Identification of disinfection by-products of selected triazines in drinking water by LC-Q-ToF-MS/MS and evaluation of their toxicity

Article abstract of DOI:10.1002/jms.1509

During the development of an on-line solid phase extraction-liquid chromatography-ultraviolet detection (SPE-LC-UV) analytical method for determination of eight selected triazines; ametryn, atrazine, cyanazine, metrybuzine, prometryn, propazin, simazine, and terbutryn, in drinking water, it was observed that the retention times of three of them (ametryn, prometryn, and terbutryn) in Milli-Q water were different from those in chlorinated Milli-Q water, indicating the formation of new products. The cause of this change was found in the oxidation of the molecules as a result of chlorination with sodium hypochlorite. Experiments performed at varying concentrations of triazines and hypochlorite showed that the extent of the reaction depended on their relative concentrations. At the maximum admissible level of 100 ng/l for individual pesticides in drinking water, no apparent transformation was observed in the absence or at low concentrations (0.05 mg/l) of hypochlorite; however, on increasing the concentration of hypochlorite to the level typically present in drinking water (0.9 mg/l) the transformation was complete. The reaction is quite fast; within 1 h the parent compound is completely degraded and after 22 h the concentrations of the by-products are constant. Investigation of the by-products by ultra performance liquid chromatography-quadrupole-time of flight- tandem mass spectrometry (UPLC-Q-ToF-MS/MS) has shown that all three triazines follow a similar transformation pathway, forming four new molecules whose structure have been elucidated. The acute toxicity of the new products was investigated using a standard method based on the bioluminescence inhibition of Vibrio fischeri, and the by-products showed a higher toxicity than that of the parent compounds. Copyright

Full text of DOI:10.1002/jms.1509

Research Article
Received: 23 June 2008
Accepted: 30 August 2008
Published online in Wiley Interscience: 25 November 2008
(www.interscience.com) DOI 10.1002/jms.1509
Identification of disinfection by-products
of selected triazines in drinking water by
LC-Q-ToF-MS/MS and evaluation of their
toxicity
a
b
a∗
a
´
Rikke Brix, Neus Bahi, Maria J. Lopez de Alda, Marinella Farre,
c
a,d
`
´
Josep-Maria Fernandez and Damia Barcelo
Duringthedevelopmentofanon-linesolidphaseextraction-liquidchromatography-ultravioletdetection(SPE-LC-UV)analytical
methodfordeterminationofeightselectedtriazines;ametryn,atrazine,cyanazine,metrybuzine,prometryn,propazin,simazine,
and terbutryn, in drinking water, it was observed that the retention times of three of them (ametryn, prometryn, and terbutryn)
in Milli-Q water were different from those in chlorinated Milli-Q water, indicating the formation of new products. The cause
of this change was found in the oxidation of the molecules as a result of chlorination with sodium hypochlorite. Experiments
performed at varying concentrations of triazines and hypochlorite showed that the extent of the reaction depended on their
relative concentrations. At the maximum admissible level of 100 ng/l for individual pesticides in drinking water, no apparent
transformation was observed in the absence or at low concentrations (0.05 mg/l) of hypochlorite; however, on increasing the
concentration of hypochlorite to the level typically present in drinking water (0.9 mg/l) the transformation was complete.
The reaction is quite fast; within 1 h the parent compound is completely degraded and after 22 h the concentrations of the
by-products are constant. Investigation of the by-products by ultra performance liquid chromatography-quadrupole-time
of flight- tandem mass spectrometry (UPLC-Q-ToF-MS/MS) has shown that all three triazines follow a similar transformation
pathway, forming four new molecules whose structure have been elucidated. The acute toxicity of the new products was
investigated using a standard method based on the bioluminescence inhibition of Vibrio fischeri, and the by-products showed
c
a higher toxicity than that of the parent compounds. Copyright ꢀ 2008 John Wiley & Sons, Ltd.
