Further Insight into Extractable (Organo)fluorine Mass Balance Analysis of Tap Water from Shanghai, China

The ubiquitous occurrence of per- and polyfluoroalkyl substances (PFAS) and the detection of unexplained extractable organofluorine (EOF) in drinking water have raised growing concerns. A recent study reported the detection of inorganic fluorinated anions in German river systems, and therefore, in some samples, EOF may include some inorganic fluorinated anions. Thus, it might be more appropriate to use the term “extractable fluorine (EF) analysis” instead of the term EOF analysis. In this study, tap water samples (n = 39) from Shanghai were collected to assess the levels of EF/EOF, 35 target PFAS, two inorganic fluorinated anions (tetrafluoroborate (BF4–) and hexafluorophosphate (PF6–)), and novel PFAS through suspect screening and potential oxidizable precursors through oxidative conversion. The results showed that ultra-short PFAS were the largest contributors to target PFAS, accounting for up to 97% of ΣPFAS. To the best of our knowledge, this was the first time that bis(trifluoromethanesulfonyl)imide (NTf2) was reported in drinking water from China, and p-perfluorous nonenoxybenzenesulfonate (OBS) was also identified through suspect screening. Small amounts of precursors that can be oxidatively converted to PFCAs were noted after oxidative conversion. EF mass balance analysis revealed that target PFAS could only explain less than 36% of EF. However, the amounts of unexplained extractable fluorine were greatly reduced when BF4– and PF6– were included. These compounds further explained more than 44% of the EF, indicating the role of inorganic fluorinated anions in the mass balance analysis.

6 mL, 30 μm).In brief, the cartridges were first conditioned by 4 mL of methanol with 0.1% NH 4 OH, 4 mL of methanol and 4 mL of ultrapure water.20 mL of ultrapure water with 0.01% NH4OH, 30 mL of ultrapure water, 4 mL of an ammonium acetate buffer solution (pH=4) and 4 mL of ultrapure water with 20% MeOH were used as the washing step when the loading was finished.After drying under vacuum for 30 min, the cartridges were eluted with 4 mL of methanol with 0.1% NH 4 OH.The extracts were then evaporated under nitrogen gas and recovery standards (RS) were spiked in the extracts of replicate 1 prior to instrumental analysis.
In order to improve TFA recovery, 50 mL of water samples (replicate 2) was extracted for quantification of ultra-short PFAS.The extraction protocol of replicate 2 followed similar procedures as replicate 1 except for skipping the washing step of ultrapure water with 0.01% NH 4 OH, because the reason for low recovery for TFA was due to the washing step aiming at removing inorganic fluoride that might have enriched onto the SPE cartridge.
In addition, individual tap water samples from the same reservoir were grouped into one pool by taking equal volumes to a final total volume of 1 L. Thus, 39 individual tap water samples were grouped into four pools to represent water samples from the 4 reservoirs.They were extracted in the same way as replicate 2 and the extracts were further used for oxidative conversion and suspect screening.The results showed the retention time in samples matched to standards in the three columns.
Moreover, the isotopic patterns of BF 4 -were found to match to natural isotopic distribution ( 11 B/( 11 B+ 10 B) was around 80%) which further supported the identification of BF 4 -(Figure S2-S4 and Table S1).

Figure 2
Figure 2 Extraction workflow for target analysis (A) and extractable fluorine mass balance analysis

Figure 6
Figure 6 Individual PFAS concentrations (ultra-short PFAS excluded) in tap water from different sources (A, B, C and D)

Solid phase extraction (SPE) for water samples
replicates were then extracted in the same protocol using Oasis WAX cartridges (Waters 150 mg,

Table 1
The identification of BF 4 -and PF 6 -

Table 2
Mass spectrometric information for target analysis

Table 4
Reproducibility (batch-to-batch variation) of PFAS (n=7) and internal standard (IS) recoveries in QC samples (homogeneously mixed water samples)

Table 5
Recoveries of IS (spiked before extractions) in real water samples (n=39)

Table 6
Recoveries of IS (spiked before extractions) in oxidative conversion (n=9)

Table 7
MDL and MQL for target analysis

Table 8
Concentrations of PFAS in tap water samples (ng/L, recovery-corrected for target analysis)

Table 9
Concentrations of EF, target PFAS and two inorganic fluorinated anions expressed in F

Table 10
Summary of identified PFAS through suspect screening with a confidence level of 3 or above in pooled tap water samples (n=4)

Table 12
The average concentrations of PFAS in tap water from four reservoirs(A, B, C and D)and the resulting perfluorooctanoic acid equivalent (PEQ) based on RPFs If an individual PFAS was not detected in some samples, these samples were excluded when estimating PEQ of this individual PFAS.Concentrations below MQL were set as MQL.