Simultaneous time-resolved aqueous haloamine measurement empowers robust kinetic model analysis

doi:10.21203/rs.3.rs-3591268/v1

Download PDF

Article

Simultaneous time-resolved aqueous haloamine measurement empowers robust kinetic model analysis

https://doi.org/10.21203/rs.3.rs-3591268/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 23 Apr, 2024

Read the published version in Nature Water →

Version 1

posted

You are reading this latest preprint version

We demonstrate the first application of Proton Transfer Time-of-Flight Mass Spectrometry (PTR-ToF-MS) for monitoring kinetics of disinfectant decay in water with 1 to 3 orders of magnitude greater sensitivity than other analytical methods. Chemical disinfection inactivates pathogens during water treatment and prevents regrowth as water is conveyed in distribution system pipes, but it also causes formation of toxic disinfection by-products (DBPs). Analytical limits have hindered kinetic models which aid in ensuring water quality and protecting public health by predicting disinfectant DBPs formation. PTR-ToF-MS, designed for measuring gas phase concentrations of organic compounds, was able to simultaneously monitor aqueous concentrations of five haloamines relevant to chloramine disinfection under drinking water relevant concentrations. This novel application to aqueous analytes opens a new range of applications for PTR-ToF-MS.

Physical sciences/Chemistry/Analytical chemistry/Mass spectrometry

Physical sciences/Chemistry/Environmental chemistry

The adoption of drinking water disinfection with free chlorine in the early 20th century has made incidences of waterborne disease outbreaks rare in the United States (U.S.). However, the ability to predict the decay of disinfectants and formation of toxic disinfection byproducts (DBPs) has remained a vexing issue for water utilities ^1,2. A recent paper indicated that 9–45 million Americans are affected annually by health-based water quality violations particularly in rural and minority communities ³. Most of the violations are due to coliform bacteria (37%) and DBP rule violations (25%), both of which are linked to issues with disinfection practices. A common way to address these issues is to use chloramines as secondary disinfectants because they form fewer regulated DBPs and persist longer in water distribution systems ^4,5. Considerable work has been done since the 1980s to develop robust kinetic models for chloramine decay⁶ and DBP formation⁷ in drinking water, but an aspect that has remained elusive and is emerging as an issue in such models is the role of bromide (Br⁻) which is common in source waters ⁸. The presence of Br⁻ in chloraminated waters leads to the formation of brominated haloamines (brominated analogues of chloramines) and brominated DBPs which are more toxic than their chlorinated counterparts and are often the driver of overall human toxicity (less abundant but more toxic DBPs form) ⁹. With increasing total dissolved solid concentrations in freshwaters and the use of alternative water sources with higher Br⁻ concentrations (e.g. seawater, municipal wastewater), accurate measurements of haloamines are needed to develop reliable models for the formation and transformation of brominated species in drinking water ^10,11. The objectives of this work are to i) demonstrate that proton transfer time of flight mass spectrometry (PTR-ToF-MS) can be used to measure concentrations of trace volatile analytes in the liquid phase and ii) to show the methodology can be used to monitor haloamine reactions in real time to expand our knowledge and prediction capabilities of haloamine formation and decay.

Most U.S. drinking water plants utilize two disinfection periods: primary for inactivating microorganisms and secondary for maintaining water quality and preventing (re)growth of pathogens in water distribution systems and premise plumbing. Chloramines are generally formed by first adding free chlorine (prechlorination) for primary disinfection followed by ammonia to form chloramines which are used as secondary disinfectants ¹². For pH ranges typical of drinking water (pH 7 to 9), monochloramine (NH₂Cl) is the most abundant chloramine species; dichloramine (NHCl₂) also forms but at lower concentrations ⁶.

In Br⁻ containing waters, prechlorination oxidizes Br⁻ to free bromine. Subsequently, when ammonia is added to form chloramines, a dynamic mixture of five haloamines forms: NH₂Cl, NHCl₂, monobromamine (NH₂Br), dibromamine (NHBr₂), and bromochloramine (NHBrCl). The formation of NH₂Br, NHBr₂, and NHBrCl is undesired compared to NH₂Cl and NHCl₂ because they decay faster ^13,14 and form greater concentrations of brominated DBPs ¹⁵. The relevant reaction kinetics among the five haloamines and in waters containing natural organic matter (NOM, the precursor to DBPs) are complex and incompletely understood, and yet are a prerequisite for developing mechanistic kinetic models to understand disinfectant residual maintenance, minimize brominated DBP formation, and manage water distribution systems in a manner that ensures public health. A better understanding of the formation of brominated DBPs is urgently needed, because recent work has shown that unregulated DBPs, such as brominated haloacetonitriles are the major drivers of toxicity in drinking water ⁹, the U.S. EPA is currently considering expanding the DBP rule to specifically regulate brominated haloacetic acids ¹⁶.

