Reaction kinetics and mechanisms of neonicotinoid pesticides with sulfate radicals

Article abstract of DOI:10.1039/c0nj00726a

The reaction kinetics and mechanisms of three neonicotinoid insecticides, imidacloprid (IMD), thiacloprid (THIA) and acetamiprid (ACT) with sulfate radicals were studied by flash-photolysis of peroxodisulfate, S ₂O₈^2-. The absolute rate constants (3 ± 1) × 10⁸, (1.1 ± 0.6) × 10⁹, and (3 ± 1) × 10⁹ M^-1 s^-1 were determined for IMD, ACT, and THIA, respectively. The reactivity and absorption spectra of the observed organic intermediates are in line with those reported for α-aminoalkyl radicals, and their absorption spectra agree very well with those estimated employing the time-dependent density functional theory with explicit account for bulk solvent effects. The mono- and di-hydroxylated oxidation products of the insecticides were identified as primary degradation products. The proposed reaction mechanism supports an initial charge transfer from the amidine nitrogen of the insecticides to the sulfate radicals. The pyridine moiety of the insecticides remains unaffected even after long irradiation times, until nicotinic acid is formed.

Full text of DOI:10.1039/c0nj00726a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
NJC
Cite this: New J. Chem., 2011, 35, 672–680
www.rsc.org/njc
PAPER
Reaction kinetics and mechanisms of neonicotinoid pesticides with sulfate
radicalsw
a
a
a
b
Marı
and Mo
´
´
a L. Dell’Arciprete, Carlos J. Cobos, Daniel O. Martire, Jorge P. Furlong
´
a
nica C. Gonzalez*
Received (in Montpellier, France) 22nd September 2010, Accepted 22nd November 2010
DOI: 10.1039/c0nj00726a
The reaction kinetics and mechanisms of three neonicotinoid insecticides, imidacloprid (IMD), thiacloprid
THIA) and acetamiprid (ACT) with sulfate radicals were studied by flash-photolysis of peroxodisulfate,
(
2ꢀ
8
9
9
ꢀ1 ꢀ1
S2O8 . The absolute rate constants (3 ꢁ 1) ꢂ 10 , (1.1 ꢁ 0.6) ꢂ 10 , and (3 ꢁ 1) ꢂ 10 M
s were
determined for IMD, ACT, and THIA, respectively. The reactivity and absorption spectra of the observed
organic intermediates are in line with those reported for a-aminoalkyl radicals, and their absorption spectra
agree very well with those estimated employing the time-dependent density functional theory with explicit
account for bulk solvent effects. The mono- and di-hydroxylated oxidation products of the insecticides were
identified as primary degradation products. The proposed reaction mechanism supports an initial charge
transfer from the amidine nitrogen of the insecticides to the sulfate radicals. The pyridine moiety of the
insecticides remains unaffected even after long irradiation times, until nicotinic acid is formed.
19 days, the elimination of the insecticide under such conditions
Introduction
might not improve the quality of the contaminated water, as
the primary products of degradation showed considerable
Neonicotinoid insecticides are agonists at nicotinic acetyl-
choline receptors and therefore show selective toxicity for
3
toxicity to Vibrio fischeri assays.
1
insects over vertebrates. In particular, acetamiprid ((E)-N-
Destruction of organic compounds contained in waters and
soils by peroxodisulfate oxidation is gaining interest as an
4,5
0
(
6-chloro-3-pyridylmethyl)-N -cyano-N-methylacetamidine, ACT
in Scheme 1) and thiacloprid (3-(6-chloro-3-pyridylmethyl)-
-thiazolidinylidene-cyanamide, THIA in Scheme 1) have been
in situ chemical oxidation (ISCO) because peroxodisulfate
2
may be easily activated to yield the highly reactive sulfate
demonstrated to be less toxic towards beneficial insects
2
such as honey bees than the widely employed imidacloprid
ꢃ
ꢀ
6
radical ions (SO
ꢃ
ꢀ
4
of neonicotinoid insecticides with SO radicals, and the
4
). Thus, the study of the reactivity
(1-(6-chloro-3-pyridylmethyl)-2-nitroimino-imidazolidine, IMD
characterization and reactivity of the organic radicals
in Scheme 1), and therefore, they are increasingly used.
Investigations on the degradation of imidacloprid, thiacloprid
generated are of environmental interest.
2ꢀ
Photolysis of S2O8
with excitation wavelengths
ꢃ
and acetamiprid mediated by hydroxyl radicals (HO ) and
lexc o 300 nm, reaction (1), is a clean source of sulfate radical
ions with high quantum yields, independent of pH in the range
toxicity evaluation of the primary oxidation products of
the insecticides showed that in all cases, the less toxic
7
–14. Flash-photolysis of S
2ꢀ
3
2
O
8
is therefore an appropriate
6
-chloronicotinic acid was found to be the major product at
method for time-resolved studies yielding information on
absolute rate constants and transient generation.
