Kinetic and mechanistic aspects of the direct photodegradation of atrazine, atraton, ametryn and 2-hydroxyatrazine by 254 nm light in aqueous solution

Article abstract of DOI:10.1002/poc.624

Atrazine (1), Atraton (2) and Ametryn (3) are photodegraded upon 254 nm irradiation, yielding 2-OH-atrazine (4) as a photoproduct. Dealkylation products are also generated, and 4-ethylamino-6-isopropylamino-(1,3,5)-triazine was also found as a photoproduct of 3. The main photoreaction is proposed to be an addition-elimination, yielding 4, which subsequently photodegrades. The ease of photodegradation depends on the electron availability at position C-2, the observed order of photoreactivity being 1 > 3 > 4 > 2. Copyright

Full text of DOI:10.1002/poc.624

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2003; 16: 498–503
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.624
Kinetic and mechanistic aspects of the direct
photodegradation of atrazine, atraton, ametryn and
2-hydroxyatrazine by 254 nm light in aqueous solution
M. E. D. G. Azenha,1 H. D. Burrows,1 M. Canle L.,2* R. Coimbra,1 M. I. FernaÂndez,2 M. V. GarcõÂa,2
M. A. Peiteado2 and J. A. Santaballa2
1Departamento de QuõÂmica, Universidade de Coimbra, 3004-535 Coimbra, Portugal
2Departamento de QuõÂmica FÂõsica e EnxenÄ erõÂa QuõÂmica I, Universidade da CorunÄ a, Ru a Alejandro de la Sota 1, E-15008 A CorunÄ a, Galicia,
Spain
Received 7 October 2002; revised 18 December 2002; accepted 10 February 2003
ABSTRACT: Atrazine (1), Atraton (2) and Ametryn (3) are photodegraded upon 254 nm irradiation, yielding 2-OH-
atrazine (4) as a photoproduct. Dealkylation products are also generated, and 4-ethylamino-6-isopropylamino-(1,3,5)-
triazine was also found as a photoproduct of 3. The main photoreaction is proposed to be an addition–elimination,
yielding 4, which subsequently photodegrades. The ease of photodegradation depends on the electron availability at
position C-2, the observed order of photoreactivity being 1 > 3 > 4 > 2. Copyright  2003 John Wiley & Sons, Ltd.
KEYWORDS: dealkylation; herbicides; photodegradation; photolysis; photohydrolysis; addition–elimination;
reaction mechanism; triazines
INTRODUCTION
cyanine derivatives14 have been shown to be
particularly efficient.
There is serious environmental concern about the levels
of s-triazine herbicides in groundwater and drinking
water supplies. Of particular importance in this respect
are chlorotriazine derivatives, such as atrazine, which
have been amongst the most widely used herbicides in
Europe and the United States.1,2 These may have a
variety of toxic effects,3 and although legislation restricts
their use, their persistence in the environment requires the
development of cheap and reliable methods for their
destruction.
Various techniques have been suggested for the
elimination of atrazine and related compounds from
drinkable water, including biological degradation,4
adsorption,4 oxidation by ozone,5,6 or the Fenton
reaction,7–9 direct and catalysed photolysis,10–14 sono-
lysis12 and radiolytic decomposition.15 Amongst the
photocatalysts tested, titanium dioxide,10,11 polyoxo-
metalates,12 riboflavin13 and porphyrin and phthalo-
However, the rational design of new treatment
methodologies requires a mechanistic understanding of
the reaction pathways involved in the degradation
process.16 The direct photolysis of atrazine in water has
been studied10,17–19 and suggested to produce 2-OH-
atrazine as the major product.10 However, little informa-
tion is available on the subsequent fate of this compound,
and mechanistic details concerning the way in which
2-OH-atrazine is generated remain obscure. In this paper
we will focus on some of these aspects, reporting a study
of the direct photolysis of atrazine (1) and related com-
pounds, atraton (2), ametryn (3) and 2-hydroxyatrazine
(4), in aqueous solutions using 254 nm light (Scheme 1).
EXPERIMENTAL
Compounds 1–4 were Pestanal certified standards from
Riedel de Ha¨en. Their purity was checked by gas
chromatography, and they were used without further
purification. Organic matter-free freshly doubly distilled
water was used to make up all solutions. In all cases the
solutions were allowed to equilibrate with air at atmos-
pheric room temperature and pressure.
