Impact statement
Our study provides critical insights into the interactions between chlorine-based disinfection treatments and microplastics, offering valuable guidance for water infrastructure material choices and disinfection practices. Key findings include:
-
• Chlorine-based disinfection processes interact with microplastics, with impacts varying by polymer type, dosage and water chemistry.
-
• Chlorination has a more significant effect on microplastics compared to chloramination.
-
• Different polymers pose varying degrees of risk when exposed to disinfection treatments.
-
• Understanding which polymers are of concern can help minimize public health risks.
This research highlights the physicochemical changes that occur in seven common polymer types when exposed to industry-relevant doses of chlorination and chloramination. These changes include biofilm formation, increased total organic carbon (TOC) levels, cracks and broken bonds, which can lead to increased threats such as serving as vectors for pathogens, formation of disinfection by-products (DBPs) and degradation into nanoplastics.
The study underscores the potential public health risks associated with more hazardous polymers, particularly polypropylene (PP), expanded polystyrene (EPS) and polyamide (PA), which showed significant degradation under chlorination. These findings are crucial for designing water treatment plants and making informed decisions about material choices, especially for applications involving chlorine-based disinfectants. By optimizing water treatment processes, we can minimize the generation of harmful sub-micron particles.
This research provides essential information for industry and policymakers and these insights should be considered in the development of the global plastic treaty. This study not only advances scientific understanding, but also has practical implications for public health and environmental policy, emphasizing the need for continued research and informed decision making in water treatment practices.
Highlights
-
• Chlorine-based disinfection treatment processes interact with microplastics.
-
• Disinfection impacts vary by polymer type, dosage and water chemistry.
-
• Chlorination impacts microplastics more than chloramination.
-
• Different polymer types pose varying degrees of risk from disinfection.
-
• Knowledge of polymers of concern can minimize public health risks.
Introduction
Plastic is a synthetic, engineered material made of long chain-like molecules called polymers and can differ greatly depending on monomer constituents, additives, density, thermal properties and other essential properties (PlasticsEurope, 2021). Some monomers and additives are hazardous, and risk leaching out of plastics as they may not always be chemically bound (Hahladakis et al., Reference Hahladakis, Velis, Weber, Iacovidou and Purnell2018; Bridson et al., Reference Bridson, Gaugler, Smith, Northcott and Gaw2021; UN. Chemicals in Plastics, 2023). Moreover, plastic polymers sorb and desorb chemical contaminants (Llorca et al., Reference Llorca2020; Fu et al., Reference Fu, Li, Wang, Luan and Dai2021; Liu et al., Reference Liu2021), pathogens (Rodrigues et al., Reference Rodrigues, Oliver, McCarron and Quilliam2019) and colonize microscopic organisms (Chen et al., Reference Chen2020; Bhagwat et al., Reference Bhagwat2021). Thus, plastic pollution poses significant threats, particularly as it is not a single stressor, but a complex mixture with different physicochemical and ecotoxicological properties (Rochman et al., Reference Rochman2015).
Nonetheless, due to its durability, versatility and affordability, plastics are ubiquitous in many varied applications including parts of water and wastewater infrastructure (Senathirajah et al., Reference Senathirajah, Kemp, Saaristo, Ishizuka and Palanisami2022). The complexity of plastic risks to the water sector is challenging, given the multitude of stressors, sources and types of plastic polymers. Water and wastewater systems are dominated by polyethylene (PE), polyvinyl chloride (PVC) and polypropylene (PP) pipe systems, and the replacement of metals, concrete and clay pipes often utilize polybutylene (PB) and crosslinked polyethylene (PEX) (Heathcote, Reference Heathcote2009). Currently, plastic pipes comprise up to 55% of global sales of water and wastewater piping, and is projected to grow (Research, P. M, 2022). The continued global expansion in plastic production, consumption and emissions is resulting in a plastic pollution disaster (Senathirajah et al., Reference Senathirajah, Bonner, Schuyler and Palanisami2023), increasing the microplastics load and likelihood for entry into water supply systems (Senathirajah and Palanisami, Reference Senathirajah and Palanisami2023).
Microplastics and nanoplastics – particles < 5 mm and < 100 nm, respectively – vary in shape and may be intentionally produced (primary microplastics) or result from the degradation or breakdown of larger plastics from exposure to the natural weathering forces through physical, chemical and/or biological processes (secondary microplastics) (Senathirajah and Palanisami, Reference Senathirajah and Palanisami2023). The persistence and transboundary characteristics of microplastics, owing to their light-weight nature and small size, enables wide dispersion, resulting in the pollution of atmospheric, aquatic and terrestrial environments and biota therein (Jambeck et al., Reference Jambeck2015; Barrett et al., Reference Barrett2020; Lau et al., Reference Lau2020). Microplastics have been found in plants (Azeem et al., Reference Azeem2021), animals and humans (Dzierżyński et al., Reference Dzierżyński2024). Impacts on various species have been investigated (Graham and Thompson, Reference Graham and Thompson2009; Lourenço et al., Reference Lourenço, Serra-Gonçalves, Ferreira, Catry and Granadeiro2017; Li et al., Reference Li2018; Verlis et al., Reference Verlis, Campbell and Wilson2018; Panebianco et al., Reference Panebianco, Nalbone, Giarratana and Ziino2019; Schwabl et al., Reference Schwabl2019), with wide-ranging impacts found to adversely and materially affect health and disrupt ecosystem function and services (Savoca et al., Reference Savoca, Wohlfeil, Ebeler and Nevitt2016; Beaumont et al., Reference Beaumont2019; Triebskorn et al., Reference Triebskorn2019; Qian et al., Reference Qian2020; Amelia et al., Reference Amelia2021; Lozano et al., Reference Lozano2021; Senathirajah et al., Reference Senathirajah2021; Kutralam-Muniasamy et al., Reference Kutralam-Muniasamy, Shruti, Pérez-Guevara and Roy2022; Landrigan et al., Reference Landrigan2023).
The main exposures routes for humans are through ingestion, inhalation or dermal contact. We could potentially be cumulatively ingesting 0.1–5 g of microplastics a week from a combination of sources, including drinking water (Feld et al., Reference Feld, da Silva, Murphy, Hartmann and Strand2021; Senathirajah et al., Reference Senathirajah2021; Semmouri et al., Reference Semmouri2022). The recent investigation using mice models on the implications of exposure to microplastics in drinking water indicated that though there were no overt signs of toxicity, immunohistochemistry detected adverse impacts ‘indicating progression toward lytic cell death via a proinflammatory pathway’ and a loss of microbiome diversity in the faecal analysis undertaken, highlighting the long-term risks of ingestion from drinking water (Szule et al., Reference Szule2022). The detection of microplastics in the human heart (Yang et al., Reference Yang2023), placenta (Ragusa et al., Reference Ragusa2021), lungs, blood and breastmilk (Kutralam-Muniasamy et al., Reference Kutralam-Muniasamy, Shruti, Pérez-Guevara and Roy2022) highlights the crucial need for effective removal of microplastics.
Although conventional water treatment plants (WTPs) and wastewater treatment plants (WWTPs) are not specifically designed to remove microplastics, studies have demonstrated that substantial amounts can be removed (Jung et al., Reference Jung, Kim, Kim, Jeong and Lee2022), 25%–90% from WTPs (Cheng et al., Reference Cheng2021; Sol et al., Reference Sol, Laca, Laca and Díaz2021) and greater than 90% in WWTPs (Jeong et al., Reference Jeong2023; Sheriff et al., Reference Sheriff, Yusoff and Halim2023). Flotation, sedimentation, filtration, coagulation–flocculation and disinfection are common treatment processes at treatment plants with varying degrees of removal efficacies. For example, sand filtration has been shown to remove 100% of microplastics > 50 μm, granulated activated carbon (GAC) filtration removed between 56% and 61% of microplastics < 10 μm (Yang et al., Reference Yang2019) and dissolved air flotation removed 79% of microplastics < 10 μm (Hidayaturrahman and Lee, Reference Hidayaturrahman and Lee2019). Coagulation–flocculation process can remove microplastics up to 92%(Zhang et al., Reference Zhang2021). Coagulants, such as aluminium salts and iron salts, polyacrylamide (PAM) and polyaluminium chloride (PAC), often used as aids to remove suspended particles and colloidal substances (Ma et al., Reference Ma2019), have been found to contribute to the microplastics loading. For instance, PAM was found in treated water when it was not detected in the raw water (Wang et al., Reference Wang, Lin and Chen2020). Electrocoagulation, an advanced oxidation process, removes the potential for contamination from coagulants, and have been found yielding > 90% removal efficiencies (Senathirajah et al., Reference Senathirajah, Kandaiah, Panneerselvan, Sathish and Palanisami2023). Oxidation processes have also been reported to reduce microplastics loads (Jeong et al., Reference Jeong2023).