long irradiation times. However, despite the fact that the half
ꢃ
life of the insecticides due to their reaction with HO radicals
2
8
ꢀ
ꢃ
ꢀ
in natural aquatic reservoirs may vary between 5 hours and
S
2
O
+ hn - 2SO
4
(1)
We report here a kinetic and mechanistic study on the reactions
a
Instituto de Investigaciones Fisicoquı´micas Teo´ricas y Aplicadas
(INIFTA), Facultad de Ciencias Exactas, Universidad Nacional de
La Plata, Casilla de Correo 16, Sucursal 4, (1900) La Plata,
Argentina. E-mail: gonzalez@inifta.unlp.edu.ar
LADECOR, Departamento de Quı´mica, Facultad de Ciencias
Exactas, Universidad Nacional de La Plata, Casilla de Correo 16,
Sucursal, (1900) La Plata, Argentina
ꢃ
ꢀ
of SO4 with neonicotinoid insecticides imidacloprid (IMD),
thiacloprid (THIA) and acetamiprid (ACT) at pH 3 and 9.
ꢃ
ꢀ
In situ generation of SO
2ꢀ
4
is performed by flash-photolysis of
b
S
2
O
8
aqueous solutions.
w Electronic supplementary information (ESI) available: The ESI
contains details on the kinetic treatment of time-resolved sulfate
radicals absorbance profiles, a table with calculated Mulliken atomic
charges and a figure showing the TD-DFT calculated spectra of the
a-aminoalkyl radicals of ACT protonated in the pyridine ring. See
DOI: 10.1039/c0nj00726a
Experimental
Imidacloprid, acetamiprid, and thiacloprid were obtained
from Aldrich and used as received. Sodium peroxodisulfate,
NaOH, and HClO4 from Merck were used without further
6
72 New J. Chem., 2011, 35, 672–680
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
Scheme 1 Chemical structures of the insecticides acetamiprid, thiacloprid, and imidacloprid.
purification. Distilled water was passed through a Millipore
ꢀ
possible thermal reactions of peroxodisulfate with the
substrates. Since no important variations in pH (o0.5 units
of pH) were observed after each flash of light or after 64 laser
pulses and considering that fresh solutions were used for
each experiment, pH may be assumed constant during the
irradiation experiments.
1
system (>18 MO cm , o20 ppb of organic carbon). The pH
of the solutions was controlled by addition of either HClO
NaOH as required.
4
or
For product detection, aqueous solutions of the pesticides
(
2 ꢂ 10ꢀ M) containing 5 ꢂ 10 M Na S O solutions were
4
ꢀ2
2
2
8
irradiated with 30 flashes of light. Product identification was
performed by HPLC-MS analysis (Agilent 1100, equipped
with a diode array and ESI/single quadrupole detector). A
C18 Restek Pinacle II column (250 mm ꢂ 2.1 mm i.d., particle
size 5 mm) was used. The eluent was a mixture of acetonitrile–
water 1 : 1 (v/v) at 0.5 mL min constant flux. Detection
limits were typically 1 ppm. The MS detector was operated in
the positive mode with mass detection in the range from 50 to
Bilinear regression analysis
For each experimental condition, several absorbance
decay profiles at different detection wavelengths were taken.
Absorbance is thus a function of wavelength and time. A
bilinear regression analysis taking advantage of the linearity of
the absorbance with both concentrations and absorption
coefficients was applied to the experimental absorption matrix
in order to retrieve information on the minimum number
of species and on their relative concentration profiles and
ꢀ
1
3
00 amu. The system used a fragmentation potential of 70 eV.
The organic substrates remaining in solution were extracted
with a 1 : 1 CH Cl –MeOH mixture, the organic phase was
2
2
26
absorption spectra.
separated and evaporated to 1/3 of its volume and further
diluted with a 1 : 1 acetonitrile–H O mixture. The injection
2
Time-dependent density functional theory calculations
volume was 10 mL.
Flash-photolysis experiments were carried out using either
conventional flash lamps or a laser as the irradiation source.
The conventional apparatus was a Xenon Co. model 720C
It has been shown that the time-dependent generalization of
the density functional theory, TD-DFT, provides remarkable
results in the calculation of the observed transition energies
of the electronically excited states of a large number of
2
5
with modified optics and electronics. Two collinear quartz
Xenon high-intensity pulsed flash tubes, Xenon Corp. P/N
2
7,28
molecules.
In particular, the theory gives a well-balanced
8
90-1128 (FWHM r 20 ms), with a continuous spectral
description for conjugated molecules and open-shell systems
2
9
distribution ranging from 200 to 600 nm and maximum
around 450 nm were used. The analysis source was a high
pressure mercury lamp (Osram HBO-100 W). The optical path
length of the 1 cm internal diameter quartz sample cell was
such as excited states of radicals. In the present study, the
hybrid B3LYP density functional was used. This approach
3
0
implies the Becke’s three-parameters exchange functional
coupled to the nonlocal correlational functional of Lee,
3
1
1
0 cm. The monochromator collecting the analysis beam
Wang, and Parr. For a more reliable comparison with the
experiments, bulk solvent effects were evaluated employing the
(
Bausch & Lomb, high intensity) was directly coupled to a
3
2
photomultiplier (RCA 1P28), whose output was fed into
a digital oscilloscope (HP 54600B). Digital data were stored
in a personal computer. The emission of the flash lamps was
filtered with an aqueous solution highly concentrated in the
corresponding organic compound in order to avoid photolysis
of the substrate.
conductor-like polarizable continuum model, CPCM, using
for the water a dielectric constant of 78.39. It should be noted
that this is a suitable approach as long as no important specific
interactions between the solute and solvent molecules are
present. All calculations have been performed with the
3
3
Gaussian 03 package.
