1,4-Dioxane is a synthetic compound considered to be one of the major emerging pollutants in the environment. This compound was historically used as a stabiliser for chlorinated solvents in industrial processes such as 1,1,1-trichloroethane (TCA) (Adamson et al., 2014; Godri Pollitt et al., 2019; Karges et al., 2018; Mohr, 2010; USEPA, 2013). Although the use of TCA was banned in 1996 by the Montreal Protocol (Arulazhagan et al., 2013; USEPA, 2017), 1,4-dioxane is still currently used in other industrial products such as paints, antifreeze, varnishes, inks, plastics, cosmetic formulations and personal care products. 1,4-Dioxane is also present as an impurity in pesticide formulations and food contact packaging (Mohr, 2010; USEPA, 2017). It is considered to be a persistent and mobile organic compound in the aquatic environment due to its intrinsic chemical properties, i.e. high solubility in water (100 mg/mL), a low octanol-water partition coefficient (log Kow at 25°C: − 0.27) and low soil adsorption coefficient (log Koc: 0.42). In addition, the persistence of this anthropogenic compound is revealed by half-lives of 2 to 5 years in groundwater and 56 days in surface water. Once released into the environment, 1,4-dioxane can be rapidly dispersed and transported through river banks and groundwater into drinking water resources and remote aquatic systems. As a result, it can end up in drinking water, potentially posing a threat to human health (Hale et al., 2020; Kim et al., 2023; Neuwald et al., 2022).
The US Environmental Protection Agency (US EPA) and the International Agency for Research on Cancer (IARC) have classified 1,4-dioxane as “probably to be carcinogenic to humans” by all routes of exposure (IARC, 1976; USEPA, 2017). Contaminated drinking water is one of consumers’ main routes of exposure to 1,4-dioxane (Doherty et al., 2023; Godri Pollitt et al., 2019; USEPA, 2017). Indeed, this chemical is completely soluble in water and does not volatilise (McElroy et al., 2019). Conventional drinking water treatments, such as activated carbon adsorption, air stripping, membrane filtration, classical oxidation and reverse osmosis, have limited effectiveness in removing 1,4-dioxane (Carrera et al., 2019; Godri Pollitt et al., 2019). Advanced oxidation and biological treatments have been reported as promising approaches for the remediation of 1,4-dioxane from contaminated water (Kikani et al., 2022; McElroy et al., 2023; Tang, 2023). However, large-scale implementation of these treatments is costly.
Drinking water guideline values for 1,4-dioxane vary widely from country to country. Canada, Japan, the Republic of Korea and the World Health Organization have set a guideline value of 50 µg/L (Godri Pollitt et al., 2019). The US Environmental Protection Agency (US EPA) set an Integrated Risk Information System (IRIS) drinking water screening concentration of 0.35 µg/L (USEPA, 2017) and a tap water regional screening level (RSL) of 0.46 µg/L (USEPA, 2013). Both levels are calculated based on an acceptable cancer risk of one in 1 million. Risk-based concentrations vary because of differences in assumed exposure pathways and exposure frequency (Broughton et al., 2019). The German Environmental Agency has been more restrictive by suggesting a guideline value of 0.1 µg/L, although 1,4-dioxane is not considered a regulated substance (Karges et al., 2018; McElroy et al., 2019). In March 2021, Germany proposed that 1,4-dioxane be included in REACH as a “substance of very high concern”. If accepted, it would be placed on the European Union (EU) list for strict regulation (ECHA, 2021).
Analytical methods for the quantification of 1,4-dioxane in water are limited, tedious and complex to implement on a large scale in laboratories. The analysis of 1,4-dioxane in water is challenging, leading to analytical methods with high quantification limits and low recovery rates (Hayes et al., 2022; USEPA, 2017). Thus, the US and German reference levels (0.35 and 0.1 µg/L) in drinking water cannot be easily achieved (Adamson et al., 2021).
