Feasibility of closing the PFAS mass balance: exploring the potential of liquid sampling atmospheric pressure glow discharge (LS-APGD) with Orbitrap mass spectrometry for neutral PFAS

Abstract

Per- and polyfluoroalkyl substances (PFASs) are anthropogenic compounds of high toxicological impact. Routine PFAS analysis is limited to a few dozen compounds from the more than 12 000 registered ones; the vast majority of them cannot be included in the analysis due to ionisation challenges. The liquid sampling-atmospheric pressure glow discharge (LS-APGD) source has been explored in a wide variety of mass spectrometric techniques, proving its versatility. PFASs have recently been included in the scope of analysis. In this study, we used LS-APGD to analyse several different PFAS forms, including some that can be done routinely, and others that are still challenging. The results show that not only the analysis of more-or-less routine PFAS compounds is possible, but LS-APGD also shows potential for analysing neutral, semi-volatile PFASs as well as unfunctionalized fluorocarbons (perfluoroalkanes). Admittedly, further optimisation is necessary to achieve the required analytical target concentrations. Comparing ski wax extracts to standards suggests that the sample matrix may influence (in this case in a positive way) the signal intensity, which needs to be taken into account in the data analysis process. Plasma discharge conditions have been optimised using a design of experiments approach. This marks the first systematic optimisation of LS-APGD in negative ionization mode for a wide variety of organic compounds. In addition, this is the first instance that LS-APGD has been shown to operate with nonpolar solvents. The body of work presented in this manuscript represents a significant expansion of the capabilities of the already versatile LS-APGD ion source.

---

1. Introduction

Per- and polyfluoroalkyl substances (PFASs) are a large group of anthropogenic chemicals of growing concern due to their toxicological impact. As such, certain PFASs are regulated by the European Union (EU) and globally.^1–3 Guidelines^4–9 for analysis of these PFASs recommend the use of liquid chromatography electrospray ionisation coupled to tandem mass spectrometry (LC-ESI-MS/MS). Due to the nature of ESI, only compounds that are ionic and thus easy to ionise can be measured with ESI, limiting the number of PFASs to a small suite of compounds (approx. 20–50 in the guidelines). However, it has been shown in many studies, with the aid of fluorine mass balance analysis, that the amount of target PFASs that can be quantified with this LC-ESI-MS/MS method is only a small fraction (∼<10%) of what can be extracted usually with methanol from different media (such as sewage water,^10–13 ski wax and soils^14–16 or animal organs^10,17,18). The remaining amounts of organofluorines that are either not contained within the list of target PFAS analysis, non-ionisable with ESI, non-eluting from the analytical column or contain particles, polymers, etc. There are other possibilities to measure volatile and/or polymeric PFASs, such as extraction with a different solvent (e.g. hexane^15,19), or the use of gas chromatography coupled to mass spectrometry (GC-MS); however, due to the tradeoff figures-of-merit of GC and LC, the sample would need to be prepared multiple times, making the process expensive and laborious. Thus, there is a need to measure a wider variety of PFASs in possibly one analysis using liquid chromatography.

It has been demonstrated that liquid sampling-atmospheric pressure glow discharge (LS-APGD) microplasma can be coupled to multiple types of mass analysers,^20–26 making it a versatile tool in research laboratories. Ionisation with LS-APGD microplasma has been shown to successfully ionise compounds that were assumed to be non-ionisable with conventional ionisation sources such as ESI or APCI for LC. Marcus et al. (2013), Zhang et al. (2014 and 2015), Zhang & Marcus (2016), Alves et al. (2020) and Williams et al. (2020)^27–33 demonstrated the versatile use of LS-APGD for organic compounds, even showing that the microplasma source can ionise compounds that lack a reactive functional group, thus usually analysed with GC-MS,^28,33,34 such as polyaromatic hydrocarbons (PAH). Although Marcus and co-workers (2023)²³ in a review collected several examples of organic compounds measured with the LS-APGD source, only recently has the initial demonstration of the ability of how the LS-APGD source might ionise acidic and neutral PFASs (such as FTOHs)³⁵ been described. Unique ionisation characteristics were noted, but no information is available on how well the LS-APGD source works in the presence of a matrix or in alternative mobile phases that might be of chromatographic relevance.

The aim of this work was to measure several ionic and neutral PFASs using LS-APGD ionisation to demonstrate the potential of this microplasma to the PFAS community both in methanol/water solution and in hexane, the latter being a completely novel implementation of the microplasma. We tested not only commercially available standards, but also investigated the matrix effect by spiking non-fluorinated ski wax with neutral PFAS compounds. In addition, the versatility of the LS-APGD ionization source is demonstrated by showing ionization in potential LC mobile phases, specifically a methanol/water mixture for reversed phase (RP) and hexane for normal phase (NP) chromatography. Comparisons are made between the spectra generated by the LS-APGD source and ESI. To the best of our knowledge, this is the first paper presenting PFAS analysis using spiked sample media and hexane with the LS-APGD ionisation source. Finally, a design of experiments (DoE) was performed to optimise plasma discharge conditions for fluorotelomer acids, sulphonic acids, and alcohols in methanol/water and perfluoroalkenes in hexane. This study is the first time LS-APGD has been demonstrated to operate with hexane as the carrier solution and is the first systematic optimisation of LS-APGD for negative ions, let alone PFASs.

