Innovative Approaches for Minimizing Disinfection Byproducts (DBPs) in Water Treatment: Challenges and Trends

Abstract

Growing concerns over public health and environmental safety have intensified the focus on minimizing harmful disinfection byproducts (DBPs) in water treatment. Traditional methods like chlorination, while effective against pathogens, often lead to the formation of DBPs, which pose significant risks. This paper explores alternative strategies to reducing DBP formation while ensuring effective disinfection. The methodology involved a bibliographic study conducted through the Scopus platform, using appropriate keywords. The initial search yielded 9576 articles from the period 2020 to 2024. The key approaches identified include advanced oxidation processes (AOPs) such as UV/H₂O₂ and ozone, which mineralize natural organic matter (NOM) and minimize chemical use and sludge production; membrane-based filtration systems, like reverse osmosis, effectively removing contaminants without chemical disinfectants, reducing DBP risks. Furthermore, conventional processes, such as coagulation and filtration, serve as crucial pretreatment steps to lower NOM levels before disinfection. Additionally, optimizing chlorine dosing, using non-chlorine disinfectants, and employing post-disinfection methods like adsorption and biological filtration further mitigate DBP formation. Finally, the integration of artificial intelligence in process optimization is emerging as a promising tool for enhancing treatment efficiency and safety. This research contributes to the development of safer, more sustainable water treatment solutions, addressing regulatory demands and public health objectives.

1. Introduction

Drinking water disinfection technology has evolved and continues to be refined to reduce the transmission of pathogens [1]. Chlorination is the primary disinfection method, a practice that dates to the early 20th century. Chlorine is globally the most commonly used disinfectant for controlling the taste, color, odor, and proliferation of microorganisms in water [2,3,4]. Its enduring popularity is owed to its efficacy in neutralizing pathogenic microorganisms responsible for waterborne illnesses [5].

Due to its effectiveness and affordability, the preference for chlorine disinfection persists in municipal water supply networks. Chlorine has long been recognized for maintaining residual concentrations to prevent microbial growth within water distribution systems. Its widespread use in disinfection has significantly reduced waterborne diseases such as cholera, typhoid fever, and dysentery, thereby safeguarding public health [6].

However, disinfectants such as chlorine, chlorine dioxide, chloramines, and ozone can react with natural organisms in the water or pollutants of human origin, such as halogenated solvents, pharmaceuticals, and pesticides, resulting in the formation of a spectrum of halogenated compounds known as disinfection byproducts (DBPs) [1]. These compounds arise from the interaction of chlorine, bromine, or iodine with naturally present organic matter in water [5]. Numerous studies have indicated that these compounds lead to adverse health effects in humans through various exposure pathways (ingestion, inhalation, dermal contact) [7,8,9,10]. Especially, DBPs are associated with an increased risk of cancer, growth retardation, spontaneous abortion, and congenital heart defects [11].

Specifically, when chlorine is introduced to water, some of it initially reacts with the inorganic and organic materials, as well as and metals present, rendering it unavailable for disinfection. This is known as the chlorine demand of water. The remaining quantity is called total chlorine and is divided into combined and free chlorine. Combined chlorine is a mild disinfectant and becomes ineffective for disinfection when it reacts with inorganic substances such as nitrates and organic nitrogen-containing molecules like urea. Free chlorine represents the remaining chlorine available to deactivate pathogens, indicating water potability. Therefore, the total chlorine required is the sum of combined chlorine and free chlorine [6].

DBPs are found in various types of water that undergo disinfection before being used or discharged. They are formed during water treatment processes, such as drinking water treatment processes, water distribution, seawater desalination, pool water regeneration, aquaculture water disinfection, ballast water treatment, hospital and domestic wastewater treatment, equipment cleaning, and others. During the discharge of disinfected recycled water, DBPs are transferred to surface waters, marine waters, and groundwater [12].

2. Methodology

This work aims to examine and collectively present the techniques that can be applied to minimize the formation of DBPs, as reported by relevant research worldwide, as the technology is evolving, including optimizations of procedures and recent innovative approaches.

The bibliographic study was performed by searching terms such as “photocatalysis”, “electrochlorination”, “ozonation”, “chloramination”, and “coagulation” on Scopus platform terms. The initial search resulted in 9576 articles (Figure 1) from 2020 to 2024 period. Further, only relevant papers were selected by meeting the criteria related to disinfection byproducts in water. Figure 1 shows that the term “adsorption” returned the highest number of articles.

Advanced oxidation processes (AOPs) and photocatalysis are widely applied, followed by activated carbon removal techniques and alternative disinfection methods. Combinations of methods are also frequently reported, while more recent advances include the use of machine learning approaches to predict/model DBP formation, and to enhance the effectiveness and aid in the selection and application of the above-mentioned methods to obtain DBP-free water.

3. DBPs

3.1. DBPs Categories

DBPs can be classified into two main groups based on their molecular composition: carbon-containing disinfection byproducts (C-DBPs) and nitrogen-containing disinfection byproducts (N-DBPs). C-DBPs, such as haloacetic acids (HAAs) and trihalomethanes (THMs), are the most detected DBPs in drinking water. On the other hand, N-DBPs, which include haloacetonitriles (HANs), halonitromethanes (HNMs), haloacetamides (HAcAms), and nitrosamines (NAs), are generally more toxic than C-DBPs [1].

Nitrogenous disinfection byproducts (N-DBPs) and iodinated disinfection byproducts (I-DBPs) are becoming increasingly concerning in water treatment. Among N-DBPs, halogenated types like HANs, HNMs, and HAcAms have attracted attention due to reports indicating their higher toxicity compared to regulated DBPs. Iodinated disinfection byproducts (I-DBPs) are generally more cytotoxic and genotoxic than their chlorinated and brominated counterparts. I-THMs were first identified in drinking water in the 1980s and have since received more attention. Besides I-THMs, various other categories of I-DBPs have been found in drinking water, including iodinated acids, different polar I-DBPs, and iodinated N-DBPs, such as iodinated HAMs (I-HAMs) [7].

Recently, heightened focus has been placed on water treatment processes, driven by growing public health and safety concerns. Various countries have implemented stringent regulations concerning both water quality and its sources. These regulations aim to guarantee drinking water’s safety by eliminating hazardous substances or minimizing their concentrations to the lowest possible levels [6].

3.2. Health Effects of DBPs

C-DBPs in tap water raise significant concerns due to their suspected carcinogenic properties [3]. Epidemiological and toxicity studies have linked prolonged consumption of chlorinated water to adverse health effects, including spontaneous abortions, congenital disabilities, and an elevated risk of gastrointestinal and urinary tract cancers. DBPs have various harmful effects, including cytotoxicity, mutagenicity, genotoxicity, carcinogenicity, and teratogenicity, posing significant health risks through different exposure routes. They have diverse precursors, including humic acid, carbohydrates, lipids, and amino acids [1,13].

Additionally, DBPs discharged into the natural aquatic environment can be toxic to aquatic organisms, such as zooplankton, algae, crustaceans, fish, and others [4]. Exposure to DBPs in drinking water over a long period, whether through drinking, breathing, or skin contact, raises the likelihood of adverse health outcomes. The International Agency for Research on Cancer (IARC) classifies specific forms of THMs and HAAs as Group 2B, indicating a potential heightened risk of liver cancer [10]. Research indicates a link between prolonged exposure to high concentrations of THMs in drinking water and a higher likelihood of developing bladder cancer. Additionally, such exposure could potentially lead to reproductive and growth-related impacts. The mechanism is believed to be related to the carcinogenic properties of certain THM compounds [14]. Furthermore, the genotoxic and overall toxic properties of more than 100 DBPs have been identified by examining thousands of water samples. Further, 22 DBPs (excluding various nitrosamines) have been subject to cancer assessments in rodent studies [14]. As a typical N-DBP, dihaloacetonitrile (DCAN) exhibits higher levels of cytotoxicity and genotoxicity when compared to other prevalent DBPs [15].

Looking back at the history of DBPs, the initial discoveries revolved around volatile compounds. In 1974, chloroform was the first DBP discovered among the volatile compounds in treated drinking water. By 1976, its link to hepatocellular carcinomas in mice and renal tumors in rats raised considerable alarm about the possible health effects of DBPs. This led to regulatory measures by the United States Environmental Protection Agency (USEPA) in 1979, targeting a group of volatile DBPs with similar structures, including THMs like bromodichloromethane and dibromochloromethane [9]. In 1986, the USEPA suggested establishing a maximum contaminant level (MCL) of 80 mg L⁻¹ for four THMs as part of Stage I regulations, aiming to later reduce it to 40 mg L⁻¹ under Stage II guidelines to mitigate human health hazards. They also enforced a cap of 60 mg L⁻¹ on HAAs, encompassing the combined concentration of five species in treated water. In contrast, the United Kingdom and the European Union set the limit for THMs at 100 mg L⁻¹ [6].

After over 40 years of research, more than 800 DBPs have been identified [1,16]. There is a reasonably good understanding of the mechanisms behind their formation and the concentrations at which they occur. Several treatment and disinfection methods have been developed and put into practice to reduce DBP levels while ensuring the safety of drinking water from a microbiological standpoint. Most developed nations utilize advanced water treatment technologies and have established regulations or guidelines concerning DBPs. For example, the USEPA lowered the permissible concentration of the combined four THMs from 100 to 80 μg L⁻¹ and regulated seven more DBPs. As a result of these research endeavors, the potential global risk of bladder cancer linked to drinking water in many regions has likely decreased compared to previous conditions. This highlights the collaborative efforts of numerous scientists, supported by both public and private sectors, working to equip regulators with the necessary tools and knowledge to improve public health. Moreover, a hypothesis has been formulated based on mechanistic studies conducted in vitro, in rodents and humans, seeking to elucidate how drinking water could contribute to cancer risk in a specific population subset with adequate exposure and a predisposing genotype. Further investigations, ranging from fundamental research to molecular epidemiology, are crucial to validate this hypothesis and explore additional models that could shed light on cancers associated with drinking water exposure. In addition to volatile THMs, the discovery of HAAs represented another milestone in the 1980s, identifying a second class of DBPs with low volatility. Dichloroacetic acid and trichloroacetic acid, commonly found HAAs, exhibit elevated concentrations in chlorinated water. The adverse health effects of THMs and HAAs prompted regulatory agencies worldwide to establish guidelines regulating their presence in treated drinking water [17].

Subsequently, attention shifted from aliphatic THMs and HAAs to aromatic DBPs. In 1987, the heterocyclic compound 3-chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone was the first aromatic DBP to be identified. Typically, heterocyclic aromatic DBPs like halopyrroles and halopyridines have a lower stability in treated drinking water due to their heteroatomic rings’ reactivity, rendering them less stable. In contrast, aromatic DBPs with phenyl structures demonstrate excellent stability. Consequently, identifying phenyl DBPs has become a prominent area of research due to their relatively high toxicity [18]. NAs, likely human carcinogens, pose a significant risk to the safety of drinking water consumers [19].

DCAN has been detected numerous times during the disinfection process, drawing considerable attention. Dissolved organic nitrogen (DON) serves as the primary precursor substance for N-DBPs, and the effective control of N-DBP sources predominantly involves the removal of DON. Relative to natural organic substances, DON typically manifests characteristics such as low molecular weight and low electrostatic charge [15].

3.3. DBPs Precursors—Natural Organic Matter (NOM)

Natural organic matter (NOM) is widely regarded as the primary precursor of DBPs. It consists of a complex blend of organic materials with diverse hydrophobic and hydrophilic components, naturally occurring in water sources due to various hydrological, biological, and geological processes. NOM’s molecular structure and chemical composition can vary depending on the location and time. NOM displays a broad spectrum of chemical compositions, molecular sizes, functional group chemistries, solubilities, and polarities. Due to its diverse composition, NOM does not have a defined set of functional groups, size distributions, acid–base properties, or chelating capabilities [20,21]. The prevailing view is that the main precursors of DBPs in NOM are hydrophobic compounds with molecular weights typically between 1 and 10 kilodaltons (kDa) [22].

The makeup of a humic substance differs based on the source water, and whether it originates from surface water (from lakes, rivers, or watersheds), wastewater, or groundwater influenced by different human activities. Given this intricate structural and compositional diversity, the process by which DBPs are formed from NOM remains unclear [23]. Organic substances are present in all water types, with the highest concentrations typically found in surface waters, lower concentrations in infiltration waters, and the weakest in groundwater. The removal of NOM during the water treatment process is a key objective. These substances add color to the water, aid in the transport of micropollutants, and can adversely affect the taste and odor of the water. Humic substances are estimated to comprise approximately half of water’s total organic carbon content [24]. The decrease in NOM can impact the efficiency of chlorine disinfection by influencing the necessary amount of disinfectant and the formation of DBPs. It is recognized that the biologically resistant fractions of humic and fulvic acids in NOM, typically originating from external sources, exhibit higher reactivity with chlorine. The reduction in NOM can impact the effectiveness of chlorine disinfection in a couple of ways: (a) decreasing NOM levels can lower the chlorine demand, meaning less chlorine is required to achieve the desired disinfection, (b) the biologically refractory humic and fulvic acid fractions from allochthonous (external) sources are more reactive with chlorine. Lowering these NOM components can reduce the formation of potentially harmful DBPs during the disinfection process. In other words, reducing the NOM content of the water can make chlorine disinfection more efficient by minimizing the chlorine dose needed and limiting the generation of undesirable DBPs as byproducts. This is a crucial consideration in water treatment processes. A conventional drinking water treatment process typically involves a series of steps, including coagulation, flocculation, sedimentation, filtration, and disinfection to treat the incoming raw water. The main goals are to reduce the total organic carbon (TOC) content and turbidity to the required levels and to produce final water with minimal taste and odor concerns. Apart from these parameters, the treated water must also meet regulatory standards like the total coliform rule, maximum residual disinfectant level (MRDL), and Stage II Disinfectants and Disinfection Byproducts (D/DBP) Rule. In recent years, numerous feasible distribution system models have been investigated to assist water utilities in making informed operational decisions. However, advanced technologies aimed at enhancing the efficiency of TOC removal have often been associated with high costs and increased complexity. Striking a balance between TOC reduction and chlorination disinfection has been a significant concern for environmental regulators and utility managers since the late 1970s, when research began to elucidate the direct relationship between the formation of THMs and the reaction between TOC and free chlorine. This sustained focus underscores the importance of optimizing the water treatment process to mitigate the formation of potentially harmful disinfection byproducts while maintaining the necessary level of microbial disinfection to protect public health [25].

