Background

Persistent, mobile and toxic (PMT) and very persistent and very mobile (vPvM) substances

For a chemical substance emitted into the environment to pose a threat to drinking water resources, it must be transported from the point of emission through natural barriers such as soils, riverbanks and aquifers, and in some cases through artificial barriers [1]. The time scales for this can vary from days, to weeks, to months and to years. Important factors controlling the overall threat are the scale of environmental emissions and whether the substance, or its transformation products, are sufficiently persistent in the environment and enough mobile in the aquatic environment to survive such a journey.

It is, therefore, clear that substances that are persistent in the environment, mobile in the aquatic environment as well as being toxic (PMT) or substances that are very persistent in the environment and very mobile in the aquatic environment (vPvM) have specific combinations of intrinsic substance properties that cause them to pose an inherent hazard to drinking water resources [2]. A persistent substance can be defined as one that remains in the same substance form in the environment over long periods of time. For regulatory standards, persistency is most often described as an inherent substance property: a degradation half-life in a given environmental media (e.g., air, water, soil, and/or sediment) under a specified set of environmental conditions (e.g., 12 °C [3]). A mobile compound is one that sorbs poorly to sediments and soils, resulting in a high potential to move and be transported through river banks, ground water and to drinking water extraction wells and remote aquatic ecosystems. Toxicity is a central consideration in chemical hazard and risk assessment and encompasses various modes of toxic action. It is most commonly described using hazard classification criteria and/or reported adverse human effects (e.g., carcinogenic, germ cell mutagenic, toxic for reproduction, etc.) [4].

There are many recent research developments and monitoring studies that have bought persistent and mobile substances to the forefront of worldwide scientific discussion [5,6,7,8]. As early as the 1990s, the scientific community realized the real hazard posed by persistent chemicals that are mobile in the aquatic environment [9,10,11]. These substance have previously gone by different acronyms: polar persistent pollutants (P3), polar persistent organic pollutants (polar-POPs) [12] and persistent, mobile organic contaminants (PMOC) [11], but all terms describe the problematic intrinsic fate and hazard properties of persistence in combination with mobility. Neumann et al., [13] first presented the names persistent, mobile and toxic substances and very persistent and very mobile substances and the corresponding acronyms PMT and vPvM in 2015. Since then, the German Environment Agency has led efforts to establish criteria to identify PMT and vPvM substances under the EU chemicals regulation REACH (EC 1907/2006 Registration, Evaluation, Authorisation and Restriction of Chemicals) [2]. The similarity with the acronyms persistent, bioaccumulative and toxic (PBT) and very persistent and very bioaccumulative (vPvB) is not accidental. It was designed to indicate from the outset the equivalency in the hazardous nature of these two sets of substances [2] and the necessity of the adoption of a hazard based regulation.

Many PMT/vPvM substances can breakthrough artificial barriers in waste water treatment plants (WWTP) [14] and drinking water treatment facilities, including through granular activated carbon (GAC) filtration, ultrafiltration, advanced oxidation processes (like ozonation) and reverse osmosis [15]. Stackelberg et al. [16] reported that even after clarification, disinfection (chlorination) and GAC filtration, many mobile substances were not effectively removed from WWTP. In their study, the removal of four mobile substances: N,N-Diethyl-meta-toluamide, nonylphenol, camphor and bisphenol A varied between 25 and 76%. Substances that are even more persistent and mobile than these, would, by definition, more easily survive disinfection and breakthrough GAC filters in fewer bed volumes. This implies that in cases, where there are both ongoing emissions of PMT/vPvM substances and incomplete removal during water treatment, environmental concentrations will increase over time as these substances circulate in the water cycle [17] and potentially become irreversible [18, 19].

The unknown extent of PMT/vPvM substances in our water resources

The EU drinking water directive (98/83/EC, amended 2015/1787) has the objective "to protect human health from the adverse effects of any contamination of water […] by ensuring that it is wholesome and clean". The EU's groundwater directive (2006/118/EC) states, "groundwater is a valuable natural resource and as such should be protected from […] chemical pollution". Moreover, the EU's water framework directive (2000/60/EC) states that "member States shall ensure the necessary protection for the bodies of water identified with the aim of avoiding deterioration in their quality to reduce the level of purification treatment required in the production of drinking water". Thus, ensuring that the sources of our drinking water are secure from any threats caused by chemicals is of the utmost importance.