A major limitation to the study of chloramine chemistry has been the lack of an analytical method that can in real time accurately measure all relevant haloamines (some rapidly form and decay) at concentrations relevant to drinking water. Current analytical methods for measuring haloamines are ultraviolet-visible (UV-Vis) spectroscopy, a combination of colorimetric tests, and membrane introduction mass spectrometry (MIMS). UV-Vis spectroscopy has been used to study haloamine chemistry but requires concentrations much higher than seen in drinking water and overlapping UV spectra make it difficult to resolve quantitatively the several species present in solution ^17,18. The indophenol¹⁹ and total chlorine DPD²⁰ colorimetric methods are widely employed by water utilities. While these two methods are often used in combination, the results can be confounding; the indophenol method specifically measures NH₂Cl, whereas the total chlorine DPD method measures the total oxidant concentration which, in the case of chloraminated waters, is the total haloamine concentration on a halogen basis ([NH₂Cl] + 2[NHCl₂] + [NH₂Br] + 2[NHBrCl] + 2[NHBr₂]). MIMS uses a semipermeable hydrophobic membrane that rejects water to introduce dissolved analytes to a mass spectrometer that identifies and quantifies analytes based on their mass to charge ratio (m/z). MIMS configurations generally use an electron ionization source which generates many fragment ions from the parent molecule. In haloamine systems, this fragmentation results in dihaloamines (NHCl₂, NHBr₂, and NHBrCl) interfering with the quantification of monohaloamines (NH₂Cl and NH₂Br) ^21,22. Fragmentation has made the simultaneous monitoring of all five haloamines with this technique impossible when there are relatively large concentrations of the three brominated haloamines, particularly NHBrCl because it interferes with both NH₂Cl and NH₂Br.

PTR-ToF-MS is an analytical technique that uses chemical ionization with hydronium ion (H₃O⁺) to measure gas phase volatile organic compounds (VOCs) as well as inorganic compounds at single parts per trillion concentrations ²³. Chemical ionization is a “soft ionization” process that results in significantly less fragmentation than electron ionization commonly used with MIMS because less energy is imparted on the parent compound when it is ionized ²⁴. One group of researchers utilized H₃O⁺ as a chemical ionization source in conjunction with MIMS and found that chloramines and NH₂Br degraded significantly within the membrane such that the signals for the parent compounds were weaker than those of decay products ²⁵; an issue that negated the advantages of chemical ionization when used in conjunction with MIMS. PTR-ToF-MS has been used to measure chloramines in the air at indoor pools ²⁶ and following cleaning with ammonia and chlorine-based cleaners ²⁷. Selected ion flow tube mass spectrometry, a similar technology to PTR-ToF-MS that is orders of magnitude less sensitive, has been used to detect and characterize NH₂Cl, NHCl₂, and NH₂Br in human breath ²⁸. The impressive sensitivity of PTR-ToF-MS for measuring gas phase compounds has the potential for correlating gas and aqueous concentrations of haloamines.

In this work, we demonstrate expanded functionality of PTR-ToF-MS by being the first to use it to examine drinking water relevant concentrations of haloamines in real time, thus enabling us to simultaneously quantify aqueous concentrations of haloamines as they form and decay. The data acquired using PTR-ToF-MS enables estimation of relevant kinetic parameters, assessment of relevant competing reactions, and validation of any already developed kinetic model. More generally, we show that PTR-ToF-MS can be used to measure concentrations of trace volatile analytes in a liquid phase via headspace analysis, thereby exposing the potential for a wide range of new applications.

2.1 Mass spectra of haloamines using PTR-ToF-MS

The protonated molecular ions used for quantification of haloamines were m/z 51.995 (NH₂ClH⁺), 85.956 (NHCl₂H⁺), 95.944 (NH₂BrH⁺), 131.903 (NHBrClH⁺), and 175.852 (NHBr₂H⁺) because they were the dominant ions. In the mass spectrum for each haloamine (Fig. 1), the protonated molecular ion for acetone, C₃H₆OH⁺ (m/z 59.049), and acetic acid, C₂H₄O₂H⁺ (m/z 61.028), are present because they are ubiquitous in the ambient air of indoor spaces and captured due to the open headspace sampling method utilized in this work ²⁷. PTR-ToF-MS sampling of ambient air in the laboratory contained peaks corresponding to acetone and acetic acid (Fig. S1). Additionally, the NHCl₂ and NHBrCl mass spectra (Fig. 1B and 1D) have significant peaks of the protonated ion of acetic acid because those experiments were buffered with acetate and performed at pH 4 and 5, respectively.

The mass spectrum for NH₂Br featured the molecular ions NH₂⁷⁹BrH⁺(m/z 95.944) and NH₂⁸¹BrH⁺ (m/z 97.942) (Fig. 1C). The mass spectrum for NHBr₂ (Fig. 1E) additionally revealed the expected molecular ions NH⁷⁹Br₂H⁺ (m/z 173.854), NH⁷⁹Br⁸¹BrH⁺(m/z 175.852), and NH⁸¹Br₂H⁺ (m/z 177.851), as well as the molecular ions of many potential fragment ions, including NH⁷⁹BrH⁺ (m/z 94.936), NH⁸¹BrH⁺ (m/z 96.935), ⁷⁹Br₂^•+ (m/z 157.837), ⁷⁹Br₂⁸¹Br^•+ (m/z 159.834), ⁸¹Br₂^•+ (m/z 161.832), NH⁷⁹Br₂⁺ (m/z 172.842), and NH⁷⁹Br⁸¹Br⁺ (m/z 174.842). With MIMS, the signals for Br₂^•+ ions have been two²⁹ and tenfold ³⁰ as high as the signal for NHBr₂, while with PTR-ToF-MS, it is only a tenth of NHBr₂ highlighting the cleaner mass spectrum obtained with PTR-ToF-MS versus MIMS. Additional discussion about the mass spectra of the other haloamines is presented in the Supplementary Material (Section S2.1).