While DBPs in drinking water pose a significant environmental health risk, the main objectives of drinking water disinfection are pathogen inactivation and the prevention of regrowth. This underscores the necessity of chlorine disinfection for the foreseeable future. Hence, striking a balance between pathogen inactivation and managing DBP formation is a complex yet crucial global concern [14]. The decrease in DBPs is typically accomplished by physically removing organic precursor materials through coagulation or enhanced coagulation along with flocculation, sedimentation, and filtration processes like granular activated carbon or membrane filtration. Utilities can also adopt alternative disinfection methods such as ozonation or ultraviolet disinfection to achieve disinfection objectives while minimizing the requirement for chlorine. However, some of these alternative disinfectants may also generate DBPs (for example, the production of bromate from ozonation) or may be less efficient in deactivating specific pathogenic microorganisms [26].

Typically, the water supply system includes coagulation and flocculation, sedimentation, and sand filtration, followed by disinfection, before the water is distributed through the network. Sedimentation and sand filtration processes are efficient in eliminating suspended solids and turbidity. However, dissolved organic matter (DOM), which consists of molecules smaller than 0.45 µm, such as dissolved organic carbon (DOC), humic acid, and fulvic acid, cannot be effectively removed by the sand filter tank. The most used disinfectants in the disinfection process are chlorine, chloramines, ozone, and chlorine dioxide.

The classical water treatment steps include the following:

• Coagulation and flocculation: This help to agglomerate small particles into larger, more easily settleable flocs.

• Sedimentation: The flocculated particles settle from the water during this sedimentation stage (Figure 2).

• Bed filtration: The water then undergoes filtration through sand (Figure 3), gravel, and other materials, effectively removing suspended solids and turbidity.

• Disinfection: Finally, the water is disinfected, commonly by using chlorine, chloramines, ozone, or chlorine dioxide, before being distributed through the supply network.

While sedimentation and sand filtration processes are efficient at removing suspended particles, they are less effective at eliminating DOM, such as humic and fulvic acids, which are smaller than 0.45 μm in size. The sand filtration step cannot efficiently remove this dissolved organic material [27].

The treatment process commonly used, granular activated carbon (GAC) adsorption (Figure 4), only removes a small portion of NOM due to its heterogeneous nature. As a result, it may not adequately address halogenated DBPs in treated water. Efforts have been focused on understanding the mechanistic formation of regulated THMs and HAAs during the chlorination of NOM to recommend effective strategies for controlling DBPs. Humic substances present in NOM contain numerous unsaturated carbon bonds and aromatic structures, particularly active phenolic groups, which significantly generate HAAs, THMs, and other halogenated DBPs [28].

It has been noted that hydrophilic and low-molecular-weight NOM tend to react preferentially with bromine and iodine, whereas hydrophobic and high-molecular-weight NOM typically result in an increased production of halogenated chlorination byproducts. NOM exhibits diverse chemical compositions, varying from simple aliphatic structures to complex-colored aromatic substances, encompassing a broad spectrum of organic compounds. A standard method of separating NOM into hydrophobic and hydrophilic fractions involves identifying hydrophobic acids, mainly known as humic substances, as the predominant component of DOC in water. On the other hand, the hydrophilic fraction is usually found in lower relative concentrations and comprises molecular groups like carbohydrates and proteins [20].

NOM, commonly present in surface waters, is the primary source of DBP precursors. Effective control of DBP formation often involves the removal of NOM. Techniques such as coagulation, activated carbon adsorption, and membrane filtration are standard methods employed in drinking water treatment plants (DWTPs) to eliminate NOM. Combining conventional and advanced treatment processes is a promising approach to improve drinking water quality. However, a systematic assessment of the efficiency of full-scale conventional and advanced treatment processes in removing DBP precursors is crucial. NOM, a complex mixture with diverse components of different sizes, structures, and properties complicates this process. These NOM components exhibit varying reactivities in DBP formation, directly impacting the types and amounts of DBPs formed. Furthermore, removal efficiency depends on the specific types of NOM components present. Therefore, a thorough characterization of NOM is crucial for accurately evaluating the removal of DBP precursors in drinking water treatment [13].

Typically, drinking water sources contain NOM, a complex mixture of organic compounds from decaying plant and animal material, and microbial activity. This includes primarily humic and fulvic acids that react with chlorine to generate various chlorinated DBPs. In addition to THMs, the most common DBP group, HAAs, are the second most prevalent. Furthermore, treated water has been found to contain various other groups of DBPs, such as HANs, haloketones (HK), chloral hydrate (CH), chloropicrin, chlorophenols, N-chloramines, halofuranones, bromohydrins, chloramino acids, and cyanogen bromide [6].

Over the past two decades, elevated levels of NOM have been observed due to global warming, acid rain, soil erosion, and water pollution [15,27]. This increase has compromised the efficacy of water treatment processes, leading to a shortened lifespan of activated carbon resources [10].

4. DBP Minimization Approaches

4.1. DBP Precursors Removal

Numerous methods for removing NOM in drinking water treatment have been investigated, including coagulation and flocculation, sedimentation (Figure 5), activated carbon adsorption, advanced oxidation processes, and membrane-based approaches [21]. Conventional drinking water treatment processes like coagulation and filtration serve as effective pretreatment steps before disinfection, helping to lower DBP formation by reducing the concentration of NOM [1,29].

Following their formation, methods such as coagulation/flocculation, adsorption, biological filtration, aquifer storage and recovery, ozonation/ultrafiltration, AOPs, or hybrid treatment processes are commonly used for removing THMs from drinking water post-disinfection.

Various methods that are aligned with biological or chemical technologies have been developed for the reduction/removal of THM levels from water, with an emphasis on the concepts of preventive action, through non-chlorine disinfectants or optimization of the chlorine dose, and treatment action that removes precursor substances with ion exchange resin or membrane filtration before chlorine addition. Following their formation, methods such as coagulation/flocculation, adsorption, biological filtration, aquifer storage and recovery, ozonation/ultrafiltration, AOPs, or hybrid treatment processes are commonly used for removing THMs from drinking water post-disinfection [30].

Innovative technologies like AOPs are being closely examined. They are becoming increasingly attractive as they can mineralize NOM instead of just capturing and transferring it to another phase. Additionally, AOPs may not necessitate significant chemical doses or pH adjustments and do not produce sludge [22].

Water treatment plants are exploring new and cost-effective advanced treatment techniques to decrease DBPs to meet regulatory requirements and tackle the challenges encountered by municipal water systems. In recent years, the prominent technologies for removing NOM from water have included enhanced coagulation (EC), electrochemical methods, advanced oxidation methods (AOM), membranes, and adsorption [10].

Recently, heightened focus has been placed on water treatment processes, driven by growing public health and safety concerns. Various countries have implemented stringent regulations concerning both water quality and its sources. These regulations aim to guarantee drinking water’s safety by eliminating hazardous substances or minimizing their concentrations to the lowest possible levels. Employing disinfection alternatives can enhance the reduction in DBPs formation [11].

4.2. Water Treatment Methods Modifications

Due to extensive water pollution, the precursors of DBPs in contaminated water sources typically consist of a mixture of autochthonous and allochthonous compounds. These include substances linked to microorganisms (such as algae and bacteria), organic matter from wastewater discharges, amino acids, and proteinaceous compounds. Generally, conventional water treatment methods encounter difficulties in effectively removing these DBP precursors. Pre-oxidation treatments like pre-ozonation and advanced post-treatments like ozone-biological activated carbon (O₃-BAC)-integrated technologies are commonly incorporated into existing drinking water treatment plants. These technologies are developed to eliminate contaminants in polluted water sources, including ammonia nitrogen, organic matter, trace pollutants like algal metabolites, odor and taste compounds, pharmaceuticals, endocrine-disrupting chemicals, and persistent organic pollutants. Improved removal of organic precursors through pre-ozonation and advanced O₃-BAC treatment significantly decreases the formation of regulated carbonaceous DBPs (e.g., chloroform) during subsequent chlorination processes. Additionally, pre-ozonation has been recognized as a practical approach for mitigating the iodinated disinfection of byproducts by converting iodide to iodate [7].

There are several options available to control DBPs concentrations. These options include removing the precursor NOM before introducing chlorine, eliminating DBPs after formation, or using alternative disinfectants. Various cost-effective strategies are implemented, aiming to reduce the formation of THMs and HAAs during the chlorination of drinking water [6]:

• Eliminating prechlorination or relocating the chlorination point: by avoiding chlorination at the initial stages of water treatment or changing the location where chlorine is introduced, the formation of THMs and HAAs can be reduced.

• Implementing enhanced coagulation practices: enhanced coagulation techniques help to remove organic precursors before chlorination, thereby reducing the formation of DBPs.

• Optimizing chlorine dosing using disinfection benchmarking: water treatment plants can ensure effective disinfection while minimizing DBP formation by closely monitoring chlorine dosage and comparing it with established benchmarks.

• Transitioning to chloramines for secondary disinfection: chloramines, compounds formed by combining chlorine with ammonia, are less reactive than free chlorine and can help decrease the formation of THMs and HAAs.

• Exploring alternative DBP minimization strategies: this includes investigating novel treatment methods or combinations of treatments aimed at reducing the formation of DBPs during water chlorination.

4.2.1. Eliminating Prechlorination

Prechlorination, the initial stage in water treatment, involves the addition of chlorine. However, a significant challenge arises due to the high concentration of NOM present in raw water. When chlorine reacts with NOM, it forms DBPs such as THMs and HAAs. Hence, it is recommended that chlorination be postponed until NOM is extracted from the water via coagulation, sedimentation, and filtration. While eliminating pre-chlorination may appear appealing, it is not feasible as it has conventionally served to inhibit the formation of biological slime in treatment plants. To tackle this challenge, water treatment facilities should consider alternative oxidants such as aeration or permanganate for oxidizing chemical contaminants such as iron. In such instances, determining and applying the optimal chlorine dosage is crucial [6].

4.2.2. Switching to Chloramines for Secondary Disinfection

Transitioning to chloramines for secondary disinfection presents an effective strategy for mitigating DBP formation, such as of THMs and HAAs. Unlike free chlorine, chloramines do not generate THMs and HAAs during chlorination. Moreover, maintaining residual disinfectant levels in the distribution system is more manageable with chloramines due to their enhanced stability compared to chlorine. However, it is vital to meticulously regulate the dosages of chlorine and ammonia to prevent issues such as di- and trichloramine formation, as well as THMs and HAAs. Inadequate control may result in the absence of residual disinfectants, fostering nitrification and microbial regrowth. One notable byproduct of chloramination is N-nitrosodimethylamine (NDMA), which is subject to specific drinking water quality standards such as 9 ng L⁻¹ in Ontario and a notification level of 10 ng L⁻¹ in California [6].

The effectiveness of precursor removal is site-specific and can vary depending on various factors, including the characteristics of the source water and the treatment methods employed. Aluminum sulfate (alum) and iron chloride (ferric) salts exhibit variable capabilities in removing NOM. In the case of alum, the pH range of 5.5–6.0 is considered optimal for NOM removal. Adding alum tends to lower pH levels, potentially facilitating in achieving the optimal pH range without the need for additional acid. However, in cases where water has extremely low or high alkalinity, adding either a base or an acid may be necessary to attain the ideal NOM coagulation pH [11].

4.3. Coagulation

Coagulation operates by destabilizing NOM particles using positively charged metals, forming larger aggregates (flocs) that facilitate the settling of organic pollutants. This method effectively removes NOM from water through three main mechanisms: colloid destabilization, precipitation, and co-precipitation. Many water treatment plants employ iron salts (ferric chloride and ferrous chloride) and alum as coagulants. These coagulants aid in the settling of NOM, reducing the solid load on filters. Coagulation efficiency is influenced by factors such as coagulation conditions, NOM characteristics, the type and concentrations of inorganic compounds, and the design and operation of the treatment plant. Research indicates that ferric chloride is more effective in removing DOC than alum. The optimal pH for coagulation is around 5.25 for ferric chloride and 5.50 for alum. Additional treatments, such as filtration, are necessary to further reduce NOM and TOC [23].

Coagulation effectively reduces the concentration of high molecular weight organic matter, such as humic acids, while more minor organic compounds are typically removed through adsorption processes using activated carbon. Activated carbon also adsorbs substances that can act as precursors to DBPs. Therefore, combining both methods is advantageous as it helps reduce the organic matter content across a broad range of molecular weights. Generally, powdered activated carbon (PAC) is introduced into the treated water during coagulation. The entry point is the rapid mixing chamber, with activated carbon added after the coagulant [24].

In the coagulation stage of the water treatment process, coagulants like aluminum sulfate and ferric chloride are commonly used. The formed precipitate is then allowed to settle in a sedimentation basin before the water moves to the filtration stage, where disinfection occurs. However, studies conducted in Missouri have shown that in many small-scale utility facilities, the floc formed is often too small or lightweight to settle effectively. As a result, it carries over to the final clarification stage and enters the filtration stage, leading to more frequent backwashing cycles. This situation also allows for higher concentrations of TOC to react with the added chlorine, resulting in elevated levels of DBPs even before the treated water exits the treatment system. Factors such as detention time in the primary and secondary basins of a two-stage water treatment process, floc density (dense floc settles more readily), raw water quality, and the type and dosage of coagulant can all influence the effectiveness of the treatment process in removing TOC. This highlights the importance of optimizing the coagulation and sedimentation stages to maximize the removal of organic precursors, as their subsequent reaction with the disinfectant can lead to potentially harmful DBPs in the finished drinking water [25].

