Chloronitramide Anion Quantitation in Tap Waters by Ion Chromatography with Electrical Conductivity and Ultraviolet Absorbance Detection

After ∼45 years since it was first detected in chloraminated water (1) and more than 30 years since its prior characterization efforts, (2−5) chloronitramide anion (Cl–N–NO₂^–) was identified as a decomposition product of inorganic chloramines. (6) In a limited survey of chloraminated tap waters in the United States, Cl–N–NO₂^– was found in all samples tested (n = 40) at a median concentration of 23 μg L^–1 and first and third quartiles of 1.3 and 92 μg L^–1, respectively. Although its toxicity is currently unknown, Cl–N–NO₂^– forms at levels comparable to some regulated disinfection byproducts (DBPs) but lower than others. For example, the Environmental Protection Agency (EPA) maximum contaminant levels (MCLs) for the mass-based sums of four trihalomethanes is 80 μg L^–1 and five haloacetic acids is 60 μg L^–1; (7) however, the chlorite MCL is over 1 order of magnitude greater, at 1 mg L^–1. The Cl–N–NO₂^– formation pathway contrasts that of most other DBPs, which are formed by reactions between a disinfectant and source water precursors, such as natural organic matter (NOM) and/or bromide. (8,9) Cl–N–NO₂^– is formed by monochloramine (NH₂Cl) or dichloramine (NHCl₂) reacting with nitronium cation (NO₂⁺), (6) a reactive nitrogen species formed during chloramine decomposition. (10,11) In this regard, Cl–N–NO₂^– is an intrinsic DBP, (12) like chlorate and perchlorate in hypochlorite feedstocks. (13−16) As chloramines are inherently unstable and autodecompose under typical conditions in drinking water systems, (17−19) Cl–N–NO₂^– is likely present in all waters where chloramines are used. This includes systems where chloramines are formed intentionally and free chlorine systems containing source water ammonia. (20) To facilitate Cl–N–NO₂^– occurrence studies and better understand its formation in tap water, analytical methods are needed for Cl–N–NO₂^– quantitation that leverage common instrumentation in drinking water laboratories.

The discovery of Cl–N–NO₂^– was supported by its quantitation in chloraminated tap waters by hydrophilic interaction liquid chromatography–ultrahigh-resolution mass spectrometry (HILIC–UHRMS). (6) This method has a limit of detection (LOD) of 0.17 μg L^–1 and a limit of quantitation (LOQ) of 0.58 μg L^–1. However, the instrumentation required for HILIC–UHRMS (i.e., orbitrap mass spectrometers) is not widely accessible due to its high capital cost and operation and maintenance requirements, limiting collection of Cl–N–NO₂^– occurrence data. Cl–N–NO₂^– reference materials were previously isolated from mixtures containing common anions (e.g., chloride, nitrite, nitrate, sulfate, phosphate) at concentrations up to ∼100 mM using ion chromatography (IC) with electrical conductivity (EC) and ultraviolet absorbance (UV) detectors. (6) Given Cl–N–NO₂^– is an anion and has a molar absorptivity (ε) peak maxima at 243 nm of ε₂₄₃ = 5,310 M^–1 cm^–1, (6) Cl–N–NO₂^– quantitation may be possible by IC-EC and/or IC-UV₂₄₃, but the applicability of these methods at μg L^–1 levels is unknown. IC methods could be a cost-effective alternative to MS-based methods and adopted at drinking water utilities and commercial water analysis laboratories, enabling widescale Cl–N–NO₂^– occurrence studies. The objective of this study is to assess IC-EC and IC-UV₂₄₃ for Cl–N–NO₂^– quantitation in tap waters at μg L^–1 levels. Cl–N–NO₂^– quantitation by IC-EC and IC-UV₂₄₃ was compared to HILIC–UHRMS across an unprecedented range of Cl–N–NO₂^– concentrations in tap waters from U.S. residential homes.

2. Materials and Methods

2.1. Ion Chromatography System

A Metrohm 850 Professional IC equipped with an autosampler, suppressor, two electrical conductivity detectors, and an 887 Professional UV/vis detector was used to isolate Cl–N–NO₂^– standard reference materials and quantify Cl–N–NO₂^– in tap waters. Two Metrosep A Supp 7 separation columns were used, each 250 mm in length with a 4.0 mm diameter, one for Cl–N–NO₂^– isolation that had been used extensively (>10,000 injections) and regenerated to maintain adequate anion separation and the second newly acquired for Cl–N–NO₂^– quantitation in drinking water matrices. Both IC columns consist of a PEEK housing packed with a poly(vinyl alcohol) carrier material with quaternary ammonium groups, 5 μm particle size, 15 MPa maximum pressure, 1.0 mL min^–1 maximum flow rate, pH range 3–12, and temperature range of 20–60 °C. Na₂CO₃ eluent concentrate (Metrohm) was diluted 100× to 3.6 mM for Cl–N–NO₂^– isolation and 25× to 14.4 mM for Cl–N–NO₂^– quantitation.