The wider environmental problem related to PMT/vPvM substances is likely larger than it appears at first glance. There are many examples of substances that are ubiquitously detected in the water cycle that can be considered as PMT/vPvM substances [1], such as melamine [20], sulfanilic acid [21] and trifluoroacetic acid [22], but owing to their differences in terms of structure and sources, they are not conceptually linked in this way. An important concern for the most mobile substances are the so-called "analytical gap" and "monitoring gap" that exist. Current analytical measurements for the most mobile substances are not widely available (the "analytical gap") and thus many go unnoticed and undetected in the water cycle ("the monitoring gap"). However, progress has been made to address this analytical gap and methods such as hydrophilic interaction liquid chromatography [23], supercritical fluid chromatography [24], mixed-mode liquid chromatography [25] and capillary electrophoresis [26] have now made it possible to analyse some of the most mobile substances. This progress also contributes to closing the monitoring gap as more of the most mobile substances can be detected. However, a recent series of studies has demonstrated that the monitoring gap is still very prevalent. In the studies, the first ranked industrial substances based on their properties of persistency and mobility, [27] second selected some of the top 57 ranked PM substances based on high emissions likelihood and for which analytical methods were rare but could be developed, and third conducted monitoring in surface and groundwater samples throughout Europe [28]. Out of these 57 substances chosen based on these considerations, 43 PM substances were detected in environmental waters, 23 of which had never been reported previously (including near ubiquitous ones, like methyl sulfate, 2-acrylamino-2-methylpropane sulfonate, benzyltrimethylammonium, benzyldimethylamine, trifluoromethanesulfonic acid, and 1,3-di-o-tolylguanidine [28]). These studies both directly demonstrated the ubiquity of PM substances and that the "monitoring gap" is a real issue. As a consequence of these two gaps the most mobile substances in the aquatic environment remain undetected, unmonitored and consequently unregulated [23, 28]. It becomes clear that a more holistic approach is needed to protect water quality and monitor, assess and manage chemical pollution of European surface waters [29].

Recent regulatory advances of PMT/vPvM substances in Europe

In the 2017 report, "Study for the strategy for a non-toxic environment of the 7th Environment Action Programme", prepared by Directorate-General for Environment (European Commission) strong concerns were raised about persistent substances that are hydrophilic, and thus mobile in water, meaning they pose particular threats to the quality of water resources [30]. The report emphasized that mobility could be considered of equivalent concern to bioaccumulation. Regulatory advances for PMT/vPvM substances have also been taken into REACH. REACH aims to ensure a high level of protection of human health and the environment" (Article 1,1) and is "underpinned by the precautionary principle" (Article 1,3). REACH provides a legislative basis to investigate hazardous properties of chemicals before market [31, 32] Through REACH, it becomes the responsibility of registrants to characterize the intrinsic hazard of substances and the risk of each of their uses over the complete life cycle. Based on this input, the European Member States and the European Chemicals Agency (ECHA) can then assess whether registered substances fulfil the criteria to be identified as substances of very high concern (SVHC) following Article 57 of REACH. Listing a substance as a SVHC is one possible pathway for authorisation or restriction under REACH.

In 2019, two substances were identified as SVHC based on their equivalent level of concern (ELoC) to persistent, bioaccumulative and toxic and/or very persistent and very bioaccumulative (PBT/vPvB) substances. One of these substances, perfluorobutane sulfonic acid (and its salts, PFBS) was identified as a SVHC owing to "very high persistence, high mobility in water and soil, high potential for long-range transport, and difficulty of remediation and water purification as well as moderate bioaccumulation in humans". These properties combined led the Member State Committee (MSC) under REACH to conclude that there is a very high potential for irreversible effects [33, 34]. The second substance HFPO-DA, commonly referred to as Gen-X, was acknowledged by the MSC to have a high potential to cause effects on wildlife and humans due to its very high persistence, mobility in water, potential for long-range transport, accumulation in plants and previously observed effects on human health and the environment [35]. In this case, the MSC unanimously agreed that in isolation, these factors are not enough to give rise to ELoC, however, that in combination, they show that there is scientific evidence of probable serious effects to the environment and humans. The same year, the organization ChemSec published an update of the SIN list (Substitute It Now list) and added a new category for PMT/vPvM substances [36]. This list is designed to flag substances they encourage industry to substitute for others and now includes 16 PMT/vPvM substances.