The expected isotopic ratios and characteristic mass defects of the haloamines detected by the PTR-ToF-MS led to high confidence of ion identification (within > 1 mDa mass accuracy). The high resolution (5 decimal points, simplified by rounding to 3 decimal points in this work) m/z for molecular and fragment ions as well as the relative mass error (Eq. 1) for all five haloamine isotopes were quantified (Table 1). The average mass deviation for the haloamines detected was 4.73 ± 2.14 ppm which shows high confidence of ion identification because ions with similar mass values can be resolved separately.

Table 1

Protonated parent ions detected by PTR-ToF-MS and their mass defects compared to exact values. Dominant ions in bold. The trihaloamine species were detected but not quantified.
Haloamine	Ion Formula	Isotopic Ratio	Exact m/z	Detected m/z	Mass Error (ppm)
Monochloramine	NH₂³⁵ClH⁺	1.00	51.99485	51.9945	6.73
Monochloramine	NH₂Cl³⁷H⁺	0.32	53.99190	53.9919	0.00
Dichloramine	NH³⁵Cl₂H⁺	1.00	85.95588	85.9564	6.05
Dichloramine	NH³⁵Cl³⁷ClH⁺	0.64	87.95293	87.9524	6.03
Monobromamine	NH₂⁷⁹BrH⁺	1.00	95.94434	95.9438	5.63
Monobromamine	NH₂⁸¹BrH⁺	0.97	97.94229	97.9422	0.92
Bromochloramine	NH³⁵Cl⁷⁹BrH⁺	0.77	129.90537	129.904	10.55
	NH³⁷Cl⁷⁹BrH⁺_, NH³⁵Cl⁸¹BrH⁺	1.00	131.90309	131.903	0.68
	NH³⁷Cl⁸¹BrH ⁺	0.24	133.90037	133.9	2.76
Dibromamine	NH⁷⁹Br₂H⁺	0.51	173.85485	173.854	4.89
	NH⁷⁹Br⁸¹BrH⁺	1.00	175.85280	175.852	4.55
	NH⁸¹Br₂H⁺	0.49	177.85075	177.851	1.41
Acetone	C₃H₆OH⁺	--	59.04914	59.0492	1.02
Acetic Acid	C₂H₄O₂H⁺	--	61.02840	61.0284	0.00
Trichloramine	N³⁵Cl₃H⁺	1.00	119.90700	119.9169	82.63
Trichloramine	N³⁵Cl₂³⁷ClH⁺	0.96	121.96000	121.914	377.66
Tribromamine	NH⁷⁹Br₃H⁺	0.34	251.76500	251.7654	1.44
	NH⁷⁹Br₂⁸¹BrH⁺	1.00	253.76300	253.7633	1.24
	NH⁷⁹Br⁸¹Br₂H⁺	0.97	255.76100	255.7613	1.05

$$relative mass error \left(ppm\right)=\frac{{m/z}_{detected}-{m/z}_{exact}}{{m/z}_{exact}}\times {10}^{6}$$

As expected, the mass spectra acquired using PTR-ToF-MS were simpler, had fewer fragment ions, and contained more distinct peaks for parent ions than those captured using MIMS. Fragment ions still formed even though chemical ionization with H₃O⁺ is a much lower energy process than electron ionization. Some fragments were likely formed due to impurities in the chemical ionization process which results in the small formation of O₂⁺ as part of the ionization gas which imparts more energy than H₃O^{+ 26}. The impact of the fragment ions from NHCl₂, NHBrCl, and NHBr₂ on the quantification of NH₂Cl and NH₂Br is further explored in Section 3.3.

2.2 Calibration curves and detection limits for haloamines using PTR-ToF-MS

Calibration curves (Fig. S2) for each haloamine were successfully developed (see discussion in Section S2.2) which demonstrates that PTR-ToF-MS can be used to effectively quantify aqueous haloamine concentrations with headspace measurements at drinking water relevant concentrations. More broadly the results show that PTR-ToF-MS can be used to measure aqueous concentrations, which no other studies have previously attempted. This opens the potential for much broader applications of PTR-ToF-MS technology to quantify low concentrations of volatile compounds in aqueous solutions.

The level of detection (LOD) and level of quantification (LOQ) using PTR-ToF-MS were determined for each haloamine (Table S1). The LOD and LOQ were estimated by the method of blank determination where the average haloamine concentration of ten sample blanks plus three and ten times the standard deviation, respectively for LOD and LOQ ³¹. The LODs of NH₂Cl, NHCl₂, NH₂Br, NHBrCl, and NHBr₂ were 0.90, 0.0023, 0.059, 0.0086, and 0.00072 µM. The LOQs of NH₂Cl, NHCl₂, NH₂Br, NHBrCl, and NHBr₂ were 1.1, 0.0045, 0.092, 0.016, and 0.0016 µM. The lowest LODs determined with MIMS for NH₂Cl, NHCl₂, NH₂Br, NHBrCl, and NHBr₂ were 0.48, 0.24, 1.44, 2.51, and 0.84 µM respectively ²². PTR-ToF-MS has LODs 1 to 3 orders of magnitude lower than MIMS for all haloamines except NH₂Cl. The improved sensitivity for the haloamines other than NH₂Cl is particularly important because they are present at much lower concentrations than NH₂Cl in chloraminated waters.