The absorption spectra were
a two-step procedure. Firstly, the
Laser flash photolysis experiments were performed by
excitation with the fourth harmonic of a Nd:YAG Litron laser
computed following
optimization of all structural parameters of each ground-state
molecule (without symmetry constraints) was performed with
the B3LYP/6-311++G(d,p) functional via analytic gradient
methods. In all the cases, real vibrational frequencies were
obtained assuring that molecular structures correspond to
energy minima. In the second step, the vertical electronic
energies, the associated wavelengths of the band maxima,
(2 ns FWHM and 6 mJ per pulse at 266 nm). The analysis
light from a 150 W Xe arc lamp was passed through a
monochromator (PTI 1695) and detected by a 1P28 PTM
photomultiplier. Decays typically represented the average of 64
pulses and were taken by and stored in a 500 MHz Agilent
Infiniium oscilloscope. The photolysis of the substrates was
observed to be negligible under the experimental conditions used.
The temperature (20 ꢁ 3 1C) was measured inside the
reactor cell with a calibrated Digital Celsius Pt-100 O thermo-
meter. Freshly prepared solutions were used in order to avoid
lmax, and oscillator strengths were computed at the CPCM-
B3LYP/6-311++G(d,p) level of theory. The expectation
2
value of the hS i operator was not greater than 0.8, close to
the exact value of 0.75 for doublet states, therefore indicating
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 672–680 673

View Article Online
that the spin contamination due to excited states is very small
for the investigated intermediates.
The DF-DFT calculations do not account for vibrational
broadening. Therefore to compare with experimental results,
simulated spectra were obtained by representing each
electronic transition with Gaussian shape functions centered
at the calculated
l
obtained considering for all computed transitions a value of
max
. The whole spectra are then
ꢀ
1
s = 2400 cm for the full width of the band at 1/e height.
Results and discussion
Kinetic measurements
Fig. 2 Apparent rate constants, kapp, vs. [Ins]
0
for IMD (black
ꢀ
2
2ꢀ
8
symbols) and THIA (open symbols) obtained in experiments with
Conventional flash photolysis of air-saturated 5 ꢂ 10 M S O
2
ꢀ3
5
9
ꢂ 10
M Na
2
S
2
O
8
aqueous solutions of pH 3 (circles) and
for ACT obtained in experiments
2 2 8
S O aqueous solutions of pH 3 (circles) and
solutions of pH either 3 or 9 showed the formation of a
transient species absorbing in the wavelength range from
(triangles). Inset: kapp vs. [Ins]
0
ꢀ3
with 5 ꢂ 10 M Na
(triangles).
3
00 to 600 nm, whose spectrum taken immediately after the
9
flash of light, is in agreement with that reported in the
2ꢀ
8
O in the presence of
ꢃ
ꢀ 3
literature for SO
4
. Photolysis of S
2
ꢀ
6
low concentrations (o1 ꢂ 10 M) of the insecticides showed
absorption traces at detection wavelengths l > 350 nm, whose
spectrum immediately after the flash of light also agrees with
2
Table 1 Bimolecular rate constants k obtained for IMD, ACT, and
THIA at pH 3 and 9
Insecticide
IMD
ACT
THIA
ꢃ
ꢀ
that of SO4 but with faster decay rates (see Fig. 1).
ꢀ
8
9
9
ꢃ
ꢀ1 ꢀ1 pH 3 (3 ꢁ 1) ꢂ 10
(1.1 ꢁ 0.6) ꢂ 10 (3 ꢁ 1) ꢂ 10
The decay of SO4 absorbance at detection l, A(l), could
be well fitted to eqn (I).
k2/M
s
8
9
9
pH 9 (3.5 ꢁ 1) ꢂ 10 (1.0 ꢁ 0.8) ꢂ 10 (4.5 ꢁ 1) ꢂ 10
A(l) = a(l) exp(ꢀkapp ꢂ t) + c(l)
The small residual absorbance [c(l) in eqn (I)] is associated
with long-living species, mainly the organic radicals formed by
reaction (2).
(I)
The k
2
values are, within the experimental error, independent
of
of pH for the three insecticides. Therefore, either the pK
a
ꢃ
ꢀ
the insecticides is out of the working pH range, or the SO4
attack on the insecticides is not significantly affected by the
acid–base centre.
