PFAS in water environments: recent progress and challenges in monitoring, toxicity, treatment technologies, and post-treatment toxicity

Abstract

There is growing awareness of the environmental presence of per- and polyfluoroalkyl substances (PFAS) and their harmful effects on animals and humans. Recent studies have revealed changes in human embryonic stem cells and maternal biomarkers, underscoring the severity and unpredictable outcomes associated with long-term exposure to PFAS. Monitoring efforts continually identify additional PFAS compounds worldwide, but a standardized and unified approach is still lacking. Traditional treatment methods such as adsorption and membrane filtration have been effective in removing 80–95% of PFAS from wastewater. However, complete removal of short-chain PFAS remains limited to a few recently developed techniques. The inability of advanced treatment methods to eliminate emerging short-chain and ultrashort-chain PFAS suggests the need for more integrated approaches that target all PFAS classes. Additionally, a few studies have discussed the potential toxicity outcomes of these treatments at both laboratory and full-scale levels. While advanced oxidative processes (AOPs) are rapidly gaining attention for degrading 90–100% of PFAS in sewage, it remains challenging to fully break down PFAS into non-toxic, mineralized products such as CO₂ and H₂O due to the strong C-F bonds and the potential toxicity of by-products in post-treated wastewater. Standardized and reliable bioassays for assessing PFAS toxicity are still under development, and current predictive models linking molecular structure to human health effects are at an early stage. This review examines the emerging health and ecological risks associated with both legacy and novel PFAS, alongside recent advances and limitations in individual and combined treatment technologies for water and wastewater. Emphasis is placed on the potential toxicity of degradation products, highlighting the need for more integrated and comprehensive toxicity assessments to guide safer PFAS remediation strategies.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40068-025-00411-9.

Introduction

Per- and polyfluoroalkyl substances (PFAS) are industrially manufactured organofluorine chemical compounds with stable carbon–fluorine bonds (Gaines 2023). Polymeric and non-polymeric PFAS are the major classes reported in the literature (Buck et al., 2011). Fluorinated polymers consist of high-molecular-weight compounds with carbon-based backbones that are partially or fully substituted with fluorine atoms (Buck et al., 2011). The nonpolymeric PFAS, which are subclassified into perfluoralkyl (fully fluorinated) and polyfluoralkyl (partially fluorinated) substances, include a broad range of chemicals with various functional groups such as perfluoroalkyl sulfonic acids (PFSAs), perfluoroalkyl carboxylic acids (PFCAs), perfluoroalkyl sulfonamides (PFSNs), fluorotelomer-based compounds (FT-PFAS such as FTOHs), as well as semi-fluorinated aliphatic compounds such as alkanes and alkenes, and other related derivatives (Buck et al., 2011). Fluorotelomer alcohols (FTOHs) are key precursors that can degrade into PFCAs and, under some conditions, into PFSAs. Structural representations of some of these classes are provided in Fig. S1 (Buck et al., 2011; Dickman & Aga 2022; Glüge et al. 2020).

These compounds possess various physicochemical properties, including chemical inertness, strong and short C-F bonds, excellent thermal resistance, and surfactant  properties (Brunn et al. 2023; Gaines 2023; Naidu et al. 2025; Rayne & Forest 2009). The carbon chain length influences their physicochemical behavior, such as hydrophobicity, bioaccumulation potential, and environmental mobility (Gaines 2023; Glüge et al. 2020; Kissa & Kissa 2001; Rayne & Forest 2009). Although there is no defined upper limit to the carbon chain length of PFAS, research has predominantly focused on compounds with C4 to C14 chains (Brunn et al. 2023; Kissa & Kissa 2001). Longer-chain PFAS, such as those with C16 carbon atoms, have been reported but remain less extensively studied (Liu et al. 2018). As per US EPA classification, PFAS are classified by chain length as either short-chain, with fewer than six carbon atoms for PFSAs and fewer than eight for PFCAs, or long-chain, with six or more carbon atoms for PFSAs and eight or more for PFCAs (Li et al. 2020; Podder et al. 2021). Shorter PFAS are shown to exhibit higher hydrophilicity and water solubility compared to their longer alternatives (Li et al. 2020). Most PFAS are anionic surfactants and can have pKa values as low as 0.4 (perfluorobutane sulfonic acid, PFBS’s pKa) and a mean pKa value of 3.8 ± 1 (perfluorooctanoic acid (PFOA’s pKa) (Burns et al. 2008; Naidu et al. 2025; Rayne & Forest 2009). Glüge et al. (2020) categorized PFAS uses into several groups based on their structure–function relationships. This structural versatility has enabled diverse industrial applications, which in turn have contributed to their widespread occurrence in surface water, groundwater, drinking water, and ecosystems globally (Glüge et al. 2020). The environmental presence of PFAS was first reported by Giesy and Kannan in 2001 (Giesy & Kannan 2001). Since then, numerous studies have documented a wide range of PFAS classes in both environmental and biological matrices (Muir & Miaz 2021). Perfluorooctane sulfonic acid (PFOS) and perfluorooctanoic acid (PFOA) have historically been among the most extensively studied PFAS worldwide (Gallen et al. 2017; Giesy & Kannan, 2002; Podder et al. 2021). However, due to their environmental persistence and well-documented adverse health effects, these compounds have been phased out or are now subject to regulatory restrictions across multiple jurisdictions (Dettori et al. 2022; Podder et al. 2021State of Per- and Polyfluoroalkyl Substances (PFAS) Report, 2025).

PFAS are often referred to as “forever chemicals” because of the inherent physicochemical properties of carbon–fluorine bonds, which resist breakdown under typical environmental conditions (Brunn et al. 2023; Wee & Aris 2023b). Consequently, they frequently bioaccumulate in the body, bio-magnify through food chains, leading to dietary exposure in humans (Brunn et al. 2023). Mounting evidence has consistently shown the adverse health effects of PFAS, especially PFOA and PFOS, on humans, animals, and ecological systems with epidemiologic studies linking them to a wide range of harmful effects, including developmental toxicity, liver damage, fat metabolic toxicity, heart disease, immune system suppression, endocrine disruption, and kidney or testicular cancer (Ankley et al. 2020; Brunn et al. 2023; Das et al. 2015; Domingo & Nadal 2019; Donat-Vargas et al. 2019; Fenton et al. 2021; Steenland & Winquist 2021; Von Holst et al. 2021).

In response to continuous efforts to address the PFAS issue, an international guideline was established in 2019 to achieve a very low limit of quantification (LOQ) in water for at least 30 different classes of PFAS (International Organization for Standardization., 2019). The ISO 21675 standard, issued by the International Organization for Standardization, establishes a limit of quantification (LOQ) of 0.0002 mg/L or lower for the majority of the listed PFAS compounds (International Organization for Standardization., 2019). However, this varies by country and depends on the analytical methods used. In Canada, updated drafts were published on PFAS regulatory limits ( Health Canada, 2018a, 2018b). Maximum acceptable concentrations (MACs) for the PFOA and PFOS were set to 0.200 and 0.600 mg/L, respectively (Health Canada, 2018a , 2018b). MACs have been devised based on the length of human exposure to the pollutants, including children. The Canadian guidelines recently included MACs for another nine PFAS detected in drinking water (Health Canada, 2018a , 2018b). According to the guidelines, the total PFAS in any drinking water sample should not exceed 0.030 mg/L (Health Canada, 2024a). US EPA guidelines also restrict the presence of PFAS in drinking water to 0.0004 mg/L (EPA 2021). European Union (EU) guidelines have also limited the concentration of all PFAS in any sample to 0.500 mg/L and the sum of 20 selected PFAS to less than 0.100 mg/L (Table S1) (Dettori et al. 2022). The US EPA and Environment and Climate Change Canada have recently added fluorotelomer sulfonic acids (FTSA) due to their associations with several health effects in humans and animals ( Health Canada, 2024b).

To reduce PFAS in the environment, various remedial strategies targeting PFAS in wastewater have been developed, with particular emphasis on combined treatment technologies (Lu et al. 2020; Wanninayake 2021). The versatility of these combined approaches has contributed not only to higher removal efficiencies but also to a better understanding of PFAS remediation pathways in contaminated water. Among these, advanced oxidation processes (AOPs) coupled with adsorption have attracted considerable attention as a practical strategy for breaking down PFAS into simpler molecules through defluorination and eventual mineralization (Lu et al. 2020). Although these technologies have demonstrated promising removal performance, recent studies suggest that the risk levels in post-treated water may be higher than in pre-treated water, likely due to the formation of more toxic degradation byproducts (Liu et al. 2024). This review highlights the recent global monitoring of PFAS, advances in toxicity assessment, and the emerging potential of combined treatment technologies to reduce PFAS contamination in water and other environmental systems. Importantly, evaluating the toxicity of treated water is a critical step toward developing effective and sustainable solutions to mitigate the harmful effects of PFAS.

