Environ. Sci. Technol. 2008, 42, 4811–4817
Degradation of Tertiary Alkylamines during Chlorination/Chloramination: Implications for Formation of Aldehydes, Nitriles, Halonitroalkanes, and Nitrosamines WILLIAM A. MITCH* AND I. MARIE SCHREIBER Department of Chemical Engineering; Yale University; Mason Lab 313b; 9 Hillhouse Avenue; New Haven, Connecticut 06520
Received December 3, 2007. Revised manuscript received March 20, 2008. Accepted March 26, 2008.
Drinking water utilities are exploring the use of waters impacted by wastewater effluents and agricultural runoff to meet the demands of growing populations. Due to the elevated organic nitrogen concentrations in these waters, the pathways responsible for transformation of organic nitrogen into toxic nitrogenous disinfection byproducts during chlorine and chloraminedisinfectionareofcurrentconcern.Tertiaryalkylamines are important functional groups in human waste products and various consumer products that can be released to drinking water supplies via wastewater effluents. We investigated degradation pathways for model tertiary alkylamines during chlorination and chloramination. Our results indicate that tertiary alkylamines degrade nearly instantaneously during chlorination to form aldehydes and secondary alkylamines quantitatively, with no significant regioselectivity. Similar results were observed during chloramination, but the observed degradation rates were much slower, with lower yields of aldehydes. As these major products were fairly stable, these results explain why tertiary amines are significant precursors of secondary nitrosamines during chloramination. Trichloronitromethane formed at very low yields during chlorination, but was not observed during chloramination; monochloronitromethane and dichloronitromethane were never detected. Despite the significant yields of aldehydes during chloramination, our results indicated low nitrile yields by the reaction between chloramines and aldehydes.
Introduction To accommodate population growth, drinking water utilities are looking beyond pristine water supplies to exploit source waters impaired by wastewater effluents and algal blooms fostered by agricultural runoff (1). Pristine water supplies feature natural organic matter (NOM) characterized predominantly by plant biopolymers (e.g., lignins) that have degraded to complex fulvic and humic acids over timescales of months to years. In impaired waters, algal organic matter (AOM), consisting of algal exudates, and wastewater effluent organic matter (EfOM) overlay the traditional NOM. Because EfOM and AOM have had less time to degrade (e.g., hours for wastewater treatment), they are anticipated to more * Corresponding author phone: (203) 432-4386; fax (203) 432-4387; e-mail:
[email protected]. 10.1021/es703017z CCC: $40.75
Published on Web 05/23/2008
2008 American Chemical Society
closely resemble biomolecules. In addition, such waters often feature higher dissolved organic nitrogen (DON) concentrations (2). Because the biomass-derived organic matter in algal- or wastewater-impacted waters has been subjected to less processing than the NOM in pristine waters, we expect important nitrogenous biochemicals to serve as significant DON constituents. Organic nitrogen moieties in the biochemicals serving as sources of AOM and EfOM are dominated by amine (N(R)3; e.g., lysine, adenine), or amide (N(R)2(CdO)R; e.g., peptide bonds, cytosine) functionalities. Amine and amide functional groups are also important in many industrial or consumer compounds likely to enter impaired waters via wastewater treatment plants (e.g., the pharmaceutical ranitidine) or agricultural runoff (e.g., the pesticide diuron). Because the electron-withdrawing nature of the amide carbonyl deactivates the amide nitrogen toward electrophilic substitution, amines are much more reactive with chlorine than amides (3), and serve as the focus of our research. The chlorine or chloramine-initiated decay of amine moieties within DON is particularly important because their degradation may form toxic nitrogenous disinfection byproducts (N-DBPs) featuring nitroso (-NO; e.g., nitrosamines), nitro (-NO2; e.g., halonitromethanes) and nitrile (-CN; e.g., cyanogen chloride and haloacetonitriles) functional groups. Drinking water concentrations resulting in a 10-6 lifetime cancer risk for N-nitrosodimethylamine (NDMA) are more than 3 orders of magnitude lower than those for the regulated trihalomethanes (THMs), such as bromoform (5). Mammalian cell geno- and cytotoxicity assays indicated that halonitromethanes and haloacetonitriles achieved similar levels of toxicity to the regulated THMs at roughly 3 orders of magnitude lower concentrations (5, 6). Given the variety of amine structures in organic matter, there is a need for a general understanding of how amines degrade in the presence of chlorine or chloramine disinfectants, and the importance of N-DBP formation. In addition, as with nitrosamines (7–9), a greater understanding of formation pathways may enable the design of disinfection systems that minimize N-DBP formation. Finally, because utilities are increasingly adopting chloramination to reduce the formation of THMs and haloacetic acids, it is important to evaluate whether the use of an inorganic nitrogen-based disinfectant enhances toxic N-DBP formation. Previously, we examined pathways by which two model primary alkylamines degraded during chlorination and chloramination (10): monomethylamine as a model precursor of cyanogen chloride and chloropicrin, two N-DBPs of current concern, and