Disinfection is a well-established standard oxidation process, utilized widely for pathogen removal and oxidative degradation of organic contaminants from both water and wastewater. Decreases in microplastics have been observed following disinfection (Jiang et al., Reference Jiang2020). Ozonation and chlorination disinfection processes have reportedly removed microplastics > 10 μm by 40%–50%, however, it was noted that finer sized microplastics increased in concentration by 5%–8%(Yang et al., Reference Yang2019). Chlorine dose ranging from 0.15 to 0.18 mgL−1 reportedly removed 0.13%–2.89% of microplastics (Luo et al., Reference Luo2023).
The oxidizing effects of chlorination of water on plastic pipes have previously been studied (Kowalska et al., Reference Kowalska, Klepka and Kowalskin.d.; Gorbunova et al., Reference Gorbunova, Gaevoi, Gerasimov, Chalykh and Kalugina2010; Mitroka et al., Reference Mitroka, Smiley, Tanko and Dietrich2013; Kelkar et al., Reference Kelkar2019; Liu et al., Reference Liu2022). Researchers have postulated that the disinfection process may in fact be generating more problematic, finer micro- and nanoplastics through the oxidative treatment process (Yang et al., Reference Yang2019; Lin et al., Reference Lin2021; Jeong et al., Reference Jeong2023; Liu et al., Reference Liu2023). Residual chlorine in contact with plastic water infrastructure accelerates ageing, fragmentation and degradation of the polymers (Whelton and Dietrich, Reference Whelton and Dietrich2009; Mitroka et al., Reference Mitroka, Smiley, Tanko and Dietrich2013; Kelkar, Reference Kelkar2017; Fujii et al., Reference Fujii2019; Kelkar et al., Reference Kelkar2019; Li et al., Reference Li2022). While oxidative effects are acknowledged, knowledge on the impacts of disinfection at operational dose rates on various polymer types is limited (Jeong et al., Reference Jeong2023).
Due to the variety of polymeric components in water infrastructure that are exposed to chlorine-based disinfected water, this study aims to add to knowledge on the impacts that chlorination and chloramination have on different plastic polymer types. We investigate the physicochemical changes resulting from chlorination and chloramination disinfection at typical dose rates on seven common polymer types as an indicator of degradation to support the identification of polymers of concern (PoC). We also strive to determine if there are preferential impacts between chlorination and chloramination. This study facilitates decision making on types of material and method of disinfection in relation to the management of microplastics.
Methods
Materials
All analytical grade chemicals, Milli-Q water, sodium hypochlorite solution (12.5% ± 2.5% Cl2 by weight), 12 M hydrogen chloride solution, 18.4 M sulphuric acid solution, 99.5% sodium thiosulphate powder, hypochlorous acid and ammonium chloride, were purchased from Merck Australia.
Specific plastic polymers are favoured for their desirable properties and are consequently used in various components of water supply infrastructure (Table 1). Micro-sized resin powders of acrylonitrile butadiene styrene (ABS), expanded polystyrene (EPS), low-density polyethylene (LDPE), polyamide (PA), polycarbonate (PC), polyethylene terephthalate (PET) and polypropylene (PP) polymers were selected due to their high use in parts of water infrastructure and detection in water environments globally (Senathirajah and Palanisami, Reference Senathirajah and Palanisami2023). As PVC is already regarded as a PoC (Lithner et al., Reference Lithner, Larsson and Dave2011; Zhang and Lin, Reference Zhang and Lin2015; Senathirajah et al., Reference Senathirajah, Kemp, Saaristo, Ishizuka and Palanisami2022; Wilcox et al., Reference Wilcox, Fox and Valliant2023), we did not investigate PVC despite its high usage as our intent was to provide evidence to contribute towards identifying PoC.
Table 1. Properties of the polymers used in this study
The seven polymeric resin powders, with a mean size of 188 μm (particles between 125 and 250 μm) were obtained from 2 M Biotech LLP, India (Pty) Limited. Varied sizes were not investigated in this experiment as size had been reported as insignificant in relation to the removal performance of disinfection (Cheng et al., Reference Cheng2021). All polymers were cleaned by immersion in 0.1% sodium dodecyl sulphate at 40°C for 1 h, then rinsed thrice with Milli-Q water, followed by immersion in 0.1 M HCl at 40°C for 1 h, then rinsed at least thrice with Milli-Q water, ensuring the solution ran free of acid and then dried under nitrogen flow before use.
One set of experiments were conducted on cleaned virgin particles to mimic new plastic infrastructure and/or fittings. For the clean virgin experiments, 1000 mg of cleaned polymer resin powder was added into 1 L of Milli-Q water, potable water (PW) or recycled water (RW) (depending on the experimental run) in a 2 L glass beaker, covered with foil and stirred for 30 min with a magnetic stirrer (1000 ppm is a conservative reflection of industry relevant conditions (WHO, 2019)). The water chemistry reflected differences between PW and RW (Supplementary Figure S1). Due to COVID safety requirements at the laboratory, we could not carry out wastewater experiments.
For the biofouled experiments (likely condition of a large proportion of particles in WTPs), the cleaned polymer particles were spiked (100 mg/100 mL) in tap water in a clean glass beaker and incubated in a closed dark room at 21°C. Quantification of biofilm biomass was measured using crystal violet (CV) method (Bhagwat et al., Reference Bhagwat2021): 1 μL CV (1% aqueous, Merck) was added to 10 mg of powdered polymer and after 45 min, rinsed with ultrapure water until the filtrate was clear. Particles were then dried for an hour and transferred to new glass tubes containing 1.5 mL of 95% ethanol and inverted to mix. After 10 min, the absorbance of the solution at 600 nm was measured using a UV Spectrophotometer (Shimadzu, UV-1800, Japan).
Disinfection experiments
Chlorination
For each polymer type (triplicate separate runs for virgin and biofouled plastics), 20 mL of the prepared microplastic suspension at 1000 mgL−1 and initial pH of 7.0 (~pH of freshwater, drinking water), 20°C was placed in a 40-mL scintillation glass vial. A stock solution of 100 mgL−1 of fresh sodium hypochlorite solution (4.5% concentrated) was prepared. From the stock solution, chlorine concentration of 2.5 mgL−1 was prepared and added to both virgin and biofouled polymer types separately. The polymers were periodically filtered using a stainless steel filter with pore size of 45 μm at intervals of 0, 0.5, 1, 2, 3, 4, 6, 12 and 24 h for further analyses. The process was repeated by adjusting the initial pH of the water sample to pH 5.0 (~pH of RO water), and subsequently to pH 8.0 (~pH of wastewater). Further, additional experiments were carried out for sodium hypochlorite concentrations of 5 and 10 mgL−1, also prepared from the stock solution.
Chloramination
Fresh monochloramine (NH2Cl) stock solution (500 mgL−1) was prepared by mixing 400 mgL−1 sodium hypochlorite (NaOCl) and 100 mgL−1 ammonium chloride (NH4Cl) for 30 min. The stock solution was used to prepare chloramine at 2, 4 and 6 mgL−1 final concentrations. NaOCl and NH2Cl stock solution concentrations were measured using a portable residual chlorine analyser. Monochloramine (20 mL) spiked water samples (PW and RW at pH 5.0, 7.0 and 8.0) were taken in 40 mL scintillation glass vials and added with both virgin and biofouled polymers (25 mgL−1) and incubated for 24 h. The polymers were periodically filtered with a stainless steel filter with pore size of 45 μm at intervals of 0, 0.5, 1, 2, 3, 4, 6, 12 and 24 h for further analyses. Experiments were carried out for monochloramine concentrations of 2, 4 and 6 mgL−1.
Characterization
We adopted an approach that utilized several analysis techniques to illustrate the physicochemical changes in PW and RW.
Physical
Disinfected polymers of both virgin and biofouled particles were air-dried and coated with ~2 nm platinum using an auto fine coater. Changes to surface morphology were observed using scanning electron microscopy (SEM) from JEOL Model No: JSM-7900F at magnification 10–10000.
Chemical
Attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectroscopy (Perkin Elmer) was used to identify the disinfection changes to the polymer surface. Treated polymers were completely dried before being placed in the ATR diamond crystal and spectra were obtained in the transmission mode of wavenumber range between 4000 and 400 cm−1 with a resolution of 4 cm−1 and an average of 32 scans. This was done before and after treatment to look for changes and carbonyl oxidation products.