ꢃ
ꢀ
- Transients
Insecticide + SO
4
(2)
Stable products identification
The apparent rate constant, kapp, is independent of wavelength
l and linearly increases with the analytical concentration of
Primary stable degradation products of the insecticides were
identified in experiments with aqueous solutions containing
the insecticide, [Ins]
in Fig. 2 for IMD, THIA, and ACT at pH 3 and 9. The slope
of these straight lines yields the bimolecular rate constants k
0
, as expected from reaction (2) and shown
ꢀ
4
ꢀ2
2
ꢂ 10 M of the insecticides and 5 ꢂ 10 M Na
2 2 8
S O after
2
,
3
0 flashes of light. The intense microsecond pulses of light
depicted in Table 1. See ESI1w for a detailed explanation of the
kinetic treatment of time-resolved data.
allowed the generation of a measurable concentration of
primary products, minimizing further reactions of these
products with sulfate radicals. Table 2 summarizes the mass
spectrometry data (MS) observed for the degradation
products of the three insecticides. Under stronger oxidation
conditions (longer irradiation times), nicotinic acid is observed
to be formed in experiments with the three insecticides.
ꢃ
ꢀ
radicals
Organic radicals formed after the reactions of SO
with the insecticides
4
In order to characterize the organic radicals formed after
2ꢀ
ꢀ
3
reaction (2), argon- or air-saturated 5 ꢂ 10
M S
2
O
8
solutions of pH > 7 with concentrations of the insecticides
ꢀ5
Z 1 ꢂ 10 M were irradiated with a conventional flash lamp.
ꢃ
ꢀ
lifetime is diminished from
Under these conditions, the SO
4
6
0 ms in the absence of insecticides to o2 ms, and the organic
intermediates formed after reaction (2) are the main transients
detected in conventional flash photolysis experiments.
Fig. 1 Absorbance traces at 450 nm obtained in experiments with
ꢀ
2
For each insecticide, several transient decay profiles were
taken at different detection wavelengths within the range
300–500 nm. A bilinear regression analysis was applied to
5
ꢂ 10 M Na
2 2 8
S O solutions in the presence (grey curve) and
ꢀ6
absence (black curve) of 1 ꢂ 10 M the IMD. The full lines stand
for the fittings to eqn (I).
6
74 New J. Chem., 2011, 35, 672–680
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
ꢀ4
Table 2 Primary degradation products formed after 30 flashes irradiation of air-saturated aqueous solutions containing 2 ꢂ 10 M of the
t
, and MS mass to charge ratios, m/z, are given, together with assigned products
ꢀ2
insecticides and 5 ꢂ 10 M Na
2
S
2
O
8
. HPLC retention times, R
Insecticide m/z
R
t
/min Product assignment
+
+
+
+
IMD
272(271 + H
)
)
)
)
1.7
1.6
2.1
2.7
N-[1-[(6-Chloro-3-pyridyl)methyl]-4,5-dihydro-5-hydroxyimidazol-2-yl]nitramide, IMD1 in Scheme 3
N-[1-[(6-Chloro-3-pyridyl)methyl]-4,5-dihydro-4,5-dihydroxyimidazol-2-yl]nitramide, IMD3 in Scheme 3
N-[1-[(6-Chloro-3-pyridyl)methyl]-4,5-dihydro-5-oxoimidazol-2,3-yl]nitramide, IMD2 in Scheme 3
3-[(6-Chloro-3-pyridyl)methyl]-4-hydroxy-2-thiazolidinylidenecyanamide, THIA2 in Scheme 4, or
2
2
88(287 + H
70(269 + H
THIA
ACT
269(268 + H
3
N-[(6-Chloro-3-pyridyl) methyl]-N-cyanoacetamidine, ACT2 in Scheme 5
-[(6-Chloro-3-pyridyl)hydroxymethyl]-2-thiazolidinylidenecyanamide, THIA1 in Scheme 4
+
+
209(208 + H
237(236 + H
)
)
2.6
6.7
0
N-[(6-Chloro-3-pyridyl) methyl]-N -cyano-N-formylacetamidine, ACT1 in Scheme 5
the absorbance matrix (see Experimental) to obtain information
on the minimum number of species. The bilinear analysis
shows that the data obtained for the different insecticides
may be described by two absorbing species: an organic
transient with absorption maxima around 330–350 nm and
4
40–450 nm, already present 100 ms after the flash of light,
which decays to yield stable products with absorption maxima
around 300–350 nm. The transient spectra obtained in
argon- and air-saturated solutions of pH 7 are shown in
Fig. 3–5 for IMD, ACT, and THIA, respectively.
For each insecticide, the absorption spectra of the transients
observed in experiments with argon and air-saturated
solutions are coincident, as shown in Fig. 3 to 5. However,
the decay of the transients strongly depends on the presence of
molecular oxygen. The decay follows a second order rate law
in argon-saturated solutions (see insets of Fig. 4 and 5 for the
transients decay of ACT and THIA, respectively) with rate
constant 2krec/e(l) (see Table 3), and a pseudofirst order law
under air saturation (see inset of Fig. 3 for the transient decay
of IMD). Considering that the dissolved molecular oxygen
concentration in air-saturated aqueous solutions at 25 1C is
Fig. 4 Absorption spectra (arbitrary units) of the transients observed
82ꢀ
ꢀ3
in conventional flash-photolysis experiments with 5 ꢂ 10 M S
2
O
ꢀ5
and 1 ꢂ 10 M ACT in argon- (J) and air-saturated (K) solutions of
pH 7. The grey line stands for the TD-DFT calculated spectrum for
the a-aminoalkyl radical ACTRX (see Scheme 5). The dotted line
stands for the calculated absorption spectrum of the a-aminoalkyl
ACTRY and the dashed line stands for that of the radical cation
ACTRC. Inset: transient and stable product (upper and lower trace,
respectively) contribution to the absorption traces at 300 nm for
experiments with the argon-saturated solutions shown in the main
figure. The fitting to a second order rate law is overlapped with the
experimental curve.