Global monitoring of PFAS and their sources

Although the industrial production of PFAS dates back to the 1950s, evidence of their environmental presence was not provided until the 1990s (Giesy & Kannan 2001; Podder et al. 2021). Figure 1 displays a graphical representation of spatial PFAS concentrations, compiled from the interpretation of all measured data from various sampling sites and circled as red (known Aqueous film forming foam (AFFF) or non-AFFF) or black (unknown sites) on a global map developed in a recent study (Ackerman Grunfeld et al. 2024). There is evidence that PFAS concentrations in most water samples collected from North America, Europe, and Australia still exceed the current US and EU regulatory drinking water standards (Ackerman Grunfeld et al. 2024). Studies across various regions have shown widespread PFAS contamination in water bodies, particularly near industrial zones and firefighting sites (Ackerman Grunfeld et al. 2024). In the US, PFAS levels in certain water sources have been reported to reach approximately 10 mg/L, likely due to their proximity to manufacturing and firefighting training sites, combined with extensive monitoring efforts that increase the detection frequency (Ackerman Grunfeld et al. 2024). The levels of PFAS in Canadian waters were lower, ranging from 0.0001 to 0.010 mg/L (Gewurtz et al. 2024). Spain, Belgium, and the Netherlands reported high levels of PFAS exceeding 1 mg/L, with Belgium showing levels as high as 2 mg/L (Ackerman Grunfeld et al. 2024; Jeong et al. 2022). China also contributed significantly to the presence of PFAS in Asia, with environmental levels reaching as high as 5 mg/L (Ackerman Grunfeld et al. 2024). Despite global phase-outs initiated since 2002, PFOA and PFOS continue to be detected in environmental and wastewater samples at concentrations exceeding regulatory limits (Baabish et al. 2021; Brunn et al. 2023; Lenka et al. 2021; Podder et al. 2021). For example, wastewater treatment plants (WWTPs) in China have reported elevated levels of these compounds. However, some studies indicate a gradual decline in their total concentrations over time (Zhou et al. 2019). Considerably high levels of 6:2 fluorotelomer thioether amido sulfonate (6:2 FtTAoS) and PFOS were detected in both influents and effluents of wastewater near airport industrial WWTPs. At the same time, C4–C7 perfluorocarboxylic acids (PFCAs) were unexpectedly found predominantly in the effluents, likely due to the total PFAS conversion during the secondary treatment process (Houtz et al. 2018; Podder et al. 2021). A side-chain PFAS polymer, perfluorooctane sulfonamide-urethane polymer (S1), was detected in 20 Canadian WWTPs at concentrations higher than other PFAS (Letcher et al. 2020). Currently, most PFAS analyzed in waters are short-chain molecules, as they exhibit enhanced mobility and the ability to relocate and spread to various locations (Brunn et al. 2023; Podder et al. 2021). Perfluorobutanoic acid (PFBA) and perfluorobutane sulfonic acid (PFBS) alone accounted for over 50% of detected short-chain PFAS in multiple monitoring studies (Brunn et al. 2023; Li et al. 2020). In one study conducted in China, PFBS and PFBA were among the most frequently detected PFAS in WWTP influents and effluents, with frequencies ranging from 95.8% to 100% (Zhou et al. 2019), while analyses in Germany detected C6 and smaller PFAS compounds in surface waters (Joerss et al. 2022; Johnson et al. 2022a, b). Beyond short-chain analogs, emerging PFAS alternatives, such as ADONA (dodecafluoro-3H-4,8-dioxanonanoate) and side-chain fluorinated polymers like S1, have been detected in European and Canadian waters, with notable accumulation even being detected in human serum (Fromme et al. 2017; Letcher et al. 2020). Table S2 summarizes some emerging classes of PFAS, primarily short-chain PFAS and their corresponding concentrations, in various environmental samples.

Fig. 1.

Analysis of the global distributions of total concentrations of 20 PFAS (subject to EU regulation, 20 PFAS in Table S1) on six continents (North America, South America, Europe, Asia, Australia, and Africa). Most of these PFAS were PFCA and PFSA linear chains ranging from C4 to C13: a The percentage of PFAS concentration on each continent was calculated based on a recent analysis of 45,000 samples from different drinking and groundwater locations. b The actual concentration of PFAS in each country

While these alternatives were introduced to replace phased-out PFAS, there is limited information on their long-term environmental fate and health risks. Despite significant progress in PFAS monitoring, large data gaps remain, particularly in South America, Russia, and the Middle East. Standardized screening methods across countries and matrices are essential for enhancing comparability and evaluating global risks. Continued research is crucial to understand the toxicity of emerging PFAS and their potential for persistence, bioaccumulation, and human exposure (Fenton et al. 2021).

Potential sources of PFAS in waterbodies

PFAS can enter aquatic environments through various pathways (Baabish et al. 2021; Gallen et al. 2017; Gonzalez et al. 2021; Lenka et al. 2022; Meng et al. 2018; Tavasoli et al. 2021). As illustrated in Fig. 2a, these include emissions from industrial activities, wastewater discharge, landfill leachate, and atmospheric deposition. Among these, industrial and municipal wastewater treatment plants (IMWWTPs) are recognized as major global contributors to PFAS contamination in regions such as the US, Canada, Europe, and Asia (EPA 2021; Gewurtz et al. 2024). In Fig. 2b, a case study from eastern China is presented, where industrial discharges accounted for approximately 78% of PFAS emissions into surface and groundwater, with legacy compounds such as PFOA and PFOS dominating. In some instances, PFAS concentrations in WWTP effluents exceed those in influents, due to the transformation of precursors such as fluorotelomer alcohols (FTOHs) into stable end-products like PFCAs and PFSAs, as stated before (Lenka et al. 2021, 2022; Podder et al. 2021; Wee & Aris 2023b). Other notable sources include AFFFs, sewage sludge, and agricultural pesticides, which together contribute additional loads of 6, 5, and 3%, respectively, in certain parts of Asia (Brunn et al. 2023; Liu et al. 2017). Sewage sludge presents further concerns, as PFAS, particularly long-chain compounds, tend to adsorb strongly to solids, making them persistent in the environment and potential contaminants of drinking water if leached into groundwater (Meng et al. 2018). A study in Switzerland found that sewage sludge from 45 WWTPs reported several long-chain PFAS at relatively high levels (up to 75 µg/kg) (Alder & Van Der Voet 2015). Landfills can contaminate waterbodies indirectly by leaching into groundwater (Fig. 2a) (Hepburn et al. 2019; Zhang et al. 2023). Atmospheric deposition has also become a significant source of PFAS contamination, with PFAS-laden aerosols dispersing globally and depositing in remote regions, such as the Arctic, where they accumulate in marine environments (Podder et al. 2021; Sha et al. 2021).

Fig. 2.

Some potential sources of PFAS in the aqueous environment. a PFAS can enter the environment through multiple pathways. b Representative information from a previous study showing the PFOS and PFOA emission sources in a major industrial country like China, calculated based on the average occurrence of PFOS and PFOA in each source

Evolving studies of the toxicity of PFAS-contaminated water on the environment

Ecological and aquatic-based studies

There is an increasing awareness of the toxicity of PFAS-contaminated water, as indicated by various published reports on their effects on plants, animals, and humans (populations or cells) (Fenton et al. 2021; Wee & Aris 2023a). The bioaccumulation of PFAS in aquatic organisms can ultimately lead to humans. Sects. "Ecological and aquatic-based studies" and "Human-based studies" discuss recent reports on the toxicity of PFAS. Table 1 lists some of their adverse effects on plant cells, animals (Zebrafish as a model animal), and human cells.

Table 1.

Representative studies on the toxicity of PFAS on human cells and different model organisms

Organism/Model	PFAS Tested	Toxicity Endpoint	Value (mg/L or Kg)	Observed Effect	References
Abutilon Theophrastus	PFOS	Plant growth, stress defense, PFOS uptake, and elemental and metabolite profiles	0, 25, or 50 mg/kg soil	Growth inhibition	Rico et al., (2024)
Daphnia magna	PFOS, PFDA, PFBA, PFNA	EC50	67.82, 163.47, > 1000, 176.8	Growth inhibition	Pietropoli et al., (2024)
	PFOS + PFAS	Mixture effect	Additive	Additive toxicity
	PFDA + PFBA + PFNA	Mixture effect	Additive	Antagonistic interaction
Raphidocelis subcapitata	PFOS, PFDA, PFBA, PFNA	EC50	26.51, 29.11, 31.29, 75.98	Growth inhibition
Xenopus laevis embryos	17 PFAS incl. PFOS, FC10-diola, and FHxSA	NOECs of PFOS, FC10-diol, FHxSA and others	2.5, 1.25, and 0.625 mg/L, 5 for all others, 100% mortality at 5 mg/L FC10-diol	Developmental toxicity	Degitz et al., (2024)
Zebrafish embryos, larvae, and juveniles	PFTrDA	Endocrine-disrupting effects (sex hormone production (concentration), transcription of steroidogenic genes)	50 (for T hormone), 0.01 (CYP17A gene)	Reduction in Testosterone, 17β-estradiol Less CYP17A mRNA expression	Jo et al., (2014)
Zebrafish embryos & larvae	PFOA, PFHxA, PFBA, PFBS	LC50	57.6b, 76.3, 83.6, 1394	Mortality, morphological changes	Wasel et al., (2021)
Zebrafish embryos	PFOA, GenX	LC50	2, 5–10	Mortality	Liu et al., (2023)
	PFOA + GenX	Mixture effect	NA
Zebrafish larvae	PFOS, PFOA, PFHxS, and PFHxA	Mortality	0.0007, 0.00175, 0.0007, and 0.0035	Mortality, overall development, developmental defects, and larval activity	Hamed et al., (2024)
Xenopus laevis embryos/larvae	1H, 1H, 10H, 10H-perfluorodecanediolc	100% Mortality	5	Acute toxicity	Degitz et al., (2024)
HepG2 cells (human liver)	PFOA, PFOS	Cell viability	NA	Additive toxicity (binary mixtures)	Ojo et al., (2021)
Homo sapiens (pediatric human liver)	PFAS, PFOS, PFOA, PFNA, PFHxS, and PFUnDA	Liver enzyme levels (188 metabolites, including amino acids)	2.38, 6.74, 0.59. 0.72 and 0.2	Higher valine, leucine, isoleucine, tryptophan, and phenylalanine	Stratakis et al., (2020)
Homo sapiens (breast)	Ten PFAS incl. PFOS and PFHxS	Estrogen receptor (ER) tumor	5.64, 0.77	Positive correlation between PFAS and ER-positive tumor	Tsai et al., (2020)
Human embryonic stem cells	PFOA, othersd	Cytotoxicity	Point of departuree	Cell death	Doris Tsai & Ford, (2024)
Human cardiomyocyte stem cells	PFOA,  othersd	Fibrosis, lipid disruption	Point of departuree	Lipid metabolism interference