According to Dong et al. [14], the effectiveness of removal through coagulation can be influenced by the type of coagulants used and the physicochemical properties of NOM. The removal of DOC under conditions of enhanced coagulation with ferric chloride can reach up to 57%, which is significantly higher than that achieved with poly aluminum chloride. High-molecular-weight dissolved organic matter and hydrophobic dissolved organic matter are removed much more quickly than low-molecular-weight dissolved organic matter and hydrophilic matter. The interactions between the coagulant and NOM particles also affect the coagulation efficiency. Compared to prechlorination followed by coagulation, the effectiveness of DOC removal with coagulation followed by post-chlorination is higher. Although more DBPs are produced during post-chlorination than pre-chlorination at pH 7.5, fewer DBPs are produced at pH 5.5, highlighting the importance of pH control during the process [14].

The effectiveness of DCAN remains limited in conventional treatment processes (such as coagulation, flocculation, sedimentation, and sand filtration) that remove DBPs of concern. The evaluation of the performance of conventional treatment processes in eliminating precursor HANs, noting an approximate removal rate of 28% for DON, yet they proved ineffective in efficiently removing HANs’ precursor substances [15]. Cuthbertson et al. observed that precursor substances of N-DBPs could be effectively eliminated if biodegradation preceded disinfection. Additionally, findings from water quality analyses have revealed that the biofiltration process exerts a positive influence on pollutant removal, particularly concerning organic micropollutants [15].

4.4. Filtration

During filtration, a filter separates suspended solids from a solid–liquid mixture as it flows through the filter medium. Filters can remove suspended solid particles, dissolved organic carbon/natural organic matter (DOC/NOM), and TOC from untreated water. Various media are utilized in filtration, including GAC, sand, fabric, and ceramic filters. The depth of a slow sand filter is significant for effectively removing DOC from untreated water. Studies have shown that the efficiency of removing solid particles varies depending on the type of filtering media, their depth, and the duration of filter operation. Biodegradation occurring within GAC filters also plays a crucial role in enhancing the efficiency of TOC removal. Initially, the DOC removal efficiency in GAC filters ranges from 77 to 81%, but after 200 days, it decreases to 13% [29].

However, the removal’s effectiveness depends on the nature of the organic matter, and despite mitigation efforts, more than 800 DBPs have been identified in treated water [1].

4.5. Advanced Water Treatment Technologies

A variety of advanced water treatment technologies [6] are also available that can provide effective TOC removal, such as the following:

• Enhanced coagulation;

• Activated carbon;

• Membrane filtration;

• Magnetic ion exchange (MIEX) process.

These advanced approaches have demonstrated the ability to reduce TOC levels significantly. However, the challenge lies in their economic feasibility, particularly for small-scale utility operations. The high capital and operational costs associated with these more sophisticated technologies often make them prohibitive for implementation by smaller water providers. This presents a significant barrier to the widespread adoption of these TOC removal techniques, even though they can improve the overall treatment process and reduce the formation of DBPs [25].

4.5.1. Enhanced Coagulation

Wastewater treatment plants (WWTPs) employ a series of complementary unit operations to remove NOM throughout the treatment process effectively. The typical approach involves combining coagulation/flocculation with filtration-based processes. Various studies have shown that optimizing coagulation and flocculation significantly influences the removal of organic components. Additionally, sand filters and GAC beds are crucial in removing specific NOM fractions. Following coagulation, ultrafiltration (UF) membranes have emerged as promising and robust methods for water quality treatment, capable of operating as hybrid systems. In this context, monitoring parameters to control membrane fouling can complement NOM characterization. Therefore, integrating enhanced coagulation models with expert rules derived from membrane experiments can enhance the performance of DWTPs [31].

DWTPs have implemented various analytical techniques to identify and eliminate NOM content and have increased the use of sensors to monitor NOM, adjusting treatment processes according to environmental conditions. Characterizing and quantifying NOM at multiple stages is essential for a resilient treatment process, allowing for evaluating treatment efficiency and determination of NOM compound reactivity levels. The waterborne NOM content is monitored using online sensors and analyzers, generating continuous large volumes of data. These data can be processed and integrated into an Environmental Decision Support System (EDSS), providing valuable insights into the capabilities of DWTP processes in reducing NOM and associated DBP formation. EDSS tools, incorporating artificial intelligence (AI), have been developed to manage system complexity and expedite decision making. These tools combine quantitative and qualitative elements, integrating mechanistic and statistical models with other AI techniques. For instance, in a practical scenario, process parameters like turbidity and UV254 were utilized to develop an artificial neural network for predicting oxidant demand in a full-scale DWTP [31].

The pH fluctuations in DOM play a crucial role in controlling the potential formation of DBPs. DOM in surface waters, transported from watershed areas during runoff events, is a significant source of DBP precursors. Moreover, the physicochemical characteristics of soil, including pH, ion content, and texture, influence and break down terrestrial DOM within the watershed. Soil pH can vary widely, from highly acidic (pH < 4) to highly alkaline (pH > 10), due to factors like rainfall patterns, temperature changes, organic matter decomposition, ammonia nitrification, and agricultural activities. The chemistry of source water, particularly its pH, impacts the reactivity of DOM with disinfectants like chlorine and influences ozone decomposition rates in AOPs.

This disruption ultimately hinders the degradation of organic matter and DBP precursors. The diverse organic matter composition presents a notable challenge in determining the removal mechanisms during the oxidation and separation processes mentioned above, necessitating the individual monitoring of the different DOM components during their elimination. In this regard, excitation–emission matrix (EEM) fluorescence spectroscopy and ultraviolet-visible (UV-Vis) spectroscopy, along with size exclusion chromatography (SEC), are valuable techniques that can effectively provide insights into the heterogeneous distribution of organic content in DOM [32].

4.5.2. Activated Carbon

Adsorption using carbon-based adsorbents has been employed since the 18th century and continues to be one of the most preferred and promising techniques for water treatment, especially for eliminating DOC and harmful contaminants at low concentrations. This method is favored for its ease of operation, minimal maintenance needs, high efficiency, and straightforward design. However, the cost-effectiveness of adsorption methods relies heavily on the accessibility and cost of carbon-rich precursor materials like wood, petroleum, coal, and coconut shells [10].

Industrial activated carbon, used for odor elimination, the filtration of hazardous materials, and water treatment, has historically consumed thousands of tons of wood, coal, and petroleum annually. Considering economic and environmental constraints, new technologies have emerged to manufacture activated carbon from cost-effective, carbon-rich sources such as agricultural waste, sewage sludge, used coffee grounds, and industrial solid waste [10].

Activated carbon is considered an effective technology for removing DBP precursors due to its extensive surface area. In a two-stage biofiltration process consisting of a sand biofilter combined with a biologically active GAC filter, it was found that this setup accounted for 60.63% ± 16.64% of the total DBP precursor removal. Using powdered activated carbon (PAC) significantly enhanced the removal of THM and N-DBP precursors. Research has shown that PAC treatment alone may not effectively prevent the formation of N-DBPs, but when combined with pre-oxidation using permanganate, it can significantly reduce the formation of both. This process also demonstrated a sufficient removal of N-DBP precursors, effectively removing low molecular weight, positively charged, and hydrophilic organic substances. The choice of activated carbon used can influence the efficiency of NOM removal. PAC has a larger surface area than GAC and can be utilized at different stages of treatment. A newly developed coconut shell-based GAC has demonstrated higher DOM removal, releasing fewer microbes, metabolites, and smaller microbial complexes than conventional GAC [14].

4.5.3. Membrane Filtration

Membrane filtration is a modern technology utilized to eliminate NOM. Integrating enhanced coagulation and ultrafiltration has shown high efficiency in NOM removal, achieving up to 59% removal rates. Furthermore, the reduction in disinfection byproduct formation potential (DBPsFP) from the combined process of enhanced coagulation and ultrafiltration was found to be 16.9% and 30.8% higher compared to enhanced coagulation alone and ultrafiltration alone, respectively. It was noted that NOM removal through the combined coagulation and ultrafiltration processes was consistently effective irrespective of pH and the dosage of polyferric chloride. To address the two primary drawbacks of membrane filtration, namely membrane fouling and limited effectiveness in removing low molecular weight NOM, pretreatment methods such as enhanced coagulation and pre-oxidation are commonly employed before the membrane filtration process [14].

Current research endeavors for eliminating DBPs through membrane filtration technology primarily concentrate on common aliphatic DBP compounds like THMs, HAAs, haloacetaldehydes (HALs), HANs, and other similar DBP species. The focus on these aliphatic DBPs stems from their widespread presence and potential adverse impacts on human health and the environment. Membrane filtration has demonstrated potential as an effective treatment method for the targeted removal of these specific DBP groups; however, the overall efficiency can be influenced by factors such as membrane characteristics, feed water composition, and operational parameters [12].

Membrane filtration technology is known for its high efficiency, minimal chemical usage, flexible operation, and the absence of secondary pollution. It is extensively employed to eliminate trace pollutants from water. Nonetheless, the efficacy of membrane filtration in removing DBPs is impacted by various factors, including membrane and DBP characteristics, feed solution composition, and operational conditions [12].

The most utilized membrane filtration processes for water treatment encompass pressure-driven membrane technologies, including microfiltration (MF), ultrafiltration (UF), nanofiltration (NF), and reverse osmosis (RO). Membrane filtration treatment has several desirable attributes in terms of DBP removal, such as high efficiency, versatility (ability to be applied to multiple water types), simplicity of operation, and flexibility. These characteristics make membrane-based processes attractive options for effectively removing DBPs [12].

4.5.4. Membrane Nanofiltration

While MF and UF processes are characterized by ease of handling and relatively low operating costs due to the low pressures required, their performance in DBP removal is limited by the large pore sizes of the membranes and the high molecular weight cut-off (MWCO) range, typically between 1 and 200 kDa. These characteristics make it challenging for MF and UF to effectively remove relatively small DBP molecules. The small molecular size of DBPs compared to the large MWCO of MF and UF membranes makes it easy for the DBP molecules to pass through the membrane pores into the permeate stream. The removal efficiency of DBPs by MF is typically around 20%, mainly due to biodegradation or adsorption. MF and UF are commonly used as pretreatment steps to eliminate particles and colloids from the feed water, aiming to prevent fouling of the subsequent NF or RO membranes rather than being the primary method for DBP removal. NF demonstrates intermediate performance between UF and RO, as NF offers better retention properties than UF by effectively rejecting multivalent ions and low-molecular-weight organic substances. At the same time, NF does not require as high an operating pressure as RO. So, in essence, NF presents a middle ground between the capabilities of UF and RO with better rejection performance than UF for specific contaminants and lower operating pressure requirements than the more energy-intensive RO process. This makes NF an attractive option, as it can provide an enhanced removal of DBPs and other organic micropollutants while operating at a reduced energy cost compared to the RO process. The intermediate nature of NF’s performance characteristics positions it as a valuable alternative membrane technology for water treatment applications where UF is insufficient, but RO may not be necessary or feasible [12].

Membrane technology has attracted considerable interest in eliminating organic micropollutants (OMP) due to its simple operation and various technological advantages, including low energy consumption, decreased carbon footprint, and minimal space requirements compared to conventional separation methods. Additionally, membrane systems provide continuous operation and can easily be integrated with current treatment technologies. The most successful commercial membrane products for water treatment include RO and NF membranes made from thin-film composite (TFC) materials. These membranes are commonly manufactured through a cost-effective, scalable, and highly reproducible interfacial polymerization (IP) process conducted from roll to roll. Interfacially polymerized membranes are known for their high-water permeability, outstanding salt rejection (>99%), robust mechanical stability, and broad pH tolerance. The utilization of TFC membranes is a significant contributing factor to pressure-driven membrane processes representing over 60% of the global desalination capacity. These TFC membranes play a crucial role in the dominance of pressure-driven membrane processes, accounting for more than 60% of the worldwide desalination capacity. Previous studies have highlighted the potential of TFC-based RO and NF membranes in removing OMPs from various water sources. However, a notable drawback of current membrane types is their limited ability to reject neutral and slight molecular weight OMPs, especially those that are highly hydrophobic or polar. The rejection of OMPs is influenced by various factors, including: (i) the intrinsic properties of the membrane, such as surface charge, pore size, molecular weight cut-off (MWCO), contact angle, and surface morphology; (ii) the characteristics of the solute, including molecular weight (Mw), acid dissociation constant (pKa), geometry, octanol–water partition coefficient (log Kow), and water solubility; (iii) the chemistry of the feed solution, such as hardness, pH, and ionic strength; and (iv) the operating conditions, including temperature and transmembrane pressure [33].

4.5.5. Magnetic Ion Exchange (MIEX) Process

Before the coagulation–flocculation processes, public water treatment involved the MIEX process as a pretreatment stage to boost the removal of precursors. With unique characteristics like a smaller bead size and a magnetic core, the MIEX resin is a robust base–anion exchange resin with exceptional ion exchange capacity and rapid separation and regeneration properties. MIEX’s capability to adsorb both organic and inorganic anions enable it to absorb DOC and bromide, thereby decreasing the formation of DBPs during chlorination. According to Ranthom et al. [34], the MIEX resin treatment, when applied at a higher dose, effectively removed both DOC and bromide, surpassing the performance of coagulation treatment in terms of total trihalomethanes (TTHMs) and HANs precursors removal. The MIEX resin and coagulation treatments preferred removing humic-like DOM over protein-like DOM, with the MIEX treatment showing a more pronounced effect. This led to treated water being enriched in protein-like DOM compared to raw water. Additionally, the MIEX resin treatment demonstrated superior efficiency in controlling the formation of di-brominated species (DBCM and DBAN) compared to coagulation treatment, highlighting its effectiveness in organic substance and bromide removal. This study reported THM concentrations of 185 µg L⁻¹ and HAN concentrations of 22.7 µg L⁻¹ in raw water, with MIEX resin treatment achieving removal efficiencies of 19–51% for THMs and 70–95% for HAN precursors.

According to Mazhar et al. [6], the MIEX treatment demonstrated a 38 to 77% reduction in TTHMs formation potential and 44 to 74% in HAAs formation potential, varying based on the water source.

