Quantitative Analysis of PFAS in Drinking Water by LC-MS/MS according to GB 5750.8-2023

Jack Wang
R&D APAC lab, Shanghai, China

Abstract

This method for the simultaneous determination of 27 PFAS (per-and polyfluoroalkyl substances) in drinking water by ultra-high performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS) was established in accordance with GB 5750.8-2023. Water samples were enriched and concentrated using a weak anion exchange (WAX) solid-phase extraction column. For all 27 analytes, an excellent linear response was observed in the range of 5 to 200 µg/L (injected standard solutions), and the correlation coefficient R² was greater than 0.9950 for each molecule. The limits of quantitation for 27 PFAS compounds in the water samples were 2.97-4.92 ng/L. The recoveries were 79.0-83.4%, and the relative standard deviations were 1.6-4.6%.

Overview of Sections:

• Introduction
• Experimental
  
• Results & Discussion
• Conclusion

Introduction

PFAS are persistent organic compounds that exist widely in the environment. Recent awareness has drawn attention to the toxicity of these substances. Liquid chromatography-mass spectrometry (LC-MS) is a common technique for the determination of perfluoroalkanes.^1-5

In the new version of the Chinese national standard for drinking water, GB 5750.8-2023⁶, sections 84 and 85 designate 11 PFAS compounds as critical target analytes for drinking water testing. Part 84 of GB 5750.8 stipulates 11 perfluorinated compounds to be analyzed, which are PFBA, PFPeA, PFHxA, PFHpA, PFOA, PFNA, PFDA, PFBS, PFHxS, PFHpS, and PFOS. The sample pretreatment method, LC-MS/MS method conditions, and qualitative and quantitative ion information are described in detail. Part 85 of GB 5750.8. requires only the determination of PFOS (Figure 1). The sample pretreatment method and other method details are the same as for section 84. The required Limit of Quantification (LOQ) for the analytes is given with 5 ng/L, and the linear concentration range for the external calibration is given with 5.0-200 µg/L (representing concentrations in the water sample of 5-200 ng/L considering the sample pre-treatment enrichment factor of 1:1000) with linearity R² >0.9950.

Figure 1. Example PFAS (chemical structure of perfluorooctanesulfonic acid, PFOS)

In this study, 27 PFAS target compounds (Table 1), including PFBA, PFMPA, PFPeA, and PFMOBA, were selected for detection and analysis in drinking water based on the guidelines in the updated standard GB 5750.8-2023 and target compounds established by the EPA. Method verification indicators and internal quality control were implemented in accordance with the provisions of the GB 5750.8-2023 method. 

Table 1. PFAS analytes in scope of this study (*marked are in GB Method)

Compound no.	Compound name	Compound abbreviation	CAS No.
1*	Heptafluorobutyric acid (75930)	PFBA	375-22-4
2	Perfluoro-3-methoxypropanoic acid (73014)	PFMPA	377-73-1
3*	Perfluorovaleric acid (73551)	PFPeA	2706-90-3
4*	Perfluorobutanesulfonic acid (93634)	PFBS	375-73-5
5	2,2,3,3,4,4-Hexafluoro-4-(trifluoromethoxy)butanoic acid	PFMOBA	863090-89-5
6	Perfluoro(2-ethoxyethane)sulfonic acid	PFEESA	113507-82-7
7	Perfluoro-3,6-dioxaheptanoic acid (04292)	3,6-OPFHpA	151772-58-6
8	1H,1H,2H,2H-Perfluorohexanesulfonic acid	4:2 FTSA	757124-72-4
9*	Perfluorohexanoic acid (93899)	PFHxA	307-24-4
10	Perfluoropentanesulfonic acid hydrate	PFPeS	2706-91-4
11	Perfluoro-2-(propyloxy)propionic acid (94275)	HFPO-DA	13252-13-6
12*	Perfluoroheptanoic acid	PFHpA	375-85-9
13*	Perfluorohexanesulfonic acid	PFHxS	3871-99-6
14	4,8-dioxa-3H-Perfluorononanoic acid	ADONA	919005-14-4
15	1H,1H,2H,2H-Perfluorooctanesulfonic acid (93497)	6:2 FTSA	27619-97-2
16*	Perfluoroheptanesulfonic acid (78049)	PFHpS	375-92-8
17*	Perfluorooctanoic acid (93973)	PFOA	335-67-1
18*	Perfluorooctanesulfonic acid (33607)	PFOS	2795-39-3
19*	Perfluorononanoic acid (05167)	PFNA	375-95-1
20*	Perfluorodecanoic acid (91367)	PFDA	335-76-2
21	Sodium 1H,1H,2H,2H-perfluorodecane sulfonate	8:2 FTSA	27619-96-1
22	N-Methylperfluorooctanesulfonamidoacetic acid	N-MeFOSAA	2355-31-9
23	Perfluoroundecanoic acid (89988)	PFUnDA	2058-94-8
24	N-Ethylperfluorooctane sulfonamidoacetic acid (94707)	EtFOSAA	2991-50-6
25	Perfluorododecanoic acid (76467)	PFDoDA	307-55-1
26	Pentacosafluorotridecanoic acid (76705)	PFTrDA	72629-94-8
27	Perfluorotetradecanoic acid (38400)	PFTeDA	376-06-7