^aThese sub-lethal concentrations were determined without the use of a buffer. ^b Table S3, ^c term used to describe the point at which a significant change in gene expression occurs (Doris Tsai & Ford 2024). ^d Table S4, ^e1H, 1H, 10H, 10H-perfluorodecane-1,10-diol (FC10-diol)

Plant-based studies

A recent study investigated the impact of PFOS on the growth of velvetleaf (Abutilon theophrasti), a spontaneous terrestrial plant, using three soil concentrations (0, 25, and 50 mg/kg, see Table 1) (Rico et al. 2024). As PFOS concentrations increased, biomass yield markedly decreased from 15 g to less than 2 g. This stunted growth was associated with disruptions in 13 out of 56 examined metabolic pathways, including arginine biosynthesis, the citric acid cycle, and fatty acid synthesis. The accumulation of PFOS in shoots and roots also inhibited the production of antioxidants such as ascorbic acid and hydroxycinnamic acids, compromising the plant’s oxidative stress defense (Rico et al. 2024). Several mechanistic factors may explain these effects. The amphiphilic nature of PFOS likely promotes its preferential accumulation in root tissues, affecting local bioavailability and toxicity (Mayakaduwage et al. 2022; Müller et al. 2016). Although the exact translocation mechanism remains unclear, it may be influenced by plant-specific traits and environmental conditions (Rico et al. 2024). Furthermore, PFOS may chelate essential elements such as Mg²⁺ ions, which were observed to redistribute from roots to leaves. The formation of PFOS–Mg(H₂O)₂–PFOS complexes could reduce the mobility and availability of PFOS in shoots (Zhao et al. 2018). Notably, changes in fatty acid metabolites at elevated PFOS concentrations may increase reactive oxygen species (ROS) production, compromising membrane integrity and structural stability. Although these findings highlight the ecological risk of PFOS bioaccumulation in vegetation, further research is needed to characterize the toxicity profiles and mechanistic effects of other PFAS classes on diverse, ecologically important plant species.

Animal-based studies (Zebrafish (Danio rerio) as a model)

Prior to human studies, animal models—especially zebrafish—have been pivotal in exploring PFAS toxicity due to their genetic similarity to humans and sensitivity to environmental chemicals (Axel Elizalde-Velázquez & Elizabeth Herrera-Vázquez 2023). Chronic exposure to PFDA and PFTrDA in zebrafish disrupted sex hormone balance and altered the expression of genes involved in steroidogenesis and reproduction, including upregulation of cyp19b and gnrhr4 and downregulation of er2b and cyp11a(Jo et al. 2014). PFDA exposure in males induced vitellogenin gene expression, suggesting estrogenic activity, while PFTrDA exposure in females reduced vitellogenin expression, correlating with lower estradiol, poorer egg quality, and reduced reproductive success (Jo et al. 2014) (Table 1). A separate study tested a PFAS mixture of PFOS, PFOA, PFHxS, and PFHxA at environmentally relevant concentrations, finding limited larval mortality (~ 23%) but significant morphological deformities after 120 h post-fertilization, including curved body axis, swim bladder defects, and edemas (Hamed et al. 2024). These effects were attributed to PFAS binding to transthyretin (TTR), which disrupted the transport of thyroid hormone (T4) and affected water permeability in embryos. The study highlighted that mixture exposures can produce more pronounced effects than individual PFAS, reflecting the greater complexity of real-world contexts, such as human blood.

Human-based studies

Epidemiological studies strongly link PFAS exposure to various human disorders (Fenton et al. 2021). Long-term exposure has been associated with altered serum and plasma biomarker levels (Canova et al. 2020; Duan et al. 2020; Pitter et al. 2020). For example, in a multi-country European study, maternal blood contained PFOS, PFOA, PFNA, PFHxS, and PFUnDA, with corresponding metabolomic changes observed in children (Stratakis et al. 2020). The children’s metabolome was also examined, and variations in the concentrations of biomarkers (glycerophospholipids and amino acids) were observed (Stratakis et al. 2020). As a result, a strong association was reported between lasting maternal exposure to PFAS during pregnancy and an increased risk of liver injury in children. Metabolic profiling provided insights into the root cause of the observed liver injury, revealing alterations in biomarkers and lipid metabolism pathways associated with liver disease. Although the precise mechanism remains unclear, it was suggested that PFAS can cross the placental barrier and accumulate in fetal tissues. Furthermore, PFAS have a strong affinity for binding to peroxisome proliferator-activated receptors (PPARs), and through this interaction, they may disrupt hepatic lipid metabolism. PFAS interactions with peroxisome proliferator-activated receptors (PPARs), and their hydrophobicity are thought to contribute to disrupted hepatic lipid metabolism, but full mechanistic understanding requires further research. Beyond liver effects, PFAS exposure has been tied to cancer risk (Steenland & Winquist 2021; Van Gerwen et al. 2024). Notably, higher PFOS and PFHxS levels correlated with ER-positive breast tumors in Taiwanese women (Tsai et al. 2020). Other studies explored associations with hormone-receptor subtypes and possible gene interactions, though mechanistic evidence remains inconclusive (Chang et al. 2023; Ghisari et al. 2017; Sonthithai et al. 2016). Figure 3 illustrates some toxicological responses triggered in humans, along with the classes of PFAS involved in these responses. PFOA and PFOS were the main PFAS associated with most of the responses, including developmental and cardiometabolic effects.

Fig. 3.

Toxicological effects of PFAS on humans and the main classes involved in triggering their response

Recent studies have shown that PFAS exposure is linked to disrupted embryonic cell fate. Six PFAS (PFOA, PFHxA, PFBA, PFBS, PFOS, and PFHxS) were tested on human embryonic stem cells in both embryoid bodies and monolayer differentiation systems (see Table 1) (Doris Tsai & Ford 2024; Zhao et al. 2024). After 16 days of exposure to low, noncytotoxic concentrations (around 2–5 mg/L), ciliogenesis was significantly inhibited, primarily due to the downregulation of the cilia-related gene IFT122, which likely impaired the intraflagellar transport (IFT-A) complex. The most severe impact was seen with early PFOS exposure. Disruptions were also identified in essential cellular pathways, including the WNT and TGF-β pathways, which are vital for embryonic development, tissue balance, and cell differentiation. Broader testing with 26 PFAS compounds further showed changes in stress response, extracellular matrix, and contractility pathways in hepatocytes and cardiomyocyte stem cells. Overall, these results suggest that PFAS toxicity during both early and later developmental stages could have long-lasting health effects on humans and might even influence future generations (Zhao et al. 2024).

As outlined in Table 1, numerous toxicological studies have investigated the effects of PFAS across a range of test organisms, including aquatic species, animal models, and human embryonic stem cells. While these studies provide valuable data on individual and mixed PFAS—especially legacy compounds like PFOA and PFOS—the findings remain highly variable and sometimes inconsistent, limiting direct comparisons and broader conclusions. For instance, although PFOS exhibited the highest toxicity toward Daphnia magna (EC₅₀ = 67.82 mg/L) and PFDA toward Raphidocelis subcapitata (EC₅₀ = 26.51 mg/L), reported LC₅₀ values for PFOA in zebrafish embryos ranged dramatically between studies (2 mg/L to 57.6 mg/L), reflecting differences in test design, exposure conditions, or organism sensitivity (Liu et al. 2023; Pietropoli et al. 2024; Wasel et al. 2021). Mixture studies further illustrate this complexity: some PFAS combinations produced additive effects (e.g., PFOS in D. magna), while others exhibited antagonistic interactions (e.g., PFDA with short-chain PFBA and PFBS). In animal models, mixture effects appeared even less predictable, underscoring the challenge of extrapolating results to real-world scenarios where organisms are exposed to complex PFAS mixtures at variable concentrations (Pietropoli et al. 2024). Although animal models offer standardized methods and shared biomarkers, human toxicity data remain sparse and largely dependent on in vitro systems using stem cell lines (Doris Tsai & Ford 2024; Fenton et al. 2021; Nguyen et al. 2024). Even then, most studies focus on a narrow group of PFAS, limiting our understanding of the broader chemical class. Moreover, some experiments rely on unrealistically high exposure concentrations (> 10 mg/L), which may overestimate ecological risk; in contrast, studies that test environmentally relevant levels are still relatively scarce (Degitz et al. 2024; Hamed et al. 2024; Stratakis et al. 2020). Testing these lower, realistic levels is essential for understanding how PFAS behave and interact in natural environments, ultimately supporting regulatory bodies in setting guidelines, safety limits, and intervention strategies. For example, zebrafish studies have contributed to defining PFAS thresholds relevant to aquatic systems, though additional models are needed to strengthen these findings (Hamed et al. 2024). Likewise, certain human case studies have helped reveal associations between PFAS exposure and disease, offering valuable reference points for regulatory decision-making.

