Critical Review on Bromate Formation during Ozonation and Control Options for Its Minimization

Homogeneous versus heterogeneous Mn(II) oxidation in peroxymonosulfate assisting chlorination: Synergistic role for enhanced Mn(II) oxidation in water treatment

Removal of Mn(II) is an essential step for addressing water discoloration in water treatment utilities worldwide. However, conventional chlorination suffers from poor oxidation of Mn(II) due to its low homogeneous oxidation kinetics. This study explored the oxidation capability of a new chemical dosing strategy employing peroxymonosulfate (PMS) to assist the chlorination process (PMS@Cl2) for effective Mn(II) oxidation. The study comprehensively explored both oxidation kinetics and underlying mechanisms associated with homogeneous and heterogeneous oxidation within the PMS@Cl2 system. At an [Mn(II)]0 of 1 mg/L, chlorination demonstrated inability in oxidizing Mn(II), with <10 % oxidation even at an elevated [Cl2] of 150 μM (∼10 mg/L). By contrast, PMS completely oxidized 100 % Mn(II) within a 30-minute reaction at a much lower [PMS] of 60 μM (kobs = 0.07 min-1 and t1/2 = 9 min), demonstrating its superior Mn(II) oxidation kinetics (over one order of magnitude faster than conventional chlorine). PMS@Cl2 exhibited an interesting synergistic benefit when combining a lower dose PMS with a higher routine dose Cl2 (loPMS@hiCl2), e.g. [PMS]:[Cl2] at 15:30 or 30:30 μM. Both conditions achieved 100 % Mn(II) oxidation, with even better values of kobs and t1/2 (0.16-0.17 min-1 and ∼4 min) relative to PMS alone at 60 µM. The synergic benefit of PMS@Cl2 was attributed to distinct functions played by PMS and Cl2 in both homogeneous and heterogeneous oxidation processes. Reactive species identification excluded the possible involvement of SO4•-, OH•, or chlorine radicals in the homogeneous oxidation of the PMS@Cl2 system. Instead, the dominant species was O2•- radical generated during the reaction of Mn(II) and PMS. Furthermore, the heterogeneous oxidation emphasized the important role of combining Cl2 dosing, which demonstrated an increased reactivity and electron transfer with the Mn-O-Mn complex, surpassing PMS. Overall, heterogeneous oxidation accelerated the oxidation kinetics of the PMS@Cl2 system by 1.1-2 orders of magnitude relative to the homogeneous oxidation of Cl2 alone. We here demonstrated that PMS@Cl2 could offer a more efficient mean of soluble Mn(II) mitigation, achieved with a relatively low routine dose of oxidant in a short reaction period. The outcomes of this study would address the existing limitations of traditional chlorine oxidation, minimizing the trade-offs associated with high residual chlorine levels after treatments for soluble manganese-containing water.

Prediction of hydroxyl radical exposure during ozonation using different machine learning methods with ozone decay kinetic parameters

The abatement of micropollutants by ozonation can be accurately calculated by measuring the exposures of molecular ozone (O3) and hydroxyl radical (•OH) (i.e., ∫[O3]dt and ∫[•OH]dt). In the actual ozonation process, ∫[O3]dt values can be calculated by monitoring the O3 decay during the process. However, calculating ∫[•OH]dt is challenging in the field, which necessitates developing models to predict ∫[•OH]dt from measurable parameters. This study demonstrates the development of machine learning models to predict ∫[•OH]dt (the output variable) from five basic input variables (pH, dissolved organic carbon concentration, alkalinity, temperature, and O3 dose) and two optional ones (∫[O3]dt and instantaneous ozone demand, IOD). To develop the models, four different machine learning methods (random forest, support vector regression, artificial neural network, and Gaussian process regression) were employed using the input and output variables measured (or determined) in 130 different natural water samples. The results indicated that incorporating ∫[O3]dt as an input variable significantly improved the accuracy of prediction models, increasing overall R2 by 0.01-0.09, depending on the machine learning method. This suggests that ∫[O3]dt plays a crucial role as a key variable reflecting the •OH-yielding characteristics of dissolved organic matter. Conversely, IOD had a minimal impact on the accuracy of the prediction models. Generally, machine-learning-based prediction models outperformed those based on the response surface methodology developed as a control. Notably, models utilizing the Gaussian process regression algorithm demonstrated the highest coefficients of determination (overall R2 = 0.91-0.95) among the prediction models.