Experimental

Standard, Sample, and Reagent Preparation

Reagent solutions, standards, and samples were prepared according to the following procedures as described in the GB method.

Reagent Preparation

• Aqueous methanol (30/70): Combine 30 mL methanol and 70 mL water and mix well (prepare freshly).
• Ammonia solution in aqueous methanol (0.1%): Add 500 μL ammonia (28.0-30.0% NH₃) to a 500 mL volumetric flask, fill to the mark with aqueous methanol and mix well (prepare fresh).
• Ammonium acetate aqueous solution (0.025 mol/L): Dissolve 0.9635 g ammonium acetate in 500 mL water, mix well, and adjust to pH 4 with acetic acid (≥99.8%).
• Ammonium acetate aqueous solution (0.005 mol/L): Dissolve 0.3854 g ammonium acetate in 1 L water and mix well. 

Standard and Sample Preparation

Standard preparation

• PFAS stock solution I (1000 mg/L): Dissolve 10.0 mg of each of the 27 PFAS reference materials in 10 mL methanol. Keep the resulting solution at 0 °C to 4 °C; it can be stored for three months.
• PFAS stock solution II (10 mg/L): Pipette 1 mL of PFAS stock solution Ⅰ into a 100 mL amber glass volumetric flask and fill up to the mark with methanol. Store at 0°C to 4°C and protected from light.
• PFAS stock solution III (1 mg/L): Pipette 1 mL of PFAS stock solution II into a 10 mL amber glass volumetric flask and fill up to the mark with methanol. Store at 0°C to 4°C and protected from light.
• PFAS standard solutions 1-5: Prepare a total of five standard working solutions (nos. 1-5) by pipetting 5 μL, 10 μL, 50 μL, 100 μL, and 200 μL, respectively, of PFAS stock solution III into five separate 1 mL vials. Fill vials up to the mark with methanol. The concentrations of PFAS in the resulting solutions are 5.0, 10.0, 50.0, 100, and 200 μg/L, respectively.

Sample preparation

• Water sample: Dissolve 4.625 g ammonium acetate in 1 L of the water sample to be analyzed and mix well. The pH of the resulting water sample is 6.8-7.0.
• Extraction by SPE:

1. Conditioning: Connect the SPE-WAX solid-phase extraction tube bed wt. 500 mg, volume 6 mL (Supelclean™ ENVI-WAX, 54057-U), to a PFAS-free Visiprep^™ SPE vacuum manifold (57030-U), and condition with 5 mL ammonia-methanol solution (0.1%), 7 mL methanol, and re-equilibrate with 10 mL water.
2. Loading: Pipette pre-treated sample (1 L) onto the SPE tube; in order not to negatively affect recovery rates, a flow rate of 8 mL/min must not be exceeded.
3. Washing: After the sample is finished (Step 2 complete), wash the sample with 5 mL ammonium acetate aqueous solution followed by 12 mL water.
4. Drying: Dry the SPE-WAX solid-phase extraction tube for 15 min under vacuum.
5. Elution: Elute with 5 mL methanol and 7 mL ammonia solution in methanol. Collect eluate in a 15 mL polypropylene centrifuge tube.
6. Evaporation & reconstitution: Blow down samples to almost complete dryness using nitrogen and a water bath (temperature ≤ 40°C). Reconstitute with aqueous methanol (30/70) to result in a 1 mL sample volume and vortex. This represents an enrichment of 1:1000.

