My Contribution to the Circular Economy

A contribution to the circular economy; fish excrete nutrients which algae take up and are then fed to Daphnia magna which is then fed back to fish.

A graphical abstract of the project, fish excrete valuable nutrients which microalgae take up.  Daphnia eat the microalgae then are fed to fish.
The basic concept in my paper is that aquaculture effluent (fish farming wastewater) is high in excreted phosphorus and nitrogen which cause eutrophication in lakes and must be managed.  Algae uptake nitrogen and phosphorus into their biomass and Daphnia magna eat algae which can then be fed back to shrimp and fish larvae. 

Project Summary

Recirculated aquaculture systems (RAS) are often seen as the future of aquaculture because of well-documented issues surrounding wild capture or open-net farming.  One engineering problem for RAS are nutrient discharges, which cause water eutrophication (water body senescence or aging and death).  Microalgae are one of the causes of eutrophication, however, microalgae directly incorporate nutrients such as nitrogen and phosphorus as well as carbon dioxide directly into their biomass.  Removing microalgae from water is extremely resource intensive but Daphnia magna eat (harvest) microalgae and are a high value fish feed.  These D. magna can then be fed to shrimp or fish larvae presenting a nice bio-circular economic production system.   This was essentially a lab scale proof of concept for this system.  The scientific details are published in the Science of The Total Environment. In future posts, I plan on going into detail some of the motivations behind this project such as eutrophication and the circular economy as well as on the ramifications of this work. 


Seriously, people are wonderful and I sincerely thank these mentors.
Standing on the shoulders of giants is grandiose for this work but I’m glad to have had tremendous help to slightly push the boundaries of knowledge outwards a little.

I owe many thanks to several people and organizations for this to include Borja Valverde Pérez who has long worked with microalgae for resource recovery from wastewater and served as the principle advisor; Xinyu Zhu who showed me many lab techniques; Marja Koski who allowed me to work on a related project and helped me tremendously with Daphnia magna and aquaculture; Irini Angelidaki who was extraordinarily generous with her resources; as well as Arnaud Dechesne, Anna Łukaszewicz, Karl Gorzelnik, Keya Mukherjee, Caitlin Nyhus, and the good people of Pease New Hampshire.

Musholm A/S was extremely generous in providing the wastewater used in this study.

The research was supported through the InWAP project grant by the Danish Innovation Foundation, Denmark and Department of Biotechnology, Government of India (Grant no. BT/IN/Denmark/61/KM/2018-19).


Disclaimers get tiring.
In the future I really should make an omnibus disclaimer then link to it from every page.

The information and views here are my own and I do not speak on behalf of the EPA in any way shape or form.  None of the information provided is intended to offer engineering, legal, financial, business, information technology, or any other possible professional advice and the information provided may contain errors or omissions.  I will not accept any responsibility or liability for how you use this information.  Use it at your own risk and take all steps necessary to ascertain that this information is correct.

The Draft WHO PFOS and PFOA Guidelines for Drinking Water Quality

WHO logo
The World Health Organization is considering comments for the updated Guidelines for Drinking Water Quality.

The World Health Organization (WHO) is considering adding PFOA and PFOS into its Guidelines for Drinking Water Quality and is seeking public comment on their background document for development of WHO Guidelines for Drinking Water Quality.  This document is undergoing rolling revisions and accepting public comment using the form here sent to by 11 November 2022.  Guidelines for Drinking Water Quality are essentially the WHO’s version of the US EPA’s Health Advisories however, are 5 orders of magnitude higher with 100 ppt individual guidelines for PFOA and PFOS as well as a 500 ppt total PFAS guideline.  This 5 order of magnitude divergence between health agencies is primarily because the US EPA used epidemiological studies to define the reference dose while the WHO primarily relied on toxicology.  Relying on toxicology, the “WHO considered that the uncertainties in identifying the key endpoint applicable to human health following exposure to PFOS and/or PFOA are too significant to derive a HBGV with confidence” (page 79) while the US EPA did not have that issue.  The intent of this article is to go over what the WHO is, why it publishes drinking water quality guidelines, and the comments I submitted.  Please remember that my views are my own and do not reflect my employer’s position!

How the WHO's Guidelines for Drinking Water Quality are organized.
A summary of how the WHO’s Guidelines for Drinking Water are organized.

What is the WHO?

The WHO serves to coordinate and guide public health policy on a global level.  It started as the Office International d’Hygiène Publique (OIHP) or The International Office of Public Hygiene of President Wilson’s failed experiment, the League of Nations.  After the League of Nations folded on 20 April 1946 many of its functions were eventually rolled into the Untied Nations including the WHO. The WHO has been heavily criticized for decades such as by its former leaders and think tanksCriticism stepped up sharply during the ongoing corona virus pandemic. Despite all this, the WHO remains virtually the only true global consensus-building public health organization.  While its influence has weakened in recent years due to other global public health actors such as the World Bank or the Bill and Melinda Gates Foundation, it still is one of the most broadly recognized and supported organizations dedicated to public health.

Why does the WHO publish drinking water quality guidelines?

The WHO itself has been publishing drinking water quality guidelines since 1958 and in fact the WHO’s grandparents were the International Sanitary Conferences (ISC) first held on 23 June 1851 which were established as a result of a cholera pandemic.  The OIHP was initially established to carry out some of the ISC’s recommendations.  Cholera is a waterborne disease; a fact made famous by John Snow and the Broad Street Pump.  By removing a pump handle Dr. Snow helped deal a blow to the miasma theory of disease and served as a founding father to epidemiology; a subject I have a case study in.

John Snow is not only a Game of Thrones character
Wrong John Snow

Outside of this historical context, access to safe drinking water was voted a basic human right by the UN; whether or not a good/service can be considered a right is a discussion for a different day.  

What are the WHO Guideline Values?

The WHO currently recommends individual provisional guidance values of 100 ppt for PFOS and PFOA as well as 500 ppt for total PFAS.  These numbers exactly mirror the EU Drinking Water Directive and are substantially higher than the US EPA recommendations which are 0.004 ppt (or 4 parts per quadrillion) and 0.02 ppt (or 20 parts per quadrillion) for PFOA and PFOS respectively.  A key reason why these values are so dramatically different is that the US EPA considered epidemiological studies while the WHO did not.

The Thinker by Auguste Rodin
The Thinker by Auguste Rodin outside the Musée Rodin in Paris, thanks Wikipedia images.

Throughout the document the WHO uses older thoughts on PFAS and consistently cites the 2016 health advisories.  The US EPA has the most current human health toxicity assessments for PFAS however, the WHO consistently lists them last. 

On page 63 the WHO states (the numbers in the quote are line numbers for that page) “PFOA exposure is associated with clear 23 evidence of carcinogenic activity in male rats based on the increased incidence of liver 24 adenomas and pancreatic acinar cell tumours.” It is therefore not possible using current evidence 25 to exclude PFOA as a human carcinogen”.  This wording reflects the US EPA’s 2016 position; more recent human and animal studies lead the US EPA to conclude that PFOA is a more potent carcinogen than previously thought.

I am not enough of a health scientist to know or provide a meaningful analysis on the health aspects of the document other than the superficial one I just provided pointing out the WHO is using older science. Normally in the debate between epidemiology and toxicology, toxicology points to a more dangerous substance than the epidemiological studies.  Here, due to the preferred analysis by these organizations it appears to be flip flopped.  I would personally prefer to err on the side of caution; especially since drinking water according to the UN is a fundamental human right.

Why I submitted comments to the WHO

Simply put the WHO generally listens if you have a well thought out valid point which conforms to their purpose.  During their last request for comment I (and probably quite a few other people) made quite a few recommendations which were incorporated into this draft document.  For example, I recommended they:

  1. Take into account Xiao’s paper which shows that PFAS can be generated during advanced oxidation processes.  This also changed their practical considerations on monitoring so was quite a substantial change to the document.
  2. Take out the octanol water partitioning coefficient (Kow) from their table of physicochemical properties because PFOA and PFOS are both hydro- and lithophobic so bifurcate into its own layer between octanol and water. Although some people report values, these values are not reflective of the ground truth since the standard test was not designed for that type of material.  They also changed some of the other values such as the boiling point and melting point because the predicted range was larger than the experimental range. 
  3. Specify that anion exchange deals with anionic PFAS since there are cationic and neutral PFAS so there were some changes to those sections as a result.
  4. Drop the “s” from the PFAS acronym for plurals because a substance cannot simultaneously be per- and polyfluorinated so the plural is already implied.
  5. Use the OECD definition of PFAS which they updated.
  6. Use updates I provided on page 71 about characterizing EPA methods 533 and 537.1

One of my comments was almost incorporated word for word on page 74 related to chain length and functional group removal efficacy; a conclusion I partially took from this Sörengård et al. paper.

Previous comments which were not accepted

Not all my previous comments were accepted.  This goes to reflect that two people can look at the same data and draw different conclusions. Especially when the topic is a matter of policy there is often no “correct” answer although there are often clear cut “incorrect” answers. I am going to use this section to highlight some of the comments I made which were rejected, why I made those comments, and potentially the reason why the comment was rejected.  

  • I recommended against using percent removal as a characterization of process efficiency. You can think of properly designed granular activated carbon or anion exchange beds as towels wiping up liquid (contaminant).  When the towel is dry and the spill is small it picks up all of it quickly.  A wet towel mopping up a large spill may not clean everything or may not clean it quickly. Try and keep that analogy in mind for a contactor.  The photo below shows one mode for thinking of breakthrough.  Better ways to characterize the efficiency are to use media usage rate or bed life.  I believe that the WHO used percent removal because it makes it easier to compare to membrane filtration although again, I disagree with this characterization.
A schema of how breakthrough occurs in soprtive processes
This diagram shows one way of thinking about breakthrough although it is one of two main conceptual processes currently in the vouge.
  • I provided references which show greater removal percentages for sorption that the WHO quotes. There are many however, the US EPA drinking water treatability database captures many such examples.  The WHO is also much more pessimistic on anion exchange than I believe is warranted based on the data.  There are both full scale, pilot, and bench studies included. I do not really know why the WHO was cautious there compared to their membrane section where results were more often accepted at face value except potentially that was due to the declining efficacy over bed life which again, is why I suggested percent removal is a bad way to measure sorption process efficiency.
  • The WHO focuses on source water monitoring.  In general, this is a good approach however, I do not agree with it for PFAS.  PFAS is not known to and is unlikely to leach or permeate into drinking water supply.   Likewise, PFAS can be generated during treatment when advanced oxidation (and potentially advanced reduction) processes are used.  If resources are constrained it makes more sense to me to monitor the finished water rather than source water if only one is to be monitored especially since source waters are unlikely to change.  The counter argument is that only specific wells could be contaminated, and it would make sense to shut down those wells and not have to treat.  If you only monitor finished water, you will miss that level of detail.  If resources are constrained and you only monitor the influent, you’ll miss that anyway.  To me, this is a good example of two reasonable policy positions based on the same facts with divergent conclusions. 
  • In the previous document, the WHO referenced replacement chemistries; specifically short chained PFAS.  I recommended that esterified replacement chemistries be mentioned as well such as HFPO-DA (also known as GenX chemicals), ADONA, and Nafion byproducts.  Instead, the WHO removed references to replacement chemistries.  This is a good compromise because there are many replacement chemistries available.

My comments on the current draft

On this new draft the comments I have submitted to the WHO using their form emailing to (11 November 2022 deadline) are:

  1. The recommendation on page 70 that PFAS monitoring is only required when there is a reasonable source nearby is not protective of public health or in the best interests of society at large.  As this document points out PFAS is capable of long-range transport and deposition.  PFAS has been found in the artic and Antarctic as well as other remote areas.  Additionally, airports; military sites; landfills; wastewater treatment plants; carpet, upholstery, wax, and paper product manufacturers; or spreading biosolids (composting) are all PFAS sources.  Rasstadt, Germany was contaminated with composting from used food containers for example.  I do not believe that you can successfully perform a risk assessment given the diverse set of sources, low toxicity thresholds, and long-range transport capabilities.
  2. Page 72 lines 8-9 state (line numbers in parathesis): “Blending or diluting PFAS contaminated source water with uncontaminated water may be a (9) cost-effective and viable option for some water systems.”  As there is ultra-low toxicity and PFOA as well as PFOS have been identified as likely carcinogens I do not believe that deliberately exposing people to these contaminants through diluting or blending is an appropriate course of action unless there are no other alternatives such as treatment (this would be an economics question) or source substitution.
  3. Page 76 provides information on PFAS removal through sorptive technologies.  I believe, and have provided references above in this article, that significantly lower treatment standards are possible.
  4. On page 16 lines 27-30 it states (line numbers in quotes): “In 2019, the US EPA conducted a broad literature search to evaluate (28) evidence for pathways of human exposure to PFOA and PFOS, and in 2021 released a draft (29) analysis that supports application of a 20 percent relative source contribution for PFOA and (30) PFOS in drinking water (US EPA, 2021b,c).” This seems to overstate the strength of the conclusion the EPA made in those health advisories.  As I mention in this article, the EPA’s position was that there was insufficient evidence to deviate from a 20% default value; not that evidence supported a particular value.
  5. On page 4 line 37-39 it states (line number in parentheses): “PFOA remains possible, even where it is no longer (38) manufactured or used due to its legacy uses, degradation of precursors, and extremely high (39) persistence in the environment and the human body.”  I believe this is also true for PFOS and that disclaimer should be added to that section as well.
  6. Page 74 lines 45-46 deal with anion exchange resin regeneration.  Greater than 95% resin regeneration has been demonstrated in literature here, here, and here although it may not be practical for utilities from a safety and/or cost prospective.
  7. On page 9, the WHO references this Boone et al study; when the study was published the conclusion that only one monitored drinking water plan exceeded the EPA’s HAs was valid however, the HAs were updated.  That changes the study conclusion from one drinking water plan exceeding the EPA’s HAs to all quantitatively monitored plants exceeding the HAs, a significantly different conclusion.

Conclusion and Key Take Aways

Over time, scientific consensus points to greater and greater awareness of the dangers PFAS pose.  Generally, we thought these chemicals were safer in the past than most scientists do today.  The WHO, in its role as a policy leader, has developed a significantly less stringent health recommendation than even the old 2016 US EPA HAs on the dangers of these chemicals by not weighting epidemiological evidence the same way.  The WHO standard exactly mirrors the EU standard under the drinking water directive.  The WHO is a reasonable organization and solicits public comment and feedback as well as generally listening to it.  I highly recommend you share your opinions with the WHO regardless of whether you agree with me and take an active role in global public health.

Further Reading

This is one of several articles on PFAS I have written.  You can read about PFAS discovery, structure, or all my articles on PFAS.

PFAS Analytic Strategies

Stylized analysis of organic compounds using liquid chromatography
Image from Rocío L. Pérez and Graciela M. Escandar in “Experimental and chemometric strategies for the development of Green Analytical Chemistry (GAC) spectroscopic methods for the determination of organic pollutants in natural waters” showing mass spectroscopy. 


