Chloramine has often been used as a chlorine alternative for trihalomethane (THM) and haloacetic acid (HAA) control. However, nitrogenous disinfection byproduct (N-DBP) formation and nitrification in distribution have always been major concerns in chloramination practices. On Kinmen Island, the high organic nitrogen content in raw water may increase the nitrogenous DBP formation. Simulated distribution system tests were conducted to explore the DBP formation kinetics in the distribution system. Lower haloacetonitrile (HAN4) formation (0.26 μg L−1) with chloramination than with chlorination (10.48 μg L−1) was observed from the 24 hours of reaction time. The nitrogen sources contributing to the dichloroacetonitrile (DCAN) formation kinetics were explored with 15N-chloramination. The results showed that nitrogen sourced from organic nitrogen was more dominant in DCAN formation with low chloramine dosage. This suggests that chloramine contributes to less DCAN formation in practice, especially for short distribution systems. In summary, the results provide evidence that simultaneous post-chloramination and pre-chlorination would be a feasible disinfection strategy applied to control regulated THM and HAA formation on Kinmen Island.

INTRODUCTION

Mono-chloramine can decrease the formation of regulated disinfection byproducts (DBPs), trihalomethanes (THMs) and haloacetic acids (HAAs). However, some studies indicated that more nitrogenous DBPs (e.g. dihaloacetonitriles and halonitromethanes) can be found in chloramination (Plewa et al. 2008; Shah & Mitch 2012). Nitrogenous DBPs have often been regarded as potentially highly genotoxic and a cytotoxic chemical molecule for mammalian cells (Plewa et al. 2003; Muellner et al. 2007; Plewa et al. 2008). Some studies indicated that chloramination caused the N-DBP ratio to increase for the total DBP formation (Dotson et al. 2009; Bond et al. 2011).

The nitrogen source in nitrogenous DBPs can originate from either inorganic/organic nitrogen in source water or mono-chloramine in disinfection processes (Margerum et al. 1994; Pedersen et al. 1999). Two nitrogenous disinfection byproduct formation mechanisms in both chlorination and chloramination were postulated, described as the ‘aldehyde-chloramine pathway’ and the ‘decarboxylation pathway’. The aldehyde-chloramine pathway relates to the reaction between aldehyde and chloramine (Figure 1) (Joo & Mitch 2007). Chlorination of the organic compound occurs initially, followed by dechlorination and aldehyde formation. The reaction between aldehyde and chloramine produces imines as intermediates. A series of dehydration and chlorination then results in chlorinated N-DBP formation. The decarboxylation pathway commonly occurs in a rich chlorine environment. Chlorination of nitrogen-containing compounds occurs at the start of the reactions. Combined with decarboxylation and dechlorination simultaneously, N-DBPs and other chlorinated disinfection byproducts may be formed (Figure 2) (Trehy et al. 1986; Hureiki et al. 1994; Pedersen et al. 1999; Joo & Mitch 2007; Yang et al. 2012). Therefore, it is important to carefully select appropriate disinfectants and disinfection practices for both regulated and non-regulated DBP formation control.
Figure 1

Primary amine propylamine chlorination reaction (Joo & Mitch 2007).

Figure 1

Primary amine propylamine chlorination reaction (Joo & Mitch 2007).

Figure 2

Amino acid aspartic acid chlorination reaction (Trehy et al. 1986).

Figure 2

Amino acid aspartic acid chlorination reaction (Trehy et al. 1986).

Regular droughts and pollution from domestic sewage has resulted in high natural organic matter and frequent algal blooms in water reservoirs on Kinmen Island. The organic compounds, normally precursors of disinfection byproducts, cannot be removed with water treatment processes completely. Therefore, DBP formation can be observed in the post-chlorination units (e.g. THMs and HAAs). The total THM concentrations in finished water after the conventional treatment process often exceeded Taiwan EPA's drinking water standard (80 μg L−1).

The main aim of this study is to estimate the effectiveness of DBP control strategies with two disinfectants (NaOCl and NH2Cl), with special focus on formation kinetics in the distribution system. Since Kinmen Island has a short water residence time (24 hours on average), the DBP formation kinetics were explored with simulated distribution system (SDS) tests.

MATERIAL AND METHODS

Sample collection

In this study, water samples from the effluent of each process of the TaiHu water treatment plant were collected in March, 2014. The TaiHu water treatment plant has conventional coagulation, aerated flotation, rapid sand filtration, slow sand filtration and chlorination. The DBP concentrations in each field sample were analyzed in the laboratory within 48 hours after sample collection. Standard 7-day DBP formation potential tests with chlorine or chloramine were conducted for waters from each process. The slow sand filtration effluent was also tested with 24 hour SDS tests to explore DBP formation in the distribution system.

