes. Typical elution time for PFOA was approximately 16.5 min. Fig. 1 shows a typical total ion chromatogram of the internal standard and the PFOA analyte. In negative ion mode, PFOA is observed as an anion at 413 [CF₃(CF₂)₆COO⁻]. The internal standard is observed at 415 [CF₃(CF₂)₅¹³CF₂¹³COO⁻].

Fig. 1 LC–MS–MS of a PTFE extract showing in the upper trace the internal standard (415 > 370), lower trace the PFOA analyte (413 > 369).

Quantitative analysis was performed using selected ion monitoring for the transition of 413 > 369 (loss of CO₂) for the analyte and 415 > 370 (loss of ¹³CO₂) for the internal standard. Any samples that fell outside the calibration were diluted appropriately and reanalyzed. A seven-point linear calibration curve was prepared (not including zero) for external calibration. Each calibrant concentration set was run in duplicate and bracketed the samples. A typical calibration curve consisted of all fourteen calibration points. With no weighting, the acceptance criterion for the calibration curve required a correlation coefficient, R, ≥0.992.

A methanol blank was run after each 100 ppb standard. The acceptance criterion was set so that the lowest level PFOA standard's area was at least five-times greater than the area of the methanol (solvent) blank. (No quantifiable area was observed in the blank.)

The limit of quantitation (LOQ) was set at 0.5 ppb, the concentration of the lowest calibration standard. A signal between the blank background and the LOQ was defined as not quantifiable (NQ).

Results and discussion

In the extraction of the PTFE polymers with the various solvents, a recovery check standard was run. The recovery check standard vessel contained just the solvent with 500 ng mL⁻¹ of PFOA added in order to determine the recovery of PFOA in the solvent. Table 1 summarizes the results of the recovery check standard for each solvent used in this study. The recoveries of PFOA for methanol, water and ethanol for both extraction methods were acceptable (80–120%). With acetonitrile acceptable recovery was observed using reflux extraction: however, using PSE as the extraction system, acetonitrile had no detectable recovery of PFOA. The results indicated that there was no recovery of PFOA using chloroform by either the reflux extraction or pressurized solvent extraction.

Table 1 Check standard percent recovery

Solvent^a	Reflux^b	PSE^bc
a Chloroform was purchased with either 1% hexane or 1% ethanol as the stabilizer and the % ethanol was increased by volume to 5%.b Results are an average of two measurments and * denotes an average of three measurements.c A second solvent extraction with methanol was made with the extraction efficiency indicated in the parenthesis
Methanol	97	95
Acetonitrile	89	4.1 (0.31)
Ethanol	104	99
Water	103	96
Chloroform–1% hexane	0.9	2.8 (0.06)
Chloroform–1% ethanol	60*	0 (58)
Chloroform–5% ethanol	42*	45 (11)

---

The initial extraction results with chloroform stabilized with 1% hexane were unexpected and additional trials using different stabilizers were performed. The chloroform modified with 1% or 5% ethanol did not extract PFOA quantitatively or reproducibly. The initial extraction in the PSE was followed by a re-extraction with methanol, showing an improvement in the recovery to 57%. This indicates that the chloroform extraction was ineffective. Recoveries for the water and alcohols ranged from 89% to 104%.

To ensure that the low sample recoveries observed in chloroform were not a result of loss during evaporation of the solvent, a sample of PFOA (500 ng mL⁻¹) in chloroform was dried overnight in the hood. An identical sample was evaporated to dryness with nitrogen and heat. These samples were reconstituted and analyzed by the analytical method. The observed recoveries were similar, 103% and 94%, respectively. This indicates that the low recoveries in chloroform extracts are not the result of losses during the evaporation or reconstitution process, but are due to inefficient extraction of PFOA in chloroform stabilized with either 1% hexane or 1% ethanol. The experiment with chloroform containing 5% ethanol showed modest recovery, indicating that the ethanol plays a role in the extraction efficiency.

Tables 2 and 3 show data for the spike recovery and the extraction results for the two different fluoropolymers (PTFE) comparing reflux extraction with pressurized solvent extraction. The results in Table 2 are comparable to those shown in Table 1. For water and alcohols under these conditions, PSE percent recoveries were generally the same as the reflux extraction method. The amount of analyte extracted and detected using the reflux extraction method appeared to be the same level for the water and alcohols (Table 3). The level of analyte detected using the water and alcohols was greater under the same PSE conditions than that with acetonitrile and chloroform. PSE is the more efficient extraction method for determining PFOA content in PTFE polymers (Table 3). The PTFE sample labeled polymer I is from a commercial lot and polymer II is the intermediate of a commercial lot, showing that the finishing steps significantly reduce the PFOA concentration in the commercial polymer. When the chloroform extraction was followed by methanol extraction and the two quantities added together (Polymer II, Table 3), the combined solvent results agreed with the water and alcohols extraction results under the same PSE conditions (1580 ppb).