Appropriate analytical methods are needed for toxicological and epidemiological studies as well as to enforce any restrictions. As alluded to in my article on PFAS structure, there is a not single definition for PFAS.  In fact from this 19 July 2022 federal register notice on the 5th Contaminant Candidate List and this proposed rule under section 8(a)(7) of the Toxic Substances Control Act it becomes clear that the EPA lacks a single PFAS definition within itself.  The EPA definitions are also different (and narrower) than the new July 2021 OECD report “Reconciling Terminology of the Universe of Per- and Polyfluoroalkyl Substances.” Many PFAS definitions are intentionally purpose written making one analytical method impossible to achieve. Additionally, there are many different PFAS types; the most commonly known ones are the anionic PFAS but there are also cationic and zwitterionic (both positive and negative functional groups). Given these two difficulties an immediate question should be how we measure these compounds.  This article intends to cover the broad strategies taken to measure PFAS and some of the methods within each as well as the methods’ strengths and weaknesses.  It focuses on anionic non-polymeric PFAS.  It may be helpful to review the article on PFAS nomenclature.


In general, there are three broad strategies for PFAS analysis:

1. Target analysis

2. Total fluorine (organic fluorine) methods

3. Non-target analysis including suspect screening and the total oxidizable precursor (TOP) assay (TOPA)

PFAS analysis is severely complicated by the prevalence of fluorochemicals in laboratory equipment particularly PTFE.  For example, PFAS is widely prevalent in HPLC solvent inlet tubing and PFAS from felt tip permanent markers used to mark containers during sampling campaigns.

Target analysis summary

Target analysis targeted methods typically involve either liquid or gas chromatography (LC or GC) coupled with mass spectrometry (MS) and require a reference standard for quantification. Standards are only available for select PFAS and only these specific substances can be quantified. Target analysis yields accurate concentrations but does not provide the total PFAS load.

Total fluorine methods summary

Fluorine is a sum of its inorganic and organic constituents; this insight forms the basis for quantification attempts by total fluorine methods.  As PFAS are organic molecules, inorganic fluorine is typically removed from the analysis and absorbable or extractable organic fluorine is measured. Due to differing chemical properties among PFAS, the extraction method chosen will not be suitable for all PFAS classes. Absorbable organic fluorine relies on choosing a suitable absorbent so suffers from the same difficulties.

Non-target analysis summary

Non-target analysis uses chromatography coupled to a high-resolution mass spectrometer and can identify previously unknown PFAS. These methods tend to rely on quadrupole time-of-flight and linear ion trap-orbitraps which are not common instruments. If reference standards are available, similar detection limits to targeted analysis can be achieved. The TOPA deserves special mention as this method performs targeted analysis then attempts to oxidize all precursor compounds into compound endpoints which have standards then performs targeted analysis again to see the difference and theoretically quantify the precursor load but not the specific precursors. These methods suffer from the high technical capacity and equipment required to employ them.

Targeted Analysis

Targeted analysis is currently the only form of analysis which can support drinking water regulatory action because it is able to quantify extremely low PFAS levels; generally around 4 parts per trillion (ppt). In general, LC-MS/MS or GC-MS/MS is used. Reference standards are only available for select PFAS so only select PFAS may be quantified. Which PFAS have standards is driven both by the ability to synthesize extremely high purity samples, regulatory emphasis, and economic viability for the producers. The main PFAS which have standards are per- and polyfluorinated acids (PFAA) – particularly the carboxylic (PFCA) and sulfonic acids (PFSA) as well as the etherified versions of these.  As a quick reminder PFCA end in “O” such as PFOA and PFSA end in “S” such as PFOS.  The etherified versions tend to have a less standardized nomenclature structure such as “GenX chemicals” or ADONA.

Targeted analysis is well established for various matrices.  For example, USEPA Methods 537.1 and 533 together cover 29 PFAS (537.1 focuses mainly on long chain PFAS and covers 18 while 533 was developed mainly to cover some more short chained PFAS and covers 25 PFAS; there are 14 PFAS which overlap between the standards).  The German Institute for Standardization (DIN) has also published standard methods DIN 38407-42 and 38414-14.  DINs are not freely available like the American methods so I will not link to a store here however, the DINs are broadly similar to the American methods.  All these methods use solid phase extraction (SPE) followed by liquid chromatography tandem mass spectrometry (LC-MS/MS). As these methods rely on solid phase extraction, the method efficacy declines with chain length.  In other words, the quantification and detection limits for short chained PFAS are higher than they are for long chained PFAS.

As mentioned in the nomenclature article, PFAS may come in “linear” or “branched” forms. These targeted methods may detect branched forms however, branched standards are not widely available and ignoring these will result in a low bias.  Linear to branched isomer ratios can also help to identify the PFAS sources. The USEPA Methods 537.1 and 533 include technical notes for dealing with branched isomers.

The DINs note that these methods can be extended to other PFAA with appropriate standard availability however, volatile fluorotelomers (like fluorotelomer alcohols) cannot be determined using these methods.

Comparison of which compounds can be detected by what EPA method
Comparison of EPA methods for PFAS determination in water; EPA 1633 (aqueous samples), EPA 537.1 (drinking water), and EPA 533 (drinking water). As you will notice the methods are broadly similar and will not work for volatile PFAS.

Total fluorine methods

Total fluorine (TF) is the sum of Inorganic Fluorine (IF) and Organic Fluorine (OF) sometimes called Total Organic Fluorine (TOF). TOF includes fluorine content of all PFAS and their precursors in the sample. TOF methods cannot however separate out organic fluorine contained in medications or other common uses from PFAS. There have been exceedingly few studies quantifying TOF fractions from PFAS compared to medications or other uses of organofluorine. Additionally, TOF methods have much higher quantification limits than targeted analysis, typically around 100 ppt. These two limitations make TOF methods suitable for screening but not protective enough of public health to seriously be used in drinking water regulation.  Remember the interim health advisories (HAs) for the apical effects of PFOA and PFOS were 0.004 and 0.02 ppt respectively, so 100 ppt is 4 or 5 orders of magnitude larger than the level at which no apical effects are expected. Additionally, organic fluorine methods only measure the fluorine while the HAs measure the full molecular weight. For example, PFOA (C8HF15O2) has a molar mass of 414.07 g/mol but the mass of fluorine only weighs approximately 285 g/mol PFOA meaning that only about 69% of each PFOA molecule would be quantified assuming perfect extraction and quantification.  The exact ratio of fluorine to the overall molecular weight is species dependent.

Organofluorine analysis requires the organic fluorine to be extracted – Extractable Organic Fluorine (EOF).  There will always be some organofluorine which cannot be extracted and some which can be extracted but cannot be quantified.  Extraction can be the same as for targeted methods. However, due to the large variability between the PFAS chemical composition, the extraction method will influence the test results. Likewise, similar to targeted methods, there is worse extraction efficiency for shorter chained PFAS and specific substances such as trifluoroacetic acid. A special EOF method is known as absorbable organic fluorine (AOF). During AOF, the sample is typically absorbed to a mixed-mode weak anion exchange SPE; again similar to targeted analysis methods.

The most common total fluorine methods are: combustion ion chromatography (CIC), particle-induced gamma-ray emission spectroscopy (PIGE), instrumental neutron activation analysis (INAA), X-ray photoelectron spectroscopy (XPS), and 19F nuclear magnetic resonance spectroscopy (19F NMR).  Other methods are under development including: inductively coupled plasma mass-spectrometry (ICP-MS), continuum source molecular absorption spectrometry (CS-MAS), defluorination with sodium biphenyl (SBP), potentiometric and fluorimetric detection, and reversed phase LC-UV or GC coupled with a flame ionization detector (FID), electron capture detector (ECD), or MS.  CIC is currently the most discussed TF method.

Non-targeted Analysis

Non-target analysis for PFAS typically uses chromatography coupled to high-resolution mass spectrometers to identify previously unknown PFAS. These methods tend to rely on quadrupole time-of-flight and linear ion trap-orbitraps observed accurate masses, isotopic fingerprints, and MS/MS fragments. Accurate masses, isotope patterns, and molecular feature fragmentation patterns obtained from high resolution mass spectroscopy (HRMS) are compared to databases with known PFAS, such as the USEPA CompTox Chemistry Dashboard, NORMAN Suspect List Exchange, MetFrag, and STOFF-IDENT. Measured chemical responses are compared across two or more sample groups in a process known as fold-change comparison.

General workflow for non-target analysis
Image from Hollender et al. in High resolution mass spectrometry-based non-target screening can support regulatory environmental monitoring and chemicals management. This shows the general workflow for non-target analysis.

Total Oxidizable Precursor Assay

The Total Oxidizable Precursor (TOP) Assay (TOPA) was developed by Houtz and Sedlak in 2012 to quantify concentrations of difficult-to-measure and unidentified precursors of PFCA and PFSA in urban runoff. Essentially a sample is measured following conventional targeted analysis methods and then samples are exposed to hydroxyl radicals generated by persulfate thermolysis under basic pH conditions and measured following conventional targeted analysis methods again. Comparing pre- and post-oxidation concentrations the PFAA precursor load can be estimated. Unfortunately, oxidation conversion yields are compound dependent and many factors affect the oxidation process. Oxidation completeness can be tracked by adding a 13C mass labelled precursor however, this again is dependent on the specific compound chosen.  Additionally, the endpoint compounds selected for targeted analysis affect the results dramatically.  Activated persulfate does not generally oxidize PFSA so this can only really provide inference on PFCA precursors.  The TOP assay is also labor intensive, requiring twice the number of samples and more than twice the amount of work to produce a single result. The TOP assay does provide an indication of PFAS precursor load without requiring as advanced equipment or training as non-targeted analysis.

TOPA summary
Image from CDM Smith which summarizes the TOPA; essentially targeted analysis is run, precursors are oxidized as much as feasible and targeted analysis is re-run. An important point is that activated persulfate does not readily oxidize PFSA so it only really provides inference on PFCA precursors.


Each strategy selected for PFAS analysis carries with it its own unique strengths and drawbacks. Under most current toxicological frameworks employed by regulatory agencies targeted analysis is the only real viable option however, this can miss many PFAS. Total fluorine methods are several orders of magnitude too imprecise to accurately protect public health but capture most PFAS.  Non-targeted analysis methods require expensive equipment, much experience, as well as being non-quantified.

Further Information

This article is one of several on PFAS. You can read about PFAS discovery, structure, nomenclature, or my other PFAS articles.

PFAS Terminology and Nomenclature

Common PFAS terms
A word cloud containing some common PFAS terms. Made using Wordit.

Per- and polyfluoroalkyl substances (PFAS) are a ubiquitous emerging public health threat. PFAS terminology can be extremely confusing because it looks like an alphabet soup of similar acronyms. This article is the second of two on PFAS terminology. The first intended to provide an overview of PFAS structure (including what counts as PFAS), this one intends to help demystify PFAS terminology and nomenclature. These are part of a broader series on PFAS including their serendipitous discovery.


PFAS are a large chemical class generally produced in industrial processes which do not produce pure substances; often impurities are introduced to the end product. An isomer is a compound with the same chemical composition as another compound but a different chemical structure.  In the PFAS world the terms “branched” and “linear” refer to differing isomers. As a short hand, people often mean all isomers of a particular compound when they name it. PFAS can also be cyclic in which case the linear/branched nomenclature breaks down.

Linear and branched non-polymeric PFAS are typically composed of two parts: a hydrophobic as well as lipophobic fluorinated carbon backbone and a typically polar hydrophilic functional group.  The suffix philic means an affinity for and phobic means fear or aversion, the prefix hydro means water and lipo is for lipid or fat. 

Structural analogues are compounds similar to other compounds but differing in one aspect. Structural analogues are also known as congeners. An example could be homologues. PFAS with the same functional groups differing in the number of carbons and fluorine atoms in the backbone are known as homologue series. For example, perfluoroalkyl carboxylic acids with 4 fully saturated carbons in the backbone to 12 fully saturated carbons in the backbone would form a homologue series.  A homologue group includes all linear and branched carboxylic acids with the same number of carbons in the backbone. For example, the C8 carboxylic homologue group includes linear perfluoroalkyloctanoic acid (PFOA), isopropyl-PFOA, and 3-methyl-PFOA; there are 89 total theoretically possible C8 carboxylic homologue congeners although in addition to the linear form. The picture below shows some C8 carboxylic homologue congeners, while technically separate compounds for ease they are often referred to as a singular entity.

Examples of branched PFOA
Image courtesy of Battelle Memorial Institute particularly Craig Hutchings and Steve Helgen. While technically different compounds these are often considered together.

Naming Systems

There are 5 main systematic methods currently in use to name fluorinated compounds: the International Union for Pure and Applied Chemistry (IUPAC) nomenclature, the perfluoro nomenclature, code numbers, Chemical Abstract Service (CAS) nomenclature, and F-nomenclature systems. Due to strong similarities, fluorinated organic compounds tend to follow hydrocarbon naming conventions.

Basic Conventions

Many PFAS are acids and may exist as a protonated or anionic form as well as a mixture of both depending on the pH. The pKa values tend to be quite small and scientists generally refer to PFAS with acid functionalities as “acids” rather than carboxylates, sulfonates, or the correct terms even though the dissociated forms may be the only relevant forms. Scientists refer to both the protonated and ionized forms by the same acronyms.  Most perfluorinated acids are environmentally anions.  Some labs report results in the acidic form, some report anions, and some a mixture.  Many times, perfluorinated carboxylic acids are reported as acids (for example perfluorooctanoic acid instead of perfluorooctanoate) and perfluoroalkyl sulfonic acids as anions (for example perfluorooctane sulfonate instead of perfluorooctane sulfonic acid).  This has caused confusion.  The labs are really measuring the concentration of perfluorinated acid anions present; when the lab reports an acid the lab adjusts for the cation. Most sulfonate standards are prepared from salts and the mass adjusted which is why they are often reported that way.  This is especially true for PFHxS, PFOS, ADONA, 9Cl-PF3ONS and 11Cl-PF3OUdS. Section 7.2.3 of EPA Method 537.1 details how to convert to adjust for this.

IUPAC Nomenclature

IUPAC Nomenclature systematically establishes unambiguous, unique, uniform, and consistent compound names based on their chemical structure.  IUPAC Nomenclature can be used for all organic compounds and is not specific to PFAS.  IUPAC Nomenclature is quite literally a book containing many precedents and steps however, the general simplified steps for IUPAC Nomenclature are:

  1. Identify the longest carbon chain.  The longest carbon chain is also known as the “parent chain.”
  2. Identify any groups off the parent chain.
  3. Number the carbons in the parent chain so that the groups off the parent chain have the lowest number.
  4. Number and name the groups just counted.
  5. Put all the name components together in alphabetical order; prefixes such as di- are not used for alphabetical order.