Water quality
Water quality parameters were measured using standard laboratory equipment. The pH and electrical conductivity were measured using (Mettler Toledo Seven Compact Duo). Triplicates of the total organic carbon (TOC) concentrations of disinfected PW and RW, treated with both chlorination and chloramination, were assessed for the seven polymers using a TOC analyser (TOC-L, Shimadzu) at three time points: 3, 15 and 30 days. We measured the concentration of metals in water using the USEPA method 3051 H by inductively coupled plasma-mass spectroscopy (ICP-MS) (Agilent 7500c).
QA and QC
All experiments were carried out while wearing 100% cotton lab coats. Sterile nitrile gloves were used during the experiments. Metal and glass equipments were preferred over plastic equipment to minimize the risk of microplastic contamination. All materials were thoroughly washed with Milli-Q water before use and between different sample processing. All glassware were cleaned by immersion in 5% HNO3 solution for 24 h, washed with tap water, then rinsed with ultra-pure water and baked at 300°C for 12 h before use (Senathirajah et al., Reference Senathirajah, Kandaiah, Panneerselvan, Sathish and Palanisami2023).
Experiments were conducted at room temperature within a biosafety cabinet to avoid atmospheric contamination. Blank samples were processed alongside the experimental samples to account for any microplastics that might be present in the laboratory environment. These blanks underwent identical processing steps, allowing background contamination detection. Some microfibers were detected in some blanks and these were characterized using FTIR which indicated that the microfibers were cellulose-based particles. Additionally, the particles under investigations were powders and not microfibers. Consequently, no corrections were required.
Filters were rinsed thoroughly with ultrapure water before being used to remove any potential contamination. After filtration, each filter was immediately placed in individual Petri dishes to preserve sample integrity until analysis.
Data analysis
Data were recorded in a database and analysed using Excel (Microsoft, Redmond, WA, USA). Mean and standard deviations were calculated. Principal component analysis (PCA) was conducted using the statistical package R (Team, 2017) to simplify the complex dataset, reducing the dimensions and signalling the factors of greatest influence. We calculated the carbonyl index (CI) using the equation: CI = A maxCarbonyl/A maxSignature (Martinez-Colunga et al., Reference Martinez-Colunga2020; Celina et al., Reference Celina, Linde and Martinez2021) for the various polymer types as presented in Table 3, where the A maxCarbonyl was the maximum absorbance between 1850 and 1650 cm−1. A semi-quantitative risk ranking method was adopted by assigning binary and ordinal scores to nine factors affecting the degradation of each polymer type. The total scores for each polymer were summed across the nine factors, followed by ranking of the polymers based on their total scores, from highest to lowest. This risk ranking methodology is often employed in risk assessments where risks are ranked based on a combination of qualitative and semi-quantitative criteria.
Results
Depending on the reaction conditions, the seven polymers demonstrated varying degrees of physical and chemical impacts resulting from chlorination and chloramination treatments.
Physical
Surface morphological changes were visible on the surface of the polymers following oxidation resulting from disinfection on both the virgin and biofouled polymers. Figure 1 provides a snapshot of the changes observed before and after treatment using the SEM after 24 h at pH 7 and 10 mgL−1 chlorination and 6 mgL−1 chloramination.
Figure 1. SEM images of the seven polymers virgin particles and biofouled particles, chlorinated and chloraminated.
The trends of the impacts observed from the experiments of both chlorination and chloramination by polymer type are summarized in Table 2 through the qualitative description of the degree and formation of flakes, crystals, deposits, debris, bumps, cracks, pits and cavities observed for all experiments.
Table 2. Summary of surface morphological impacts on seven polymer types observed following chlorination and chloramination
It becomes evident that chlorination on virgin polymers led to the greatest changes, with notable changes on the biofouled particles. The reduction in flakes and layers following treatment on biofouled particles was due to reaction of the disinfectants with the biofilm, resulting in the reduction of biofilm. The SEM images were aligned with measurements of biofilm growth on polymer types, following the order PA > LDPE > EPS > PC > PP > PET > ABS (Figure 2).
Figure 2. Biofilm growth over 3 weeks on seven microplastic polymer types.
Lesser impacts were observed following chloramination (except for virgin PA and PC), and more changes were observed, particularly with the formation of debris/residue on the surfaces of particles in RW than in PW. In terms of polymers, overall EPS, PP, LDPE and PA presented more surface morphological changes, with PET presenting the least.
Chemical
FTIR spectroscopy was utilized to capture the chemical changes of the seven polymers in PW and RW, subjected to chlorination and chloramination at three dosages each at three pH conditions, and the spectra of the three most impacted polymers are depicted in Figure 3. The remaining spectra are included in Supplementary Figure S2. The virgin microplastics displayed characteristic absorption peaks for each polymer.
Figure 3. FTIR spectra for three polymers treated by disinfection processes at three industry relevant dose rates (a_1: EPS Chlorinated and a_2: EPS Chloraminated; b_1: PA Chlorinated, b_2: PA Chloraminated; c_1: PP Chlorinated and c_2: PP Chloraminated [colours darken with increasing dosage, finer lines represent PW, thicker lines represent RW, warm colours represent virgin particles, cool colours represent biofouled particles, solid lines represent pH 7, dotted lines represent pH 5 and dashed lines represent pH 8]).
The PCA eigenvectors indicated that the highest coefficients that affected the variability were the polymer type, then particle condition (Supplementary Table S1 and Supplementary Figure S7).
Water quality
The TOC concentration increased over time for virgin and biofouled particles for all polymers as the dose increased for chlorination and chloramination treatments (Figure 4). Both treatments demonstrated a consistent trend in the order of increase of TOC according to ABS < EPS < LDPE < PC < PET < PP < PA. Only slight differences were observed between PW and RW, with more TOC in PW when virgin polymers were exposed to chlorination, and more TOC in RW when biofouled polymers were exposed to chloramination. In the first 15 days, chlorination treatment resulted in slightly higher TOC than chloramination in PW and RW. TOC was then observed to be higher in the chloraminated samples in RW and PW at 30 days. While the chlorination treatment resulted with the greatest TOC from biofouled polymers at pH 7 and from virgin polymers at pH 5, the pH 5 conditions affected the chloraminated virgin particles at operationally low dose rates the least. Generally, pH was not seen to be a key influencer on the amount of TOC.
Figure 4. Mean total organic carbon concentration following chloramination (dosages 2, 4, 6 ppm) and chlorination (dosages 2.5, 5, 10 ppm) in potable water (PW) and recycled water (RW) after 3, 15 and 30 days for virgin polymers and biofouled polymers.
Discussion
Chlorination and chloramination
WTPs and WWTPs predominantly use conventional treatment processes including flotation, coagulation–flocculation, sedimentation, filtration and disinfection. Chlorination and chloramination are well-established disinfection treatment processes for both potable water and wastewater, for removing pathogens (including bacteria, viruses and fungi) and oxidative degradation of organic contaminants (Li et al., Reference Li2022). The treatments target the regulated waterborne microbial pathogens which have similar density and size such as Cryptosporidium (1.05–1.04 gcm−3, 4–7 μm), Giardia (1.18 gcm−3, 10–20 μm), Escherichia coli (1.08 gcm−3, 1 μm), Naegleria (1.14 gcm−3, 7–35 μm) and Legionella (1.21 gcm−3, 0.3-20 μm) (Flesher et al., Reference Flesher, Jennings, Lugowski and Kasper1982; Lowrey, Reference Lowrey1985; Medema et al., Reference Medema, Schets, Teunis and Havelaar1998; Pernin et al., Reference Pernin, Pélandakis, Rouby, Faure and Siclet1998; Füchslin et al., Reference Füchslin, Kötzsch, Keserue and Egli2010; Deksne et al., Reference Deksne2022) as unregulated microplastics (0.85–1.42 gcm−3, 0.1–5000 μm).
Chlorine, known for its strong oxidative property, is utilized in gaseous form or hypochlorite salts for safer handling and lower associated risks (Voukkali and Zorpas, Reference Voukkali and Zorpas2015; NHMRC, 2022). Chlorine hydrolyses to form hypochlorous acid (HOCl) when added to water. The HOCl then dissociates to introduce hydrogen ions (H+) and hypochlorite ions (OCl−) at a rate dependent on pH and temperature.
The free chlorine residual – that is the sum of HOCl and OCl− – is the amount of chlorine available to keep the water safe. It is recommended that free chlorine residuals ≥ 0.2 mgL−1 are maintained throughout the distribution system (NHMRC, 2022). Chlorine doses up to 4 mgL−1, not exceeding 5 mgL−1, are considered safe (NHMRC, 2022). We investigated dose rates of 2.5, 5 and 10 mgL−1 and observed that increasing dosage increased impact on all polymers.