2.3 Impact of dihaloamine fragments on quantification of monohaloamines

The dihaloamines (NHCl₂, NHBrCl, and NHBr₂) have been shown to form fragments (Section 3.1) that contribute to the molecular ions for NH₂Cl (m/z 51.995) and NH₂Br (m/z 95.944). The contribution from dihaloamines to the fragments at m/z 51.995 and 95.944 must be accounted for to ensure all haloamines are accurately quantified in a mixture. A set of interference calibration curves was developed to account for contribution of the fragment ions from NHCl₂, NHBrCl, and NHBr₂ to those two m/z values (Fig. 2). All the interference calibration curves were fit linearly (R² > 0.92) with an intercept of zero, except for NHCl₂ (Fig. 2A). The nonzero intercept for NHCl₂ is likely due to small amounts of HOCl (which can be present at the low pH value of 4 that NHCl₂ was formed in this work) being ionized and interfering at m/z 51.995. This phenomenon did not impact the NHBrCl interference calibration curve (Fig. 2B) because less HOCl would be present at the higher pH of 5 and HOCl would rapidly react with Br⁻ present in the NHBrCl solutions.

PTR-ToF-MS has much lower levels of fragmentation than MIMS, which makes it a more viable option for collecting robust kinetic data of haloamines. The contribution of 10 µM of dihaloamines to the signals of monohaloamines for PTR-ToF-MS and MIMS is compared in Table 2. The fragments formed during chemical ionization using PTR-ToF-MS are approximately 2 to 20% of those formed from electron ionization by MIMS. The much lower level of fragmentation of NHBrCl and NHBr₂ is particularly important because these dihaloamines can be present in much greater concentrations than NHCl₂ in solutions containing Br⁻.

Table 2

Contribution of 10 µM of dichloramine (NHCl₂), bromochloramine (NHBrCl), and dibromamine (NHBr₂) to signals associated with monochloramine (NH₂Cl) and monobromamine (NH₂Br)
	PTR-ToF-MS (this work)		MIMS (Pope, 2006)		MIMS (Hu et al., 2021)
Dihaloamine	NH₂Cl	NH₂Br	NH₂Cl	NH₂Br	NH₂Cl	NH₂Br
NHCl₂	0.3 µM	-	17 µM	-	5.2 µM	-
NHBrCl	3.2 µM	1.4 µM	50 µM	7.8 µM	NM*	NM*
NHBr₂	-	3.4 µM	-	28 µM	-	21 µM
* Not measured (NM)

2.4 Effectiveness of PTR-ToF-MS at monitoring dynamic aqueous haloamine kinetics

While the preceding discussion highlighted the ability to accurately measure haloamine compounds at environmentally relevant concentrations, a primary goal of this work is to demonstrate the potential for this method to guide model development and predict haloamine formation and decay. Two kinetic experiments were performed, with and without NOM (the major precursor to DBPs found in all source water), to evaluate the effectiveness of using PTR-ToF-MS to measure haloamines in a dynamic mixture. Experiments were designed to simulate the chloramination process used in practice; chloramines are commonly formed with prechlorination followed one minute later by ammonia addition to form chloramines. During each experiment, PTR-ToF-MS was used to simultaneously monitor the five haloamine concentrations. The indophenol and total chlorine DPD colorimetric methods were also used to measure NH₂Cl and total haloamine ([NH₂Cl] + 2[NHCl₂] + [NH₂Br] + 2[NHBrCl] + 2[NHBr₂]) concentrations, respectively.

The effectiveness of using PTR-ToF-MS for monitoring haloamine concentrations was assessed by comparing the results to the colorimetric methods in both kinetic experiments. The NH₂Cl and total haloamine concentrations measured using PTR-ToF-MS were in good agreement with those determined with the established colorimetric methods with differences ranging from 0.4 to 3.7 µM (4.1 to 13%) for the experiment without NOM (Fig. 3A) and 0.03 to 6.2 µM (0.1 to 37%) for the experiment with NOM (Fig. 3B). The consistency between the two total haloamine measurements shows that the mass balance on the haloamines measured using PTR-ToF-MS is correct. Accurate NH₂Cl measurements highlight the ability to account for fragment ions because the most significant impact of fragmentation occurs when NHBrCl interferes with NH₂Cl quantification (Table 2), and NHBrCl forms rapidly during the kinetic experiments (Fig. 4). Additionally, the consistency between the PTR-ToF-MS and colorimetric method measurements in the NOM experiment shows that a low concentration (2 mg/L as C) of organic carbon, typical of natural waters, does not significantly impact or interfere with the PTR-ToF-MS method for quantifying haloamines.

2.5 Implications for existing haloamine models

The kinetic experiment monitored by PTR-ToF-MS shows that, in the absence of NOM, the concentrations of the various haloamines changed substantially during the short 30-minute experiments (Fig. 4). NH₂Cl and NH₂Br were rapidly consumed to form NHBrCl and NHBr₂ and the dihaloamines then decayed. NHCl₂ was excluded from Fig. 4 because it was present at very low concentrations (< 0.1 µM) and did not change substantially. The experimental data were also compared to the most widely used haloamine model (Fig. 4, Table S2) and another model developed by Pope using MIMs (Fig. S3, Table S3). The two kinetic models yield similar simulated concentrations of haloamines; therefore, we will focus on the comparison to the kinetic model developed by Luh and Mariñas (L&M model) ¹⁸.