For example, C2F4 is tetrafluoroethene, shown below, is the simplest perfluorinated alkene (fully saturated carbon backbone with no functional groups).

Picture from Wikidata. There are two carbons and a double bond so it is named “ethene.” Tetrafluoro comes from the 4 fluorine atoms attached to the carbons.

Perfluoro Nomenclature

Perfluoro Nomenclature only applied to perfluoroalkyl substances.  It is commonly used to quickly identify fully fluorine saturated carbon compounds. For shorter substances using IUPAC naming conventions is still easy and understandable. Most chemists would immediately understand that hexafluorobenzene (C6F6 – shown below) has a cyclic fully saturated carbon backbone. However, for longer compounds such as pentadecafluorooctanoic acid it is difficult to readily tell which is the motivation for Perfluoro Nomenclature.  The proper IUPAC name for that compound is actually: 2,2,3,3,4,4,5,5,6,6,7,7,8,8,8-pentadecafluorooctanoic acid which is fairly cumbersome to use further motivating Perfluoro Nomenclature.  The pentadecafluoro refers to the 15 fluorine atoms, octanoic acid refers to a carboxylic acid functional group that is off the 8th carbon in a chain.  This compound is pictured under hexafluorobenzene.

Picture from Wikimedia. Hexafluoro denotes the 6 fluorine atoms coming off the carbons in the benzine ring. By Buck et al’s or one of the OECD’s PFAS definitions this compound would not be included however, it would be under Glüge et al’s definition. The preferred IUPAC name for this compound is 1,2,3,4,5,6-hexafluorobenzene.
Pentadecafluorooctanoic acid
Pentadecafluorooctanoic acid courtesy of PubChem.

Perfluoro Nomenclature uses the prefix perfluoro along with standard hydrocarbon nomenclature. This simplified naming system has been in place since the 1950s and is the most common naming system among practitioners. The acronyms developed from this system follow a simple pattern; they all start with the prefix PF for perfluoro then have a root based on the carbon backbone, and a suffix denoting the functional group.  The table below gives the roots for C1-C14 and the two most common functional group abbreviations are A for carboxylic acids as well as S for sulfonic acids.

Carbon backbone length naming roots
Table of roots for specific carbon backbone lengths.
Carboxylic and sulfonic functional groups
The carboxylic acid (left) and sulfonic acid (right) functional groups. The R represents some organic chain.

The acronym itself provides a well-defined structure for the PFAS compound.  For instance perfluorobutanesulfonic acid or PFBS is shown below.  The front letters PF stand for per-fluoroalkyl, the B stands for but-, -ane shows there are only single bonds in the carbon backbone which is implied, and the S is for sulfonic acid.  PFBA, PFHpA, and PFDA are also shown.

Perfluorobutanesulfonic acid or PFBS
Perfluorobutanesulfonic acid or PFBS. The sulfonic acid functional group is circled in dashed yellow, the carbons are in red, numbered so that the first carbon is nearest the functional group. By convention, the unlabeled connectors are carbon atoms. Carbon always forms 4 total bonds, if 4 bonds are not shown the carbon by general agreement is bonded to a hydrogen.
Examples of PFAS acronyms
The table shows the appropriate root color coded, carboxylic acid groups are circled in dashed blue and sulfonic acid groups are circled in dashed yellow. PFBS is perfluorobutanesulfonic acid, PFHpA is perfluoroheptanoic acid, and PFDA is perfluorodecanoic acid. PFBA is perfluorobutanoic acid however, butanoic is sometimes replaced by butyric acid in older works.

Should it be necessary to refer to a specific salt for a perfluorinated compound it is customary to write the salt’s common abbreviation in front of the standard perfluorinated abbreviation then drop the acid from the name.  For instance, the ammonium salt of PFOA, ammonium perfluorooctanoate, is often abbreviated APFO. While not strictly included in the perfluorinated system, some polyfluoroalkyl compounds can degrade into perfluorinated compounds. A polyfluoroalkyl compound is one where the carbon backbone is not fully fluorinated. An important subset of polyfluoroalkyl substances are fluorotelomers. Fluorotelomers are substances made from the most common industrial process for perfluorinated compounds, telomerization. In telomerization a transfer agent, known as a telogen, reacts with a polymerizable taxogen (monomer) to produce a telomer.  Generally in commercial production, perfluoroethyl (pentafluoroethyl) iodide as a telogen reacts with tetrafluoroethylene oligomers as taxogens, most commonly tetrafluoroethylene, to produce telomer perfluoroalkyl iodide polymers.  The perfluoroalkyl iodide polymers are then converted into other substances.

Picture from the Big Chemical Encyclopedia showing pentafluoroethyl iodide reacting with tetrafluoroethylene oligomers to produce perfluoroalkyl iodide polymers.

Similar to perfluorinated acids, fluorotelomer substances are typically named for their functional group.  Fluorotelomers often are abbreviated FT followed by their functional group.  There are often numbers preceding the abbreviation such as 8:2. The first number represents the total perfluorinated carbon atoms and the second number the unsaturated carbons atoms.

8:2 FTOH
By convention the unlabeled connectors are carbon atoms. Carbon atoms always have 4 bounds and as this field was developed from hydrocarbon chemistry the hydrogens are typically not shown unless part of a functional group. The carbon atom labeled “2” in green for instance has two hydrogens attached to it. An alcohol group is a hydroxyl connected to an aliphatic carbon.

Code numbers

Systematic names are fantastic in precision but can be difficult to use.  Unless you work with compounds frequently, memorizing the prefix structures and extra rules, which were glossed over in this article can represent an annoying time commitment.  Additionally, it can be error prone especially to infrequent or new users.  Industrial PFAS manufacturers such as ICI Fluoropolymers, ISC, DuPont, and 3M developed inhouse shorthands for PFAS.  As the inhouse systems were not meant for the general public this initially caused confusion. In Organofluorine Chemistry dichlorodifluoromethane (CF2Cl2) is used as an example to illustrate this. Dichlorodifluoromethane was known as Freon-12 by DuPont, Arcton-6 by ICI, and Isceon-122 by ISC. By 1957 a standardized code number system based on DuPont’s system was adopted and formalized as The American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) Standard 34 which was last updated in 2019. DuPont’s system was first developed by Albert Henne, Thomas Midgley, and Robert McNary in 1929.


Dichlorodifluoromethane (CF2Cl2), better known as Freon-12, was formerly a popular chlorofluorocarbon halomethane refrigerant now banned under the Montreal Protocol. It is now only allowed as a fire suppressant in submarines and aircraft. Picture from the Gas Encyclopedia by Air Liquide.

In the ASHRAE system for pure compounds, a letter designates the primary purpose of the compound, R for refrigerant, P for aerosol propellant, S for solvent, and the numbers are based on the chemical formula.  Sometimes the letter is replaced with a  trade name such as Freon.  For blends, numbers are assigned sequentially based on ASHRAE data review completion date. For an ASHRAE designation, the first digit from the right is the number of fluorine atoms.  The second digit from the right is the number of hydrogen atoms plus one.  The third digit is the number of carbon atoms minus one; if this is a zero then that number would be omitted. The fourth digit is the number of unsaturated carbon-carbon bonds, in saturated compounds this number is omitted. If an upper-case B or I appears the number immediately following it is the number of bromine or iodine atoms respectively. There are often lowercase letters following the numeric designations.  The first appended letter is for substitutions on the central carbon atom, an x represents a chlorine substitution, y for fluorine, and z for hydrogen. The second appended letter designates substitutions on the terminal methylene carbon, =CCl2 is -a, =CClF is -b, =CF2 is -c, =CHCl is -d, =CHF is -e, and =CH2 is -f. Isomers are also designated by a Z or an E appended to the name for “zusammen” or “entgegen” respectively. Zusammen is German for together and entgegen is opposite; they represent cis and trans isomers respectively. 

Cyclic compounds (R300 series), ethers, inorganic fluids (R700 series), and miscellaneous organic compounds (R600 series) have additional rules.  The series are specified so the 000 series are methane based compounds, 100 are ethane based, 200 are propane based, and 300 are cyclic. The 400 series are zeotropes and 500 are azeotropes. The 600 series are other organic compounds. The 700 series are inorganic refrigerants such as ammonia, also known as R-717. The 1000 series which are unsaturated organic compounds such as the new R-1234yf refrigerant. R-1234yf is used for about half of all automobiles built after 2018 and it breaks down into short chained perfluorinated carboxylic acids; specifically, trifluoroacetic acid.

2,3,3,3-Tetrafluoropropene also known as HFO-1234yf and R-1234yf.  It has a greenhouse gas (GHG) warming potential less than that of carbon dioxide. R-134a, which R-1234yf  helped replace, has a GHG warming potential of 1,430 as a comparison. Picture from Wikipedia.
The compound on the left is R-1234ye(E) and the compound on the right is R-1234ye(Z).
The compound on the left is R-1234ye(E) and the compound on the right is R-1234ye(Z).  The part which makes the compound an isomer is circled in dotted blue lines. R-1234 has other possibilities and chemical structures as well which can be further specified by the letters following the numbers. As an example R-1234yc is CH2FCF=CF2.

There are other code systems in use that are more field dependent as well.  For instance, chlorofluorocarbons have a numbering system as do halogenated fire extinguishers.  The Halon numbering system is in comparison very easy; there are always four numbers and left to right the digits represent carbon, fluorine, chlorine, and bromine atoms present in the molecule.  For instance, Halon-1211 (CF2ClBr also known as Freon-12B1) and Halon-1301 (CF3Br which would also be known as R-13B1).  On 1 January 1994 the United States banned the import and production of halons 1211 and 1301 under the Clean Air Act to comply with the Montreal Protocol. These halons are still allowed to be used and recycled halons may be purchased.  Both Halon 1211 and 1301 have Class A, B, and C ratings. Halon-1211 is a “streaming agent,” and more commonly used in hand-held extinguishers because it discharges mostly as a liquid stream.  Halon-1301 is a “flooding agent” and discharges mostly as a gas, allowing it to penetrate tight spaces and behind obstacles making it ideal for enclosed spaces commonly found in aircraft.  The United States owns about 40% of global halon-1301 supply and it will likely stay in use for many years worldwide.  Halon-1301 is critical as it was formerly the only material usable in enclosed spaces where humans were present. 

CAS nomenclature

CAS stands for Chemical Abstract Service (CAS) and it is a division of the American Chemical Society. A CAS number is a unique identifier assigned by CAS to every chemical substance. It includes isotopes, mixtures, alloys, and substances which have unknown or variable composition. CAS numbers do not contain information about the substance structure unlike all previously discussed naming systems, simplified molecular-input line-entry system (SMILES), or the IUPAC International Chemical Identifier (InChI) system.

The same PFAS chemical in different forms will have differing chemical properties because of structural differences.  For example, PFOS (perfluorooctanesulfonic acid) in its acid form (C8HF17O3S) has a CAS number of 1763-23-1, its potassium salt (C8F17KO3S) 2795-39-3, and its ammonium salt (C8H4F17NO3S) 29081-56-9.


F-nomenclature was accepted by the American Chemical Society as an alternative to perfluoro nomenclature because some early fluorochemistry pioneers disliked perfluoro nomenclature. JH Simons in particular disliked perfluoro nomenclature. Simons invented the electrochemical fluoridization processes which was the first industrial process for producing perfluorinated compounds.

In F-nomenclature perfluoropentanoic acid (PFPeA – a C5 fully saturated carboxylic acid) would be F-pentanoic acid. As an aside, C5H10O2 which is pentanoic acid, is also known as valeric acid because it is the cause of the unpleasant smell in valerian.

Picture of valerian by Ivar Leidus.  The unpleasant smell these flowers give off is caused by pentaonic acid.

Some elements of this naming system have survived.  For instance, the symbol -R which indicates an organic backbone sometimes is written -RF to indicate a perfluoro backbone. Also, placing an F internal to a benzene ring structure is a shorthand saying that all the bonds off the ring not otherwise indicated are fluorine instead of hydrogen.  A forerunner to F-nomenclature used a phi (φ) but was generally similar to F-nomenclature. 

Other less common nomenclature systems

In the 1970s, J.H. Simons, an early fluorochemistry pioneer, advocated for placing “for” before the final syllable in standard hydrocarbon nomenclature to convey it was perfluorinated. Organofluorine Chemistry gives the following examples: CF4 as methforane and the perfluorinated version of ethene, CF2=CF2 would be ethforene. 

Conclusion and Key Take Aways

Most PFAS naming systems are designed to reveal information about the basic chemical structure or function and to facilitate communication.  In industrial settings where not many PFAS are used simultaneously it can be easier to use codes to cut down on potential errors common in systematic names.  It is important to understand the basic names within your specific use.

This article is one of several on PFAS. You can read about PFAS discovery, structure, or all my articles on PFAS.


PFAS Structure

Common PFAS uses
This image shows some common uses of PFAS. Image courtesy of Carollo Engineers; particularly Cayla Cook and Eva Steinle-Darling.

Per- and polyfluoroalkyl substances (PFAS) are a ubiquitous emerging public health threat. PFAS terminology can be extremely confusing because it looks like an alphabet soup of similar acronyms. This article is the first of two on PFAS terminology. This one intends to provide an overview of PFAS structure, the second one intends to help demystify PFAS terminology and nomenclature. My previous article on PFAS was on Dr. Roy Plunkett’s discovery of PTFE, a tetrafluoroethylene fluoropolymer.


The EPA’s comptox database lists over 9,252 PFAS; this is too large a class to name on individual bases. A systematic naming convention has evolved so that much information is conveyed with just the name. There are 5 main ways to name fluorinated compounds: the perfluoro nomenclature, F-nomenclature, code numbers, the International Union for Pure and Applied Chemistry (IUPAC) nomenclature, and Chemical Abstract Service (CAS) nomenclature systems. Additionally, fluorine chemistry as a field has existed long before those five systems were created and naming was not really standardized until Buck et al tried to harmonize conventions in 2011. The idea of this article is to provide an overview of common PFAS groupings, the second article will go into the five naming conventions.


PFAS is a general nonspecific name encompassing a group of substances.  There is no clear agreed upon definition of what exactly counts as PFAS.  The Organization of Economically Developed Countries (OECD), the United States Environmental Protection Agency (EPA), and the World Health Organization (WHO) have all used various definitions.  In the paper meant to harmonize conventions, Buck defined PFAS as “highly fluorinated aliphatic substances that contain 1 or more carbon atoms on which all the hydrogen substituents (present in the nonfluorinated analogues from which they are notionally derived) have been replaced by F atoms, in such a manner that they contain the perfluoroalkyl moiety CnF2n+1–.” This definition was the first clear PFAS definition. A moiety means part of a molecule. The OECD/United Nations Environment Program (UNEP) pointed out that Buck’s definition left out many substances commonly recognized as PFAS such as perfluorodicarboxylic acids and recommended the –CnF2n– moiety instead, although there is still debate surrounding this. Buck’s definition was updated by Glüge et al and informed the OECD/UNEP definition.  Glüge’s definition expands Buck’s to cover perfluorocarbon chains with functional groups on both ends, aromatic substances with perfluoroalkyl moieties on side chains, and fluorinated cycloaliphatic substances.