• Spiking experiments: For the determination of method recovery (%) and precision, prepare two samples by mixing 1 L of drinking water sample with 30 μL and 50 μL of PFAS stock solution III, respectively. The concentration of PFAS in the two samples is 30 ng/L (utilized for the analysis of precision) and 50 ng/L (recovery).

LC-MS/MS Analysis 

Contamination by PFAS compounds originating from the HPLC system and materials used in analysis is a concern. Therefore, a highly retentive Ascentis^® Express PFAS Delay column was placed in front of the injector to provide retention of PFAS compounds originating from the mobile phase system across various mobile phase conditions. It delays the instrument background and prevents interference with peaks originating from the analyzed/injected samples.

For the separation, a Fused-Core^® Ascentis^® Express PFAS HPLC column was used, which is application tested for PFAS analysis to provide reliable and efficient separations.

The obtained sample extracts and standards were analyzed by LC-MS/MS using the parameters described in Tables 2 & 3.

Table 2. LC-MS/MS Conditions used for PFAS analysis in drinking water

LC Conditions
Instrument:	WatersTM Acquity I-class plus and WatersTM Xevo TQS UPS-tandem mass spectrometer equipped with an electrospray ionization (ESI) source.
Column:	Ascentis® Express 90 Å PFAS, 2.7 μm, 10 cm × 2.1 mm I.D. (53559-U)
Delay column:	Ascentis® Express 90 Å PFAS Delay, 2.7 μm, 5 cm × 3.0 mm I.D. (53572-U)
Detection:	MS/MS, ESI(-) (Table 3), MRM transition, see Table 4
Mobile phase:	[A] 5 mM Ammonium acetate in water; [B] Methanol
Gradient:	Time (min)	% B	% A
	0	33.0	67.0
	18.0	98.0	2.0
	18.1	100	0
	21.0	100	0
	21.1	33.0	67.0
	26.0	33.0	67.0
Flow rate:	0.30 mL/min
Column temp.:	35 °C
Injection:	5.0 μL

Table 3. MS Conditions used for PFAS detection

MS Parameter
Polarity:	ESI (-)
Capillary voltage:	4.0 kV
Atomization gas flow:	3.0 mL/min
Heater flow	10.0 mL/min
Heat block temperature:	400 °C
DL tube temperature:	250 °C
Interface temperature:	300 °C
Scan:	Multiple reaction monitoring (MRM) See Table 4

Results & Discussion

Drinking water samples were pre-teated with ammonium acetate, prepared by solid-phase extraction (SPE), and analyzed by UHPLC-MS/MS, with the MS being operated in MRM mode. The 27 PFAS target analytes were quantified using an external standard calibration. In Figure 2, the result for an unspiked drinking water sample is displayed. Table 4 lists the MRM transition used and the chromatographic data obtained by the analysis of a spiked drinking water sample.

Figure 2. LC-MS/MS chromatogram of an unspiked drinking water sample.

Table 4. Chromatographic data and MRM transitions for the analysis of a drinking water sample spiked with PFAS at a concentration of 50 ng/L of each PFAS

Compound	Retention time (min)	MRM Transitions m/z
PFBA	4.70	213→169
PFMPA	4.79	327→307
PFPeA	5.65	263→219
PFMOBA	5.93	449→80
PFBS	6.25	299→80
4:2FTSA	6.32	230→85
3,6-OPFHpA	6.45	295→80
PFHxA	6.57	313→269
PFEESA	6.65	349→80
HFPO-DA	6.89	285→169
PFPeS	7.11	315→135
PFHpA	7.44	363→319
ADONA	7.62	377→251
PFHxS	7.89	399→80
6:2FTSA	8.03	498→78
PFOA	8.22	413→369
PFHpS	8.58	279→85
PFNA	8.91	463→419
PFOS	9.19	499→80
8:2FTSA	9.40	549→80
PFDA	9.52	513→469
N-MeFOSAA	9.78	570→419
PFUnDA	10.03	563→519
EtFOSAA	10.04	584→419
PFDoDA	10.47	613→569
PFTrDA	10.85	663→619
PFTeDA	11.15	713→669

