Phototransformation of propiconazole in aqueous media

Article abstract of DOI:10.1021/jf010253r

The photolysis of propiconazole in pure water, in water containing humic substances, and in natural water was investigated. The reaction rates were determined, and the main photoproducts were identified with the help of HPLC-mass spectrometry and by NMR.

Full text of DOI:10.1021/jf010253r

J. Agric. Food Chem. 2001, 49, 5377−5382
5377
P h ototr a n sfor m a tion of P r op icon a zole In Aqu eou s Med ia
Delphine Vialaton,† J ean-Franc¸ois Pilichowski,† Daniela Baglio,‡ Ana Paya-Perez,‡
Bo Larsen,‡ and Claire Richard*,†
Laboratoire de Photochimie Mole´culaire et Macromole´culaire, UMR CNRS nï. 6505,
Ensemble Universitaire des Ce´zeaux, F-63177 Aubie`re CEDEX, France, and J oint Research Center,
European Commission, I-21020 Ispra (VA), Italy
The photolysis of propiconazole in pure water, in water containing humic substances, and in natural
water was investigated. The reaction rates were determined, and the main photoproducts were
identified with the help of HPLC-mass spectrometry and by NMR. The quantum yield for direct
photolysis was 0.11 ( 0.01 at the maximum of absorption (269 nm). Photocyclization after HCl
elimination and photohydrolysis of the cyclized intermediate were the main reaction pathways at
254 nm. By contrast, oxidation prevailed over dechlorination in simulated or natural solar light.
Humic substances (10 mg‚L-1) and naturally occurring chromophores contained in natural water
enhanced the rate of propiconazole photodegradation in solar light. Half-life in J une in Clermont-
Ferrand (latitude 46°N) was found to be 85 ( 10 h in pure water and 60 ( 10 h in natural water;
showing that photodegradation of propiconazole in natural waters involves both direct photolysis
and photoinduced reactions.
Keyw or d s: Propiconazole; direct photolysis; photoinduced transformation; humic substances;
natural water; solar light
INTRODUCTION
relative importance of the two pathways depends on the
pollutant structure. To get a better insight into the
mechanism of propiconazole phototransformation, we
tried to obtain relevant information on both photoprod-
ucts and photochemical kinetics under various experi-
mental conditions. In a first step, we studied the direct
photolysis of propiconazole using 254-nm radiation,
artificial light simulating the solar light, and natural
solar light. Identification of the main photodegradation
products was based on 1H NMR and HPLC-mass
spectrometric (MS) analyses. Preliminary and detailed
NMR studies (1H and 13C) on propiconazole itself were
necessary. In a second step, we evaluated the ability of
propiconazole to undergo humic-substances-mediated
phototransformation when exposed to artificial or natu-
ral solar light. Finally, propiconazole was irradiated in
natural water with solar light as close to natural
conditions as possible.
The presence of pollutants in surface waters poses
health and environmental risks. The European Union
has listed the persistent and potentially toxic com-
pounds for which more data are needed regarding their
fate in the environment. Propiconazole (()-1-[2-(2,4-
dichlorophenyl)-4-propyl-1,3-dioxolan-2-ylmethyl]-1H-
1,2,4 triazole figures in this priority list (EEC/414/91).
It is a systemic foliar fungicide with a broad range of
activity. The toxicity is moderate: the acute oral LD50
is around 1500 mg/kg in rats. There is no report of long-
term human exposure to propiconazole. This compound,
however, has been classified as a possible human
carcinogen (group C) by the U.S. Environmental Protec-
tion Agency (EPA, document no. 96-29020, 1996), on the
basis of data from rodent studies.
Little is known on the fate of this compound in the
environment. According to the literature (1), the com-
pound’s half-life in aerobic soils is 40-70 days, showing
that intrinsic biodegradation processes are relatively
slow. In surface waters, the half-life is shorter, suggest-
ing that photochemical degradation might occur.
MATERIALS AND METHODS
Ch em ica ls a n d Wa ter Sa m p les. Propiconazole (98,4%
purity) was purchased from Riedel de Hae¨n-Fluka (Saint
Quentin Fallavier, France) and used as received. It was a
mixture of four stereoisomers because of the presence of two
asymmetric carbons, i.e. two couples of enantiomers (RR+SS)
and (RS+SR). Humic acids (technical grade) were supplied by
Aldrich. Water was purified using a Milli-Q device (Millipore).
