Formation of N-nitrosodimethylamine (NDMA) from reaction of monochloramine: A new disinfection by-product

Article abstract of DOI:10.1016/S0043-1354(01)00303-7

Studies have been conducted specifically to investigate the hypothesis that N-nitrosodimethylamine (NDMA) can be produced by reactions involving monochloramine. Experiments were conducted using dimethylamine (DMA) as a model precursor. NDMA was formed from the reaction between DMA and monochloramine indicating that it should be considered a potential disinfection by-product. The formation of NDMA increased with increased monochloramine concentration and showed maximum in yield when DMA was varied at fixed monochloramine concentrations. The mass spectra of the NDMA formed from DMA and ¹⁵N isotope labeled monochloramine (¹⁵NH₂Cl) showed that the source of one of the nitrogen atoms in the nitroso group in NDMA was from monochloramine. Addition of 0.05 and 0.5mM of preformed monochloramine to a secondarily treated wastewater at pH 7.2 also resulted in the formation of 3.6 and 111ng/L of NDMA, respectively, showing that this is indeed an environmentally relevant NDMA formation pathway. The proposed NDMA formation mechanism consists of (i) the formation of 1,1-dimethylhydrazine (UDMH) intermediate from the reaction of DMA with monochloramine followed by, (ii) the oxidation of UDMH by monochloramine to NDMA, and (iii) the reversible chlorine transfer reaction between monochloramine and DMA which is parallel to (i). We conclude that reactions involving monochloramine in addition to classical nitrosation reactions are potentially important pathways for NDMA formation. Copyright

Full text of DOI:10.1016/S0043-1354(01)00303-7

Water Research 36 (2002) 817–824
Formation of N-nitrosodimethylamine (NDMA) from reaction
of monochloramine: a new disinfection by-product
Junghoon Choi, Richard L. Valentine*
Department of Civil and Environmental Engineering, University of Iowa, Iowa City, IA 52242-1527, USA
Received 1 February 2001; accepted 1 June 2001
Abstract
Studies have been conducted specifically to investigate the hypothesis that N-nitrosodimethylamine (NDMA) can be
produced by reactions involving monochloramine. Experiments were conducted using dimethylamine (DMA) as a
model precursor. NDMA was formed from the reaction between DMA and monochloramine indicating that it should
be considered a potential disinfection by-product. The formation of NDMA increased with increased monochloramine
concentration and showed maximum in yield when DMA was varied at fixed monochloramine concentrations. The
1
5
15
mass spectra of the NDMA formed from DMA and N isotope labeled monochloramine ( NH Cl) showed that the
2
source of one of the nitrogen atoms in the nitroso group in NDMA was from monochloramine. Addition of 0.05 and
0
3
.5 mM of preformed monochloramine to a secondarily treated wastewater at pH 7.2 also resulted in the formation of
.6 and 111 ng/L of NDMA, respectively, showing that this is indeed an environmentally relevant NDMA formation
pathway. The proposed NDMA formation mechanism consists of (i) the formation of 1,1-dimethylhydrazine (UDMH)
intermediate from the reaction of DMA with monochloramine followed by, (ii) the oxidation of UDMH by
monochloramine to NDMA, and (iii) the reversible chlorine transfer reaction between monochloramine and DMA
which is parallel to (i). We conclude that reactions involving monochloramine in addition to classical nitrosation
reactions are potentially important pathways for NDMA formation. r 2002 Elsevier Science Ltd. All rights reserved.
Keywords: Disinfection by-product (DBP); N-nitrosodimethylamine (NDMA); Monochloramine; 1,1-dimethylhydrazine (UDMH);
Chlorine transfer
1
. Introduction
California has established a temporary action level of
2
0 ng/L for NDMA in drinking water [4].
Nitrosamines have been found in many food products
Nitrosamines are a class of compounds, many of
which are carcinogenic, mutagenic, and teratogenic [1,2].