Individual and combined remediation approaches of PFAS in water and wastewater

Since the 1980s, research on PFAS remediation in freshwater and wastewater has grown rapidly in response to rising concerns about toxicity (Ateia et al. 2019; Dickman & Aga 2022; Lu et al. 2020; Yadav et al. 2022). As shown in Fig. 4, this has led to the development of diverse treatment technologies influenced by PFAS properties, cost, location, and environmental factors. While traditional methods encompass physical, chemical, and biological treatments, recent publications have predominantly featured chemical and physical approaches, with slower progress in biological and AOP-based methods (Yadav et al. 2022). Newer strategies increasingly combine treatments—such as physicochemical, biophysical, and biochemical methods—implemented either in tandem (concentrating PFAS before destruction) or in parallel (sequentially combining similar processes) (Lu et al. 2020). Table 2 illustrates some examples of the recent combined treatment techniques of PFAS and their degradation efficacies.

Fig. 4.

The research progress on PFAS use and treatment. a A timeline of the research progress from the early 1940s until 2023. b Common PFAS treatment technologies in wastewater used since 1981 (the values were calculated based on the estimated number of total studies performed on representative treatment technologies described in the review)

Table 2.

Parameters and degradation/mineralization efficiency of some representative and recent combined technologies

Target PFAS	Individual or mixture	Combined technology	Technology Type	Reactive species	Reaction conditions (conc., Time, pH, species, catalyst)	Scale	% Removala/Degradation efficiency	Mineralization indicators (% defluorination)	Refs.
PFOA, PFHxS, PFBS, 6:2FTS, GenX	Individual	Two stages: 1) Photocatalysis (PC) (WO3/TiO2) 2) photocatalysis Ozonation (PCO) UVA/visible light	Physiochemical (parallel)	O2•−	0.25 mg/L, 4 h, Neutral, O3, 0.2 g/L	Lab	Stage 1: PC 3–26% Stage 2: PCO 23%, 19%, 9%, 24%, 4%	N/A	Lashuk et al., (2022)
PFOA	Individual	Two stages: 1)Adsorptiona 2)Photocatalysis (UVC-254-sulfate)	Physiochemical (tandem)	Sulfate radicals SO4•−	100 mg/L, 24 h, 5.2, Sulfate, 0.4 g/L	Lab	Stage 1: 94.8%–99.1% Stage 2: 100%	83.1%	Ren et al., (2023)
PFOA	Individual	Two stages: 1)Photocatalysis (UVC-sulfite) 2)UVC-sulfite/Adsorption (biochar surfactant)	Physicochemical (parallel)	Sulfite radicals SO3•−	0.5–1 mg/L, 2 h, 6–7, BSS = 0.5 mM, Sulfite	Lab	Stage 1: < 27% Stage 2: > 69.6%	8.3%	J. He et al., (2024a, b)
PFOA, PFOS, PFBS, PFBA	Individual and mixture	Two stages: 1)Adsorption 2)Photocatalysis (In, Fe, or Ga/TNTs @AC, UV254)	Physiochemical (parallel)	O2•−, OH•	0.2 mg/L, 2 h, 8, 1 g/L	Lab and pilot	Stage 1:90% Stage 2: 95% 35% (mixture)	(Fe) 53.9% (In, Ga) < 10%,	Junker et al., (2024)
29 PFASb Long PFAS, short PFAS	Individual	Three stages: 1)Ultrafiltration 2)Foam fractionation 3)Electrochemical process	Physiochemical (tandem)	N/A	0.005 mg/L (influent), 0.002 mg/L (effluent), 0.015 mg/L (foam) 0 and 9 h 7.5, 7.6	pilot	Stage 1: 60% (groundwater) 59% (leachate) Stage 2: 86% (long) 31% (short)	40%	Smith et al., (2023a, b)
15 PFAS PFCAs, PFSAs, FOSA, 6:2 FTS	Mixture	Two stages: 1)Nanofiltration 2)Activated carbon (AC) or ion exchange (IX)	Physical physical (parallel)	N/A	< 0.006–0.110 mg/L, 35 weeks, 7.7–7.9	pilot	Stage 1: ~ 99% Stage 2: IX > GAC	N/A	Franke et al., (2019)
7 PFAS	Mixture	Two stages: 1)Nanofiltration 2)Activated carbon (AC)	Physical Physical (parallel)	N/A	0.0027–0.160 mg/L, 15 weeks, 7	pilot	Stage 1: > 98% Stage 2: IX > GAC	N/A	Franke et al., (2021)
PFOA	Individual	Two stages: 1)Nanofiltration 2)Electric field	Physiochemical (parallel)	N/A	0.5 mg/L, 48 h, N/A	Pilot	Stage 1: 45% Stage 2: 97%	⁓90%	Ji et al., (2023a, b)

^a(bark particles with grafted amine groups)

^bTable S5 shows the PFAS used, which are classified into short PFCAs, short PFSAs, long PFCAs, and other PFAS (Smith et al. 2023a, b)

Physical treatment

Physical treatment methods are widely investigated for PFAS remediation because they can effectively reduce PFAS concentrations in water (Fig. 4). These approaches primarily rely on sorption (e.g., activated carbon, ion exchange resins) or separation processes (e.g., membrane filtration) to capture PFAS from the aqueous phase (DiGuiseppi et al. 2024). However, a key challenge is that these methods do not degrade PFAS; instead, they transfer them into secondary waste streams, such as spent adsorbents or retentate concentrates, which then require further disposal or destructive treatment. The mechanisms, strengths, and limitations of these physical processes are discussed in more detail in Sects. "Activated carbon (AC)" and "Membrane technology".

Activated carbon (AC)

Activated carbon treatment has been extensively tested on various PFAS classes, with performance governed mainly by two primary factors: the hydrophobicity of PFAS and the adsorbent pore size distribution (He et al. 2024a, b; Murray et al. 2019). In one study, granular activated carbon (GAC) was evaluated for its capacity to remove 15 distinct long- and short-chain PFAS from groundwater. Results showed effective removal of long-chain PFAS, such as PFOS and PFHxS, whereas short-chain compounds like PFBA and PFPeA were poorly adsorbed (Rodowa et al. 2020).

This trend has been consistently observed across other studies, suggesting that higher hydrophobicity generally enhances adsorption onto GAC surfaces when comparing PFAS with similar functional groups (Rodowa et al. 2020). The underlying mechanism is likely related to the greater hydrophilic character of short-chain PFAS, which tend to remain dissolved in water rather than partitioning onto the carbon surface. Furthermore, long-chain PFAS preferentially adsorb into larger mesopores ranging from 0.002 to 0.05 µm, while short-chain PFAS can better access micropores with diameters below 0.002 µm. At full scale, GAC was successful at eliminating perfluoroalkyl carboxylic acids (PFAAs) with six or more carbon atoms and perfluoroalkyl sulfonic acids (PFSAs) with four or more carbon atoms in surface water (Kempisty et al. 2022).

Another form of activated carbon, which has been extensively tested in both bench- and large-scale applications, is colloidal activated carbon (CAC) (Niarchos et al. 2023a, b; Niarchos et al. 2023a, b). Although CAC has been successfully applied to soil-contaminated sites and aquifers with longevity exceeding three to four decades, a recent field study observed an increase in PFAS concentrations resulting from a decrease in the treatment’s effectiveness (Niarchos et al. 2023a, b). This may be due to site-specific factors, such as the site’s hydrology and groundwater chemistry (Hakimabadi et al. 2023). While activated carbon has proven effective for removing PFAS from water with relatively low energy requirements, it still faces significant limitations (Co 2023; DiGuiseppi et al. 2024). Regeneration and disposal of spent adsorbents are technically challenging and can incur additional costs, while also raising environmental concerns due to the concentrated PFAS load retained in the adsorbent, which may present unknown risks if re-released into the environment (DiGuiseppi et al. 2024). Sect. "Thermal treatment (ThT)" discusses possible waste management practices. Moreover, its removal efficiency declines when addressing a broader spectrum of PFAS, particularly those with ultrashort-chain congeners (Dickman & Aga 2022). In practical applications, the cost and performance of GAC are closely linked to the specific PFAS mixture and concentration in the treated water. For example, treating wastewater containing 110 ng/L (sum of seven PFAS) using GAC followed by incineration was estimated to cost up to $13 million over 20 years (Ling 2024). However, comprehensive field-scale assessments of long-term cost-effectiveness remain scarce (Kempisty et al. 2022). In terms of energy demand, a recent evaluation by the Minnesota Pollution Control Agency assigned GAC a relative score of 3 on a scale of 1–3, indicating comparatively lower energy consumption (Co 2023).