Natural water was collected in J une 1999 from Barrage de
Villerest (Loire, France) using inactinic bottles. The pH was
8.5 and the dissolved organic carbon (DOC) concentration was
7.2 mg‚L-1. Samples were stored at 4 °C and filtered through
0.45-µm filters (cellulose acetate, Sartorius) before use. Solu-
tions containing humic substances were adjusted to pH 7.0
by using phosphate buffers.
The phototransformation of a pollutant in surface
water may result from light absorption by the pollutant
itself (direct photolysis) or may be photoinduced by the
dissolved natural organic matter (NOM) or nitrate ions
present in water, as these chromophores are known to
photoproduce reactive species (indirect photolysis) (2-
8). Numerous studies have been reported in the litera-
ture (see for example refs 9-16) showing that the
* To whom correspondence should be addressed. Tel. +33-
(0)473407142. Fax: +33(0)473407700. E-mail: claire.richard@
univ-bpclermont.fr.
P h otolyses. Monochromatic Irradiations. Aqueous solu-
tions of propiconazole (9.0 × 10-5 - 3.0 × 10-4 mol‚L-1) were
irradiated at 254 nm using a device equipped with up to six
germicidal tubes (Mazda TG 15W) and a quartz cylindrical
† Ensemble Universitaire des Ce´zeaux.
‡ European Commission.
10.1021/jf010253r CCC: $20.00 © 2001 American Chemical Society
Published on Web 10/13/2001

5378 J. Agric. Food Chem., Vol. 49, No. 11, 2001
Vialaton et al.
reactor (14-mm i.d.). Quantum yield of photolysis was mea-
sured at 270 ( 3 nm using a 900-W xenon lamp equipped with
a monochromator Spex 1681. Potassium ferrioxalate was used
as a chemical actinometer.
Polychromatic Irradiations. Aqueous solutions of pure prop-
iconazole (1.5 × 10-5 mol‚L-1) and mixtures of propiconazole
(1.5 × 10-5 mol‚L-1) and humic substances (10 mg‚L-1) were
irradiated using a “solarbox” (Cofranegra, Italy) equipped with
a xenon lamp and a special glass filter restricting the trans-
mission of wavelengths below 290 nm. The spectral distribu-
tion of this device was similar to that of solar light, and the
average irradiation intensity was 750-800 W/m2. The tem-
perature was set at 25 °C. The solutions were irradiated for
60 h in a 130-mL Pyrex glass cylindrical reactor sealed with
caps. Aliquots of 1 mL were removed at selected intervals and
analyzed by HPLC-MS.
Solutions of propiconazole (1.5 × 10-5 mol‚L-1) in pure
water, in neutral water containing humic substances added
at a level of 10 mg‚L-1, and in natural water were filled into
Pyrex glass reactors, attached on a rack inclined by about 15°
from horizontal, and exposed to solar light between J une 2nd
and 8th, 1999 at Clermont-Ferrand (46° N, 3° E). At selected
intervals, aliquots of 0.5 mL were removed from the three
solutions and analyzed by HPLC-UV.
An a lytica l P r oced u r es. Losses of propiconazole were
monitored by HPLC-UV using a Waters apparatus equipped
with two pumps (model 510), an autosampler, a photodiode
array detector (model 996), and a C18 reversed-phase column
(4.6 mm × 250 mm, Spherisorb S5 ODS2, Waters). The eluent
was a mixture of water acidified with 0.1% of orthophosphoric
acid (A) and acetonitrile (B) at a constant flow of 1 mL‚min-1
.
The following gradient was used: mixture of 50% A and 50%
B kept constant for 5 min, linear increase of B content to 90%
in 5 min, then decrease of B to 50% in 5 min, and mixture of
50% A and 50% B kept constant till the end of the analysis
(20 min).