Calibration

The external calibration for all 27 PFAS using 5 PFAS standard solutions (c = 5.0, 10.0, 50.0, 100, and 200 μg/L) provided R² linearity values between 0.9982 and 0.9999.

As an example, the results of the calibration for PFBA are displayed in Figure 3 and Table 5. All other PFAS provided similar results (see the table in the appendix).

Figure 3. Calibration curve obtained by the analysis of PFBA standard solutions 1-5 (c = 5.0, 10.0, 50.0, 100, and 200 μg/L).

Table 5. Calibration data for PFBA as an example obtained by the analysis of PFAS standard solutions 1-5 (c = 5.0, 10.0, 50.0, 100, and 200 μg/L)

Sample	PFBA concentration (μg/L)	Area
PFAS standard solution 1	5.0	1161
PFAS standard solution 2	10.0	2998
PFAS standard solution 3	50.0	12990
PFAS standard solution 4	100	27980
PFAS standard solution 5	200	57960

Data Precision and Recovery (%)

The drinking water sample with a PFAS concentration of 30 ng/L was used for evaluation of method precision (n=7), while the second sample (PFAS concentration 50 ng/L) was used to assess method recovery (n=6). Summarized results obtained are shown in Table 6. For precision, the %RSD ranged from 1.6 to 4.6%, and the recovery rates were between 79.0 and 83.4%.

Table 6. Average precision %RSD (n=7) for a 30 ng/L spiked sample and the average recovery (n=6) for a 50 ng/L spiked sample

Compound	Precision 30 ng/L Sample (n=7)		Average Recovery 50 ng/L Sample (n=6)
	Average (µg/L)	RSD (%)	%
PFBA	27.3	1.6	81.3
PFMPA	27.0	2.1	80.9
PFPeA	26.7	2.4	82.5
PFBS	26.8	2.4	82.6
PFMOBA	26.2	2.9	82.1
PFEESA	27.5	3.1	79.9
3,6-OPFHpA	27.1	2.9	80.4
4:2 FTSA	26.0	3.7	83.4
PFHxA	27.9	1.9	82.8
PFPeS	26.6	2.5	80.9
HFPO-DA	28.1	3.5	81.1
PFHpA	26.3	3.8	83.1
PFHxS (µg/L)	26.7	3.7	81.7
ADONA	27.3	1.6	82.4
6:2 FTSA	28.1	2.6	81.3
PFHpS	25.3	2.2	80.6
PFOA	27.9	3.8	82.0
PFOS	27.5	2.6	79.0
PFNA	26.4	1.2	82.8
PFDA	27.0	2.9	81.5
8:2 FTSA	27.8	3.7	80.5
N-MeFOSAA	27.8	2.8	82.7
PFUnDA	28.2	3.4	81.0
EtFOSAA	26.5	4.6	80.5
PFDoDA	26.9	2.0	82.8
PFTrDA	27.7	3.8	81.3
PFTeDA	26.7	4.1	80.5

Sensitivity

For the sensitivity determination for the LC method, the baseline noise of a blank drinking water sample was employed: 3N/X was used to determine Limit of Detection (LOD), and 10N/X was used to determine Limit of Quantification (LOQ). LODs ranged from 0.99-1.64 µg/L and LOQ from 2.97-4.92 µg/L for the LC method (Table 7) representing concentrations in the water samples of 0.99-1.64 ng/L and 2.97-4.92 ng/L respectively.

Table 7. LOQ & LOD of 27 PFAS compounds in water samples.