4.6. Advanced Oxidation Processes (AOPs)

Numerous technological methods for water and wastewater treatment have been put forward and, among these, AOPs have garnered growing acknowledgement for their effectiveness in water disinfection and decontamination [35]. Traditional AOPs have shown an efficient removal of chemical and microbial pollutants from water. Furthermore, AOPs provide a way to combine two distinct processes into a single treatment step, water disinfection and decontamination, thereby improving the cost-effectiveness of treatment [36].

The primary mechanism in most AOPs involves the in-situ generation of highly reactive species, such as the hydroxyl radical (·OH), the superoxide radical (O₂·⁻), and the sulfate radical (SO₄·⁻), which can degrade organic pollutants and pathogens. Currently, hydroxyl radical-based processes are extensively studied among various AOPs. The hydroxyl radical has a higher oxidation potential (approximately 2.72 V) than other common disinfectants/oxidants like ozone, chlorine, and chlorine dioxide. It is non-selective in attacking target molecules, making it suitable for treating contaminated water. Hydroxyl radical-based AOPs can be classified into Fenton or Fenton-like processes, ozonation, photolytic-like processes, and photocatalytic-like processes, depending on the method used to generate the radical species.

Nevertheless, the effectiveness of hydroxyl radical-based AOPs can be restricted in intricate water matrices containing hydroxyl radical scavengers, like the carbonate/bicarbonate anion and NOM. As a result, there has been an increasing interest in utilizing reactive species with greater selectivity, such as the sulfate radical [9].

4.7. Alternative to Chlorine Disinfectants

Employing disinfection alternatives can enhance the reduction in DBPs formation. Exploring alternatives to traditional chlorine disinfectants has become imperative in the quest for more sustainable and effective water treatment solutions. There are several alternatives that can be considered for various applications, including chloramination, ozonation, UV irradiation.

4.7.1. Chloramination

Chloramine is an alternative to chlorine due to its extended persistence in water distribution systems and its lower production of regulated DBPs, such as HAAs and THMs, compared to chlorine [3]. Even though chloramine is frequently employed as a disinfectant to minimize the formation of regulated DBPs, its interactions with dissolved organic compounds in water result in the production of nitrogenous DBPs (N-DBPs), which are often more harmful than THMs and HAAs [37].

Bromide, a naturally occurring anion commonly found in raw water at concentrations ranging from less than 0.01 to 4 mg L⁻¹, plays a significant role in chloramination processes. During chloramination, chlorine or chloramine oxidizes bromide to form hypobromous acid/hypobromite (HOBr/OBr⁻) or bromamines (e.g., NH₂Br, NHBrCl, and NHBr₂), respectively. These compounds can react with NOM in water, leading to the formation of brominated DBPs. Toxicological investigations have revealed that brominated DBPs demonstrate elevated cytotoxicity, genotoxicity, and developmental toxicity compared to their chlorinated counterparts. Furthermore, epidemiological studies have emphasized the greater concerns associated with brominated DBPs over chlorinated DBPs [3].

The chloramination of various organic compounds has been extensively documented, including organophosphorus pesticides, nitrogenous organic compounds, algal organic matter, herbicides, and oxytetracycline. It has been noted that the levels of iodinated DBPs are higher during chloramination than chlorination. Both chlorine and chloramines can oxidize iodide to hypoiodous acid (HOI). In chlorination, the reactions that lead to the formation of iodate occur much faster than those resulting in the formation of iodinated DBPs. Conversely, in chlorination, the reactions leading to the formation of iodite and iodate proceed much slower than those leading to the formation of iodinated DBPs [4].

4.7.2. Ozonation

Ozone water disinfection was initially introduced in France in 1886. Ozone is a highly versatile and efficient disinfectant due to its potent oxidative characteristics and ability to be utilized at different stages within the treatment process [23].

Ozone selectively and rapidly reacts with electron-rich compounds such as activated aromatic compounds, olefins, neutral amines, and reduced sulfur species. In natural waters and wastewater effluents, DOM typically acts as the primary consumer of ozone. During ozonation, hydroxyl radicals (•OH) are generated in situ as secondary oxidants through ozone decomposition processes. These hydroxyl radicals exhibit less selectivity than ozone, with second-order rate constants generally ranging from 10⁹ to 10¹⁰ M⁻¹ s⁻¹, primarily through OH-addition or H-abstraction reactions. In environmental applications, the concentration of •OH, either in steady-state or transient form, is controlled by its main scavengers, such as DOM, carbonate (CO₃²⁻), and bicarbonate (HCO₃⁻) [38].

Both ozone and sodium hypochlorite are powerful oxidizing agents that play vital roles in pre-oxidation and post-disinfection processes. These agents can degrade toxic organic compounds, algae, and microorganisms, thereby improving the treatment efficiency of WWTPs and DWTPs. However, their interactions with NOM and halogen anions, commonly found in natural waters, ultimately form oxidation/disinfection byproducts [39].

Ozone can chemically react with different substances in natural water through direct transformation. Another chemical process after ozonation is the indirect reaction of free radicals produced due to the breakdown of ozone molecules in water. The initial ozonation process can increase the average size of colloidal particles, converting dissolved organic matter into colloidal particles and improving the removal of organic matter or turbidity in sedimentation and filtration processes. Furthermore, the initial ozonation can modify the structure of organic compounds in the water, convert dissolved organic matter into colloidal particles, and enhance turbidity removal during sedimentation [24].

The dosage of ozone used in drinking water treatment can vary depending on the type of water, typically ranging from 2 to 5 mg L⁻¹. Due to its rapid decomposition under standard drinking water treatment conditions, ozone primarily acts as a primary disinfectant. Subsequently, following ozonation, chlorination or chloramination is commonly utilized as a secondary disinfection step. When ozone doses are at 1 or 2 mg L⁻¹, there is an observed increase in the formation of THMs, dihaloacetic acids, and trihaloacetic acids during subsequent chlorination, while higher ozone doses result in reduced formation. Additionally, chloral hydrate and trichloronitromethane formation potentials rise with higher ozone doses, ranging from 1 to 6 mg L⁻¹. In the case of subsequent chloramination, the concentrations of dihaloacetic acids, THMs, and HANs exhibit varying trends, with some DBPs decreasing and others increasing after ozonation. Moreover, in the presence of bromide in raw water, ozonation has been observed to modify the speciation of DBPs such as THMs, HAAs, and dihaloacetonitriles during subsequent chlor(am)ination, favoring the formation of more brominated compounds over chlorinated products [3].

Limited studies have been conducted to identify, quantify, and evaluate the toxicity of ozone DBPs (ODBPs) compared to those formed by chlorination. ODBPs are generally categorized into two main groups: (1) inorganic ODBPs, which primarily form based on the bromide levels in the raw water, and (2) organic ODBPs, which mainly form based on the characteristics of NOM, ozone dosage, and contact time. Various types of organic ODBPs have been recognized, including aldehydes, ketones, carboxylic acids (CA), hydroxy acids, alcohols, esters, ketoaldehydes, aldoacids, ketoacids, and alkanes, [15,27].

The toxicity information regarding these organic ODBPs is scarce, which raises concerns about the safety of ozonated water. During ozonation, ozone and hydroxyl radicals (generated from the decomposition of ozone) can oxidize NOM, resulting in the creation of readily biodegradable ODBPs. The amount and types of ODBPs generated are highly unpredictable. The formation of ODBPs is significantly influenced by various factors, particularly the composition and concentration of NOM, disinfection conditions (such as ozone dose and contact time), and other treatment processes at WWTPs. Over 60% of the organic compounds remain unidentified during the ozonation disinfection process. Among the identified organics (40%), 26–33% is classified as carbonaceous aromatic byproducts (CABPs). According to a study, 67% of CABPs were removed during water treatment processes, while other classes of ODBPs were eliminated at rates exceeding 80–85%. The higher formation levels and lower removal rates of CABPs in WWTPs have sparked increased research interest in CABPs [23].

During the ozonation of bromide (Br−)-containing waters, bromate (BrO₃⁻) can be produced through a complex mechanism involving reactions with both ozone and hydroxyl radicals (•OH). Bromate is classified as a potential human carcinogen, and the World Health Organization (WHO) suggests a drinking water standard of 10 μg L⁻¹. Although no defined bromate standard for wastewater exists, Switzerland, for example, has put forward an environmental quality standard of 50 μg L⁻¹ [38].

4.7.3. Ultraviolet (UV) Irradiation

Compared to chlorination, UV irradiation is a physical treatment method that does not leave residual disinfecting capability. As a result, UV disinfection is typically complemented with other oxidizing processes such as chlorination, chloramination, or the addition of hydrogen peroxide (H₂O₂) to ensure adequate disinfecting power in drinking water distribution systems. Combining UV irradiation with another oxidant is often utilized as an AOP to eliminate persistent organic pollutants. A novel AOP, the UV/chlorine process has effectively removed various contaminants, including trichloroethylene, desethylatrazine, sulfamethoxazole, carbamazepine, and diclofenac. Additionally, more ·OH radicals can be generated in the UV/chlorine process compared to the UV/H₂O₂ process due to the higher molar absorptivity and quantum yield of HOCl. The formation potential of DBPs in the UV/chlorine system may significantly differ from those formed during UV irradiation or chlorination alone, owing to distinct reaction mechanisms [2].

AOPs, such as UV/H₂O₂, UV/O₃, and UV-chlorine, are recognized as promising methods for removing organic pollutants from water solutions by generating powerful hydroxyl radical oxidation species (HO-). Among these AOPs, the UV/H₂O₂ process has been implemented in full-scale water treatment facilities, but it requires large amounts of H₂O₂ to achieve the desired efficiency due to the low molecular absorption coefficient of H₂O₂. The ultraviolet-chlorine process has been considered as an emerging alternative solution to the UV/H₂O₂ process, as, in addition to HO-, reactive chlorine species (RSC) such as Cl·, Cl₂·⁻, and ClO· are also produced by the UV-chlorine process [40].

4.7.4. Electrochlorination

Out of various disinfection methods, only chlorination provides long-term water disinfection within distribution networks and equipment. In situ electrochlorination offers a user-friendly approach that tackles issues associated with transporting and storing hazardous chemicals. This method has several benefits, including easy process automation and utilizing renewable energy sources, making it well-suited for decentralized water disinfection. Incorporating electrochlorination into water reuse systems is a safety measure that ensures water disinfection and facilitates safe water recycling, thereby decreasing the need for freshwater resources. Nevertheless, in situ electrochlorination poses a similar risk of generating organic DBPs as traditional chlorination. Electrochemical water treatment systems have attracted considerable attention for their ability to effectively eliminate persistent pollutants and pathogens and their safety and operational convenience. Microorganism deactivation is accomplished by producing disinfecting agents during the electrochemical treatment process. This approach shows potential for decentralized treatment in rural regions, as it can operate autonomously without reliance on water and electrical grids. In these systems, chloride anions (Cl⁻) are oxidized on the anode surface, producing electrogenerated chlorine (Cl₂(aq)), which then reacts with water to form hypochlorous acid (HOCl). Ιn situ electrochlorination has the potential to reduce resource consumption and reduce the formation of organic byproducts, as lower doses of electrogenerated chlorine can achieve similar levels of inactivation compared to chemical treatments [41].

4.7.5. Solar Water Disinfection (SODIS)

The WHO has identified infections caused by pathogens resistant to antimicrobials, such as antibiotic-resistant bacteria (ARB), as a global health threat. The situation worsens when pathogenic bacteria carrying antibiotic resistance genes (ARG) transfer this resistance trait, via ARGs, to other bacteria in the water, thereby preserving an antimicrobial property even after disinfection. Hence, there is an urgent need for new water disinfection techniques capable of combating these emerging pathogens while being cost-effective and environmentally sustainable [42].

The WHO has acknowledged SODIS as a simple and effective method for treating water contaminated with pathogenic microorganisms and reducing the occurrence of diarrheal diseases through sunlight exposure. The SODIS technique involves placing untreated water in transparent plastic containers (typically 1 to 2-L PET bottles) and exposing them to direct sunlight for a minimum of 6 h (on cloudy days, the exposure time should be extended to 48–72 h). Subsequently, the water should be kept in the SODIS bottles and consumed within the next 24 h. The microbicidal action of SODIS is attributed to generating reactive oxygen species when the chromophores of microorganisms absorb UVA and UVB photons. The process’s effectiveness is enhanced by the rise in temperature resulting from sun exposure. SODIS is particularly appealing from both an economic and environmental standpoint for several reasons: it has minimal operational costs for users (primarily involving bottle replacement), is extremely easy to use and relies solely on sunlight, preserves the organoleptic properties of water, does not necessitate additional chemicals, and does not leave behind residues [42].

Furthermore, SODIS has been demonstrated to be effective against numerous waterborne pathogens, including Escherichia coli, Salmonella spp., Vibrio cholerae, Enterococcus faecalis, bacteriophage MS2, hepatitis A virus, and Cryptosporidium parvum. However, SODIS does have several drawbacks that have been the focus of recent studies aimed at addressing them: (i) water treated with SODIS, as mentioned, must be consumed within 24 h to prevent the reactivation of pathogenic organisms after sun exposure; (ii) the sunlight’s effect on the plastic material can lead to the release of chemical compounds into the drinking water. SODIS is of significant interest from an economic and environmental perspective for several reasons: it has low operational costs for users (primarily involving bottle replacement), is extremely easy to use, and relies solely on sunlight. It does not alter the taste or odor of water, requires no additional chemicals, and leaves no residues. Additionally, SODIS has been demonstrated to be effective against a wide range of waterborne pathogens, including Escherichia coli, Salmonella spp., Vibrio cholerae, Enterococcus faecalis, bacteriophage MS2, hepatitis A virus, and Cryptosporidium parvum [42].

4.7.6. Photocatalysis

Photocatalytic water treatment has gained significant attention in research, focusing on material synthesis, understanding the mechanism, and designing microreactors. The process has several advantages, including operating under mild conditions, consuming low energy, and having a high potential for mineralization. These qualities appeal to cost-constrained situations, such as in underdeveloped and energy-stressed regions. As a result, researchers are rapidly advancing photocatalytic water treatment technologies to optimize and scale up the process for practical use. The potential to effectively treat water using minimal resources and under ambient conditions has positioned photocatalysis as a promising solution for addressing water quality challenges in areas with limited infrastructure and financial constraints [43].