Substantial discrepancies are apparent between the L&M model and the data collected using PTR-ToF-MS, which highlights the significant deficiencies in existing kinetic models to describe haloamine formation and decay at conditions representative of actual chloramine application to drinking water (e.g., pH 7–9, < 4 mg Cl₂/L, and chloramines formed with a prechlorination step). The model greatly underpredicts NH₂Cl decay and NHBrCl formation, while also overpredicting NHBr₂ formation. The predicted dominant brominated haloamine was NHBr₂ while the PTR-ToF-MS data showed that NHBrCl was the dominant brominated haloamine after the first two minutes. Therefore, if one relied on the L&M model to understand brominated DBP formation, the model estimated dominant brominated haloamine, NHBr₂, could be incorrectly associated with DBP formation instead of NHBrCl.

The differences between the L&M model and PTR-ToF-MS data reflect the limitations of current analytical methods, which required Luh and Mariñas to perform experiments at conditions atypical of chloramination. They combined NH₂Cl (at a concentration more than threefold higher than the highest allowable chloramine concentration of 4 mg Cl₂/L in drinking water) with a fivefold excess of Br⁻ to form NHBrCl (example shown in Fig. S4). This approach resulted in a set of reactions that applies to their experimental conditions but does not translate to more representative chloramination conditions as exhibited in Fig. 4. PTR-ToF-MS can overcome the previous issues by providing more representative experimental data to develop a truly comprehensive haloamine kinetic model for guiding and optimizing disinfection practices when using chloramines. Additionally, PTR-ToF-MS can monitor the relevant haloamine species when NOM is present which is advantageous for studying haloamine reactions with NOM that lead to DBP formation (Fig. S5).

The use of PTR-ToF-MS to sample the headspace of an aqueous solution and directly correlate that to dissolved concentrations opens a wide range of new potential uses for this analytical technology. Many VOCs from anthropogenic sources (e.g. petroleum products, plastics, solvents, paints, adhesives, refrigerants) end up in ground, surface, and drinking waters ^32,33. Measurement of VOCs in water generally requires sample preparation (e.g., extractions, concentrating), separation by gas chromatography, and then quantification via mass spectrometry, flame ionization detection, or electron capture detection ³⁴. Headspace sampling utilizing PTR-ToF-MS has the potential to significantly simplify analysis of VOCs in water and increase sample throughput and improve sensitivity. Additional work is necessary to test the effectiveness of PTR-ToF-MS for the wide range of VOCs in water, but the work in this study demonstrated a proof of concept that rapid, sensitive headspace sampling can be used to directly determine aqueous concentrations of volatile analytes in real time.

The novel haloamine kinetic data collected in this work highlights a particularly useful application of PTR-ToF-MS because not only are aqueous concentrations being measured but reactions kinetics are being assessed in real-time. The data demonstrated major deficiencies in existing kinetic models, illustrating their limited usefulness in predicting disinfectant residual maintenance and DBP formation, and more broadly as tools for managing water distribution systems. Because PTR-ToF-MS allows simultaneous, real-time measurement of the five important haloamines, future experimental work and modeling will be able for the first time to evaluate kinetic models for all important species. Additionally, this method has the potential to also study the simultaneous formation of DBPs which would provide even greater insight into the formation mechanisms of DBPs. Such a study would allow revisions to existing kinetic models to close fundamental knowledge gaps, thereby enabling a better understanding of Br^– impacted waters and providing more robust predictive tools for managing drinking water treatment plants and distributions systems.

The research presented was not performed or funded by EPA and was not subject to EPA’s quality system requirements. The views expressed in this article are those of the author(s) and do not necessarily represent the views or the policies of the U.S. Environmental Protection Agency. Any mention of trade names, manufacturers or products does not imply an endorsement by the United States Government or the U.S. Environmental Protection Agency. EPA and its employees do not endorse any commercial products, services, or enterprises.

6.1 Synthesis of haloamines

Reagent grade chemicals and ultrapure water (18.2 MΩ·cm, Milli-Q, Millipore) were used to prepare all stock haloamine solutions and dilutions and wrapped in foil to protect against UV induced decay. The haloamines were made by combining hypochlorite (OCl⁻) or hypobromite (OBr⁻) with ammonia chloride (NH₄Cl) at various halogen to nitrogen ratios (X₂/N) and pH values. A 4.99% sodium hypochlorite solution (NaOCl) solution was used to make hypobromite ion (OBr⁻) solutions. The OCl⁻ concentration was determined before use by measuring the absorbance on a Hach DR6000 Spectrophotometer (Hach Company, Loveland, CO) at 292 nm, using a molar absorptivity (ε_OCl-,292nm) of 362 M^-1cm^-1 ³⁵. The OBr⁻ stock solution was prepared by combining OCl⁻ with Br⁻ at a Br⁻/Cl₂ molar ratio of 1.05. The exact OBr⁻ stock solution concentration was determined by monitoring the absorbance for OBr⁻ at 329 nm (ε_OBr-,329nm = 332 M^-1cm^-1)³⁶.