Keywords: triazines; disinfection by-products; UPLC-QToF; Vibrio fischeri; chlorination; identification
Introduction
chlorinated, s-triazines do not react with chlorine, the non-
chlorinated s-triazines react very quickly with this compound.[9]
One of the major concerns regarding the disinfection by-
products (DBPs) from chlorination is that NaClO is known to
produce genotoxic DBPs[10] and can thus increase the acute
toxicity of these practically nontoxic s-triazines. Also, the major
chlorinated degradation products of Atrazine, Simazine, and
Propazine are considered to be endocrine-disrupting chemicals
by the U.S. Environmental Agency, and thus, monitoring these
compounds is compulsory.[11] This raises the question of the
Triazines are, because of their high use and physical-chemical
characteristics which allow for mobility and moderate persistence,
among the most frequently reported pesticides in ground water.[1]
Even though newer pesticides (including triazines) are relatively
labile and as such do not persist long in the environment, there
is such a widespread use that it is expected that there will always
be some level of exposure.[2] The lability of the pesticides also
makes it very important to monitor their transformation products
as these can at times be present in concentrations higher than that
of the parent compound.[3] This fact has been recognized by the
European Union (EU) and the World Health Organization (WHO),
who have both included specific transformation/degradation
products in their guidelines for drinking water quality.[4–6]
∗
Correspondence to: Maria J. Lopez de Alda, Department of Environmental
Chemistry, IIQAB-CSIC, C/ Jordi Girona 18–26, 08034 Barcelona, Spain. E-
mail: mlaqam@cid.csic.es
a
Department of Environmental Chemistry, IIQAB-CSIC, C/Jordi Girona 18-26,
08034 Barcelona, Spain
Owing to their labile nature, there is also a risk of transformation
during the disinfection process of drinking water. Sodium
hypochlorite (NaClO) is one of the most used disinfectants
worldwide,[7] underlining the importance of identifying possible
degradation products resulting from this procedure. In the
disinfection process, hypochlorous acid (HClO) is the main
responsible reagent for pathogen destruction, but both HClO and
ClO− react with organic compounds giving addition, substitution,
or oxidation products.[8] A further interesting fact is that while
`
b
CTM Centre Tecnologic, Area de Tecnologia Ambiental (ATA), Av. Bases de
Manresa n◦ 1, 08242 Manresa, Spain
c
SIA Enginyers, Monturiol 16, baixos, 08018 Barcelona, Spain
`
d
Catalan Institute for Water Research (ICRA), Parc Científic i Tecnologic de la
Universitat de Girona, Edifici Jaume Casademont, C/Pic de Peguera, 15, 17003
Girona, Spain
c
J. Mass. Spectrom. 2009, 44, 330–337
Copyright ꢀ 2008 John Wiley & Sons, Ltd.

Identification of DBPs of selected triazines
Name
Prometryn
Terbutryn
Ametryn
H
H
H
S
N
N
S
N
N
N
S
N
Structure
N
N
N
N
N
N
NH
NH
NH
C10H19SN5
241.1361
886-50-0
Empirical formula
Molecular weight
CAS no.
C10H19SN5
241.1361
7287-19-6
C9H17SN5
227.1205
834-12-8
Figure 1. Structures of the studied compounds.
possible endocrine-disrupting properties of DBPs of the remaining
triazines, thus making their identification even more vital.
The current paper deals with the transformation of three
triazines (prometryn, terbutryn, and ametryn) as a result of the
chlorination of drinking water. Out of the eight tested compounds
(ametryn, atrazine, cyanazine, metrybuzine, prometryn, propazin,
simazine, and terbutryn), only these three were degraded in the
presence of hypochlorite. Figure 1 shows their chemical structure,
empirical formula, molecular weight, and chemical abstract
number (CAS) number. It has been investigated at which level of
chlorination,thedegradationoccursandthedegradationproducts
havebeenidentifiedbyultraperformanceliquidchromatography-
quadrupole-time of flight- tandem mass spectrometry (UPLC-Q-
ToF-MS/MS). Furthermore, the acute toxicity of the degradation
products has been determined by the bioluminescence inhibition
of the marine bacteria Vibrio fischeri.
On-line SPE-LC-UV analyses
Fully automated on-line solid phase extraction (SPE) was per-
formed with a Symbiosis Environ system from Spark Holland
(Emmen, The Netherlands) consisting of a Symbiosis Environ (P-2)
unitequippedwithanautosamplerEndurance,ahighpressuredis-
penser(HPD)using amixmodeoptionandanAutomaticCartridge
Exchange (ACE). Aqueous standards (50 ml) were preconcentrated
(at 5 ml/min) on polymeric cartridges PLRP-s (12.5 mg crosslinked
styrene-divinylbenzene polymer, 15–25 µm particle size, from
SparkHolland),previouslyconditionedwith2 mlacetonitrile(ACN)
and 2 ml water (flow rate 5 ml/min). After washing with 1 ml water
(5 ml/min), elution to the chromatographic system was carried out
with the liquid chromatography (LC) mobile phase. The chromato-
graphic system consisted of an Agilent 1100 LC pump (Palo Alto,
CA, USA) connected in series with an Agilent 1100 photodiode
array (UV-DAD) detector (Milford, MA).