3. Results and Discussion

3.4. Tap Water Comparison of IC and HILIC–UHRMS

Figure 2 shows Cl–N–NO₂^– concentrations in the 12 tap waters measured by IC-EC and IC-UV₂₄₃ peak area compared to the HILIC–UHRMS method developed in prior work. (6) Figure s30 shows these data for peak height.

Environ. Sci. Technol. Lett. 2026, 13, 2, 275–280: Figure 2. Ion chromatography-electrical conductivity (IC-EC, Panels A and B) and -ultraviolet absorbance at 243 nm (IC-UV243, Panels C and D) comparison testing with HILIC–UHRMS for Cl–N–NO2– quantitation in 12 waters which included eight chloraminated tap waters (TX-1, PA, TX-2, SC, CA, TX-3, OK, and MN), three chlorinated tap waters (AR, GA-1, and GA-2), and one synthetic groundwater (GW) with no added disinfectant. (A) IC-EC measured by peak area vs Cl–N–NO2– determined by HILIC–UHRMS, (B) IC-EC residuals by peak area, (C) IC-UV243 measured by peak area vs Cl–N–NO2– determined by HILIC–UHRMS, and (D) IC-UV243 residuals by peak area. The solid black line in Panels A and C is the 1:1 line and shown with the correlation coefficient, R2. The average residual ( 𝑟𝑖⎯⎯⎯ ; n = 12) in μg L–1 is shown in Panels B and D as a measure of the linear regression model bias and accuracy with values closer to zero indicating lesser overall bias.

Cl–N–NO₂^– quantified using the IC methods tracked the 1:1 line across the range of Cl–N–NO₂^– concentrations in the tap waters from not detected to 315 μg L^–1, indicating they were strongly correlated with the HILIC–UHRMS measurements. For IC-EC, the R² (n = 12) was 0.991 (Figure 2A) and the average residual (𝑟𝑖⎯⎯⎯) was +3.2 μg L^–1, indicating an overall positive bias. For IC-UV₂₄₃, the R² (n = 12) was 0.997 and the 𝑟𝑖⎯⎯⎯ was 0.0 μg L^–1, indicating no overall bias. Therefore, the accuracy and precision of Cl–N–NO₂^– quantitation by IC-UV₂₄₃ exceeded that of IC-EC for the tap waters tested. This is the first report, to the best of our knowledge, of Cl–N–NO₂^– presence in chlorinated tap waters. Although the Cl–N–NO₂^– concentrations are lower compared to chloraminated tap waters, this finding suggests that Cl–N–NO₂^– exposure is more widespread and additional formation pathways such as breakpoint chlorination could be relevant, as previously speculated. (6) The Cl–N–NO₂^– concentration maximum observed in the chloraminated tap waters (315 μg L^–1) is greater by a factor of about three compared to that in prior work, (6) illustrating a pressing need for additional occurrence studies to capture its full range in water systems.

We demonstrated that IC-EC and IC-UV₂₄₃ can be used to accurately quantify Cl–N–NO₂^– in tap waters with minor modifications to the IC system (i.e., 5 mL injection loop and 14.4 mM Na₂CO₃ eluent) using standards formulated in 1 mM borate buffer at pH 9. While the lowest LOD (5.8 μg L^–1) and LOQ (8.2 μg L^–1) are greater than those for HILIC-UHRMS (0.17 and 0.58 μg L^–1, respectively), the IC methods are accurate within an environmentally relevant range, with Cl–N–NO₂^– at levels up to about 315 μg L^–1. Spike recovery experiments indicate matrix effects were negligible for Cl–N–NO₂^– quantitation by IC-EC and IC-UV₂₄₃ for concentrations <50 μg L^–1 (i.e., P_{<50 μg L^–1} > 0.05). IC-based quantitation methods were strongly correlated with HILIC–UHRMS (R² ≥ 0.991) and unbiased for quantitation by IC-UV₂₄₃ peak area. Cl–N–NO₂^– may continue to form in tap waters containing a chloramine residual, particularly at pH ≲ 8 where chloramines decompose more readily. (29) A quenching agent/procedure has not yet been identified for chloramines that does not impact Cl–N–NO₂^–. Thus, the time between sample collection and analysis should be minimized (<1 day, if possible) to get the most accurate assessment of Cl–N–NO₂^– occurrence until if/when a suitable sample preservation procedure is developed. Nevertheless, the IC methods developed and validated here can be adopted by water utilities and testing laboratories for widespread Cl–N–NO₂^– monitoring campaigns.