Disinfectant preparation

Sodium hypochlorite stock solution (Sigma Aldrich, USA) was used as the chlorination oxidant. The chloramine stock solution preparation was according to Hu et al. with slight modification. Sodium hypochlorite and ammonium chloride (Isotec, USA) were mixed with a mass ratio of Cl2:N equal to 3.5:1 (or molar ratio 0.69:1), and the pH was adjusted to pH 8.5 with 0.1N sodium hydroxide solution (Hu et al. 2010).

DBP analysis

DBPs in the water samples were concentrated by liquid–liquid extraction (adapted from USEPA method 551.1 and 552.3) before being determined by a gas chromatography-electron capture detector (GC-μECD, 6890N, Agilent Technologies, USA). Methyl tert-butyl ether (MTBE, MACRON, USA) was used as the extraction solvent. Three mL solvent were added to a dechlorinated 30 mL water sample in a 40 mL glass vial, then 10 g sodium sulfate salt (MACRON, USA) were also added into the solution. After vigorous mixing, for method 551.1, the upper solvent layer was transferred to clean sample vials for further gas chromatographic analysis. For method 552.3 (HAAs), samples were derivatized with 10% (v/v) sulfuric acid in methanol solution before gas chromatographic analysis.

To investigate nitrogen sources of N-DBP, chloramine with the 15N isotope labeled was used in chloramination experiments. Isotopic chloramine preparation and analysis were according to Yang et al. The preparation method was the same as above, except that isotopic ammonium chloride (15NH4Cl) was used as the ammonium source. Isotopes 14N and 15N nitrogenous disinfection byproduct dichloroacetonitrile (DCAN) formation was analyzed by gas chromatography mass spectrometry (GC-MS, 5975C MSD, Agilent Technologies, USA) with Electron-Impact Ionization mode (Yang et al. 2010).

RESULTS AND DISCUSSION

Water treatment processes

Pre-chlorination before the coagulation process is practiced at the TaiHu water treatment plant and some DBP formation occurred (Figure 3). Trihalonitromethanes (TCNM) were not detected in these samples collected from the field. THM concentrations increased substantially at the final disinfection (Figure 3(a)). This suggested that THMs have not been removed by the water treatment processes, and the formation exceeded Taiwan EPA's regulation standards for disinfection (showed as a dotted line, Figure 3(a)). The slow sand filtration process removed haloacetonitriles and HAAs from pre-chlorination effectively. However, DBPs in the disinfection effluent suggested that the existence of DBP precursors still caused substantial DBP formation during post-chlorination in the finished water.
Figure 3

DBP concentrations in the effluents of each process. (a) THM4, (b) HAN4, (c) HAA9 (CF: coagulation/flotation; RF: rapid sand filtration; SF: slow sand filtration; DI: disinfection).

Figure 3

DBP concentrations in the effluents of each process. (a) THM4, (b) HAN4, (c) HAA9 (CF: coagulation/flotation; RF: rapid sand filtration; SF: slow sand filtration; DI: disinfection).

Simulated distribution system test

Chloramination and chlorination trials need to be conducted with different dosages and reaction times, and the finished water distribution time (lower than 24 hours in Kinmen) and chlorine residual were often considered. Comparison of the two different disinfectants was also based on the actual conditions. SDS tests were performed to compare chlorine with chloramine for the slow sand filtration effluent sample.

According to the Taiwan EPA drinking water regulation, the free chlorine residual must be maintained at 0.2–1.0 mg-Cl2 L−1 in the distribution system. In the study, 24 hours of water residence time was assumed in the distribution system, and 0.2–1.0 mg-Cl2 L−1 of free chlorine residual and total chlorine residual must be maintained in the 24-hour chlorination and chloramination tests, respectively. The following SDS test was conducted with the slow sand filtration effluent to explore the possible DBP formation in the distribution system. Chlorination with an initial dosage of 4 mg-Cl2 L−1, and chloramination with an initial dosage of 1 mg-Cl2 L−1 were obtained through preliminary trials to find the adequate initial dosage which complies with Taiwan EPA's distribution water chlorine residual requirements.