Separation of the triazines was achieved on a Mediterranean
Sea C18 column (150 × 4.6 mm, 5 µm) from Teknokroma (Sant
Cugat del Valle´s, Spain) using acetonitrile/water at 1 ml/min as
the mobile phase. The gradient started with 35% ACN, which was
held isocratic for 18 min, increased to 80% in 22 min, was held
under these conditions for 5 min, returned back to the starting
conditions over 5 min, and followed by 5 min equilibration. The
last triazine is eluted at 36.6 min and corresponds to Terbutryn.
Total analysis time was 55 min and the detection was performed
at 220 nm.
Experimental
Reagents and solutions
Acetonitrile (ACN) and extra pure water was purchased from
Scharlau Chemie SA (Spain), and were filtered through 0.45 µm
filters before being used. Methanol high performance liquid
chromatography (HPLC) gradient grade 240 nm far UV, NaClO
solution 15% w/v extra pure and sodium sulfite, extra pure, Ph Eur
BP E221 were all purchased from Scharlau Chemie SA (Spain).
For the liquid chromatography-quadrupole-time of flight- tan-
dem mass spectrometry(LC-Q-ToF-MS/MS) studies, water (CHRO-
MASOLV Plus for HPLC) was from Sigma–Aldrich (Madrid,
Spain) and acetonitrile (CHROMASOLV LC-MS grade) was from
Riedel–de–Ha¨en (Seelze, Germany).
High purity standards of ametryne, prometryn, and terbutryn
were purchased from Riedel–de–Hae¨n (Seelze, Germany) and
SIGMA–ALDRICH Laborchemikalien GmbH (Seelze, Germany).
Stock standard solutions (1000 mg/l) for each triazine were
prepared in methanol and stored at 4 ◦C in the dark. Working
solutions of the individual standards at 1 mg/l and mixtures of all
of them were prepared in the range 50–2000 ng/l by appropriate
dilution of the stock standard solutions.
UPLC-Q-ToF-MS/MS analysis
UPLC was performed on a Waters Acquity UPLC system (Waters
Corp., Mildford, MA) equipped with a binary solvent delivery
system, an autosampler and a UV detector. Chromatographic
separation was performed on a 50 × 2.1 mm Waters Acquity
C18 1.7 µm column. The injection volume was 10 µl and the
flow rate was 0.35 ml/min. Triazines were analyzed using positive
ionization. The elution was performed by an ACN/water gradient.
The gradient was linearly increased from 10 to 90% ACN in 10 min,
then increased to 95% over 2 min. Total run time, including re-
equilibration of the column to the initial conditions, was 16 min.
Mass spectrometry was performed on a Q-ToF-Micro (Waters
Corp.). The nebulization gas (nitrogen) was set to 600 l/h at a
temperatureof350 ◦C;theconegas(nitrogen)wassetto50l/h,and
the source temperature to 120 ◦C. The capillary and cone voltages
were set to 3000 and 30 V, respectively. For MS experiments, the
Q-ToF-Micro instrument was operated in a wide pass quadrupole
Solutions for the identification study of the transformation
products were made by mixing 0.8 ml triazine in water (10 mg/l)
with 0.2 ml hypochlorite solution (60 mg/l); obtaining final
concentrations of 8 mg/l triazine and 12 mg/l hypochlorite.
Samples were left at room temperature for the duration of each
experiment.
c
J. Mass. Spectrom. 2009, 44, 330–337
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/jms

R. Brix et al.
Figure 2. On-line SPE-LC-UV analysis of aqueous standard mixtures of the eight triazines (2 µg/l) prepared in (a) Milli-Q water, (b) tap water, (c) Milli-Q
water spiked with 1 mg/l hypochlorite, and (d) tap water previously spiked with sodium sulphite.
mode with the time of flight (TOF) data being collected between
m/z 50 and 1000 (PI) and low collision energy of 10 eV. Data
were collected in the centroid mode, with a scan accumulation
time of 1 s. The MS/MS experiments were performed at variable
collision energies (10–40 eV). All analyses were performed using
an independent reference spray (lockSpray) to ensure accuracy
and reproducibility. Val-Tyr-Val was used as the lock mass (m/z
380.2029) at a flow rate of 10 µl/min. The lockSpray frequency was
set at 11 s, meaning that every 11 s the flow from the lockspray
was introduced into the mass spectrometer for 1 s, thus giving the
c
www.interscience.wiley.com/journal/jms
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
J. Mass. Spectrom. 2009, 44, 330–337

Identification of DBPs of selected triazines
Prometryn-8ppm-hipo-0hrs-40
4.09
100
Prometryn
8.00
100
90
6.07
0
80
70
60
50
40
30
20
10
0
2.00
4.00
6.00
Prometryn-8ppm-hipo-1hrs-40
4.11
100
0
4.89
2.00
4.00
6.00
8.00
8.00
8.00
8.00
8.00
Prometryn-8ppm-hipo-5hrs-40
0.9 mg/l HClO
0.48 mg/l HClO
0.12 mg/l HClO
0.06 mg/l HClO
0 mg/l HClO
4.11
100
0
4.95
4.92
5.18
Ametryn
Prometryn
5.53
3.39
Terbutryn
2.00
4.00
6.00
Prometryn-8ppm-hipo-22hrs-40
Figure 3. Formation of disinfection by-products of ametryn, prometryn,
and terbutryn as a function of the concentration of hypochlorite.
