the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 09 Jan 2025
Measurement report: Per- and polyfluoroalkyl substances (PFAS) in particulate matter (PM10) from activated sludge aeration
Abstract. Environmental pollution with per- and polyfluoroalkyl substances (PFAS), commonly referred to as “forever chemicals”, received significant attention due to their environmental persistence and bioaccumulation tendencies. Effluents from wastewater treatment plants (WWTPs) have been reported to contain significant levels of PFAS. Wastewater treatment processes such as aeration have the potential to transfer PFAS into the atmosphere. However, understanding their fate during sewage treatment remains challenging. This study aims to assess aerosolisation of PFAS during WWTP process. Special emphasis is given to new generation and legacy PFAS (e.g., perfluorooctanesulfonic acid (PFOS) and perfluorooctanoic acid (PFOA)) as they are still observed in sewage after years of restrictions. Particulate matter with aerodynamic size ≤10 µm (PM10) collected above a scaled-down activated sludge tank treating domestic sewage for a population >10,000 people in the UK were analysed for a range of short-, medium- and long-chain PFAS. Eight PFAS including perfluorobutanoic acid (PFBA), perfluorobutanesulfonic acid (PFBS), perfluoroheptanoic acid (PFHpA), perfluorohexanesulfonic acid (PFHxS), PFOA, perfluorononanoic acid (PFNA), PFOS and perfluorodecanoic acid (PFDA) were detected in the PM10. The presence of legacy PFOA and PFOS in PM10 samples, despite being restricted for over a decade, raises concerns about their movement through domestic and industrial sewage cycles. The total PFAS concentrations in PM10 were 15.49 pg m-3 and 4.25 pg m-3 during Autumn and Spring campaigns, respectively. PFBA was the most abundant PFAS, suggesting a shift towards short chain PFAS use. Our results suggest that WWT processes such as activated sludge aeration could aerosolise PFAS into PM.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Atmospheric Chemistry and Physics.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.- Preprint
(567 KB) - Metadata XML
-
Supplement
(166 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-3952', Anonymous Referee #1, 25 Jan 2025
In their manuscript, the authors present results from collecting PM10 samples from a scaled-down activated sludge treatment tank and analyzing the samples for several emerging and legacy PFAS. The study was conducted in October 2023 and repeated in March 2024 in the UK, and domestic sludge from a regional WWTP was used. Overall, the manuscript is well written and appropriately structured, and it fills a gap in the literature regarding the study of aerosolization of PFAS during WW treatment. However, there are a few issues that the authors should address.
Major issues:
- From the brief description in the abstract and the slightly more extensive description in the Methods section (L. 107-112), the setup of the scaled down (what is the scale?) in conjunction with the “parent WWTP” is unclear and I strongly suggest adding more detail, including the flowrate of the sludge that is supplied to the smaller tank and how it is connected to the main WWTP.
- In the conclusion of the abstract and also in the Results & Discussion section, the concerns associated with aerosolization of PFAS from WWTP remains vague and should be summarized more explicitly, given the persistence of PFAS in the environment and their ability to travel long distances. As other (industrial) sources become more restricted, the focus likely has to shift to sources like WWTPs. Please add this discussion to the manuscript.
- In conjunction with the previous comment, in L. 83 the authors state “…a knowledge gap exists regarding the atmospheric fate of PFAS…”, which I suggest rephrasing as “…a knowledge gap exists regarding the release of PFAS into the atmosphere…” to better fit the context of this manuscript.
- Please provide the LODs and LOQs for this study in the SI.
- As part of the discussion in L. 179-189 it should be stated that PFAS levels were not measured in the sludge itself, which makes any explanation for the observed trends highly speculative. This fact is touched upon in L. 222-224 and again in L. 246-251, but needs to be mentioned earlier and more explicitly.
- I suggest moving the three sentences from L. 246-251 to L. 181, because the discussion provided in these sentences is important for context at this earlier point.
- 190-200: The authors are discussing different general sources of the detected PFAS, but are not clear about how these PFAS might have been transferred into the sewage sludge. Are the authors aware of any of those industries in their sampling region? What about levels of PFAS in the local drinking water supply, which likely makes up a large portion of the wastewater? Laundry water may also be a source, as PFAS have been detected in dryer lint.
- 225-232: The discussion of the diurnal trends seems incomplete: The only clearly observable trend appears to be that a larger number of PFAS are present above LOD/LOQ during the night compared to the day. Why is that? If a compound is detected during the day, its level is ballpark similar to the corresponding night measurement. PFOA appears to be the only exception with clear spikes on some days. Maybe people do more of their laundry in the evening, run the dishwasher, produce more PFAS containing WW? Please revise this section.