2. Experimental

Samples were prepared as ∼10 μg PFAS mL⁻¹ solution of selected PFASs in either a 70 : 30 methanol : ultrapure water solution or in hexane. The compounds selected represent ionic and neutral PFAS varieties. Trifluoroacetic acid, perfluorobutanoic, hexanoic, heptanoic, nonanoic and decanoic acids (TFA, PFBA, PFHxA, PFHpA, PFNA, and PFDA) and perfluorohexanesulfonic acid (PFHxS) were purchased as neat solids from Merck and collectively referred to as ionic PFASs. 8 : 2 fluorotelomerealcohol (FTOH) was purchased from Fluorochem. A mixture containing 4 : 2–10 : 2 FTOH in methanol was purchased from Chiron. Perfluoroheptane (C₇) and perfluorooctane (C₈) were purchased from abcr. The FTOH and perfluoroalkanes are collectively referred to as neutral PFASs.

Source condition optimisation was performed in order to get the highest intensity peaks for a selection of PFASs. The methanol soluble PFASs (TFA, PFHpA, PFOA, PFNA, PFDA, perfluoropropanesulfonic acid (PFPrS) and 8 : 2 FTOH) were included into one experiment, while perfluoroheptane dissolved in hexane was in another experiment. Optimisation was performed on (1) the electrode gap between electrodes (at 3 levels), (2) source position between Orbitrap and solution electrodes (at 3 levels), (3) current (at 4 levels), (4) helium sheath gas flow rate (at 5 levels) and (5) solution flow rate (at 4 levels). A summarising table can be found in Table 1. A design of experiments approach to plasma optimisation allows for the investigation of a single parameter's contribution to analyte responsivity without overlooking important inter-parametric effects. With a traditional one-variable-at-a-time approach (OVAT), inter-parametric effects can easily be overlooked. A Design of Experiments (DoE) model has been applied. The Logworth value is derived from the negative logarithm of the p-value (−log 10(p)), providing a statistical measure of factor significance within the model. A higher Logworth value indicates a stronger contribution of the corresponding factor or interaction to the response variable, while terms below a predefined threshold (commonly Logworth < 1.3, corresponding to p > 0.05) are considered statistically insignificant and removed to refine the model. This approach helps eliminate noise and improves model robustness by retaining only the most influential parameters.^36,37 While DoEs for LS-APGD source parameter optimisation have been performed previously, these are the first DoEs in negative mode for organic compounds and the first DoE using hexane as the carrier solution.

Table 1 MS conditions and optimisation parameters used for methanol soluble and hexane soluble PFASs

	Methanol soluble	Hexane soluble
MS conditions
Scan range (Da)	100–600	100–400
Resolution	70 000
Microscans	5
AGC target	3 × 10^6
Max inj. time (ms)	100
Capillary temperature (°C)	150
Injection volume (μL)	20
Optimisation parameters
Electrode gap (mm)	0.5
	1.0
	1.5
Source position (mm)	4.0
	7.0
	10
Current (mA)	15
	20
	30
	40
Gas flow (mL min^−1)	200	350
	300	400
	400	500
	500	600
	600
Solution flow (μL min^−1)	10
	20
	30
	40

---

To investigate potential matrix effects, a ski wax (Toko non-fluorinated ski wax), which has been previously shown to not contain organofluorines¹⁵ was extracted as follows: 0.2 g of ski wax was weighed into 15 mL centrifuge tubes and then either 5 mL of 70 : 30 methanol : water or 5 mL of hexane were added. The resulting mixture was sonicated for an hour and then left overnight. On the following day, the mixtures were centrifuged, and the supernatant was decanted and then measured. In two separate vials, one containing 0.2 g wax was spiked with the mixture containing PFBA, PFHxA, PFHpA, PFNA, PFDA, PFHxS and the FTOH mixture and extracted with 70 : 30 methanol : water, while the other was spiked with perfluoroheptane and perfluorooctane, and extracted with hexane as described above.

The LS-APGD Orbitrap coupling has been described previously.^32,33 Standard operating conditions were employed throughout this exploratory effort unless otherwise stated. The helium gas flow was set to 500 mL min⁻¹ and the plasma voltages were set to ∼550 V. The analyte mixture is introduced into the inner capillary using a 20 μL min⁻¹ flow rate via a syringe pump. The LS-APGD source was coupled to an LTQ XL Orbitrap, without any modifications apart from exchanging the electrospray source. The system was operated in the negative ion trap and negative Fourier transform MS modes.³⁴ Plasma optimisation studies were conducted using a Thermo Q Exactive, again with no modifications other than the removal of the ESI source. No collisional dissociation of any kind was used during this study.