4.08
100
5.21
3.41
software the possibility to perform ongoing correction of the exact
mass of the analyte. Data were averaged over 10 spectra/min. The
Masslynx software (Waters Corp.) has a feature that calculates all
possible elemental compositions from the accurate mass. When
using previous knowledge and basic organic chemical rules to
eliminate nonfeasible formulae, this is a strong tool for forming
hypotheses about the identity of an unknown compound. The
final identification can then be performed based on the accurate
mass measurements of the parent ions and fragments obtained in
MS/MS experiments.
0
2.00
4.00
6.00
6.00
6.00
Prometryn-8ppm-hipo-50hrs-40
4.09
100
4.95
5.18
3.39
0
2.00
4.00
Prometryn-8ppm-hipo-100hrs-40
4.09
100
4.97
Acute toxicity
5.18
3.39
0
For the acute toxicity evaluation, a method based on the
bioluminescence inhibition of the marine bacteria V. fischeri,
NRR-11177 was applied. This marine bacterium naturally emits
light, thanks to the action of the enzyme bacterial-luciferase. The
bioluminescence is directly proportional to the metabolic status of
the cell. A toxic substance will cause changes in the cellular state,
which may be on different levels – cell wall, cell membrane, the
electron transport chain, enzymes, cytoplasmatic constituent, but
in all these cases changes are rapidly reflected in a decrease of the
light production that can be easily measured by a photomultiplier
in a luminometer. This work was carried out using the ToxAlert100
(Merck).
2.00
4.00
Figure 4. Kinetics of formation of prometryn by-products in the presence
of hypochlorite.
Results and Discussion
Transformation of triazines in the presence of chlorine
Figure 2 shows the on-line solid phase extraction-liquid
chromatography-ultraviolet detection (SPE-LC-UV) analysis of
aqueous standard mixtures of the eight triazines (2 µg/l) pre-
pared in Milli-Q water, tap water, Milli-Q water spiked with 1 mg/l
hypochlorite, and tap water previously spiked with sodium sul-
fite (12 mg/l, in order to block the hypochlorite). As it can be
seen, the peaks corresponding to three of the eight target tri-
azines, namely, ametryn (RT 30.95 min), prometryn (35.62 min),
and terbutryn (36.44 min), which are visible in Milli-Q water, dis-
appear in the chlorinated solutions (tap water and Milli-Q water
added with HClO), but remain the same in the solution pre-
pared in tap water previously treated with sodium sulfite (to
block any residual hypochlorite). The disappearance of these
three peaks goes hand-in-hand with the appearance of three
new peaks at shorter retention times (7.49, 11.97, and 13.04 min)
which correspond to the formation of three main transformation
products.
The percentage of inhibition (%I) is determined by compar-
ing the response given by a saline control solution to that
corresponding to the sample.
%I = [1 − (dilution light/control light)] × 100
Measuring the inhibition values in a wide number of sample
concentrations it is possible, by fitting an inhibition curve, to
calculate the concentration where 50% of the specie is affected by
this substance (EC50).
Main advantages of this method are rapid, robust, cost effective,
and standard responses, and it has been applied in different works
for the characterization of organic pollutants,[12,13] degradation
products, and mixtures.[14]
c
J. Mass. Spectrom. 2009, 44, 330–337
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/jms

