Environ. Sci. Technol. 2006, 40, 2559-2564
Cyanogen Bromide Formation from the Reactions of Monobromamine and Dibromamine with Cyanide Ion HONGXIA LEI, ROGER A. MINEAR, AND BENITO J. MARIN ˜ AS* Department of Civil and Environmental Engineering, University of Illinois at Urbana-Champaign, Urbana, Illinois 61801
Cyanide ion (CN-) was found to react with monobromamine (NH2Br) and dibromamine (NHBr2) according to the reactions NH2Br + CN- + H2O f NH3 + BrCN + OHand NHBr2 + CN- + H2O f NH2Br + BrCN + OH- with respective reaction rate constants of 2.63 × 104 M-1 s-1 and 1.31 × 108 M-1 s-1. These values were found to be 105106 times greater than those for the corresponding reactions between chloramine species and CN-. As a result, bromamines, even if present at relatively low concentrations, would tend to outcompete chloramines in reacting with CN-, and thus, the formation of BrCN would predominate that of ClCN through these reaction mechanisms. The NH2Br reaction was found to be general-acid-catalyzed. The third-order catalysis rate constants for H+, H2PO4-, HPO42-, H3BO3, and NH4+ correlated linearly with their corresponding acid dissociation constants, consistent with the Brønsted-Pedersen relationship. The NHBr2 reaction did not undergo catalysis. A model was developed to predict the concentrations of bromamines over time on the basis of the above two reactions with CN- and bromamine formation/decomposition reactions previously reported.
Introduction Cyanogen halides are disinfection byproducts (DBPs) found in drinking water treated with ozone, free chlorine, and monochloramine. The presence of cyanogen chloride (ClCN) in disinfected water was reported in a survey of 35 water utilities across the United States (1), and cyanogen bromide (BrCN) has subsequently been found to be a DBP of ozonation (2), chloramination (3, 4), and chlorination (5). Several pathways have been reported for the formation of ClCN in drinking water. Depending on the mechanism, the nitrogen in ClCN may come from chloramines or from aliphatic amino acids and heterocyclic compounds. Humic acids (6), amino acids (7), proteins, peptides (8), purines (9), and model compounds containing functional groups similar to those found in NOM (6, 10, 11) react with free chlorine or chloramines to form ClCN. Monochloramine (NH2Cl) has also been shown (12) to react with formaldehyde (HCHO), a common DBP, forming cyanide ion as an intermediate compound that reacts irreversibly with additional chloramines to produce ClCN (13). The reactions responsible for BrCN formation in drinking water are not as well-known as those for ClCN. Similar to the pathway initiated by the reaction between NH2Cl with HCHO, * Corresponding author phone: (217)333-6961; fax: (217)333-6968; e-mail:
[email protected]. 10.1021/es0519942 CCC: $33.50 Published on Web 03/18/2006
2006 American Chemical Society
TABLE 1. Absorption Characteristics of Relevant Compounds species
λ (nm)
E (M-1 cm-1)
OClOBrNH2Br
292 (max) 329 (max) 278 (max) 232 278 232 (max)
362 (17) 332 (18) 425 (16) 82 (16) 715 (16) 2000 (16)
NHBr2
one possible mechanism could be a pathway initiated by the reaction of monobromamine (NH2Br) with HCHO. Although bromamines are not used for drinking water treatment, they could form during ozonation and chlorination of waters containing bromide ion (Br-) (5, 14, 15). NH2Br would then react with HCHO, ultimately forming BrCN following a pathway parallel to that of ClCN formation. Furthermore, even if NH2Br was not able to compete with NH2Cl for the relatively low concentrations of HCHO typically available, BrCN could still be produced from the reaction between bromamines and the intermediate compound cyanide ion (CN-) produced from NH2Cl and HCHO (12). The objective of this study was to investigate the kinetics of the reactions of NH2Br and NHBr2 with CN-. An additional objective was to develop a kinetic model incorporating the rate expressions from this study with those previously reported for the decomposition of bromamines (16) and to apply the model to predict bromamine decomposition in the presence of CN-.