F igu r e 1. 1H NMR data of propiconazole.
trum was necessary to distinguish between the two couples of
resonances of the triazole ring protons (the results are sum-
marized in Figure 1). For 13C peaks assignments, J -modulated,
distortionless enhancement by polarization transfer (DEPT),
heteronuclear correlation (HETCOR), and heteronuclear mul-
tiple bond correlation (HMBC) were used to attribute all the
resonances (Table 1). To be noticed are the particular proper-
ties of the H-C bonds of the triazole ring: undecoupled 1H/
13C correlation allowed measurement of the coupling constants.
1J (CH)3 and 1J (CH)5 are very high: 212 and 207 Hz, respec-
tively. These unusual values (17) fit well with the fact that in
the conventional J -modulated spectrum, C3 and C5 atoms
resonances occur on the unexpected “wrong” side. Moreover,
chemical shifts and coupling constants deduced from spectra
were checked to fit well with those given in the literature (18).
Oth er An a lytica l Meth od s. UV spectra were recorded on
a Cary 3 Varian spectrophotometer. DOC measurements were
performed by Institut Louise Blanquet (Clermont-Ferrand,
France) using a standard titration method.
An a lysis of P h otop r od u cts. Irradiation of propiconazole
in the different experimental conditions led to the formation
of five detectable photoproducts. Products named 1 and 2 could
be isolated and analyzed by 1H NMR (Table 2) and HPLC-
APCI-MS (Table 3). Products 3, 4, and 5, produced in smaller
amounts, were analyzed by HPLC-APCI-MS (Table 3) through
injection of crude reactional mixtures. In addition, 5 could be
produced independently by hydrolysis. The detailed procedures
are described below.
The mass spectrometric analyses of propiconazole and its
photoproducts were performed by HPLC-MS using a Thermo
Separation Products series gradient pump equipped with a 4.6
mm (i. d.) × 250 mm Alltech column packed with 5-µm Alltima
C18 reversed-phase material, and thermostated at 20 °C. A
mixture of water (C) and methanol (D) was used as mobile
phase at a constant flow rate of 1 mL/min. The gradient
program was the following: constant mixture of 50% C and
50% D for 5 min, then linear increase of D content to 90% in
10 min, and a mixture of 10% C and 90% D kept constant till
the end of the analysis (25 min). The Finnigan MAT LCQ mass
spectrometer was equipped with a standard atmospheric
pressure chemical ionization (APCI) interface operating in
positive ion mode. The parameter optimization was performed
using the infusion technique at 25 µL/min with a solution of
propiconazole in water (1.5 × 10-5 mol‚L-1). The mass selected
in this operation was m/z ) 342 corresponding to [M+H]+. The
following optimized conditions were obtained: APCI vaporizer
temperature 450 °C; capillary temperature 150 °C; source
voltage 6 kV; capillary voltage 25 V; source current 5 µA;
sheath gas flow (N2) 80 arbitrary units; auxiliary gas flow (He)
10 arbitrary units. The MS analyses were carried out in full
scan mode, scanning the range from m/z ) 60 to m/z ) 450 in
1.1 s. Disappearance of propiconazole and formation of pho-
toproducts were measured by HPLC-APCI-MS using inte-
grated selected ion monitoring (SIM) signals corresponding to
[M+H]+ at the detected compounds.
NMR Sp ectr oscop y. Proton and carbon NMR spectra were
collected on both 400 MHz AC 400 and 300 MHz DSX 300
Bruker spectrometers, depending on the program used. The
NMR analysis of propiconazole was made difficult by the
presence of the two couples of enantiomers (RR+SS) and the
assignment of the resonances derived from the combination
of several experiments. Concerning proton spectra, the main
difficulty was to attribute the resonances of the 1,3-dioxolane
group. Phase sensitive COSY-45 (correlation spectroscopy)
sequence confirmed the multiple double irradiation experi-
ments in the knowledge of connections in each stereoisomer.
NOESY (nuclear Overhauser enhancement spectroscopy) spec-
Com p ou n d 1. A 100-mL solution of propiconazole (1.0 ×
10-4 mol‚L-1) in acetonitrile was irradiated at 254 nm for 2
min in the device equipped with six tubes, yielding 1 as the
single detectable photoproduct. Acetonitrile was evaporated
until the sample volume was reduced to 1 mL. Compound 1
was separated from propiconazole by HPLC fractionation using
a Beckmann HPLC equipped with a UV detector and a C18
reversed-phase column (4.6 mm × 250 mm, RP18 5-µm
Lichrospher 100, Merck) by using isocratic conditions (35%
water/65% acetonitrile) at a constant flow rate of 1.5 mL‚min-1
.