R. Brix et al.
212.1508
100
4.15
100
170.1027
213.0952
128.0570
234.0993
235.0414
171.0104
171.9828
128.9423
214.0988
210.9979
256.0975
250.0204
278.0493
268.0991
82.0216 86.0365
123.0507
139.0164
194.7845
191.0609
97.9969
155.7958 164.9721
224.2336
2
69.0540
112.9678
0
70
80
90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280
100
280.1194
4.96
281.0419
282.1026
258.1389
265.0996
272.0856
240.1265
212.1405
199.0598
355.1210
283.0912
243.1160
302.1054
314.0905 320.0779
196.1341
231.0672
361.0871
183.0637
170.1043
152.0967 156.0459
216.1004
336.0958
139.0191
0
140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360
296.1140
100
274.1345
297.0799
298.1026
275.0990
276.1182
312.0883
314.0735
299.1155
311.0248
322.8605
324.8557
326.8597
253.9883256.1054
264.9867 268.8996
280.1212 283.8280
288.0298
295.0690
302.8369
318.0998
262.0487
271.1222
334.0675
n
0
255
260
265
270
275
280
285
290
295
300
305
310
315
320
325
330
335
314.0829
100
3.45
5.23
0.35
278.1041
263.0804
316.0813
264.0767
265.0602
0
272.0349
268.8984
324.8524
322.8595
274.0746
280.1155
305.1111
317.0379
318.0806
0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00 7.50 8.00 8.50 9.00 9.50 10.0010.5011.0011.50
299.0355
296.1058
308.9998
292.0938
283.8384 290.8988
326.8529
33
252.0344
256.0574
332.8627
280.9890
262.0408
301.0049
258.9041
310.8725
320.0185
0
255
260
265
270
275
280
285
290
295
300
305
310
315
320
325
330
Figure 5. Chromatogram and MS spectra of degradation products of prometryn.
The peak appearing at 32.7 min in the chromatograms in
Fig. 2 (b), (c) and (d) is due to a previous contamination of the
HPLCequipmentandevenoccurredinchromatogramsofstandard
solutions prepared in Milli-Q water.
Time development of the formation of disinfection by-
products
To study the time development of the formation of DBPs
and simultaneously identify their structure, 12 mg/l hypochlorite
was added to the samples with 8 mg/l triazine in water and
these samples were injected into the UPLC-Q-ToF-MS/MS system
at different time intervals during 1 week. Figure 4 shows the
development of the by-products of prometryn.
Determination of the effect of the concentration of hypochlo-
rite
For a further investigation of the effect of chlorination, standard
mixtures of the triazines (10 mg/l) were prepared with 0.9 and
12 mg/l hypochlorite (usual concentrations in drinking water are
between0.5and1 mg/l). Resultsshowedthatametryn, prometryn,
and terbutryn (10 ppm) were only degraded in the samples with
12-mg/l hypochlorite. It was suspected that the reason for the
apparent lack of degradation with 0.9-mg/l hypochlorite was
because of the comparatively high concentration of triazines.
To test this hypothesis, solutions containing 100 ng/l of
each triazine were prepared in different hypochlorite solutions
with concentrations ranging from 0 to 0.9 mg/l. The results
(Fig. 3) confirmed that without the presence of hypochlorite, no
degradation takes place and that degradation increases with the
hypochlorite concentration and at 0.9 mg/l of hypochlorite, the
triazines are fully degraded.
From Fig. 4 it can be seen that prometryn is fully degraded
within 1 h leading to the formation of one by-product. It can also
be noted that after 5 h, the presence of a second new product can
be detected. Furthermore, after 22 h a third and a fourth product
can be observed, but with much lower responses. There is no
significant change after 22 h. An identical pattern was observed
for terbutryn and ametryn with the only difference being that the
second, third and fourth by-products developed to a lesser extent
for ametryn.
The relatively fast reaction times imply that the reaction
products are likely to be present at the time that the drinking
water arrives to the consumer as the average residence time of
the drinking water in the distribution net has been reported to be
36–48 h.[15]
c
www.interscience.wiley.com/journal/jms
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
J. Mass. Spectrom. 2009, 44, 330–337

Identification of DBPs of selected triazines
NH
NH
N
N
N
N
N
+
C
H N OS + Na
10 19 5
O
S
C
N
H
NH
+
Mw = 280.3497
NH
S
+
Na
+
H
N S
5
10 20
Degradation
+
C H N OS + Na
NH
9
16 5
Mw = 242.3685
Fractionation
Mw = 265.3149
N
N
O
S
N
NH
+
+
C
H
N O S + Na
Na
10 19
5 2
Mw = 296.1157
Degradation
NH
N
Degradation
N
N
NH
N
O
+
C
H N OSCl + Na
10 18 5
S
NH
Mw = 314.0836
+
Na
N
N
O
O
N
S
ClH C
2
+
H
Na
Degradation
Fractionation
Fractionation
H
N
HN
N
NH
NH
N
NH+
N
N
N
N
O
O
Fractionation
OH
C H N O
Mw = 212.15
N
H
N
S
N
H
N
S
9
18
5
+
+
+
Na
Na
+
C
H N OS + Na
10 17 5
C H N OS + Na
9
14 5
Fractionation
Mw = 263.0817
Mw = 278.1052
HN
N
NH
2
H N
2
N
NH
2
N
+
NH+
N
NH
OH
OH
C H N O
C H N O Mw = 170.1042
6
12 5
Mw = 128.0572
3
6 5
Figure 6. Reaction pathway of prometryn.