Particularly, N-Nitrosodimethylamine (NDMA) has
been classified as a probable human carcinogen by the
US Environmental Protection Agency [3]. Risk assess-
ments from the US EPA identify a theoretical 10ꢀ
lifetime risk level of cancer from NDMA exposures as
as well as soils, wastewater, and drinking water [5,6]. It
is generally thought that nitrosamine formation involves
N-nitrosation, a reaction between nitrosatable amines
and nitrite [7,8]. Therefore, NDMA is likely to be found
especially where both secondary amines and nitrite
occur. Environmentally oriented NDMA occurrence
and formation studies have been generally empirical in
nature and have focussed primarily on determining if,
not how, nitrosamines may be formed in water. Most
studies are predicated on the assumption that nitrite is a
required reactant and have added it to water to
determine an NDMA formation potential. Ayanaba
and Alexander [9] demonstrated the formation of
6
0.7 ng/L [3]. There is no state or federal drinking water
maximum contaminant level, however the State of
*
Corresponding author. Tel.: +1-319-335-5653; fax +1-319-
35-5660.
E-mail addresses: junghoon-choi@uiowa.edu (J. Choi),
richard-valentine@uiowa.edu (R.L. Valentine).
3
0043-1354/02/$ -see front matter r 2002 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 4 3 - 1 3 5 4 ( 0 1 ) 0 0 3 0 3 - 7

818
J. Choi, R.L. Valentine / Water Research 36 (2002) 817–824
NDMA in the lake water at neutral pH values when
dosed with nitrite and dimethylamine (DMA) and
trimethylamine. Studies have also investigated the
influence of other precursors or fulvic acid on nitrosa-
mine formation with added nitrite [10,11].
Recently, NDMA was found in highly purified
wastewaters intended for recycle as well as some treated
drinking waters while absent in the influent streams [4].
The current investigations in California indicate that
NDMA could be more commonly observed in treated
waters than previously suspected, which suggests that
NDMA occurrences may be related to treatment and
disinfection processes.
Our studies were conducted to investigate the
hypothesis that NDMA is a disinfection by-product
specifically produced by the reaction of monochlora-
mine and DMA in the absence of nitrite. Monochlor-
amine is purposely produced as a disinfectant, which
may also form in chlorinated water in the presence of
ammonia. We selected DMA as a potential precursor
because it is ubiquitous in surface and wastewater
concentration of 0.1 mM. This study used preformed
monochloramine additions to the DMA containing
solutions as opposed to forming it in the presence of
DMA and ammonia by addition of HOCl. This was
done to simplify interpretation of the results by avoiding
the need to consider initial competing reactions invol-
ving HOCl. The concentration of ammonia was also
varied from 0.14 to 1.0 mM to obtain varied mono-
chloramine Cl/N molar ratios (0.1–0.7). The mixtures
were reacted in the dark for 0.5–40 h. An experiment was
also conducted to evaluate the relative importance of
nitrosation by reacting sodium nitrite with DMA in the
absence of monochloramine. DMA was also reacted
1
5
2
with isotopically labeled NH Cl to identify the source
of the nitrogen atoms in the nitroso group of NDMA.
Lastly, NDMA formation potential in collected secon-
darily treated wastewater was evaluated by adding 0.05
and 0.5 mM of monochloramine.
2
.3. Preparation and analysis of monochloramine
[
12,13]. Other alkylamines or pesticides may also
decompose to give rise to potential precursors of
NDMA [9]. A series of experiments were conducted to
identify the formation of NDMA from monochloramine
and DMA. Unlike N-nitrosation that requires nitrite, we
propose a new NDMA formation mechanism that
involves monochloramine. A kinetic model was devel-
oped and used to validate the proposed mechanism.
Monochloramine stock solutions were prepared in
mM bicarbonate buffer by adding predetermined
4
amount of sodium hypochlorite solution into ammo-
nium sulfate solution. The pH was adjusted to
approximately 9.6 by addition of NaOH and aged in
1
5
the dark for at least 1 h before use. N isotope labeled
15
monochloramine ( NH Cl) stock solutions were pre-
2
1
5
pared using N isotope labeled ammonium sulfate. The
concentration of monochloramine was measured using
the N,N–diethyl–p–phenylenediamine ferrous ammo-
nium sulfate (DPD-FAS) method [14].