Methods

Based on the aforementioned identification of these two PMT/vPvM substances, this work investigated whether a general case could be made to show that all PMT/vPvM substances can be considered as ELoC to PBT/vPvB substances and as such identified as SHVC under REACH. To achieve this, categories were developed based on previous recommendations to identify SVHC under the 'equivalent level of concern' route (for example as has been done for skin sensitizers [37] and neurotoxicants and immunotoxicants [38]).

These recommendations introduce four human health effect categories (i. possible serious health effects, ii. irreversibility of health effects, iii. delay of health effects, and iv. quality of life impacted), as well as two other effect factors (i. societal concern, and ii. is derivation of a safe concentration possible). To compare the ELoC of PMT/vPvM substances to PBT/vPvB substances, the assessment categories for human health were used directly; however, the other factor categories were altered and new categories were added. The other factors categories were changed to the following: (i. increased societal costs, (ii) negative effect on resources, and (iii) do emissions need to be minimized. Nine "environmental effect" categories were developed as follows: (i) irreversible exposure, (ii) irreversible effect, (iii) intergenerational exposure and effect, (iv) unknown/uncertain spatial scale, (v) disparity between point of release and point of effect, (vi) unknown/uncertain temporal scale, (vii) uncertain/difficult to predict long term fate and toxic effects, (viii) harmful to the aquatic environment, and, (ix) potential to reach remote pristine areas.

The development of these 16 categories covering health effects (4 categories), environment effects (9 categories) and other effects (3 categories) was aided by discussions and written consultations with a broad group of stakeholders during scientific and regulatory meetings and workshops. The following were of significance for the dialogue: the 16th meeting of ECHA’s PBT expert group in September 2017, the Risk Management Expert Meeting in October 2017, the Society of Environmental Toxicology and Chemistry Europe Conferences in 2015, 2016, 2017, 2018 and 2020 and several workshops organised by the German Environmental Agency.

These 16 categories were then applied to three case study substances: PFBS, HFPO-DA and 1,4-dioxane. The first two substances were chosen due to their recent identification as SHVC and 1,4-dioxane owing to its listing in the public activities coordination tool (PACT) in 2019 for regulation under REACH. Following this, a general conceptual comparison using the defined categories was carried out to assess the ELoC between PMT/vPvM substances and PBT/vPvB substances.

Results and discussion

Case study I: Perfluorobutanesulfonic acid (PFBS) and its salts

PFBS belongs to the compound class of per- and polyfluoroalkyl substances (PFAS) that have a hydrophobic, alkylated, fluorine-saturated carbon-chain with a hydrophilic head attached at a terminal end [39]. A recent report has stated that over 4600 CAS registry numbers are associated with PFAS that may have been on the global market, including many that have at least one perfluoroalkyl moiety but were not commonly recognised as PFASs previously [40]. The use of PFAS has become more stringent in recent years, and currently C11-14 perfluoroalkyl carboxylic acids (PFCAs), perfluorooctanoic acid (PFOA), perfluorononanoic acid (PFNA), perfluorohexane sulfonic acid and its salts (PFHxS) are all under some form of regulation. As a result of this, long chain PFAS have been replaced with short chain PFAS. Short chain PFAS [41] are less bioaccumulative, but their high persistence, aquatic mobility [42] and unknown ecotoxicological effects renders them of concern. PFBS contains 4 carbon atoms and was introduced to market following the phase out of perfluorooctanesulfonic acid (PFOS) [43]. In water, PFBS is a strong acid which is predominantly dissociated as an anion.

In December 2019, PFBS and its salts were identified as SVHC in accordance with Article 57(f) based on scientific evidence of probable serious effects to the environment and human health which give rise to an ELoC to substances listed in Articles 57 (d) and (e) of REACH. Table 1 presents a summary of the argumentation agreed upon by the MSC [33, 34] to show that PFBS and its salts present an ELoC to PBT/vPvB substances.

Table 1 ELoC of PFBS (IUPAC name (1,1,2,2,3,3,4,4,4-Nonafluorobutane-1-sulfonic acid) compared to PBT/vPvB substances
Full size table

REACH dossiers were used to prepare the table, and literature references to support statements can be found within these dossiers [33, 34]. Both the neutral and potassium salt forms have been registered under REACH (EC numbers 206–793-1 and 249–616-3, respectively).

