Per and Poly-fluoroalkyl Substances (PFAS) are synthetic organofluorine compounds, characterised by C-F bonds. Due to the low molecular polarity and strong C-F bond energy, PFAS are stable compounds. Thus, they have been used in a variety of industries around the globe, including in firefighting, aerospace and vehicles, and also distributed in our daily consumer products such as food packaging, fabrics, and carpets [1]. Therefore, PFAS are prevalent in many water bodies globally, like the River Thames in the UK, especially near industrial sites. However, the solubility means that PFAS can travel into nearby drinking water sources, threatening 13 million lives.

It is impossible to understate the destructive nature of PFAS when unleashed into drinking water sources. Their impacts are demonstrably fatal in humans and animals, impairing immune function and reproductive outcomes, increasing the risk of different cancers, and prompting organ failure. Antibody concentration in young children decreased 49% for every 1 ng/mL of consumed PFAS, total concentrations of >5.7ng/mL indicated a 2.2 times increase in hypothyroid disease [2]. Contemporary research also indicates dramatic increases to the risk of childhood leukaemia (1.39 times) [3], breast cancer (1.47 times) [4], kidney cancer (1.74 times) [5], and thyroid cancer (1.56 times) [6], for every linear doubling of PFAS concentration in blood.

Perfluorooctane sulfonic acid (PFOS) and perfluorooctanoic acid (PFOA) are the most common forms of PFAS, accounting for >50% of PFAS from water samples in Hong Kong and Shenzhen [7] and >80% in Europe. They are long-chained, containing a >8 carbon aliphatic chain. Long chained PFAS exhibit greater toxicity and accumulation, where they can remain in human bodies for 4-5 years [8]. Furthermore, compared to other long chained PFAS such as PFHxS, PFOS and PFOA pose additional harms. Both the International Agency for Research on Cancer (IARC) and the US Environmental Protection Agency (EPA) have designated PFOA a potential human carcinogen. For example, PFOA exposure indicates a 35% increase in developing testicular cancer, and every increase in 1ng/mL in the mother results in a 30 g reduction in baby mass [2]. 55ng/mL of PFOS resulted in a 350% increase in hepatocellular carcinoma [9], and alanine transaminase enzymes, both indicative of liver cancer, had 3.55 times increased production at high PFOS volumes [10].

Figure 1: Molecular Structure of PFOA (left) and PFOS (right)

Despite recent advancements in trying to phase out PFAS in industry, the occurrence of PFAS in the environment is expected to increase [11]. Though both the US and EU have implemented policies to regulate PFAS production, no such effort has been made in the UK. Thus, policy and technology to tackle PFAS pollution is of utmost importance [12], especially considering that 14.3% of the population have had adverse health effects because of different PFAS [13]. The cost to Europe as a whole thus sits at £46-75 billion [14].

However, the treatment of PFAS pollution in the UK is hindered in two major ways. First, current approaches to PFAS detection are expensive and highly specialised, with very few facilities in the UK that offer the service. The average cost of concentration measuring is £110. This means that identification of PFAS hot spots – a prerequisite to addressing the issue – is challenging. Therefore, both policies for PFAS regulation and the distribution of PFAS-remediating technology suffers in efficiency. Liberalising access to geospatial data is a struggle that must be overcome. Second, emerging technologies towards PFAS are unclear in their efficacy. Emerging filtration technologies, namely nanofiltration, ion exchange and reverse osmosis, are often limited in efficacy by fouling [15]. Implementation of Reverse Osmosis plants can cost $99 million; single use resins cost $44,164 [16]. Furthermore, they are most suitable for industrial applications, and are rather inaccessible to home settings. Though literature identifies Granular Activated Carbon (GAC) as a potential treatment mechanism, its prominence and efficacy as point-of-use filters are undermined by a lack of research. Specifically, GAC comes in a variety of sizes; there is no available research on the impact of those sizes on the suitability of GAC within such filters and when those filters need to be replaced.

Thus, we present here a novel approach to tackling PFAS contamination in the Thames Valley region: a concurrent detection and filtration mechanism at an affordable price. Detection is achieved with a geospatial neural network and filtration optimised through analysing the size and depth of GAC. We consider an effective geospatial representation of regional PFAS concentration to be one that corresponds to any known values, and allows any user to access data easily. Accordingly, we have designed a random forest-KNN ensemble that presents data continuously and interactively. The model has been verified by our own empirical analysis of PFAS concentration at locations along the River Thames, demonstrating concordant results, thus indicating our model’s success.