2
. Materials and methods
2
.1. Chemicals
N-nitrosodimethylamine (NDMA, 100 mg/mL in
methanol) was obtained from Chem Service. Deuterated
2.4. NDMA analysis
N-nitrosodimethylamine (d -NDMA, 1000 mg/mL in
6
methanol) was obtained from Protocol Analytical
NDMA was determined by isotope dilution gas
chromatography/mass spectrometer (GC/MS) method
[15]. Prior to extraction, all 1 L samples are dosed with
1
5
15
Supplies. N isotope labeled ammonium sulfate ( N
2
,
9
8%+) was obtained from Cambridge Isotope Lab, Inc.
Sodium nitrite, 1,1-dimethylamine hydrochloride, and
ammonium sulfate were obtained from Aldrich. Sodium
hypochlorite solution was obtained from Fisher Scien-
tific. All other chemicals used in these experiments were
analytical laboratory grade.
6
the isotopically labeled d -NDMA as an internal
standard. 1 L sample is added with 200 mg of carbonac-
eous adsorbent (Ambersorb 572, Aldrich) and extracted
by shaking the solution for 1 h at 200 rpm. Ambersorb
beads are vacuum filtered onto a glass fiber filter, and
dried in air for 30 min. Beads are transferred to a 2 mL
amber vial where beads are soaked with 0.5 mL of
methylene chloride for 20 min before analysis. A 95 mL
aliquot of methylene chloride extract is injected into
GC/MS (Finnigan MAT) equipped with Large Volume
Injector (Optic2). NDMA is quantified based on the
mass detection of the characteristic molecular ion
(m=z ¼ 74:048) of NDMA and molecular ion of d6-
NDMA (m=z ¼ 80:086). The MDL at the 99% con-
fidence level was determined to be 2.4 ng/L.
2
.2. NDMA formation reactions
Experiments were conducted in batch 1 L sealed
bottles at 251C. All reaction solutions were prepared
using deionized water buffered with 1 mM bicarbonate,
and adjusted to approximately pH 7 by acid addition.
DMA was added from 0.01 to 4.0 mM at fixed
monochloramine concentration of 0.1 mM. Monochlor-
amine was added from 0.01 to 2.0 mM at fixed DMA

J. Choi, R.L. Valentine / Water Research 36 (2002) 817–824
819
3
. Results and discussion
3
.1. NDMA formation studies
Initial experiments compared NDMA formation by
the reaction of DMA with nitrite to that from the
reaction of DMA with monochloramine (Fig. 1).
NDMA was not found as a contaminant in any of the
individual reactant solutions. Approximately 12 mg/L of
NDMA was formed after 24 h from the reaction of
0
.1 mM of DMA and 0.1 mM of preformed monochlor-
amine, while about 2 mg/L of NDMA was produced by
the reaction of 0.1 mM of DMA with 0.1 mM of nitrite.
In the absence of nitrite, the formation of NDMA is
attributed to monochloramine with the reaction of
DMA. No significant amount of NDMA was produced
by the addition of HOCl to a solution of DMA,
presumably because of the absence of free ammonia.
NDMA formation continued over 40 h period reach-
ing 18 mg/L and the formation potential did not appear
to be exhausted after 40 h as shown in Fig. 2. The DPD-
FAS titration and UV absorbance at 244 nm indicated
that monochloramine concentration decreased from
Fig. 2. NDMA formation as a function of time. 0.1 mM of
DMA was reacted with 0.1 mM of monochloramine. The pH
was adjusted to 7.070.1 using 1 mM bicarbonate buffer.
Solutions were kept in the dark at 251C. Model results are
calculated NDMA concentrations based upon reactions and
rate constants shown in Table 1.
0
.1 mM to approximately 0.06 mM after 40 h. Included
on this figure and several that follow (Figs. 3–5), are
lines showing predicted NDMA concentrations based
upon the proposed reaction mechanism and kinetic
modeling to be discussed in subsequent subsections.