Chloramination is an alternative chlorine-based disinfectant utilized by the water sector. Monochloramine (NH2Cl) is increasingly opted for disinfection treatment due to its less aggressive nature, slower decay rate, stability and persistence and less production of taste and odour issues compared to free chlorine (Marchesi et al., Reference Marchesi2020; NHMRC, 2022). The controlled addition of ammonia to HOCl produces monochloramine.
As chloramination doses up to 4 mgL−1 are considered safe (Lytle et al., Reference Lytle2021; NHMRC, 2022), we investigated dose rates of 2, 4 and 6 mgL−1 and observed increasing impacts on all polymers as dosage increased. Our results indicated the greatest changes from chlorination and lesser impacts following chloramination (PA and PC excepted). The chlorination dosage of 5 mgL−1 revealed more impacts than the chloramination dosage of 6 mgL−1, affirming that monochloramine is less aggressive on polymeric materials. Similarly, monochloramine is found to be effective for pathogens such as Naegleria, Legionella, E. coli and Vibrio. However, for the inactivation of more resistant microorganisms such as Giardia and Cryptosporidium, chlorination is more effective (Gagnon et al., Reference Gagnon2004; NHMRC, 2022).
Dynamics of microbes, contaminants and polymer properties
Microplastics act as substrates for complex microbial communities, bacteria and taxa to grow and be transported (Gregory, Reference Gregory2009; Zettler et al., Reference Zettler, Mincer and Amaral-Zettler2013; Kirstein et al., Reference Kirstein2016; Shi et al., Reference Shi, Xia, Wei and Ni2022) (Supplementary S1). Naegleria, Legionella (WHO, 2019), Giardia, Cryptosporidium (Zhang et al., Reference Zhang2022; Zhong et al., Reference Zhong2023), E. coli (Rodrigues et al., Reference Rodrigues, Oliver, McCarron and Quilliam2019; Duong et al., Reference Duong, Park and Maeng2021) and Vibrio (Zettler et al., Reference Zettler, Mincer and Amaral-Zettler2013; Kirstein et al., Reference Kirstein2016; Naik et al., Reference Naik, Naik, D’Costa and Shaikh2019; Rodrigues et al., Reference Rodrigues, Oliver, McCarron and Quilliam2019) as well as hazardous algae and cyanobacteria (Zettler et al., Reference Zettler, Mincer and Amaral-Zettler2013; Naik et al., Reference Naik, Naik, D’Costa and Shaikh2019) have been identified on plastic polymers (Li et al., Reference Li, McDonald, Sathasivan and Khan2019). This stresses the importance of adequate treatment especially as microplastics colonized by biofilm not only provides bacterial source, but also emits scents that promotes ingestion (Savoca et al., Reference Savoca, Wohlfeil, Ebeler and Nevitt2016).
Additionally, microbial communities secrete extracellular polymeric substances and anionic molecules, further advancing the formation of biofilms (Senathirajah et al., Reference Senathirajah, Kemp, Saaristo, Ishizuka and Palanisami2022). The formation of biofilm is also regulated by hydrophobicity, due to greater adherence to hydrophobic surfaces than hydrophilic surfaces (Pinto et al., Reference Pinto, Langer, Hüffer, Hofmann and Herndl2019; Duong et al., Reference Duong, Park and Maeng2021; Guo et al., Reference Guo2022). Hence, different communities form preferentially on different polymer types, and our results indicated decreasing abundance of biofilm in the order PA > LDPE > EPS > PC > PP > PET > ABS (Figure 2). Of note, PA has been reported to have the highest biofilm structural diversity (Ramsperger et al., Reference Ramsperger2020).
To manage biofilm, it is common industry practice to periodically flush and shock-dose water distribution networks at high concentrations (Senathirajah and Palanisami, Reference Senathirajah and Palanisami2023). Monochloramine is an effective biocide disinfectant for controlling biofilm growth (Marchesi et al., Reference Marchesi2020; NHMRC, 2022). Notwithstanding, excessive free ammonia in the system can result with chloramines causing biological nitrification. The resulting accumulation of nitrite nitrogen poses human health hazards and promotes biofilm or sediment formation, which facilitates the large-scale reproduction of bacteria, compromises water quality and increases consumption of dissolved oxygen (DO), consequently lowering pH (Liu et al., Reference Liu, Liu and Ding2020). Not surprisingly, our water quality results reflected lower levels of DO at pH 5 in RW compared to the PW. The RW had more impurities, as indicated by the lower DO, higher electrical conductivity values and metal concentrations (Supplementary Figure S1). DO levels affects the types of microbial communities (aerobic/anaerobic) that form on polymers and thus biotic degradation rates (Chamas et al., Reference Chamas2020), viz. effective management of microbial communities can reduce degradation.
The presence of biofilm is additionally problematic as it facilitates greater adsorption of contaminants, through electrostatic interactions, ion exchange, complexation, diffusion through cell walls and membranes or precipitation mechanisms (Qi et al., Reference Qi2021). Notably, biofilm is not the only consideration in relation to adsorption. For example more biofilm formed on PE and PET than PP, but higher abundance of antibiotic resistant bacteria (ARG) was found in the biofilm formed on PP (Guo et al., Reference Guo2022). The effects of ARGs are a significant threat to water supply systems, particularly following disinfection (Shi et al., Reference Shi, Xia, Wei and Ni2022; Shi et al., Reference Shi2022). Chlorination and chloramination are reported to increase abundance and dissemination of ARGs (Ghordouei Milan et al., Reference Ghordouei Milan, Mahvi, Nabizadeh and Alimohammadi2022; Shi et al., Reference Shi, Xia, Wei and Ni2022). This is compounded by the presence of microplastics which provides sites for spreading ARGs (Arias-Andres et al., Reference Arias-Andres, Klümper, Rojas-Jimenez and Grossart2018; Zhang et al., Reference Zhang, Lu, Wu, Wang and Luo2020; Shi et al., Reference Shi2022), and exacerbated by biofilm which enriches sites by horizontal gene transfers in natural ecosystems (Arias-Andres et al., Reference Arias-Andres, Klümper, Rojas-Jimenez and Grossart2018) as evident in PE, PET and PP (Zettler et al., Reference Zettler, Mincer and Amaral-Zettler2013; Guo et al., Reference Guo2022). This predicament is not exclusive to ARGs, but also observed with other emerging contaminants (ECs) such as per and poly-fluoroalkyl substances (PFAS) and polycyclic aromatic hydrocarbons (PAHs). Chlorine disinfection processes have been identified to promote the formation of PFOS (Xiao et al., Reference Xiao, Hanson, Golovko, Golovko and Arnold2018) and chlorinated derivatives of PAHs (Cl-PAHs) (Pinto et al., Reference Pinto2014; Xu et al., Reference Xu2018), while microplastic particles, enhanced by the presence biofilm (Bhagwat et al., Reference Bhagwat2021), provide sites for adsorption and dissemination (Emecheta et al., Reference Emecheta2022).
Microplastics provide a platform to bioaccumulate ECs and act as vectors (Shen et al., Reference Shen2019; Amelia et al., Reference Amelia2021). The binding mechanisms between contaminants and microplastics vary by polymer type, with hydrophobic interaction, hydrogen bonding, van der Waals forces and electrostatic interactions being the main mechanisms, and affects the toxicity of the particle (Kelkar et al., Reference Kelkar2019). Polymer properties such as hydrophobicity, crystallinity, polarity, size and density impact adsorption capacity (Li et al., Reference Li, Zhang and Zhang2018; Wang et al., Reference Wang2021; Xu et al., Reference Xu, Wang and Sun2021; Rai et al., Reference Rai, Sonne, Brown, Younis and Kim2022) (Supplementary S2). The biotic and abiotic degradation processes are slower when adsorption is less (Min et al., Reference Min, Cuiffi and Mathers2020). To an extent, noting that the physicochemical properties of both the contaminant and microplastics influences the adsorption capacity (Rochman et al., Reference Rochman, Hentschel and Teh2014; Gao et al., Reference Gao2019), degradation susceptibility follows adsorption capacity.
Degradation mechanisms
Depending on its unique properties (Table 1), polymers respond and degrade differently (Ofridam et al., Reference Ofridam2021). Polymer degradation can be classified by oxidation and hydrolysis mechanisms (Fotopoulou and Karapanagioti, Reference Fotopoulou, Karapanagioti, Takada and Karapanagioti2019), which deteriorates and modifies the mechanical, physical and chemical properties, leading to depolymerization by cleavage of molecules into oligomers, dimers or monomers and finally mineralization (Fotopoulou and Karapanagioti, Reference Fotopoulou, Karapanagioti, Takada and Karapanagioti2019). Disinfection increases the susceptibility of polymers to oxidative degradation, with free radicals attacking polymer chains to produce oxidation products (Supplementary S3). Indicators of oxidation include polymer chain scission, formation of oxygen-containing functional groups and vinyl groups (CH2=CH), leaching of organic products and broadened molecular weight distribution (Jeong et al., Reference Jeong2023).