Experimental Section Reagents. All solutions were prepared with distilled deionized (DDI) water. Hypochlorite (OCl-) solutions were prepared by diluting reagent grade sodium hypochlorite (∼4 to 6%). Actual OCl- concentrations were measured spectrophotometrically at the maximum absorbance wavelength (λmax) of 292 nm using the OCl- molar absorptivity (17) listed in Table 1. Hypobromite (OBr-) solutions were prepared by reacting OCl- with bromide ion (Br-) at pH 11 (16) and standardized spectrophotometrically at the λmax of 329 nm using the OBrmolar absorptivity (18) also listed in Table 1. Borate buffers were prepared from reagent grade boric acid, and phosphate buffers, from reagent grade Na2HPO4 and NaH2PO4. The pH of the buffer solutions was adjusted to target values with NaOH or HCl solutions. Stock CN- solutions with a concentration of 0.2 M were prepared by dissolving analytical grade NaCN into NaOH solutions of pH 11.0 and standardized against AgNO3 solution with p-dimethylaminobenzalrhodanine as indicator. Ammonia stock solutions were prepared from analytical grade NH4Cl. Unless ammonia was used as a default buffer, a 0.2 M stock solution of buffer (phosphate or borate) was added to achieve a target stock buffer concentration of 0.04 M, which resulted in a buffer concentration of 0.01 M after mixing OBr-, ammonia, and CN- solutions. Reagent grade NaClO4 was used to adjust the ionic strength of all final reaction mixtures to 0.1 M. Experimental Methods. The pH of all solutions was measured with a Corning 340 digital pH meter equipped with a combination three-in-one electrode for automatic temperature compensation. The molar H+ ion concentration, [H+], was calculated by dividing the activity {H+} ) 10-pH by an activity coefficient calculated from the extended DeByeHu ¨ ckel equation and the Gu ¨ ntelberg approximation (19). The ionic strength effect on dissociation constant (Ka) of the VOL. 40, NO. 8, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2559
TABLE 2. Summary of Experiments Performed to Study Bromamine Reactions with Cyanide Ion, at 25 °C, Ionic Strength of 0.1 M (NaClO4) exptl set
pH
CT,OBr,0 (mM)
CT,CN,0 (mM)
CT,NH3,0 (mM)
buffer
CT,PO4 or CT,H3BO3 (M)
MB-1 MB-2a MB-2b MB-3 MB-4 MB-5 MB-6 MB-7 DB-1 DB-2 DB-3 DB-4 DB-5
9.02 8.06 8.99 8.69 8.51 to 9.46 8.61 9.44 10 6.23 6.27 to 6.42 6.49 6.67 to 6.77 6.87
0.50 0.50 0.50 0.50 0.50 0.10 to 0.60 0.50 0.10 to 0.60 0.25 to 0.5 0.50 0.2 to 0.6 0.49 0.25 to 0.50
1 1 1 2 1 1 0.5 to 5 5 0.8 0.15 to 0.5 0.8 0.15 to 0.50 0.8
5.0 5.0 5.0 1 to 12 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0
borate phosphate phosphate borate borate borate borate borate phosphate phosphate phosphate phosphate phosphate
0.005 to 0.04 0.005 to 0.03 0.005 to 0.03 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01
various acid-base chemicals used was taken into account by using Ka values reported in solutions with the same ionic strength of 0.1 M (20). The reactions of bromamines with CN- at 25 °C were studied with a SX.18MV Stopped-Flow Reaction Analyzer (Applied Photophysics Ltd., Leatherhad, Surrey, United Kingdom), which had capability for sequential mixing of three reactants and a time resolution limit of 1 ms. Because of their low chemical stability, bromamines were produced in the first mixing stage of the stopped flow system by reacting OBr- and ammonia solutions, which rapidly formed NH2Br (21). Ammonia-to-bromine molar ratios of 10-100 and a pH in the range of 6-10 were used to control the dominant species of bromamines to be either NH2Br or NHBr2. At pH above 8, NH2Br was predominant and relatively stable, and its reaction with CN- was monitored spectrophotometrically in the second stage of the stopped-flow system. At