Environ. Sci. Technol. 2010, 44, 700–706
Influence of the Method of Reagent Addition on Dichloroacetonitrile Formation during Chloramination ELEANOR L. HAYES-LARSON AND WILLIAM A. MITCH* Department of Chemical Engineering, Environmental Engineering Program, Yale University, Mason Lab 313b, 9 Hillhouse Avenue, New Haven, Connecticut 06520
Received August 17, 2009. Revised manuscript received October 19, 2009. Accepted November 30, 2009.
Formation of dichloroacetonitrile (DCAN) in natural waters was evaluated for different disinfection scenarios, including application of free chlorine, preformed monochloramine or dichloramine, or formation of chloramines in situ by addition of free chlorine and ammonia in either order. Formation of DCAN was highest during free chlorination. Regardless of the order of ammonia or chlorine addition, DCAN formation was consistently higher over 1-2 day contact times when chloramines were formed in situ than when preformed chloramines were applied. During in situ chloramine formation, organic amine precursors effectively competed with ammonia to react with free chlorine, forming organic dichloramine intermediates to nitrile formation. Combined with previous research indicating that application of preformed monochloramine reduced nitrosamine formation, the results indicate that application of preformed monochloramine could provide an inexpensive alternative for chloraminating utilities to significantly reduce the exposure to DCAN and nitrosamines for consumers located within 1-2 days of water travel time from the treatment plant. This technique would be applicable in situations where chloramination is used alone (e.g., chlorination of non-nitrified secondary wastewater effluents during municipal wastewater recycling), or combined with primary disinfectants other than free chlorine (e.g., ozonation, chlorine dioxide, or ultraviolet light).
1. Introduction To satisfy growing populations, utilities are exploiting water supplies impaired by municipal wastewater effluents or algal blooms promoted by nutrient-rich agricultural runoff. These waters feature elevated concentrations of dissolved organic nitrogen (DON) (1). There is growing concern regarding the potential of DON components to serve as precursors for nitrogenous disinfection byproducts (N-DBPs). Categories of N-DBPs that have raised particular concern due to their toxicity include nitrosamines, halonitriles (e.g., cyanogen chloride and haloacetonitriles), and halonitromethanes (2-4). For example, dichloroacetonitrile (DCAN) was ∼300 times more cytotoxic than its haloacetic acid analogue, dichloroacetic acid, in mammalian cell assays (4). Previous studies with model compounds have demonstrated that nitrile formation can occur during the chlorination or chloramination of primary amine moieties via the * Corresponding author: telephone (203) 432-4386; fax (203) 4324387; e-mail
[email protected]. 700
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 2, 2010
formation of organic dichloramine intermediates. As rate constants for chlorine transfer to organic amines are orders of magnitude higher for free chlorine than for monochloramine (e.g., 6.1 × 107 M-1 s-1 (5) vs 0.21 M-1 s-1 (6) for reaction with dimethylamine), DCAN formation rates generally are higher for free chlorination than chloramination. In the case of free amino acids, where the R-amino moiety is proximal to a carboxylic acid, concerted decarboxylation of the organic dichloramine intermediate (i.e., decarboxylation combined with chloride loss), coupled with hydrochloric acid elimination forms nitriles (7-16). With free aspartic acid, subsequent chlorination of the carbon adjacent to the nitrile results in DCAN formation (Scheme 1 (11)). Where the model primary amine is not proximal to a carboxylic acid group, elimination of two hydrochloric acids from the organic dichloramine intermediate forms a nitrile, but nitrile formation is much slower than observed for free amino acids (i.e., half-lives of days instead of hours during chlorination) (17). Many utilities are exploring altering their disinfection schemes to include chloramination to reduce the formation of regulated trihalomethanes and haloacetic acids. The influence of different chloramination disinfection schemes on nitrile formation has not been fully explored. Previous research with glycine as a model organic precursor implied that addition of preformed monochloramine might reduce cyanogen chloride formation compared with addition of free chlorine to ammonia-containing solutions, but that the effect might not be important in natural organic matter models for natural waters. Applying the two chloramination techniques to solutions containing glycine within an Aldrich humic acid matrix, one study found that addition of free chlorine to ammonia-containing solutions produced ∼40% more cyanogen chloride, one of the simplest nitriles, after 24 h than did addition of preformed monochloramine; however, no significant formation was observed from the Aldrich humic acid alone (18). The enhanced formation was attributed to effective competition by glycine with ammonia for reaction with free chlorine to form an organic chloramine intermediate. A later study by the same group found no significant differences in cyanogen chloride or DCAN formation after 3 d for different methods of chloramine application to Suwannee River Natural Organic Matter (NOM)-containing solutions (19). Other research suggested that dichloramine might play a role in promoting nitrile formation. Several studies evaluated formation of cyanogen chloride during chloramination by adding chlorine to solutions containing ammonia and humic acids (20), amino acids (21), and proteins (22) as model organic precursors. In all cases, cyanogen chloride formation increased as the chlorine to ammonia molar ratio approached the 1.5:1 molar ratio characteristic of breakpoint chlorination, but declined at higher molar ratios. Explanations were not provided for this behavior. However, since dichloramine formation is maximized near this 1.5:1 molar ratio (23, 24), the cyanogen chloride results suggested that dichloramine may be important. Previously, we demonstrated that nitrosamine formation could be controlled by modifications to the method by which the chlorine and ammonia reagents are added under typical chloramination conditions (25). We had defined two nitrosamine formation pathways relevant to chloramination. Under typical chloramination conditions where the chlorine to ammonia molar ratio is