From the 24 hours of retention time SDS tests (Figures 4 and 5 for chlorination and chloramination, respectively), lower THM4 and HAA9 formation in chloramination was observed (25.4 μg L−1 and 5.7 μg L−1 within 24 hours of reaction time for THM4 and HAA9 formation, respectively), compared with chlorination (159.9 μg L−1 and 133.6 μg L−1 within 24 hours of reaction time for THM4 and HAA9 formation, respectively), which complied with Taiwan EPA's drinking water DBP regulation. Meanwhile, chloramination with low dosage condition (applied in the SDS test) did not promote more haloacetonitriles formation (TCNMs were not detected in this study). Lower HAN4 formation (0.26 μg L−1 within 24 hours of reaction time) from chloramination was observed, which was lower than with chlorination (10.48 μg L−1 within 24 hours of reaction time).
Figure 4

SDS test results during chlorination. (a) THMs, (b) HAAs, (c) HANs.

Figure 4

SDS test results during chlorination. (a) THMs, (b) HAAs, (c) HANs.

Figure 5

SDS test results during chloramination. (a) THMs, (b) HAAs, (c) HANs.

Figure 5

SDS test results during chloramination. (a) THMs, (b) HAAs, (c) HANs.

For the range of disinfectant dosages, the haloacetonitrile formation kinetic curve indicated that higher haloacetonitrile formation was found in chlorination than in chloramination (Figures 4(c) and 5(c)). However, the simulation results still need to be modified after considering the true conditions in the distribution pipes. Flow rate, pipe material and micro-organisms can probably contribute to change the chlorine residual decay rate, THM4 and HAA9 formation, etc. (Rossman et al. 2001; Tung & Xie 2009).

Nitrogen source tracking

From SDS tests, HAN4 formation was lower during chloramination than during chlorination. The nitrogen source of HAN formation may come from either soluble nitrogen in source water or chloramine. In this study, nitrogen sources of HAN formation were explored by isotopic studies. Diluted TaiHu raw water (dissolved organic carbon = 4 mg-C L−1, pH 7) was used as the target sample in the nitrogen source tracking experiments.

DCAN is one major specie of haloacetonitriles. By GC-MS quantification, DCAN could be grouped into 14N-DCAN and 15N-DCAN. 14N-DCAN meant the nitrogen source came from the original substrate existing in the water sample, while 15N-DCAN meant the nitrogen source came from the external isotopic chloramine involved in the DCAN formation reaction.

Nitrogen isotope tracking implied that the substrate-nitrogen precursor in water dominated for the lower disinfectant dosage (3 mg-Cl2 L−1 dosage for Figure 6(a)). In the 7 day formation potential test results, the 14N-DCAN to 15N-DCAN ratio was 48.9:50.2, results for other dosages are given in Table 1. 15N-DCAN formation linearly depended on mono-chloramine exposure (chlorine residual times contact time, CT value) despite the initial disinfectant dosage; while 14N-DCAN formation was second-order correlated with the CT value (Figure 6(b)). Similar results were found from the study of Chuang & Tung (2015). The key difference between 14N-DCAN and 15N-DCAN is the formation of intermediate compounds. Some intermediates formation was included in the 15N-DCAN formation reaction, and limited reaction pathway exit (Chuang 2013). The 14N-DCAN formation reaction could be explained with second-order reactions between chloramine and organic compounds (Gallard & von Gunten 2002). The different reaction order for 14N-DCAN and 15N-DCAN may be important for water treatment practice. With a lower dosage, nitrogen from natural substrates in water is probably more dominant in DCAN formation. Under the same conditions, the effect on DCAN formation reaction with chloramine was less dominant. However, other nitrogen sources of N-DBPs need to be investigated (e.g. HAcAms, TCNM, NDMA, etc.). More detailed information about chloramine as a nitrogen source involved in DBP formation could provide more chloramine risk evaluation.
Table 1

DCAN formation and isotopic species composition ratio with different chloramine dosages

DCAN Conc. at day 7 (μg L−1-mg C L−1)Ratio (14N:15N) at day 7 (%)
12 mg Cl2 L−1 0.413 35.1: 64.9 
6 mg Cl2 L−1 0.242 44.6: 55.4 
3 mg Cl2 L−1 0.153 49.8: 50.2 
DCAN Conc. at day 7 (μg L−1-mg C L−1)Ratio (14N:15N) at day 7 (%)
12 mg Cl2 L−1 0.413 35.1: 64.9 
6 mg Cl2 L−1 0.242 44.6: 55.4 
3 mg Cl2 L−1 0.153 49.8: 50.2 
Figure 6

(a) With contact time as scale, 14N-DCAN and 15N-DCAN formation curve with initial chloramine dosage of 3 mg-Cl2L−1. (b) With exposure (CT) as scale, 14N-DCAN and 15N-DCAN formation distribution.