Fig. 3 Absorption spectra (in arbitrary units) of the transients
observed in conventional flash-photolysis experiments with:
ꢀ5
ꢀ3
82ꢀ
(
J) 1 ꢂ 10 M S
2
O
and 2 ꢂ 10 M IMD argon-saturated
Fig. 5 Absorption spectra (arbitrary units) of the transients observed
ꢀ3
ꢀ
3
82ꢀ
ꢀ5
82ꢀ
solutions of pH 7 and (K) 5 ꢂ 10 M S
2
O
and 2 ꢂ 10 M IMD
in conventional flash-photolysis experiments with 5 ꢂ 10 M S
2
O
ꢀ5
air-saturated solutions of pH 7. The grey line stands for the TD-DFT
and 1 ꢂ 10 M THIA in argon- (J) and air-saturated (K) solutions
of pH 7. The dashed grey lines stand for the TD-DFT calculated
spectrum for the a-aminoalkyl radicals THIARX (curve a) and
THIARY (curve b) (see Scheme 2). The grey solid line stands for
the linear combination 0.9 ꢂ a + 0.5 ꢂ b of curves a and b.
The dashed line stands for the calculated absorption spectrum of
the radical cation THIARC. Inset: transient contribution to the
absorption traces at 320 nm for experiments with the argon-saturated
solutions shown in the main figure. The fitting to a second order rate
law is overlapped with the experimental curve.
calculated spectrum for the a-aminoalkyl radical IMDRX
(see Scheme 3). The dotted line stands for the calculated absorption
spectrum of the a-aminoalkyl radical situated on the methylene bridge,
IMDRY, and the dashed line stands for that of the radical cation
IMDRC. Inset: transient and stable product (upper and lower traces,
respectively) contribution to the absorption traces at 420 nm for
experiments with the air-saturated solutions shown in the main
figure. The fitting to a first order rate law is overlapped with the
experimental curve.
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 672–680 675

View Article Online
Table 3 Second order decay rate constants 2krec/e, rate constants for the reaction with molecular oxygen, kO2, lower limit absorption coefficients,
min
TD-DFT
e
shown in Schemes 3–5, respectively
, and theoretical values for the absorption coefficient, e
, for the a-aminoalkyl radicals IMDRX, THIARX and THIARY, and ACTRX
a
Insecticide
IMDRX
ACTRX
THIARX + THIARY
ꢀ
1
6
5
5
2
k
rec/e(l)/cm s
3 ꢂ 10 (350 nm)
3 ꢂ 10 (300 nm)
5 ꢂ 10 (320 nm)
ꢀ
1
ꢀ1
7
5
5
k
O2/M
min
s
ꢀ1
1.2 ꢂ 10
1200 (350 nm)
00 (450 nm)
700 (350 nm)
80 (450 nm)
5 ꢂ 10
(4–6) ꢂ 10
ꢀ1
4
4
e
e
(l)/M cm
1.2 ꢂ 10 (300 nm)
1 ꢂ 10 (320 nm)
4
TD-DFT
ꢀ1
ꢀ1
4
4
(l)/M cm
1 ꢂ 10 (300 nm)
1.45 ꢂ 10 (320 nm)
3
a
Corresponds to a mixture of 0.9 moles of THIAX and 0.5 moles of THIAY.
.5 ꢂ 10ꢀ M, the bimolecular rate constants kO2 for the
4
2
reaction of the transients with molecular oxygen are obtained
(
Table 3).
Assuming that the observed transients are the only organic
intermediates formed after reaction (2), and considering that
2
ꢀ
ꢃ
ꢀ
S
2
O
8
competes with the insecticide for SO
4
with reaction
, the concentration of the
organic transients formed is on the order of the photo-
7
ꢀ1 ꢀ1
s
rate constant kPS E 1 ꢂ 10 M
ꢃ
ꢀ
generated [SO
2
PS ꢂ [S
4
] times the factor k ꢂ [Ins] +
ꢃ
ꢀ
4
]). The concentration of SO may be calcu-
2
ꢂ [Ins]/(k
2
ꢀ
k
2
O
8
lated from the absorption traces at 450 nm obtained in
experiments in the absence of the insecticides, but otherwise
ꢀ1
ꢀ1 7
identical conditions, taking eSO4ꢀ(450 nm) = 1600 M cm
From these values and the organic transient absorption extra-
.
polated at time zero, a lower limit for their absorption
min
coefficients, e , is obtained, as depicted in Table 3. Taking
min
the e
values, the recombination rate constants 2krec are in
9
ꢀ1 ꢀ1
the range (3–5) ꢂ 10 M
s
for the three insecticides.