NDMA formation was studied as a function of
monochloramine concentration at a fixed DMA con-
centration of 0.1 mM (Fig. 3). NDMA formation
increased with increasing monochloramine concentra-
tion from 0.01 to 2 mM. The formation appeared to
reach an apparent plateau with addition of approxi-
mately 2 mM of monochloramine.
Fig. 3. NDMA formation after 24 h as a function of mono-
chloramine concentration. DMA concentration was fixed at
0
.1 mM. The pH was adjusted to 7.070.1 using 1 mM
bicarbonate buffer. Solutions were kept in the dark at 251C.
Model results are calculated NDMA concentrations based
upon reactions and rate constants shown in Table 1.
NDMA formation showed a maximum in yield when
DMA was varied (0.01–4.0 mM) at fixed monochlor-
amine concentrations of 0.1 and 0.5 mM (Figs. 4 and 5).
The maximum occurred when the ratio of DMA to
monochloramine was approximately 1.0 mM. The
amount of NDMA produced rapidly decreased as the
ratio of DMA to monochloramine was further increased
beyond 1.0 mM.
Fig. 1. NDMA formation after 24 h as a function of added
component. The concentration of each compound is 0.1 mM.
The pH was adjusted to 7.070.1 using 1 mM bicarbonate
buffer. Solutions were kept in the dark at 251C.

820
J. Choi, R.L. Valentine / Water Research 36 (2002) 817–824
Fig. 4. NDMA formation after 24 h as a function of DMA
concentration. Monochloramine concentration was fixed at
Fig. 6. NDMA formation as a function of monochloramine Cl/
N molar ratio. Both concentrations of monochloramine and
DMA are 0.1 mM. Solutions were kept in the dark at 251C. The
pH was adjusted to 7.070.1 using 1 mM bicarbonate buffer.
Line connects data points. No model results shown.
0.1 mM. The pH was adjusted to 7.070.1 using 1 mM
bicarbonate buffer. Solutions were kept in the dark at 251C.
Model results are calculated NDMA concentrations based
upon reactions and rate constants shown in Table 1.
significant as that observed when varying the ratio of
DMA to monochloramine. It should be noted that
decomposition of monochloramine occurred over this
time period (less than 30%) at pH 7. The rate of
decomposition of monochloramine tends to increase
with increasing Cl/N ratio [16].
1
5
DMA was reacted with preformed NH Cl and the
2
mass spectra of NDMA peak was examined. Fig. 7b
shows that the mass spectra of NDMA produced from
1
4
NH Cl and DMA was identical to that of commer-
2
1
4
cially obtained NDMA (Fig. 7a) made from
Fig. 7a). Both spectra show a parent molecular ion
peak for NDMA at m=z ratio of 74. However, the mass
N
(
1
5
spectra of NDMA from the reaction of NH Cl and
2
DMA (Fig. 7c) results in a parent molecular ion at m=z
ratio of 75. The one mass unit shift from 74 to 75 shows
Fig. 5. NDMA formation after 24 h as a function of DMA
concentration. Monochloramine concentration was fixed at
15
that one of nitrogen atoms in NDMA is N and must
15
be from NH
2
Cl because DMA was not labeled, i.e.
0.5 mM. The pH was adjusted to 7.070.1 using 1 mM
15
(
of 42 is due to CH =N=C H which is formed during the
CH
3 2
) N
N=O. Note that the mass peak at m=z ratio
bicarbonate buffer. Solutions were kept in the dark at 251C.
Model results are calculated NDMA concentrations based
upon reactions and rate constants shown in Table 1.
+
2
2
characteristic fragmentation of DMA [17] and this peak
1
5
2
was not affected by NH Cl. Hence, we conclude that
the formation of NDMA in this system is from the
reaction of DMA and monochloramine.