Retention time in condition of HPLC-APCI-MS analysis: 15
min; λmax ) 218 and 260 nm.

Photolysis of Propiconazole in Aqueous Media
J. Agric. Food Chem., Vol. 49, No. 11, 2001 5379
F igu r e 2. HPLC-APCI-MS chromatograms of aqueous propiconazole (1.5 × 10-5 mol‚L-1) irradiated (a) at 254 nm, and (b) in
simulated solar light.
Ta ble 1. 13C Ch em ica l Sh ifts of P r op icon a zole (CDCl3 a s In ter n a l Refer en ce: δ ) 77 p p m , d ow n field fr om TMS;
Ca lcu la ted Reson a n ces Wer e Obta in ed fr om Ch em Win d ow Softw a r e)
Ta ble 2. Differ en ces in 1H NMR Sign a ls betw een
P r op icon a zole a n d Com p ou n d s 1 a n d 2
min in the device equipped with six tubes gave m/z ) 256,
258, and 260. The same product was obtained on acidic
hydrolysis of propiconazole using HClO4 at a level of 0.5 mL
in 5 mL of acetonitrile. Retention time in condition of HPLC-
APCI-MS analysis: 11.7 min; λmax ) 252 nm with shoulder
at 290 nm.
δ ppm (multiplicity)
propiconazole
1
2
H8a; H8b 3.17 (t); 3.90 (t) 3.75 (t); 4.30 (t) between 1 and 1.5
H9
3.92 (t)
4.25 (t)
between 1 and 1.5
-
H6a; H6b 4.71 (d); 4.79 (d) 4.40 (d); 4.50 (d)
RESULTS
H20
H5
7.54 (d)
8.27 (s)
7.87 (s)
8.10 (d)
none
7.95 (s)
8.10 (d)
none
8.00 (s)
3.70 (s)a
5.10 (m)a
Dir ect P h otolysis a t Sh or t Wa velen gth . Propi-
conazole shows a maximum of absorption at 269 nm
(ꢀ ) 200 mol-1‚L‚cm-1). The quantum yield of photolysis
at this wavelength was found to be 0.11 ( 0.01 in
aerated medium.
H3
CH3
CHn-OH
-
-
-
-
a
New resonances.
Com p ou n d 2. A 200-mL solution of propiconazole (4.5 ×
10-5 mol‚L-1) in pure water was irradiated at 254 nm for 2
min in the device equipped with six tubes. Photoproducts were
extracted with three 5-mL portions of dichloromethane. Then,
the solvent was removed by evaporation to dryness. The
resulting mixture was dissolved in 1 mL of a water-acetoni-
trile (50/50, v/v) mixture. Compound 2 was isolated by HPLC
fractionation in the same conditions as 1. Retention time in
condition of HPLC-APCI-MS analysis: 12 min; λmax ) 240
nm.
Com p ou n d s 3 a n d 4. A 100-mL aliquot of an aqueous
solution containing propiconazole (8.0 × 10-5 mol‚L-1) and
K2S2O8 (1.0 × 10-3 mol‚L-1) was irradiated at 254 nm in the
device equipped with one tube. HPLC-APCI-MS analysis
indicated that 3 and 4 each have several isomers and show
isotopic clusters at m/z ) 356, 358, and 360 for the former,
and m/z ) 358, 360, and 362 for the latter. In both cases,
isomers were eluted within the range 12.4-14.2 min.
Com p ou n d 5. HPLC-APCI-MS analysis of a solution of
propiconazole (4.5 × 10-5 mol‚L-1) irradiated at 254 nm for 2
Figure 2 shows the typical HPLC-APCI-MS chro-
matogram of a solution of propiconazole irradiated at
254 nm for 2 min. The two main photoproducts, com-
1
pounds 1 and 2, could be identified on the basis of H
NMR (Table 2) and HPLC-APCI-MS (Table 3) data.