Hydrolysis, a reaction with water, results with the cleavage of chemical bonds (Speight, Reference Speight and Speight2017). Generally, polymers containing larger molecular weights do not readily hydrolyse (Benítez et al., Reference Benítez2013; Chamas et al., Reference Chamas2020) and degrade more slowly (Plota and Masek, Reference Plota and Masek2020). Also polymers with main functional group consisting of amide bonds, such as PA, are known to be vulnerable to hydrolysis (Bernstein et al., Reference Bernstein, Derzon and Gillen2005), and have weak chlorine resistance (Cran et al., Reference Cran, Bigger and Gray2011; Idrees and Umar, Reference Idrees and Umar2022; Du et al., Reference Du2023). Our results of PA aligned with previous studies (Cran et al., Reference Cran, Bigger and Gray2011; Idrees and Umar, Reference Idrees and Umar2022; Du et al., Reference Du2023) and demonstrated significant changes when disinfected. Hydrolysis reactions first occur in the amorphous parts of the polymer which leads to an increased concentration of carboxylic acid chain ends (Chamas et al., Reference Chamas2020). Thus, amorphous structures (e.g. PA and EPS) are more susceptible to degradation through hydrolysis, chain scission, crack propagation and mechanical breakdown than crystalline structures such as LDPE and PET (Getor et al., Reference Getor, Mishra and Ramudhin2020; Plota and Masek, Reference Plota and Masek2020). The long-chain, highly branched crystalline structure of LDPE has an affinity to degrade from cross-linking and chain scission reactions. Accordingly, we saw less degradation in LDPE and PET compared to PA and EPS. PP was reported to degrade three times faster than PS due to its crystallizability (Nakatani et al., Reference Nakatani, Ohshima, Uchiyama and Motokucho2022). PP is susceptible to oxidative degradation even when exposed to solar radiation (Rouillon et al., Reference Rouillon2016) (UV photodegradation mechanism) which presents similar surface degradation traits as oxidation from chlorine-based disinfection.
Our biofouled particles results reflected that hydrophilic polymers are more susceptible to degradation (Sun et al., Reference Sun2021) due to hydrolysis, as well as higher surface energies and microbial attachment to the surface, which accelerates degradation (Chamas et al., Reference Chamas2020). Moreover, the adsorption of metals and hydrophilic organics increases following UV photodegradation (Liu et al., Reference Liu2019), implying that the adsorption of metals onto microplastics could increase following disinfection. This is of particular risk in wastewater environments, where heavy metals are prevalent (Supplementary S4). The metal cations may directly adsorb onto the polymer’s surfaces, co-precipitate or adsorb onto hydrous oxides (Bradney et al., Reference Bradney2019). Chlorination is effective in precipitating metals and is often used in water treatment to facilitate filtration. Thus, the implications warrant further research. Interestingly, precipitates formed during chloramination have higher solubility in water (Torrey, Reference Torrey2013).
Water chemistry, particularly ionic strength and pH influences adsorption capacity, notably more on aged plastics compared to virgin plastics (Li et al., Reference Li, Zhang and Zhang2018; Godoy et al., Reference Godoy, Blázquez, Calero, Quesada and Martín-Lara2019; Guo and Wang, Reference Guo and Wang2019; Liu et al., Reference Liu2019; Huang et al., Reference Huang2020; Bhagwat et al., Reference Bhagwat2021). Hence, we investigated three typical pH conditions (reverse osmosis water: pH ~5, drinking water: pH ~7, wastewater: pH ~8) for all seven polymers commonly found in water and wastewater systems (Okoffo et al., Reference Okoffo, O’Brien, O’Brien, Tscharke and Thomas2019; Johnson et al., Reference Johnson2020; Raju et al., Reference Raju2020; Zhang et al., Reference Zhang2020; Senathirajah et al., Reference Senathirajah, Kemp, Saaristo, Ishizuka and Palanisami2022; Vega-Herrera et al., Reference Vega-Herrera2022; Zhu et al., Reference Zhu, Hao, Zhang and Lan2023) with biofouled and virgin particles (Supplementary S5). Degradation, chain scission and reduction in molecular weight can occur in acidic conditions, resulting from the hydrolysis of ester bonds in the polymer backbone, or basic conditions, which could cause deprotonation of the carboxylic acid. Our results indicated that maximum impacts for chlorination were observed at pH 7. The weak acidic and basic conditions showed slight differences with marginally more impacts at pH 8 for chloramination. This aligns with the maximum availability of HOCl lying in the pH range of 6.5–7.5 and maximum monochloramine stability around pH 8.
Noteworthy, due to its strong oxidizing properties, HOCl accelerates the degradation of plastics. And if not fully mineralized, directly increases the toxicity as particle size decreases (Jeong et al., Reference Jeong2023) as the surface area to volume ratio of the particle increases, increasing the adsorption capacity from the additional sites for the adsorption of contaminants (Wagner et al., Reference Wagner2014; Gao et al., Reference Gao2019; Zhu et al., Reference Zhu, Hao, Zhang and Lan2023) and accumulation of toxins (Davidson, Reference Davidson2012). The similarity in size to various zooplankton and phytoplankton increases the risk to be mistakenly ingested by many species (Graham and Thompson, Reference Graham and Thompson2009; Davison and Asch, Reference Davison and Asch2011; Browne et al., Reference Browne, Niven, Galloway, Rowland and Thompson2013; Van Cauwenberghe et al., Reference Van Cauwenberghe, Claessens, Vandegehuchte and Janssen2015; Lourenço et al., Reference Lourenço, Serra-Gonçalves, Ferreira, Catry and Granadeiro2017; Li et al., Reference Li2018; Panebianco et al., Reference Panebianco, Nalbone, Giarratana and Ziino2019; Schwabl et al., Reference Schwabl2019; Zhang et al., Reference Zhang, Wang and Kannan2019; Zhang et al., Reference Zhang, Wang, Trasande and Kannan2021), with growing concern with decreasing size and its increasing adsorptive capacity. Smaller sized particles are also able to be dispersed (through air (Zhang et al., Reference Zhang2020), water (Ryan, Reference Ryan2015) and land (Selonen et al., Reference Selonen2020)) and translocated within organisms (Senathirajah et al., Reference Senathirajah2021) more easily (Thevenon et al., Reference Thevenon, Carroll and Sousa2014) (Supplementary S2).
Particle density also directly impacts the dispersion and distribution pathways, thereby influencing the physical, chemical and biological interactions (Liu et al., Reference Liu, Zhan, Wu, Li, Wang and Gao2020) and subsequent fate in terms of degradation, accumulation and bioavailability. Particle density varies by polymer type and can be modified by oxidative degradation following exposure to disinfectants. Our results showed considerable impacts following disinfection on the lesser dense polymers, EPS and PP compared to more dense polymers, PET and PC. The appearance of fine micro-scale cavities (Table 2) implies a reduction in particle density. Similar to previous findings of PET (Zhu et al., Reference Zhu, Hao, Zhang and Lan2023), we observed more obvious rough surfaces with fragmentation, debris and cracks on the oxidized particles compared to the virgin particles (Table 2). Fragmentation suggests cleavage of backbones, the formation of side chains and oxygen-containing functional groups and potential release of additives (Zhu et al., Reference Zhu, Hao, Zhang and Lan2023) (Supplementary S6). The physicochemical impacts on the surfaces increases the specific surface area of the particle which increases adsorption capacity (Supplementary S2). Following disinfection, we observed these changes, along with increase in carbonyl groups, which also facilitates and enhances the adsorption of disinfection by-products (DBPs) on microplastics (Zhu et al., Reference Zhu, Hao, Zhang and Lan2023).
Disinfection by-products
Chlorine and chloramines react with dissolved organic matter (DOM) to produce carcinogenic DBPs (Torrey, Reference Torrey2013; Wei et al., Reference Wei, Gao and Li2017; Zhu et al., Reference Zhu, Hao, Zhang and Lan2023). However, because chloramines are not as reactive as chlorine, they react less with DOM to produce less DBPs (Marchesi et al., Reference Marchesi2020; Lytle et al., Reference Lytle2021; NHMRC, 2022; Ghanadi et al., Reference Ghanadi, Kah, Kookana and Padhye2023; Zhu et al., Reference Zhu, Hao, Zhang and Lan2023) (Supplementary S7). Microplastics also forms DBPs during disinfection (Ghanadi et al., Reference Ghanadi, Kah, Kookana and Padhye2023; Qadafi et al., Reference Qadafi, Rosmalina, Pitoi and Wulan2023), and the appearance of cavities potentially implying the release of DBPs into water (Lin et al., Reference Lin2021).