3. Results and discussion

3.1 Ionic PFAS standards

The concentration of the standards ranged from 1.9 μg PFAS mL⁻¹ (TFA) to 32 μg PFAS mL⁻¹ (PFHpA) in methanol : water. These compounds are readily ionisable with ESI, making them a perfect benchmark for comparison between the two ion sources. In ESI, perfluorocarboxylic and sulfonic acids produce a quasi-molecular ion [M–H]⁻. In the case of perfluorocarboxylic acids, the primary fragments are the [M–COOH]⁻ fragment followed by secondary daughter ion fragments at m/z 119, m/z 169 and m/z 219 via fluorine migration shifts.³⁸ In the case of perfluorosulfonic acids, the preferred fragments from [M–H]⁻ are [SO₃]⁻ and [SO₃F]⁻ (m/z 80 and 99 respectively).³⁸ When analysing PFSA and PFCA compounds with LS-APDG, the mass spectra obtained (Fig. 1a–d and g) reveal a comparable pattern with the [M–H]⁻ species being clearly visible, suggesting that PFSA and PFCA fragment via the same mechanism in both ESI and LS-APGD. Diagnostic fragments, such as the so-called “quantifier” fragments for perfluorocarboxylic and sulfonic acids, [M–COOH]⁻ and [SO₃]⁻ (m/z 79.95811), are also clearly visible in the corresponding spectra. Additional diagnostic fragments, such as [C₂F₅]⁻ or [C₃F₇]⁻, are also present for PFDA and PFHxA, confirming that the fragmentations indeed belong to PFAS compounds (Table S1†). Additionally, it should be noted that lock masses were not used to improve the mass accuracy in these studies. These demonstrate that LS-APGD produces reproducible spectra for ionic PFASs with comparable fragmentation patterns to ESI.

Fig. 1 (a–g) Spectral composition of different PFASs; (a–d) ionic perfluorocarboxylic acids, (e and f) neutral fluorotelomer alcohols and (g) ionic perfluorosulfonic acid in 70 : 30 methanol : water, using LS-APGD.

3.2 Neutral PFAS standards

The concentration of standards was around 8 μg PFAS mL⁻¹ prepared either in methanol : water or in hexane solvent. Neutral PFAS compounds such as fluorotelomers are usually not susceptible to ESI. Only a handful of research groups have shown the analysis of such PFASs with atmospheric pressure ionisation (API)^39–41 or with ESI,⁴² with most of the researchers using the adduct formation of FTOH with acetate to yield a negative [M + CH₃COO]⁻ complex,⁴⁰ making the comparison between ESI and LS-APGD less straightforward.

Measuring 6 : 2 and 8 : 2 FTOH showed the presence of [M–H]⁻ species (Fig. 1e and f), though the main fragments are of the form [M–C₂H₄OH]⁻. These are the same species found in the previous studies of the LS-APGD source, where Goodwin et al.³⁵ presented a straight forward mechanism for the structural rearrangement. The common loss of that entity yields an anion that represents the fluoroalkyl chain and thus a direct measure of the base molecule. Given the fact that the observed fragment ion is the result of a complex rearrangement rather than a simple loss of protons, observing the same species on two non-equivalent Orbitrap systems provides strong support for the general utility of the ionization method.

Previously it was seen that the mobile phase composition used for the analysis of FTOHs could influence the fragments that were formed in APCI.⁴⁰ Using methanol/water and the API source leads to the formation of [M–H]⁻ ions,⁴⁰ which agrees with our study. Ayala-Cabrera and co-workers,⁴⁰ who used API, discussed two possibilities for fragmentation of FTOHs: (1) ions observed at odd m/z values likely originate from the loss of HF units and CO and (2) ions observed at even m/z values are likely the products of in-source fragmentations of [M]⁻˙ followed by the successive loss of HF units. Trier and co-workers,⁴² who used ESI for FTOH determination, also showed a similar fragmentation pattern from the [M–H]⁻ ion. Their proposed fragmentation pattern shows the losses of HF, CO, CH₂O and F₂. In our case, apart from the [M–H]⁻ species, several different diagnostic fragments were observed, such as [C₃F₇]⁻, [C₄F₉]⁻, [C₅F₁₁]⁻ and [M-C₂H₄OH]⁻ (Fig. 1e and f). However, the fragments are not formed through the loss of HF or other units, but through the successive loss of [CF₂]⁻, which can be explained by the fact that the plasma is more energetic than the ionisation sources API and ESI.

In order to validate the feasibility of LS-APGD to ionise non-polar PFAS, we prepared two fluorocarbons (perfluoroalkanes) in hexane with a concentration of around 170 mg mL⁻¹ and analysed them. To reiterate, the operation of LS-APGD in a non-polar solvent such as hexane had never been demonstrated before. Previously, it was found that n-perfluoroalkanes break down under an electron impact (EI) source in gas chromatography to a series of perfluoroalkenyl ions [C_nF_2n+1]⁺ because of the detachment of F⁻ and [CF₂]⁻, while the [CF₃]⁻ fragment is the same for all perfluoroalkane homologues.⁴³ Upon analysing the two perfluoroalkanes in the positive mode with the LS-APGD source, no peaks were visible at their corresponding m/z values (Fig. S2†), and thus the analysis was repeated in the negative ion mode. Both C₇ and C₈ perfluoroalkanes were visible and produced mass spectra (Fig. 2a), albeit with low intensity despite the high concentrations used. It needs to be mentioned again that the same discharge conditions were used throughout the analysis for all PFAS. Here using the same conditions with the non-polar solvent as what was used in the case of methanol : water. Interestingly, in the case of C₇ perfluoroalkane, beyond the [M]⁻ ion mass, an additional peak at m/z 388.08 (plus ∼0.1 Da) was observed. The same solution was measured using the ESI source, where there were no peaks found at the corresponding m/z values (Fig. 2b). Hexane and methanol/water blanks were also investigated for the presence of FTOHs and perfluoroalkanes, but no peaks were observed, suggesting that the observed peaks in the standard spectrum indeed originate from the PFAS standards (Fig. S3†).