The influence of excess ammonia was investigated by
varying ammonia concentration from 1.0 to 0.14 mM
Surface and wastewaters may contain DMA or
related compounds, and thus NDMA may also be
formed by the reaction with monochloramine. This
potential was demonstrated in secondarily treated
wastewater at pH 7.2 by addition of preformed
monochloramine. Addition of 0.05 and 0.5 mM mono-
chloramine resulted in the NDMA formation of 3.6 and
111 ng/L, respectively after 24 h. None was present in the
absence of monochloramine addition.
(
Cl/N molar ratio of 0.1–0.7) (Fig. 6). Both monochlor-
amine and DMA concentrations are fixed at 0.1 mM.
This range was chosen because it is in the sub-
breakpoint region and the monochloramine is relatively
stable over the 24 h reaction time period. The amount of
NDMA formed slightly increased with decreasing
ammonia (increasing Cl/N ratio) at fixed monochlor-
amine of 0.1 mM. The influence, however, was not as

J. Choi, R.L. Valentine / Water Research 36 (2002) 817–824
821
Fig. 8. Proposed reaction scheme for NDMA formation from
DMA and monochloramine.
classical nitrosation, which requires nitrite as a nitrosat-
ing agent [7,8]. Depending on the relative stability of
nitrite in the presence of monochloramine [18], however,
both pathways may take place in parallel to form
NDMA. The proposed key reactions when preformed
monochloramine is an initial reactant are shown
schematically in Fig. 8 and listed in Table 1. Also listed
in Table 1 are rate constants used in the reaction
modeling.
The critical NDMA formation reactions consist of (i)
the formation of 1,1-dimethylhydrazine (UDMH) inter-
mediate from the reaction of DMA with monochlor-
amine followed by, (ii) the oxidation of UDMH by
monochloramine to NDMA, and (iii) the reversible
chlorine transfer reaction between monochloramine and
DMA which is parallel with (i).
By means of a modification of the Raschig synthesis,
it has been shown that monochloramine reacts
with DMA in aqueous solution to produce UDMH
(
Reaction 3 in Table 1) [21–24].
The oxidation of UDMH has been studied because it
is widely used in rocket propellant fuels. NDMA was
found when UDMH was deliberately exposed to the
atmosphere [25,26] and oxidized by sodium hypochlorite
[27]. However, we hypothesize that UDMH would be
preferentially oxidized by monochloramine as shown in
Reaction 4 in Table 1 under these reaction conditions.
Oxidation of UDMH by HOCl cannot be ruled out but
its involvement is not likely due to its extremely low
concentration determined by the equilibrium of mono-
chloramine hydrolysis in the presence of excess ammo-
nia. This is also supported by the lack of significant
effect of ammonia concentration. Increasing it by a
factor of 7 increased NDMA by only a factor of 2 at
most (Fig. 6). The oxidation of UDMH by dissolved
oxygen is also probably not significant in the presence of
monochloramine. Additional studies showed that the
amount of NDMA formed after 24 h was independent of
oxygen content. Its formation in nitrogen- and oxygen-
sparged solutions containing 0.1 mM of monochlora-
mine and 0.1 mM of DMA was essentially identical, 11.6
and 12.1 mg/L, respectively.
Fig. 7. The mass spectra of NDMA from (a) commercially
+
obtained NDMA showing a parent m=z peak (M ) at 74, (b)
14
2
the reaction of NH Cl and DMA showing a parent m=z peak
+
15
(
2
M ) at 74, (c) the reaction of NH Cl and DMA showing a
+
parent m=z peak (M ) at 75.