Historical Note

As mentioned in the PFAS discovery article, PFAS were initially investigated to replace sulfur dioxide, methyl chloride, ammonia, and other extremely dangerous early refrigerants. This may jog your memory; the first major public PFAS concerns were mainly focused on perfluorocarbons under the 1987 Montreal and 1997 Kyoto Protocols.  The Montreal Protocol aimed to limit chlorofluorocarbons.  The Kyoto Protocol aimed to limit 7 greenhouse gas (GHG) categories and stated that climate change is real and driven by anthropogenic emissions. The GHG categories included perfluorocarbons (PFCs) and hydrofluorocarbons (HFCs).

Kyoto Protocol countries by agreement status
Photo courtesy of Wikimedia Commons Green: Annex B parties with binding targets in the second period, Purple: Annex B parties with binding targets in the first period but not the second, Blue: non-Annex B parties without binding targets, Yellow: Annex B parties with binding targets in the first period but which withdrew from the Protocol, Orange: Signatories to the Protocol that have not ratified, Red: Other UN member states and observers that are not party to the Protocol.

Unfortunately, early 2000 environmental scientists adopted the PFC acronym for PFAS instead of the perfluorocarbons identified in the Kyoto Protocol. There is still quite some confusion caused by this. While PFC is used in turn of the millennium PFAS literature it should be avoided now.

General Organofluorine Concepts

Carbon (C) bonded to fluorine (F) is the class basis for PFAS.  The C-F bond is extremely persistent and the key factor for the problems PFAS causes as well as its desirable technical properties. Carbon can form up to 4 bonds with other atoms.  The first broad distinction is between polymers and non-polymers. A polymer is a molecule made up of repeating subunits called monomers. Polymers often serve as the feedstock to produce other PFAS chemicals, the most widely recognized fluoropolymer is polytetrafluoroethylene (PTFE) which Dr. Roy Plunkett discovered on 6 April 1938.

Structure of PTFE repeating monomer
Picture courtesy of Wikipedia Commons. This depicts PTFE. The brackets show the monomer tetrafluoroethylene, the n symbolizes that this is a repeating structure.

Functional groups often serve as the basis for classification systems. A functional group is a moiety made of a specific structure which causes characteristic chemical reactions which are similar no matter what the rest of the molecular composition is. PFAS chemicals can also exist in multiple states such as acids, anions, cations, and salts which have important implications for their physical and chemical properties. Anionic forms are the environmentally prevalent form and the general form most authors refer to. Some legislation, particularly European Registration, Evaluation, Authorisation, and Restriction of Chemicals (REACH), defines substance to include impurities and stabilizers of the main PFAS compound.

Broad non-polymeric structure

Non-polymeric PFAS include perfluorinated and polyfluorinated substances.  When carbon is bound only to fluorine except for functional groups it is called perfluorinated and the compound is saturated with F. When some carbon bonds include things other than functional groups or fluorine it is called polyfluorinated and is considered unsaturated with respect to F; hydrogen is the most common other bond. As PFAS stands for “per- and polyfluoroalkyl substances” and no single compound can be both unsaturated and saturated with respect to fluorine simultaneously “PFAS” is by default a plural acronym.  Both per- and poly-fluoroalkyl substances are subdivided into classes where the functional groups behave relatively similarly.  Two large classes are fluoroalkyl acids and fluoroalkane sulfonamides. These large classes are then subdivided again based on functional groups/moieties.  For example, perfluoroalkyl acids are further subdivided into carboxylic, sulfonic, sulfinic, phosphoric, phosphinic, and other acid types.  There are about 425 common moiety-based subdivisions within PFAS.

PFAS categorizations
Picture courtesy of The Interstate Technology Regulatory Council. This photo goes over some broad PFAS categories; while only 5 moiety based groups are shown here there are currently over 425 acknowledged nonpolymer moiety classification groups.

In 2017, Wang et al published an iconic chart covering the number of papers published on various PFAS groups reproduced below.

Number of peer reviewed articles between 2002 and 2016 for various PFAS.
Picture from Wang et al. This shows some broad helpful PFAS classifications. PFCA stands for perfluorinated carboxylic acids, PFSA for perfluorinated sulfonic acids, PFPA for perfluorinated phosphoric acids, PFPiA for perfluorinated phosphinic acids, PFECA for perfluoropolyether carboxylic acids, PFESA for perfluoropolyether sulfonic acids, and PASF for perfluoroalkane sulfonyl fluorides. Examples of groups left off this classification include perfluoroalkyl iodides (PFAI) and many others. The articles mentioned cover 2002 to 2016.

Conclusion and Key Take Aways

There is no clear definition on what PFAS encompass, even within regulatory organizations. There are about 9,252 PFAS substances which are broadly divided into polymeric and non-polymeric substances.  Non-polymeric PFAS subclasses are based on carbon chain saturation with fluorine into full saturated (perfluoroalkyl) or non-fully saturated (polyfluoroalkyl) groups. Those groups are subdivided based on moieties/functional groups so that compounds which behave similarly are classified together. An upcoming article will cover the five common classification systems for individual compounds.

Other articles in the series include PFAS Discovery

Further Reading on PFAS classifications

Covid-19 Wastewater Update

Wastewater and public health potentials
Photo from Farkas, K., Hillary, L. S., Malham, S. K., McDonald, J. E., & Jones, D. L. (2020). Wastewater and public health: the potential of wastewater surveillance for monitoring COVID-19. Current Opinion in Environmental Science & Health.

Since my previous article on wastewater based epidemiology (WBE) for SARS-Cov-2 monitoring, there have been rapid developments. According to the World Health Organization’s 12 September 2020 update there have been over 28 million confirmed cases and 900,000 deaths worldwide making it a serious global pandemic. For comparison, last year about 1.7 million people acquired AIDS and 700,000 died. There is growing evidence that built environmental systems, particularly ventilation systems and residential plumbing systems, contribute to SARS-Cov-2 spread.

SARS-Cov-2 in the Gastrointestinal Track

The pooled SARS-Cov-2 viral RNA prevalence in stool samples from clinically confirmed cases is only estimated to be around 50% although estimates range from 15% to 84% in this meta-study and review. These studies unfortunately generally did not have many participants; between 9 and 4,243, with most studies having under 60 participants. Likewise, SARS-Cov-2 loads and viral RNA in fecal samples reported between 1,000 and 10,000,000 SARS-Cov-2 copies per fecal milliliter; one study had 153 participants, where only 44 participants (29%) had viral RNA present, while the other studies all had under 50 total participants.  That study indicated that there was, broadly speaking, a traceable general shedding pattern. During the initial SARS-Cov outbreak in 2002-2003 and MERS-Cov outbreak in 2012, viral RNA was still present in stool samples over 30 days after the illness. Similarly, patients with SARS-Cov-2 in their stool continued to shed RNA viral positive fecal samples after showing negative respiratory/nasopharyngeal samples. The estimated continued positive shedding duration and percentage still shedding varied greatly but reported means vary between 11 days and 5 weeks in 20% to greater than 70% of patients that had positive stool samples. There is limited evidence to suggest that viral RNA in stool comes from live infectious viruses instead of deactivated or destroyed viruses however, testing for the live virus is difficult to do and few people try. Most studies suggest that SARS-Cov-2 in urine is rare however, some studies report its presence past negative throat swabs.

Funny image
This guy has about a 50% chance of containing SARS-Cov-2 when excreted from an infected individual

SARS-Cov-2 in potable water distribution

It is extremely unlikely that SARS-Cov-2 can remain viable in potable water systems, especially in the US where 0.2 mg/L chlorine residual minimum must be at temporally farthest tap. While I could not find information on SARS-Cov-2’s survival in chlorinated water, other human coronaviruses are highly susceptible to chlorination. Likewise, I could not find information on SARS-Cov-2’s survival in non-chlorinated tap water which dominates Europe however, other human coronaviruses showed a three log removal (99.9% removal) at 23°C (73.4°F) in 10 days; at 4°C (39.2°F) human coronaviruses do not show a three log removal after greater 100 days. These results are not particularly helpful.  Cold inlet tap water’s temperature is normally 10-15.5°C (50-60°F) but can vary from 3.6-“jacuzzi temperature” 39°C (38.6- ≈100°F) in the United States (low value is Anchorage, Alaska and the high value is Death Valley, California). The temperature depends on several factors: water age, water source (surface or ground), season, processed water storage, pipe depth, and ambient air temperature. In aggregate however, I cannot derive a scenario where SARS-Cov-2 would proliferate enough in potable water systems to make someone sick through showering for instance.

SARS-Cov-2 in sewers

Similar to potable water, I was unable to find information specific to SARS-Cov-2 however, information on other human coronaviruses is available. Other human coronaviruses die rapidly in wastewater with three log removal (99.9%) occurring between 2 and 4 days for all temperatures. I do not believe there is a general standard time for sewage to reach treatment plants however, most sewers are designed with a self-cleaning velocity that should be reached daily (between 0.6 m/s and 1 m/s mainly dependent on specific gravity and pipe diameter) and are generally capped at 3 m/s during max flow to prevent erosion. Rochester, NY takes about 24 hours for sewage to reach its treatment facilities which is normal and a decent average proxy. All reputable sources agree that standard wastewater treatment processes, which are designed for virus and bacteria inactivation among other things, inactivate SARS-Cov-2. Likewise, dilution occurs in sewers which should increase the minimum infective dose by lowering the virus’ concentration.

SARS-Cov-2 in residential plumbing

Sewers, unlike potable water, are not generally pressurized and are ventilated to eliminate smells. This little distinction is critical.  Circumstantial evidence reported in the Annals of Internal Medicine indicated that 9 people became sick with SARS-Cov-2 from fecal aerosols. This is not the first time that a respiratory disease has been tied to sewage waste vents. The 2003 SARS outbreak at Amoy Gardens in Hong Kong was implicated in 321 cases and 43 deaths. During China’s ultra-strict lockdown, Kang complied camera footage indicating no contact between the sick apartment members and the newly infected group who lived on different floors. Among more than 200 air and surface samples collected, the only ones testing positive for SARS-CoV-2 came from the 15th floor family’s apartment and a vacant apartment’s bathroom on the 16th floor directly above. Tracer gas piped into the 15th floor apartment’s drainpipe exited in the 25th and 27th floor apartment bathrooms. Generally, there is a plumbing “trap” (shaped like a U or P) that has water in it to block smells from rising. These however, can dry out leaving a transmission route for disease. Drying out can occur from non-use or air pressure surges. The ethane tracer gas presence indicates that these traps dried out. Contact tracing and other standard causal patterns did not reveal leads. One team member on Kang’s study indicated that there could also be three other outbreak incidents related to waste vent gases. However, while compelling, there is no iron clad evidence and it is possible the disease was contracted elsewhere. Mechanical bathroom exhaust fans and outdoor air conditions can lead to a favorable environment for SARS-Cov-2 to spread through bathroom exhaust. There should be appropriate caution reading these findings. Many factors must fall into place for this kind of residential transmission. For instance, the proposed transmission route relies on viral infectivity in fecal droplets and aerosols. However, building wastewater systems are a potential reservoir for many other viruses and bacteria, even in the absence of SARS-CoV-2.

SARS-Cov-2 in toilets

Virus-containing fecal aerosols can be produced during toilet flushing after index patient use. These bioaerosols can settle onto surfaces and remain infective. There was a case where a South Korean woman most likely contracted Covid-19 from an airplane toilet. She self-quarantined in complete isolation for three weeks before the flight, did not use public transport to get to the airport, wore an N-95 mask for the entire flight except a visit to the bathroom, all passengers sat two meters (six feet) from each other during boarding, and quarantined for two weeks by South Korean officials on landing. The one asymptomatic sick passenger on the plane used the toilet before her. The most likely transmission route was encountering contaminate surfaces because the airplane used high-efficiency particulate arresting systems. According to Dr. Joseph Allen from Harvard’s T.H. Chan School of Public Health, about 1,000,000 additional particles per air cubic meter are generated when a toilet is flushed with the lid up. These particles can settle on surfaces or linger in the air until someone breaths them in.

Protecting yourself

There are some easy common-sense protective measures you can take to protect yourself. Ensure bathrooms you use are well ventilated, turn on an exhaust fan when entering a bathroom and leave it on when you leave. Make sure the P or U trap isn’t dried out; a bad smell indicates a dry trap. Close the lid when flushing the toilet to help prevent bioaerosols from spreading. Clean and disinfect bathroom surfaces. Most importantly, wash your hands when leaving the bathroom, then try and use a paper towel to touch surfaces including the door handle on your way out.

Potential WBE Advances

To date SARS-Cov-2 Wastewater Based Epidemiology (WBE) relies on the same analytical platforms used in clinical diagnostic testing (eg PCR or antigen testing). WBE does not need to be limited to the monitoring the infectious agent’s nucleic acid or antigens. WBE could target endogenous biomarkers that are significantly elevated in diseased states. This could reduce analytical costs and broaden availability (through immunoassays) or better serving as leading infection indicators (earlier alerts). Urine (as opposed to fecal) biomarkers would also simplify sampling and sample preparation. Since Covid-19 can cause extensive inflammatory damage, biomarker for systemic oxidative stress such as the prostaglandin-like class of substances called isoprostanes are currently being proposed. These biomarkers may be more universally excreted among infected individuals, better track the infection severity, have tighter per-capita excretion ranges (allowing for better case count calibration and estimation), and avoiding a potential under-appreciated problem with using PCR, where RNA fragments may not be originating from viable virus, but rather from virus remnants (litter) from cleared infections. That last issue could overestimate infection incidence or intensity. It is also speculated that patient repeat infection reports are caused by this.

WBE could also be used to test hypotheses involving correlating various community-wide population demographics with the magnitude and duration of SARS-CoV-2 measurements to probe inter-community disparities such as race, culture, income, healthcare availability, and occupation. WBE data could also be examined for correlations with drug manufacturer geographic prescribing data — notably for drugs suspected to improve or exacerbate Covid-19 therapeutic outcomes. WBE could also determine which SARS-CoV-2 subtypes dominate in given populations.

WBE Other Shortcomings

In addition to the difficulties I outlined in my first article on WBE, I have learned about some additional difficulties. Population size estimations are difficult because populations fluctuate due to travel and commuters. The standard approach to this is to measure certain endogenous biomarkers such as cortisol or cotinine then calculate those as daily loads normalized to population sizes. However, some unique population fluctuations have negligible catchment impacts leading to higher uncertainties in smaller populations. Other standard population estimating wastewater parameters used such as Chemical Oxygen Demand, Biochemical Oxygen Demand, or ammonia can reduce uncertainties but can be strongly influenced by the wastewater’s composition. Another is that biomarkers must be relatively stable not only in the sewer system but also through the sampling and storage processes.