The instrumental analysis of 1,4-dioxane in water is essentially carried out by gas chromatography coupled to mass spectrometry (GC-MS). For aqueous environmental samples, different extraction techniques have been reported over the years, as reviewed by Sun et al. (2016) and McElroy et al. (2019). Liquid-liquid extraction (LLE) is suggested by EPA methods 8260 and 8270. However, the LLE technique is laborious and solvent-consuming. In addition, both standard methods have limits of quantification (LOQ) ranging from 1 to 3 µg/L (McElroy et al., 2019), which is far from the US regulatory level. Moreover, a comparison of the data obtained between US EPA method 8270 and method 8260 showed that the former method produces three times lower concentrations of 1,4-dioxane (without isotope dilution) than the latter (Chiang et al., 2016). Alternatively, US EPA standard methods 522 and 541 for the preconcentration of 1,4-dioxane in drinking water can be performed using solid-phase extraction (SPE) with charcoal adsorbents (USEPA, 2008; USEPA, 2015). This extraction is reported to be useful for achieving LOQs for 1,4-dioxane in water below 1 µg/L (Carrera et al., 2017). Previous studies have reported LOQs for this method ranging from 0.034 to 0.5 µg/L in surface and drinking water (Carrera et al., 2017; Karges et al., 2020; Stepien and Püttmann, 2013). However, SPE is also a laborious extraction technique. Alternatively, faster techniques such as purge and trap (P&T) and headspace have been used in several studies (Hayes et al., 2022). The lowest 1,4-dioxane LOQs reported were 0.15 µg/L for P&T (Sun et al., 2016) and 0.07 µg/L for headspace (Hong et al., 2014), both calculated in ultrapure water. Previous studies have also investigated the performance of solid-phase micro-extraction (SPME) for extracting 1,4-dioxane from water. In contrast to previous techniques, higher LOQs ranging from 0.5 to 1.2 µg/L have been demonstrated using different fibre coatings and incubation conditions (Nakamura and Daishima, 2005; Shirey and Linton, 2006). It has been found that SPME extraction efficiency can decrease when there are other volatile organic compounds (VOC) in the water sample that have more affinity than 1,4-dioxane, and their sorption competitively displaces 1,4-dioxane (Li et al., 2011).
The occurrence of 1,4-dioxane in the aquatic environment has been investigated in only a few countries. In the US, the third unregulated contaminant monitoring rule programme (UCMR 3) collected data on several contaminants from US public drinking water supplies (USEPA, 2020). Covering three years, the data set showed that 1,4-dioxane was present in 21% of the public water systems sampled (4,864 public water systems). Compared with other contaminants investigated in UCMR 3, the detection frequency of 1,4-dioxane was relatively high, resulting in 6.9% of observed concentrations exceeding the US EPA reference concentration of 0.35 µg/L (Adamson et al., 2017; Adamson et al., 2021). More recently, the systematic occurrence of 85 volatile compounds was investigated in aquifers feeding US public water supplies. With a limit of detection of 0.35 µg/L for 1,4-dioxane, its frequency of detection was 0.5% of the groundwater sampled (nine wells) in the California coastal basins (Bexfield et al., 2022). In Germany, groundwater contaminated with 1,4–dioxane has also been reported at sites where chlorinated solvents were previously used or produced (de Boer et al., 2022; Karges et al., 2018). The maximum observed concentration was 152 µg/L at one site. In 2017–2018, the same research group conducted a study of 1,4-dioxane in surface water and associated treated water after the implementation of mitigation measures at several sites. A decrease in 1,4-dioxane was observed, with concentrations below 10 µg/L in surface water and 1.68 µg/L in drinking water (Karges et al., 2022). In Spain, a survey conducted in 2015 showed 1,4-dioxane concentrations between 5.7 and 11.6 µg/L in groundwater from the Llobregat River. The concentrations of 1,4-dioxane observed were far from the LOQ of the analytical method at 50 ng/L in surface water (Carrera et al., 2017; Carrera et al., 2019). In China, from May 2018 to April 2019, 15 sampling sites were investigated along a river that supplies the city of Shanghai City. Surface water samples showed 100% of detection of 1,4-dioxane with a maximum concentration of 8.3 µg/L (Wang et al., 2022). In France, 1,4-dioxane surface water monitoring data are collected via the Naïades information portal (Naïades, 2023). However, depending on the laboratory in charge of the determination, LOQs ranged from 0.5 to 15 µg/L. Consequently, 1,4-dioxane was quantified in only 0.3% of the surface waters analysed. Concentrations over 20 µg/L were recorded in a river likely tainted by discharge from a known pharmaceutical plant. Given the environmental problems and the growing concern for human health posed by 1,4-dioxane and the lack of data on its occurrence in France, it is clearly necessary to assess the potential exposure of the French population to this compound in drinking water. The objectives of this study were therefore: (i) to develop and validate an analytical method for 1,4-dioxane in natural water matrices with a LOQ of 0.15 µg/L (below the US EPA guideline of 0.35 µg/L) and (ii) to carry out a sampling campaign to determine the presence of 1,4-dioxane in public water supplies throughout France.