Chlorination generated intensive nanoscale cavities in PS (Lin et al., Reference Lin2021), which corresponded with our findings where disinfection generated cavities (Figure 1). In the instance of ABS and PA, the smoother surfaces after treatment could imply that DBPs were potentially leached into the water due to the oxidants causing breakage of bonds. Disinfection promotes the leaching of DOM from microplastics which compounds the formation of DBPs (Lin and Su, Reference Lin and Su2022; Xu et al., Reference Xu2022; Zhou et al., Reference Zhou2022). Oxygen functional groups on the surface of polymers following oxidative degradation by disinfection further increases the formation of DBPs (Miao et al., Reference Miao2023) as demonstrated by previous studies on PP (Hao et al., Reference Hao2022) and HDPE (Mitroka et al., Reference Mitroka, Smiley, Tanko and Dietrich2013). Organic intermediates have also been detected following chlorination (Mitroka et al., Reference Mitroka, Smiley, Tanko and Dietrich2013). Reactive chlorine (Cl• and ClO•) and reactive oxygen species (•OH) react with microplastics to form DBP precursors (Miao et al., Reference Miao2023). In essence, microplastics act as DBP precursors, facilitate formation and form DBPs, and act as vectors for DBPs.
In our study, TOC, representing the concentration of organic carbon, was utilized as an indicator of DOM (Shetty and Goyal, Reference Shetty and Goyal2022) to signal the likelihood of DBPs production. Both treatment processes displayed elevated TOC levels, with chlorination exhibiting a greater increase. TOC levels increased with dose, suggesting that greater degradation at higher doses resulted with more carbon in water. Not surprisingly, more TOC was recorded from the experiments with biofouled particles and in RW. The resulting greater concentrations of TOC up to 15 days in chlorinated samples is attributed to the more aggressive nature of hypochlorites compared to chloramines. The greater persistence of chloramine explains the higher TOC in samples at 30 days. Our results indicated increasing TOC in the order ABS < EPS < LDPE < PC < PET < PP < PA, reinforcing that leaching of DOM, and thus potential formation of DBPs, are also dependent on polymer type, which aligns with previous findings (Lee et al., Reference Lee, Romera-Castillo, Hong and Hur2020; Ghanadi et al., Reference Ghanadi, Kah, Kookana and Padhye2023).
Spectral revelations
FTIR analysis presents the chemical transformations by detecting the functional groups at distinct bands and changes in bonds (Supplementary S8). In general, vibrational energy decreases when the strength of bonds decreases. Our results (Figure 3 and Supplementary Figure S2) indicated that the transmittance of untreated microplastics tended to be higher than treated microplastics for all seven polymers, suggesting a weakening of bonds in the polymer structure post-disinfection. The changes in spectra provided evidence that oxidation led to degradation (Table 3).
Table 3. FTIR analysis of seven common polymer types found in water and wastewater
Our results indicated that bond cleavage transpired following disinfection, as seen by the decrease in peak intensities at ~1250–1050 cm−1 (stretching vibration of C–O–C bond) and ~1750–1500 cm− 1 (stretching vibration of C=O bond) on the treated polymers. Cleavage is mostly by β-fission processes to form the fragments with aldehyde, ketone and vinyl end groups (Vohlídal, Reference Vohlídal2021). The results of the spectra in the range 1650–1850 cm−1, with a shoulder and broadening of the band, indicated the production of oxygen-containing carbonyl functional groups including α,β-unsaturated ketone, α,β-unsaturated aldehyde, saturated ketone and saturated aldehyde. The new peaks around 1720 cm−1 correspond to the C=O stretching vibration, and the band around 1745 cm−1 can be associated with the C=O stretching vibration of a ketone group of antioxidants (Fujii et al., Reference Fujii2019).
Carbonyl groups increase as the degree of oxidation increases. Our results indicated that the degree of oxidation increased with dose, as greater formation of carbonyl groups (1650–1850 cm−1) and hydroxyl groups (3200–3650 cm−1)(Li et al., Reference Li, Chen, Yang, Jiang and Dan2017; Senathirajah et al., Reference Senathirajah, Kandaiah, Panneerselvan, Sathish and Palanisami2023) were seen with increasing dose rates. The changes in intensities at ~3300 cm−1 indicated reduced functional hydroxyl groups. The formation of –O-H stretching at 3450 cm−1 broadened absorption indicated changes in the hydrogen bonding between the –O-H groups. The peaks with nominal changes indicated that those structures remained. Distinct variations by polymer type with unique characteristic peaks (Veerasingam et al., Reference Veerasingam2021; Miao et al., Reference Miao2022) were observed (Table 4).
Table 4. Common characteristic peaks observed in the FTIR analysis by polymer types
Studies exposing HDPE pipes to chlorinated water demonstrated reduced antioxidant levels, increased crystalline content and the formation of oxygen-containing functional groups and vinyl groups, surface carbonyl bond formation, polymer chain scission, cracking and broadened molecular weight distribution (Whelton and Dietrich, Reference Whelton and Dietrich2009; Samarth and Mahanwar, Reference Samarth and Mahanwar2021) congruous to our findings of degradation. Consistent exposure to chlorinated water accelerated ageing conditions for HDPE, identified first by the carbonyl functional group with FTIR absorbance at 1715 cm−1, believed to be the precursors to microcracks (Mitroka et al., Reference Mitroka, Smiley, Tanko and Dietrich2013). LDPE demonstrated a higher rate of photo-oxidative degradation than HDPE due to LDPE’s higher frequency of reactive branch points (Chamas et al., Reference Chamas2020), thereby explaining our results (Table 3). LDPE may also contain a small number of unsaturated (C═C) bonds, vinylidenes, in the main chain or at the chain ends that are readily oxidized by radicals to unstable hydroperoxides, then converted to more stable UV-absorbing carbonyl groups (Chamas et al., Reference Chamas2020). The new peak on the LDPE appears around 1760 cm−1, representing the C=O bonds of the tensile vibration of carbonyl groups. We found LDPE degradation to be less than PP.
Degradation of PE (Gulmine et al., Reference Gulmine, Janissek, Heise and Akcelrud2003) and PP (Damodaran et al., Reference Damodaran2015) led to the growing intensity of the carbonyl (C=O) stretching vibration band of saturated aliphatic ketones. It is acknowledged that PP is very vulnerable to oxidative degradation by chain scission due to the higher tertiary carbon content of PP (Samarth and Mahanwar, Reference Samarth and Mahanwar2021), and its aliphatic structure with methyl pendant groups hanging off the main polymer chain, creating more space between chains. The intensity of the peak of the methyl pendant group in PP decreased indicating the weakening of bonds. New peaks on the PP spectrum were assigned to the carboxyl and aldehyde groups.
HDPE and PP were reported to be less resistant to chlorine than PS (Kelkar et al., Reference Kelkar2019). Aligned with another study, we also found EPS produced more organics than LDPE (Ghanadi et al., Reference Ghanadi, Kah, Kookana and Padhye2023) and that EPS degraded at typical operational dose rates (5 mgL−1)(Kelkar et al., Reference Kelkar2019). The study also reported that PS formed C-Cl bonds at high dose rates, increasing the toxicity of the particle (Kelkar et al., Reference Kelkar2019).
PA and PET have been reported to degrade more easily by oxidation than LDPE (Sun et al., Reference Sun2021). However, our results indicated that only PA was more easily degraded than LDPE. Usually, hydrocarbon polymer skeletons with O or N heteroatoms degrade more easily than polymers made of C skeletons alone made from aliphatic chains, as demonstrated by the PA results. Aliphatic polymers have fewer electrons than aromatic polymers which makes them targets for oxidants and more prone to oxidative degradation. The presence of hydroxyl and carboxylic acid functional groups commonly found on aliphatic chains makes aliphatic polymers more susceptible to degradation through chemical reactions including hydrolysis and esterification. During the oxidation reaction, polar carboxyl, aldehyde and hydroxyl groups are added to the surface (Jia et al., Reference Jia, Asahara, Hsu, Asoh and Uyama2020). The changes of functional groups induces changes of polarity which affect hydrophilicity (Kaczmarek et al., Reference Kaczmarek, Kamińska, Światek and Rabek1998), viz. polarity reduces hydrophobicity (Jia et al., Reference Jia, Asahara, Hsu, Asoh and Uyama2020; Xu et al., Reference Xu, Wang and Sun2021) and, in the process, increases susceptibility to hydrolysis.