Photocatalysis with TiO₂ nanoparticles under ultraviolet light (λ < 387 nm) generates hydroxyl radicals (OH•) and superoxide radicals (O₂•⁻) for oxidation and reduction, respectively. Additionally, the photocatalyst’s heterogeneous nature allows for physical adsorption. This process is crucial in mitigating DBPs through decomposition, mineralization, reductive dechlorination, and physical removal of organic precursor substances [26].

TiO₂ photocatalysis can be integrated into a DBP mitigation strategy in various ways, as summarized in Table 1.

In water treatment facilities, it can be used to remove and mineralize precursor substances, preventing the formation of DBPs through pre-chlorination. It can also mineralize any remaining DBP precursor compounds and destroy actual DBPs before water is released into the distribution network. However, it is important to note that re-chlorination would be necessary after the application of photocatalysis due to the destruction of residual chlorine. It is worth considering that implementing advanced oxidation water treatment, like photocatalysis, may pose economic challenges in situations with relatively high flow rates [26].

Photocatalytic water treatment offers a significant advantage as it only requires water and oxygen as substrates for oxidation. However, despite extensive theoretical research in this area, the practical application of photocatalysis is hindered by several roadblocks. One major challenge is the limited yield of oxidative species due to the recombination of photogenerated electron–hole pairs within the photocatalyst. This limitation has slowed the transition from theoretical research to real-world implementation [43].

One promising approach to enhance photocatalytic oxidation efficiency is combining photocatalysis with chemical oxidants. This strategy has been proposed as an effective means to overcome the limitations imposed by the inherent photogenerated electron–hole recombination within the photocatalyst [43]. Additionally, the photocatalysis process is often combined with other methods, such as coagulation, membrane technologies, and others. It is also gaining popularity for the direct removal of DBPs and their formation potential [44].

The process of heterogeneous photocatalysis combined with oxidant activation requires three key elements: an energy source, a heterogeneous photocatalyst, and a water-soluble oxidant. Research typically utilizes ultraviolet light (UV, wavelengths between 10 and 400 nm) and visible light (Vis, wavelengths between 400 and 780 nm) as the primary energy sources. Different types of lamps, such as low-pressure mercury lamps, LED lamps, and xenon lamps, can generate these energy sources [44]. Heterogeneous photocatalysts are crucial for facilitating the photochemical and redox reactions in this process, as they capture light energy, support charge separation and transfer, and ultimately enable the utilization of the produced charges. The efficiency of the photocatalyst is determined by its ability to absorb light, separate the photogenerated electron–hole pairs, and promote the desired redox reactions [43].

AOPs, such as TiO₂ photocatalysis, are based on generating non-specific radical species capable of oxidizing a wide range of pollutants. In photocatalysis, UV light with a wavelength of less than 387 nm irradiates a semiconductor photocatalyst, producing highly reactive radical species such as •OH [22,26]. Additionally, the heterogeneous nature of TiO₂ allows for physical adsorption on its surface, contributing to a multifaceted treatment approach. The effectiveness of photocatalysis in mitigating DBP depends on the concentration and composition of NOM in the influent, which can vary significantly as untreated water undergoes various physicochemical treatment processes [22].

In many instances, the significant contribution to DBP formation comes from hydrophilic and lower molecular weight fractions. The application of photocatalytic treatment tends to alter the composition of NOM towards these components. According to Mayer et al. [22], there was an observed shift in the hydrophilic fraction from 15 to 90% after undergoing photocatalytic treatment. Consequently, even after photocatalysis, DBPs can continue to form and surpass initial levels. Mayer et al. [22] also noted a rise in TTHM formation by as much as 75% when employing limited photocatalysis (65 kWh m⁻³) for raw water treatment. However, with higher energy levels, there was a substantial decrease in DBP formation, leading to a 95% reduction in TTHMs. Unfortunately, this enhanced performance required a significant amount of energy, which could make it economically unfeasible for applications in high-flow water treatment.

By optimizing sequential treatment processes with restricted TiO₂ photocatalysis after conventional treatment methods like adsorption and filtration, it is possible to focus on eliminating/destroying both hydrophobic and hydrophilic fractions. This approach can effectively reduce DBP formation and simultaneously minimize the usage of system inputs such as chemicals and energy [22].

Towards the conclusion of the previous century, photocatalysis surfaced as an alternative or supplementary technology to conventional disinfection methods. Photocatalysis, categorized as an AOP, can be classified into heterogeneous and homogeneous processes. In both scenarios, sunlight can be harnessed to stimulate the production of hydroxyl radicals (•OH), known for their potent disinfectant characteristics. In water, pathogen inactivation occurs during heterogeneous photocatalysis with a photocatalyst and UV radiation. Although several catalysts have been experimented with, TiO₂ remains the most extensively researched in disinfection procedures. Nevertheless, when sunlight is employed as the energy source for heterogeneous photocatalysis, the efficiency of the process is diminished. This is because photocatalysts are predominantly activated by UV radiation, which accounts for only 3–5% of sunlight. The efficiency of the process significantly improves when the catalyst is in the form of suspended micro- or nanoparticles. These particles can be immobilized to prevent nanoparticle contamination; however, immobilization restricts large-scale applications compared to homogeneous processes. The most extensively studied process in homogeneous solar photocatalysis is solar photo-Fenton treatment. This treatment involves a series of reactions with dissolved iron as the catalyst, hydrogen peroxide and solar radiation. Solar photo-Fenton can deactivate pathogens that are highly resistant to conventional disinfection methods, including Bacillus subtilis, phytopathogenic fungi, antimicrobial-resistant bacteria and genes, and viruses found in surface water and wastewater [42].

Presently, the advancement of solar photochemical processes, particularly those utilizing solar light, is regarded as being of dynamic importance. AOPs founded on photocatalysis have been demonstrated to be a compelling tool for eliminating chemical and biological contaminants from water. They have garnered considerable attention. Remarkably, heterogeneous photocatalysis (HP) is described as “chemical reactions induced by a solid material (photocatalyst) that absorbs appropriate radiation and remains unchanged throughout the process”. TiO₂ and ZnO are the most commonly utilized semiconductors as photocatalysts for solar applications in environmental settings. The primary advantage of these technologies is their ability to eliminate/reduce pollutants through mineralization, unlike other traditional processes where the pollutant is simply transferred to the environment. Hydroxyl radicals (HO•, E₀ = 1.9–2.8 V vs. NHE) and other highly oxidizing species like the superoxide anion (O₂•-) and hydroperoxide (HO₂•) are responsible for oxidizing organic pollutants. In environmental applications, the sulfate radical anion, SO₄•⁻ (E₀ = 2.5–3.1 V vs. NHE), can also play a role when persulfate (S₂O₈²⁻) or peroxymonosulfate (HSO₅⁻) are used as oxidants. Hydroxyl radicals (HO•) and sulfate radical anions (SO4•⁻) have half-lives of 10–3 and 30–40 μs, respectively. While both typically engage in electron transfer reactions, hydroxyl radicals can also react through hydrogen atom abstraction [30].

As per Perez-Lucas et al. [30], even though THMs are highly volatile and photodegradable compounds (with removal percentages ranging from 86% to 54% for TCM and TBM after 4 h of stirring in the dark), the application of solar heterogeneous photocatalysis using commercially available photocatalysts like the semiconductor materials ZnO and TiO₂, in combination with a potent oxidant such as Na₂S₂O₈, significantly improves their elimination in drinking water. The primary advantage of AOPs is their ability to restrict or reduce pollutant content through mineralization, unlike conventional methods that merely transfer pollutants from one location to another. The removal percentages after the photolytic test (4 h) varied between 63 and 92% for trichlormethane (TCM) and tribromomethane (TBM). When photocatalysts were utilized, ZnO/Na₂S₂O₈ exhibited greater effectiveness compared to TiO₂/Na₂S₂O₈ in degrading chlorinated species (TCM, bromodichloromethane-BDCM, and dibromochloromethane-DBCM), while no significant differences were observed for TBM.

Using photocatalysis before the disinfection step can reduce the levels of NOM and other DBP precursors, ultimately limiting the formation of DBPs in the disinfected water. Sinha and Ghosal [44] observed a 57% reduction in river water’s THMFP after 12 h of photocatalysis using fluidized TiO₂ particles. More than 90% reductions in THMFP and haloacetic acid formation potential (HAAFP) using a mechanically driven photocatalytic reactor within 60 min of irradiation time have also been reported. Integrated photocatalysis into a microfiltration reactor significantly reduced THM formation potential (66–69%) for synthetic and natural water samples. Employing a nanofiltration membrane after the photocatalytic reactor further reduced the formation potential. In another study, the THMFP of the surface water was significantly increased within 15 min of photocatalysis. The photocatalytic process elucidated the observation, degrading higher, more aromatic compounds into smaller, less aromatic ones. Limited photocatalysis, therefore, leads to incomplete oxidation, which subsequently creates intermediate compounds that are more reactive to chlorine. Another pilot-scale study on photocatalysis and enhanced coagulation showed that limited photocatalysis increased THM formation, while extended photocatalysis significantly minimized THMFP. However, the energy consumption was very high in the latter case, rendering the process impractical for full-scale plant implementation.

On the other hand, the photocatalytic reduction of THMs requires significantly less energy compared to a decrease in their precursors. Therefore, to balance the mitigation of DBPs, process factors such as chemical dosage, irradiation time, and energy consumption need to be carefully studied and optimized. Another study investigated the photosensitized photodegradation of humic acid (HA). It was reported [44] that the photocatalysis of an HA solution with TiO₂/UV after 2 h resulted in a 20% reduction in THMFP. On the other hand, the irradiation of an HA solution in the presence of TiO₂ under visible light increased the THMFP of HA. The results suggest that the HAs were chemically transformed in the homogeneous solution under visible light, thereby increasing the THMFP. In another study, the HAAFP of water irradiated with UV light for 30 min was 1.22 times higher than the HAAFP of the raw water. However, it was reduced to 66% because of 60 min of photocatalytic oxidation. Comparing the performance of UV/TiO2 treatment using TiO₂ suspension and nanostructured TiO₂-coated thin film, the reduction in THMFP for the raw water was around 30% for both the suspended and the fixed UV/TiO₂ treatments. However, in the case of the HAA9FP, the suspended TiO₂ treatment was more effective (90% reduction) than the fixed treatment. It has also been reported that the formation of THMs in chlorinated water was reduced using photocatalysis before chlorination. The decrease in their formation was attributed to the change in the chemical structure of their precursors, which ultimately became less susceptible to chlorine reactivity. A few studies have also investigated the role of the photocatalytic process in the direct degradation of DBPs. It was observed that the TiO₂ photocatalyst supported on diatomite in nanoscale effectively degraded Br-THMs under Xe light irradiation. The results showed that the Br-THMs were debrominated stepwise. The degradation rate increased significantly with the increase in the degree of substitution of Br-THMs by bromine because the C-Br bond was less stable than the C-Cl bond. THMs were observed to be degraded with the TiO₂ catalyst in combination with sodium persulfate (Na₂S₂O₈) oxidant under solar light. The process effectively degraded tribromomethane (TBM), as only 4% of the initial TBM was recovered in the water after 240 min of TiO₂/Na₂S₂O₈ treatment. In the case of TCM, around 30% remained after 60 min of treatment [44].

5. Machine Learning Approaches

Machine learning (ML) is a specialized discipline within computer science and artificial intelligence that focuses on developing machines that learn from data and make predictions without explicit programming. In the field of ML, the learning process entails utilizing statistical analysis to examine input data and generate predictions. In most cases, ML refers to particular algorithms that enhance themselves to analyze large datasets. The efficiency of developed algorithms is directly proportional to the size of the dataset.

Algorithms for ML are becoming more and more useful for a variety of uses. ML approaches are applied in several research topics, including analytical chemistry [45,46], ophthalmology [47], the food and agricultural sector [48], psychology [49], water quality control [50], and many more.

Asteris et al. [45] established and suggested an ML approach and demonstrated its robustness and value as a tool for scientists, researchers, engineers, and practitioners in monitoring water systems and designing wastewater treatment plants.

Analytical chemistry has undergone a radical transformation because of ML, which has profoundly altered how data assessment is evaluated and understood. The advancements in this subject have greatly improved the capacity to use various statistical and mathematical tools, also known as algorithms, to extract useful information, identify patterns, and make precise forecasts.

Many water distribution networks face challenges when related to monitoring emerging DBPs. Utilizing empirical models to forecast challenging-to-measure water quality parameters may offer a viable alternative to traditional water sampling and analysis [51,52]. Empirical models can be utilized to forecast the contents of specific contaminants (known as response) by easily accessible data on other factors related to water quality and operational conditions (known as predictors).

A proposed tackle could be the use of machine learning-based predictive modeling. To create machine learning models that can forecast DBPs, it is crucial to have access to extensive datasets that include information on water quality.

The datasets should include ample water quality information to facilitate an analysis of the intricate correlation between predictors and the response. However, gathering information on newly formed DBPs in water distribution networks can be challenging at times, and the creation of ML models often encounters restrictions due to limited data availability [53].

Recent studies [50,54,55] have indicated that various machine learning techniques, such as regression trees, support vector regression (SVR), and Gaussian process regression (GPR), can effectively monitor emerging contaminants in water distribution networks and optimize disinfection operations to reduce the occurrence of these contaminants. Hence, assessing the efficacy of various ML methods in forecasting and developing DBPs while dealing with limited datasets is valuable.

According to Peng et al. [56], modern ML techniques might exhibit more accuracy than linear or log-linear power functions when forecasting the development of DBPs. Many researchers [56,57,58,59] have demonstrated that ANNs outperformed multivariate power functions in forecasting the generation of DBPs. Nevertheless, using other machine learning techniques to predict the development of disinfection byproducts (DBPs) is infrequently documented [56].