The NH₂Cl stock solution was prepared by adding OCl⁻ dropwise to a well-mixed pH 9 ammonia solution at a Cl₂/N molar ratio of 0.6. The NH₂Cl stock solution concentration was determined by measuring the absorbance at 243 nm (ε_NH2Cl,243nm = 461 M^-1cm^-1)³⁷. The NHCl₂ stock solution was prepared by buffering a NH₂Cl solution with 10 mM acetate and rapidly dropping the pH to 4. The NHCl₂ solution was aged for at least 4 hours to maximize formation of NHCl₂. The NHCl₂ and NH₂Cl concentrations were determined by reconciling overlapping UV spectra at 243 (ε_NH2Cl,243nm = 461 M^-1cm^-1 and ε_NHCl2,243nm = 235 M^-1cm^-1) and 294 nm (ε_NH2Cl,294nm = 15 M^-1cm^-1 and ε_NHCl2,294nm = 282 M^-1cm^-1).

The NH₂Br stock solution was prepared by combining ammonia and OBr⁻ solutions at a Br₂/N molar ratio of 1:1000 at pH 9 (the large excess of ammonia buffered the solution). The NHBr₂ stock solution was prepared by combining ammonia and OBr⁻ at a Br₂/N molar ratio of 1:2 at pH 7.2 in 10 mM phosphate buffer. The NH₂Br and NHBr₂ concentrations in each stock solution were determined by reconciling overlapping UV spectra at 232 (ε_NH2Br,232nm = 82 M^-1cm^-1 and ε_NHBr2,232nm = 2000 M^-1cm^-1) and 278 nm (ε_NH2Br,278nm = 425 M^-1cm^-1 and ε_NHBr2,278nm = 715 M^-1cm^-1)³⁸.

The NHBrCl stock solution was prepared by combining a NH₂Cl solution with HOBr at a NH₂Cl/HOBr molar ratio of 3:2 at pH 5 in 10 mM acetate buffer. At these conditions, NHBrCl will rapidly form and is sufficiently stable for analysis within minutes ³⁹. The NHBrCl concentration was equal to the HOBr added, and the NH₂Cl concentration was equal to half the NHBrCl concentration.

6.2 PTR-ToF-MS operating parameters and sampling procedure

A Vocus 2R PTR-ToF-MS (Aerodyne, Inc., Billerica, MA, USA) was used to measure haloamine concentrations. The internal parameters used in PTR-ToF-MS greatly impact ion sensitivities and detection transmission efficiencies ⁴⁰. The following optimized parameters were used: focused ion molecule reaction (FIMR) pressure = 2.3 mbar, FIMR temperature = 120 °C, big segmented quadrupole (BSQ) voltage = 275 V, H₃O⁺ ion source flow rate = 15 sccm, FIMR front voltage = 650 V, FIMR rear voltage = 25 V. These settings resulted in E/N = 155 Td, where E is the electrical field strength and N is the gas number density, which is sufficiently high to prevent excessive water clusters in high humidity samples. The PTR-ToF-MS parameters used in this study are commonly used in studies ^41,42 but were further optimized in the current work for NH₂Cl, NHCl₂, and NH₂Br signals by adjusting FIMR parameters including temperature.

Headspace sampling was used to measure haloamine concentrations in standards for calibration curves and samples from kinetic experiments. 5 mL of a solution was pipetted from a bulk standard or experiment into a 2 dram (7.4 mL) screw-top vial (Fisherbrand vial N51A, Fisher Scientific, Inc.), and a ⅛-inch Teflon line (length of ~20 cm) was connected to the Vocus inlet and held 1 cm above the liquid surface of the solution (it took approximately 30 seconds to transfer samples for measurement on the PTR-ToF-MS). The Teflon line was held in the vial until the Vocus signal stabilized, approximately 15-20 seconds. The PTR-ToF-MS was operated at 1 Hz time resolution and with sufficient flow pressure (using a LI-COR 850 CO₂ monitor as a pump downline) to result in <1 second response time. All standard headspace measurements were completed in at least triplicate except in the case of NHBr₂ which will be discussed later.

Data processing of the Vocus concentration data was done using PTRwid (version v_003_jul_01_2021) with the configuration file customized to cause all five haloamines and their halogen isotopes to be added to the unified mass list⁴³. The protonated m/z ratio (as measured by the Vocus) of the isotopes of each haloamine, volatile buffer, and acetone are shown in Table 1. The PTRwid data processing resulted in a concentration time series at 1 Hz time resolution which was further analyzed using MATLAB (Mathworks, Portola Valley, CA, USA). Headspace sampling was performed until a consistent mass spectrum was observed for 15-20 seconds.

6.3 Calibration curves

The NH₂Cl, NHCl₂, and NH₂Br stock solutions were used to make a series of dilutions to make standards for calibration or the mass spectrometer. The NHBrCl standards were made by combining NH₂Cl and HOBr for each individual standard. NHBr₂ is inherently unstable and rapidly decays so it was impossible to make a series of standards diluted from a stock solution. Therefore, the standard curve for NHBr₂ was made by simultaneously monitoring the decaying concentration of the NHBr₂ stock solution using UV-Vis spectroscopy and then immediately pipetting a sample for analysis. This procedure was then repeated every few minutes on the decaying NHBr₂ solution to obtain the data needed to make the NHBr₂ calibration curve.