the lower pH range of 6-7, a portion of NH2Br disproportionated rapidly to NHBr2 in the first stage of the stopped-flow system (16). The resulting mixture of NH2Br and NHBr2 was reacted with CN- in the second stage. Because both bromamines were present in these experiments, their concentrations were determined by measuring UV absorbance at the λmax values of NH2Br and NHBr2, that is, 278 and 232 nm, respectively. The corresponding molar absorptivities at these wavelengths are listed in Table 1. Preliminary experiments performed by also monitoring absorbance at the additional wavelength of 258 nm, the λmax values of NBr3, confirmed that this third bromamine species was below detection for the range of experimental conditions investigated (16). Experimental Matrix. Bromamine reactions with CNwere investigated at two pH ranges. Experiments were performed at pH 8-10 (sets MB-1-7) to investigate the reaction when NH2Br was the only measurable bromamine species (Table 2). Experiments were performed to evaluate catalysis effects due to borate buffer (MB-1) and phosphate buffer (MB-2) and the role played by the concentrations of ammonium ion (MB-3), H+ (MB-4), total bromine (MB-5,7), and total cyanide (MB-6). Experimental sets were also run at pH 6-7 (sets DB-1-5) to study the reaction in the presence of both NH2Br and NHBr2 (Table 2). Experiments were designed to evaluate the effects of total bromamine (DB1,3,5) and cyanide (DB-2,4) concentrations on the observed reaction rates.
Results and Discussion NH2Br Reaction with CN-. NH2Br produced from the reaction between HOBr and NH3 disproportionated reversibly into NHBr2 and NH3 according to the acid-catalyzed reaction (16)
2NH2Br a NHBr2 + NH3 2560
9
(1)
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 8, 2006
and then both bromamine species underwent decomposition into unknown products according to the irreversible reactions (16)
NH2Br + NHBr2 f products
(2)
2NHBr2 f products
(3)
Because the decomposition of NH2Br required the presence of NHBr2 (eqs 2 and 3), and NHBr2 formed more readily at low pH and low total ammonia concentration due to predominance of the forward disproportionation reaction (eq 1), experimental sets MB-1-7 were designed at relatively high pH and in the presence of relatively high ammonia concentration (Table 2) so that NH2Br would be the predominant bromamine species. Under these conditions and drawing a parallel to reactions reported for the chloramine system (13), the main reaction that would take place in the presence of CN- would be the irreversible reaction k1
NH2Br + CN- + H2O 98 NH3 + BrCN + OH-
(4)
The corresponding rate expression for reaction 4 would be
d[NH2Br] dCT,OBr dCT,CN ≈ ) ) -k1[NH2Br][CN-] (5) dt dt dt in which CT,OBr ≈ [NH2Br] is the total bromine concentration, CT,CN ) ([HCN] + [CN-]) is the total cyanide concentration, and k1 is the second-order rate constant. [CN-] can be expressed as a function of CT,CN by
[CN-] )
Ka CT,CN [H+] + Ka
(6)
in which Ka ) 10-9.01 (20) is the acid-base dissociation constant for hydrogen cyanide (HCN) at ionic strength of 0.1 M and 25 °C, and CT,CN can be expressed as a function of CT,OBr ≈ [NH2Br] with the stoichiometric expression
CT,CN ) CT,CN,0 - CT,OBr,0 + [NH2Br]
(7)
Substituting eqs 6 and 7 into eq 5, and integrating the resulting differential equation gives
(
ln
[NH2Br]
CT,CN,0 - CT,OBr,0 + [NH2Br]
)
)
( )
Ka CT,OBr,0 t + ln (8) -k1(CT,CN,0 - CT,OBr,0) + CT,CN,0 [H ] + Ka
FIGURE 3. Plot of rate constants calculated from the slopes of linear regressions in Figure 2 for experimental set MB-1, and similar plots (not shown) for experimental sets MB-2a,b and MB-3 against the corresponding acid species ([H3BO3], [NH4+]) or total buffer (CT,PO4) concentration.