Figure 6

(a) With contact time as scale, 14N-DCAN and 15N-DCAN formation curve with initial chloramine dosage of 3 mg-Cl2L−1. (b) With exposure (CT) as scale, 14N-DCAN and 15N-DCAN formation distribution.

In summary, SDS results indicated that for all regulated DBPs, THM4 and HAA9, the concentration in distribution systems where chloramination is used can comply with Taiwan EPA's drinking water standards, and a lower formation of HAN4 than with chlorination was observed. This provides evidence that pre-chlorination and post-chloramination may be a possible strategy for DBP control in the TaiHu treatment plant. However, other non-regulated DBPs should be explored based on similar experiments, and their formation should be included in future disinfection strategic planning.

CONCLUSIONS

DBP formation potentials from chlorination and chloramination were explored for waters from the TaiHu treatment plant. The results showed that:

  1. It is possible to control THM4 and HAA9 by chloramination. In addition, HAN4 formations were lower in chloramination than in chlorination. The results implied that chloramination may be a feasible disinfection strategy for the TaiHu water treatment plant.

  2. For DCAN, one of the major species of HANs found in our samples, the precursors mainly originated from the nitrogen source originally existing in the water; the nitrogen contributed from chloramine did not dominate the DCAN formation in chloramination.

  3. For compliance purposes, pre-chlorination plus post-chloramination may be a possible disinfection method to control regulated DBPs. However, formation of nitrosamines may be a concern if chloramination is practiced. Further studies may be needed to explore the formation of nitrosamine under the proposed disinfection strategy.

ACKNOWLEDGEMENT

The authors acknowledge the financial support from the Taiwan Ministry of Science and Technology (project numbers: ‘NSC 100-2628-E-002-022-MY3’ and ‘MOST 103-2119-M-002-004’).

REFERENCES

REFERENCES
Chuang
Y. H.
2013
The Formation Kinetics of Haloacetonitriles and Halonitromethanes during Chloramination
.
Doctoral Thesis
,
Graduate Institute of Environmental Engineering, National Taiwan University
,
Taipei
,
Taiwan ROC
.
Hu
J.
Song
H.
Addison
J. W.
Karanfil
T.
2010
Halonitromethane formation potentials in drinking waters
.
Water Research
44
(
1
),
105
114
.
Hureiki
L.
Croué
J. P.
Legube
B.
1994
Chlorination studies of free and combined amino acids
.
Water Research
28
(
12
),
2521
2531
.
Margerum
D. W.
Schurter
L. M.
Hobson
J.
Moore
E. E.
1994
Water chlorination chemistry: nonmetal redox kinetics of chloramine and nitrite ion
.
Environmental Science & Technology
28
(
2
),
331
337
.
Muellner
M. G.
Wagner
E. D.
McCalla
K.
Richardson
S. D.
Woo
Y. T.
Plewa
M. J.
2007
Haloacetonitriles vs. regulated haloacetic acids: are nitrogen-containing DBPs more toxic?
Environmental Science & Technology
41
(
2
),
645
651
.
Pedersen
E. J.
Urbansky
E. T.
Marinas
B. J.
Margerum
D. W.
1999
Formation of cyanogen chloride from the reaction of monochloramine with formaldehyde
.
Environmental Science & Technology
33
(
23
),
4239
4249
.
Plewa
M. J.
Wagner
E. D.
Jazwierska
P.
Richardson
S. D.
Chen
P. H.
McKague
A. B.
2003
Halonitromethane drinking water disinfection byproducts: chemical characterization and mammalian cell cytotoxicity and genotoxicity
.
Environmental Science & Technology
38
(
1
),
62
68
.
Plewa
M. J.
Muellner
M. G.
Richardson
S. D.
Fasanot
F.
Buettner
K. M.
Woo
Y. T.
McKague
A. B.
Wagner
E. D.
2008
Occurrence, synthesis, and mammalian cell cytotoxicity and genotoxicity of haloacetamides: an emerging class of nitrogenous drinking water disinfection byproducts
.
Environmental Science & Technology
42
(
3
),
955
961
.
Rossman
L. A.
Brown
R. A.
Singer
P. C.
Nuckols
J. R.
2001
DBP formation kinetics in a simulated distribution system
.
Water Research
35
(
14
),
3483
3489
.
Trehy
M. L.
Yost
R. A.
Miles
C. J.
1986
Chlorination byproducts of amino acids in natural waters
.
Environmental Science & Technology
20
(
11
),
1117
1122
.