Peng et al. [56] specifically constructed a prediction model of chloral hydrate, effectively addressing the existing gaps in information. The random forest regression and support vector regression algorithms outperformed the classic stepwise multiple linear regression approach in each DBP species’ training and testing sets. This was seen by their lower RMSE and MAPE values, as well as their higher R2 values. The primary reason for this is the capacity of random forest regression and SVR to accurately represent the intricate and nonlinear connections between water quality parameters that impact the creation of DBPs. UV254 was the most influential water quality measure in predicting the target DBPs.

Hu et al. [54] focused on developing multiple ML models to forecast the occurrence of DBPs in water distribution networks accurately. Eleven ML and model regularization techniques were evaluated and categorized into four groups: (a) the first group consisted of two advanced non-parametric regression methods: support vector regression (SVR) and GPR; (b) the second group comprised three regression tree-based methods: boosting, bagging, and random forest (RF); (c) the third group consisted of three linear regression-based methods: MLR, Lasso, and principal components regression (PCR); and (d) the fourth group included three neural network (NN) based methods: Bayesian regularized NN (BRNN), resilient back-propagation (RPROP) NN, and GRNN. The study conducted by Hu et al. [54] utilized eleven ML algorithms, categorized into four groups, to create prediction models for three often-seen emergent disinfection byproducts in surface water distribution networks. The models were constructed and optimized using the identical dataset, and their predictive efficacy was evaluated and compared. The results indicate that the SVR and GPR models showed superior performance compared to other models in terms of both prediction accuracy and stability. SVR demonstrated superior accuracy and stability in predicting DCAN and TCP, whereas GPR proved to be the most effective model for CPK prediction. The GRNN model in the NN category had highly favorable prediction performance, with the BRNN and standard RPROP NN models following closely behind. The random forest (RF) technique is considered the most effective method for predicting emergent DBPs within the category of regression trees. Linear regression-based models provide the lowest level of prediction accuracy, suggesting that basic multivariate linear models are inadequate for capturing the intricate patterns observed in the water quality data. It is worthwhile to investigate the performance of other advanced machine learning approaches, such as extreme learning machines, Bayesian additive regression trees, and extreme gradient boosting, in predicting new DBPs in the future. Furthermore, future modeling work can incorporate other influential water quality indicators, operational factors, and DBP species in addition to the variables already considered in the study performed by Hu et al. [54].

6. Recent Research Trends

Table 2 presents the recent research trends via selected publications outlined in the relevant scientific literature during the last five years. During scientific literature search, it can be observed that for all sectors described during the above sections, there has been a significant increase in the number of publications and reported progress/advances as well as in the variety of applied/tested techniques.

DBPs continue to be a crucial issue related to human health and environmental protection; the technological achievements since their discovery 50 years ago are impressive and, currently, there are evolving novel methods that, when combined/optimized, can be of high value towards achieving actual DBP removal/minimization in treated waters. Major developments have been performed in recent years, from DBP precursors and formation examination and interpretation to the detection of different compounds at trace levels and toxicity testing. Adsorption techniques, AOPs, and alternative disinfectants have been applied, and they have shown very promising results for water quality improvement. Currently, machine learning approaches that are developed and applied can also contribute to the confrontation of DBP issues, through valorizing datasets obtained to estimate/model DBPs concentrations, therefore assisting in the selection and application of the most effective minimization techniques.

7. Conclusions and Perspectives

The formation of toxic DBPs during water treatment has been comprehensively documented over the last five decades. However, the detailed mechanisms behind the formation of individual DBP compounds, the identification of emerging DBP categories, and the properties of their precursors remain critical topics of investigation. These challenges, combined with the exploration of various water treatment strategies, continue to push research forward. The ultimate goal is to deliver drinking water that is entirely “free” from DBPs, or at least to reduce their concentrations to levels that are both minimized and safe for public health.

Recent research efforts are increasingly focused on the development and optimization of cutting-edge techniques, including the application of novel disinfectants, adsorbents, AOPs, and innovative modifications or combinations of treatment technologies. The objective is to improve efficiency in DBP removal while maintaining water safety. Simultaneously, the rise of AI and ML has opened new horizons in water treatment, offering powerful tools for predictive modeling, process optimization, and real-time monitoring. These advanced methods are becoming integral in fine-tuning treatment processes, predicting DBP formation, and improving overall water quality control, ensuring that public health is safeguarded with greater precision and efficiency than ever before. The synergy between novel treatment methods and AI-driven approaches holds tremendous potential for revolutionizing water treatment and achieving a future where DBPs are effectively controlled or eliminated.

Author Contributions

Conceptualization, S.K.G. and A.D.N.; methodology, S.K.G. and A.D.N.; validation, S.K.G., A.D.N. and D.E.A.; formal analysis, S.K.G., A.D.N. and D.E.A.; investigation, S.K.G., A.D.N. and D.E.A.; resources, S.K.G., A.D.N. and D.E.A.; data curation, S.K.G., A.D.N. and D.E.A.; writing—original draft preparation, S.K.G., A.D.N. and D.E.A.; writing—review and editing, S.K.G., A.D.N. and D.E.A.; visualization, S.K.G., A.D.N. and D.E.A.; supervision, S.K.G., A.D.N. and D.E.A.; project administration, S.K.G. and A.D.N. All authors have read and agreed to the published version of the manuscript.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data are contained within the paper.

Conflicts of Interest

The authors declare no conflicts of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Tables

Figure 1. Number of articles covering the years 2020–2024 returned for each searched keyword.

Figure 2. Sedimentation tank.

Figure 3. Sand filtration beds.

Figure 4. GAC filter adsorber.

Figure 5. Sedimentation tank effluent.

Table 1

Incorporation of photocatalysis in DBP mitigation.

Precursor Removal	Combined Use with	Efficiency Improvements	Direct Removal of DBPs
During prechlorination	Chemical oxidants	Solar photo-Fenton	Best results for brominated species
Before distribution network	Coagulation	TiO2 and ZnO semiconductors
	Membranes
Limitations
NOM shift to hydrophilic fractions → favors DBPs formation
Increased energy consumption → increased costs
Rechlorination needed when applied in the end of the process

Table 2

Current research trends in DBPs minimization research.

DBPs Minimization	Techniques	Reference
Precursors and/or DBPs removal
Coagulation	Enhanced coagulation	[60]
Coagulation	Low-level ferrous iron	[61]
Coagulation	Tri-protonated ferrate as preoxidant/coagulant	[62]
Various techniques	Pre-ozonation, coagulation–sedimentation, sand filtration, and ozone combined with biological activated carbon (O3-BAC)	[63]
Various techniques	Adsorption, boiling, membrane filtration	[64]
Adsorption	Anion-exchange resin adsorption followed by electrolysis	[65]
Adsorption	Carbon nanomaterial-based adsorbents	[66]
Adsorption	Graphene oxide/ferrihydrite	[67]
Adsorption	Granular activated carbon	[68]
Adsorption	Nanoscale silver supported on activated carbon	[69]
Adsorption	GAC	[70]
Adsorption	Hydrolyzed polyacrylonitrile UF membrane	[71]
Adsorption	Biological activated carbon	[72]
Adsorption	Biological biochar and activated carbon filters	[73]
Adsorption	Nanofiltration	[74]
AOPs	Heterogenous photocatalysis followed by granular activated carbon	[68]
AOPs	Pre-oxidation by ozone, permanganate, and ferrate	[75]
AOPs	Pre-oxidation by ozone, permanganate, and ferrate	[76]
AOPs	Solar heterogeneous photocatalysis	[30]
AOPs	Pre-oxidation by ozone, chlorine dioxide, permanganate, and ferrate	[64]
Alternative disinfectants/modifying treatment
	Preformed monochloramine	[77]
	Optimizing Cl2 contact time	[78]
	UV/chlorine and VUV/chlorine as ultrafiltration membrane pretreatment	[79]
	UV/NH2Cl	[80]
	Monochloramine	[81]
	O3, Fe(VI), Mn(VII), and ClO2	[64]
	Cl2/ClO2	[82]
	Ozone	[83]
AI/machine learning
	Membrane design	[84]
	Ultrafiltration processes	[85]
	Real-time monitoring of Cl2 dosage	[86]
	Adsorption on nanocomposite material	[87]
	Optimal coagulant dose by artificial neural network fuzzy inference system (ANFIS)	[88]
	Optimized coagulation process	[89]
	Nanofiltration membrane performance	[90]
	Prediction of physicochemical water quality characteristics	[91]
	Model chlorine, chloramine, and chlorine odor intensity	[92]

References

1. Wu, T.; Karimi-Maleh, H.; Dragoi, E.N.; Puri, P.; Zhang, D.; Zhang, Z. Traditional Methods and Biosensors for Detecting Disinfection By-Products in Water: A Review. Environ. Res.; 2023; 237, 116935. [DOI: https://dx.doi.org/10.1016/j.envres.2023.116935] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37625534]

2. Guo, Z.-B.; Lin, Y.-L.; Xu, B.; Huang, H.; Zhang, T.-Y.; Tian, F.-X.; Gao, N.-Y. Degradation of Chlortoluron during UV Irradiation and UV/Chlorine Processes and Formation of Disinfection by-Products in Sequential Chlorination. Chem. Eng. J.; 2016; 283, pp. 412-419. [DOI: https://dx.doi.org/10.1016/j.cej.2015.07.074]

3. Han, J.; Zhai, H.; Zhang, X.; Liu, J.; Sharma, V.K. Effects of Ozone Dose on Brominated DBPs in Subsequent Chlor(Am)Ination: A Comprehensive Study of Aliphatic, Alicyclic and Aromatic DBPs. Water Res.; 2024; 250, 121039. [DOI: https://dx.doi.org/10.1016/j.watres.2023.121039]

4. Tian, F.-X.; Xu, B.; Lin, Y.-L.; Hu, C.-Y.; Zhang, T.-Y.; Xia, S.-J.; Chu, W.-H.; Gao, N.-Y. Chlor(Am)Ination of Iopamidol: Kinetics, Pathways and Disinfection by-Products Formation. Chemosphere; 2017; 184, pp. 489-497. [DOI: https://dx.doi.org/10.1016/j.chemosphere.2017.06.012] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28618281]

5. Golfinopoulos, S.; Nikolaou, A. The Screening of Disinfection By-Products in Large and Small Water Systems in Greece. Desalination Water Treat.; 2017; 58, pp. 80-85. [DOI: https://dx.doi.org/10.5004/dwt.2017.0352]

6. Mazhar, M.A.; Khan, N.A.; Ahmed, S.; Khan, A.H.; Hussain, A.; Rahisuddin,; Changani, F.; Yousefi, M.; Ahmadi, S.; Vambol, V. Chlorination Disinfection By-Products in Municipal Drinking Water—A Review. J. Clean. Prod.; 2020; 273, 123159. [DOI: https://dx.doi.org/10.1016/j.jclepro.2020.123159]

7. Zhang, Y.; Chu, W.; Yao, D.; Yin, D. Control of Aliphatic Halogenated DBP Precursors with Multiple Drinking Water Treatment Processes: Formation Potential and Integrated Toxicity. J. Environ. Sci.; 2017; 58, pp. 322-330. [DOI: https://dx.doi.org/10.1016/j.jes.2017.03.028]

8. Wang, Y.; Li, L.; Sun, Z.; Dong, H.; Yu, J.; Qiang, Z. Removal of Disinfection By-Product Precursors in Drinking Water Treatment Processes: Is Fluorescence Parallel Factor Analysis a Promising Indicator?. J. Hazard. Mater.; 2021; 418, 126298. [DOI: https://dx.doi.org/10.1016/j.jhazmat.2021.126298]

9. Lei, X.; Xie, Z.; Sun, Y.; Qiu, J.; Yang, X. Recent Progress in Identification of Water Disinfection Byproducts and Opportunities for Future Research. Environ. Pollut.; 2023; 337, 122601. [DOI: https://dx.doi.org/10.1016/j.envpol.2023.122601]

10. Tafvizi, H.; Chowdhury, S.; Husain, T. Low Cost Activated Carbon for Removal of NOM and DBPs: Optimization and Comparison. Water; 2021; 13, 2244. [DOI: https://dx.doi.org/10.3390/w13162244]

11. Badawy, M.I.; Gad-Allah, T.A.; Ali, M.E.M.; Yoon, Y. Minimization of the Formation of Disinfection By-Products. Chemosphere; 2012; 89, pp. 235-240. [DOI: https://dx.doi.org/10.1016/j.chemosphere.2012.04.025] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/22591848]

12. Xia, S.; Liu, M.; Yu, H.; Zou, D. Pressure–Driven Membrane Filtration Technology for Terminal Control of Organic DBPs: A Review. Sci. Total Environ.; 2023; 904, 166751. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2023.166751] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37659548]

13. Wang, L.-S.; Wei, D.-B.; Wei, J.; Hu, H.-Y. Screening and Estimating of Toxicity Formation with Photobacterium Bioassay during Chlorine Disinfection of Wastewater. J. Hazard. Mater.; 2007; 141, pp. 289-294. [DOI: https://dx.doi.org/10.1016/j.jhazmat.2006.07.001] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/16916574]

14. Dong, H.; Zhang, H.; Wang, Y.; Qiang, Z.; Yang, M. Disinfection By-Product (DBP) Research in China: Are We on the Track?. J. Environ. Sci.; 2021; 110, pp. 99-110. [DOI: https://dx.doi.org/10.1016/j.jes.2021.03.023]

15. Gui, X.; Zhang, H.; Ji, B.; Ma, J.; Xu, M.; Li, Y.; Yan, M. Study on the Control of Dichloroacetonitrile Generation by Two-Point Influent Activated Carbon-Quartz Sand Biofilter. Membranes; 2022; 12, 137. [DOI: https://dx.doi.org/10.3390/membranes12020137]

16. Richardson, S.; Ternes, T.; Van, D. Water Analysis: Emerging Contaminants and Current Issues. Anal. Chem.; 2018; 90, pp. 398-428. [DOI: https://dx.doi.org/10.1021/acs.analchem.7b04577]

17. EC (European Community) Council. Directive (EU) 2020/2184 of the European Parliament and of the Council of 16 December 2020 on the Quality of Water Intended for Human Consumption (Recast) (Text with EEA Relevance); European Community: Maastricht, The Netherlands, 2020; pp. 1-62.