TABLE 3. Summary of Measured and Predicted Acid-Assisted Reaction Rate Constants, k1,HA, for the Reaction between NH2Br and CNFIGURE 1. Effect of boric buffer (MB-1), phosphate buffer (MB2a,b), and initial total ammonia (MB-3) concentrations on experimental (symbols) and predicted (lines) monobromamine concentration traces resulting from reacting NH2Br with CN- at µ ) 0.1 M (NaClO4), 25.0 ( 0.1 °C, and the additional specific conditions indicated in each plot.
k1,HA, M-2s-1 acid, HA
pKa(20)
H2O H+ H2PO4H3BO3 NH4+ HPO42H3PO4 H2CO3 HCO3-
15.52b
-1.72 6.72 9.01 9.29 11.74 2 6.16 10
P/ Q
measa
2/3 3/2 2/3 1/4 4/4 1/3 3/2 2/3 1/3
(4.74 ( 0.10) × (5.58 ( 0.30) × 1012 (9.22 ( 0.02) × 107 (1.59 ( 0.06) × 106 (3.02 ( 0.39) × 105 (8.57 ( 1.38) × 105
predicted 102c
2/3 1/3
3.2 × 1010 1.3 × 108 5.8 × 105
a k 1,HA measured in this study at ionic strength of 0.1 M (NaClO4) and 25 °C. b pKa for H2O given as -log(Kw/55.5). c Values for water-assisted rate constant given in terms of k1,H2O/55.5.
served with experimental sets MB-1-3 as depicted in Figure 1. The overall rate constant, k1, could then be expressed as
FIGURE 2. Plot of data from experimental set MB-1 according to eq 8. Results obtained for experimental sets MB-1-3 are shown in Figure 1, and those for sets MB-4-7 are presented in Figure S-1 of the Supporting Information section. Consistent with the preceding discussion, NH2Br was the only bromamine species detected, both initially and throughout the entire time period during which the reaction with CN- was monitored. As depicted in the various plots of Figures 1 and S-1, buffer, pH and reactant concentration effects were observed. To further analyze these effects, each data set was replotted according to eq 8. Representative results from these efforts are shown in Figure 2 for the case of experimental set MB-1. As depicted in Figure 2 and consistent with eq 8, each data set plotted linearly. The slope of each line, equal to k1(CT,CN,0 - CT,OBr,0)Ka/([H+] + Ka), was used to calculate the rate constant, k1. Catalysis Effects on the NH2Br Reaction with CN-. Drawing a parallel with the occurrence of acid catalysis for the reaction between NH2Cl and CN- (13), we assumed that acid catalysis was also responsible for the increase in apparent rate constant with increasing buffer concentration depicted in Figure 2 for experimental set MB-1 and also generally ob-
k1 ) k1,0 +
∑(k
1,HA[HA])
)
k1,0 + k1,H+[H ] + k1,H3BO3[H3BO3] + k1,H2PO4-[H2PO4-] + +
k1,HPO42- [HPO42-] + k1,NH4+[NH4+] + ‚‚‚ (9) in which k1,0 is the rate constant for water-assisted catalysis and k1,HA is the catalysis rate constant for general acid HA. A plot of rate constant k1 values against the corresponding boric acid concentration is shown in Figure 3 for experimental set MB-1. As depicted in the Figure, a linear plot that was consistent with the assumption that acid catalysis was taking place was obtained. In accordance with eq 9, the slope of the linear plot in Figure 3 was determined to obtain the k1,H3BO3 value listed in Table 3. The data for experimental sets MB-2a,b were analyzed following the same approach as that described for MB-1. The resulting rate constant k1 values are plotted against the total phosphate buffer concentration CT,PO4 ≈ ([H2PO4-] + [HPO42-]) in Figure 3. Two experimental sets at different pH were run because both H2PO4- and HPO42- could produce measurable acid catalysis, and the data sets at the two different pH values allowed the determination of the two unknown catalysis rate contants, k1,H2PO4- and k1,HPO42-, from the two slopes of the linear plots in Figure 3 after modifying the two phosphate buffer terms in eq 9 to the form VOL. 40, NO. 8, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2561