Another shortcoming is wastewater itself makes it extremely difficult to extract and quantify biomarkers and chemicals. PCR inhibitors include fats and proteins, as well as humic and fulvic acids. New digital PCR techniques use Poisson distributions, via partitioning samples into reaction wells to lessen these effects.

Previously Unmentioned Successes

WBE can distinguish differences between prescription and consumption of a pharmaceutical. Investigating parent compounds to metabolites ratios or ratios between compound enantiomers in wastewater can distinguish human excretion from direct pharmaceutical disposal in sewers. This distinction ability is important because prescriptions do not necessarily correlate to use. Delayed prescribing is a strategy where doctors prescriptions available but ask patients to delay using them to see if symptoms improve. These initiative successfully reduced antibiotic use in New Zealand, Norway and England; WBE can distinguish how many antibiotics were actually used as opposed to prescribed.

WBE can minimize the tests required to uncover positive cases. Clinical tests need to continually increase test coverage. The ratio between tests required to uncover a single case and total tests is generally the most direct infection extent indicator. A low ratio (when using random sampling) points to a high incidence of infection and therefore the need for more intensive testing until the ratio significantly increases (where increasing testing amounts are required to confirm additional cases). This indicates increasing success in containment or mitigation measures. However, diagnostic tests are never intended for mass surveillance. The tests are generally time-consuming and costly as well as exposing the test administrator. There are two alternatives: increase conventional testing or minimize the tests required to reveal positive cases. Pooled testing procedures increases testing capacity and throughput, especially for PCRs. Pre-targeting subpopulations can help with minimizing the rests required as well. These methods conserve diagnostic tests. Using WBE then can be akin to using a forward observer to improve artillery’s accuracy. This would greatly reduce the demand for diagnostic testing and reduce supply-chain shortages caused by insufficient manufacturing capacity. The metric of success for WBE when used for targeting the use of clinical diagnostic testing would be lower ratios for “Tests Administered” per “Case Confirmed” (counter intuitively, maximize the positivity test rate).

WBE may also be the only way to infer the uninfected population as well as provide perspective on how well diagnostic testing reflects the total population.

Corrections to Previous Article

In my previous article, I mentioned that WBE started around 2001. In the 1980s, Finland, Israel, and Senegal all successfully analyzed sewage samples to assess circulating polio.


You can probably catch Covid-19 from public toilets and in star-crossed circumstances from your neighbor’s toilet. WBE research is developing but remains much more difficult than analyzing for chemicals such as illegal drugs because there are differences in viral shedding patterns, total shedding, viral attenuation during sewer travel, and determining statistically representative sampling. Even in other applications, matrix separations pose difficulties for WBE.  WBE is still an effective epidemiology tool to rapidly monitor disease spread and trends, especially when paired with other contemporary measures. The preponderance of evidence suggests that CoVs are less stable in the environment than other enteric viruses. Water recycling guidelines may have to be revised in light of emergent diseases and viral shedding into sewer systems. Effective surveillance systems are key for the rapid intervention and infectious disease control. WBE is the most effective and cheap near real-time tool available to communities.

Further Reading

  • IWA’s Information resources on water and COVID-19
  • Chan, K. H., Poon, L. L., Cheng, V. C. C., Guan, Y., Hung, I. F. N., Kong, J., … & Peiris, J. S. M. (2004). Detection of SARS coronavirus in patients with suspected SARS. Emerging infectious diseases10(2), 294.
  • Cha, S., & Smith, J. (2020). Explainer: South Korean findings suggest ‘reinfected’ coronavirus cases are false positives. Reuters.
  • Cheung, K. S., Hung, I. F., Chan, P. P., Lung, K. C., Tso, E., Liu, R., … & Yip, C. C. (2020). Gastrointestinal manifestations of SARS-CoV-2 infection and virus load in fecal samples from the Hong Kong cohort and systematic review and meta-analysis. Gastroenterology.
  • Foladori, P., Cutrupi, F., Segata, N., Manara, S., Pinto, F., Malpei, F., … & La Rosa, G. (2020). SARS-CoV-2 from faeces to wastewater treatment: What do we know? A review. Science of the Total Environment743, 140444.
  • Gundy, P. M., Gerba, C. P., & Pepper, I. L. (2009). Survival of coronaviruses in water and wastewater. Food and Environmental Virology1(1), 10.
  • Heller, L., Mota, C. R., & Greco, D. B. (2020). COVID-19 faecal-oral transmission: Are we asking the right questions?. Science of The Total Environment, 138919.
  • Hovi, T., Shulman, L. M., Van Der Avoort, H., Deshpande, J., Roivainen, M., & De Gourville, E. M. (2012). Role of environmental poliovirus surveillance in global polio eradication and beyond. Epidemiology & Infection140(1), 1-13.
  • Kaiser, Jocelyn (2020) Can you catch COVID-19 from your neighbor’s toilet? Science Magazine
  • O’Brien, J. W., Choi, P. M., Li, J., Thai, P. K., Jiang, G., Tscharke, B. J., … & Thomas, K. V. (2019). Evaluating the stability of three oxidative stress biomarkers under sewer conditions and potential impact for use in wastewater-based epidemiology. Water research, 166, 115068.
  • Petrie, B., Youdan, J., Barden, R., & Kasprzyk-Hordern, B. (2016). New framework to diagnose the direct disposal of prescribed drugs in wastewater–a case study of the antidepressant fluoxetine. Environmental Science & Technology, 50(7), 3781-3789.
  • Wolfel, R., Corman, V. M., Guggemos, W., Seilmaier, M., Zange, S., Müller, M. A., … & Hoelscher, M. (2020). Virological assessment of hospitalized cases of coronavirus disease 2019. Nature. https://doi. org/10.1038/s41586-020-2196-x.
  • Wu, Y., Guo, C., Tang, L., Hong, Z., Zhou, J., Dong, X., … & Kuang, L. (2020). Prolonged presence of SARS-CoV-2 viral RNA in faecal samples. The lancet Gastroenterology & hepatology5(5), 434-435.

Perchlorate Regulation

A thorough look at the EPA’s perchlorate final action

99.3% pure potassium perchlorate
99.3% pure potassium perchlorate containing a small amount of cabosil to minimize clumping and keep the mixture free-flowing. This is typical of fireworks; courtesy of

The EPA recently declined to regulate perchlorate under the Safe Drinking Water Act (SDWA) and promulgated this press release. The Federal Register Final Action Notice is here (RIN:2040-AF28). In general, this decision has been heavily criticized; for instance, a New York Times’ headline read “E.P.A. Won’t Regulate Toxic Compound Linked to Fetal Brain Damage.” The ranking Senate Environment and Public Works Committee member, Senator Tom Carper (D-DE) even said “ [the] EPA has abdicated its responsibility to set federal drinking water standards for a chemical long known to be unsafe, instead leaving it up to states to decide whether or not to protect people from it.” It begs the question why would an agency charged to protect human and environmental health decline to regulate a known hazard? Perhaps more surprisingly, the nation’s water safety and utility advocate, the American Water Works Association (AWWA) agreed with the EPA’s ruling. This article will examine what perchlorate is, what it means to be regulated under the SDWA, perchlorate’s regulatory history, and what the EPA, as well as advocates and critics had to say about the recent ruling. As this is slightly longer than a typical article each section is meant to be relatively self-contained.

What is Perchlorate?

Perchlorates are chemical compounds containing ClO4. The most common commercial perchlorates are ammonium perchlorate (NH4ClO4), perchloric acid (HClO4), potassium perchlorate (KClO4), sodium perchlorate (NaClO4), and lithium perchlorate (LiClO4).  The perchlorate anion may also be bound to other alkali or alkaline earth metals. Perchloric Acid (HClO4) is stronger than sulfuric and nitric acids and is the most common precursor for other perchlorates. Perchlorates can form naturally in the atmosphere leading to trace amounts in precipitation; especially in west Texas, New Mexico, and Northern Chile in the Atacama Desert. Perchlorate is often used as solid-state rocket fuel and is also used in fireworks, flares, gunpowder, and explosives. Surprisingly, given its highly reactive nature, perchlorate can persist in the environment for years. By far the most manufactured perchlorate is ammonium perchlorate for its use in aerospace and defense; it is manufactured more by mass than all other perchlorates combined.

Perchlorate’s uses

The space shuttle’s booster rocket was about 70% ammonium perchlorate and the rest was powdered aluminum or elastomeric binders. Ammonium perchlorate was also responsible for the PEPCON disaster on 4 May 1988 which caused 2 fatalities, 372 injuries, and about $100 million 1988 dollars damage ($218 million in 2020 dollars). The disaster’s damage radius was approximately 10 miles (16 km) and equivalent to about one TNT kiloton or the same yield as a small tactical nuclear weapon.  The blast was caught on video by Dennis Todd and has been used in many TV shows. It can be viewed here.

Perchlorates’ use in rocket fuel extends to military applications. Most submarine launched intercontinental ballistic missiles use ammonium perchlorate boosters; because of this many countries, including the US, consider exact perchlorate production, import, and export figures confidential or a state secret. However in 2008, the Department of Defense did publish an estimated use between 6 and 8 million perchlorate pounds annually.

Perchlorate is also used in temporary adhesives, electrolysis baths, batteries, air bag ignitors, matches, desiccants (drying agents), etching agents, electropolishing, ion-exchange chromatography, oxygen candles (used on submarines and in spacecraft), cleaning agents, and oxygen generating systems. Much like the rat poison Warfarin, perchlorates are sometimes used in medical applications. Perchlorate is used to treat overactive thyroids and to counter the drug amiodarone’s side effects. Additionally, perchlorate is used to block radioactive technetium uptake during medical imaging of the brain, blood, and placenta.

Perchlorate in Agriculture

Chilean nitrate fertilizer contained high natural perchlorate levels (0.12-0.26% by weight) and perchlorate is found where this was used; especially on Long Island, New York. Studies have also uncovered perchlorates in synthetic fertilizers in concentrations from 1,800 μg/g to 4,200 μg/g. Likewise, perchlorates have been used in weed killers and leguminous plant growth promoters. Perchlorates have also been used to increase poultry and other farm animal’s weights although there is no evidence perchlorates increase human weight.

Perchlorate, Mars, and Space Travel

Magnesium perchlorate (Mg(ClO4)2) is manufactured in smaller amounts than other perchlorates however, it has been found on Mars in large quantities (>0.6% by weight in soil). Magnesium perchlorate forms the basis for several theories that liquid water can be found on Mars at or below the surface because it has a high solubility (99.3 g/100 mL) and freezing point depression/boiling point elevation falls into colligative properties. A colligative property depends on the quantity of solutes in a solvent rather than the solute’s specific chemical nature. Magnesium perchlorate’s presence on Mars has also led to theories that Mars could be a natural rocket refueling station. Unfortunately, 0.6% by weight in soil is equivalent to 6×106 ppb and would be exceedingly toxic to human settlement.

Where is perchlorate found?

Perchlorates were identified in about 5% of community water systems in the US based on 1997 and 1998 American Water Works Association led studies performed on the EPA’s behalf. A community water system is a water system that serves at least 25 people or 15 service connections; there are about 52,110 community water systems in the United States as of 7 April 2014 according to the CDC. The studies were performed before Standard Method 314.0 was adopted by the EPA and indicated a need for analytical refinements. The Environmental Working Group, a nonprofit environmental activist group that is often labeled as alarmist, suggests that perchlorate was detected in 375 water utilities serving 12.1 million Americans in its Tap Water Database. During EPA mandated tests in 2001 and 2003 under the 1st Unregulated Contaminant Monitoring Rule (UCMR1) perchlorate was found in drinking water systems serving 16.6 million Americans.

Perchlorates were found in at least 60 of the 1,335 sites on the National Priority List (NPL) as of 7 August 2020. There is a good chance the number of NPL sites perchlorate is found at will increase as more sites are evaluated; in 2006 there were only 49 NPL sites with perchlorate remediation out of 1,581 total NPL sites.

As perchlorate is highly soluble, relatively stable, and highly mobile in water and has a low vapor pressure, perchlorates do not volatilize from water or soil to air, leach readily into groundwater, and travel large distances from initial contamination. Perchlorate is only weakly absorbed so its movement through soil is generally un-retarded. These two properties indicate that perchlorate travels rapidly and far in water. For instance, the Olin Flare Facility in Morgan Hill California created a plum which stretched more than 10 miles (16 km). You can read more about the Olin Flare Facility here.

In addition to the UCMR1 studies, Arizona, California, and Texas each completed their own state-based studies summarized in the table below. Texas’ study is particularly interesting because no credible anthropogenic perchlorate source could explain the contamination’s scale or level leading Texas Tech researchers to propose natural contamination.

State based supplements to UCMR1 summay

The AWWA, EPA, and ATSDR (Agency for Toxic Substances and Disease Registry) all agree that perchlorate is not widely dispersed to the environment by rocket combustion. When it is used as a fuel perchlorate is destroyed in the combustion process. Most environmental releases come from anthropogenic releases (manufacturing accidents), fireworks, fertilizer application, and natural formation.

Perchlorate has been detected in most states, the District of Columbia, Puerto Rico, and the Mariana Islands. A contamination map from this 2010 Government Accountability Office (GAO) report is given below.

Perchlorate contamination map from this 2010 GAO report

Surprisingly, potassium perchlorate is also an FDA approved additive in food container rubber gaskets, not to exceed 1% under 21 CFR 177.1210(b)(5). There are perchlorate contamination indications in food. Leafy plants such as tobacco and lettuce and highly vascular plants such as tomatoes have been shown to uptake perchlorate in significant concentrations (upto 164.6 mg/kg dry weight in green flue-cured tobacco). Perchlorates are also found in household bleach that is stored for long times.  

Perchlorates health effects

Perchlorates mainly exhibit thyroid toxicity in humans. Perchlorates also partially inhibit thyroidal iodine uptake leading to one medical application for them. Perchlorates are conjectured to lead to hypothyroidism and goiter although this remains unproven. While unproven, perchlorates were used to treat hyperthyroidism for many years leading credence to the suspicion. Side effects from perchlorate treatment included skin rashes, nausea, and vomiting. Some patients developed severe red-blood cell shortages leading to death (fatal aplastic anemia). It is speculated that perchlorate effect depends on gender, exposure length, and dietary iodine consumption.

Perchlorate hurts the thyroid by inhibiting iodine’s transport into thyroid follicle cells by competitive binding to sodium/iodine symporter which catalyzes Na+ and I‘s transfer. Perchlorate’s inhibition limits iodine levels needed to produce thyroxine (T4) and triiodothyronine (T3). Perchlorate is considered an endocrine disrupting compound because it effects T4 and T3. While perchlorate’s effects are reversible, development problems caused by inadequate circulating hormones are not; leading to conjectured child brain developmental effects and the New York Times’ headline at this article’s start.