Ion exchange (IX) resin

Ion exchange (IX) resin has been tested for PFAS removal and has been shown to eliminate various PFAS classes effectively (Dixit et al. 2021; Ellis et al. 2025; Liu et al. 2022a, b). IX treatment works by exchanging ions of the same type and replacing them with other ions of the same charge. It is designed to purify high volumes of water (DiGuiseppi et al. 2024). Since most PFAS are present in water as anionic species, ion exchange (IX) resins are designed with cationic functional groups that effectively bind to these anionic moieties, such as carboxylate (–CO₂⁻) or sulfonate (–SO₃⁻) groups (Ellis et al. 2025). During this process, low-affinity anions, typically chloride ions (Cl⁻), are displaced into the aqueous phase to maintain charge balance. IX resins can be broadly categorized into “single-use” resins and reusable resins. Recent developments in PFAS-selective resins have primarily focused on single-use designs, which often exhibit improved selectivity and removal efficiency, particularly for long-chain PFAS that are otherwise challenging to capture (Cheng & Knappe 2024). However, these single-use systems introduce practical challenges, including the high cost of frequent replacement and the disposal of spent resins that now contain concentrated PFAS loads. Although regenerable resins have not yet been widely applied in large-scale water treatment, they appear more promising for targeting short-chain PFAS and may offer better economic feasibility compared to single-use resins (Ellis et al. 2025). More recently, research has also focused on developing effective regeneration strategies for single-use PFAS-selective resins, which could further reduce operational costs and improve sustainability (Ellis et al. 2025). Treatment of PFAS has been carried out on multiple fabricated anionic/single-use IX resins, some of which are commercially available, patented resins, and sold under different commercial names (Boyer et al. 2021; Dixit et al. 2021). One critical factor influencing the performance of these resins is their structural design and porosity. For instance, IRA67—a polyacrylic gel resin functionalized with tertiary amine groups—showed relatively low adsorption capacities for PFOS, ranging from 4 to 5 mmol/g (Dixit et al. 2021; Gao et al. 2017). IRA67 showed a poor adsorption capacity of 4–5 mmol/g for PFOS (Dixit et al. 2021). In contrast, IRA958, which incorporates quaternary ammonium functional groups onto a polyacrylic macro-porous matrix, achieved significantly higher uptake capacities, reportedly up to 5251 mmol/g for PFOS. Several other commercial IX resins have also demonstrated efficient PFAS removal, particularly at short empty bed contact times, highlighting their potential for practical applications. Despite their effectiveness, the long-term sustainability of IX applications for PFAS removal will largely depend on advances in regeneration techniques, cost reduction strategies, and the development of guidelines for managing the PFAS-enriched waste stream. Some reports have estimated the cost of using IX combined with incineration to treat PFAS-contaminated leachate. Over 20 years, the estimated cost of treating ten PFAS at a concentration of 794 ng/L was approximately $ 2.4 million (Ling 2024). Energy consumption is also relative to the type of treatment that is combined with the IX (Co 2023).

Membrane technology

Membrane treatment studies aimed to concentrate long-chain and short-chain PFAS and remove them from water and wastewater streams (Taher et al. 2024). The primary processes regulating the flow and removal of PFAS across membranes were pore size exclusion, hydrophobic, and electrostatic mechanisms, with pore size exclusion being the primary controlling mechanism (Taher et al. 2024). RO possessing a size of < 1 nm can act as an effective barrier to PFAS with molecular weights larger than their pore sizes (Taher et al. 2024). Removal of PFAS using reverse osmosis (RO) was first attempted in a study using four commercial polyamide RO membranes (ESPA3, LFC3, BW30, and SG) (Mastropietro et al. 2021; Tang et al. 2006). A removal efficiency of 99% PFOA, ranging from 0.5 to 1600 ppm from water, was achieved (Mastropietro et al. 2021; Tang et al. 2006). Recently published work on the RO process demonstrated its efficiency in removing 90–99% of long-chain PFOA and PFOS, as well as 60–80% of short-chain PFBS (Johnson et al. 2022a, b). Other recent studies have also confirmed the feasibility of this approach in targeting and removing PFAS (Ma et al. 2024). Similarly, nanofiltration (NF) membranes with a pore size of 1–10 nm have been reported to exhibit high removal performance over a range of PFAS (Ma et al. 2024). Comparative studies have been conducted on RO and NF, showing comparable PFAS removal rates (Ma et al. 2024; Mastropietro et al. 2021). Several operational and environmental factors can influence the effectiveness of membrane-based PFAS treatment systems. Key factors, including PFAS physicochemical properties, water matrix composition, PFAS type, and operational conditions (pH, temperature, and mixing rate), govern membrane removal performance (Liu et al. 2022a, b). Multivalent cations (Mg²⁺, Ca²⁺) and pH adjustments can boost rejection by enhancing bridging and electrostatic repulsion, while surface coatings (polyelectrolyte (poly diallyldimethylammonium) chloride, PDADMAC) improve short-chain PFAS capture (Taher et al. 2024; Zhi et al. 2025). This layer-by-layer assembled NF membrane enhanced the removal of short-chain PFBA-PFHpA, PFBS, PFMOPrA, PFMOBA, and GenX by 86.1–98.1% (Zhi et al. 2025).

However, NF/RO membranes suffer from fouling, scaling, and retentate‐management issues necessitating backflushing, cleaning, and integration with post-treatment systems such as magnetic activated carbon (MAC) adsorption, catalytic defluorination, direct flow filtration or coagulation to degrade and/or mineralize concentrated PFAS streams (see Fig. 5) (Fennell et al. 2024; Sim et al. 2024). Although membrane treatment alone costs roughly $0.28–1.13/m³, adding retentate treatment can increase expenses to $ 13.10/m³ (Taher et al. 2024). RO is ranked 1–2 on a relative energy scale, indicating high energy demand (Co 2023).

Fig. 5.

Degradation processes in combination with high-pressure membrane technologies. 1) Membrane filtration was used with (2 adsorption (Sim et al. 2024); with 3) Oxidative degradation (Fennell et al. 2024); with 4) Direct flow filtration or with 5) Coagulation used as an additional filtration step after 2) (Prajapati et al. 2025)

Other emerging physical treatments

Nanomaterial-mediated treatments have shown promise for PFAS removal from water through multi-stage mechanisms. These typically start with adsorption, which is often coupled with destructive technologies such as photocatalysis, as detailed in the hybrid treatment Sects. "Physicochemical treatment" and "Biochemical treatment". Recent studies have tested a variety of emerging nanomaterial-based catalysts—including nanosensors, nanomembranes, nano-metal oxides, and hybrid nano-metal oxides—for the treatment of PFOA and other PFAS (see Sect. "Photocatalytic-based treatment") (Fernandes et al. 2025; Manayil Parambil et al. 2025). In parallel, foam fractionation and ozone fractionation are environmentally benign separation techniques that rely on concentrating PFAS at the air–liquid interface, where the stabilization of the bubble film produces a foam layer. This foam layer can significantly concentrate PFAS, with commercial ozone-fractionation systems reporting removal efficiencies exceeding 99% (Hussain et al. 2025). However, short-chain PFAS are less efficiently removed by this technique (Malovanyy et al. 2025; Smith et al. 2023a, b).

Chemical treatment

Chemical treatments are powerful, destructive remediation methods for PFAS that can break them down into shorter-chain compounds or fully mineralize them into harmless byproducts. However, their broader application has been limited by high costs and the need for potentially toxic additives. These limitations are further discussed in the following sections (Sects. "Photolysis" and "Thermal treatment (ThT)").

Photolysis

Previous and current studies have examined the direct UV light irradiation of PFAS, leading to significant degradation and partial defluorination (Xin et al. 2023). Direct photolysis is driven by UV light, where target molecules absorb photons, typically at wavelengths below 320 nm, resulting in the formation of electronically excited reactive species (Verma et al. 2024). Early photolysis studies employing a xenon-mercury lamp of 220–460 nm proposed that PFAS degradation proceeds via decarboxylation of the carboxylate group, forming radicals that may then undergo defluorination or HF elimination (Eqs. 1–4) (Hori et al. 2004). A recent study by Xin et al. tested a series of PFCAs (C3–C10), FTUCAs, and GenX under far UVC irradiation (222 nm) and found that photolysis primarily cleaved the C–C bond in the carboxylate group, generating carboxylate radicals (^•COO^–) and perfluoroalkyl radicals (C_nF_2n+1•) (Eq. 1) (Xin et al. 2023). These radicals could further hydrate to form C_nF_2n+1OH (Eq. 2); however, the following Eqs. 3 and 4 were unlikely, as subsequent cleavage of C–F bonds and HF elimination may not occur, indicating that complete defluorination was limited under these conditions (Xin et al. 2023).

$$C_n{F}_{2n+1}CO{O}^{\bullet }\to C_n{F}_{2n+1}^{\bullet }{+}^{\bullet }CO{O}^{-}$$ 1

$$C_n{F}_{2n+1}^{\bullet }+H_{2}O\to C_n{F}_{2n+1}OH+H^{+}$$ 2

$$C_n{F}_{2n+1}OH\to C_{n-1}{F}_{2n-1}COF+H^{+}+F^{-}$$ 3

$$C_{n-1}{F}_{2n-1}COF+H_{2}O\to C_{n-1}{F}_{2n-1}CO{O}^{-}+2H^{+}+F^{-}$$ 4

Three main PFAS classes—8:2 FTUC, PFCAs, and GenX—were studied, with 8:2 FTUC showing the highest decay rate (81%). The primary degradation products identified were PFHpA, PFHxA, PFPeA, and PFBA from PFOA, and 8:2 FTUC. In contrast, GenX mainly degraded to FTA and PFPrA. While the study successfully reported high degradation of these PFAS, only 31% defluorination was achieved. It was suggested that a combination of oxidants or reductants with UV can be applied to enhance the defluorination rate (Verma et al. 2024; Xin et al. 2023). Operational conditions, such as UV light intensity (UVA, UVB or UVC), temperature, pH, or salt type, may significantly influence the degradation efficiency of these systems. For instance, increasing the temperature during PFOS photoreduction from 78 °C to 102 °C enhanced removal rates from 20% to more than 80%, as more hydrated electrons accumulated in the gas phase, increasing the frequency of direct collisions with PFOS molecules (Lyu et al. 2015; Verma et al. 2024). While some treatments show promising results for a few long-chain PFAS, their effectiveness against a broader range, especially ultrashort-chain congeners, remains uncertain (Verma et al. 2024). Limited cost analyses suggest that photochemical treatments may be more energy- and cost-intensive than physical methods, such as GAC or IX, primarily due to their reliance on high-energy UVC sources. Combining light with conventional methods has been explored to enhance efficiency and reduce costs (Feijoo et al. 2023).