Achieving realistic ozonation conditions with synthetic water matrices comprising low-molecular-weight scavenger compounds

Ozonation is used worldwide for drinking water disinfection and increasingly also for micropollutant abatement from wastewater. Identification of transformation products formed during the ozonation of micropollutants is challenging due to several factors including (i) the reactions of both oxidants, ozone and hydroxyl radicals with the micropollutants, as well as with intermediate transformation products, (ii) effects of the water matrix on the ozone and hydroxyl radical chemistry and (iii) the generation of oxidation by-products. In this study, a simple approach to achieve realistic ozonation conditions in the absence of dissolved organic matter has been developed. It is based on composing synthetic water matrices with low-molecular-weight scavenger compounds (phenol, methanol, acetate, and carbonate) that mimic the chemical interactions of ozone and hydroxyl radicals with real water matrices. Synthetic waters composed of only four low-molecular-weight compounds successfully replicated two lake waters and two secondary wastewater effluents, matching instantaneous ozone demand, ozone and hydroxyl radical exposures in the initial phase, as well as the ozone evolution in the second phase of the ozonation process. The synthetic water matrices also reproduced the effects of temperature and pH changes observed in real waters. The abatement of two micropollutants, bezafibrate and atrazine, and the formation of the corresponding transformation products during ozonation were in agreement for synthetic and real waters. Furthermore, the kinetics and extent of bromate formation during ozonation in synthetic water were comparable to real lake water and wastewater. This supports the robustness of the proposed approach because bromate formation is very sensitive to the interplay of ozone and hydroxyl radicals. Furthermore, with the novel reaction system, a significant effect of hydroxyl radicals scavenging by carbonate on bromate formation was demonstrated. Overall, the herein-developed approach based on synthetic water matrices allows to perform realistic ozonation studies including both oxidants, ozone and hydroxyl radicals, without the constraints of using dissolved organic matter.

Virus inactivation in low ozone exposure water reuse applications

In drinking water applications, an ozone exposure (Ct) based framework has been historically used to validate ozone disinfection. However, significant viral inactivation can be achieved with little to no measurable ozone exposure. Additionally, ozone exposure depends on multiple water quality variables as well as the calculation/ozone measurement method used. In this study, we evaluated alternative ozone monitoring frameworks as well as the impact of water quality variables on ozone decay kinetics and virus/coliform inactivation. Here we show that both change in UV254 absorbance and applied O3:TOC were well correlated with viral inactivation and these frameworks were resilient to changes in water quality. Both increasing temperature (12-30 ⁰C) and pH (5.5-8.4) was shown to significantly increase the ozone decay rate and decreased the resulting ozone exposure by as much as ∼90% in the case of pH. However, due to the increased reaction rate of ozone with viruses at elevated temperature and pH, there was only a minor impact (∼20% in the case of pH) in overall disinfection performance for a given O3:TOC. These frameworks were also considered for variable source water with TOC (5-11 mg/L) and TSS (1.2-5.8 mg/L). Change in UV254 absorbance or applied ozone dose (mg/L) were the strongest indicators of disinfection performance for source waters of variable TOC, however site-specific testing may be needed to apply this framework. Challenge testing with influent nitrite indicated that ozone disinfection performance is significantly impacted (>50% reduction in inactivation) in the presence of nitrite thus enforcing the importance of accounting for this value in the applied ozone dose. Multi-point ozone dissolution was investigated as an alternative ozone application method that may present a benefit with respect to overall disinfection performance especially if nitrite was present. Developing and validating these alternative monitoring frameworks and ozone application methods is imperative in water reuse applications where unnecessary elevated ozone exposure may lead to harmful byproduct formation.