Identification of disinfection by-products
marked with a star in Fig. 6. The added sodium is a typical result of
positive ionization using an electrospray.
The specification of the Q-ToF-Micro states a maximum mass
error of 5 ppm for the range of masses discussed in this paper.
If we calculate the number of possible elemental compositions
valid for the mass 280.1194 with a window of 5 ppm, without
any restrictions on the type or number of atoms, we get more
than 130 possible elemental compositions. However, if the type
and number of atoms is restricted by the information from the
In Fig. 5, an enlarged chromatogram of the degradation products
of prometryn is shown along with the MS spectra of the four peaks.
It can be seen that the two major products formed have masses
of 280.1194 Da and 296.1140 Da, respectively and the two minor
products have masses of 212.1508 Da and 314.0829 Da.
The mass 280.1194 Da could correspond to an oxygenation of
the sulphur, resulting in the following ion: [C10H19N5OS + Na]+,
c
J. Mass. Spectrom. 2009, 44, 330–337
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/jms

R. Brix et al.
4.62
100
300
300.0904
100
3.49
302.0892
264.0993
249.0840
216.0975 241.9072
128.9642
139.0272
303.0072330.9568340.9066
4.21
4.00
374.8993402.8564
413.3096
98.0031
443.9989
171.9745
0
123.0502
0
1.00
1.00
1.00
1.00
2.00
2.00
2.00
2.00
3.00
5.00
5.00
5.00
5.00
6.00
6.00
6.00
6.00
7.00
7.00
7.00
7.00
8.00
8.00
8.00
8.00
9.00 10.00 11.00 12.00
4.23
100
282
100
282.1234
283.1006
298.0943
3.51
198.1490
260.1327
241.9720
139.0240
128.9630
404.0264
334.8843140.9484
374.8983
413.2887
0
212.1514
170.1149
452.0364
98.0030
0
3.00
4.00
3.49
9.00 10.00 11.00 12.00
100
266
266.1116
100
267.1216
341.1146
96.0562114.0711
0
251.0918
226.1222044.1275
300.0924
355.0547
384.6649
446.0628
422.0324528.0522
128.9588
176.0271882.1444
0
3.00
2.75
4.00
9.00 10.00 11.00 12.00
100
198
100
198.1523
156.0994
3.49
199.1492
220.1327
4.25
128.9644 157.0921
252.1093
295.0405
278.0495
86.0424
114.0732
429.9883
448.8697
334.9679
340.9607
390.0268
0
171.9758
0
3.00
4.00
9.00 10.00 11.00 12.00
100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440
m/z
Time
Figure 7. Ion extracted chromatograms and MS spectra of degradation products of ametryn in the presense of hypochlorite.
spectra (e.g. no halogens) and the possible degradation scenarios
under the experimental conditions (i.e. maximum three oxygen
atoms, ten carbon atoms, five nitrogen atoms and no phosphor
or fluorine), the calculation leads to only one possible elemental
composition. This fact alone can be used as a confirmation of the
suggested elemental composition, but in the current work it has
been sought to always have a second fact for final confirmation.
In order to verify the hypothesis that the mass 280.1194 Da
corresponds to the ion: [C10H19N5OS + Na]+, an MS/MS run
was performed. In this run, the quadrupole was filtering ions,
permitting only the ion with mass 280 Da to pass to the collision
cell where 20 eV was applied. In this run, the molecule fractured
with a neutral loss of 15 Da. This could correspond to a loss of
a methyl unit. Literature confirms that this fractionation pattern
strongly indicates an oxygenation of the sulphur.[16]
The mass 296.1140 Da could correspond to a further oxygena-
tion of the sulphur, resulting in the following ion: [C10H19N5O2S +
Na]+ (marked with a square in Fig. 6). This hypothesis could not
be confirmed by MS/MS as it was not possible to fractionate the
molecule. But the hypothesis is supported by the fact that the
mass 274.1345 Da was found at a low abundance. This mass cor-
responds to the following ion: [C10H20N5O2S]+ which is the same
molecule as above, but ionized with hydrogen instead of sodium,
thus confirming the hypothesis.
The fourth transformation product has the mass 314.0829 Da.
Again it can be seen from the mass spectra (Fig. 5, bottom right)
that the molecule has been fragmented already in the cone.