Case study II: HFPO-DA (3: 2,3,3,3-tetrafluoro-2-(heptafluoropropoxy)propanoic acid) its salts and its acyl halides (tradename GenX)

HFPO-DA also belongs to the PFAS compound class and is used as a processing aid for producing fluoropolymers with many applications, such as fluoropolymer resins, wire cables and coatings. The production history and volumes of HFPO-DA are unknown, although it has been suggested that production began in 2009 [45]. HFPO-DA was first detected in environmental samples from the Cape Fear River, North Carolina in 2018, likely from a point source release from a production plant [46]. HFPO-DA has also been detected in other countries in relation to known local production and use. HFPO-DA has been detected in the Xiaoqing River System in China [47], in America [48], in the Lower Rhine River [47] and in rivers in close proximity to a production plant in the Netherlands [49].

QSAR modelling and ready or inherent biodegradability tests have shown that HFPO-DA does not exhibit any primary biodegradation, is not readily or inherently biodegradable and is not structurally transformed under the experimental test conditions. Experimental data is limited, and the bioaccumulation potential of HFPO-DA is uncertain. HFPO-DA has a very low adsorption potential to organic carbon and other solids and a low volatility meaning it has a highly mobility in the aquatic environment and can be transported to areas far from the point of release. According to the classification and labelling in accordance with the CLP Regulation (Regulation (EC) 1272/2008), HFPO-DA has the hazard classifications Acute Toxicity 4, Skin Corrosion/Irritation 1B, Eye Damage/Irritation 1, and specific target organ toxicity – single exposure (STOT SE) 3.

In June 2019, HFPO-DA and its salts were identified as SVHC in accordance with Article 57(f). The MSC acknowledged that HFPO-DA has a high potential to cause effects in wildlife and in humans through the environment due to its very high persistence, mobility in water, potential for long-range transport, accumulation in plants and observed effects on human health and the environment. Table 2 presents the argumentation used by the MSC to show how HFPO-DA present an ELoC to PBT/vPvB substances [35].

Table 2 ELoC of HFPO-DA compared to PBT/vPvB substances
Full size table

REACH dossiers were used to prepare the table, and literature references to support statements can be found within these dossiers [35].

Case Study III: 1,4-dioxane CAS No. 123–91–1

1,4-dioxane has a plethora of uses. It is a solvent in the production of lacquers, varnishes, cleaning and detergent preparations, dyes, antifreeze, adhesives, cosmetics, deodorant fumigants, shampoos, emulsions and polishing compositions, polyester manufacturing, pulping of wood, extraction medium for animal and vegetable oils, laboratory chemical (eluent in chromatography), cassettes, plastic and rubber, insecticides as well as a stabilizer for 1,1,1-trichloroethane [7, 50] 1,4-dioxane is prevalent in groundwater at industrial due to its historic use as a stabilizer [51] and in sewage water resulting from consumer use of surfactants.

1,4-dioxane is non-volatile with a very low Henry's Law constant (4.8 × 10 − 6 atm·m3/mol at 25 °C). However, 1,4-dioxane has a high water solubility (100 mg/mL), a low octanol–water partitioning coefficient (log Kow =  − 0.27) and adsorbs only weakly to organic matter with a log Koc of 0.42 [50]. 1,4-dioxane has a half life of 2 to 5 years in groundwater and 56 days in surface water highlighting the persistent nature of this substance [52]. Taken together these properties imply that 1,4-dioxane is readily leached into water systems and transport through even subsurface environment, which has been demonstrated by point-source releases of 1,4-dioxane causing plumes which have the potential to contaminate both nearfield and distal water bodies [53].

1,4-dioxane is currently listed with the hazard classes under the Classification and labelling in accordance with the CLP Regulation: "Flammable liquids 2", "Carcinogenic 2", "Eye Damage/Irritation 2", and "STOT SE 3". In addition, a recent opinion from the Committee for Risk Assessment (RAC) has proposed that 1,4-dioxane should be classified as Carcinogenic 1B [50]. The properties suspected of causing cancer, may cause respiratory inhalation and causes serious eye irritation are properties of ecotoxicological concern. Table 3 shows the ELoC of 1,4-dioxane to PBT/vPvB substances.