- I recommend adding a Limitations section, which should include at a minimum the following considerations: only one location was sampled during limited time frame, the sewage sludge was not analyzed to investigate the relationship between PFAS concentrations in PM10 and in the sludge, PM10 concentrations were not measured nor was the PM10 further characterized, and neutral PFAS (e.g., FTOHs, FOSEs) were not included in the analysis.
Minor issues:
- I understand that this manuscript has been submitted as a measurement report, but does that categorization have to be part of the title? The "Measurement report" part is missing from the title in the SI, so at least the title should be made consistent.
- Throughout the manuscript, I suggest replacing several instances of “aerodynamic size” with “aerodynamic diameter”, including in the abstract.
- 114: Define “near” – how far from the aeration tank was the sampler installed?
- 118: Define “day” and “night” – what were the time frames? And why was sampling so much shorter during the day compared to the night?
- 128: Regarding the importance of sampling artifacts, especially for PFAS, please consider Chang et al. (2024), “Indoor air concentrations of PM2.5 quartz fiber filter-collected ionic PFAS and emissions to outdoor air: findings from the IPA campaign” (https://doi.org/10.1039/D4EM00359D).
- 131: Delete “slightly”.
- 139: What were the filters prewashed with?
- Tables S3 and S4 mention that the results were blank corrected. Please add a brief description about the process to the Methods section.
- Section 2.5: Were field blanks collected?
- 208-209: The sentence about the pH value is difficult to understand. Please rephrase.
- 218: Period missing after “concentration”.
- 218-219: Rain water may also dilute the wastewater, depending on the PFAS sources.
- 252 and L. 257: It is my understanding that Weinberg et al. measured PFAS in TSP and not “estimated” the concentrations. “Estimated” indicates theoretical/modeling results. Please revise.
- 261: Why the parentheses around “up to 1.31 pg m-3”?
- 307: Replace “could potentially represent” with “likely represent”.
- 320: Replace “filed” with “field”.
- SI: I recommend including the full author list in the SI instead of using “Jishnu Pandamkulangara Kizhakkethil et al.”.
- SI: Please include table and figure captions in the table of content.
- SI: Please list CAS RNs or other unique identifiers in Table S1 together with the PFAS name and abbreviation.
- SI: Please indicate clearly in the captions of Tables S3 and S4 that none of the other targeted PFAS were detected in any sample. If any were detected at least once, I recommend including the concentration in the provided tables.
Citation: https://doi.org/10.5194/egusphere-2024-3952-RC1 -
AC2: 'Reply on RC1', Ivan Kourtchev, 28 Feb 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2024-3952/egusphere-2024-3952-AC2-supplement.pdf
-
RC2: 'Comment on egusphere-2024-3952', Anonymous Referee #2, 28 Jan 2025
In this article, the authors measure the concentrations of 15 legacy and emerging PFAS in PM10 collected from a scaled down activated sludge aeration tank. This study was conducted at two different time points, Oct 2023 and March 2024 using domestic sludge from a wastewater treatment plant in the UK. This manuscript is well written and fills a knowledge gap regarding PFAS aerosolization from domestic sludge wastewater treatment processes. I recommend this paper for publication following revision.
General Comments:
I strongly recommend including both the LOD and LOQ values in the SI. I also recommend reporting the lab, field, and system blanks as well.
In line 90, the authors mention screening from 15 PFAS but then in line 97 state that the EPA 533 PAR mix, which contains 25 PFAS, was used. Please list which PFAS were targeted and provide justification for why some were targeted, and others were not.
Please provide further clarity regarding the scaled-down AS tank as well as the sample collection. How much smaller was the scaled-down tank compared to the large-scale WWTP? How far from the tank and how high above the rim of the tank was the MiniVol placed?
Why is there such a difference in the sampling time for day and night samples? Would the short day sampling periods (1.4 h to 5.7 h) account for the low detection frequency for day samples, especially as compared to the night?
Line 135: Were the filters spiked with IS and then the 5mL of methanol added? Or were the filters sitting in methanol and then, when ready for extraction, the solution of methanol with a filter was spiked with IS?
Section 2.5 – What were the recoveries/extraction efficiencies for the targeted PFAS? Please report these in the SI.