Compound no.	Compound	LOD (ng/L)	LOQ (ng/L)
1	PFBA	1.05	3.15
2	PFMPA	1.13	3.39
3	PFPeA	1.14	3.42
4	PFBS	1.08	3.24
5	PFMOBA	1.61	4.83
6	PFEESA	1.62	4.86
7	3,6-OPFHpA	1.15	3.45
8	4:2 FTSA	1.13	3.39
9	PFHxA	1.59	4.77
10	PFPeS	0.99	2.97
11	HFPO-DA	1.16	3.48
12	PFHpA	1.09	3.27
13	PFHxS	1.11	3.33
14	ADONA	1.52	4.56
15	6:2 FTSA	1.55	4.65
16	PFHpS	1.07	3.21
17	PFOA	1.61	4.83
18	PFOS	1.09	3.27
19	PFNA	1.04	3.12
20	PFDA	1.07	3.21
21	8:2 FTSA	1.09	3.27
22	N-MeFOSAA	1.11	3.33
23	PFUnDA	1.64	4.92
24	EtFOSAA	1.02	3.06
25	PFDoDA	1.12	3.36
26	PFTrDA	1.09	3.27
27	PFTeDA	1.10	3.30

Conclusion

A method for the simultaneous determination of 27 PFAS in drinking water by UHPLC-MS/MS using a Fused-Core^® Ascentis^® Express PFAS column was developed. An Ascentis^® Express PFAS Delay column was installed to prevent instrument background interferences. Water samples were enriched and concentrated by SPE using a Supelclean^® ENVI-WAX weak anion exchange tube. 

The 27 PFAS showed for the external calibration a good linear relationship in the range of 5-200 µg/L (representing the applied sample preparation water sample concentration of 5-200 µg/L), and the correlation coefficients R2 were greater than 0.9950. The recoveries were 79.0-83.4%, the relative standard deviations were 1.6-4.6%, and the limits of quantitation (LOQ) for the LC method for 27 PFAS compounds were 2.97-4.92 μg/L, representing concentrations in the water samples of 2.97-4.92 ng/L. 

The method performed excellently regarding selectivity, peak shape, and necessary retention and can therefore be used for the detection and analysis of PFAS in drinking water in accordance with the GB 5750.8-2023 standard with high accuracy and reliability.

Explore our PFAS standards portfolio and Solvents for PFAS analysis.

REFERENCES

1. Llorca M, Farré M, Picó Y, Barceló D. 2009. Development and validation of a pressurized liquid extraction liquid chromatography–tandem mass spectrometry method for perfluorinated compounds determination in fish. Journal of Chromatography A. 1216(43):7195-7204. https://doi.org/10.1016/j.chroma.2009.06.062

2. Li Z, Li C, Wen Z, Yan H, Zou C, Li Y, Tian L, Lei Z, Li H, Wang Y, et al. 2021. Perfluoroheptanoic acid induces Leydig cell hyperplasia but inhibits spermatogenesis in rats after pubertal exposure. Toxicology. 448152633. https://doi.org/10.1016/j.tox.2020.152633

3. Arvaniti OS, Asimakopoulos AG, Dasenaki ME, Ventouri EI, Stasinakis AS, Thomaidis NS. Simultaneous determination of eighteen perfluorinated compounds in dissolved and particulate phases of wastewater, and in sewage sludge by liquid chromatography-tandem mass spectrometry. Anal. Methods. 6(5):1341-1349. https://doi.org/10.1039/c3ay42015a

4. Vavrouš A, Vápenka L, Sosnovcová J, Kejlová K, Vrbík K, Jírová D. 2016. Method for analysis of 68 organic contaminants in food contact paper using gas and liquid chromatography coupled with tandem mass spectrometry. Food Control. 60221-229. https://doi.org/10.1016/j.foodcont.2015.07.043

5. Pérez F, Nadal M, Navarro-Ortega A, Fàbrega F, Domingo JL, Barceló D, Farré M. 2013. Accumulation of perfluoroalkyl substances in human tissues. Environment International. 59354-362. https://doi.org/10.1016/j.envint.2013.06.004

6. GB 5750.8-2023 Standard test methods for drinking water - Part 8: Indicators of organic matter. . [Internet]. Available from: http://down.foodmate.net/standard/sort/3/137087