Table 3 ELoC of 1,4-dioxane compared to PBT/vPvB substances
Full size table

PBT/vPvB substances as precedent for the identification of PMT/vPvM substances under REACH

When concerns were raised about the identification and unintended ecotoxicological effects of DDT [66] and the detection of PCBs in top predators [67] it became clear that such PBT/vPvB substances can reach threshold concentrations at which (eco)toxic effects occur, unless emissions are controlled or minimized. Since then, PBT/vPvB substances have been included in a large amount of regulation [4] including REACH. Annex XIII provides criteria to be used to identify PBT/vPvB substances and these should be applied to all substances manufactured or produced at 10 or more tonnes per year and for all constituents present greater than 0.1% (though certain substances are exempted, such as isolated intermediates, as described in Article 14(2)). In ECHA’s PBT/vPvB guidance it is stated that: "PBT or vPvB substances may have the potential to contaminate remote areas that should be protected from further contamination by hazardous substances resulting from human activity because the intrinsic value of pristine environments should be protected" […] "the effects of such accumulation are unpredictable in the long-term" […] "such accumulation is in practice difficult to reverse as cessation of emission will not necessarily result in a reduction in substance concentration" [68]. PBT/vPvB substances have also been acknowledged to present a planetary boundary threat if released in substantial quantities, as their removal from the environment can be irreversible [69].

The same is true for PMT/vPvM substances; however, there is a regulatory gap in REACH to address these substances. The case studies above show that PMT/vPvM substances have an ELoC to PBT/vPvB substances due to the scientific evidence of probable serious effects to human health or the environment. The difference between these substance groups is their exposure route. In cases, where emissions are continuous, mobile substances can accumulate in (semi-)closed drinking water cycles as waste water is recycled to drinking water and as well as in pristine remote environments, far from the site of release, via surface water and groundwater transport. Similarly, under prolonged emissions, bioaccumulative substances are able to accumulate in the food chain and also in pristine remote environments via migrating biota and in some cases atmospheric transport [70]. For both PMT/vPvM and PBT/vPvB substances, the long term and long-range transport and risk of exposure is very difficult to determine in advance and with sufficient accuracy. Owing to the complex nature of water systems and food chains as well as a lack of modelling tools, risks posed by these substances are most often identified retrospectively. In addition, reversing emissions (and, therefore, effects) is difficult due to the high persistence of these substances. Reservoirs of emitted substances in commercial products, landfills, polluted soil and aquifers, for instance, can act as continuous sinks to the environment over long time scales. Thus, there is the potential for risks to persist over multiple generations. In extreme cases, where exposure of either PMT/vPvM or PBT/vPvB substances reaches harmful levels, such as chronic toxicity, they would pose irreversible health and environmental effects that could result in impaired quality of life or ecosystem function. Releases of both PMT/vPvM and PBT/vPvB substances places pressure on society from an economic and resource point of view, since a cessation of emissions does not necessarily lead to a reduction in concentrations in the long term. Based on this, Table 4 provides a summary of the criteria that can be used to demonstrate that all PMT/vPvM substances can be considered an ELoC as PBT/vPvB substances, and as such should be identified as SVHC under Article 57(f).

Table 4 Conceptual comparison showing that PMT/vPvM substances pose an equivalent level of concern as PBT/vPvB substances
Full size table

Conclusion

The conceptual comparison using the 16 categories developed here supports the assumption that PMT/vPvM substances cause an equivalent level of concern as PBT/vPvB substances and as such should be regulated under Article 57 in REACH. Currently this can be done using Article 57 (f) as shown above, or it could be achieved via the introduction of two new articles: Article 57 (g) for PMT substances and Article 57 (h) for vPvM substances. At the time of writing, a Motion for Resolution has been put forward by the European Parliament for such an inclusion [71]. For practical guidance towards this inclusion, the German Environment Agency has recently published criteria and an assessment procedure that can be used under REACH to identify PMT/vPvM substances [2]. These criteria could be added to Annex XIII of REACH and used by ECHA to publish a Guidance on Information Requirements and Chemical Safety Assessment for a PMT/vPvM assessment. Using these established PMT/vPvM substance criteria in Article 57 and Annex XIII would allow the distinct and unambiguous identification of PMT/vPvM substances as SVHCs. This would provide the chemical industry with a defined framework to follow, as for PBT/vPvB substances, and would allow the substitution of hazardous substances and uses under REACH.