Fig. 2 Mass spectra of C₇ and C₈ perfluoroalkanes in hexane using LS-APGD in the negative mode (a) and mass spectra of C₇ and C₈ perfluoroalkanes in hexane using ESI and ionisation sources in the negative mode (b).

As final comments, in the case of FTOHs and perfluoroalkenes, the intensities were observed to decrease with increasing mass in the homologue series, due to the lower ionisation potential across the homologue series, which is expected to occur with ionic PFASs as well.

3.3 Design of experiments model

It is important to perform optimisation to better understand the feasibility of LS-APGD at more relevant sample concentrations. Given that many plasma discharge conditions need optimisation, previous studies^{25,33,44–46} have indicated that a design of experiments (DoE) approach is a powerful way to thoroughly test parameter combinations across the experimental design space to determine optimal conditions without resorting to testing every possible parameter configuration. All user-defined plasma conditions have been included in the experimental design space. These factors include interelectrode gap, current, sheath gas flow rate, solution flow rate, and source position relative to the mass spectrometer entrance. The design also included second-order terms (i.e., current × current) as well as cross terms (i.e., current × solution flow rate). Fig. 3 presents Logworth values for the full and reduced DoE models. The full model includes all terms investigated during the DoE, while the reduced model has had low contributing terms (those with a Logworth < 1.3, indicating that the term is outside of the 95% confidence interval) removed to reduce noise from the model. After removing the terms contributing to the noise, the reduced model shows the most important factors contributing to analyte response. Correlation coefficients (actual vs. predicted) were found between 0.81 and 0.91 for each analyte in the reduced model, indicating good agreement between the model predicted response and observation. In the methanol soluble PFAS model, the most significant parameter is the solution flow. Other relevant factors include current × solution flow and current × gas flow cross terms. The electrode gap remained in the reduced model due to it being part of the cross terms that are significant. In the hexane soluble PFAS model, the most significant factor is the position × position cross term, although gas flow, electrode gap, current and position are also significantly influencing the model. A table showing the optimised conditions for all 8 methanol soluble PFAS as a mix and as individual compounds as well as for hexane soluble PFAS can be found in Table 2.

Fig. 3 Full and reduced DoE models for methanol (a) and hexane (b) soluble PFAS.

Table 2 Optimised conditions for methanol soluble PFAS (all 8 compounds), hexane soluble PFAS and individual PFAS

	C7 fluoroalkane^a	TFA^b	PFHpA^b	PFOA^b	PFNA^b	PFDA^b	PFPrS^b	PFHxS^b	8 : 2 FTOH^b	Compromised^b
a In hexane. b In methanol.
Electrode gap (mm)	0.5	0.5	0.5	0.5	0.5	0.5	0.5	0.5	0.5	0.5
Position (mm)	7	4	7	4	4	7	4	4	10	7
Current (mA)	15	40	40	40	40	40	40	40	40	40
He gas flow (mL min^−1)	600	200	200	200	200	200	200	200	200	200
Solution flow (μL min^−1)	40	40	40	40	40	40	40	40	40	40

---

3.4 Analysis of methanol-soluble PFAS

The calculated models for the individual analytes show no trend in chain length; moreover, for the 8 analytes, only one parameter (source position) changes between the “individual analyte” models (Table 3). Comparing the individual analyte model to the “all-analyte” model, the predicted impact decreases minimally (3–28%), thus indicating that the same set of plasma discharge conditions should be suitable for a wide variety of PFAS. The optimised conditions were tested for sensitivity and repeatability by injecting (n = 10) the mixture of the 8 PFAS at 3 μg mL⁻¹ (8 : 2 FTOH) and at 30 ng mL⁻¹ (all other PFAS). Repeat injections (n = 10) showed a relative standard deviation (RSD%) of 8–18% for perfluorocarboxylic acids (PFCAs) and perfluorosulfonic acids (PFSAs), and 25% for 8 : 2 FTOH. PFSAs and PFCAs showed high sensitivity with the LS-APGD source, with counts in the 10⁶−10⁸ region/ng PFAS compound. Interestingly, using ESI, PFCAs have higher intensity compared to PFSAs using standards at the same concentration, while with LS-APGD, PFSAs show a higher sensitivity, possibly due to the different plasma chemistry. However, a deeper exploration is necessary to identify the exact cause of this phenomenon. Additionally, using ESI, the ionisation efficiency decreases with increasing chain length, resulting in longer PFAS having lower intensity peaks for similar concentrations. Using LS-APGD, however, this trend is not observed. The sensitivity observed for 8 : 2 FTOH was much lower than for PFCAs and PFSAs, with counts being in the 10⁴ region/ng PFAS compound (Fig. 4 and Table 3). However, this still represents a notable advancement compared to ESI, where FTOHs are undetected as [M–H]⁻, highlighting the capability of the LS-APGD source to ionise neutral, non-polar PFAS.