ꢀ
attack on the insecticides should be
ꢃ
4
Several sites of SO
considered. Sulfate radical attack on the pyridine moiety of the
insecticides yielding a hydroxylated pyridine ring may be
neglected, since neither the observed reaction products, nor
the spectra of the observed transients support this reaction
7
Scheme 2 Mesomeric forms of the amidine group in the insecticide
thiacloprid and charge transfer to sulfate as the initial reaction
pathway leading to the formation of the radical cation THIARC,
followed by H-elimination from the latter to yield a-aminoalkyl
radicals THIARX and THIARY.
pathway. On the other hand, the rate constants for the sulfate
5
radical H-abstraction from C–H bonds are (10 to 10 M
7
ꢀ1 ꢀ1 8
s )
much smaller than those measured here for the reactions of
ꢀ
with the insecticides. The one-electron charge transfer
ꢃ
SO
4
reaction from an amine nitrogen to sulfate radicals to yield a
radical cation in the N atom has been reported to take place
9
with aliphatic amines, pyrimidines, cyanuric acid, pur-
10
11
atom and decreases with the ring size (if they are cyclic as in
1
2
13
14
ines, arginine and tryptophan. Since the three insecticides
have amino-type nitrogen atoms in the amidine moiety and the
IMD and THIA) in the sequence C
6 7 8 5
> C > C , C . The
occurrence of the mesomeric form 2 of the imido group (see
1
9
values obtained for k
2
are on the order of those reported for
Scheme 2) supports protonation in the imino nitrogen.
Strong electron withdrawing substituents such as NO and
the electron transfer reaction from an unprotonated amine-
9
type nitrogen to the sulfate radicals, the charge transfer
2
CN in the imino nitrogen of IMD and of ACT and THIA,
reaction must be considered. Since the one-electron reduction
15
respectively, are expected to considerably diminish the pK of
a
ꢃ
ꢀ
o
potential for SO4 is E (SO /SO4 ) = 2.5 V vs. NHE
ꢃ
ꢀ
2ꢀ
the insecticides. Considering that, within the experimental
error, no effect of pH in the range from 3 to 9 was observed
4
and the reduction potentials for IMD, ACT, and THIA were
1
6
reported to be lower than 1.2 V, the variation in the standard
free energy for an electron transfer reaction involving the
insecticide and the sulfate radicals is thermodynamically
a
on the reaction rate, the pK of the imidine nitrogen in the
three insecticides is expected to be o3.
Nitrogen-centered radical cations are not prone to efficiently
0
ꢀ1
allowed, as DET
G
o ꢀ96 kJ mol . Scheme 2 shows
react with molecular oxygen, as observed here for the organic
7
,14
the proposed reaction including the resonance structures of
7
the amidine group for the insecticide thiacloprid.
transients.
In fact, many literature reports agree in the fact
+
1
that H atoms in position a to the N atom are easily lost as H
to yield an a-aminoalkyl radical highly stabilized by the free
18
Amidines are strong bases (pK
a
= 6.5–9.6), whose basicity
9
,10,13
depends to a great extent on the substituents at the imino-nitrogen
electronic pair of the vicinal nitrogen.
a-Aminoalkyl
6
76 New J. Chem., 2011, 35, 672–680
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
radicals show significant absorbance up to 500 nm and are
known to undergo diffusion controlled self-reaction at 298 K
The coincidence obtained between the experimental transient
spectra and those calculated for the a-aminoalkyl radicals RX,
except for THIA, indicates that the observed transients may be
assigned to these radicals. The transient spectrum observed for
THIA (shown in Fig. 5) can be reproduced from the calculated
spectra if 1.8 mol of the a-aminoalkyl radical THIARX are
2
in iso-octane solvent and to efficiently react with molecular
0
3
,21,22
oxygen,
in agreement with the behavior of the transients
herein observed. Therefore, considering that the amine-type
nitrogen atom in the amidine moiety of IMD, ACT and
THIA, all present H-atoms a to nitrogen, identification of
the observed transients as a-aminoalkyl radicals is strongly
suggested. Scheme 2 shows the reaction paths leading to the
formation of two possible a-aminoalkyl radicals in C1 and C2
of the radical cation.
formed per mol of THIARY.
min
The values of e
the observed transients and the e
predicted from the experimental traces of
TD-DFT
obtained for the
a-aminoalkyl radicals RX from theoretical calculations are,
within the error, on the same order of magnitude (see Table 3).
Therefore, the a-aminoalkyl radicals IMDRX, ACTRX,
THIARX and THIARY are the main organic transients
detected.