k1 ≈
(
)
Ka [H+] CT,PO4 + ‚‚‚ + k1,HPO42‚‚‚ + k1,H2PO4+ Ka + [H ] Ka + [H+] (10)
in which Ka is the dissociation constant for the acid-base reaction H2PO4- a HPO42- + H+ (Table 3). The resulting k1,H2PO4- and k1,HPO42- values are listed in Table 3. Notice that, as expected, the catalysis rate constant for the stronger acid H2PO4- is significantly higher than that for the weaker acid HPO42-. Experimental data for set MB-3 also revealed that ammonium ion had a small though measurable catalysis effect, as depicted by the corresponding linear plot in Figure 3. The slope of the line was used to calculate the catalysis rate constant k1,NH4+, in accordance with eq 9. The resulting value is listed in Table 3. The data for experimental sets MB-4-7 (Figure S-1 in Supporting Information section) were also plotted according to eq 8 to determine the catalysis rate constant for H+; however, the concentrations of the other acids present in experimental sets MB-4-7 (i.e., H3BO3, NH4+) changed when the pH was varied. But because the catalysis rate constants k1,H3BO3 and k1,NH4+ were already known from the data analysis of experimental sets MB-1,3 and the concentration of H3BO3 and NH4+ could be calculated at each pH using the corresponding Ka values (Table 3), the catalysis rate constant k1,H+ could be obtained from the slope of the line resulting from plotting (k1 - k1,H3BO3[H3BO3] - k1,NH4+[NH4+]) versus [H+], as depicted in Figure 4. The slope (k1,H+) and intercept (k1,0) obtained by linear regression are listed in Table 3. It is interesting that k1,0 ) 2.63 × 104 M-1 s-1 and k1,H+ ) 5.58 × 1012 M-2 s-1 obtained for the reaction between NH2Br and CN- are more than 106 and 102 times higher, respectively, than the corresponding rate constants k1,0 ) 1.96 × 10-2 M-1 s-1 and k1,H+ ) 4.32 × 1010 M-2 s-1 reported for the reaction of NH2Cl with CN- (13). Accordingly, if the concentration of NH2Br was >1% as compared to that of NH2Cl, NH2Br would outcompete NH2Cl in reacting with CN-, and thus, the formation of BrCN would be more pronounced than that of ClCN at pH above 8. Linear Free Energy Relationship for NH2Br Reaction with CN-. Rate constant k1,HA values for the six acid species (H2O, H+, H2PO4-, HPO42-, H3BO3, and NH4+) for which catalysis effects were determined could be correlated with their acid dissociation constants using the linear free energy relationship
( )
( )
k1,HA P log ) log GA + R log P KaQ
(12)
in which P is the maximum number of protons the acid form can donate, Q is the maximum number of protons the conjugate base of the acid can accept, and R and GA are parameters characteristic of the reaction. Using corresponding Ka, P, and Q values listed in Table 3, the resulting Brønsted-Pedersen plot is presented in Figure 5. The equation resulting from linear regression of the data is also given in the Figure. The R value of 0.58 indicates a relatively moderate degree of proton transfer during the reaction, which together with the occurrence of general acid catalysis supports the presence of the intermediate NH3+Br (22), which is formed by the loose binding of H+, coming from the various acids involved, to NH2Br (23, 24), which then reacts with CN- to form the transition state (13) shown in Scheme 1. NHBr2 Reaction with CN-. NHBr2 was produced by reacting OBr- with NH3, first producing NH2Br, which then quickly disproportionated into NHBr2 and NH3 according to eq 1. Drawing a parallel to the reaction involving dichloramine 2562
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 8, 2006
FIGURE 4. Plot of calculated rate expression terms (k1,0 + k1[H+]) for experimental sets MB-4-7 against the corresponding H+ concentration.