Table 3 Intensities of perfluorocarboxylic and sulfonic acids, fluorotelomer alcohol and C₇ perfluoroalkane with LS-APGD using the optimised conditions

PFAS	[M–H]^− intensity with LS-APGD (counts per ng)
a Counts per mg.
TFA	255 735
PFHpA	1 979 430
PFOA	924 742
PFNA	905 306
PFDA	2 663 604
PFPrS	24 378 800
PFHxS	33 334 686
8 : 2 FTS	6175
C7 perfluoroalkane	11 434^a

---

Fig. 4 Log scale representation of sensitivity of the 8 methanol-soluble PFAS using the optimised conditions for all analytes in counts per ng PFAS. This figure clearly shows that the LS-APGD source is compound specific.

3.5 Analysis of hexane-soluble PFAS

After the optimisation of the parameters, a solution containing 170 mg per mL perfluoroheptane in hexane was injected repeatedly (n = 10) to get information on reproducibility and sensitivity. The RSD of ten repeat injections was found to be 25%, which agrees with the results observed with the methanol soluble PFAS. The sensitivity was found to be 11 counts per μg perfluoroheptane, which is significantly lower than what was observed with the methanol soluble PFAS. Fig. 6 shows that the [M]⁻ peak is not the most dominant peak in the spectrum, suggesting the presence of complex plasma chemistry. The most intense peaks were annotated based on accurate mass as different fluorine containing fragments (Table 4) possibly due to the reaction with air that was entrained in the plasma, leading to the formation of oxygen and nitrogen containing fluoroalkane fragments. However, there are peaks that cannot be elucidated easily, and thus future studies should focus on the identification of peaks in this spectrum and on elucidating the mechanisms underlying their formation within the helium micro-plasma environment.

Table 4 Annotation of fragmentation of the perfluorocarbon C₇F₁₆

Measured m/z	Annotated formulae	Theoretical m/z	Error (ppm)
	C7F16	387.97445
368.9760	C7F15^−	368.97605	0.13
196.9838	C4F7O^−	196.9837	0.31
229.9853	C4F8OHN^−	229.9852	0.37
179.9885	C3F6OHN^−	179.9884	0.51

---

3.6 Spiked wax extracts

Non-fluorinated waxes were spiked with various PFAS compounds (4 : 2 FTOH, PFHxS, TFA, PFHxA, and PFDA) as described in the Experimental section in methanol : water and hexane solvents. The resultant concentrations in the extract were between 0.2 μg mL⁻¹ (TFA) and 3.1 μg mL⁻¹ (PFHxA). As expected, no PFAS-related peaks were observed in the mass spectrum of the non-fluorinated wax. Spiking the non-fluorinated wax with the perfluoroalkanes, however, showed high intensity peaks at the corresponding m/z for C₇ and C₈ perfluoroalkanes (m/z 387.97 and 437.96), TFA (m/z 112.98), PFHxA (m/z 268.98), PFHxS (m/z 398.93), PFDA (m/z 512.96) and 4 : 2 FTOH (m/z 263.01) (Fig. S5 and S6†). Comparing the signal intensities in the spiked wax and a reference standard solution, it can be seen in Fig. 5 and S7† that spiked wax showed ∼4–5 times higher intensity peaks than the standard solution, suggesting a possible influence of the sample matrix. One explanation for this could be that with the introduced matrix, there is an increase in the electron density in the glow discharge. Further investigation of this aspect would be necessary to provide insight into the ionisation mechanism; however, this was not within the scope of this study. Nonetheless, the influence of the matrix should be considered during quantification. For comparison, the ionisation source was changed back to ESI, and the same standards were analysed; however, no peaks were observed for either FTOHs or the perfluoroalkanes (Fig. 2b), showing that LS-APGD ionises non-polar compounds more effectively than ESI. Thus, the LS-APGD source very clearly provides information about PFAS populations not afforded by the ESI source.

Fig. 5 Comparison of mass spectra of perfluoroalkanes in hexane and in spiked wax extracted with hexane, using LS-APGD in the negative mode. Spiked and extracted wax have higher intensities despite having the same concentration as the standard, showing the positive matrix effect.

Fig. 6 Mass spectrum of perfluoroheptane in hexane, showing the peaks formed in the source.