6.4 Kinetic experiments

Two kinetic experiments, one without NOM and one with NOM, were performed to assess the effectiveness of using PTR-ToF-MS to measure haloamine mixtures undergoing formation and decay. To maximize brominated haloamine formation, prechlorination used Br⁻ and free chlorine concentrations at the upper range that occurs in drinking water treatment ⁴⁴ before ammonia addition to form haloamines. For the experiment without NOM, a solution of 2 mg/L as Br⁻ (25 μM) in 10 mM phosphate buffer at pH 7.2 was prepared and dosed with 4 mg/L as Cl₂ of free chlorine (56 μM). The solution was mixed for 1 minute, the median prechlorination duration⁴⁵, and then ammonia was added at a Cl₂/N molar ratio of 0.6 to form haloamines. The same procedure was repeated for the experiment with NOM except that 2 mg/L as C of Upper Mississippi River NOM (International Humic Substance Society, St Paul, MN, USA) was also present in solution before free chlorine addition. Ammonia addition initiated the start of a kinetic experiment because within milliseconds the monohaloamines, NH₂Cl and NH₂Br, form. Monitoring continued for 30 minutes using PTR-ToF-MS as well as the indophenol and total chlorine DPD colorimetric methods. Samples were continuously taken from the bulk sample and put in the screw-top vials to be analyzed by the PTR-ToF-MS and samples were taken at approximately 2, 10, 20, and 30 minutes to be analyzed using the colorimetric methods.