3
.2. Proposed reaction mechanism
We propose a new NDMA formation mechanism that
involves monochloramine that probably occurs in
chloraminated water containing DMA or other
potential precursors. This mechanism is unlike that of

822
J. Choi, R.L. Valentine / Water Research 36 (2002) 817–824
Table 1
Proposed reactions and rate constants for NDMA formation from DMA and monochloramine
Rate constant at pH 7 (251C) (M sꢀ1
ꢀ
1
)
Reference
Reaction
NH2Cl þ ðCH3Þ NH - ðCH3Þ NCl þ NH3
k1
k1 ¼ 1:4 ꢁ 10ꢀ1
(1)
(2)
(3)
(4)
[19]
[20]
2
2
k2
ꢀ3
ðCH3Þ NCl þ NH3 - NH2Cl þ ðCH Þ NH
k2 ¼ 5:83 ꢁ 10
2
3 2
NH2Cl þ ðCH3Þ NH - ðCH Þ NNH2 þ H þ Clꢀ
k3
þ
k3 ¼ 1:28 ꢁ 10
ꢀ3
This model
This model
2
3 2
ðCH3Þ NNH2 þ 2NH2Cl þ H2O - ðCH Þ NNO þ 2NH þ 2Clꢀ
k4
þ
4
k4 ¼ 1:11 ꢁ 10
ꢀ1
2
3
2
Aside from NDMA formation, the reaction of
order in both DMA and monochloramine. Presumably
any intermediates would be rapidly oxidized to NDMA.
The rate constants in Table 1 were obtained either
from literature or estimated from the data. The direct
acid catalyzed chlorine transfer rate constant (k1)
applicable at pH 7 was obtained from the literature
[19]. The rate constant characterizing chlorine transfer
from DMCA to ammonia (k2) was estimated using the
value for methylamine [20] and a correlation between
rate constants and the basicity [19,20]. The rate
constants characterizing the formation of UDMH (k3)
and its oxidation to NDMA (k4) were estimated
simultaneously by minimizing the errors between
measured and predicted NDMA concentrations on the
data sets shown in Figs. 2 and 3. Finally, the set of
differential equations was solved using Scientistt.
The model appears to adequately predict NDMA
concentrations within about 20% of the measured
values over a 40 h time period (Fig. 2). The model
predicted a plateau in NDMA formation with mono-
chloramine concentration (Fig. 3) consistent with the
expected limitation of DMA concentration due to
DMCA formation (DMA limiting). It also captured
the maximum in NDMA formation occurring near the
ratio of DMA to monochloramine of approximately 1
(Figs. 4 and 5). The maximum is hypothesized to occur
because when the DMA to monochloramine ratio
becomes approximately greater than 1, the amount of
monochloramine available to oxidize UDMH is rapidly
exhausted with increasing DMA due to chlorine transfer
(monochloramine limiting).
UDMH with monochloramine may form other products
such as tetramethyltetrazene, methylenedimethylhydra-
zine, and formaldehyde dimethylhydrazone [28]. How-
ever, due to lack of data, we have not included the
formation of these products. Including the formation of
these products may affect the apparent rate constants
characterizing UDMH formation (k3) and oxidation to
NDMA (k4) in the model (Table 1).
Instead of UDMH formation, the reaction between
monochloramine and DMA also leads to reversible
chlorine transfer from monochloramine to DMA to
form dimethylchloramine (DMCA) [19,20,29,30] as
shown in reactions (1) and (2) in Table 1. Thus, DMCA
formation by chlorine transfer would play a significant
role in the kinetics and serve to reduce the rate of
NDMA formation. The transfer of active chlorine
between monochloramine and DMA can occur by direct
chlorine transfer or by chloramine hydrolysis with
subsequent N-chlorination [20]. Since direct chlorine
transfer is subject to general acid catalysis especially in
the near neutral and acidic pH range [29,30], the rate of
chlorine transfer reaction is expected to increases as pH
decreases. In this simplified model, chlorine transfer via
chloramine hydrolysis is not considered. However, the
pathways involving chloramine hydrolysis may be
important under other reaction conditions especially
when free chlorine addition is practiced.
A reaction of DMCA with ammonia producing
UDMH, although possible [31,32], is probably not
significant under these reaction conditions. Otherwise,
NDMA formation would also continue to increase
(
instead of reaching a plateau, Fig. 3) under conditions
where essentially all of the DMA is converted to
DMCA.
4. Conclusions
NDMA can be directly formed by the reaction of
monochloramine with DMA. Given that drinking
waters and wastewaters may contain DMA or related
compounds as well as ammonia, NDMA should, there-
fore, be considered a potential disinfection by-product.