4. Conclusion

In this feasibility study, we explored uncharted territories and presented the analysis of not only ionic and semi-polar PFAS, but also non-polar perfluorinated alkanes. These compounds need much more investigation in combination with liquid sampling ionisation sources such as ESI or APCI, to benefit from the versatility and potential of LS-APGD microplasma for PFAS analysis. The microplasma source can handle diverse solvents compatible with different liquid chromatography modes such as methanol/water for RP and hexane for NP separations. Future work should focus further on identifying microplasma operating conditions for the highest sensitivity using additional PFAS compound groups (such as fluorotelomer sulphates, sulfonamides, sulfonamidoethanols, etc.), ultrashort chain PFAS compounds, and replacement chemistries (such as GenX) to enable a comprehensive PFAS analysis. Connection to reverse phase chromatography with a methanol/water gradient and to normal phase chromatography with hexane should also be explored. Analytical figures of merit and quantification utilising ¹³C isotope labelled PFAS also need to be investigated deeper.

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors thank the University of Graz for contributing to the publication. DB and VG are funded by the Empa Advanced Analytical Technologies innovation credit.

References

1. United Nations Environment Programme, Decision SC-4/17.Listing of perfluorooctane sulfonic acid, its salts and perfluorooctanesulfonylfluoride, in Conference of the Parties to the Stockholm Con-vention on Persistent Organic Pollutants, Geneva (CH), 2009, http://www.pops.int/Portals/0/download.aspx?d=UNEP-POPS-COP.4-SC-4-17.English.pdf, last accessed: 14. Nov. 2024.
2. U.S. Environmental Protection Agency, EPA and 3M Announce Phaseout of PFOS News Release, Washington (DC), 2000, available at: https://archive.epa.gov/epapages/newsroom_archive/newsreleases/33aa946e6cb11f35852568e1005246b4.html, last accessed: 14. Nov. 2024.
3. U.S. Environmental Protection Agency, 2010/15 PFOA stewardship program, 2006, available at: https://www.epa.gov/assessing-and-managing-chemicals-under-tsca/fact-sheet-20102015-pfoa-stewardship-program, last accessed: 14. Nov. 2024.
4. L. Rosenblum and S. C. Wendelken, Method 533 Determination of Per- and Polyfluoroalkyl Substances in Drinking Water by Isotope Dilution Anion Exchange Solid Phase Extraction and Liquid Chromatography/Tandem Mass Spectrometry, U.S. Environmental Protection Agency, Washington, DC, 2019, last accessed: 14. Nov. 2024.
5. J. A. Shoemaker, P. Grimmett and B. Boutin, Method 537 Determination of Selected Perfluorinated Alkyl Acids in Drinking Water by Solid Phase Extraction and Liquid Chromatography/Tandem Mass Spectrometry (LC/MS/MS), U.S. Environmental Protection Agency, Washington, DC, 2008, last accessed: 14. Nov. 2024.
6. J. Shoemaker and D. Tettenhorst, Method 537.1 Determination of Selected Per- and Polyflourinated Alkyl Substances in Drinking Water by Solid Phase Extraction and Liquid Chromatography/Tandem Mass Spectrometry (LC/MS/MS), U.S. Environmental Protection Agency, Washington, DC, 2020, last accessed: 14. Nov. 2024.
7. Analysis of Per- and Polyfluoroalkyl Substances (PFAS) in Aqueous, Solid, Biosolids, and Tissue Samples by LC-MS/MS, U.S. Environmental Protection Agency, Washington, DC, 2024, last accessed: 14. Nov. 2024.
8. SW-846 Test Method 8327: Per-and Polyfluoroalkyl Substances (PFAS) by Liquid Chromatography/Tandem Mass Spectrometry (LC/MS/MS), U.S. Environmental Protection Agency, Washington, DC, 2019, last accessed: 14. Nov. 2024.
9. Other Test Method 45 (OTM-45) Measurement of Selected Per- and Polyfluorinated Alkyl Substances from Stationary Sources, U.S. Environmental Protection Agency, Washington, DC, 2024, last accessed: 14. Nov. 2024.
10. A. Kärrman, T. Wang, R. Kallenborn, A. M. Langseter, S. M. Grønhovd, E. M. Ræder, J. L. Lyche, L. Yeung, F. Chen, U. Eriksson, R. Aro and F. Fredriksson, PFASs in the Nordic Environment : Screening of Poly- and Perfluoroalkyl Substances (PFASs) and Extractable Organic Fluorine (EOF) in the Nordic Environment, TemaNord NV - 2019:515, Nordisk Ministerråd, Copenhagen,  DOI:10.6027/TN2019-515, last accessed: 14. Nov. 2024.
11. A. Kärrman, L. W. Y. Yeung, K. M. Spaan, F. T. Lange, M. A. Nguyen, M. Plassmann, C. A. De Wit, M. Scheurer, R. Awad and J. P. Benskin, Can determination of organofluorine (EOF) be standardized? First interlaboratory comparisons of EOF and fluorine mass balance in sludge and water matrices, Environ. Sci.: Processes Impacts, 2021, 23(10), 1458–1465.