Proposed structures are given in Figure 3. Mass spec-
trometric analyses proved that these compounds are
monochlorinated through the presence of M and M+2
in the ratio 1:3. The molecular ions obtained correspond
to the loss of HCl in the former case and to the loss of
HCl plus addition of H2O in the latter. Unlike propi-
conazole that loses triazole by MSn collision-induced
dissociation (CID), compound 1 loses the aliphatic chain
and part of dioxolane, and compound 2 loses first H2O
and then the dehydrated alkyl chain R. These results
indicate a higher degree of bonding of the triazole-ring
in the photoproduct than in propiconazole. The 1H NMR
spectrum of compound 1 indicates that the missing

5380 J. Agric. Food Chem., Vol. 49, No. 11, 2001
Vialaton et al.
Ta ble 3. HP LC-AP CI-MS Da ta of P r op icon a zole a n d P h otop r od u cts in P ositive Mod e Ion (MS-MS F r a gm en ts
Obta in ed on th e Ita licized Molecu la r Ion )
molecular ions m/z
major fragments (relative abundance)
propiconazole
342 [M+H]+-344-346
306 [M+H]+-308
273 (24); 237 (48); 205 (54); 159 (100)
220 (100)
1
2
3
4
5
324 [M+H]+-326
306 (100); 238 (84)
256 (80); 159 (4)
356 [M+H]+-358-360
358 [M+H]+-360-362
256 [M+H]+-258-260
187 (100); 159 (22)
F igu r e 3. Structure of identified photoproducts.
F igu r e 4. Phototransformation of propiconazole (1.5 × 10-5
mol‚l-1) in simulated solar light. Plot of ln(c/c0) vs the irradia-
tion time: b, pure water; O, water containing humic acids (10
mg/L).
proton is the homologous of H5 of triazole, as a conse-
quence of a six-membered new ring formation. All other
signals fit well with the proposed cyclic structure. The
1H NMR spectrum of compound 2 is greatly simplified
compared to that of propiconazole: signals from diox-
olane ring are not observed, and the typical double AB
spectrum from the (CH2sN) groups (one for each
isomer) is also missing. On the other hand, new signals
appear: a singlet at 3.70 ppm corresponding to three
equivalent protons and a multiplet at 5.10 ppm. The
singlet is attributed to the methyl group linked to
triazole, and the multiplet is attributed to a proton
geminated with a hydroxyl group in the alkyl chain R.
Product 5 was identified to the ketone (Figure 3) on
the basis of several results. By MSn CID fragmentation,
this compound loses first the triazole ring, just as
propiconazole, and then CO. Moreover, its UV spectrum
is very similar to that of 2,4-dichloroacetophenone (λmax
) 252 nm with shoulder at 290 nm). This assignment
was confirmed by a complementary experiment: the
acidic hydrolysis of propiconazole, that is known to
remove 1,3-dioxolane used as a carbonyl protecting
group producing therefore the ketone, also yielded 5.
The two other photoproducts named 3 and 4 have the
particularity to show several isomers by HPLC-APCI-
MS eluted at retention times lying between 12.4 and
14.2 min (Figure 2). They have isotopic clusters at m/z
) 356, 358, and 360 for the former and m/z ) 358, 360,
and 362 for the latter and are therefore dichlorinated
photoproducts. The molecular mass of 3 [M+14+H]+
may correspond to the addition of an oxygen atom with
loss of two protons, whereas that of 4 [M+16+H]+
corresponds to the addition of an oxygen atom only. It
can be proposed that 3 and 4 are oxidation photoprod-
ucts: 3 might be a ketone and 4 might be an alcohol.
P h ototr a n sfor m a tion in Sim u la ted or Na tu r a l
Sola r Ligh t. Results are shown in Table 4.
Ta ble 4. Kin etic P a r a m eter s Mea su r ed in Sim u la ted a n d
Na tu r a l Sola r Ligh t
first-order rate constant
measured in
half-life
measured in
solar light (h)
conditions
pure water
humic substances
(10 mg‚L-1
natural water
simulated solar light (h-1
)
5 × 10-3
85 ( 10
73 ( 10
7 × 10-3
)
60 ( 10
propiconazole (1.5 × 10-5 mol‚L-1) in simulated solar
light led to a decrease in concentration equal to 30%
after 60 h. The plot of ln(c/c0), where c represents the
concentration of propiconazole at time t and c0 repre-
sents the initial concentration, against the irradiation
time was linear in accordance with a first-order kinetics
(Figure 4). The apparent first-order rate constant was
equal to 5.0 × 10-3 h-1
.