Ultrasonication (US)

Benefiting from the role of sonochemistry in treatment studies, recent efforts documented successful attempts in PFAS removal (Rokhsar Talabazar et al. 2024; Yadav et al. 2022). Cavitation, generated by the build-up of high pressure and temperature from applying high-frequency ultrasound in aqueous solutions, produces unstable bubbles that eventually collapse, forming localized hotspots reaching approximately 5000 °C and 500 bar. At these hotspots, PFAS molecules can accumulate at the air–water interface and undergo pyrolytic degradation (Fang et al. 2024). PFOA has been studied using a combination of oxidants and sonication technology, and Eqs. 5–8 were proposed as the mechanisms describing the degradation processes (Cao et al. 2020). A study shows that the use of persulfate, a well-known and commonly used oxidant for advanced oxidation processes (AOPs), resulted in a 99% degradation of PFOA. In Eq. 1, bubbles were formed by the generated low-frequency energy, resulting in cavitation. The cavitated bubbles then result in the generation of hydroxyl and hydrogen radicals as depicted in Eq. 2. The hydroxyl radical is generated and then reacts with the persulfate as the bubble and cavitation continue to generate hydroxyl ions and sulfate radicals (Eq. 3). The hydroxyl radical reacts with the persulfate due to its well-known high redox potential, indicating a high affinity for accepting electrons. Finally, the PFOA reacts with sulfate radical, leading to the generation of mineralized products of F⁻, SO₄²⁻, and CO₂ (Eq. 4).

$$H_{2}O)))\to Cavitationbublles$$ 5

$$H_{2}O)))\to O{H}^{\bullet }+H^{\bullet }$$ 6

$$O{H}^{\bullet }+S{O}_4^{2-})))\to O{H}^{-}+S{O}_4^{\bullet -}$$ 7

$$PFOA+S{O}_4^{\bullet -}\to F^{-}+S{O}_4^{2-}+C{O}_2$$ 8

Other oxidants used in conjunction with US, apart from persulfate, are periodate (Lee et al. 2016), salt ions (Phan Thi et al. 2014), surfactant (Lin et al. 2016), and permanganate (Hu et al. 2018). Operational factors like pH, temperature, additives, and gases also affect cavitation intensity, with higher temperature and pH showing lower degradation efficiencies (Panda et al. 2025). Despite these advances, treatment costs and energy consumption remain high, specifically at larger-scale applications; for example, a pilot-scale acoustic reactor system required over $19 m⁻³ using 12 kW transducers (Gole et al. 2018).

Thermal treatment (ThT)

Thermal degradation (ThT) is widely explored as a destructive treatment for PFAS-contaminated residuals such as sludge, retentate, and spent adsorbents generated by GAC and IX systems (Zgonc et al. 2023). Techniques, including incineration, smouldering, supercritical water oxidation (SCWO), plasma processes, and hydrothermal alkaline treatment (HALT), have been tested, with some, such as incineration and SCWO, already reaching field-scale applications (Chen et al. 2025). During thermal treatment, at high temperatures (typically 500–1000 °C), PFAS undergo thermal bond scission of the strong carbon–fluorine bonds, leading to defluorination and mineralization into inorganic fluoride ions (Hao et al. 2021). Under basic conditions, the mechanism likely proceeds through nucleophilic substitution, where OH⁻ replaces F⁻ on the PFAS carbon chain, forming hydroxylated intermediates. These unstable intermediates then undergo C–C bond cleavage, releasing CO₂, which subsequently reacts with NaOH to form HCO₃⁻ or CO₃²⁻. While this process can achieve nearly complete breakdown, low-molecular fluorinated by-products may still form, and treatment in the presence of materials like GAC can lead to the generation of additional or unknown by-products (Wang et al. 2024; Weitz et al. 2024). Advances such as HALT have achieved over 90% degradation of various PFAS classes from AFFF sources at moderate temperatures (150–250 °C) (Austin et al. 2024; Hao et al. 2021). The use of alkaline additives, such as Ca(OH)₂, can further increase PFOS removal rates from approximately 49% to as high as 98% (Abou-Khalil et al. 2024). Smouldering, though less extensively studied, has been enhanced by adding GAC as a fuel source, where flameless burning causes a self-sustaining exothermic reaction that generates heat to drive the PFAS degradation process (Fournie et al. 2023). SCWO has been recognized by the US EPA for its high efficiency (> 99% removal) at lower operational temperatures (~ 650 °C) in field demonstration studies (Rosansky et al. 2024). SCWO is driven by the ability of an oxidation reaction to mineralize the organic compounds under temperatures and pressures beyond the supercritical point of water (Wei et al. 2021). It has been demonstrated by different groups that SCWO is an economical technology that can target highly concentrated streams (Zhang et al. 2020). Plasma treatment has similarly demonstrated effectiveness at lower temperatures (Mbanugo et al. 2025; Topolovec et al. 2024; Zhang et al. 2024a, b). Nonetheless, these treatments may generate potentially toxic gaseous and solid by-products, and costs remain a significant challenge for large-scale implementation. For instance, combining IX with incineration to treat landfill leachate containing approximately 794 ng/L of 11 PFAS compounds was estimated to cost around $2.3 million over a 20-year period (Ling 2024). While the energy consumption varies depending on the type of thermal treatment, it is generally considered moderate to low and arbitrarily assigned a relative scale of 2–3 (Co 2023).

Biological treatment

Bioremediation of PFAS-contaminated water has emerged as a more sustainable alternative, owing to its eco-friendly nature, low chemical demand, and reduced operational costs (Zhang et al. 2022). Several microbial species and phytoremediation have been tested for treating PFAS-contaminated wastewater (Zhang et al. 2022). However, biodegradation has generally not proven to be an effective treatment method for PFAS transformation, mainly due to the long treatment durations required, often ranging from several days to weeks (D’Agostino & Mabury 2017; Huang & Jaffé, 2019; Kim et al. 2014; Zhang et al. 2022). The following Sects. ("Microbial transformations", "Phytoremediation", and "Enzymatic transformations"4.3.1–4.3.3) discuss specific examples and highlight the key limitations of current PFAS bioremediation strategies.

Microbial transformations

More often reported under aerobic circumstances, bacterial species isolated from sludge, soil, or sediments were utilized for PFAS transformation (Zhang et al. 2022). Biological degradation of PFAS remains limited, as natural processes are generally unable to effectively cleave the strong C–F bond (Fang et al. 2024). For example, aerobic biotransformation by Pseudomonas YAB-1 achieved partial removal of PFOA without releasing fluoride. Some anaerobic systems show better defluorination; Acidimicrobium sp. strain A6 reached 67.6% defluorination over 18 days, producing fluoride and shorter-chain PFAS (Ruiz-Urigüen et al. 2022). It is hypothesized that A6 partially defluorinates PFOA, followed by further chain shortening by heterotrophic bacteria. Additional studies have investigated the biodegradation of FTOHs, FTSAs, and FTCAs, indicating different extents of fluoride release (Li et al. 2018; Wu et al. 2024). Overall effectiveness depends on factors like PFAS concentration, redox conditions, pH, temperature, and microbial community composition (Zeeshan et al. 2025). Recently, enzymes such as Fluoroacetate Dehalogenase (FAcD) have shown promise in C–F bond cleavage (Harris et al. 2024). While some pilot-scale and sludge-based applications have been tested (Gottardo et al., 2023; Nordhagen, 2022), biological methods alone remain insufficient, often requiring integration with other technologies, raising costs and energy demands (Gottardo et al. 2023; Nordhagen 2022). For instance, combining microbial treatment with GAC and incineration to manage wastewater PFAS contamination over 20 years could cost around $41 million (Ling 2024).

Phytoremediation

Phytoremediation is another natural and environmentally safe method for in-situ removal of PFAS from groundwater and soil, or reducing their concentrations (Savvidou et al. 2024). Due to the bioaccumulative nature of PFAS in plant tissues, aquatic and terrestrial plants have been explored as a potential medium for PFAS removal (Mayakaduwage et al. 2022; Savvidou et al. 2024). Analysis of both aquatic and terrestrial species reveals distinct uptake mechanisms: terrestrial plants rely on phytoextraction, adsorbing PFAS in roots, translocating them via vascular transport, and sequestering them in shoots, whereas aquatic species predominantly employ phytofiltration, in which root surfaces act as adsorption and precipitation sites (He et al. 2023; Mayakaduwage et al. 2022). In controlled experiments, seven common weed species achieved up to 41.4 wt % removal of short-chain PFAS (PFBA, PFPeA), selectively partitioning these contaminants into aerial biomass (He et al. 2023). Uptake efficiency correlates not only with fluorocarbon chain length but also with functional head groups, molecular size, and hydrophilicity, indicating that both physicochemical properties of the contaminant and plant physiology critically govern translocation and storage (He et al. 2023; Mayakaduwage et al. 2022; Müller et al. 2016). These variations in the plant uptake process demonstrate the adaptability of this method in breaking down both types of PFAS (Qiao et al. 2021; Savvidou et al. 2024). Nevertheless, phytoremediation faces significant operational constraints, including prolonged treatment periods (days to weeks), phytotoxic effects at elevated PFAS concentrations, complex uptake dynamics that vary by compound class, and the downstream challenge of safely handling contaminated biomass (Mayakaduwage et al. 2022). Pilot- and full-scale phytoremediation has been demonstrated using constructed wetlands in the US, China, and elsewhere, where systems operated over 6 years have successfully treated PFAS-spiked tap water and wastewater (Savvidou et al. 2024). However, limited cost data exist, and the high expenses of PFAS analysis with these systems hinder their large-scale implementation (Savvidou et al. 2024).