n-propylamine as a model for longer chain primary alkylamines. We found that the nitrogen atoms of model primary alkylamines were relatively rapidly dichlorinated by free chlorine or inorganic chloramines. The dichlorinated monomethylamine and n-propylamine subsequently degraded with half-lives of 12 and 3 days, respectively, on the timescales of drinking water distribution systems. We examined three degradation pathways for dichlorinated primary amines: elimination of two hydrochloric acids to form nitriles (the dehydrohalogenation pathway), elimination of one hydrochloric acid followed by hydrolysis of the chlorinated imine intermediate to yield an aldehyde (the hydrolysis pathway), and oxidation and subsequent chlorine addition to the R-carbon to form halonitroalkanes (the oxidation pathway). For both monomethylamine and n-propylamine, the oxidation pathway was relatively minor, featuring yields of ∼0.008% with application VOL. 42, NO. 13, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
4811
SCHEME 1
of free chlorine but an order of magnitude higher with inorganic chloramines. The hydrolysis and dehydrohalogenation pathways were more important, particularly for npropylamine, where yields of propionaldehyde and propionitrile reached 20 and 30%, respectively. Tertiary alkylamines may be important N-DBP precursors. NDMA precursors are particularly associated with EfOM (11–14). While the ideal model precursor for NDMA formation, dimethylamine, occurs in urine (15), we previously demonstrated that it was rapidly removed during secondary wastewater treatment and was not an important precursor in municipal wastewater effluents (13). Trimethylamine also occurs in human urine (15). We demonstrated that tertiary alkylamines with dimethylamine functional groups, such as trimethylamine, could serve as significant NDMA precursors during chloramination (13). Secondary nitrosamine formation from tertiary alkylamine precursors implies degradation by secondary alkylamine intermediates. Further degradation by similar pathways might yield primary alkylamine precursors of nitriles and halonitroalkanes. Although Ellis and Soper (16) had suggested that chlorination of trimethylamine forms dimethylamine and formaldehyde via dehydrohalogenation and subsequent hydrolysis of a chlorinated trimethylamine intermediate (Scheme 1), most of their experiments were conducted at low pH (i.e., 2-3) with an excess of tertiary amines. The reaction timescales, the tendency for specific alkyl functional groups to be preferentially removed (i.e., the regioselectivity) and product yields of this pathway under environmentally relevant conditions were not clear. Moreover, formation of N-DBP products and the application of inorganic chloramines were not evaluated. The purpose of this study was to understand how tertiary alkylamines degrade during chlorination and chloramination, and to assess the importance of tertiary alkylamines as N-DBP precursors. First, we examined whether trimethylamine was an important NDMA precursor in municipal wastewater effluents. Next, we characterized the time scale and major products of the degradation of tertiary alkylamines during both chloramination and chlorination. We also evaluated the potential for tertiary alkylamines to form nitrile and halonitroalkane N-DBPs. We used these data to evaluate whether altering disinfection systems to chloramination will enhance N-DBP formation when applied to tertiary alkylamine-containing waters.
Materials and Methods Materials. Cambridge Isotope Laboratories d3-monomethylamine (98%), d6-dimethylamine (98%), d9-trimethylamine (98%), and 15N-ammonium chloride (98%), Protocol Analytical trichloronitromethane (1 g/L), Orchid Helix Biotech chloronitromethane (90-95%) and dichloronitromethane (95%), Chem Service chloroacetonitrile, dichloroacetonitrile, and trichloroacetonitrile (all 100 µg/mL), and Acros trimethylamine hydrochloride (98%), triethylamine hydrochloride (99+%), and diethylmethylamine (98%) were used as received. Carboxen-PDMS (75 µm) solid phase micro extraction (SPME) fibers were obtained from Supelco. All other reagents were reagent grade or described previously (10). Most experiments were conducted in deionized water (18 MΩ) produced with a Millipore Elix 10/Gradient A10 water purification system. All glassware was baked for 3 h at 400 °C in a muffle furnace to remove organic contaminants prior to use. 4812
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 13, 2008
Model Tertiary Amine Decay Experiments. Preparation of stock oxidant solutions was described previously (7). Most experiments were conducted at room temperature (22 °C) under headspace-free conditions in 25 mL glass screw-cap vials with PTFE-lined septa. A typical experiment involved the syringe injection of a tertiary amine or formaldehyde through the septa into 25 mL phosphate-buffered deionized water containing appropriate concentrations of chlorine or chloramines. Residual oxidants in sample aliquots were quenched as follows. For the analysis of amines, cyanogen chloride, and the chlorinated analogues of haloacetonitriles and halonitromethanes, oxidants in samples were quenched with 1 mM of ascorbic acid and potassium iodide; iodide functions as a catalyst for organic chloramine dechlorination. For cyanogen chloride, haloacetonitrile and halonitromethane analyses, glacial acetic acid was injected to lower the pH to 3.5-4.0, where these compounds are stable (17). For the analysis of aldehydes, oxidants in samples were quenched with 1 mM sodium thiosulfate and