The pH values of the different solutions were not far
from those typical of riverine water (ca 6), due to their
typical macroscopic pK1 value, ca 5.20–22 pH measure-
ments were carried out at 298.0 K using a combined glass
*Correspondence to: M. Canle L., Departamento de Qu´ımica F´ısica e
Enxen˜er´ıa Qu´ımica I, Universidade da Corun˜a, Ru´a Alejandro de la
Sota 1, E-15008 A Corun˜a, Galicia, Spain.
E-mail: mcanle@udc.es
Contract/grant sponsor: Ministerio de Ciencia y Tecnologı´a; Contract/
grant number: PPQ 2000-0449-CO2-01; Contract/grant number:
HP1999-0078.
Contract/grant sponsor: Secretarı´a Xeral de Investigacio´n e Desen-
volvemento, Xunta de Galicia; Contract/grant number: PGIDTOO-
MAM 10303PR.
Copyright  2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 498–503

DIRECT PHOTODEGRADATION OF TRIAZINE HERBICIDES
499
constants are average values obtained from replicated
experiments, whose reproducibility was within 5%.
Samples used for product analysis were directly
extracted from the reaction mixture at different irradia-
tion times, and stored frozen at ca 253 K prior to analysis.
The samples were used directly, without any further
work-up procedure.
Product analysis were carried out by high-performance
liquid chromatography (HPLC) and gas chromatogra-
phy–mass spectrometry (GC–MS). Changes in the ClÀ
content of the photosylate were recorded for 1 using a
properly calibrated ClÀ selective electrode. In the case of
3 the photosylate was analysed for the presence of
methanethiol by derivatization with monobromobi-
mane23 and subsequent measurement of the fluorescence
emission in a single-beam, 1 cm pathlength Aminco
Bowman Series 2 luminescence spectrometer.
HPLC analyses were performed with a Waters system
equipped with a flow unit, automatic injection, column
oven with temperature control and a photodiode-array
UV–visible detector, using a detection wavelength of
220 nm. The flow-rate was 1.2 mL minÀ1 and 100 mL of
the sample were injected in all cases. The linearity of
response of the detector to all analysed products was
tested. A reversed-phase 250 mm  4.6 mm i.d. column,
packed with C-18 Intertsil (5 mm), was used. The mobile
phases consisted of acetonitrile and 5 mM acetic acid
(pH ꢀ 4.5), using the solvent gradient 0:100 (0 min),
75:25 (15 min), 100:0 (20 min), 0:100 (25 min).
GC–MS analyses were carried out using a DB-XLB
(60 m  0.25 mm i.d., 0.25 mm film thickness) capillary
column, with splitless injection and a Thermo Finnigan
Trace GC 2000 apparatus, equiped with a CTC Analitics
GCPAL injector. Mass spectrometric detection was
carried out using a Thermo Finnigan PolarisQ apparatus
equiped with an ion trap in the full-scan mode and by
selecting the different molecular ions characteristic of the
compounds under study.
The main photoproducts observed are listed in Table 1.
To complement and reinforce the experimental studies,
electronic structure calculations were carried out with the
Gaussian 98 suite of programs.24 The theoretical model,
denoted B3LYP, is based on density functional
theory,25,26 and the 6–31G(d,p) basis set was used. Full
geometry optimization was carried out in the gas phase, a
frequency calculation on each optimal geometry resulting
in no imaginary frequencies.
Scheme 1. Compounds used in this study
electrode, previously calibrated with commercial buffers
of pH 7.02 Æ 0.01 and 4.00 Æ 0.01. The accuracy
achieved in the measurement of pH was within Æ0.02
units.
Spectrophotometric measurements using a continuous
flow of solution were made using a Milton Roy Spectro-
nic 3000 double-beam array spectrophotometer. A
10 mm pathlength and 1.0 mL capacity quartz cells was
used. All other spectrophotometric measurements in
which solution flow was not needed were taken on a
double-beam Kontron Uvikon 941 Plus spectro-
photometer, using standard quartz cells with 10 mm
pathlength and 3.5 mL capacity. Both systems were
thermostated to within Æ0.1 K by water flow.