18. DeMarini, D.M. A Review on the 40th Anniversary of the First Regulation of Drinking Water Disinfection By-Products. Environ. Mol. Mutagen.; 2020; 61, pp. 588-601. [DOI: https://dx.doi.org/10.1002/em.22378]

19. Nawrocki, J.; Andrzejewski, P. Nitrosamines and Water. J. Hazard. Mater.; 2011; 189, pp. 1-18. [DOI: https://dx.doi.org/10.1016/j.jhazmat.2011.02.005]

20. Riyadh, A.; Peleato, N.M. Natural Organic Matter Character in Drinking Water Distribution Systems: A Review of Impacts on Water Quality and Characterization Techniques. Water; 2024; 16, 446. [DOI: https://dx.doi.org/10.3390/w16030446]

21. Mustereț, C.P.; Morosanu, I.; Ciobanu, R.; Plavan, O.; Gherghel, A.; Al-Refai, M.; Roman, I.; Teodosiu, C. Assessment of Coagulation–Flocculation Process Efficiency for the Natural Organic Matter Removal in Drinking Water Treatment. Water; 2021; 13, 3073. [DOI: https://dx.doi.org/10.3390/w13213073]

22. Mayer, B.K.; Daugherty, E.; Abbaszadegan, M. Disinfection Byproduct Formation Resulting from Settled, Filtered, and Finished Water Treated by Titanium Dioxide Photocatalysis. Chemosphere; 2014; 117, pp. 72-78. [DOI: https://dx.doi.org/10.1016/j.chemosphere.2014.05.073] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24972073]

23. Pulicharla, R.; Proulx, F.; Behmel, S.; Sérodes, J.-B.; Rodriguez, M.J. Trends in Ozonation Disinfection By-Products—Occurrence, Analysis and Toxicity of Carboxylic Acids. Water; 2020; 12, 756. [DOI: https://dx.doi.org/10.3390/w12030756]

24. Karwowska, B.; Sperczyńska, E. Organic Matter and Heavy Metal Ions Removal from Surface Water in Processes of Oxidation with Ozone, UV Irradiation, Coagulation and Adsorption. Water; 2022; 14, 3763. [DOI: https://dx.doi.org/10.3390/w14223763]

25. Poleneni, S.R.; Inniss, E.; Shi, H.; Yang, J.; Hua, B.; Clamp, J. Enhanced Flocculation Using Drinking Water Treatment Plant Sedimentation Residual Solids. Water; 2019; 11, 1821. [DOI: https://dx.doi.org/10.3390/w11091821]

26. Gerrity, D.; Mayer, B.; Ryu, H.; Crittenden, J.; Abbaszadegan, M. A Comparison of Pilot-Scale Photocatalysis and Enhanced Coagulation for Disinfection Byproduct Mitigation. Water Res.; 2009; 43, pp. 1597-1610. [DOI: https://dx.doi.org/10.1016/j.watres.2009.01.010]

27. Sriboonnak, S.; Induvesa, P.; Wattanachira, S.; Rakruam, P.; Siyasukh, A.; Pumas, C.; Wongrueng, A.; Khan, E. Trihalomethanes in Water Supply System and Water Distribution Networks. Int. J. Environ. Res. Public Health; 2021; 18, 9066. [DOI: https://dx.doi.org/10.3390/ijerph18179066]

28. Jiang, J.; Han, J.; Zhang, X. Nonhalogenated Aromatic DBPs in Drinking Water Chlorination: A Gap between NOM and Halogenated Aromatic DBPs. Environ. Sci. Technol.; 2020; 54, pp. 1646-1656. [DOI: https://dx.doi.org/10.1021/acs.est.9b06403]

29. Latif, S.; Alim, M.A.; Rahman, A.; Haque, M.M. A Review on Chlorination of Harvested Rainwater. Water; 2023; 15, 2816. [DOI: https://dx.doi.org/10.3390/w15152816]

30. Pérez-Lucas, G.; Martínez-Menchón, M.; Vela, N.; Navarro, S. Removal Assessment of Disinfection By-Products (DBPs) from Drinking Water Supplies by Solar Heterogeneous Photocatalysis: A Case Study of Trihalomethanes (THMs). J. Environ. Manag.; 2022; 321, 115936. [DOI: https://dx.doi.org/10.1016/j.jenvman.2022.115936]

31. Suquet, J.; Godo-Pla, L.; Valentí, M.; Verdaguer, M.; Martin, M.J.; Poch, M.; Monclús, H. Development of an Environmental Decision Support System for Enhanced Coagulation in Drinking Water Production. Water; 2020; 12, 2115. [DOI: https://dx.doi.org/10.3390/w12082115]

32. Siddique, M.S.; Lu, H.; Xiong, X.; Fareed, H.; Graham, N.; Yu, W. Exploring Impacts of Water-Extractable Organic Matter on Pre-Ozonation Followed by Nanofiltration Process: Insights from pH Variations on DBPs Formation. Sci. Total Environ.; 2023; 876, 162695. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2023.162695] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36898544]

33. Alonso, E.; Sanchez-Huerta, C.; Ali, Z.; Wang, Y.; Fortunato, L.; Pinnau, I. Evaluation of Nanofiltration and Reverse Osmosis Membranes for Efficient Rejection of Organic Micropollutants. J. Membr. Sci.; 2024; 693, 122357. [DOI: https://dx.doi.org/10.1016/j.memsci.2023.122357]

34. Ranthom, N.; Khongnakorn, W.; Jutaporn, P. MIEX Resin and Enhanced Coagulation Treatment of High-Bromide Natural Water: Chlorine Reactivity and DBP Precursors Removal. J. Environ. Chem. Eng.; 2023; 11, 111497. [DOI: https://dx.doi.org/10.1016/j.jece.2023.111497]

35. Fatta-Kassinos, D.; Vasquez, M.I.; Kümmerer, K. Transformation Products of Pharmaceuticals in Surface Waters and Wastewater Formed during Photolysis and Advanced Oxidation Processes—Degradation, Elucidation of Byproducts and Assessment of Their Biological Potency. Chemosphere; 2011; 85, pp. 693-709. [DOI: https://dx.doi.org/10.1016/j.chemosphere.2011.06.082]

36. Lofrano, G.; Ubaldi, F.; Albarano, L.; Carotenuto, M.; Vaiano, V.; Valeriani, F.; Libralato, G.; Gianfranceschi, G.; Fratoddi, I.; Meric, S. et al. Antimicrobial Effectiveness of Innovative Photocatalysts: A Review. Nanomaterials; 2022; 12, 2831. [DOI: https://dx.doi.org/10.3390/nano12162831]

37. Sun, Z.; Chen, Z.; Chung Lan Mow, M.C.; Liao, X.; Wei, X.; Ma, G.; Wang, X.; Yu, H. Chloramine Disinfection of Levofloxacin and Sulfaphenazole: Unraveling Novel Disinfection Byproducts and Elucidating Formation Mechanisms for an Enhanced Understanding of Water Treatment. Molecules; 2024; 29, 396. [DOI: https://dx.doi.org/10.3390/molecules29020396]

38. Wünsch, R.; Hettich, T.; Prahtel, M.; Thomann, M.; Wintgens, T.; Gunten, U. von Tradeoff between Micropollutant Abatement and Bromate Formation during Ozonation of Concentrates from Nanofiltration and Reverse Osmosis Processes. Water Res.; 2022; 221, 118785. [DOI: https://dx.doi.org/10.1016/j.watres.2022.118785] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35949072]

39. Zhu, X.; Hao, C.; Zhang, M.; Lan, B. Enhanced Adsorption of Bromoform onto Microplastic Polyethylene Terephthalate Exposed to Ozonation and Chlorination. Molecules; 2023; 28, 259. [DOI: https://dx.doi.org/10.3390/molecules28010259] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36615452]

40. Wang, C.; Ying, Z.; Ma, M.; Huo, M.; Yang, W. Degradation of Micropollutants by UV–Chlorine Treatment in Reclaimed Water: pH Effects, Formation of Disinfectant Byproducts, and Toxicity Assay. Water; 2019; 11, 2639. [DOI: https://dx.doi.org/10.3390/w11122639]

41. Atrashkevich, A.; Alum, A.; Stirling, R.; Abbaszadegan, M.; Garcia-Segura, S. Approaching Easy Water Disinfection for All: Can in Situ Electrochlorination Outperform Conventional Chlorination under Realistic Conditions?. Water Res.; 2024; 250, 121014. [DOI: https://dx.doi.org/10.1016/j.watres.2023.121014]

42. Ballesteros, M.; Brindley, C.; Sánchez-Pérez, J.A.; Fernández-Ibañez, P. Worldwide Research Trends on Solar-Driven Water Disinfection. Int. J. Environ. Res. Public Health; 2021; 18, 9396. [DOI: https://dx.doi.org/10.3390/ijerph18179396] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34501986]

43. Su, R.; Zhu, Y.; Gao, B.; Li, Q. Progress on Mechanism and Efficacy of Heterogeneous Photocatalysis Coupled Oxidant Activation as an Advanced Oxidation Process for Water Decontamination. Water Res.; 2024; 251, 121119. [DOI: https://dx.doi.org/10.1016/j.watres.2024.121119] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38219690]

44. Sinha, R.; Ghosal, P.S. A Comprehensive Appraisal on Status and Management of Remediation of DBPs by TiO2 Based-Photocatalysts: Insights of Technology, Performance and Energy Efficiency. J. Environ. Manag.; 2023; 328, 117011. [DOI: https://dx.doi.org/10.1016/j.jenvman.2022.117011] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36525732]

45. Asteris, P.G.; Alexakis, D.E.; Tsoukalas, M.Z.; Gamvroula, D.E.; Guney, D. Machine Learning Approach for Rapid Estimation of Five-Day Biochemical Oxygen Demand in Wastewater. Water; 2023; 15, 103. [DOI: https://dx.doi.org/10.3390/w15010103]

46. Pérez-Beltrán, C.H.; Pérez-Caballero, G.; Andrade, J.; Cuadros-Rodríguez, L.; Jiménez-Carvelo, A. Non-Targeted Spatially Offset Raman Spectroscopy-Based Vanguard Analytical Method to Authenticate Spirits: White Tequilas as a Case Study. Microchem. J.; 2022; 183, 108126. [DOI: https://dx.doi.org/10.1016/j.microc.2022.108126]

47. Li, S.; Li, M.; Wu, J.; Li, Y.; Han, J.; Song, Y.; Cao, W.; Zhou, X.-T. Developing and Validating a Clinlabomics-Based Machine-Learning Model for Early Detection of Retinal Detachment in Patients with High Myopia. J. Transl. Med.; 2024; 22, 405. [DOI: https://dx.doi.org/10.1186/s12967-024-05131-9]

48. Othman, S.; Mavani, N.R.; Hussain, M.A.; Rahman, N.A.; Ali, J.M. Artificial Intelligence-Based Techniques for Adulteration and Defect Detections in Food and Agricultural Industry: A Review. J. Agric. Food Res.; 2023; 12, 100590. [DOI: https://dx.doi.org/10.1016/j.jafr.2023.100590]

49. Xu, Z.; Pan, Z.; Wang, Y.; Zhang, Y.; Leng, P. Research on a Machine Learning-Based Adaptive and Efficient Screening Model for Psychological Symptoms of Community Correctional Prisoners. Sci. Rep.; 2024; 14, 9890. [DOI: https://dx.doi.org/10.1038/s41598-024-60181-9]

50. Xia, L.; Han, Q.; Shang, L.; Wang, Y.; Li, X.; Zhang, J.; Yang, T.; Liu, J.; Liu, L. Quality Assessment and Prediction of Municipal Drinking Water Using Water Quality Index and Artificial Neural Network: A Case Study of Wuhan, Central China, from 2013 to 2019. Sci. Total Environ.; 2022; 844, 157096. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2022.157096]

51. Chowdhury, S.; Mclellan, P.J. Models for Predicting Disinfection Byproduct (DBP) Formation in Drinking Waters: A Chronological Review. Sci. Total Environ.; 2009; 407, pp. 4189-4206. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2009.04.006]

52. Shahi, N.K.; Maeng, M.; Dockko, S. Models for Predicting Carbonaceous Disinfection By-Products Formation in Drinking Water Treatment Plants: A Case Study of South Korea. Environ. Sci. Pollut. Res.; 2020; 27, pp. 24594-24603. [DOI: https://dx.doi.org/10.1007/s11356-019-05490-7] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31243657]

53. Hong, H.; Zhang, Z.; Guo, A.; Shen, L.; Sun, H.; Liang, Y.; Wu, F.; Lin, H. Radial Basis Function Artificial Neural Network (RBF ANN) as Well as the Hybrid Method of RBF ANN and Grey Relational Analysis Able to Well Predict Trihalomethanes Levels in Tap Water. J. Hydrol.; 2020; 591, 125574. [DOI: https://dx.doi.org/10.1016/j.jhydrol.2020.125574]

54. Hu, G.; Mian, H.R.; Mohammadiun, S.; Rodriguez, M.J.; Hewage, K.; Sadiq, R. Appraisal of Machine Learning Techniques for Predicting Emerging Disinfection Byproducts in Small Water Distribution Networks. J. Hazard. Mater.; 2023; 446, 130633. [DOI: https://dx.doi.org/10.1016/j.jhazmat.2022.130633] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36610346]

55. Li, H.; Zhou, B.; Xu, X.; Huo, R.; Zhou, T.; Dong, X.; Ye, C.; Li, T.; Xie, L.; Pang, W. The Insightful Water Quality Analysis and Predictive Model Establishment via Machine Learning in Dual-Source Drinking Water Distribution System. Environ. Res.; 2024; 250, 118474. [DOI: https://dx.doi.org/10.1016/j.envres.2024.118474]

56. Peng, F.; Lu, Y.; Wang, Y.; Yang, L.; Yang, Z.; Li, H. Predicting the Formation of Disinfection By-Products Using Multiple Linear and Machine Learning Regression. J. Environ. Chem. Eng.; 2023; 11, 110612. [DOI: https://dx.doi.org/10.1016/j.jece.2023.110612]