Structurally, the chemical bonding, including polar attraction between nitrile groups and the aromatic chains in the styrene group and hydrocarbon backbone, is strong in ABS. Polymers with aromatic rings are more resistant to degradation, which our results corroborate. The absence of intensity changes in the carbonyl range indicated the lack of production of carbonyl functional groups in ABS. Increases were seen in the C-O stretching vibrations between 1000 and 1400 cm−1 and the SEM displayed small nanoscale cavities over the surface. ABS is susceptible to nitric and sulphuric acids and by aldehyde, ketones, esters and chlorinated hydrocarbons (PubChem, 2023). Limited information is available regarding the contribution to nitrification from the nitrile group upon degradation of ABS.
The continuous exposure of PC membranes to chlorine ions in solution degraded the cross-linked polymeric membrane structure (Idrees and Umar, Reference Idrees and Umar2022). The changes in intensity of our C=O peak around the carbonyl range of PC aligned with previous findings (Shi et al., Reference Shi2021).
Carbonyl index
FTIR is commonly used to monitor oxidation reactions by determining the carbonyl index (CI) to evaluate the extent of degradation of polymers (Senathirajah et al., Reference Senathirajah, Kandaiah, Panneerselvan, Sathish and Palanisami2023). The CI reflects the degree of oxidation and degradation, indicated by a decrease in the CI (Fotopoulou and Karapanagioti, Reference Fotopoulou, Karapanagioti, Takada and Karapanagioti2019) (CIuntreated > CI disinfected).
As with previous studies (Kowalski et al., Reference Kowalski, Reichardt and Waniek2016; Fujii et al., Reference Fujii2019), the onset of material degradation was observed, as demonstrated by the CI values, after exposure to oxidation particularly at the high dose rate, with fluctuations in the initial 6–12 h, but with a net increase in oxidative degradation with longer immersion time (24 h). The observed increases in CI might be explained by the interference of other contaminants particularly in non-polar, hydrophobic, molecular structures made with a C backbone of aliphatic chains such as LDPE with great gaps between chains that readily adsorbs organic matter (Senathirajah et al., Reference Senathirajah, Kandaiah, Panneerselvan, Sathish and Palanisami2023). Similar to previous findings, the CI values of untreated LDPE were greater than chlorinated LDPE (Sun et al., Reference Sun2021). PA resulted in a decrease in CI, aligning with oxidative degradation resulting from electrocoagulation treatment (Senathirajah et al., Reference Senathirajah, Kandaiah, Panneerselvan, Sathish and Palanisami2023). PP, PA and EPS demonstrated the most degradation in terms of CI, attributable to the susceptibility of their functional groups to undergo oxidation reactions. It has been proposed that CI can be used to predict the decreases in molecular weight as a function of oxidation level (Benítez et al., Reference Benítez, Rodríguez and Casas2021). Thus, it should also serve as an indicator of the extent of DBPs formation, as the release of by-products leads to a reduction in molecular weight. Only PET and PC did not indicate the formation of hydroxyl groups, potentially insinuating lesser DBPs formation from these polymers.
Polymers of concern
Our results are aligned with previous investigations that demonstrate different impacts of disinfection on different polymer types. Our FTIR results indicated heterogeneous degradation as evident by bond cleavage, hydrolysis of ester bonds and changes to functional groups, including carbonyl and hydroxyl, on particles following chlorine-based disinfection. The SEM imagery validated this by showcasing the appearance of cracks, flakes, pits and cavities on the surface of the polymers indicating that chlorination and chloramination induced degradation to varying degrees. Our SEM results highlight that chlorinated EPS, PA, PP and PET present higher adsorptive risks. Increasing the concentrations of both disinfectants resulted in increased degradation. Nominal differences were seen between the RW and PW in the FTIR spectra, but SEM images revealed more impacts on chlorinated polymers in RW which is not surprising as surface morphological changes occur before changes at the molecular level. Polymer properties, including the polymeric molecular structure, affected the susceptibility of polymers to degradation (Plota and Masek, Reference Plota and Masek2020). There was general agreement with previous findings that initially chlorination aggressively attacked polymers, but over longer exposure, the polymers exposed to monochloramine exhibited greater impacts (Torrey, Reference Torrey2013). The PCA showed a decline after the second component (Supplementary Table S1 and Supplementary Figure S7), suggesting that polymer type and biofouling were most significant. In order to assess the PoC, a semi-quantitative approach was adopted. A higher score was assigned to degradation factors with greater impact.
Scoring the seven polymers investigated resulted with PA, PP and EPS as the three highest PoC (Table 5). This comports with the findings of the combined ranking utilising 21 criteria that signalled that PS, PVC, PP and PA were top four high-risk PoC (Senathirajah et al., Reference Senathirajah, Kemp, Saaristo, Ishizuka and Palanisami2022). We did not investigate PVC as it is already a recognized PoC (Lithner et al., Reference Lithner, Larsson and Dave2011; Zhang and Lin, Reference Zhang and Lin2015; Senathirajah et al., Reference Senathirajah, Kemp, Saaristo, Ishizuka and Palanisami2022; Wilcox et al., Reference Wilcox, Fox and Valliant2023). While our findings have certain limitations, assessing the impacts of chlorination and chloramination on common polymers helps to target PoC for efficacious management responses. By highlighting the differences of impacts on polymer types, we enable informed decisions in relation to the choice of materials, treatment processes and operational practices. The results provide evidence for risk-based decision-making to ensure safe environmental and public health. The knowledge, spotlighting PoC could be utilized during deliberations for the development of the United Nations’ Treaty to end plastic pollution.
Table 5. Semi-Quantitative Assessment of Degradation Factors for 7 Polymers to Determine Polymers of Concern.
Hazard assessment – The big picture
Environmental, social and economic impacts are the pillars of sustainability and are interconnected with cascading effects (Figure 5). Disinfection treatment has implications across these pillars, thus we considered all three risks of harm to inform comprehensive decision making (Supplementary S9).
Figure 5. Potable water, wastewater and biosolids risks of harm – factors contributing to the risks encompassing the pillars of sustainability.
Planet – Environmental
Microplastics, fragmented and degraded, readily disperse and circulate in the environment. Weathering forces alters their properties, affecting buoyancy (Gregory, Reference Gregory2009; Erni-Cassola et al., Reference Erni-Cassola, Zadjelovic, Gibson and Christie-Oleza2019) and interactions with other contaminants (Rai et al., Reference Rai, Sonne, Brown, Younis and Kim2022) and organisms (Graham and Thompson, Reference Graham and Thompson2009; Davison and Asch, Reference Davison and Asch2011). The discharge of oxidized microplastics from WWTPs impact ecosystem biodiversity and microbial functions. Our study provided evidence that disinfection corroded the surface and generated flakes, cracks, pits and nanoscale cavities which compound the impacts of buoyancy. Floating microplastics impede sunlight penetration, reducing photosynthesis, threatening organism survival and accelerating climate change (Senathirajah et al., Reference Senathirajah, Bende-Michl, Bhubalan and Palanisami2023). They alter biogeochemical cycles (Wang et al., Reference Wang2016; Romera-Castillo et al., Reference Romera-Castillo, Pinto, Langer, Álvarez-Salgado and Herndl2018; Hu et al., Reference Hu, Shen, Zhang and Zeng2019), compromise water quality and form toxic DBPs (Lin and Su, Reference Lin and Su2022; Xu et al., Reference Xu2022; Zhou et al., Reference Zhou2022). They sorb and desorb contaminants, increasing their toxicological burden (Shen et al., Reference Shen2019; Amelia et al., Reference Amelia2021), and pose risks to ecosystems and the human food web (Alimba and Faggio, Reference Alimba and Faggio2019; Ebere et al., Reference Ebere, Wirnkor and Ngozi2019; Lee et al., Reference Lee, Lee and Kwon2019; Ding et al., Reference Ding2020; Zhang et al., Reference Zhang, Han, Sun and Wang2020; Senathirajah et al., Reference Senathirajah2021) (Supplementary S6).