A Merck-Schuhardt TNN 15/32 irradiation system was
used for irradiation, supplying essentially monochro-
matic light at 254 nm. In order to carry out continuous
spectrophotometric monitoring of the photodegradation,
inlet and outlet valves were coupled to one of the quartz
cells, and a peristaltic pump, allowing a flow-rate
between 1 and 1000 ml hÀ1, was used to obtain a smooth,
non-turbulent flow inside the cell (ca 12 mL minÀ1). The
lamp was jacketed with a glass container and thermo-
stated to within Æ0.1 K by water flow.
All relevant data were converted into ASCII code and
analysed using ad hoc software packages. Reported rate
Table 1. Main photoproducts observed upon 254 nm steady-state irradiation
Detection method
Fluorescence
Compound
GC–MS
HPLC
1
2
3
—
—
—
MeSH
2-Hydroxyatrazine
2-Hydroxyatrazine
2-Hydroxyatrazine
Desisopropylatraton
4-ethylamine-6-isopropylamine-1,3,5-triazine
Copyright  2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 498–503

500
M. E. D. G. AZENHA ET AL.
leading to other reactions competing with photosubstitu-
tion of ClÀ. Furthermore, irradiation was seen to lead to a
decrease in the pH of the solution that was not
compensated by the buffering effect of 1 or its photo-
products.
The products of the initial stages of photolysis were
studied by HPLC. Upon 60 min of irradiation with
254 nm light of a 51 mM solution of 1 in 5% MeOH–H2O,
loss of 1 (retention time tR = 16.3 min) was found to be
accompanied by the simultaneous quantitative recovery
of a product (99%, tR = 12.7 min), that was identified as 4
by comparison with an authentic sample, in agreement
with previous reports.10,29 Kinetic analysis of these
results was complicated by consecutive photodegradation
of this photoproduct. However, the rate constant for
formation of 4 as measured by HPLC appears to be
markedly lower than the rates of photodegradation
obtained from the UV spectral changes, suggesting
parallel routes for atrazine photodegradation. In support
of this, some HPLC peaks due to minor components were
also observed. These have not yet been identified,
although the GC–MS results show no signs of de-
alkylated products.
Figure 1. Spectral changes observed at various times from
5 min to 1 h following 254 nm irradiation of 51 mM atrazine
(pH ꢀ 6.0, 5% MeOH, T = 298.0 K, 1 h of irradiation). Insets:
(i) change in absorbance observed at 223 nm and (ii) change
in [ClÀ] observed upon irradiation as a function of irradiation
time
At longer times, degradation of the major primary
photoproduct, 4, was observed by both UV–visible
spectrophotometry and HPLC. The photodegradation of
4 was also studied in separate experiments. Its absorption
spectrum, highly dependent on the acidity of the medium
(Fig. 2), showed absorption maxima at 215 nm in basic
medium (pH ꢀ 12.2, ꢀ ꢀ 45800 l molÀ1 cmÀ1), and at l
<200 nm and also 240 nm in acidic medium (pH ꢀ 2.3,
ꢀ(238 nm) ꢀ 15200 l molÀ1 cmÀ1). On this basis, the two
macroscopic pKa values observable for 4 under common
acidity conditions were obtained from the change in
absorbance at both 215 and 240 nm (Fig. 2, inset),
exhibiting values pK1 = 5.2 Æ 0.1 and pK2 = 10.8 Æ 0.4,
RESULTS
The UV absorption spectrum of 1 in aqueous solution
(pH ꢀ 6, 5% MeOH) shows absorptions at 263 and
223 nm, with molar absorption coefficients 3500 and
34400 l molÀ1 cmÀ1, in good agreement with previous
reports on this and the related compound terbuthyla-
zine.10,19 The 223 nm band has previously been assigned
to a p to p* transition.27 The nature of the longer wave-
length transition is, as yet, unclear. However, Mason has
assigned the band observed at 260 nm in sym-triazine in
water to the n to p* transition.28 Upon irradiation with
254 nm light, these bands were seen to decrease in
intensity and new absorptions grew in at 238 and ca
205 nm. In the initial stages of the reaction, good
isosbestic points were observed (Fig. 1), indicating clean
interconversion to the primary products. Upon prolonged
photolysis (up to 8 h) the absorbance at 238 nm was seen
to decrease and a less well defined spectrum was
observed.