Enzymatic transformations

Extracellular enzymatic treatments have shown some capacity to degrade PFAS, although generally with limited efficiency (Colosi et al. 2009; Yang et al. 2022). A typical route has been proposed for the enzymatic treatment of PFAS, with some differences in the radical species formed (Harris et al. 2024). Enzymes can degrade PFAS by generating reactive oxygen species (ROS) that can further react with long-chain PFAS, fragmenting them into shorter-chain PFAS (Fig. S2) (Kumar et al. 2023). For example, catalyzed oxidative humification reactions (ECOHRs) laccase-mediated oxidation removed 59% of PFOS over 162 days in the presence of redox mediator, hydroxybenzotriazole (HBT) (Harris et al. 2024; Luo et al. 2018; Marciesky et al. 2023). HBT was activated by laccase, generating free radicals that targeted the PFOS’s C–C bonds, resulting in decarboxylation and formation of shorter-chain PFAA radicals, which further reacted through radical chain reaction to generate fluorinated and nonfluorinated by-products (Luo et al. 2018). Not all intermediates have been identified so far, which makes pathway elucidation incomplete (Kumar et al. 2023). More approaches are required to reveal more species and define more detailed mechanistic pathways (Kumar et al. 2023). Enzymatic PFAS removal typically achieves less than 70% PFAS transformation, with far less defluorination, likely due to slow reaction kinetics, enzyme sorption losses, and other factors related to the inherent properties of the enzyme (Luo et al. 2018). Integrating enzymes with more destructive treatment steps can accelerate PFAS breakdown and improve overall removal rates (Long et al. 2024; Zhang et al. 2022). Additionally, while enzymatic PFAS degradation shows promise at the bench scale, pilot-scale demonstrations remain scarce. Scaling up these treatments demands rigorous assessment of enzyme stability, kinetics, and reactor design, as well as downstream processing considerations. Moreover, the premium cost of highly specific biocatalysts raises significant economic hurdles, calling into question the feasibility of large-scale implementation (Harris et al. 2024).

Recent progress in combined treatment techniques

Over the past two decades, there has been a marked increase in research on combined treatment approaches aimed at enhancing individual process efficiencies and maximizing PFAS defluorination rates (Dirani et al. 2024; Lu et al. 2020; Rousis et al. 2024). As shown in Fig. 6, among different treatment methods, physicochemical treatments are the most widely studied, especially after combined treatments of similar types. Numerous studies published between 2011 and 2023 have reported on physicochemical treatment methods (see Fig. 6). In contrast, other PFAS treatment methods, such as biochemical or biophysical approaches, have generally been found less effective. Selected examples of these treatments are discussed in Sects. "Physicochemical treatment", "Biophysical treatment", and "Biochemical treatment". Table 2 shows examples of the latest combined treatments employed to treat PFAS in water surfaces.

Fig. 6.

Number of papers used in combined treatment techniques since 1981 and until 2023 (calculated based on the treatment approaches described in the manuscript). Activated carbon or membrane processes combined with either AOPs, direct photolysis, or sonication were among the physicochemical treatments found using the Google Scholar search engine. The search engine identified several popular biochemical treatment methods, including bacterial or phytoremediation treatment in conjunction with AOPs. The search for biophysical treatment involved combining physical treatments like membrane processes or activated carbon with bacterial treatment or phytoremediation

Physicochemical treatment

Photocatalytic-based treatment

The addition of photocatalysts to accelerate the rate of photolytic-based reactions has been thoroughly investigated as an effective AOP treatment to degrade PFAS in waterbodies (Verma et al. 2024). Light-sensitive materials, activated by UV or visible light with photon energy equal to or exceeding their band gap ≥ E_g (energy band gap) participate in the degradation pathways of PFAS (Eqs. 9–13) (Cao et al. 2024). The process produces perfluoroalkyl radicals and CO₂ (via holes, Eq. 11) or carboxylate cations and hydroxyl radicals (Eq. 12), ultimately yielding defluorinated products, HF, CO₂, and H⁺ (Eqs. 13 and 14).

$$h^{+}+H_{2}O\to O{H}^{\bullet }+H^{+}$$ 9

$$e^{-}+O_2\to O_2^{\bullet -}$$ 10

$$C_n{F}_{2n+1}CO{O}^{-}+h^{+}\to C_n{F}_{2n+1}^{\bullet }+C{O}_2$$ 11

$$C_n{F}_{2n+1}CO{O}^{-}+O{H}^{\bullet }\to C_n{F}_{2n+1}CO{O}^{+}+O{H}^{-}$$ 12

$$C_n{F}_{2n+1}^{\bullet }+O{H}^{\bullet }\to C_n{F}_{2n}O+HF$$ 13

$$C_n{F}_{2n+1}CO{O}^{+}\to C_n{F}_{2n+1}+C{O}_2+H^{+}$$ 14

Metal-based photocatalysts such as TiO₂ and Fe₃O₄ nanoparticles have been explored to enhance UVC-assisted PFAS degradation. However, unmodified TiO₂ alone has shown limited performance, requiring further modification to achieve practical efficiencies (Verma et al. 2024). Coupling TiO₂/WO₃ with ozonation achieved only modest removal (25%) of six PFAS, and the identities and toxicity of the resulting byproducts remain largely unknown (Lashuk et al. 2022). Metal-free photocatalysts, such as graphitic carbon nitride (g-C₃N₄, GCN), benefit from visible-light activation, potentially reducing energy demands compared to UVC-driven systems (Navidpour et al. 2024). Nonetheless, GCN generally performs less effectively than TiO₂ under UVC, as it produces fewer reactive radicals. Reaction efficiency is also highly sensitive to operational parameters, including photocatalyst dosage, pH, temperature, atmospheric conditions, and matrix composition (DiGuiseppi et al. 2024). Fig. S3 compares the PFAS degradation mechanisms mediated by TiO₂ and by GCN. Moreover, photocatalysts exhibit varying affinities for different PFAS classes, thereby constraining their broad applicability (Gar Alalm & Boffito 2022). Advancing this chemical treatment with oxidants or reductants may help increase the efficiency of the catalysts. Furthermore, scalability is limited by uncertainties in photocatalyst stability and reusability, which directly affect cost-effectiveness. In addition, safe disposal or regeneration of spent photocatalysts remains an unresolved challenge (DiGuiseppi et al. 2024).

Electrochemical-based treatment

Electrochemical (redox) processes (EC) remain widely investigated as an AOP method for treating PFAS concentrates produced by upstream physical separations due to their minimum material requirements, low waste generation, and large-scale applicability (Table 2) (Mirabediny et al. 2023; Veciana et al. 2022). The mechanistic pathways for the EC-mediated degradation of PFAS are driven by the oxidation potential of the electrodes, which assist in the generation of ROS, triggering the formation of PFAS radicals (Gar Alalm & Boffito 2022). Generally, the electrodes can oxidize PFAS molecules directly (oxidation at the electrode) or indirectly (oxidation by mediators) (Sivagami et al. 2023). One of the common degradation pathways of PFOA oxidation begins with the radicalization of the chain, converting it into perfluoroalkyl radicals (Eq. 15). The radical then undergoes Kolbe decarboxylation to form another radical (Eq. 16), which then reacts with O₂ to form perfluoro-heptyl peroxy radicals and gradually breaks them down into shorter-chain PFAS ultimately leading CO₂ and HF (Eqs. 17, 18 and 19).

$$C_7{F}_{15}CO{O}^{-}+M\to C_7{F}_{15}CO{O}^{\bullet }+e^{-}$$ 15

$$C_7{F}_{15}CO{O}^{\bullet }\to C_7{F}_{15}^{\bullet }+C{O}_2$$ 16

$$C_7{F}_{15}^{\bullet }+O_2\to C_7{F}_{15}O{O}^{\bullet }$$ 17

$$C_7{F}_{15}O{O}^{\bullet }+R^{\prime }CO{O}^{\bullet }\to C_7{F}_{15}{O}^{\bullet }+R^{\prime }C{O}^{\bullet }+O_2$$ 18

$$C_7{F}_{15}{O}^{\bullet }+\to C_6{F}_{13}{O}^{\bullet }+C{F}_2$$ 19

Large-scale electrochemical (EC) treatment faces challenges due to high operational costs, low PFAS concentrations, and low conductivity in real water systems (Veciana et al. 2022). To improve performance and scalability, EC has been combined with low-cost carbon-based methods (Trzcinski & Harada 2024). A remediation technique based on an expandable graphite intercalation compound (GIC) in conjunction with an electrochemical reaction was studied to degrade PFOA from contaminated water, achieving 99.4% PFOA removal in under 30 min, and producing PFHpA, PFHxA, and PFBA as byproducts (Trzcinski & Harada 2024). It was suggested that PFOA degradation followed the same degradation pathways stated above (Eqs. 15–19) (Trzcinski & Harada 2024). High electrocatalyst costs, susceptibility to poisoning, and operational expenses challenge large-scale EC applications, although energy demand is relatively modest, often just a few kWh/m³, compared to more energy-intensive methods like ultrasonication (Trzcinski & Harada 2024).