The distribution of photoproducts was different from
that obtained at a shorter wavelength. In particular, the
formation of photoproducts 1 and 2 was found to be
drastically reduced, while that of products 3 and 4 was
favored. Comparing the peak area obtained by integrat-
ing HPLC-APCI-MS signals in the SIM mode enabled
us to get ratios between the area measured after
irradiation in simulated solar light and the area mea-
sured after irradiation at 254 nm. We found 0.07 for
compound 1, 0.02 for compound 2, and 2.19 for com-
pounds 3 and 4. As seen on the chromatogram b of
Figure 2, the major photoproducts in solar light are
compounds 3 and 4, whereas compounds 1 and 2 prevail
at 254 nm (chromatogram a of Figure 2). The kinetics
of photoproducts formation was monitored by HPLC-
APCI-MS. From the shape of the formation curves
(Figure 5), it can be deduced that compounds 1, 3, and
4 are primary photoproducts, whereas compound 2 is a
secondary photoproduct.
In Pure Water. The absorption of propiconazole at
wavelengths longer than 290 nm is very weak. Photo-
transformation, however, occurred. The irradiation of

Photolysis of Propiconazole in Aqueous Media
J. Agric. Food Chem., Vol. 49, No. 11, 2001 5381
Sch em e 1
When exposed to solar light, propiconazole underwent
a slow photodegradation (Figure 6). The half-life in J une
was estimated at 85 ( 10 h.
In Natural Water. In a last step, propiconazole was
dissolved in natural water and exposed to solar light.
As illustrated by Figure 6, the disappearance was faster
in natural water than in pure water, showing that
propiconazole was subject to enhanced photodegradation
due to indirect processes in natural water. The half-life
was estimated as 60 ( 10 h in natural water compared
to 85 h in pure water.
In the Presence of Humic Substances. The presence
of humic substances (10 mg‚L-1) increased the rate of
propiconazole disappearance when irradiated both in
simulated (Figure 4) and natural solar light (Figure 6).
In the former case, the apparent first-order rate con-
stant deduced from the plot ln(c/c0) vs t was found to be
equal to 7 × 10-3 h-1, i.e., 40% higher than that in pure
water. In the latter, the half-life was estimated as 73 (
10 h, compared to 85 ( 10 h in pure water.
Photoproducts were similar to those observed in pure
water, but the distribution pattern was different. The
presence of humic substances reduced the formation
rate of compounds 1 and 2 (Figure 5a) but enhanced
that of 3 and 4 (Figure 5b). These quantitative data
confirm that humic acids are able to photoinduce the
degradation of propiconazole.
DISCUSSION AND CONCLUSION
We show that propiconazole photodegrades in solar
light. The phototransformation is faster by about 30%
in natural water than in pure water. It proves that both
direct photolysis and photoinduced processes contribute
to the degradation reaction under environmentally
relevant conditions. Humic matter contained in natural
waters is likely to be involved in the photoinduced
reactions as the phototransformation of propiconazole
was shown to be enhanced by humic acids.
An important part of the work was devoted to product
studies using APCI mass spectrometry and high-field
1
NMR H and 13C. The analytical aspects of this work
reaction are interesting in several ways.
The full attribution of NMR signals needed prelimi-
nary and detailed studies on propiconazole itself. The
two isomers of propiconazole both could be well char-
acterized (Figure 1).
Propiconazole undergoes several ways of phototrans-
formation on excitation in pure water. We prove that
photo-cyclization accompanied by elimination of HCl
occurs, especially upon irradiation at short wavelength.