FIGURE 5. Brønsted-Pedersen plot for catalysis rate constants k1,HA for the reaction between NH2Br and CN- and corresponding regression line and equation obtained with the experimentally obtained values (solid symbols). Open symbols are values calculated with the regression line for the various species of interest. (13), the following reaction could take place in the presence of CN-; k2
NHBr2 + CN- + H2O 98 NH2Br + BrCN + OH-
(13)
however, not all of the NH2Br is converted into NHBr2 (eq 1), and furthermore, NH2Br is a product of the reaction between NHBr2 and CN- (eq 13). Consequently, the possibility that both bromamine species could react simultaneously with CN- needs to be addressed. Results for experimental sets DB-1,3,5 designed to assess the effect of total bromamine concentration are presented in Figure 6, and those for sets DB-2,4 performed to study the effect of cyanide concentration are shown in Figure S-2 of the Supporting Information section. As depicted in these
SCHEME 1
FIGURE 6. Effect of initial total bromine concentration (experimental sets DB-1,3,5) on experimental (symbols) and predicted (lines) bromamine concentration traces resulting from reacting NH2Br/NHBr2 mixtures with CN- at µ ) 0.1 M (NaClO4), 25.0 ( 0.1 °C and the additional specific conditions indicated in Table 2. Figures, a significant portion of the reaction between NHBr2 and CN- took place within a few milliseconds. Because the reaction between NH2Br and CN- was slower, the early portion of the kinetic curves in Figures 6 and S-2 could be used to determine the reaction rate for eq 13. The rate expression for reaction 13 would be
d[NHBr2] dCT,CN ≈ ) -k2[NHBr2][CN-] dt dt
(14)
Substituting [CN-] in terms of CT,CN (eq 6) and the stoichiometric expression
CT,CN ) CT,CN,0 - [NHBr2]0 + [NHBr2]
(15)
into eq 14 and integrating results in
(
ln
[NHBr2]CT,CN,0
)
) [NHBr2]0(CT,CN,0 - [NHBr2]0 + [NHBr2]) Ka (CT,CN,0 - [NHBr2]0)t (16) -k2 + [H ] + Ka
The data for experimental set DB-1 are represented according to eq 16 in Figure 7. As depicted in the figure, all four data sets plotted linearly within the time range of 1-10 ms, with subsequent apparent drops in reaction rate due to NHBr2 depletion and NH2Br reacting with CN-. Data obtained at times below 1 ms, the resolution limit of the stopped flow system, were disregarded. According to eq 16, the slope of the linear regression of the initial data (i.e., t ) 1-10 ms) in Figure 7 was equal to k2Ka/([H+] + Ka); however, because pH , pKa or [H+] . Ka for all experimental sets DB-1-5, the slope could be assumed to be approximately equal to the simplified expression k2Ka/[H+]. Because the pH was constant for all four experimental runs of DB-1, a unique regression line should have been obtained. The variability in linear regression slope ((10%) depicted in Figure 7 was likely associated with small experimental errors in pH and reactant concentration measurements. The slopes obtained for experimental sets DB-1-5, obtained following the same approach depicted in Figure 7 for set DB-1, are plotted against the inverse of [H+] in Figure 8. As depicted in the Figure, a linear plot was obtained, and the slope of the linear regression shown would be equal to
FIGURE 7. Plot of data from experimental set DB-1 according to eq 16.