12. R. Aro, U. Eriksson, A. Kärrman, F. Chen, T. Wang and L. W. Y. Yeung, Fluorine mass balance analysis of effluent and sludge from nordic countries, ACS ES&T Water, 2021, 1(9), 2087–2096.
13. V. Müller, A. Kindness and J. Feldmann, Fluorine mass balance analysis of PFAS in communal waters at a wastewater plant from Austria, Water Res., 2023, 244, 120501.
14. P. Roesch, C. Vogel, T. Huthwelker, P. Wittwer and F. G. Simon, Investigation of per- and polyfuoroalkyl substances (PFAS) in soils and sewage sludges by fuorine K-edge XANES spectroscopy and combustion ion chromatography, Environ. Sci. Pollut. Res., 2022, 29, 26889–26899.
15. V. Müller, L. C. A. Costa, F. S. Rondan, E. Matic, M. F. Mesko, A. Kindness and J. Feldmann, Per and polyfluoroalkylated substances (PFAS) target and EOF analyses in ski wax, snowmelts, and soil from skiing areas, Environ. Sci.: Processes Impacts, 2023, 25(10), 1926–1936.
16. B. Tan, T. Wang, P. Wang, W. Luo, Y. Lu and K. Y. Romesh, et al., Per and polyfluoroalkyl substances in soils around the Nepali Koshi River: Levels, distribution, and mass balance, Environ. Sci. Pollut. Res., 2014, 21(15), 9201–9211.
17. D. Herzke, V. Nikiforov, L. W. Y. Yeung, B. Moe, H. Routti, T. Nygård, G. W. Gabrielsen and L. Hanssen, Targeted PFAS analyses and extractable organofluorine – Enhancing our understanding of the presence of unknown PFAS in Norwegian wildlife, Environ. Int., 2023, 171, 107640.
18. T. Schröder, V. Müller, M. Preihs, J. Borovicka, d. V. Gonzalez, A. Kindness and J. Feldmann, Fluorine mass balance analysis in wild boar organs from the Bohemian Forest National Park, Sci. Total Environ., 2024, 922, 171187.
19. R. Gonzalez de Vega, M. Plassmann, D. Clases, K. Zangger, V. Müller, E. Rosenberg, A. Reimann, L. Skedung, J. P. Benskin and J. Feldmann, A multi-platform approach for the comprehensive analysis of per- and polyfluoroalkyl substances (PFAS) and fluorine mass balance in commercial ski wax products, Anal. Chim. Acta, 2024, 1314, 342754.
20. S. Konegger-Kappel, B. T. Manard, L. X. Zhang, T. Konegger and R. K. Marcus, Liquid sampling-atmospheric pressure glow discharge excitation of atomic and ionic species, J. Anal. At. Spectrom., 2014, 30(1), 285–295.
21. R. K. Marcus, H. W. Paing and L. X. Zhang, Conceptual Demonstration of Ambient Desorption-Optical Emission Spectroscopy Using a Liquid Sampling-Atmospheric Pressure Glow Discharge Microplasma Source, Anal. Chem., 2016, 88(11), 5579–5584.
22. R. K. Marcus, B. T. Manard and C. D. Quarles, Liquid sampling-atmospheric pressure glow discharge (LS-APGD) microplasmas for diverse spectrochemical analysis applications, J. Anal. At. Spectrom., 2017, 32(4), 704–716.
23. R. K. Marcus, E. D. Hoegg, K. A. Hall, T. J. Williams and D. W. Koppenaal, Combined atomic and molecular (CAM) ionization with the liquid sampling-atmospheric pressure glow discharge microplasma, Mass Spectrom. Rev., 2023, 42(2), 652–673.
24. T. J. Williams and R. K. Marcus, Coupling of the liquid sampling-atmospheric pressure glow discharge (LS-APGD) ionization source with a commercial triple-quadrupole mass spectrometer, J. Anal. At. Spectrom., 2019, 34(7), 1468–1477.
25. T. J. Williams and R. K. Marcus, Development of an integrated, single electrode liquid sampling – atmospheric pressure glow discharge microplasma ionization source, Spectrochim. Acta, Part B, 2021, 179, 105994.
26. T. J. Williams, E. D. Hoegg, J. R. Bills and R. K. Marcus, Roles of collisional dissociation modalities on spectral composition and isotope ratio measurement performance of the liquid sampling – atmospheric pressure glow discharge/orbitrap mass spectrometer coupling, Int. J. Mass Spectrom., 2021, 464, 116572.
27. R. K. Marcus, C. Q. Burdette, B. T. Manard and L. X. Zhang, Ambient desorption/ionization mass spectrometry using a liquid sampling-atmospheric glow discharge (LS-APGD) ionization source, Anal. Bioanal. Chem., 2013, 405(25), 8171–8184.
28. L. X. Zhang, B. T. Manard, S. K. Kappel and R. Kenneth Marcus, Evaluation of the operating parameters of the liquid sampling-atmospheric pressure glow discharge (LS-APGD) ionization source for elemental mass spectrometry, Anal. Bioanal. Chem., 2014, 406(29), 7497–7509.
29. L. X. Zhang, B. T. Manard, B. A. Powell and R. K. Marcus, Preliminary Assessment of Potential for Metal-Ligand Speciation in Aqueous Solution via the Liquid Sampling-Atmospheric Pressure Glow Discharge (LS-APGD) Ionization Source: Uranyl Acetate, Anal. Chem., 2015, 87(14), 7218–7225.