The kinetics of the initial process were followed by
studying the decrease in the absorption of 1 at 223 nm
and the build-up of the new absorption at 238 nm. In both
cases, the kinetics fitted good first-order behaviour, with
rate constants (2.30 Æ 0.03) Â 10À3 sÀ1 (223 nm, Fig. 1,
inset i) and (2.5 Æ 0.2) Â 10À3 sÀ1 (238 nm), which are
identical within experimental error. The values are also
close to those previously reported for the direct
photolysis of aqueous solutions of 1.10 The irradiation
was also found to lead to a release of ClÀ, which was
monitored as indicated previously (Fig. 1, inset ii),
showing a slightly lower rate constant, although within
Figure 2. Spectra of 4 under different conditions of acidity.
(&) Basic medium; (~) acidic medium; (*) pH ꢀ 6. Inset:
determination of the two macroscopic pKa values; (&)
215 nm; (*) 240 nm
the same order of magnitude [(1.8 Æ 0.3) Â 10À3
s
À1].
This difference may be associated with photolysis
Copyright  2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 498–503

DIRECT PHOTODEGRADATION OF TRIAZINE HERBICIDES
501
Figure 3. Spectral changes observed at various times from
5 min to 4 h following 254 nm irradiation of 50 mM 2-OH-
atrazine (pH ꢀ 6.0, 5% MeOH, T = 298.0 K, 3 h of irradia-
tion). Inset: change in absorbance observed at 215 nm as a
function of irradiation time
Figure 4. Spectral changes observed at various times from
3 min to 4 h following 254 nm irradiation of 51 mM atraton
(pH ꢀ 6.0, 5% MeOH, T = 298.0 K, 4 h of irradiation)
in good agreement with those previously obtained using
capillary zone electrophoresis and capillary isoelectric
focusing methods.22
(Fig. 5). The photodegradation followed first-order
kinetics (Fig. 5, inset i), with a rate constant of
(2.2 Æ 0.5) Â 10À3
s
À1. HPLC studies showed that 4
Upon steady-state irradiation at 254 nm of a 50 mM
solution of 4 (pH ꢀ 6, 5% MeOH–H2O), spectral changes
were observed (Fig. 3), with the formation of new bands
at ca 204 and 238 nm. The initial spectrum at pH ꢀ 6
(Fig. 3) results clearly from combination of the absorp-
tion bands mentioned earlier (Fig. 2). As for 1, irradiation
led to a decrease in the pH of the solution that was not
compensated by the buffering effect of 4 or its photo-
products, consequently affecting the shape of the
observed spectra (Fig. 3). This notwithstanding, a photo-
degradation rate of (6.2 Æ 0.6) Â 10À4 sÀ1 was estimated
from the observed absorbance changes.
was the main photoproduct, in agreement with the
literature.29 GC–MS studies on the photolysate evidenced
the formation of 4-ethylamino-6-isopropylamino-1,3,5-
triazine. Fluorimetric measurements upon derivatization
of the photolysate with monobromobimane showed an
increase in the intensity of fluorescence during the first
30 min, followed by a slight and slower decrease (Fig. 5,
inset ii).
Table 2 summarizes the different rate constants
obtained for the 254 nm-initiated photodegradation of
triazines 1–4. No obvious linear free energy relationship
To help understand the mechanism of photodegrada-
tion of these compounds, the photodegradation of
aqueous solutions of two other (1,3,5)-s-triazine deriva-
tives, 2 and 3, was also studied. Solutions of 51 mM of 2 in
5% MeOH–H2O, pH ꢀ 6.0, show an absorption spectrum
with maximum at 220 nm (ꢀ ꢀ 33075 l molÀ1 cmÀ1), and
a weak shoulder at 250 nm (ꢀ ꢀ 2115 l molÀ1 cmÀ1).
Upon irradiation, a decrease in the absorbance at 220 nm
was observed, with corresponding increases in absor-
bances at ca 250 nm and <200 nm (Fig. 4). From the
decrease in absorbance at 220 nm, a rate constant of
(2.57 Æ 0.04) Â 10À5 sÀ1 was determined. HPLC studies
showed that the 2-hydroxy derivative was one of the
photoproducts, in agreement with the literature.29 In
addition, GC–MS experiments on the photolysate showed
that desisopropylatraton was formed by dealkylation.