Robust Activity and Stability of P-Doped Fe-Carbon Composites Derived from MOF for Bromate Reduction

Iron-based materials are effective for the reductive removal of the disinfection byproduct bromate in water, while the construction of highly stable and active Fe-based materials with wide pH adaptability remains greatly challenging. In this study, highly dispersed iron phosphide-decorated porous carbon (Fe2P(x)@P(z)NC-y) was prepared via the thermal hydrolysis of Fe@ZIF-8, followed by phosphorus doping (P-doping) and pyrolysis. The reduction performances of Fe2P(x)@P(z)NC-y for bromate reduction were evaluated. Characterization results showed that the Fe, P, and N elements were homogeneously distributed in the carbonaceous matrix. P-doping regulated the coordination environment of Fe atoms and enhanced the conductivity, porosity, and wettability of the carbonaceous matrix. As a result, Fe2P(x)@P(1.0)NC-950 exhibited enhanced reactivity and stability with an intrinsic reduction kinetic constant (kint) 1.53-1.85 times higher than Fe(x)@NC-950 without P-doping. Furthermore, Fe2P(0.125)@P(1.0)NC-950 displayed superior reduction efficiency and prominent stability with very low Fe leaching (4.53-22.98 μg L-1) in a wide pH range of 4.0-10.0. The used Fe2P(0.125)@P(1.0)NC-950 could be regenerated by phosphating, and the regenerated Fe2P(0.125)@P(1.0)NC-950 maintained 85% of its primary reduction activity after five reuse cycles. The study clearly demonstrates that Fe2P-decorated porous carbon can be applied as a robust and stable Fe-based material in aqueous bromate reduction.

Kinetic modelling assisted balancing of organic pollutant removal and bromate formation during peroxone treatment of bromide-containing waters

The peroxone process (O3/H2O2) is reported to be a more effective process than the ozonation process due to an increased rate of generation of hydroxyl radicals (•OH) and inhibition of bromate (BrO3-) formation which is otherwise formed on ozonation of bromide containing waters. However, the trade-off between the H2O2 dosage required for minimization of BrO3- formation and effective pollutant removal has not been clearly delineated. In this study, employing experimental investigations as well as chemical modelling, we show that the concentration of H2O2 required to achieve maximum pollutant removal may not be the same as that required for minimization of BrO3- formation. At the H2O2 dosage required to minimize BrO3- formation (<10 µg/L), only pollutants with high to moderate reactivity towards O3 and •OH are effectively removed. For pollutants with low reactivity towards O3/•OH, high O3 (O3:DOC>>1 g/g) and high H2O2 dosages (O3:H2O2 ∼1 (g/g)) are required for minimizing BrO3- formation along with effective pollutant removal which may result in a very high residual of H2O2 in the effluent, causing secondary pollution. On balance, we conclude that the peroxone process is not effective for the removal of low reactivity micropollutants if minimization of BrO3- formation is also required.

PFAS removal from landfill leachate by ozone foam fractionation: System optimization and adsorption quantification

Landfills are the primary endpoint for the disposal of PFAS-laden waste, which subsequently releases PFAS to the surrounding environments through landfill leachate. Ozone foam fractionation emerges as a promising technology for PFAS removal to address the issue. This study aims to (i) assess the effectiveness of the ozone foam fractionation system to remove PFAS from landfill leachate, and (ii) quantify equilibrium PFAS adsorption onto the gas-water interface of ozone bubbles, followed by a comparison with air foam fractionation. The results show that ozone foam fractionation is effective for PFAS removal from landfill leachate, with more than 90 % long-chain PFAS removed. The identified operating conditions provide valuable insights for industrial applications, guiding the optimization of ozone flow rates (1 L/min), dosing (43 mg/L) and minimizing foamate production (4 % wettability). The equilibrium modelling reveals that the surface excess of air bubbles exceeds that of ozone bubbles by 20-40 % at a corresponding PFAS concentration. However, the overall removal of PFAS from landfill leachate by ozone foam fractionation remains substantial. Notably, ozone foam fractionation generates foamate volumes 2 - 4 times less, resulting in significant cost savings for the final disposal of waste products and reduced site storage requirements.