Line 225: This section seems incomplete and highly speculative. PFBA was detected during the day, twice, in the October sampling period and only once was it higher than the night concentration. I don’t know that it can be claimed that the differences are attributable to diurnal variations when the sampling periods for day and night are so different. For PFHpA and PFHxS, the measured concentrations during the night sampling are so low (how close to the LOD/LOQ?) that it might simply be that the sampling time during the day was not long enough to collect sufficient mass to be quantified.
I recommend adding a limitations section or at least paragraph on the limitations of the study. It may be beneficial to include the following: wastewater was not analyzed for PFAS, targeted PFAS was limited to 15 out of thousands, and collection of both gas- and particle-phase PFAS. Studies (see Ao et al., 2024 - 10.1016/j.jhazmat.2023.133018) have also detected polyfluoroalkyl phosphate esters (PAPs) at high concentrations in household dust as well as in food-contact materials, cosmetics, and other consumer products. They’ve also be shown to biodegrade into PFOA (8:2 diPAP) and to other PFCAs (see Lee et al., 2010 - https://doi.org/10.1021/es9028183 and Liu and Liu 2016 - 10.1016/j.envpol.2016.01.069). Additionally, while the authors note that anonymity of the WWTP limits the environmental data they can share, can the authors provide any comment on possible nearby sources of ambient PFAS (e.g., other fluorochemical manufacturing plants or point sources). Wind direction and wind speed data could have also been informative.
The authors may also find it useful to expand their discussion to include the implications of detecting PFAS in PM10 from aerosolized domestic waste. What does this mean for long-range transport and human exposure? As more stringent regulations are placed on emissions of PFAS from major fluorochemical plants, domestic-related emissions are likely to become more important.
Specific Comments:
Line 35: Specify types of cancer associated with exposure to PFAS.
Lines 185 – 187: Does the composition of the contaminated water (particularly the organic content of wastewater) also influence the degree of aerosolization?
Line 201: I think it’s interesting that PFOA was detected in all day samples during the October period but not in the March samples. The authors comment on differences in sewage composition affecting the measured concentrations in PM10, but what about domestic activities that occur in the Fall vs the Spring that may contribute to this? Are the authors able to provide insight into this seasonal difference beyond the sewage differences? I recognize this is probably quite difficult as there are many different sources of PFAS, but this line of thinking could yield interesting theories and questions. However, it may be that the authors simply state (if they agree) that this suggests that there are seasonal variations in household activities that may affect sewage concentrations.
Line 249: Is it possible that PFBA and PFBS, which are volatile, are present in the gas-phase and sorbed to the GFF? Or is this unlikely. Can the authors provide insight into this?
Line 252 and Line 257: Are the values estimated or measured by Weinberg et al. (2011)?
Line 285: Specify the two cities where Lin et al., (2022) and Qiao et al. (2024) sampled. It seems a bit like apples and oranges to specify Ontario, Canada and then all of China.
Line 287 – 290: Is this a useful comparison? Are the two studies comparing the same number and types of PFAS? Perhaps it would be more comparable to sum report the total PFAS concentrations for only the matched PFAS. Also, are the LOD/LOQs for this study and Weinberg et al. (2011) comparable?
Lines 301 – 302 – add in the polyfluoralkylphosphate esters – See Ao et al., 2024 (DOI provided in previous comment).
SI Tables S3 and S4 – list the CAS number for each compound
SI Tables S4 and S5 – I assume the reported SD is the SD for replicate injections? Please state in this tables. I also recommend replacing CSB with the measured value as this is more informative, especially if the authors include the blank concentrations, LOD, and LOQ values in the SI.
Technical Corrections:
Line 43: …”PFAS that are thought to be less…” – change are to were; considering lines 45 – 47 states that studies have shown that replacement PFAS have similar adverse health effects as long chain counterparts.
Line 60: WWTP – define at first use in main text
Line 62: TSP – define at first use
Line 70: change depend to depends -> “the distribution of PFAS depend[s] on the type…”
Line 129: change transition to partition
Line 281: change size to diameter
Citation: https://doi.org/10.5194/egusphere-2024-3952-RC2 -
AC1: 'Reply on RC2', Ivan Kourtchev, 28 Feb 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2024-3952/egusphere-2024-3952-AC1-supplement.pdf
-
AC1: 'Reply on RC2', Ivan Kourtchev, 28 Feb 2025
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 222 | 45 | 11 | 278 | 22 | 9 | 8 |
- HTML: 222
- PDF: 45
- XML: 11
- Total: 278
- Supplement: 22
- BibTeX: 9
- EndNote: 8
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1