Solutions of 51 mM of 3 in 5% MeOH–H2O, pH ꢀ 6.1,
show an absorption maximum at 222 nm (ꢀ ꢀ 36575
l molÀ1 cmÀ1) and a weaker band at 270 nm (ꢀ ꢀ 3918
l molÀ1 cmÀ1). Upon irradiation a decrease in absorbance
at 222 nm was observed to be accompanied by the
formation of a new absorption at shorter wavelengths
Figure 5. Spectral changes observed at various times from
5 min to 5 h following 254 nm irradiation of 51 mM 3
(pH ꢀ 6.10, 5% MeOH, T = 298.0 K, 5 h of irradiation).
Insets: (i) change in absorbance observed at 220 nm and (ii)
change in the intensity of ¯uorescence emission at 480 nm
upon derivatization of the photolysate with monobromobi-
mane (lexc = 394 nm) as a function of irradiation time
Copyright  2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 498–503

502
M. E. D. G. AZENHA ET AL.
Table 2. Rate constants obtained spectrophotometrically for
the photodegradation of triazines 1±4
Compound
k (sÀ1
)
(2.3 Æ 0.03) Â 10À3 a
(2.5 Æ 0.2) Â 10À3 b
(2.57 Æ 0.04) Â À103À5
(2.2 Æ 0.5) Â 10
1
2
3
4
(6.2 Æ 0.6) Â 10À4
a
Detection at l = 223 nm.
Detection at l = 238 nm.
b
Scheme 2. Mechanism of photodegradation proposed for
compounds 1 and 2
was found between these rate constants and the sub-
stituent inductive or steric effects. A study of the
photodegradation with a wider range of 1,3,5-triazine
herbicides is planned in which both kinetic and quantum
yield data will be measured to check on the existence of
any useful structure-reactivity correlations.
Theoretical calculations carried out at the B3LYP/6–
31G(d,p) level allowed us to analyse the Mulliken
charges at position 2 of the different triazines, evidencing
in all cases an electronic deficiency in this position (Table
3).
known to have very high oxidation potentials.32 Further-
more, homolytic dissociation is unlikely to be an
important mechanism in the case of triazines, since
2-hydroxy products are observed on photolysis of a wide
variety of 1,3,5-triazines having very different dissocia-
tion energies for the 2-substituent bonds.21,33 We there-
fore believe that the main mechanism involves addition–
elimination. Support for this comes from charge calcula-
tions. Table 3 shows that all 1–4 are electronically
defficient at C-2 which is in agreement with the addition
of water/HOÀ at such a position. Moreover, the electron
deficiency is less important for 2 and 4, in line with the
fact that the photodegration is slower for these deriva-
tives, as evidenced by the observed rate constants (Table
2).
DISCUSSION
The photodegradation of four related 1,3,5-triazines was
studied in aqueous solutions using 254 nm light. Two
main degradation pathways are observed, photosubstitu-
tion at the 2-position, leading generally to the 2-hydroxy
derivative, and dealkylation on the amine side-chains.
Photohydroxylation at the 2-position has also been
observed with various other 1,3,5-triazines,30 and seems
to be a common reaction for this class of compounds.
Three mechanisms can in principle be considered for this:
(i) homolytic dissociation of the carbon–substituent bond
at the 2-position, followed by reaction of the triazinium/
triazinyl radical with water or HOÀ, (ii) addition of water/
HOÀ followed by elimination of the other substituent or
(iii) a charge-transfer process involving the formation of
the triazinium radical cation, followed by nucleophilic
attack of water/HOÀ. Intermediacy of charge-transfer
species has been suggested in the photosubstitution of 1-
nitronaphthalene by hydroxide ion.31 However, the
mechanism is unlikely in this case since triazines are
Therefore, the behaviour observed for 1 and 2 can be
rationalized in terms of the mechanism shown in Scheme
2, which would explain the increase in [ClÀ] for 1, and
also as the decrease in the pH of the medium observed for
all 1–4.