Promises and challenges of advanced oxidation processes (AOPs)

Many other AOPs have been reported to degrade PFAS. Figure 7 illustrates some examples of AOPs used for degrading PFAS from water, including UV-assisted ozonation, UV-assisted persulfate oxidation (PS), photocatalysis, electrochemical processes, and mechanochemical treatment. PS oxidation represents one of the most effective AOP methods, and it has been reported that the PS mechanism starts with the generation of sulfate and hydroxyl radicals, SO₄^•− and OH^•, which can be used to attack PFAS, promoting the radicalization and subsequent stepwise mechanisms (Mojiri et al. 2023). Other mechanistic pathways of the described AOPs in Fig. 7 can be found in the SI Appendix C. Although detailed mechanistic pathways for mechanochemical treatment have not yet been experimentally validated, Ateia et al. have proposed a schematic diagram outlining a potential degradation pathway (Ateia et al. 2021). The capacity to target PFAS without generating any secondary waste is a distinctive characteristic of AOP treatment methods (Mojiri et al. 2023). Numerous studies suggest that AOPs offer advantages over alternative treatments, thanks to their rapid radical-driven reactions that enable faster degradation (Babu Ponnusami et al. 2023a; Ersan et al. 2024). Some AOPs have progressed beyond bench-scale to pilot and limited full-scale trials, yet large-scale implementation is still limited (Ersan et al. 2024). Importantly, AOPs can also convert unknown PFAS into identifiable products, supporting better monitoring and targeting strategies (Ersan et al. 2024). However, challenges such as high operational costs, incomplete mineralization leading to toxic by-products, and resource-intensive designs still constrain broader implementation (Babu Ponnusami et al. 2023b). While some configurations, like H₂O₂/O₃/UV, are relatively economical (~ $0.01/L), addressing cost and scalability remains essential for broader application (Satyam & Patra 2025).

Fig. 7.

Five common advanced oxidation processes designed and employed to target PFAS. 1) Ozonation assisted by H₂O₂, 2) Fenton process assisted by electrochemical or photocatalytic process, 3) Electrochemical-based reactions, 4) Persulfate (PMS) oxidation, and 5) Tribo- or mechanochemical reaction

Biophysical treatment

PFAS treatment has also been piloted and, in some cases, implemented at a large scale through biophysical hybrid systems; however, their overall effectiveness and scalability remain limited (Lu et al. 2020). These systems may suffer from the inability of both treatments to achieve synergistic effects in PFAS removal (Lu et al. 2020). For instance, a treatment train combining reverse osmosis and foam fractionation with laccase-immobilized membranes achieved only modest degradation efficiencies of 35% for PFOA and PFOS (Broman et al. 2021). Similarly, a thermophilic anaerobic digester (AD) tested over 3 months failed to degrade PFAS alone; the addition of GAC improved average removals to 61% for PFHpA, PFOA, PFOS, and PFNA (Deligiannis et al. 2024). In another study, higher PAC dosages (300 mg/L) in plug-flow reactors reduced total PFAS from 0.675 mg/L to 0.412 mg/L over 20 days, achieving 75% and 68% removal for PFOS and PFOA, but only 32% for the short-chain PFBS (Gottardo et al. 2023). Notably, both studies highlight that PFAS degradation efficiency tends to increase with carbon chain length, underlining persistent challenges in treating short-chain PFAS. Although large-scale applications of these two separation techniques can be cost-effective and practically implementable, the key challenge lies in the long treatment times by biological processes and the need for more efficient destructive processes such as incineration. A recent theoretical estimate projected that constructing and operating a 20-year treatment facility to treat wastewater effluent containing 110 ng/L (sum of seven PFAAs) using a membrane bioreactor followed by GAC and incineration would cost approximately $13 million (Ling 2024).

Biochemical treatment

Combining chemical methods such as AOPs with biological processes has been previously explored for other contaminants, yet applications targeting PFAS are still limited and mainly experimental (Oller et al. 2011). A recent innovative study integrated constructed wetlands with microbial fuel cells (MFCs), where GAC served as electrodes to promote bio-electrochemical degradation of PFAS (Ji et al. 2023a, b). A system combining wetlands with microbial fuel cells (MFCs) was designed to test their potential in PFAS degradation (Ji et al. 2023a, b). This system, which incorporated Cyperus alternifolius and gravel layers, achieved PFAS removal through adsorption onto the GAC electrodes, plant uptake, and enhanced microbial activity. Notably, closed-circuit systems demonstrated an over 96% removal efficiency for PFOA and PFOS, approximately 10% higher than open-circuit setups, demonstrating the contribution of the electric field and the secretion of extracellular polymeric substances (EPS) by electroactive bacteria. However, PFAS toxicity still inhibited key microbial enzymes, and these systems have only been tested at the bench scale, with no reported operational costs or energy demands (Ji et al. 2023a, b). Despite their promise as greener, lower-energy alternatives to conventional physical and chemical treatments, questions remain regarding scalability, long-term stability, and real-world applicability (Ji et al. 2023a, b).

Toxicity of post-treated water and wastewater

Despite rapid advances in PFAS treatment technologies, assessments of residual toxicity in treated effluents remain limited (Table S6). Many destructive methods fragment PFAS into shorter‑chain congeners, yet often fail to achieve complete defluorination or detoxification. Notably, PFOS degradation can yield byproducts that are demonstrably more harmful than the parent compound, inducing both acute and chronic toxicity in fish, daphnids, and green algae (Furtado et al. 2021). The persistent focus on removal efficiency over ecotoxicological outcomes has left sublethal effects and the risks of transformation products largely unquantified. Emerging life-cycle assessment models underscore this gap, calling for harmonized, comprehensive toxicity evaluations that span treatment modalities and PFAS variants (Li et al. 2022; Liu et al. 2024; Wee & Aris 2023b). Figure 8a shows the treatment processes of PFAS in one study in WWTPs in China (Liu et al. 2024), and Fig. 8b presents risk‑model–predicted toxicity profiles for key degradation products (Hu et al. 2023).

Fig. 8.

The treatment processes of PFOS and PFOA in China WWTPs and the toxic DPs formed. a Processes and common toxic DPs found in their plant. b The toxicity assessment is used to evaluate their toxic effects on humans. To estimate the risk level of each PFAS, values resulting from quantitative structure–activity relationship (QSAR) modeling were used to calculate the Toxicological Priority Index (ToxPi) score (Liu et al. 2024)

Current gaps in understanding PFAS

Until now, it is unclear how PFAS are distributed over numerous sites worldwide, especially in developing nations, due to the lack of sophisticated analytical and monitoring methods (Brunn et al. 2023). Despite the substantial efforts devoted to this topic, there are still limitations that are summarized in the following points:

• Limited data exist on PFAS distribution in developing countries, and the environmental or ecological factors driving their uneven global presence remain poorly understood.

• More research is needed to develop cost-effective and environmentally friendly treatment technologies, especially those targeting short- and ultrashort-chain PFAS.

• Combining biological treatments, such as enzymatic processes, with chemical or physical methods could enhance removal efficiency, but these hybrid systems remain largely unexplored.

• A deeper mechanistic understanding of how short- and ultrashort-chain PFAS degrade under combined treatment technologies is still lacking.

• Toxicity evaluations of treated water and byproducts are insufficient, particularly regarding long-term and sublethal effects on diverse organisms; more standardized testing and modeling are needed to better assess these risks.

Emerging trends and future directions

The urgent demand for effective PFAS treatment has driven the exploration of emerging destructive technologies, including plasma, HALT, advanced reduction processes (ARPs), and electron beam irradiation, which show potential to complement or improve upon conventional approaches (DiGuiseppi et al. 2024; Soker et al. 2025; Tshangana et al. 2025; Zhang et al. 2024a, b). Other methods, such as nanobubble technology, nanomaterial-functionalized membranes, and mechanochemical treatments, are also being tested at the bench scale, particularly targeting PFAS-concentrated streams when combined with other methods (such as plasma or GAC) (Luo et al. 2025; Zhang et al. 2024a, b). However, comprehensive cost evaluations and large-scale feasibility assessments of these methods remain limited and need further exploration. Singular hybrid systems capable of both adsorption and in situ destruction present an appealing strategy for minimizing residual PFAS and operational complexity (Tshangana et al. 2025). Their potential benefits should be integrated with machine learning tools to provide real monitoring approaches. These tools can also be used to support the prediction of PFAS toxicity and bioavailability, offering tools to better assess potential ecological and human health risks of treated effluents (Yin et al. 2025).

Conclusions

Despite ongoing efforts to detect and characterize emerging PFAS across various environmental matrices, high levels and a wide range of PFAS—especially short-chain compounds—are still reported in several regions worldwide. Using specific animal models has greatly advanced our understanding of the harmful health effects linked to PFAS exposure. However, fully understanding their mechanisms of action and connecting these effects to human health remains difficult, as such models may not replicate the complex, long-term molecular responses in human cell receptors. Despite these challenges, it is essential to address the issue now by developing new treatment strategies aimed at reducing PFAS impacts. Among the methods studied, electrochemical treatments have shown significant promise in decreasing PFAS concentrations in water. At the same time, hybrid and engineered systems are being developed as potentially more efficient and affordable options. To ensure their effectiveness in real-world conditions, these treatment technologies need to be thoroughly tested through comprehensive approaches that identify degradation by-products and then evaluate the toxicity of these by-products in both synthetic and actual water samples.

Author contributions

AA prepared the original draft and presentation. OOA was the project administrator. OOA, HRH, and GA were involved in the conceptualization, reviewing, and editing. GA provided the resources and funding.

Funding

Natural Sciences and Engineering Research Council of Canada (NSERC) [grant ID 735239].

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent of publication

Not applicable.

Competing interests

The authors declare no competing interests.