It should also be noted that the characteristic pattern for the
presence of chloro is present for the parent compound, but not
for the fragments. The parent compound of this spectra is most
likely another degradation product of the 280 Da sulfoxide. The
degradation lead to a substitution of a hydrogen atom for a chloro
atom; resulting in the molecule C10H18N5OSCl (marked with a
diamond in Fig. 6). This thesis is supported by the presence of
fragments with masses 278.1041 Da (loss of HCl) and 263.0804 Da
(loss of HCl and CH3).
For terbutryn, the same transformation products are seen as for
prometryn, with the same masses, as terbutryn has the same mass
as prometryn. Ametryn is undergoing the same transformations,
but to a much lower extent (see Fig. 7).
The formation of the first three degradation products can be
confirmedbytheliterature,[9,17,18] however,thefourthdegradation
product has previously not been mentioned in the literature.
Toxicology
The V. fischeri toxicity studies were carried out with 30 min
exposure time. It was found that EC50 values were very high
(18–26 mg/l - see table 1), indicating very low acute toxicity which
is in correspondence with a previous literature.[19,20]
After treatment with hypochlorite, the acute toxicity increased
considerably (e.g. prometryn becomes five times more toxic), but
EC50s still remained rather high (5.2–15 mg/l) and were still not
considered toxic. However, these results indicate that in future
works, the synergistic effects with other compounds present in
the same samples should be studied.
The mass 212.1508 Da is likely a further degradation of the
sulfone above (296 Da) to an alcohol, C9H17N5O (marked with a
triangle in Fig. 6). In this case, there is no need to perform MS/MS
to confirm the structure since this molecule is already fragmented
by the cone voltage, as can be seen in the mass spectra shown
in Fig. 5. The fragmentation corresponds to the loss of first one
C3H6 group resulting in the ion [C6H12N5O]+, mass 170.1027 Da
and later two C3H6 groups leading to the ion [C3H6N5O]+, mass
128.0570 (see Fig. 6).
c
www.interscience.wiley.com/journal/jms
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
J. Mass. Spectrom. 2009, 44, 330–337

Identification of DBPs of selected triazines
liquid chromatography with diode array detection, International
Journal of Environmental Analytical Chemistry 1996, 65(1–4), 53.
[4] Council of the European Communities, Directive 98/83/EC, Official
Journal of the European Communities 1998, L 330, 32.
[5] Council of the European Communities, Directive 2000/60/EC,
Official Journal of the European Communities 2000, L 327, 1.
Table 1. Cell toxicity of the three triazines before and after
treatment with hypochlorite
EC50 in
EC50 in H2O
Compound
Milli-Q + 2% NaCl
hypochlorite + 2% NaCl
[6] World Health Organisation (WHO), Guidelines for drinking-
water quality, third edition, incorporating first addendum,
Annex 4 – Chemical summary tables, http://www.who.int/
water sanitation health/dwq/en/gdwq3 ann4tab.pdf May 2008.
[7] G. Baxter 1994, Chlorine disinfection – the industry standard. Dis-
Ametryn
Prometryn
t-Butryn
18 1 mg/l
26 1 mg/l
24 1 mg/l
9.2 3 mg/l
5.2 2 mg/l
15 3 mg/l
infection of Potable Water – International Specialised Conference,
Kruger National Park, South Africa.
Conclusions
[8] C. G. Zambonin, I. Losito F. Palmisano, Liquid chromatogra-
phy/electrospray ionisation sequential mass spectrometric
identification of the main chlortoluron by-products during
water disinfection using chlorine, Rapid Communications in Mass
Spectrometry 2000, 14(10), 824.
[9] A. Lopez, G. Mascolo, G. Tiravanti, M. Santori R. Passino, Ox-
idation of sulfur-containing s-triazines during groundwater
hypochlorination, Water Science and Technology 1994, 30, 53.
[10] L. Guzzella, S. Monarca, C. Zani, D. Feretti, I. Zerbini, A. Buschini,
P. Poli, C. Rossi S. D. Richardson, In vitro potential genotoxic
effects of surface drinking water treated with chlorine and
alternative disinfectants, Mutation Research – Genetic Toxicology
and Environmental Mutagenesis 2004, 564(2), 179.
[11] H. Jiang, C. Adams, N. Graziano, A. Roberson, M. McGuire D. Khiari,
Occurrence and Removal of Chloro-s-Triazines in Water Treatment
Plants, Environ. Sci. Technol. 2006, 40(11), 3609.
[12] M. Farre D. Barcelo, Toxicity testing of wastewater and sewage
sludgebybiosensors,bioassaysandchemicalanalysis,TrAC – Trends
in Analytical Chemistry 2003, 22(5), 299.