Table 2 Check standard percent recovery from PTFE samples

Solvent	Reflux temperature/°C	Polymer I		Polymer II
		Reflux	PSE	Reflux	PSE
a Chloroform stabilized with 1% hexane was used for these extractions.b The results reported is the sum of the initial extraction using chloroform stabilized with 1% ethanol followed by a second extraction with methanol.
Methanol	65	93.1	88.4	92.9	89.2
Acetonitrile	82	91.3	0	86.4	7.9
Ethanol	78	99.1	85.5	104	92.5
Water	100	104	104	104	106
Chloroform	61	3.1^a	0^a	36.1^b	59.5^b

---

Table 3 Concentration of PFOA (ppb) extracted from PTFE polymers

Solvent	Reflux temperature/°C	Polymer I		Polymer II
		Reflux	PSE	Reflux	PSE
a NQ is nonquantifiable, indicating that the observed signal was less than the lowest calibrant.b Chloroform stabilized with 1% hexane was used for these extractions.c The result reported is the sum of the initial extraction using chloroform stabilized with 1% ethanol followed by a second extraction with methanol.
Methanol	65	42	140	485	1420
Acetonitrile	82	46	NQ^a	412	241
Ethanol	78	46	135	535	1100
Water	100	46	59	493	1040
Chloroform	61	NQ^ab	NQ^ab	59^c	1580^c

---

Conclusions

Ethanol, methanol and water efficiently extracted PFOA from PTFE polymers. PSE was able to extract more analyte in a shorter time period than reflux extraction with water, methanol and ethanol. Acetonitrile as the solvent gave reasonable recoveries with reflux extraction but not with PSE. Chloroform gave unacceptable results by either extraction method. This result supports the use of water, ethanol, or methanol for extracting PFOA from PTFE matrices.

Acknowledgements

The authors thank Charles F. Koerting, Donald A. Wilson Jr. and Gregory A. Urove for their advice and assistance.

References

1. C. A. Moody and J. A. Field, Environ. Sci. Technol., 1999, 33, 2800.
2. K. J. Hansen, L. A. Clemen, M. E. Ellefson and H. O. Johnson, Environ. Sci. Technol., 2001, 35, 766.
3. Z. Kuklenyik, J. A. Reich, J. S. Tully, L. L. Needham and A. M. Calafat, Environ. Sci. Technol., 2004, 38, 3698.
4. K. Kannan, S. Corsolini, J. Falandz, G. Fillman, K. S. Kumar, B. G. Loganathan, M. A. Mohd, J. Olivero, N. V. Wouwe, J. H. Yang and K. M. Aldous, Environ. Sci. Technol., 2004, 38.
5. K. Inoue, F. Okada, R. Ito, S. Kato, S. Sasaki, S. Nakajima, A. Uno, Y. Saijo, F. Sata, Y. Yoshimura, R. Kishi and H. Nakazawa, Environ. Health Perspect., 2004, 112, 1204.
6. M. M. Schultz, D. F. Barofsky and J. A. Field, Environ. Eng. Sci., 2003, 20, 487.
7. K. Kannan, S. Corsolini, J. Falandysz, G. Oehme, S. Focardi and J. P. Giesy, Environ. Sci. Technol., 2002, 36, 3210.
8. J. W. Martin, M. M. Smithwick, B. M. Braune, P. F. Hoekstra, D. C. G. Muir and S. A. Mabury, Environ. Sci. Technol., 2004, 38, 373.
9. J. W. Martin, S. A. Mabury, K. Kannan, U. Berger, P. De Voogt, J. Field, J. Franklin, J. P. Giesy, T. Harner, D. C. G. Muir, B. Scott, M. A. Kaiser, U. Jarnberg, K. C. Jones, H. Schroeder, M. Simcik, C. Sottani, B. Van Bavel, A. Karrman, G. Lindstrom and S. Van Leeuwen, Environ. Sci. Technol., 2004, 38, 248A.
10. J. G. Drobny, Technology of Fluoropolymers, CRC, Boca Raton, FL, 2001.
11. The Society of the Plastics Industry, Inc., Guide to the Safe Handling of Fluoropolymer Dispersions, Washington, DC, 2001.
12. E. Kissa, Fluorinated Surfactants and Repellents, Marcel Dekker, NY, 2nd edn., 2001.
13. H. J. Vandenburg, A. A. Clifford, K. D. Bartle, J. Caroll and I. D. Newton, Analyst, 1999, 124, 397.
14. The Society of the Plastics Industry, Inc., Detecting and Quantifying Low Levels of Fluoropolymer Polymerization Aids–A Guidance Document, Washington, DC, 2003.
15. U.S. Food and Drug Administration, Guidance for Industry: Preparation of Food Contact Notifications and Food Additive Petitions for Food Contact Substances: Chemistry Recommendations, Rockville, MD, 2002.

---