57. Godo-Pla, L.; Emiliano, P.; Poch, M.; Valero, F.; Monclús, H. Benchmarking Empirical Models for THMs Formation in Drinking Water Systems: An Application for Decision Support in Barcelona, Spain. Sci. Total Environ.; 2021; 763, 144197. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2020.144197]

58. Kulkarni, P.; Chellam, S. Disinfection By-Product Formation Following Chlorination of Drinking Water: Artificial Neural Network Models and Changes in Speciation with Treatment. Sci. Total Environ.; 2010; 408, pp. 4202-4210. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2010.05.040]

59. Lin, J.; Chen, X.; Ansheng, Z.; Hong, H.; Liang, Y.; Sun, H.; Lin, H.; Chen, J. Regression Models Evaluating THMs, HAAs and HANs Formation upon Chloramination of Source Water Collected from Yangtze River Delta Region, China. Ecotoxicol. Environ. Saf.; 2018; 160, pp. 249-256. [DOI: https://dx.doi.org/10.1016/j.ecoenv.2018.05.038]

60. Wang, P.; Ding, S.; Xiao, R.; An, G.; Fang, C.; Chu, W. Enhanced Coagulation for Mitigation of Disinfection By-Product Precursors: A Review. Adv. Colloid Interface Sci.; 2021; 296, 102518. [DOI: https://dx.doi.org/10.1016/j.cis.2021.102518]

61. Cassol, G.S.; Shang, C.; Li, J.; Ling, L.; Yang, X.; Yin, R. Dosing Low-Level Ferrous Iron in Coagulation Enhances the Removal of Micropollutants, Chlorite and Chlorate during Advanced Water Treatment. J. Environ. Sci.; 2022; 117, pp. 119-128. [DOI: https://dx.doi.org/10.1016/j.jes.2022.03.022]

62. Diaz, M.; Daer, S.; Pype, M.L.; Keller, J.; Doederer, K.; Ledezma, P. Tri-Protonated Ferrate: A Novel Electro-Synthesised Pre-Oxidant/Coagulant for NOM and DBP Precursors Removal in Drinking Water Production. J. Water Process Eng.; 2024; 57, 104570. [DOI: https://dx.doi.org/10.1016/j.jwpe.2023.104570]

63. Lin, Q.; Dong, F.; Miao, Y.; Li, C.; Fei, W. Removal of Disinfection Byproducts and Their Precursors during Drinking Water Treatment Processes. Water Environ. Res.; 2019; 92, pp. 698-705. [DOI: https://dx.doi.org/10.1002/wer.1263] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31643120]

64. Li, J.; Song, Y.; Jiang, J.; Yang, T.; Cao, Y. Oxidative Treatment of NOM by Selective Oxidants in Drinking Water Treatment and Its Impact on DBP Formation in Postchlorination. Sci. Total Environ.; 2023; 858, 159908. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2022.159908]

65. Yin, T.; Wu, Y.; Shi, P.; Li, A.; Xu, B.; Chu, W.; Pan, Y. Anion-Exchange Resin Adsorption Followed by Electrolysis: A New Disinfection Approach to Control Halogenated Disinfection Byproducts in Drinking Water. Water Res.; 2019; 168, 115144. [DOI: https://dx.doi.org/10.1016/j.watres.2019.115144]

66. Gusain, R.; Kumar, N.; Ray, S.S. Recent Advances in Carbon Nanomaterial-Based Adsorbents for Water Purification. Coord. Chem. Rev.; 2020; 405, 213111. [DOI: https://dx.doi.org/10.1016/j.ccr.2019.213111]

67. Wu, Z.; Zhang, Y.; Jiang, J.; Pu, J.; Takizawa, S.; Hou, L.; Yang, Y. Insights into Graphene Oxide/Ferrihydrite Adsorption as Pretreatment during Ultrafiltration: Membrane Fouling Mitigation and Disinfection by-Product Control. J. Hazard. Mater.; 2022; 436, 129098. [DOI: https://dx.doi.org/10.1016/j.jhazmat.2022.129098] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35569372]

68. Zhang, C.; Maness, J.; Cuthbertson, A.; Kimura, S.; Liberatore, H.; Richardson, S.; Stanford, B.; Sun, M.; Knappe, D. Treating Water Containing Elevated Bromide and Iodide Levels with Granular Activated Carbon and Free Chlorine: Impacts on Disinfection Byproduct Formation and Calculated Toxicity. Environ. Sci. Water Res. Technol.; 2020; 6, pp. 3460-3475. [DOI: https://dx.doi.org/10.1039/D0EW00523A]

69. Soliman, A.M.; Elshorbagy, W.; Maraqa, M.A.; Al-Issai, L.M.; Sheikh, E.S.E.; Elhaty, I.A.; Ayesh, A.I.; Pál, T. In-Situ, Facile and Green Preparation of Nanoscale Silver Supported on Activated Carbon: Disinfection Properties and Removal of Inorganic DBPs from Drinking Water. Environ. Nanotechnol. Monit. Manag.; 2022; 17, 100621. [DOI: https://dx.doi.org/10.1016/j.enmm.2021.100621]

70. Lau, S.; Feng, Y.; Gu, A.; Russell, C.; Pope, G.; Mitch, W. Cytotoxicity Comparison between Drinking Water Treated by Chlorination with Postchloramination versus Granular Activated Carbon (GAC) with Postchlorination. Environ. Sci. Technol.; 2023; 57, pp. 13699-13709. [DOI: https://dx.doi.org/10.1021/acs.est.3c03591]

71. Siddique, M.S.; Song, Q.; Xiong, X.; Fareed, H.; Graham, N.; Yu, W. Hydrolyzed Polyacrylonitrile UF-Membrane for Surface and TAP Water Treatment: Influence on DBPs Formation and Removal. Chem. Eng. J.; 2023; 471, 144314. [DOI: https://dx.doi.org/10.1016/j.cej.2023.144314]

72. Wang, Y.; Li, L.; Dong, H.; Wang, Q.; Liu, S.; Wu, M.; Yu, J.; Qiang, Z. Removal of Carbonaceous and Nitrogenous Disinfection By-Product Precursors in Biological Activated Carbon Process of Drinking Water: Is Service Life a Pivotal Factor?. Chem. Eng. J.; 2023; 465, 142875. [DOI: https://dx.doi.org/10.1016/j.cej.2023.142875]

73. Youngwilai, A.; Khan, E.; Phungsai, P.; Therdkiattikul, N.; Limpiyakorn, T.; Mhuantong, W.; Ratpukdi, T.; Supanchaiyamat, N.; Hunt, A.J.; Ngernyen, Y. et al. Comparative Investigation of Known and Unknown Disinfection By-Product Precursor Removal and Microbial Community from Biological Biochar and Activated Carbon Filters. Water Res.; 2024; 261, 121994. [DOI: https://dx.doi.org/10.1016/j.watres.2024.121994] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38955037]

74. Zheng, W.; Chen, Y.; Zhang, J.; Peng, X.; Xu, P.; Niu, Y.; Dong, B. Control of Chlorination Disinfection By-Products in Drinking Water by Combined Nanofiltration Process: A Case Study with Trihalomethanes and Haloacetic Acids. Chemosphere; 2024; 358, 142121. [DOI: https://dx.doi.org/10.1016/j.chemosphere.2024.142121] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38677607]

75. Lin, Y.; Lou, Y.; Li, A.; Wei, Y.; Li, H.; Zhou, M.; Li, Y. Effects of Pre-Oxidation by Ozone, Permanganate and Ferrate on Generation and Toxicities of Disinfection Byproducts. Int. J. Environ. Sci. Technol.; 2021; 19, pp. 1-16. [DOI: https://dx.doi.org/10.1007/s13762-021-03526-5]

76. Rougé, V.; von Gunten, U.; Allard, S. Efficiency of Pre-Oxidation of Natural Organic Matter for the Mitigation of Disinfection Byproducts: Electron Donating Capacity and UV Absorbance as Surrogate Parameters. Water Res.; 2020; 187, 116418. [DOI: https://dx.doi.org/10.1016/j.watres.2020.116418]

77. Allard, S.; Kristiana, I.; Andringa-Bate, C.; Joll, C.A. Alternative Application of Preformed Monochloramine as a Drinking Water Disinfectant for Redosing in Long Drinking Water Distribution System Servicing Remote Locations. Water Res.; 2020; 185, 116083. [DOI: https://dx.doi.org/10.1016/j.watres.2020.116083]

78. Bloodgood, M.A.; Chowdary, S.A.; Daiber, E.J.; Shi, H.; Granger, C.O.; Richardson, S.D. A Balancing Act: Optimizing Free Chlorine Contact Time to Minimize Iodo-DBPs, NDMA, and Regulated DBPs in Chloraminated Drinking Water. J. Environ. Sci.; 2022; 117, pp. 315-325. [DOI: https://dx.doi.org/10.1016/j.jes.2022.05.024]

79. Chen, S.; Zhao, Z.; Cui, F.; Liu, B. Comparative Study of UV/Chlorine and VUV/Chlorine as Ultrafiltration Membrane Pretreatment Techniques: Performance, Mechanisms and DBPs Formation. J. Hazard. Mater.; 2023; 459, 132249. [DOI: https://dx.doi.org/10.1016/j.jhazmat.2023.132249]

80. Zhang, J.; Wang, Q.; Dong, L.; Rao, P.; Zhang, X.; Gao, N.; Yin, X. Mechanism, Kinetics and DBPs Formation of UV/NH2Cl Process on Sulfathiazole Removal in Aqueous Solution. Water Resour. Ind.; 2023; 30, 100221. [DOI: https://dx.doi.org/10.1016/j.wri.2023.100221]

81. Liu, H.; Lv, H.; Xu, H.; Rao, D.; Zhang, J.; Sun, B. Is Monochloramine Pre-Oxidation a Viable Strategy for Enhancing the Treatment Efficiency of Algae-Laden Water with Conventional Drinking Water Treatment Process?. Chemosphere; 2024; 352, 141312. [DOI: https://dx.doi.org/10.1016/j.chemosphere.2024.141312]

82. Wang, D.; Chen, X.; Luo, J.; Shi, P.; Zhou, Q.; Li, A.; Pan, Y. Comparison of Chlorine and Chlorine Dioxide Disinfection in Drinking Water: Evaluation of Disinfection Byproduct Formation under Equal Disinfection Efficiency. Water Res.; 2024; 260, 121932. [DOI: https://dx.doi.org/10.1016/j.watres.2024.121932] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38906077]

83. Dowdell, K.S.; Olsen, K.; Paz, E.F.M.; Sun, A.; Keown, J.; Lahr, R.; Steglitz, B.; Busch, A.; LiPuma, J.J.; Olson, T. et al. Investigating the Suitability of Online Flow Cytometry for Monitoring Full-Scale Drinking Water Ozone System Disinfection Effectiveness. Water Res.; 2024; 257, 121702. [DOI: https://dx.doi.org/10.1016/j.watres.2024.121702]

84. Rall, D.; Schweidtmann, A.M.; Aumeier, B.M.; Kamp, J.; Karwe, J.; Ostendorf, K.; Mitsos, A.; Wessling, M. Simultaneous Rational Design of Ion Separation Membranes and Processes. J. Membr. Sci.; 2020; 600, 117860. [DOI: https://dx.doi.org/10.1016/j.memsci.2020.117860]

85. Krippl, M.; Dürauer, A.; Duerkop, M. Hybrid Modeling of Cross-Flow Filtration: Predicting the Flux Evolution and Duration of Ultrafiltration Processes. Sep. Purif. Technol.; 2020; 248, 117064. [DOI: https://dx.doi.org/10.1016/j.seppur.2020.117064]

86. Dias, C.G.; Librantz, A.F.H.; dos Santos, F.C.R. Modeling and Simulation of an Intelligent System for Dosage Control of Post-Chlorination in Water Treatment Plants. Eng. Sanit. Ambient.; 2020; 25, pp. 323-332. [DOI: https://dx.doi.org/10.1590/s1413-41522020173961]

87. Yang, J.; Du, Q.; Ma, R.; Khan, A. Artificial Intelligence Simulation of Water Treatment Using a Novel Bimodal Micromesoporous Nanocomposite. J. Mol. Liq.; 2021; 340, 117296. [DOI: https://dx.doi.org/10.1016/j.molliq.2021.117296]

88. Narges, S.; Ghorban, A.; Khotanlou, H.; Khazaei, M. Prediction of the Optimal Dosage of Coagulants in Water Treatment Plants through Developing Models Based on Artificial Neural Network Fuzzy Inference System (ANFIS). J. Environ. Health Sci. Eng.; 2021; 19, pp. 1543-1553. [DOI: https://dx.doi.org/10.1007/s40201-021-00710-0]

89. Jayaweera, D.; Aziz, N. An Efficient Neural Network Model for Aiding the Coagulation Process of Water Treatment Plants. Environ. Dev. Sustain.; 2022; 24, pp. 1069-1085. [DOI: https://dx.doi.org/10.1007/s10668-021-01483-0]

90. Hu, J.; Kim, C.; Halasz, P.; Kim, J.F.; Kim, J.; Szekely, G. Artificial Intelligence for Performance Prediction of Organic Solvent Nanofiltration Membranes. J. Membr. Sci.; 2021; 619, 118513. [DOI: https://dx.doi.org/10.1016/j.memsci.2020.118513]

91. Setshedi, K.; Mutingwende, N.; Ngqwala, N. The Use of Artificial Neural Networks to Predict the Physicochemical Characteristics of Water Quality in Three District Municipalities, Eastern Cape Province, South Africa. Int. J. Environ. Res. Public Health; 2021; 18, 5248. [DOI: https://dx.doi.org/10.3390/ijerph18105248]

92. Mao, R.; Zhang, K.; Zhang, Q.; Xu, J.; Cen, C.; Pan, R.; Zhang, T. Joint Majorization of Waterworks and Secondary Chlorination Points Considering the Chloric Odor and Economic Investment in the DWDS Using Machine Learning and Optimization Algorithms. Water Res.; 2022; 220, 118595. [DOI: https://dx.doi.org/10.1016/j.watres.2022.118595] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35613482]