Oxidation processes and me

This publication summarizes my journey in the field of chemical oxidation processes for water treatment over the last 30+ years. Initially, the efficiency of the application of chemical oxidants for micropollutant abatement was assessed by the abatement of the target compounds only. This is controlled by reaction kinetics and therefore, second-order rate constant for these reactions are the pre-requisite to assess the efficiency and feasibility of such processes. Due to the tremendous efforts in this area, we currently have a good experimental data base for second-order rate constants for many chemical oxidants, including radicals. Based on this, predictions can be made for compounds without experimental data with Quantitative Structure Activity Relationships with Hammet/Taft constants or energies of highest occupied molecular orbitals from quantum chemical computations. Chemical oxidation in water treatment has to be economically feasible and therefore, the extent of transformation of micropollutants is often limited and mineralization of target compounds cannot be achieved under realistic conditions. The formation of transformation products from the reactions of the target compounds with chemical oxidants is inherent to oxidation processes and the following questions have evolved over the years: Are the formed transformation products biologically less active than the target compounds? Is there a new toxicity associated with transformation products? Are transformation products more biodegradable than the corresponding target compounds? In addition to the positive effects on water quality related to abatement of micropollutants, chemical oxidants react mainly with water matrix components such as the dissolved organic matter (DOM), bromide and iodide. As a matter of fact, the fraction of oxidants consumed by the DOM is typically > 99%, which makes such processes inherently inefficient. The consequences are loss of oxidation capacity and the formation of organic and inorganic disinfection byproducts also involving bromide and iodide, which can be oxidized to reactive bromine and iodine with their ensuing reactions with DOM. Overall, it has turned out in the last three decades, that chemical oxidation processes are complex to understand and to manage. However, the tremendous research efforts have led to a good understanding of the underlying processes and allow a widespread and optimized application of such processes in water treatment practice such as drinking water, municipal and industrial wastewater and water reuse systems.

Highly efficient reduction of bromate by vacuum UV/sulfite system

Bromate (BrO3-), a worldwide regulated by-product after ozone disinfection, is often detected in bromide-containing water, and has a strict limit of 10 μg L-1 in potable water. BrO3- degradation by advanced reduction processes (ARPs) has gained much attention because of efficient removal and easy integration with ultraviolet disinfection (UV at 254 nm). In the vacuum UV (VUV, 185/254 nm)/sulfite system, the elimination kinetics of BrO3- increased by 9-fold and 15-fold comparing with VUV alone and UV/sulfite system. This study further demonstrated the hydrated electron (eaq-) works as the dominant species in BrO3- degradation in alkaline solution, while in the acidic solution the H• became a secondary reactive species besides eaq-. Hence, the influences of pH, sulfite concentration, dissolved gas and water matrix on effectiveness of degradation kinetics of BrO3- was explored in details. With increasing pH, the proportion of SO32- species increased and even became the major ones, which also correlated well with the kobs (min-1) of BrO3- degradation. The stability of eaq- also climbs with increasing pH, while that of H• drops significantly. Higher sulfite dosage favored a more rapid degradation of BrO3-. The presence of dissolved oxygen inhibited BrO3- removal due to the scavenging effect of O2 toward eaq- and transformed VUV/sulfite-based ARP to an advanced oxidation process (AOP), which was ineffective for BrO3- removal. BrO3- removal was inhibited to varying degrees after anions (e.g., bicarbonate (HCO3-), chloride (Cl-), nitrate (NO3-)) and humic acid (HA) being added.