30. L. X. Zhang and R. K. Marcus, Mass spectra of diverse organic species utilizing the liquid sampling-atmospheric pressure glow discharge (LS-APGD) microplasma ionization source, J. Anal. At. Spectrom., 2016, 31(1), 145–151.
31. M. R. Alves, J. S. Sauer, K. A. Prather, V. H. Grassian and C. L. Wilkins, Liquid Sampling-Atmospheric Pressure Glow Discharge Ionization as a Technique for the Characterization of Salt-Containing Organic Samples, Anal. Chem., 2020, 92(13), 8845–8851.
32. T. J. Williams, J. R. Bills and R. K. Marcus, Mass spectrometric characteristics and preliminary figures of merit for polyaromatic hydrocarbons via the liquid sampling-atmospheric pressure glow discharge microplasma, J. Anal. At. Spectrom., 2020, 35(11), 2475–2478.
33. T. J. Williams and R. K. Marcus, Coupling the liquid sampling-atmospheric pressure glow discharge, a combined atomic and molecular (CAM) ionization source, to a reduced-format mass spectrometer for the analysis of diverse species, J. Anal. At. Spectrom., 2020, 35(9), 1910–1921.
34. C. Masucci, K. O. Nagornov, A. N. Kozhinov, K. Kraft, Y. O. Tsybin and D. Bleiner, Evaluation of atmospheric-plasma-source absorption mode Fourier transform Orbitrap mass spectrometry for chlorinated paraffin mixtures, Anal. Bioanal. Chem., 2024, 416(23), 5133–5144.
35. J. V. Goodwin, C. Masucci, D. Bleiner and R. K. Marcus, Initial demonstration of microplasma ionization/Orbitrap mass spectrometry for molecular screening of perfluorinated compounds, J. Anal. At. Spectrom., 2024, 39(10), 2353–2362.
36. D. C. Montgomery, Design and Analysis of Experiments, John Wiley & Sons, 9th edn, 2017.
37. M. J. Anderson and P. J. Whitcomb, DOE Simplified: Practical Tools for Effective Experimentation, CRC Press, 4th edn, 2020.
38. G. Arsenault, A. McAlees, R. McCrindle and N. Riddell, Analysis of perfluoroalkyl anion fragmentation pathways for perfluoroalkyl carboxylates and sulfonates during liquid chromatography/tandem mass spectrometry: evidence for fluorine migration prior to secondary and tertiary fragmentation, Rapid Commun. Mass Spectrom., 2007, 21(23), 3803–3814.
39. B. Szostek, K. B. Prickett and R. C. Buck, Determination of fluorotelomer alcohols by liquid chromatography/tandem mass spectrometry in water, Rapid Commun. Mass Spectrom., 2006, 20(19), 2837–2844.
40. J. F. Ayala-Cabrera, F. Javier Santos and E. Moyano, Negative-ion atmospheric pressure ionisation of semi-volatile fluorinated compounds for ultra-high-performance liquid chromatography tandem mass spectrometry analysis, Anal. Bioanal. Chem., 2018, 410(20), 4913–4924.
41. J. F. Ayala-Cabrera, E. Moyano and F. J. Santos, Gas chromatography and liquid chromatography coupled to mass spectrometry for the determination of fluorotelomer olefins, fluorotelomer alcohols, perfluoroalkyl sulfonamides and sulfonamido-ethanols in water, J. Chromatogr. A, 2020, 1609, 460463.
42. X. Trier, J. H. Christensen and W. M. A. Niessen, Negative electrospray ionisation of fluorotelomer alcohols (FTOH) and FTOH-derived acrylate surfactants by liquid chromatography coupled to accurate (tandem) mass spectrometry - DTU Findit, 39th International Symposium on High-Performance-Liquid-Phase Separations and Related Techniques (HPLC2013), 2013, last accessed: 27. August 2024.
43. N. Kagramanov, Algorithms for fragmentation of n-alkanes and n-peroforalcanes, Fluorine Notes, 2020, 3–4,  DOI:10.17677/FN20714807.2020.01.02.
44. H. W. Paing, T. J. Bryant, C. D. Quarles Jr and R. K. Marcus, Coupling of Laser Ablation and the Liquid Sampling-Atmospheric Pressure Glow Discharge Plasma for Simultaneous, Comprehensive Mapping: Elemental, Molecular, and Spatial Analysis, Anal. Chem., 2020, 92(18), 12622–12629.
45. J. V. Goodwin, B. T. Manard, B. W. Ticknor, P. Cable-Dunlap and R. K. Marcus, Initial Characterization and Optimization of the Liquid Sampling-Atmospheric Pressure Glow Discharge Ionization Source Coupled to an Orbitrap Mass Spectrometer for the Determination of Plutonium, Anal. Chem., 2023, 95(32), 12131–12138.
46. S. Shrestha, J. V. Goodwin, B. T. Manard and R. K. Marcus, Parametric optimization of the liquid sampling-atmospheric pressure glow discharge ionization source coupled to an Orbitrap mass spectrometer for neodymium isotope ratio determinations, Int. J. Mass Spectrom., 2025, 508, 117385.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ja00136f

---