FIGURE 8. Plot of the slopes of linear regressions in Figure 7 for experimental set DB-1 and similar plots (not shown) for experimental sets DB-2-5 against the corresponding inverse H+ concentration. k2Ka. Using the value pKa ) 9.01 (20) reported for the acid HCN at 25 °C and ionic strength of 0.1 M, the resulting secondorder rate constant value for reaction 13 was k2 ) 1.31 × 108 M-1 s-1. VOL. 40, NO. 8, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2563
Experiments performed (data not shown) to check the occurrence of catalysis for reaction 13 revealed that, consistent with the observation for the parallel reaction between NHCl2 and CN- (13), the reaction of NHBr2 with CN- did not undergo acid catalysis. It is interesting to notice that the k2 value found in the present study for the reaction of NHBr2 with CN- is more than 105 times greater than the value k2 ) 576 M-1 s-1 reported for the reaction of NHCl2 with CN(13). Consequently, it would be expected that bromamines would also outcompete chloramines in reacting with CN- at the lower pH range of 6-7. General Model for Bromamine Decomposition in the Presence of CN-. The concentration of bromamines as a function of time in the presence of CN- could be generally predicted by expanding the predictive model developed by Lei et al. (16) for reactions 1-3 through incorporating terms corresponding to reactions 4 and 13 in the overall rate expressions for bromamine kinetics. The predicted curves obtained with such expanded model for all experimental sets (MB-1-7 and DB-1-5) are plotted in Figures 1, 6, S-1, and S-2. As shown in the Figures, the predictions were in generally good agreement with the data except for reaction times below ∼1 ms, the time required for mixing to take place inside the stopped-flow cell. In general, reaction 4 was predominant for the entire reaction time for experimental sets MB-1-7 (Figures 1 and S-1). In contrast, although reaction 13 was predominant at early reaction times (1-10 ms) for experimental sets DB-1-5 (Figures 6 and S-2), the effects of reactions 1-4 were observed at later reaction times.
Acknowledgments This research was supported by a grant from the U.S. Environmental Protection Agency Science to Achieve Results (STAR) program (Grant No. R826830-01). Mention of trade names or commercial products does not constitute endorsement or recommendation for use. The scientific views expressed are solely those of the authors and do not necessarily reflect those of U.S. EPA.
Supporting Information Available
(7)
(8)
(9)
(10)
(11)
(12)
(13) (14)
(15) (16) (17) (18)
Additional information concerning results and modeling for experimental sets MB-4-7 and DB-2,4 can be found in the Supporting Information. This material is available free of charge via the Internet at http://pubs.acs.org.
(19)
Literature Cited
(21)
(1) Krasner, S. W.; McGuire, M. J.; Jacangelo, J. G.; Patania, N. L.; Reagan, K. M.; Aieta, E. M. The occurrence of disinfection byproducts in U.S. drinking water. J. Am. Water Works Assoc. 1989, 81, 41-53. (2) West, M. J.; Lieu, T. K.; Krasner, S. W. Analytical determination of cyanogen bromide by electron-capture gas chromatography. In Proceedings of Water Quality Technology Conference; AWWA: Orlando, Fla, 1991. (3) Sclimenti, M. J.; Hwang, C. J.; Speitel, J., G. E.; Diehl, A. C. The simultaneous determination of cyanogen chloride and cyanogen bromide in chloraminated waters by a simplified microextraction GC/ECD technique. In Proceedings of Water Quality Technology Conference; AWWA: San Francisco, CA.; 1994. (4) Xie, Y.; Reckhow, D. A. A rapid and simple analytical method for cyanogen chloride and cyanogen bromide in drinking water. Water Res. 1993, 27, 507-511. (5) Heller-Grossman, L.; Idin, A.; Limoni-Relis, B.; Rebhun, M. Formation of cyanogen bromide and other volatile DBPs in the disinfection of bromide-rich lake water. Environ. Sci. Technol. 1999, 33, 932-937. (6) Ohya, T.; Kanno, S. Formation of cyanide ion or cyanogen chloride through the cleavage of aromatic rings by nitrous acid or chlorine. X. Pathway of cyanogen chloride formation in the