[13] M. Farre, C. Goncalves, S. Lacorte, D. Barcelo M. F. Alpendurada,
Pesticide toxicity assessment using an electrochemical biosensor
with Pseudomonas putida and a bioluminescence inhibition assay
with Vibrio fischeri, Analytical and Bioanalytical Chemistry 2002,
373(8), 696.
[14] M. Farre, D. Asperger, L. Kantiani, S. Gonzalez, M. Petrovic
D. Barcelo, Assessment of the acute toxicity of triclosan and methyl
triclosan in wastewater based on the bioluminescence inhibition of
Vibrio fischeri, Analytical and Bioanalytical Chemistry 2008, 390(8),
1999.
Three triazines, i.e. ametryne, prometryne, and terbutryn, out of
eight tested, have been shown to react with hypochlorite in water
and form four main DBPs, which in theV.fischeri toxicity studies are
moretoxicthantheparentcompounds.Attypicalhypochloriteand
pesticides concentrations (0.5–1 mg/l and 100 ng/l, respectively)
in drinking water the transformation of the parent compounds
into the corresponding by-products is complete. In addition, the
reactionisquitefast;within1 htheparentcompoundiscompletely
degraded and after 22 h the concentrations of by-products are
constant, which means that the reaction products are likely to be
presentatthetimethatthedrinkingwaterarrivestotheconsumer.
The four main degradation products have been identified with
UPLC-Q-ToF-MS/MS of which only three have previously been
described in the literature. Because some s-triazines are suspected
endocrine-disrupting compounds and, further, that DBPs formed
during water chlorination are often genotoxic, the identified by-
products should be subjected to a more profound investigation
in terms of toxicity to evaluate potential risks of exposure through
chlorinated drinking water.
Acknowledgements
This work has been supported by the EU Project MODELKEY
[GOCE 511237] and by the Spanish Ministry of Science and
Innovation (CTM2005-24255-E, BIO2005-00840, CTM2006-26227-
E/TECNO, and CGL2007-64551/HID) and reflects the author’s views
only. The EU is not liable for any use that may be made of the
information contained in it. Waters (Cerdanyola del Valle`s, Spain)
is acknowledged for the gift of LC columns.
[15] S. Eichler, R. Christen, C. Holtje, P. Westphal, J. Botel, I. Brettar,
A. Mehling M. G. Hofle, Composition and dynamics of bacterial
communities of a drinking water supply system as assessed by
RNA- and DNA-based 16S rRNA gene fingerprinting, Applied and
Environmental Microbiology 2006, 72(3), 1858.
[16] P. Wright, A. Alex, D. Gibson, R. Jones P. Macrae, Characterisation of
sulphoxidesbyatmosphericpressureionisationmassspectrometry,
Rapid Communications in Mass Spectrometry 2005, 19(14), 2006.
[17] A. Lopez, G. Mascolo, G. Tiravanti R. Passino 1997, Degradation of
herbicides (ametryn and isoproturon) during water disinfection by
means of two oxidants (hypochlorine and chlorine dioxide). Water
Science and Technology. 35, 129.
References
[18] G. Mascolo, A. Lopez, R. Passino, G. Ricco G. Tiravanti, Degradation
of sulphur containing s-triazines during water chlorination, Water
Research 1994, 28(12), 2499.
[19] M. J. Ruiz, L. Lopez-Jaramillo, M. J. Redondo G. Font, Toxicity
assessment of pesticides using the Microtox test: Application to
environmental samples, Bulletin of Environmental Contamination
and Toxicology 1997, 59(4), 619.
[20] C. Gaggi, G. Sbrilli, A. M. Hasab El Naby, M. Bucci, M. Duccini
E. Bacci, Toxicity and hazard ranking of s-triazine herbicides using
Microtox, two green algal species and a marine crustacean,
Environmental Toxicology and Chemistry 1995, 14(6), 1065.
[1] R. F. Spalding, M. E. Exner, D. D. Snow, D. A. Cassada, M. E. Burbach
S. J. Monson, Herbicides in Ground Water beneath Nebraska’s Man-
agement Systems Evaluation Area, J Environ Qual 2003, 32(1), 92.
[2] D. B. Barr, J. R. Barr, V. L. Maggio, R. D. Whitehead, M. A. Sadowski,
R. M. Whyatt L. L. Needham,
A multi-analyte method for the
quantification of contemporary pesticides in human serum and
plasma using high-resolution mass spectrometry, Journal of
Chromatography B 2002, 778(1–2), 99.
[3] S. Dupas, S. Guenu, V. Pichon, A. Montiel, B. Welte M. C. Hennion,
Long-term monitoring of pesticides and polar transformation prod-
uctsingroundwaterusingautomatedon-linetrace-enrichmentand
c
J. Mass. Spectrom. 2009, 44, 330–337
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/jms