Such a reaction was quite unexpected even though other
examples of photo-cyclization accompanied by elimina-
tion of HCl or HBr already have been reported in the
literature (19-22). The mechanism of cyclization is not
yet clear. The concerted removal of H and Cl atoms is
F igu r e 5. Formation of 1, 2 (a) and 3 + 4 (b) over time upon
irradiation of propiconazole (1.5 × 10-5 mol‚L-1) in simulated
solar light. (4): 1 in pure water; (0): 2 in pure water; (9): 2 in
water containing humic substances (10 mg/L); (O): 3 + 4 in
pure water; (b) 3 + 4 in water containing humic substances
(10 mg/L).
F igu r e 6. Phototransformation of propiconazole (1.5 × 10-5
M) in solar light. b pure water; O water containing humic acids
(10 mg/L); 2 natural water.

5382 J. Agric. Food Chem., Vol. 49, No. 11, 2001
Vialaton et al.
(7) Thomas-Smith, T. E.; Blough, N. V. Photochemical
formation of hydroxyl radical by constituents of natural
waters. Environ. Sci. Technol. 2001, 35, 2721-2726.
(8) Zafiriou, O. C.; True, M. B. Nitrate photolysis in
seawater by sunlight. Marine Chem. 1979, 8, 33-42.
(9) Mille, M. J .; Crosby, D. G. Pentachlorophenol and 3,4-
dichloroaniline as models for photochemical reactions
in seawater. Marine Chem. 1983, 14, 111-120.
(10) Skurlatov, Y. I.; Zepp, R. G.; Baughman, G. L. Photolysis
rates of (2,4,5-trichlorophenoxy)acetic acid and 4-amino-
3,5,6-trichloropicolinic acid in natural waters. J . Agric.
Food Chem. 1983, 31, 1065-1071.
(11) Vialaton, D.; Richard, C.; Baglio, D.; Paya-Perez, A.-B.
Phototransformation of 4-chloro-2-methylphenol in wa-
ter: influence of humic substances on the reaction. J .
Photochem. Photobiol. A 1998, 119, 39-45.
(12) Mushtaq, M.; Chukwudebe, A. C.; Wrzesinski, C.; Allen,
L. R. S.; Luffer-Atlas, D.; Arison, B. H. Photodegradation
of emamectin benzoate in aqueous solution. J . Agric.
Food Chem. 1998, 46, 1181-1191.
conceivable, because molecular models show that the
chlorine atom attached to C16 is very close to the
hydrogen atom H5. The compound 2 that has the mass
of 1+18 and shows a secondary formation kinetic is
likely to result from a multistep and uncomplete hy-
drolysis of the dioxolane group of product 1 (Scheme 1).
This reaction, which is not observed in the dark, is most
probably light-induced. The formation of ketone 5 is less
surprising because this compound corresponds to the
hydrolysis of the 1,3-dioxolane protecting group. The
analysis of irradiated mixtures by HPLC-APCI-MS
allowed the detection of photoproducts 3 and 4 that have
the particularity to each have several isomers. They are
likely to be oxidation photoproducts; however, full
assignment could not be achieved. The mechanism of
direct photolysis is clearly wavelength dependent: pho-
tocyclization is the major pathway at short wavelength,
whereas oxidation mainly occurs in solar light.
The photoinduced transformation by humic sub-
stances yields mainly compounds 3 and 4. Oxidant
species photogenerated by humic material seem able to
abstract electrons or hydrogen atoms from propiconazole
with subsequent formation of oxidation products. The
oxidation is more difficult to explain in the case where
propiconazole is irradiated in pure water. Oxidation of
triplet excited states of propiconazole might take place.
As an alternative, photoinduction involving primary
photoproducts might occur. For example, compound 5
might photoinduce the oxidation of propiconazole as it
is known that excited triplet states of aromatic ketones
have oxidant properties (23-25).
(13) Corin, N.; Backlund, P. H.; Kulovaara, M. A. M. Pho-
tolysis of the resin acid dehydroabietic acid in water.
Environ. Sci. Technol. 2000, 34, 2231-2236.
(14) Welker, M.; Steinberg, C. Rates of humic substances
photosensitized degradation of microcystin-LR in natu-
ral waters. Environ. Sci. Technol. 2000, 34, 3415-3419.
(15) Krieger, M. S.; Yoder, R. N.; Gibson, R. Photolytic
degradation of florasulam on soil and water. J . Agric.
Food Chem. 2000, 48, 3710-3717.