Perchlorate was shown to significantly increase systolic blood pressure (the top number in blood pressure readings) which is the maximal arterial pressure when the heart beats. Differences in systolic and diastolic (the bottom number in blood pressure readings or the arterial pressure between beats) are known as pulse pressures. A normal pulse pressure is about 40 mm Hg or less; pulse pressures 60 or greater are considered risk factors for cardiovascular disease, stroke, and heart attacks.

There is some indication that perchlorate may have pulmonary toxicity such as inflammatory infiltrates, alveolar collapse, subpleural thickening, and lymphocyte proliferation.

Perchlorate’s Carcinogenicity

Perchlorates are not considered to be carcinogens by the Department of Health and Human Services or the International Agency for Research on Cancer. It is also conjectured that perchlorate may cause changes in children’s brain development.

The Safe Drinking Water Act (SDWA)

The SDWA, Public Health Services Act Title XIV, is the major federal law protecting drinking water and was first enacted in 1974 with major revisions in 1986 and 1996. The original 1974 act established a system where states, once vetted, are responsible for the SDWA’s implementation and enforcement; this is known as primacy. Primacy may also be granted to Indian Tribes, Territories, and the District of Columbia. In cases where primacy has not been granted to a state, tribe, territory, or district the EPA retains primacy. The EPA has primacy in every Indian tribe except the Navajo Nation, in Wyoming, and the District of Columbia. In every case where primacy was requested it has been granted. To request primacy local regulations at least as stringent as national requirements must be adopted, adequate enforcement procedures must be developed (including monitoring and inspections), and administrative penalty authority must be adopted. Primacy agencies must also conduct water system inventories, maintain records and compliance data, and make reports as EPA may require as well as developing safe water provision emergency plans. Although the SDWA is the major federal law, other laws including the Clean Water Act and various Water Infrastructure Improvement Acts also modify drinking water protection. The SDWA applies to the approximately 152,700 American water systems which supply water to at least 25 people or 15 service connections.

Monitoring Requirements under the SDWA

Primacy agencies may follow the Standardized Monitoring Framework (SMF) or the Alternative Monitoring Guidelines. The SMF was meant to simplify, standardize, and consolidate drinking water monitoring requirements. The SMF was established 30 January 1991 and included 9-year fixed compliance cycles each subset into three periods. The Alternative Monitoring Guidelines allow primacy agencies to grant utilities: monitoring waivers, surrogate sampling and reduced nitrate monitoring.

Under the SMF community water systems serving >10,000 would collect 4 quarterly perchlorate samples during the second compliance period of the fourth compliance cycle (January 1, 2023 through December 31, 2025). Community water systems serving 10,000 or fewer people and non-transient noncommunity water systems would collect 4 quarterly samples during the third compliance period of the fourth compliance cycle (January 1, 2026 through December 31, 2028). The EPA estimated that 60% of surface water systems and 10% of groundwater systems would be ineligible for waivers to reduce monitoring requirements.

SDWA Amendments

The first major SDWA amendments in 1986 were mainly designed to speed the EPA’s contaminant regulation and groundwater protection. The 1986 amendments were extremely ambitious and attempted to make up for lost time since the original act. The 1996 amendments recognized the 1986 amendments were too aspirational and slowed down the required contaminant regulation pace as well as overhauling the entire regulation process. The 1996 amendments gave us the current risk-based regulatory approach used today. Additionally, the 1996 amendments created state-based operator certification programs.

Chemical Regulation under the SDWA

The 1996 SDWA amendments established three requirements for a chemical to be regulated under the Safe Drinking Water Act (SDWA):

  1. The substance must have an adverse health effect
  2. It must occur or have a substantial likelihood to occur at adverse public health frequencies and levels
  3. The EPA’s Administrator must judge regulation to present a meaningful opportunity for health risk reduction

Once the EPA Administrator determines to regulate a contaminant the EPA must propose a rule within 24 months and promulgate a National Primary Drinking Water Regulation (NPDWR) within 18 months after proposal. New regulations generally become effective three years after promulgation and up to two additional years may be provided if capital improvements are needed. Systems serving 3,300 or fewer people may be given an additional 9 years beyond the compliance deadline.  The EPA must review and strengthen as appropriate each drinking water regulation every six years. When developing regulations, EPA is required to use the best available, peer reviewed science, supporting studies and data as well as make publicly available a risk assessment document that discusses estimated risks, uncertainties, and studies used in the assessment.

The Unregulated Contaminant Monitoring Rule (UCMR)

The 1996 SDWA amendments also created the Unregulated Contaminant Monitoring Rules. Every 5 years the EPA must publish an unregulated contaminant candidate list (CCL) for chemicals which are known or anticipated to both occur and cause deleterious health effects in public water systems. Additionally, every 5 years the EPA must publish a 30 contaminant or less list that requires a monitoring program by public water systems known as the Unregulated Contaminant Monitoring Rule (UCMR). CCLs and UCMRs are required on a quinquennium basis and are given sequential numbers to refer to the different lists; for example the Third Unregulated Contaminant Monitoring Rule (UCMR3) was published on 2 May 2012 and UCMR4 was published on 20 December 2016. Every 5 years, the EPA is required to make a regulator determination (whether to regulate) for at least 5 of the 30 chemicals under the CCL. Generally, UCMR monitoring is from all public water systems serving more than 10,000 people and 800 representative public water systems serving 10,000 or fewer people. Perchlorate was the first drinking water contaminant that EPA has proposed to regulate in nearly 24 years under the provisions of the Safe Drinking Water Act Amendments of 1996.

Maximum Contaminant Levels (MCL) and a Maximum Contaminant Level Goals (MCLG)

A maximum contaminant level (MCL) is a legally enforceable regulation under the Safe Drinking Water Act. A maximum contaminant level goal (MCLG) is an aspirational non-enforceable public health objective rather than a regulatory standard. For non-carcinogens the MCLG is based on the reference dose. The reference dose essentially is a conservative estimate of the daily allowable contaminant consumption a person can have without an expected adverse health effect during a typical lifetime. As the US adheres to a linear no threshold radiation policy, the MCLG for carcinogens is always zero. MCLs and MCLGs often coincide however, they do not always. Since MCLs are legally enforceable they may not match MCLGs because of difficulties in contaminant measurement, lacking treatment technologies, or treatment cost outweighing public health benefits. MCLs must be set as close as “feasible” using best available technology, treatment techniques, or other means (considering costs) to the MCLG. The EPA may ignore the feasible level if the feasible level could lead to an increase in health risk by increasing other contaminant’s concentration or interfering with treatment processes used to comply with other SDWA requirements. In such cases the overall health risk must be minimized.

For contaminants the EPA wants to regulate but there is no economical or technically viable threshold (MCL), the EPA creates “treatment technique rules” such as the Lead and Copper Rule which are enforceable procedures to minimize risk.

SDWA Cost Benefit Analysis

Criteria three for regulating a contaminant under the SDWA in the 1996 amendments, the EPA Administrator’s judgement to present meaningful health risk reduction, comes with it a required cost-benefit analysis. EPA must publish a “health risk reduction and cost analysis.” For each drinking water standard and each alternative standard being considered, EPA must publish and take comments on quantifiable and nonquantifiable health risk reduction benefits and costs. EPA may promulgate an interim standard without first preparing a health risk reduction and cost analysis or determination whether a regulation’s benefits justify the costs if the Administrator believes that a contaminant presents an urgent public health threat.

If the EPA Administrator determines that the benefits do not justify the costs, the EPA may promulgate a standard that maximizes health risk reduction benefit at a justifiable cost provided by the benefits. The EPA generally has set standards based on technologies affordable for large communities however, P.L. 104-182 requires the EPA to list any technologies or procedures that are affordable for small public water systems serving populations of 10,000 or fewer. If EPA does not identify “compliance” technologies that are affordable for small systems, then the agency must identify small system “variance” technologies or other means that may not achieve the MCL but are protective of public health. The SDWA allows for variances and exceptions based on costs.

Perchlorate Regulation

As of 11 August 2020, drinking water perchlorate is regulated in Arizona, California, Massachusetts, New Jersey, New York, and Texas. In addition, Illinois, Maryland, Nevada, New Mexico, and Wisconsin have advisory levels. The Government Accountability Office states that 10 states have established advisory or health-based perchlorate goals. Perchlorates can be classified as a D001 Resource Conservation and Recovery Act (RCRA) hazardous wastes under 40 CFR 261.23 based on perchlorate’s reactivity characteristic depending on specific circumstances. The EPA has also established a 55 mg/kg residential soil screening level (SSL) and a 720 industrial SSL under Superfund. The Department of Transportation regulates perchlorates under its 49 CFR 172.101 hazardous material table. Various states also list perchlorate as a hazardous substance such as Rhode Island and Pennsylvania.

Perchlorate’s Regulatory history

In 1995, the EPA established a provisional 4 – 18 ppb provisional reference dose range.

In 1997, the EPA first discovered perchlorate contamination in Nevada. At that time, 1,000 pounds (454 kg) per day of perchlorate entered Lake Mead and the Colorado River through contaminated groundwater. Lake Mead provides drinking water to residents of southern Nevada. The contamination originated from the only two perchlorate-manufacturing facilities in the United States at the time.

In August 1997, the Nevada Division of Environmental Protection selected 18 parts per billion (ppb) as the recommended action level for cleanup pending a more current risk assessment. Likewise, California’s Department of Public Health selected 18 ppb action level based on the EPA’s 1995 provisional reference dose range.

In 1998, the EPA published its first draft perchlorate assessment and perchlorate was added to the EPA’s first Contaminant Candidate List (CCL1) indicating that the EPA might be interested in regulating perchlorate. In 1999 perchlorate was included on the EPA’s first Unregulated Contaminant Monitoring Rule (UCMR1). Perchlorate was rolled over into CCL2 in 2005 along with 50 other CCL1 contaminants because the EPA was waiting on monitoring results and health effect studies. Without a regulatory determination perchlorate was rolled into a draft CCL3 in 2008.

In January 2002, the EPA revised its draft assessment with a 1 ppb reference dose (RfD); this assessment was widely criticized although it did cause California to lower its action level to 4 ppb. California selected 4 ppb as that was the 1995 EPA’s lower provisional reference dose range and the lowest level that the era’s analytical methods could detect.  After a second peer review which was also not well received by the scientific community, The Department of Defense, The National Aeronautics and Space Administration, The Department of Energy, and the EPA asked the National Academy of Sciences to provide an assessment which it rendered in January 2005 recommending a 0.0007 mg/(kg * day) RfD.

In 2001, perchlorate was detected in Massachusetts’ Military Reservation’s (MMR) groundwater at 600 ppb. The site was investigated from the 1980s when groundwater contamination was first discovered. In 1996 in response to the contamination discovered in Cape Cod’s Aquifer the first of four EPA SDWA administrative orders was issued forcing the Impact Area Ground Water Study Program’s creation. Cape Cod’s aquifer has 6 lenses: Sagamore, Monomoy, Nauset, Chequesset, Pamet, and Pilgram.  MMR is situated directly on top of Cape Cod’s aquifer’s most productive part: the Sagamore Lens. About 450 million gallons per day (MGD) flow through Cape Cod’s aquifer with nearly 60% of water flux due to the Sagamore Lens. The Sagamore Lens is the only drinking water source for Cape Cod’s 200,000 year-round and 500,000 seasonal residents. In 1997 the second administrative order was issued; this order mandated that the Army stop all training at MMR. I cannot understate how far reaching this was for the Army. In the “pre-war era” then the Army existed primarily to train. Stopping that main function was akin to ordering Coca-Cola to stop producing soft drinks. That order was the primary driving force behind the Department of Defense’s extensive perchlorate response. The Bourne Water District Officially requested state guidance in March 2002. Massachusetts officially recommended that Bourne should set a 1 ppb limit based on the EPA draft assessment. In 2003, the Massachusetts Department of Environmental Protection established a scientific advisory committee. In February 2004 the committee recommended, and Massachusetts adopted, a 1 ppb advisory level which was below the 4 ppb detection limit available at the time.

In 2003, a federal court in California found that Superfund applied because perchlorate is ignitable and a characteristic hazardous waste. California’s legislature enacted AB 826, the Perchlorate Contamination Prevention Act of 2003, requiring California’s Department of Toxic Substances Control (DTSC) to adopt regulations specifying best management practices for perchlorate and perchlorate-containing substances.

In March 2004, the California EPA Office of Environmental Health Hazard Assessment (OEHHA) set a 6 ppb Public Health Goal (PHG). A PHG is the contaminant level in drinking water that does not pose a significant risk to health. Unlike EPA’s Drinking Water Equivalent Level, OEHHA’s PHG level accounts for exposures to a contaminant from sources besides drinking water. It was not a regulatory requirement. Interestingly, the EPA and the University of California peer reviewed this document even though the EPA had initially suggested a lower reference dose from only water sources.

On 18 February 2005 the EPA set a 0.0007 mg/(kg * day) reference dose (RfD) for perchlorate consistent with the January 2005 National Academy of Sciences report. A (RfD) is a scientific estimate of a daily exposure level that is not expected to cause adverse health effects in humans with a 10 fold conservative uncertainty factor (meaning the level expected to not cause adverse health effects in health humans (No Observed Effect Level – NOEL) is actually 0.007 mg/(kg * day); the 10 fold uncertainty is to cover scientific error such as variability across life-stages, individuals, or genders and protect sensitive sub-groups such as fetuses). EPA’s RfD translated to a 24.5 ppb Drinking Water Equivalent Level (DWEL). In dilute water a part per billion is equivalent to a microgram per liter μg/L. A Drinking Water Equivalent Level, which assumes that all contamination comes from drinking water and is the contaminant concentration an average citizen will have no adverse effect with a margin of safety. Exposures above the DWEL are not necessarily considered unsafe because of the built-in safety margin. EPA’s Superfund cleanup program issued 24.5 ppb cleanup guidance based on the RfD.

In July 2006 Massachusetts became the first state to adopt a regulatory standard when it promulgated a 2 ppb perchlorate state drinking water standard. To arrive at a drinking water standard, the department considered information on the availability and feasibility of testing and treatment technologies, as well as data that demonstrated that perchlorate can enter drinking water as a by-product of hypochlorite (bleach) solutions used as disinfectants. The department chose to set the standard at a level that did not create disincentives for public water systems to disinfect their water supplies. The department determined that a maximum contaminant level of 2 parts per billion would provide the best overall protection of public health, considering the benefits of disinfection, while retaining a margin of safety to account for uncertainties in the available data.