People – Social
Microplastics, detected in various human tissues, can translocate and accumulate (Schwabl et al., Reference Schwabl2019; Ibrahim et al., Reference Ibrahim2021; Kannan and Vimalkumar, Reference Kannan and Vimalkumar2021; Horvatits et al., Reference Horvatits2022; Jenner et al., Reference Jenner2022; Leslie et al., Reference Leslie2022; Ragusa et al., Reference Ragusa2022; Feng et al., Reference Feng2023; Pironti et al., Reference Pironti2023; Yang et al., Reference Yang2023; Zhao et al., Reference Zhao2023). They contribute to diseases including inflammation, cancers, immune disruption, endocrine disruption, reproductive problems, birth defects, mental health issues, developmental problems, neurological problems, respiratory problems, cardiovascular diseases and lung disease (Jeong and Choi, Reference Jeong and Choi2020; Senathirajah et al., Reference Senathirajah2021; Kumar et al., Reference Kumar2022; Feng et al., Reference Feng2023; Landrigan et al., Reference Landrigan2023; Senathirajah and Palanisami, Reference Senathirajah and Palanisami2023). Long-term exposure to DBPs has serious health ramifications (Waller et al., Reference Waller, Swan, Delorenze and Hopkins1998; Grellier et al., Reference Grellier2010; Grellier et al., Reference Grellier, Rushton, Briggs and Nieuwenhuijsen2015; Wright et al., Reference Wright, Evans, Kaufman, Rivera-Núñez and Narotsky2017), echoing the adverse outcomes from microplastic exposure (Landrigan et al., Reference Landrigan2023). This flags the urgency to protect public health. The financial strain of medical care can be immense and affordability for access to health care could further increase the disparities of wealth inequities and social injustices.
Prosperity – Economic
Compromised water quality can potentially limit economic growth and threaten environmental and human well-being. The economic costs associated with microplastics and DBPs are speculated to exceed US$100 billion/year (Landrigan et al., Reference Landrigan2023) and millions of USD/year,(Chowdhury et al., Reference Chowdhury, Rodriguez and Sadiq2011) respectively, for North America alone. Remediation costs are substantial, estimated at US$162 billion for WTPs and US$176 billion for WWTPs (Senathirajah and Palanisami, Reference Senathirajah and Palanisami2023). Biosolids and treated wastewater disperse microplastics further, impacting soil fertility and crop yields. Microplastics in agriculture can impact soil fertility and alter soil porosity and hydrology, leading to reduced crop yields resulting in less productivity and compromise revenue generation (Gong and Xie, Reference Gong and Xie2020; Senathirajah and Palanisami, Reference Senathirajah and Palanisami2023; Tran et al., Reference Tran2023). Plastics, used in water industry infrastructure, degrade under chlorination and chloramination, leading to asset failure. Substituting plastic materials can be costly and assessments need to be undertaken to ensure that alternatives have no unintended consequences.
Future work
An audit of water infrastructure would identify polymer types and hotspots in the water supply cycle, enabling targeted interventions.
Investigations into safe inhibitors could reduce degradation and extend infrastructure longevity without unintended consequences.
Our study indicates a need for safe dose rates to manage biofilm and reduce polymer degradation risks.
Existing standards like Australian AS/NZS 4020 and American ASTM Standard D6284 do not require manufacturers to disclose plastic ingredients or testing data. Given the risks associated with DBPs and microplastics, our research calls for additional regulations considering microplastics and greater transparency.
Further research on DBPs and DBP precursors from high-use PoC would help determine mitigation measures. Risks of DBPs interacting with microplastics should be quantified. UV and ozone, while not generating DBPs, do interact with microplastics, warranting further study. Strengthening regulations of the Australian (NHMRC, 2022) and American (EPA, U, 2023) Safe Drinking Water Acts considering microplastics would protect public health.
From a public health perspective, it is crucial to develop better detection methods and optimize treatment mechanisms to reduce microplastic exposure. Implementing upstream and downstream strategies with appropriate incentives would yield effective outcomes (Lau et al., Reference Lau2020; Miranda et al., Reference Miranda, Silva and Pereira2020).
Conclusion
This study reveals that chlorination and chloramination at industry-relevant dose rates causes physicochemical changes to seven common polymer types. Our results and discussion further highlight differences between polymer types such as biofilm formation, increased TOC levels, cracks and broken bonds, which poses increased threats including serving as vectors for pathogens, increased DBP formation and particle degradation leading to nanoplastic formation. The changes stress potential public health risks, especially of concern given that the more hazardous polymers showed greater threats. PP, EPS and PA were the most affected polymers, showing significant degradation when exposed to chlorination and, to a lesser extent, chloramination. Therefore, when designing treatment plants and making decisions about material choices, it is advisable to consider the risks of polymer types, especially for water treatment applications involving contact with chlorine-based disinfectants. We emphasize the need to optimize water treatment processes by disinfection type, adjustment of dose rates and pH levels, to minimize the generation of sub-micron particles. This information is crucial for guiding industry and policymakers in developing standards and interventions to reduce harmful particle production. Additionally, these findings should be considered during the development of the global plastic treaty to ensure informed decision making and effective holistic management strategies.
Abbreviations
- ABS
-
acrylonitrile butadiene styrene;
- ARG
-
antibacterial resistant gene;
- ATR-FTIR
-
attenuated total reflectance Fourier transform infrared;
- Cd
-
cadmium;
- CIP
-
ciprofloxacin antibiotic;
- Cl
-
chlorine;
- DDE
-
dichlorodiphenyldichloryethylene;
- DDT
-
dichlorodiphenyltrichloroethane;
- DO
-
dissolved oxygen;
- DOC
-
dissolved organic carbon;
- DOM
-
dissolved organic matter;
- EC
-
emerging contaminant;
- ECHA
-
European Chemicals Agency;
- EPS
-
expanded polystyrene;
- GAC
-
granulated activated carbon
- HDPE
-
high-density polyethylene;
- LDPE
-
low-density polyethylene;
- PA
-
polyamide;
- PAC
-
polyaluminium chloride;
- PAM
-
polyacrylamide;
- PAHs
-
polycyclic aromatic hydrocarbons;
- PB
-
polybutylene;
- PC
-
polycarbonate;
- PCA
-
principal component analysis;
- PE
-
polyethylene;
- PET
-
polyethylene terephthalate;
- PEX
-
crosslinked polyethylene;
- PFAS
-
perfluoroalkyl and polyfluoroalkyl substances;
- PFOS
-
perfluorooctanesulphonate;
- PoC
-
polymers of concern;
- PP
-
polypropylene;
- PVC
-
polyvinyl chloride;
- PS
-
polystyrene;
- PU
-
polyurethane;
- PW
-
potable water;
- RW
-
recycled water;
- SEM
-
scanning electron microscope;
- TOC
-
total organic carbon;
- UV
-
ultraviolet;
- WTP
-
water treatment plant;
- WWTP
-
wastewater treatment plant
Open peer review
To view the open peer review materials for this article, please visit http://doi.org/10.1017/plc.2025.6.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/plc.2025.6.
Acknowledgements
The Research Training Program Scholarship offered by the Commonwealth of Australia through the University of Newcastle is duly acknowledged. The authors also acknowledge the Water Research Australia and the Water Corporation of Western Australia for their ongoing support. The authors express sincere gratitude to Vikash, Tarinee and Nikesh Vimalan for their ongoing support.
Author contribution
Kala Senathirajah: conceptualization, methodology, investigation, formal analysis, writing – original draft, review and editing Raji Kandaiah and Logheswaran Panneerselvan: investigation, writing – review and editing, Kyana Young: Writing – review & editing, Thava Palanisami: supervision, funding acquisition.
Financial support
This project was partially funded as part of the Water Research Australia PhD Scholarship. The authors declare that they have no known competing financial interests or personal relationships that could have influenced the work reported in this article.
Competing interests
All authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.
Open access
Comments
23 August 2024
Dear Editor-in-Chief,
I am writing to submit our research article, ‘Disinfection impacts: Effects of different disinfection treatments on common polymer types to guide the identification of polymers of concern in the water industry’, for consideration to be published in your highly esteemed scientific journal.
We would like to present a study on the impacts of chlorination and chloramination, which are common disinfection treatment processes for water and wastewater treatment. Microplastics are prevalent throughout the water supply cycle including in water and wastewater. Our study investigates the fate and transformation on microplastic particles resulting from chlorination and chloramination on seven polymer types to enable targeted mitigation measures. Our research will make a novel contribution to the water industry and potentially contribute knowledge to development of the global treaty to end plastic pollution by highlighting polymers of concern. These findings will underpin treatment decisions and contribute to future decision-making by the water sector when addressing microplastic pollution.
We believe that the findings will appeal to the scientific community, the policy makers and also the general public who refer to your journal for most up to date information on critical environmental topics.
Each of the authors confirms that this manuscript has not been previously published and is not currently under consideration by any other journal. All the authors have approved the contents and agree to the Cambridge Prisms: Plastics' submission policies. In addition, all authors have no conflict of interest, financial or otherwise.
Thank you for your time and consideration.
Sincerely,
Kala Senathirajah