(16) Vialaton, D.; Baglio, D.; Paya-Perez, A.-B.; Richard, C.
Photochemical transformation of acifluorfen under labo-
ratory and natural conditions. Pestic. Manage. Sci. 2001,
57, 372-379.
(17) Gunther, H. La spectroscopie de RMN; Masson: Paris,
France, 1993.
(18) Silverstein, R. M.; Bassler, G. C.; Morill, T. C. Spectro-
metric Identification of Organic Compounds; Wiley:
New York, 1991.
(19) Swaminathan, K. S.; Ganesh, R. S.; Venkatachalam, C.
S.; Balasubramanian, K. K. A novel photochemical cycli-
sation of 4-chloro-3-(N-aryliminomethyl) (2H)benzopy-
rans and benzothiopyrans. Tetrahedron Lett. 1983, 24,
3653-3657.
(20) Raucher, S.; Bray, B. L.; Lawrence, R. F. Synthesis of
(()-Catharanthene, (+)-Anhydrovenblastine and (-)-
Anhydrovencovaline. J . Am. Chem. Soc. 1987, 109, 442-
446.
Su p p or tin g In for m a tion Ava ila ble: Carbon spectra,
J -modulated, COSY, NOESY, (in order to distinguish between
H3 and H5, through their different interactions with (CH2)6,
depending on their relative space proximity), traditional and
undecoupled HETCOR (over one bond), and heteronuclear
multiple bond correlation (HMBC, set to long-range correla-
tion) were acquired with gradient selection, if necessary. This
material is available free of charge via the Internet at http://
pubs.acs.org.
LITERATURE CITED
(1) Tomlin, C. The Pesticide Manual, 10th ed.; The Royal
Society of Chemistry and The Crop Protection Council:
Surrey, U.K., 1994.
(2) Cooper, W. J .; Zika, R. G.; Petasne, R. G.; Fischer, A.
M. Sunlight-induced photochemistry of humic sub-
stances in natural water: major reactive species. ACS
Symp. Ser. 1989, 219, 333-361.
(3) Hoigne´, J .; Faust, B. C.; Haag, W. R.; Scully, F. E., J r.;
Zepp, R. G. Aquatic Humic Substances as sources and
sinks of photochemically produced transcient reactants.
ACS Symp. Ser. 1989, 219, 363-381.
(4) Canonica, S.; J ans, U.; Stemmler, K.; Hoigne´, J . Trans-
formation kinetics of phenols in water: Photosensitiza-
tion by dissolved natural organic material and aromatic
ketones. Environ. Sci. Technol. 1995, 29, 1822-1831.
(5) Aguer, J .-P.; Richard, C.; Andreux, F. Comparison of the
photoinductive properties of commercial, synthetic and
soil-extracted humic substances. J . Photochem. Photo-
biol. A 1997, 103, 163-168.
(21) Gupta, S.; Bhakuni, D. S. Photochemical synthesis of
aporphines. Synth. Commun. 1988, 18, 2251-2258.
(22) Netto-Ferreira, J . C.; Scaiano, J . C. A laser flash
photolysis study of the mechanism of the photocyclisa-
tion of R-chloro-O-methyl-acetophenons. J . Am. Chem.
Soc. 1991, 113, 5800-5803.
(23) Das, P. K.; Encinas, M. V.; Scaiano, J . C. Laser flash
photolysis study of the reaction of carbonyl triplets with
phenols and photochemistry of p-hydroxy-propiophe-
none. J . Am. Chem. Soc. 1981, 103, 4154-4162.
(24) Murov, S. L.; Carmichael, I.: Hug, H. G. Handbook of
Photochemistry, 2nd ed.; Marcel Dekker: New York,
1993.
(25) Canonica, S.; Hellrung, B.; Wirz, J . Oxidation of phenols
by triplet aromatic ketones in aqueous solution. J . Phys.
Chem. A 2000, 104, 1226-1232.
Received for review February 26, 2001. Revised manuscript
received August 27, 2001. Accepted September 5, 2001.
(6) Vaughan, P. P.; Blough, N. V. Photochemical formation
of hydroxyl radical by constituents of natural waters.
Environ. Sci. Technol. 1998, 32, 2947-2953.
J F010253R