Rosario-ortiz, B. F., Rose, J., Speight, V., Gunten, U. Von & Scnoor, J. How do you like your tap water? Science 351, 912–914 (2016).
Sedlak, D. L. & von Gunten, U. The chlorine dilemma. Science 331, 42 LP – 43 (2011).
Allaire, M., Wu, H. & Lall, U. National trends in drinking water quality violations. Proceedings of the National Academy of Sciences 115, 1–6 (2018).
American Water Works Association. 2017 Water Utility Disinfection Survey Report. (2018).
World Health Organization. Monochloramine in drinking-water: background document for development of WHO guidelines for drinking-water quality. (2004).
Valentine, R. L. & Jafvert, C. T. Reaction scheme for the chlorination of ammoniacal water. Environmental Science and Technology 26, 577–586 (1992).
Duirk, S. E. & Valentine, R. L. Modeling dichloroacetic acid formation from the reaction of monochloramine with natural organic matter. Water Research 40, 2667–2674 (2006).
Amy, G., Siddigui, M., Zhai, W., DeBroux, J. & Odem, W. Survey of bromide in drinking water and impacts on DBP formation. (1994).
Allen, J. M. et al. Drivers of disinfection byproduct cytotoxicity in U.S. drinking water: should other DBPs be considered for regulation? Environmental Science & Technology 56, 392–402 (2022).
Dietrich, A. M. & Burlingame, G. A. Critical review and rethinking of USEPA secondary standards for maintaining organoleptic quality of drinking water. Environmental Science & Technology 49, 708–720 (2015).
Vidic, R. D., Brantley, S. L., Vandenbossche, J. M., Yoxtheimer, D. & Abad, J. D. Impact of shale gas development on regional water quality impact of shale gas development. Science 340, (2013).
Kirmeyer, G., Martel, K., Thompson, G. & Radder, L. Optimizing Chloramine Treatment. (2004).
Wahman, D. G., Speitel, G. E. & Katz, L. E. Bromamine decomposition revisited: a holistic approach for analyzing acid and base catalysis kinetics. Environmental Science & Technology 51, 13205–13215 (2017).
Trofe, T. W., Inman, G. W. & Donald Johnson, J. Kinetics of monochloramine decomposition in the presence of bromide. Environmental Science and Technology 14, 544–549 (1980).
Pope, P. G. & Speitel, G. E. Reactivity of bromine-substituted haloamines in forming haloacetic acids. ACS Symposium Series 995, 182–197 (2008).
USEPA. Fact Sheet: Aouncement of Completion of EPA’s Third Six-Year Review of Existing Drinking Water Standards (EPA 815-F-16-010). (2016).
Brodfuehrer, S. H., Wahman, D. G., Alsulaili, A., Speitel, G. E. & Katz, L. E. Role of carbonate species on general acid catalysis of bromide oxidation by hypochlorous acid (HOCl) and oxidation by molecular chlorine (Cl2). Environmental Science & Technology 54, 16186–16194 (2020).
Luh, J. & Mariñas, B. J. Kinetics of bromochloramine formation and decomposition. Environmental Science and Technology 48, 2843–2852 (2014).
Hach Company. Method 10020, Chloramine ( Mono ) and Nitrogen , Free Ammonia. (2019).
Hach Company. Method 10070, Chlorine, Free and Total, High Range. (2018).
Pope, P. G. Haloacetic acid formation during chloramination : role of environmental conditions , kinetics , and haloamine chemistry. (University of Texas at Austin, 2006).
Hu, W., Lauritsen, F. R. & Allard, S. Identification and quantification of chloramines, bromamines and bromochloramine by membrane introduction mass spectrometry (MIMS). Science of the Total Environment 751, (2021).
Blake, R. S., Monks, P. S. & Ellis, A. M. Proton-transfer reaction mass spectrometry. Chemical Reviews 109, 861–896 (2009).
Thompson, J. M. Introduction. in Mass Spectrometry 1–56 (Pan Stanford Publishing Pte. Ltd., 2018).
Roumiguières, A., Bouchonnet, S. & Kinani, S. Challenges and opportunities for on-line monitoring of chlorine-produced oxidants in seawater using portable membrane-introduction Fourier transform-ion cyclotron resonance mass spectrometry. Analytical and Bioanalytical Chemistry 413, 885–900 (2021).
Wu, T. et al. Real-time measurements of gas-phase trichloramine (NCl3) in an indoor aquatic center. Environmental Science & Technology 55, 8097–8107 (2021).
Arata, C. et al. Volatile organic compound emissions during HOMEChem. Indoor Air 31, 2099–2117 (2021).
Senthilmohan, S. T. et al. Detection of monobromamine, monochloramine and dichloramine using selected ion flow tube mass spectrometry and their relevance as breath markers. Rapid Communications in Mass Spectrometry 22, 677–681 (2008).
Pope, P. G. Haloacetic acid formation during chloramination : role of environmental conditions , kinetics , and haloamine chemistry. (University of Texas at Austin, 2006).
Hu, W., Lauritsen, F. R. & Allard, S. Identification and quantification of chloramines, bromamines and bromochloramine by membrane introduction mass spectrometry (MIMS). Science of the Total Environment 751, (2021).
Shrivastava, A. & Gupta, V. B. Methods for the determination of limit of detection and limit of quantitation of the analytical methods. Chronicles of Young Scientists 2, 21–25 (2011).
DeSimone, L. A. & Peter B. McMahon, M. R. R. The quality of our Nation’s waters: Water quality in principal aquifers of the United States, 1991-2010. (2015).
Zogorski, J. S. et al. The quality of our nations’s waters: volatile organic compounds in the nation’s ground water and drinking-water supply wells. (2006).
Demeestere, K., Dewulf, J., Witte, B. De & Langenhove, H. Van. Sample preparation for the analysis of volatile organic compounds in air and water matrices. Journal of Chromatography A 1153, 130–144 (2007).
Furman, C. S. & Margerum, D. W. Mechanism of chlorine dioxide and chlorate ion formation from the reaction of hypobromous acid and chlorite ion. Inorg Chem 37, 4321–4327 (1998).
Troy, R. C. & Margerum, D. W. Non-metal redox kinetics: hypobromite and hypobromous acid reactions with iodide and with sulfite and the hydrolysis of bromosulfate. Inorg Chem 30, 3538–3543 (1991).
Kumar, K., Day, R. A. & Margerum, D. W. Atom-transfer redox kinetics: general-acid-assisted oxidation of iodide by chloramines and hypochlorite. Inorg Chem 25, 4344–4350 (1986).
Lei, H., Mariñas, B. J. & Minear, R. A. Bromamine decomposition kinetics in aqueous solutions. Environ Sci Technol 38, 2111–2119 (2004).
Brodfuehrer, S. H., Goodman, J. B., Wahman, D. G., Speitel, G. E. & Katz, L. E. Apparent reactivity of bromine in bromochloramine depends on synthesis method: implication bromine chloride and molecular bromine as important bromine species. Journal of Environmental Engineering 148, (2022).
Blomdahl, D. C. Optimization of volatile organic compound sensitivity of Vocus PTR-ToF-MS for quantifying disinfection byproducts. (University of Texas at Austin, 2021).
Hall, E. C. et al. Varying humidity increases emission of volatile nitrogen-containing compounds from building materials. Build Environ 205, 108290 (2021).
Sreeram, A., Blomdahl, D., Misztal, P. & Bhasin, A. High resolution chemical fingerprinting and real-time oxidation dynamics of asphalt binders using Vocus Proton Transfer Reaction (PTR-TOF) mass spectrometry. Fuel 320, 123840 (2022).
Holzinger, R. PTRwid: A new widget tool for processing PTR-TOF-MS data. Atmos Meas Tech 8, 3903–3922 (2015).
Amy, G., Siddigui, M., Zhai, W., DeBroux, J. & Odem, W. Survey of bromide in drinking water and impacts on DBP formation. (1994).
AWWA Disinfection Systems Committee. Committee Report : Disinfection Survey , Part 2 - Alternatives , experiences , and future plans. American Water Works Association 100, 110–124 (2008).

There is NO Competing Interest.

SupplementaryMaterial.docx

Download PDF

Journal Publication

published 23 Apr, 2024

Read the published version in Nature Water →

Version 1

posted

You are reading this latest preprint version

Simultaneous time-resolved aqueous haloamine measurement empowers robust kinetic model analysis

Status:

Journal Publication

Version 1

Abstract

Figures

1. Introduction

2. Main Text

2.1 Mass spectra of haloamines using PTR-ToF-MS

2.2 Calibration curves and detection limits for haloamines using PTR-ToF-MS

2.3 Impact of dihaloamine fragments on quantification of monohaloamines

2.4 Effectiveness of PTR-ToF-MS at monitoring dynamic aqueous haloamine kinetics

2.5 Implications for existing haloamine models

3. Conclusion

4. Disclaimer

Methods

6.1 Synthesis of haloamines

6.2 PTR-ToF-MS operating parameters and sampling procedure

6.3 Calibration curves

6.4 Kinetic experiments

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1