In the case of 3, an initial increase in the concentration
of thiol (MeSH) was observed, followed by a loss of
MeSH. This could be due to the photolysis of MeSH to
Á
yield MeS , with subsequent reduction of the thiol
concentration. Furthermore, the detection of 4-ethylami-
no-6-isopropylamino-(1,3,5)-triazine indicates homoly-
sis of the C—S bond followed by H abstraction by the
resulting triazinyl radical. This observation is in agree-
ment with previous findings for 2-methylthio-(1,3,5)-
triazines.27,34 All the observations for 3 can be rational-
ized in terms of the mechanism shown in Scheme 3.
Despite the fact that 4 is the one of the main photo-
products of photodegradation of 1–3, the same being
valid for analogous (1,3,5)-triazine herbicides with
different substituents,16 no detailed product analyses are
available for its photodegradation. From the observations
for 1–3, dealkylation in the lateral chain could be
expected.
Table 3. Calculated Mulliken charges at position 2[B3LYP/
6±31G(d,p)]
Compound
Charge at position 2
1
2
3
4
0.266696
0.685101
0.272177
0.656304
For this mechanistic route, dealkylation of the amino
groups, the mechanism is less clear. Theoretical calcula-
tions show that the lone electron pairs on the side-chain
nitrogen atoms are included in the electron delocalization
Copyright  2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 498–503

DIRECT PHOTODEGRADATION OF TRIAZINE HERBICIDES
503
REFERENCES
1. The Agrochemical Handbook (3rd edn). Royal Society of
Chemistry: Cambridge, 1994.
2. Hileman B. Chem. Eng. News 1996: 23.
3. Belluck DA, Benjamin SL, Dawson T. In Pesticide Transforma-
tion Products. Fate and Significance in the Environment,
Somasundaram L, Coats JR (Eds). American Chemical Society:
Washington, DC, 1991.
4. Bonnet MC, Welte B, Montiel A, Dore M. Environ. Technol. 1991;
12: 217.
5. de Laat J, Dore M, Suty H. Rev. Sci. Eau 1995; 8: 23.
6. Acero JL, Stemmler K, von Gunten U. Environ. Sci. Technol.
2000; 34: 591.
7. Arnold SM, Hickey WJ, Harris RF. Environ. Sci. Technol. 1995;
29: 2083.
8. Larson RA, Schlauch MB, Marley KA. J. Agric. Food. Chem.
1991; 39: 2057.
9. Balmer ML, Sulzberger B. Environ. Sci. Technol. 1999; 33: 2418.
10. He´quet V, Gonza´lez C, Le Cloirec P. Water Res. 2001; 35: 4253.
11. Pelizzetti W, Carlin V, Minero C, Gra¨tzel M. New J. Chem. 1991;
15: 351.
12. Hiskia A, Ecke M, Troupis A, Kokorakis A, Henig H,
Papaconstantinou E. Environ. Sci. Technol. 2001; 35: 2358.
13. Cui H, Hwang H-M, Zeng K, Glover H, Yu H, Liu Y.
Chemosphere 2002; 991.
Scheme 3. Mechanism of photodegradation proposed for
compound 3
of the heterocyclic ring, thus facilitating a potential
dealkylation of the alkyl side-chains. Therefore, it is
possible that dealkylation is a relatively minor route with
this excitation source (254 nm), resulting from some
dissociative pathway ocurring at higher excitation
energies. Joint experimental and theoretical studies are
in progress in order to compare the effects of different
excitation wavelengths to test this and to analyse in detail
the reaction mechanism.
14. He´quet V, Cloirec PL, Gonza´lez C, Meunier B. Chemosphere
2000; 41: 379.
15. Angelini G, Bucci R, Carnevaletti F, Colosimo M. Radiat. Phys.
Chem. 2000; 59: 303.
16. Burrows H, Canle LM, Santaballa JA, Steenken S. J. Photochem.
Photobiol. B 2002; 67: 71.
17. Schmidt S, Mattusch J, Werner G. Z. Chem. 1989; 29: 239.
18. Torrents A, Anderson BG, Bilboulian S, Johnson WE, Hapeman
CJ. Environ. Sci. Technol. 1997; 31: 1476.
19. Sanlaville Y, Schmitt P, Mansour M, Meallier P. Fresenius’
Environ. Bull. 1997; 6: 418.