This was demonstrated by observing NDMA formation
from the addition of monochloramine to a secondarily
treated wastewater. Both this reaction and classical
nitrosation reactions are therefore potentially important
pathways for the formation of NDMA. These findings
3
.3. Kinetic modeling
The reactions shown in Table 1 and corresponding
rate expressions were used to develop the set of
differential rate expressions describing the reaction
mechanism. We assumed that reactions (1)–(3) were
elementary and could be described by simple second
order reactions. Reaction 4 is clearly not elementary. We
assumed, however, that the rate limiting reaction is first

J. Choi, R.L. Valentine / Water Research 36 (2002) 817–824
823
suggest that any test developed to ascertain the
NDMA formation potential should consider the reac-
tion with monochloramine in addition to the reaction
with nitrite. Since a variety of related nitrogenous
substances are quite common in some waters, the
formation of nitrosamines other than NDMA is also
suspected.
The model is simplified but captures several important
aspects. These include the rate limiting formation of
UDMH and its subsequent oxidation to NDMA.
Additionally, the chlorine transfer reaction and the
formation of DMCA must play a significant role in the
kinetics by consuming both monochloramine and
DMA, which would act to eventually reduce NDMA
formation. Consideration of additional reactions will be
required to provide a more detailed description of this
model system and its applicability to water and waste-
water treatment systems.
[8] Williams DHL. Nitrosation. New York: Cambridge
University Press, 1988.
[
9] Ayanaba A, Alexander M. Transformations of methyla-
mines and formation of a hazardous product, dimethylni-
trosamine in samples of treated sewage and lake water.
J Environ Qual 1974;3:83–9.
[
[
10] Graham JE, Andrews SA, Farquhar GJ, Meresz O.
Factors affecting NDMA formation during drinking water
treatment. In: Water quality technology conference.
New Orleans, LA: AWWA, 1995. p. 757–72.
11] Weerasooriya SVR, Dissanayake CB. The enhanced
formation of N-nitrosamines in fulvic-acid mediated
environment. Toxicol Environ Chem 1989;25:57–62.
[12] Sacher F, Lenz S, Brauch HJ. Analysis of primary and
secondary aliphatic amines in waste water and surface
water by gas chromatography-mass spectrometry
after derivatization with 2,4-dinitrofluorobenzene or
benzenesulfonyl chloride.
5–93.
J Chromatogr A 1997;764:
8
[
[
13] Smith TA. The occurrence, metabolism and functions of
amines in plants. Bio Rev 1971;46:201–41.
14] APHA, AWWA, WEF. Standard methods for the
examination of water and wastewater, 18th ed. Washing-
ton, DC: APHA, 1992.
Acknowledgements
The researchers are grateful for financial support for
this work by grants from the Iowa State Water
Resources Research Institute, Environmental Protection
Agency (Star grant), and the American Water Works
Association Research Foundation. We would also
like to thank Dr. Rhodes Trussell for his helpful
comments.
[
15] Taguchi VY, Jenkins SWD, Wang DT, Palmentier JFP,
Reiner EJ. Determination of N-nitrosodimethylamine by
Isotope Dilution, High-resolution Mass Spectrometry.
Can J Appl Spectrosc 1994;39:87–93.
[16] Jafvert CT, Valentine RL. Reaction scheme for the
chlorination of ammoniacal water. Environ Sci Technol
1992;26:577–86.
[
17] Libbey LM, Scanlan RA. Mass spectrometry of N-
nitrosamines. In: Karasek FW, Hutzinger O, Safe S,
editors. Mass spectrometry in environmental sciences.
New York: Plenum Press, 1985. p. 537–49.
References
[
[
[
18] Vikesland P, Ozekin K, Valentine RL. Monochloramine
decay in model and distribution system waters. Water Res
[
[
1] Loeppky RN, Micheljda CJ. Nitrosamines and related
N-Nitroso compounds: chemistry and biochemistry.
Washington, DC: ACS, 1994.
2
001;35:1766–76.