For any point-of-use system design, we identified several important considerations, beyond the efficacy of the filter. First, flow rate must be carefully monitored, both to have minimal impact on customer use and to prevent backwash into the system, clogging the pipes. Second, the filtration does not introduce additional contaminants. Third, the lifespan of the filter is crucial; it provides information on the churn rate and ensures sufficient filtration efficacy is maximised. Given these parameters, we designed a low-cost, effective, and long-lasting GAC Filter.

[1]  CHEM Trust UK. The “Forever Chemicals”. Sept. 2019. url: (visited on 12/05/2024).

[2]  Suzanne E. Fenton et al. “Per- and Polyfluoroalkyl Substance Toxicity and Human Health Review: Current State of Knowledge and Strategies for Informing Future Research”. In: Environmental Toxicology and Chemistry 40.1 (2021). First published: 05 October 2020, pp. 113–133. doi: 10.1002/etc.4890. url:

[3]  RR Jones et al. “Maternal serum concentrations of per-and polyfluoroalkyl substances and childhood acute lymphoblastic leukemia”. In: Journal of the National Cancer Institute 116.5 (May 2024), pp. 728–736. doi: 10.1093/jnci/djad261.

[4]  X Li et al. “Perfluoroalkyl substances (PFASs) as risk factors for breast cancer: a case-control study in Chinese population”. In: Environmental Health 21.1 (Sept. 2022), p. 83. doi: 10.1186/s12940-022-00895-3.

[5]  MS Seyyedsalehi and P Boffetta. “Per-and Poly-fluoroalkyl Substances (PFAS) Exposure and Risk of Kidney, Liver, and Testicular Cancers: A Systematic Review and Meta-Analysis”. In: Medicina del Lavoro 114.5 (Oct. 2023), e2023040. doi: 10.23749/mdl. v114i5.15065.

[6]  Li Jing and Zhixiong Shi. “Per- and polyfluoroalkyl substances (PFAS) exposure might be a risk factor for thyroid cancer”. In: EBioMedicine 72 (Oct. 2023). Open Access, p. 104866. doi: 10.1016/j.ebiom.2023.104866. url: ebiom.2023.104866.

[7]  Ziyad Abunada, Motasem YD Alazaiza, and Mohammed JK Bashir. “An overview of per-and polyfluoroalkyl substances (PFAS) in the environment: Source, fate, risk and regulations”. In: Water 12.12 (2020), p. 3590.

[8]  Fan Li et al. “Short-chain per-and polyfluoroalkyl substances in aquatic systems: Occurrence, impacts and treatment”. In: Chemical Engineering Journal 380 (2020), p. 122506.

[9]  JA Goodrich et al. “Exposure to perfluoroalkyl substances and risk of hepatocellular carcinoma in a multiethnic cohort”. In: JHEP Reports 4.10 (Aug. 2022), p. 100550. doi: 10.1016/j.jhepr.2022.100550.

[10]  Elizabeth Costello et al. “Exposure to per- and Polyfluoroalkyl Substances and Markers of Liver Injury: A Systematic Review and Meta-Analysis”. In: Environmental Health Perspectives 130.4 (Apr. 2022), p. 046001. doi: 10.1289/EHP1009. url: https://doi. org/10.1289/EHP1009.

[11]  url: https:/ /

[12]  Emiliano Panieri et al. “PFAS molecules: a major concern for the human health and the environment”. In: Toxics 10.2 (2022), p. 44.

[13]  Risks of PFAS for human health in Europe (Signal). Published 16 Apr 2024, Modified 17 Apr 2024. European Environment Agency. Apr. 2024. url:

[14]  Alissa Cordner et al. “The True Cost of PFAS and the Benefits of Acting Now”. In: Environmental Science & Technology 55.14 (July 2021). Erratum in: Environ Sci Technol. 2021 Sep 21;55(18):12739, pp. 9630–9633. doi: 10.1021/acs.est.1c03565.

[15]  Nour AlSawaftah et al. “A comprehensive review on membrane fouling: Mathematical modelling, prediction, diagnosis, and mitigation”. In: Water 13.9 (2021), p. 1327.

[16]  Anderson C Ellis et al. “Life cycle assessment and life cycle cost analysis of anion exchange and granular activated carbon systems for remediation of groundwater contaminated by per-and polyfluoroalkyl substances (PFASs)”. In: Water Research 243 (2023), p. 120324.

(Visited 60 times, 1 visits today)