In October 2007, California promulgated a 6 ppb regulatory standard. By law, the California Department of Public Health is required to set a drinking water standard as close to the public health goal as is economically and technologically feasible. California found that large water systems could meet the standard with $18 per customer annual costs while annual costs for smaller systems ranged from $300 to $1,580. The Department of Public Health proposed economic based variances for systems serving less than 10,000 customers if the estimated annual treatment cost per household exceeds 1% of the median household income in the community served.

On 10 October 2008, under President George W. Bush and EPA Administrator Marcus Peacock, the EPA published a preliminary determination not to regulate perchlorate. Administrator Peacock found that there was no meaningful opportunity for health risk reduction.

In January 2009, the EPA issued a 15 μg/L Interim Health Advisory for perchlorate (EPA 822-R-08-25) based on the Office of Water’s analysis to assist state and local officials in addressing contamination while the EPA conducted its perchlorate risk reduction evaluation.

In April 2010 the EPA’s Office of the Inspector General released a report critiquing the risk assessment process and procedures used by the EPA to develop and derive the perchlorate RfD.

On 11 February 2011, under President Barrack Obama and EPA Administrator Lisa Jackson, the EPA reversed this decision and decided to regulate perchlorate however, it provided no specific regulatory proposals. This decision caused me personally much consternation as I was writing my bachelor’s thesis on modeling perchlorate in groundwater using the Complex Variable Boundary Element Method and forced me to re-write several sections just before it was due!

In September 2012 the US Chamber of Commerce submitted a request for correction under the Information Quality Act regarding the regulatory decision. The EPA formed a Scientific Advisory Board that recommended many changes to the data under the decision. The Scientific Advisory Board recommended that the EPA abandon the standard RfD based MCL approach in favor of physiologically based pharmacokinetic/pharmacodynamic modeling based on mode of action. The EPA accepted and followed this advice.

On 18 February 2016, the National Defense Resource Council (NDRC), filed a complaint against the EPA essentially saying the EPA was failing in its mandate because the EPA had not yet proposed a perchlorate limit. The NDRC is a non-profit environmental advocacy group which started as a Scenic Hudson Preservation Conference outgrowth; oddly enough this group was created to block Consolidated Edison plans for a powerplant on Storm King Mountain right by West Point, New York.

On 18 October 2016 the NDRC and EPA reached consent degree in which the EPA did not admit to failing to comply with statutory obligations. The EPA entered the consent decree to prevent judicial interference with the remedial plan that it preferred. The EPA also expressly reserved all discretion under the SDWA and general administrative law principles. Basically, the NDRC simply wanted the EPA to hurry up and propose specific limits. Its important to note that the EPA could NOT have promised the NDRC any specific regulatory changes because that would violate the Administrative Procedures Act (APA). Agencies may agree to consider rule-making changes and to adopt regulations required by law, but they generally will not make substantive commitments concerning the content of regulations that are subject to APA requirements. The consent decree required the EPA to issue national drinking water regulation for perchlorate by 19 December 2019. The EPA asked, and the NDRC agreed, to extend the deadline until 19 June 2020.

On 23 May 2020 EPA Administrator Andrew R. Wheeler signed a proposed rule issued on 26 June 2019 setting a maximum contaminant level (MCL) coincidental with a maximum contaminant level goal (MCLG) of 56 micrograms per liter (μg/L). Alternatives included in the proposal were 18 μg/L or 90 μg/L MCLs coinciding with MCLGs, establishing an MCLG of 18, 56, or 90 with a National Drinking Water Standard, and withdrawing its 2011 decision to regulate perchlorate.

On 18 June 2020, after the public comment period, the EPA issued a press release about the final action declining to regulate perchlorate under the Safe Drinking Water Act. The NDRC was given until 9 July 2020 to challenge the scientific footing the EPA used; which it did. On 21 July 2020, the EPA published this final action. It is likely that the NDRC will sue the EPA to try and force a different outcome.

The EPA’s Stated Opinion

On 18 June 2020, the EPA issued a press release about the final action declining to regulate perchlorate under the Safe Drinking Water Act. On 21 July 2020, the EPA published this final action. EPA Administrator Andrew Wheeler said “State and local water systems are effectively and efficiently managing levels of perchlorate. Our state partners deserve credit for their leadership on protecting public health in their communities, not unnecessary federal intervention.” The EPA cited three main criteria for declining to regulate perchlorate:

  1. Californian and Massachusetts’ state level regulation of perchlorate (these two states constituted about 60% of what would have been exceedances in the new rule; the exceedances were based on the old UCMR1 monitoring campaign updated with current conditions from Massachusetts and California)
  2. Current remediation efforts to address perchlorate contamination especially in Nevada contaminating the Colorado River and Lake Meade as well as overall decreasing perchlorate levels
  3. Improved storage and handling procedures for drinking water disinfectants

Oddly enough, in the actual final regulatory action the third criteria cited was not mentioned. Improved bleach storage and handling only appeared in the press release.

The EPA’s analysis found that nationwide perchlorate regulation costs would significantly outweigh benefits and decided to focus its limited resources on more immediate and significant public health concerns.  The EPA also cited historical precedent.  By the EPA’s analysis the pesticide Aldrin (a DDT component) would have affected more water systems than perchlorate does, and the EPA declined to regulate Aldrin in 2003. The EPA also estimated that implementing a perchlorate regulation would cost $9.5-18M while the benefits only ranged from $0.3-3.7M. The EPA argued that perchlorate’s infrequent occurrence at concerning levels imposes high monitoring and administrative cost burdens on states and public water systems without rendering net tangible benefits.

The EPA wrote: “It is of paramount importance that water systems (particularly medium, small, and economically distressed systems) focus their limited resources on actions that ensure compliance with existing NPDWRs and maintain their technical, managerial, and financial capacity to improve system operations and the quality of water being provided to their customers, rather than spending resources monitoring for contaminants that are unlikely to occur.” While the EPA finds perchlorate dangerous above certain exposure levels however it doesn’t occur often enough to warrant the regulatory program costs.

Key Stakeholder Comments


On 2 July 2007 the AWWA, building on earlier letters from 2 February 2005 and 27 May 2005, recommended the EPA to regulate perchlorate. The AWWA stated that “National compliance costs for a perchlorate MCL ranging from 2 to 24 [ppb] is smaller than estimated compliance costs for other drinking water regulations.”

In the response period to the request for comment on the final action the AWWA submitted a comment supporting no regulation for perchlorate. Absent withdrawal, the AWWA requested that the EPA adjust monitoring requirements due to the significant burden the current requirements would impose on utilities and primacy agencies. The AWWA pointed out that the 2011 consent decree did not require the EPA to regulate perchlorate only to propose a regulation for perchlorate. The AWWA said the “EPA would be acting in an arbitrary and capricious manner if it finalized a perchlorate regulation because the costs of all three of the proposed MCLGs exceed the benefits.” Michigan v. EPA was cited as an example where the Supreme Court overturned a rule where “EPA refused to consider whether the costs of its decision outweighed the benefits.” The AWWA suggested that there were significant technical flaws in the proposed MCLGs and stated that the new pharmacokinetic/pharmacodynamic modeling the EPA used was a flawed method and criticized its transparency as well as a sensitivity analysis’ lack. As an example, the AWWA cited that the epidemiological data used by the EPA was drawn entirely from non-U.S. populations.  However, the AWWA charges that the measured variability between individuals and subpopulations is larger than the small perturbations in fT4 and clinical effects considered in the EPA analysis. If the EPA had decided to regulate perchlorate the AWWA requested that the EPA transition utilities to a 9-year monitoring cycle after a year below the MCL.


The NDRC naturally was displeased with the ruling. The NDRC falsely asserts that the EPA was required to regulate perchlorate under the 2016 consent decree.

The American Academy of Pediatrics

The American Academy of Pediatrics (AAP), a non-profit 67,000 member professional organization for primary care pediatricians, pediatric medical subspecialists, and pediatric surgical specialists dedicated to the health, safety, and well-being of infants, children, adolescents, and young adults requested a lower NDWPR than currently exists. The AAP cited that Children born with even mild, subclinical deficiencies in thyroid function may have lower IQs, higher chances of being diagnosed with attention deficit/hyperactivity disorder (ADHD), and visuospatial difficulties. The AAP ended their comment saying:

[we are] particularly concerned that EPA is considering withdrawing its 2011 determination to regulate perchlorate, relinquishing national oversight over a chemical with well-established health risks in drinking water. This would set a precedent inconsistent with EPA’s stated mission to protect public health. AAP urges the EPA to set a stronger MCLG [maximum contaminant level goal] for perchlorate that is based on all available evidence of potential harms to protect public health. A lower MCLG will allow EPA to generate reporting data that more accurately portrays the populations at risk and to better protect vulnerable populations

American Academy of Pediatrics comment on perchlorate’s final action

Primacy Agencies

California, Massachusetts, New York, and New Jersey, states which all set lower state-based limits on perchlorate, were highly critical of the EPA’s methods and also noted that the traditional EPA risk methods would have set an 8 ppb limit.

The Salt River Pima-Maricopa Indian Community from Arizona stated:

Withdrawal of regulation will only encourage the industry to abandon any preventative measures to contain current contamination

Salt River Pima-Maricopa Indian Community comment on perchlorate’s final action

My Personal Thoughts

The whole debate seems to center on cost-benefit analysis. Something ignored in all the cost-benefit analysis I’ve seen so far is the tendency for compliance costs to come down overtime with monitoring method development, investment, new treatment technologies, and research. It is unlikely however that costs will significantly drop absent that demand driven development. My time at Camp Edwards as a cadet showed me fireworks platforms can create dangerous plumes in small towns from seemingly innocuous widespread traditions. Massachusetts did not attribute the fireworks plume because that would force the county or state to pay the cleanup costs for that plume instead of the Department of Defense. There are probably many smaller water systems without awareness. The 2010 GAO report states that the EPA does not have a perchlorate tracking system and that perchlorate’s nationwide extent is unknown. It seems likely that larger water systems will be less likely to be contaminated and more cheaply treat perchlorate. The overall background perchlorate concentration falling is mainly due to regulation, as the EPA states particularly in California, Massachusetts, and Nevada. The FDA should also be leaned on to develop more robust perchlorate consumption numbers and most likely remove perchlorate from food contact surfaces. Any perchlorate threshold the EPA sets should incorporate total cumulative dose. With all this in mind however, US water systems are already poor and don’t even spend enough to replace or maintain infrastructure. Overall, I like the Californian approach where a relatively strict criteria was set with easy waiver procedures. Since that was not set the health advisory needs to be updated.  


Perchlorate can be included on future CCLs; it most likely should be as well. Without the regulatory spotlight perchlorate is unlikely to gather more evidence for or against regulation. To date no new chemical regulations have been passed since the 1996 SDWA amendments (although the EPA proposed PFAS regulation in March 2020). This may be indicative that the regulatory process is not working although it is positive that the EPA does not waste precious resources regulating trivial contaminants (there are 13 SDWA chemicals which have had zero violations – these were mandated by the 1986 SDWA amendments and should probably be de-listed).

References and Further Reading

Wastewater and Covid-19 Surveillance

Screenshot of Biobot report
Report from Biobot on Livingston County, MI on Covid-19 from wastewater

Covid-19 is currently a hot topic, environmental health and engineering is no exception. Wastewater is now in international news because of it! This Reuters article from 19 June 2020 for example shows that researchers found RNA from Covid-19 in Milan and Turin’s wastewater in December 2019 before China reported the first cases on 31 December 2019! The Italian National Institutes of Health examined 40 sewage samples collected in northern Italy between October 2019 and February 2020 and found that samples in Milan and Turin from 18 December 2019 showed SARS-Cov-2. Monitoring sewage for health purposes is known as “wastewater-based epidemiology” (WBE).

Early WBE

Using wastewater to track populations is not a new idea. It was first proposed by Christian Daughton in 2001 to track illicit drug use. You can read his paper here. As a former wastewater teacher of mine, COL Timmes, liked to say: “everyone passes through us.” Generally, he meant that you can’t easily hide from the central sewage system. In more polite terms raw wastewater is a reservoir of excretion products such as: parent compounds, metabolites, and genetic material. The earliest widespread use of WBE (then called “sewage epidemiology”) was in 2005 to monitor for illicit drugs which you can find here. After this early case WBE gained traction. At least Australia, Belgium, Germany, Ireland, Italy, the Netherlands, Norway, Spain, South Korea, the United Kingdom, and the United States use WBE to monitor illicit drug use. After this initial use WBE started to take off in public health circles and WBE started to be used to track broader chemical public health indicators, for instance alcohol consumption in Norway, counterfeit medicine distribution in the Netherlands, and even tobacco use in Italy.

Environmental engineers and public health officials eventually realized that any excreted substance that has known kinetic pathways in wastewater could be used to reverse engineer the initial concentration. All these early methods focused on chemicals and were based around mass spectrometry. WBE was then and is still used to study exposure to chemicals or pollutants such as pesticides, herbicides, and flame retardants. After the sewer’s viability as a surveillance network was established, someone around 2008 realized with some work they could use quantitative polymerase chain reaction methods (qPCR) to amplify, detect, and quantify genetic material.

WBE basics

WBE’s popularity continues to increase because exclusive reliance on testing of individuals is slow, costly, and generally impractical. WBE also often serves as a disease early warning indicator because asymptomatic or prodromal individuals typically don’t get tested and there may be underdiagnosis. In cases like this WBE serves as an unbiased community prevalence estimator. This is especially true with Covid-19 whose asymptomatic period is about a fortnight. Ultimately, WBE allows near real-time cheap monitoring of health indicators such as obesity, diabetes, drug use, microbial antibiotic resistance, and disease outbreak. Its use in disease outbreaks offers particularly rich data on genetic diversity of outbreaks and phylogenic analysis can reveal viral ancestry.

In Australia, the University of Queensland has been linking census data to wastewater samples across the country to see the interrelationship between wastewater chemicals and social and economic measures of a population. Doing that opened the study of socioeconomic influences on chemical consumption. This study showed that caffeine consumption is associated with aspects of financial capability and educational attainment in Australia for instance.

WBE success

WBE is successful in sentinel surveillance providing early outbreak warnings and in determining the efficacy of public health interventions. It is remarkably sensitive at picking up infections and viral load in wastewater. For polio for instance, WBE sensitivity is estimated at about 1 case per 10,000 uninfected people. WBE also allows spatial sensitivity by moving “upstream.” WBE can detect variations in circulating strains through phylogenic analysis allowing for comparisons between region and viral genomic evolution. Another important benefit of WBE is that it enables disease prevalence gauging by circumventing individual stigmatization which can arrive from clinical diagnosis (early AIDS research for instance).