potassium iodide. For amine analyses, samples were transferred to 40 mL glass vials and injected with d3-monomethylamine, d6dimethylamine, and d9-trimethylamine as internal standards for primary, secondary, and tertiary amines, respectively. The pH was adjusted to 12 using sodium hydroxide, a Teflonlined magnetic stir bar was added and the vial was capped with a Teflon-lined septum. While stirring, the gas-phase headspace was extracted with a syringe-mounted CarboxenPDMS solid phase microextraction (SPME) fiber injected through the septum for 20 min at room temperature. Samples were analyzed by gas chromatography (Varian CP-Volamine 60 m x 0.32 mm) mass spectrometry (Varian 2000 system) using methanol chemical ionization (GC/MS/CI). The column temperature program was 50 °C held for 5 min, and ramping to 160 at 25 °C/min held for 10 min. Injection port and detector temperatures were 220 and 170 °C, respectively. The SPME fiber was desorbed in the injection port for 3 min. For cyanogen chloride, haloacetonitrile and halonitromethane analyses, most samples were transferred to 40 mL glass vials and extracted by shaking for 10 min into 4 mL MtBE. Samples were analyzed by gas chromatography (Alltech AT-1701 30 m × 0.25 mm × 1 µm column) with electron capture detection (GC-ECD) against authentic standards of haloacetonitriles and halonitromethanes or standards of cyanogen chloride made freshly in 25 mL deionized water by mixing sodium hypochlorite with sodium cyanide in a 1:20 molar ratio. The column temperature program was 40 °C held for 15 min, and ramping to 120 °C at 20 °C/min held for 4 min. Injection port and detector temperatures were 220 and 280 °C, respectively. For samples treated with 15N-labeled chloramines, cyanogen chloride was analyzed using the same technique used for amines, except that no internal standard was used, the pH was not adjusted during extraction and electron ionization mass spectrometry was used for detection. Aldehydes were analyzed by HPLC with UV detection (365 nm) following derivatization with 2,4-dinitrophenylhydrazine (18). Wastewater Sample Experiments. Grab samples were collected in fluorinated containers from the effluents of the primary and secondary treatment units at the Town of Wallingford wastewater treatment plant (CT) and the secondary treatment unit at the City of Norwalk wastewater treatment plant (CT) prior to application of disinfectants. Composite samples were unavailable in the volumes needed
TABLE 1. Summary of Experimental Conditions amine type trimethylamine
oxidant µM
50 50 50 50 50 50 40 40 40 40 triethylamine 50 diethylmethylamine 50
FIGURE 1. NDMA formation potential vs added trimethylamine (TMA) concentrations in a sample obtained from the Wallingford municipal wastewater treatment plant spiked with various concentrations of TMA. Error bars represent one standard deviation of experimental replicates (n ) 2). to conduct the experiments. Trimethylamine was analyzed as described above. NDMA was analyzed by isotope dilution analysis as described previously (11). Briefly, after injection of d6-NDMA, 1 L samples were extracted with 0.4 g Ambersorb 572 resin beads for 3 h. The beads were filtered, air-dried, and extracted into 5 mL of methylene chloride. The methylene chloride extracts were blown down to 1 mL using nitrogen gas, and analyzed by gas chromatography tandem mass spectrometry using methanol chemical ionization. NDMA precursor concentrations were evaluated after application of monochloramine (2 mM) for 10 days at pH 6.9; the concentration of NDMA formed under these extreme conditions was used as a surrogate for NDMA precursors (19).
Results Importance of Trimethylamine as a NDMA Precursor in Municipal Wastewater Effluents. The trimethylamine concentration in the primary wastewater effluent sample from the Wallingford plant was 900 nM. However, the concentration in the Wallingford and Norwalk secondary effluents were below the 200 nM method detection limit. To estimate the contribution of trimethylamine as a NDMA precursor, we measured NDMA precursor concentrations after spiking concentrations of trimethylamine (125-1000 nM) into aliquots of the Wallingford secondary effluent. After plotting precursor concentrations against total added trimethylamine concentrations (Figure 1), the maximum contribution of trimethylamine to NDMA formation at the two wastewater plants can be estimated using the slope of the regression line (720 ng/L/µM trimethylamine) in Figure 1. Since the unspiked samples could have contained a maximum of 0.2 µM trimethylamine (the method detection limit), the maximum contribution of trimethylamine to NDMA formation in the unspiked samples would be 144 ng/L. The unspiked secondary effluent samples from the Wallingford and Norwalk treatment plants formed 323 and 755 ng/L NDMA during the precursor analysis. Therefore, the maximum contribution of trimethylamine to NDMA formation at the two plants would be 30% at Wallingford and 19% at Norwalk. Accordingly, although trimethylamine is a constituent of urine (15), we must examine other compounds, including tertiary amines with dimethylamine functional groups (13), to account for NDMA formation. Decay of Trimethylamine during Chlorination. Because Cl[+1] transfer from free chlorine to amines is fast, chlorinated trimethylamine would rapidly form (16). However,
type HOCl HOCl HOCl HOCl HOCl HOCl NHCl2 NHCl2 NH2Cl NH2Cl HOCl HOCl
Cl:amine
amine
µM mole ratio pH
half-life
50 100 150 200 200 200 200 200 400 400 200 200
1 2 3 4 4 4 10 10 10 10 4 4
7 7 7 7 5 9 5 7 7 9 9 9