To help identify the nature of the observed transients, the
TD-DFT spectra of a-aminoalkyl radicals in carbon atoms
1
and 2 and the radical cation (now on referred to as RX,
RY and RC, respectively) were calculated for each of the
insecticides, as also shown for comparison in Fig. 3 to 5. The
spectra of the a-aminoalkyl radicals containing the conjugated
acid of the pyridine group were also calculated to evaluate the
influence of pH on the transient spectra. Interestingly,
protonation of the pyridine nitrogen does not significantly
affect the radical spectra of the a-aminoalkyl radicals
The theoretical calculations show that the difference in
energy DE between a-aminoalkyl radicals RX and RY is
ꢀ1
28.6, 24.8, and 28.6 kJ mole for ACT, THIA and IMD,
respectively. Therefore, formation of the a-aminoalkyl
radicals RY is expected to be favored. Based on this result,
the experimental observation that mainly radicals RX are
formed may be a consequence of kinetic limitations precluding
the formation of RY. The Mulliken atomic charges in C2 of
(
see ESIw, Fig. S1 for the corresponding spectra for ACT).
Scheme 3 Mechanism for the reaction of IMD with sulfate radical anions. Transients in brackets are proposed, but not detected.
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 672–680 677

View Article Online
2
1,23
radical cation RC (see Scheme 2) are more negative than in C1
+
reducing properties
and may further react with molecular
for the three insecticides (see ESIw, Table S1). Therefore, H
elimination from C1 in RC is expected to take place more
oxygen to yield peroxide radicals (reaction path c), which
upon further disproportionation yield the hydroxyl and
keto derivatives of IMD, compounds IMD1 and IMD2,
respectively. Further oxidation of compound IMD1 yields
the dihydroxylated product IMD3 (reaction path d).
readily than from C2.
The kO2 values reported for several a-aminoalkyl radicals
9
ꢀ1 ꢀ1 22
s . The k values
are within the range (0.04–3) ꢂ 10 M
O2
5
ꢀ1 ꢀ1
observed for ACT and THIA are of 5 ꢂ 10 M
s
, therefore
As already discussed, Scheme 2 shows sulfate radical attack
to THIA yielding a-aminoalkyl radicals THIARX and
THIARY with a 1.8 : 1 ratio. Both a-aminoalkyl radicals
may further react with molecular oxygen to yield peroxyl
radicals (reactions paths e and f in Scheme 4), which
disproportionate to yield compounds THIA1 and THIA2,
and undetected carbonyl derivatives. Despite that only one
chromatographic peak was observed, either from compound
THIA1 or THIA2, the formation of both compounds cannot
be neglected because of the shape of the observed transient
spectrum (vide supra).
suggesting an electron-deficient a-aminoalkyl radical
centre in these compounds. It may be expected that a higher
delocalization of the unpaired electron in ACT and THIA
a-aminoalkyl radicals, compatible with
a
shift of the
2
0,22
absorbance maximum to the visible,
further stabilizes the
radical diminishing the rate of reaction with molecular oxygen.
Reaction pathways
Based on the detected intermediates and the observed reaction
ꢀ
ꢃ
products, a pathway for the primary steps of the SO
oxidation of the insecticides may be proposed.
4
The primary oxidation steps for ACT are shown in Scheme 5.
As already proposed for the other insecticides, a charge transfer
pathway from the aminic nitrogen in ACT to sulfate radicals
yields sulfate anions and the radical cation ACTRC (reaction
Scheme 3 shows the primary oxidation steps proposed for
IMD. As discussed before, a charge transfer pathway leads to
the formation of sulfate anions and the radical cation of
+
path g), which upon elimination of H leads to a-aminoalkyl
+
IMD (reaction path a), which upon elimination of H leads
radical ACTRX (reaction path h). Reaction of the latter with
molecular oxygen yields peroxyl radicals (reaction path i)
which disproportionate to the aldehyde (compound ACT1),
to a-aminoalkyl radical IMDRX (reaction path b). The
ꢃ
a-aminoalkyl radicals (–HC –Nz) are known by their
Scheme 4 Mechanism for the thermal reactions of the a-aminoalkyl radicals THIARX and THIARY. Transients and stable compounds in
brackets are proposed, but not detected.
6
78 New J. Chem., 2011, 35, 672–680
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
Scheme 5 Mechanism for the reaction of ACT with sulfate radical anions. Transients and stable compounds in brackets are proposed, but not
detected.
molecular oxygen, and a hydroxymethylamine (product not
detected). Hydrolysis of the latter (and/or of ACT1) yields the
demethylated product ACT2 and formaldehyde (and/or formic
acid), in agreement with the proposed demethylation mechanism
formed. The absorption spectra of the latter radicals estimated
employing the time-dependent density functional theory with
explicit account for bulk solvent effects are in great accordance
with those measured experimentally. The amidine nitrogen of
ꢃ
3
ꢃ
ꢀ
initiated by HO radicals.
the molecule is the preferred site of attack by SO radicals in
4
The mechanisms proposed for the oxidation of the three
insecticides involve peroxyl radical formation, which are
the three insecticides. In fact, the pyridine group is not
oxidised until an advanced oxidation stage of the insecticide
to 6-chloro-nicotinic acid is attained, in agreement with
reported results on the oxidation of the nicotinoid insecticides
2
4
known to absorb below 300 nm. Because the high absorption
of the solutions below 300 nm hinders any transient detection
at these wavelengths, these radicals could not be detected.
3
with hydroxyl radicals.