CONCLUSION
20. Dean JR, Wade G, Barnabas IJ. J. Chromatogr. A 1996; 733: 295.
21. Lide DR (Ed). CRC Handbook of Chemistry and Physics: a Ready-
reference Book of Chemical and Physical Data. CRC Press: Boca
Raton, FL, 2000.
Compounds 1–3 undergo 254 nm-initiated photodegra-
dation, leading to 4 as one of the main photoproducts, as
shown by HPLC. Dealkylation products were also
observed by GC–MS in the case of 2, and 4-ethyl-
amino-6-isopropylamino-(1,3,5)-triazine in the case of 3.
An addition–elimination mechanism is proposed for 1–3,
leading to 4, which would subsequently undergo photo-
degradation. The observed order of reactivity upon
irradiation is 1 > 3 > 4 > 2, in accordance with the
electron availability at position C-2. No linear free
energy relationship was found between the observed rate
constants and substituents inductive or steric effect.
22. Schmitt P, Poiger T, Simon R, Freitag D, Kettrup A, Garrison AW.
Anal. Chem. 1997; 69: 2559.
23. Fahey RC, Newton GL. Methods Enzymol. 1987; 143: 85.
24. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA,
Cheeseman JR, Zakrzewski VG, Montgomery JA, Stratmann RE,
Burant JC, Dapprich S, Millam JM, Daniels AD, Kudin KN, Strain
MC, Farkas O, Tomasi J, Barone V, Cossi M, Cammi R, Mennucci
B, Pomelli C, Adamo C, Clifford S, Ochterski J, Petersson GA,
Ayala PY, Cui Q, Morokuma K, Malick DK, Rabuck AD,
Raghavachari K, Foresman JB, Cioslowski J, Ortiz JV, Stefanov
BB, Liu G, Liashenko A, Piskorz P, Komaroni I, Gomperts R,
Martin RL, Fox DJ, Keith T, Al-Laham MA, Peng CY,
Nanayakkara A, Gonza´lez C, Challacombe M, Gill PMW, Johnson
BG, Chen W, Wong MW, Andre´s JL, Head-Gordon M, Replogle
ES, Pople JA. Gaussian 98. Gaussian: Pittsburgh, PA, 1998.
25. Lee C, Yang W, Parr RG. Phys. Rev. B 1988; 37: 785.
26. Becke AD. J. Chem. Phys. 1993; 98: 5648.
Acknowledgements
27. Pape BE, Zabik J. J. Agric. Food. Chem. 1970; 18: 202.
28. Mason SF. J. Chem. Soc. 1959; 1240.
29. Marcheterre L, Choudhry GG, Webster GRB. Environ. Contam.
Toxicol. 1988; 103: 61.
30. Ferna´ndez Pe´rez MI. PhD Thesis, Departamento de Qu´ımica F´ısica
e Enxen˜er´ıa Qu´ımica I, Universidade da Corun˜a, 2002.
31. Martins LJA, Kemp TJ, Formosinho SJ, Branco JS, Fernandes
MM. J. Chem. Soc., Faraday Trans. 1991; 87: 3617.
32. Azenha MEDG, Burrows HD, Canle LM, Coimbra R, Ferna´ndez
MI, Rodr´ıgues AE, Santaballa JA, Steenken S. Chem. Commun.
2003; 112.
R.C. acknowledges an EU Socrates–Erasmus grant
received to stay in A Corun˜a. M.I.F. and M.A.P. worked
under contract within research projects PPQ2000-0449-
C02-01 and the Acciones Integradas bilateral program
HP1999-0078 of the Ministerio de Ciencia y Tecnologı´a
(Spain) and PGIDTOOMAM10303PR of the Secretarı´a
Xeral de Investigacio´n e Desenvolvemento, Xunta de
Galicia (Spain), respectively. We also acknowledge the
Centro de Supercomputacio´n de Galicia for providing
computing facilities and Dr. F. Ferna´ndez for technical
support.
33. Kerr JA. Chem. Rev. 1966; 66: 465.
34. Barcelo´ D, Durand G, de Bertrand N, Albaige´s J. Sci. Total
Environ. 1993; 132: 283.
Copyright  2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 498–503