SARS-Cov-2 Simplified WBE Procedure

In general all WBE follows the same process: pretreatment, concentration, recovery, secondary concentration, then detection. Detection normally means either molecular analysis or traditional culturing. In an International Water Association (IWA) webinar on 19 June 2020 Charles Gerba, an environmental microbiologist at the Water, Energy, and Sustainable Technology Center (WEST) in the University of Arizona provided an outline of how they were testing:

  • Gather a 500 mL to 1 L sample of wastewater (grab or composite was not specified)
  • Take a 100-250 mL aliquot to process
  • Spike some samples with 229E to test efficiency
  • Store at -80°C for future analysis
  • Centrifuge to remove solids because some virus are lost to solids – in general about 100 mL would spin down to 1-3 mL
  • RT-qPCR: biomarkers (gene targets) N1, N2, N3, E229. Normally N2 and E229 are used to ensure the signal is specific enough. N1 and N3 are typically dropped

Difficulties in WBE Interpretation

WBE sounds amazing and it truly is. It has already been used successfully to track public health threats from polio to alcohol and all these achievements for a field under 20 years old. Its full potential isn’t even near realized at this point. However, there are several issues in the field. The largest is the lack of standardization and inability to compare results between testing facilities. These two factors are intrinsically linked but one will not necessarily solve the other. Another set of issues revolve around tying total loads to population numbers.

Difficulties with standardization

WBE is still a new field. It has not decided upon standards for many common procedures yet. For instance, some areas preform pre-process techniques to lower the risk of catching Covid-19 from working with SARS-Cov-2. Different pre-processing techniques such as pasteurization or filtration, will produce different signal drops.

Even the sample collection is very different. In wastewater there are typically two kinds of sampling: grab and composite. Grab sampling reflects a discrete point in time and space; composite sampling essentially is several grab samples pooled together at regular time or spatial intervals. Composite sampling is the most common in wastewater because varying flow patterns cause hydraulic surges followed by intermittent periods of low to no flow. However, that does not necessarily make it the best method for WBE.

The solids amount in the wastewater can also reduce efficacy of RT-qPCR methods; what phase to analyze (particulate or liquid) can affect results. Likewise, different inhibitors used for sample shipment may reduce the signal strength. The specific method chosen as a standard unfortunately must consider cost as well as effectiveness and test time. Likewise decontamination procedures between tests must be considered.

Difficulties with linking viral loads to population cases

Sewers undergo infiltration and inflow (i/i). Infiltration is where groundwater enters the sewer system through joints or breaks, inflow is where water is channeled into a sewer from various sources into the sewer such as downspouts. Without getting too deep, there are combined, separate, and merged sewer systems referring to surface runoff or sewage removal. Most large cities have merged systems were sewers were initially built as combined but started providing separate runoff and sewage systems. In short, a remote lab won’t necessarily have the proper infrastructural or weather contextualization to interpret the RNA signal in testing.

Another significant hurdle for disease monitoring is figuring out each disease’s excretion pattern. While it may seem reasonable that a greater number of sick people or sicker people excrete a higher viral load this is not always the case. Extrapolating the viral load to clinical cases becomes complicated. If the disease already has a well known viral shedding pattern and spread pattern with significant effort based, on where in the outbreak a disease is, you can get a correlation however it would be predicated upon many assumptions. For diseases with well defined correlations between degree of illness and viral shedding combined with disease transmission knowledge it is not possible to distinguish between one moderately sick person and two or more asymptomatic people with any degree of precision. With novel diseases only trend analysis is possible. Given the unknowns around viral shedding it becomes difficult to determine how the RNA signal drop corresponds with prevalence drops in the local community. It also becomes difficult to determine how strong the signal change needs to be to differentiate from statistical noise.

Correlating viral loads with clinically identified cases becomes even more challenging because of variable excretion rates during the infection, temporal delays, inconsistent spatial variability due to travel leading to use of multiple wastewater treatment systems, i/i, inactivation during transport, or infrequent, absent or inadequate clinical testing. Genomic instability in wastewater, sampling variability (grab/composite), and viral concentration efficiency differences compound these problems.

Where the sample was taken from, for instance from the sewage network or treatment plant, is also believed to effect viral recovery making comparisons difficult. The type of upstream user, for example domestic or industrial, will make a large difference as well. Areas with more septic systems then become harder to check. Likewise there is a divide between smaller more rural populations and larger cities; cities tend to create more normalization and may not necessarily be compared to their rural counterparts.

Practical difficulties with WBE

The best monitoring schedule at what frequency and spatial resolutions remain open questions which most likely vary across diseases. Likewise, who pays for the monitoring is an important consideration. Currently, WEST’s price list is between $350 and $1,250 per sample depending on how exactly they perform and analyze the sample. The quantification level can be tricky as well since most PCR techniques were developed for the clinical setting instead of an environmental one. There is also a privacy issue with this sort of monitoring.


WBE is an amazing tool for disease monitoring but is better suited to looking at trends because direct comparisons across catchments remains elusive. Since some aspects rely on data individual to specific catchments (recent precipitation, sewer condition, length of sewer and viral decay in sewer transport etc…) direct comparisons between viral loads may never really be achieved.

Further Resources

US EPA on Coronavirus in water and wastewater

Research Centers:


  • Kitajima, M., Ahmed, W., Bibby, K., Carducci, A., Gerba, C. P., Hamilton, K. A., … & Rose, J. B. (2020). SARS-CoV-2 in wastewater: State of the knowledge and research needs. Science of The Total Environment, 139076
  • Nemudryi, A., Nemudraia, A., Surya, K., Wiegand, T., Buyukyoruk, M., Wilkinson, R., & Wiedenheft, B. (2020). Temporal detection and phylogenetic assessment of SARS-CoV-2 in municipal wastewater. medRxiv : the preprint server for health sciences, 2020.04.15.20066746
  • Venugopal, Anila, Harsha Ganesan, Suresh Selvapuram Sudalaimuthu Raja, Vivekanandhan Govindasamy, Manimekalan Arunachalam, Arul Narayanasamy, Palanisamy Sivaprakash et al. “Novel Wastewater Surveillance Strategy for Early Detection of COVID–19 Hotspots.” Current Opinion in Environmental Science & Health (2020)
  • Ahmed, W., Angel, N., Edson, J., Bibby, K., Bivins, A., O’Brien, J. W., … & Tscharke, B. (2020). First confirmed detection of SARS-CoV-2 in untreated wastewater in Australia: A proof of concept for the wastewater surveillance of COVID-19 in the community. Science of The Total Environment, 138764
  • Gracia-Lor, E., Castiglioni, S., Bade, R., Been, F., Castrignanò, E., Covaci, A., … & Lai, F. Y. (2017). Measuring biomarkers in wastewater as a new source of epidemiological information: Current state and future perspectives. Environment international, 99, 131-150
  • Xagoraraki, I., & O’Brien, E. (2020). Wastewater-based epidemiology for early detection of viral outbreaks. In Women in Water Quality (pp. 75-97). Springer, Cham

PFAS Discovery

Reenactment of the 1938 discovery of Teflon. Left to right: Jack Rebok, Robert McHarness, and Roy Plunkett
Reenactment of the 1938 discovery of Teflon. Left to right: Jack Rebok, Robert McHarness, and Roy Plunkett Photo courtesy of Hagley Museum and Library

Per- and polyfluoroalkyl substances (PFAS) are a widely used class of chemicals. You’re probably familiar with some of the popular brand names employing these chemicals such as Teflon, Gortex, and Dockers Stain Defender. As our understanding of PFAS has evolved it is becoming an emergent public health threat. This article is the first in a series serving to provide information on PFAS and it will cover their discovery. Other articles in the series will cover concern, regulation, treatment, environmental fate, contaminated sites, consumer protection, uses and provide sources for further information.


Per- and polyfluoroalkyls substances (PFAS) are a family of synthetic substances covering over 4,700 chemicals which have a number of deleterious health effects attributed to them. As of 7 June 2020 California, Connecticut, Colorado, Minnesota, North Carolina, New Hampshire, New Jersey, and Vermont all have state health guidelines for some PFAS. New Jersey’s limits just recently started on 1 June 2020. The US Environmental Protection Agency (EPA) has established a health advisory of 70 parts per trillion of combined PFAS.  While PFAS is not directly regulated under the Toxic Substances Control Act (TSCA) it is monitored under the Significant New Use Rule (SNUR). On 20 February 2020 the EPA proposed a supplemental SNUR for PFAS.

Like most things, an exact beginning is hard to quantify. PFAS’ story could start in a letter dated 26 August 1812 when André-Marie Ampère wrote to Humphry Davy postulating the existence of fluorine, or in 1869 when Dmitri Mendelvee positioned fluorine in the periodic table. An equally rational choice would be 1886 when Henri Moissan first isolated elemental fluorine in France leading to his award of the 1906 Nobel Prize in chemistry. Another serious contender would be Thomas Midgley and A. L. Henne’s 1928 invention of Freon and other chlorofluorocarbons (CFCs) in their Fridgidaire laboratory, which at the time was a General Motor’s subsidiary. I am choosing to start PFAS’ story with Roy Plunkett’s 1938 discovery of Teflon and the birth of fluoropolymers.  


On 6 April 1938 at the Chemours Jackson Laboratory in New Jersey Dr. Roy Plunkett discovered polytetrafluoroethylene (PTFE) by accident while researching new CFC refrigerants. While CFCs are now banned for their deleterious atmospheric effects under the Montreal Protocol, at the time they were used to replace ammonia and sulfur dioxide refrigerants which killed dozens of workers annually. PTFE is better known by its brand name: Teflon. Although CFCs are perfluorinated compounds, PTFE was the first discovered chemical in the class of Per- and polyfluoroalkyl substances.

After work on 5 April 1938, Dr. Plunkett and his assistant Jack Rebok reacted tetrafluoroethylene (TFE) with hydrochloric acid then compressed the mixture into metal cylinders and froze it overnight. The next morning on the 6th of April, Jack Rebok placed one of the cylinders onto a balance then opened the stop valve. Only 990 grams of TFE came out of a supposedly 1 kg container. Puzzled by the mass balance, Dr. Plunkett tipped the cylinder over and a white powder fell out. Then Dr. Plunkett stuck a metal wire to try and get more of the substance out. He was unable to get much out that way so eventually Jack Rebok suggested to cut the flask open.

Photo of lab notebook page where Plunkett recorded the discovery of PTFE
Photo of lab notebook page where Plunkett recorded the discovery of PTFE from Kinnane, A. (2002). DuPont: From the banks of the Brandywine to miracles of science.

In Dr. Plunkett’s article The History of Polytetrafluoroethylene: Discovery and Development he wrote:

On the morning of April 6, 1938, Jack Rebok, my assistant, selected one of the TFE (tetrafluoroethylene) cylinders that we had been using the previous day and set up the apparatus ready to go. When he opened the valve — to let the TFE gas flow under its own pressure from the cylinder — nothing happened…We were in a quandary. I couldn’t think of anything else to do under the circumstances, so we unscrewed the valve from the cylinder. By this time it was pretty clear that there wasn’t any gas left. I carefully tipped the cylinder upside down, and out came a whitish powder down onto the lab bench. We scraped around some with the wire inside the cylinder…to get some more of the powder. What I got out that way certainly didn’t add up, so I knew there must be more, inside. Finally…we decided to cut open the cylinder. When we did, we found more of the powder packed onto the bottom and lower sides of the cylinder.

Instead of ignoring the powder Dr. Plunkett started experimenting on it and discovered PTFE is highly resistant to corrosive acids, has excellent performance in extreme temperatures, and does not dissolve in solvents. This along with its slippery nature lead to DuPont sending PTFE to its central research department. However, no real commercially viable use was found for PTFE.

Ironically, World War II saved Teflon from oblivion. On its own PTFE, what Dr. Plunkett discovered, is a relatively useless polymer. It melts at around 327°C (≈620°F) and under that temperature sits in a ball of nonflowing gel. PTFE does not dissolve in anything and does not react with acids, bases, or solvents and at the time cost about $100 per pound (about $1,820 per pound or $4.02 per gram in 2020 dollars) to manufacture.

The Manhattan Project was the US effort to develop the atomic bomb and the savior of PTFE. The Manhattan Project needed corrosion resistant materials to separate U-235 from U-238 using differential diffusion of UF6. After Lieutenant General Leslie Groves heard of PTFE’s inertness, he verified it could separate U-235 from U-238. Then following LTG Groves’ request, the US Patent Office placed PTFE under a “Secrecy Order” and it was referred to only as “K-416.” Following military interest, DuPont patented PTFE in 1941 and registered the trade name Teflon in 1944.  The secrecy order lasted until 1946; by that time the Manhattan Project had paid for a great deal of research that otherwise would not have been carried out on the polymer and its manufacturing cost dropped tremendously.

Dr. Plunkett is also famous for leading DuPont’s team which added tetraethyllead (CH3CH2)4Pb to gasoline (which was phased out under the Clean Air Act) and for significant improvements to freon (a CFC refrigerant).

Brief PFAS Use Examples

The examples of PFAS extend across all facets of human life, from everyday household cookware to aerospace and electronics. PFAS’ widespread use takes advantage of its beneficial properties: chemical resistance, thermal stability, cryogenic properties, low friction coefficients, low surface energies, low dielectric constants, high volume and surface resistivities, and flame resistance. Once it was used to separate U-235 from U-238, the Manhattan Project immediately started finding other uses for PTFE. For example, PTFE was also used in the Manhattan Project in the antenna cap of proximity fuses thanks to its electrical insulating property and invisibility to Doppler radar. Fuel tank coatings used PTFE because of its resistance to low temperatures. After World War II, Teflon was turned to human well-being and started being used in catheters because of its low friction coefficient. PTFE was also used as insulation for wires and cables. It was even used during the Statue of Liberty’s renovations.  PTFE often serves as a precursor for other PFAS chemicals which were ubiquitous until their dangers were realized. PFAS are used in fire fighting foams, ski wax, stain-resistant materials (rugs, clothing, furniture, sprayable stain protectors), cookware, outdoor gear, cosmetics, shaving cream, sunscreen, shampoo, and myriad other applications.


While it may seem easy to villainize Dr. Plunkett for his discovery’s degradation of the environment and damage to human health it is critical to remember Dr. Plunkett in the context of this time. Early refrigerants included sulfur dioxide and ammonia; both regularly poisoned people. His contributions to tetraethyllead boosted octane levels enabling, among other things, advanced plane flight and jets. His work introduced numerous new products and processes which are widely used in medicine, refrigeration, aerosol, electronic, plastics, and aerospace. Several of his innovations are of critical importance to national defense.  There was also less awareness of the dangers of persistent chemicals to humans and the environment.

Further References

The EPA has designated Lahne Mattas-Curry as a point of contact and can be reached at

For more on the history and discovery of PFAS:

Other articles in the series include PFAS Structure

The EPA’s PFAS website is

The EPA Long-Chain Perfluoroalkyl Carboxylate and Perfluoroalkyl Sulfonate Chemical Substances; Significant New Use Rule; Supplemental Proposal


NJ’s new PFAS rule – a group of faculty, post-doctoral scholars, graduate students, and undergraduates affiliated with the Social Science Environmental Health Research Institute at Northeastern University operating on an NSF grant