2564
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 8, 2006
(20)
(22) (23)
(24)
reactions of 1-naphthol and 4-phenylimidazole with chloramine. Chem. Pharm. Bull. 1988, 36, 4095-4102. Hirose, Y.; Maeda, N.; Ohya, T.; Nojima, K.; Kanno, S. Formation of cyanogen chloride by the reaction of amino acids with hypochlorous acid in the presence of ammonium ion. Chemosphere 1988, 17, 865-873. Hirose, Y.; Maeda, N.; Ohya, T.; Nojima, K.; Kanno, S. Formation of cyanogen chloride by the reaction of peptides and proteins with hypochlorous acid in the presence of ammonium ion. Chemosphere 1989, 18, 2101-2108. Ohya, T.; Kanno, S. Formation of cyanide ion or cyanogen chloride through the cleavage of aromatic rings by nitrous acid or chlorine. XI. On the reaction of purine bases with hypochlorous acid in the presence of ammonium ion. Chemosphere 1989, 19, 1835-1842. Kanno, S.; Nojima, K.; Ohya, T. Formation of cyanide ion or cyanogen chloride through the cleavage of aromatic rings by nitrous acid or chlorine. IV. On the reaction of aromatic hydrocarbons with hypochlorous acid in the presence of ammonium ion. Chemosphere 1982, 11, 663-667. Kanno, S.; Nojima, K.; Ohya, T. Formation of cyanide ion or cyanogen chloride through the cleavage of aromatic rings by nitrous acid or chlorine. V. On the reaction of aromatic amines or phenolic compounds with hypochlorous acid in the presence of ammonium ion. Chemosphere 1982, 11, 669-673. Pedersen, E. J., III; Urbansky, E. T.; Marin ˜ as, B. J.; Margerum, D. W. Formation of cyanogen chloride from the reaction of monochloramine with formaldehyde. Environ. Sci. Technol. 1999, 33, 4239-4249. Schurter, L. M.; Bachelor, P. P.; Margerum, D. W. Nonmetal redox kinetics: mono-, di-, and trichloramine reactions with cyanide ion. Environ. Sci. Technol. 1995, 29, 1127-1134. von Gunten, U.; Hoigne, J. Bromate formation during ozonization of bromide-containing waters: interaction of ozone and hydroxyl radical reactions. Environ. Sci. Technol. 1994, 28, 12341242. Song, R.; Amy, G.; Westerhoff, P.; Minear, R. A. Bromate minimization during ozonation. J. Am. Water Works Assoc. 1997, 89, 69-78. Lei, H.; Marin ˜ as, B. J.; Minear, R. A. Bromamine decomposition kinetics in aqueous solution. Environ. Sci. Technol. 2004, 38, 2111-2119. 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. 1998, 37, 4321-4327. Troy, R. C.; Margerum, D. W. Nonmetal redox kinetics: hypobromite and hypobromous acid reactions with iodide and with sulfite and the hydrolysis of bromosulfate. Inorg. Chem. 1991, 30, 3538-3543. Snoeyink, V. L.; Jenkins, D. Water Chemistry; John Wiley & Sons: New York, 1980. Smith, R. M.; Martell, A. E. Critical Stability Constants; Plenum Press: New York, 1976; Vol. 4. Wajon, J. E.; Morris, J. C. Bromamination chemistry: rates of formation of NH2Br and some N-bromamino acids. In Water Chlorination: Environmental Impact and Health Effects; Jolley, R. L., Brungs, W. A., Crumming, R. B., Jacobs, V. A., Eds.; Ann Arbor Science: Ann Arbor, MI, 1980; Vol. 3, pp 171-181. Inman, G. W., Jr.; Johnson, J. D. Kinetics of monobromamine disproportionation-dibromamine formation in aqueous ammonia solutions. Environ. Sci. Technol. 1984, 18, 219-224. Cristina, S. P.; Azure, M. T.; Workman, H. J.; Gray, E. T., Jr. Decomposition of bromamines in aqueous solution: preliminary report of the decomposition kinetics and disproportionation of NH2Br. In Water Chlorination: Chemistry, Environmental Impact and Health Effects; Jolley, R. L., Bull, R. J., Davis, W. P., Katz, S., Roberts, M. H., Jr., Jacobs, V. A., Eds.; Lewis Publishers: Chelsea, MI, 1985; Vol. 5, pp 763-774. Valentine, R. L.; Jafvert, C. T. Comment on “Kinetics of monobromamine disproportionation-dibromamine formation in aqueous ammonia solutions”. Environ. Sci. Technol. 1985, 19, 286-287.
Received for review October 7, 2005. Revised manuscript received January 11, 2006. Accepted January 16, 2006. ES0519942