19] Isaac RA, Morris J. Transfer of active chlorine from
chloramine to nitrogenous organic compounds. 1. Ki-
netics. Environ Sci Technol 1983;17:738–42.
2] O’Neill IK, Borstel RCV, Miller CT, Long J, Bartsch H.
N-Nitroso Compounds: occurrence, biological effects and
relevance to human cancer, IARC Scientific Publication
No. 57. Oxford University Press: Lyon, 1984.
20] Yoon J, Jensen JN. Distribution of aqueous chlorine with
nitrogenous compounds: chlorine transfer from organic
Chloramines to ammonia. Environ Sci Technol
[
[
3] US EPA. N-nitrosodimethylamine CASRN 62-75-9, In-
tegrated Risk Information Service (IRIS) Substance File,
1
993;27:403–9.
1997.
[
21] Audrieth LF, Colton E, Jones MM. Formation of
hydrazine from t-butyl hypochlorite and ammonia. J Am
Chem Soc 1954;76:1428–31.
4] California Department of Health Services. California
drinking water: NDMA-related activities, 2000.
www.dhs.cahwnet.gov/org/ps/ddwem/chemicals/NDMA/
NDMAindex.htm,.
[
[
22] Cahn JW, Powell RE. The raschig synthesis of hydrazine.
J Am Chem Soc 1954;76:2565–7.
[
[
5] Mills AL, Alexander M. Factors affecting dimethylnitro-
samine formation in samples of soil and water. J Environ
Qual 1976;5:437.
23] Jones MM, Audrieth LF, Colton E. Studies on the raschig
synthesis of hydrazine: the reaction between aqueous
chloramine and ammonia solutions. J Am Chem Soc
1955;77:2701–2.
6] Jobb DB, Hunsinger RB, Meresz O, Taguchi VY. A study
of occurrence and inhibition of formation of N-nitrosodi-
methylamine (NDMA) in the Ohsweken water supply. In:
Water quality technology conference. Toronto, Ont:
AWWA, 1992. p. 103–32.
[24] Rowe RA, Audrieth LF. Preparation of some N-disub-
stituted hydrazines by reaction of chloramine with
secondary amines. J Am Chem Soc 1956;78:563–4.
[25] Lunn G, Sansone EB, Andrews AW. Aerial oxidation of
hydrazines to nitrosamines. Environ Mol Mutagenesis
1991;17:59–62.
[
7] Challis BC, Kyrtopoulos SA. Rapid formation of carcino-
genic nitrosamines in aqueous alkaline solutions. Br J
Cancer 1977;35:693–6.

824
J. Choi, R.L. Valentine / Water Research 36 (2002) 817–824
[
[
26] Lunn G, Sansone EB. Oxidation of 1,1-dimethylhydrazine
UDMH) in aqueous solution with air and hydrogen
peroxide. Chemosphere 1994;29:1577–90.
between chloramine and methylamine. Inorg Chem
1989;28:3808–13.
(
[30] Ferriol M, Gazet J, Adad MS. Kinetics and mechanisms of
chlorine transfer from chloramine to amines in aqueous
medium. Int J Chem Kinet 1991;23:315–29.
27] Castegnaro M, Brouet I, Michelon J, Lunn G, Sansone
EB. Oxidative destruction of hydrazines produces N-
nitrosamines and other mutagenic species. Am Ind Hyg
Assoc J 1986;47:360–4.
[31] Yagil G, Anbar M. The kinetics of hydrazine formation
from Chloramine and ammonia.
1962;84:1797–803.
J Am Chem Soc
[
28] Sisler HH, Kren RM, Utvary K. The reaction of
Chloramine with 1,1-Dimethylhydrazine, formation of
Tetramethyl-2-tetrazene. Inorg Chem 1969;8:2007–8.
29] Ferriol M, Gazet J, Adad MS. Chlorine transfer from
chloramine to amines in aqueous medium. 1. Reaction
[32] Randolph CL, Meyer RE. 1,1-Dimethylhydrazine.
US Patent 3050560, USA, 1962.
[