The observed primary oxidation products for the
insecticides have been reported to retain certain toxicity in
3
V. fischeri assays. Therefore, detoxification of soils and
Conclusions
The insecticides IMD, ACT, and THIA chemically react with
8
contaminated waters with peroxodisulfate-based ISCO
technologies requires large enough oxidant residence times
to assure the elimination of toxic primary oxidation products
and improve the quality of the contaminated media.
sulfate radical anions with rate constants of (3 ꢁ 1) ꢂ 10 ,
9
9
ꢀ1 ꢀ1
M s at pH 3,
(
1.1 ꢁ 0.6) ꢂ 10 , and (3 ꢁ 1) ꢂ 10
respectively. The reactions involve a charge transfer from the
insecticide to sulfate radicals, characteristic for tertiary
amines. Proton elimination from the a carbon to the N atom
yields a-aminoalkyl radicals detected as the main transients
In this work, theory and experiments played complementary
roles, with theory providing a framework within which
the empirical results could be interpreted. As a consequence,
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 672–680 679

View Article Online
a detailed, well supported reaction mechanism for the inter-
action of sulfate radicals with neonicotinoid insecticides is
postulated.
16 M. L. Dell’Arciprete, L. Santos-Juanes, A. Arques, R. F. Vercher,
´
A. M. Amat, J. P. Furlong, D. O. Martire and M. C. Gonzalez,
Catal. Today, 2010, 151, 137–142.
1
1
1
2
2
2
7 A. A. Aly and A. M. Nour-El-Din, Special Issue Reviews and
Accounts, ARKIVOC, 2008, I, 153–194.
8 V. G. Granik, Russ. Chem. Rev. (Engl. Transl.), 1983, 52,
Acknowledgements
6
69–703.
9 E. Raczynska and J. Oczczapowicz, Tetrahedron, 1985, 41,
175–5179.
This research was supported by Consejo Nacional de
5
Investigaciones Cientı
Agencia Nacional de Promocio
PICT 2007 # 00308). M.L.D. thanks CONICET for a
´
ficas
y
Te
´
cnicas (CONICET) and
0 P. Marriott, R. Castelhan and D. Riller, Can. J. Chem., 1982, 60,
274–278.
1 K. O. Hiller and K. D. Asmus, J. Phys. Chem., 1983, 87,
´
n Cientı
´
fica y Tecnologica
´
(
3
682–3688.
2 J. Lalevee, B. Graff, X. Allonas and J. P. Fouassier, J. Phys. Chem.
A, 2007, 111, 6991–6998.
23 E. Baciocchi, T. Del Giacco and A. Lapi, Org. Lett., 2004, 6,
791–4794.
graduate studentship. M.C.G. and C.J.C. are research
members of CONICET. D.O.M. is a research member of
´
Comision de Investigaciones Cientıficas de la Provincia de
´ ´
4
Buenos Aires (CIC), Argentina.
2
4 C. von Sonntag and H. Schuchmann, in The Chemistry of Free
Radicals, ed. Z. B. Alfassi, Wiley, Chichester, 1997.
5 M. L. Alegre, M. Gerones, J. A. Rosso, S. G. Bertolotti,
´
2
References
´
A. M. Braun, D. O. Martire and M. C. Gonzalez, J. Phys. Chem.
A, 2000, 104, 3117–3125.
1
2
3
K. Matsuda, S. D. Buckingham, D. Kleier, J. J. Rauh, M. Grauso
and D. B. Satelle, Trends Pharmacol. Sci., 2001, 22, 573–580.
T. Iwasa, N. Motoyama and J. T. Ambrose, Crop Prot., 2004, 23,
´
26 E. A. San Roman and M. C. Gonzalez, J. Phys. Chem., 1989, 93,
3532–3536.
27 K. B. Wiberg, R. E. Stratman and M. J. Frisch, Chem. Phys. Lett.,
1998, 297, 60–64.
28 W. J. McElroy and S. J. Waygood, J. Chem. Soc., Faraday Trans.,
1990, 86, 2557–2564.
29 S. Hirata and M. Head-Gordon, Chem. Phys. Lett., 1999, 302,
375–382.
30 A. D. Becke, Phys. Rev. A: At., Mol., Opt. Phys., 1988, 38,
3098–3100.
3
71–378.
M. L. Dell’Arciprete, L. Santos-Juanes, A. Arques Sanz,
R. Vicente, A. M. Amat, J. P. Furlong, D. O. Martire and
´
M. C. Gonzalez, Photochem. Photobiol. Sci., 2009, 8, 1016–1023.
(a) C. Liang, C. J. Bruell, M. C. Marley and K. L. Sperry,
Chemosphere, 2004, 55, 1225–1233; (b) P. A. Block,
R. A. Brown, D. Robinson, Proceedings of the fourth international
conference on the remediation of chlorinated and recalcitrant
compounds, 2004; (c) K.-C. Huang, Z. Zhao, G. E. Hoag,
A. Dahmani and P. A. Block, Chemosphere, 2005, 61, 551–560.
D. Amarante, Pollut. Eng., 2000, 32, 40–42.
4
31 C. Lee, W. Wang and R. G. Parr, Phys. Rev. B: Condens. Matter,
1988, 37, 785–789.
5
6
32 V. Barone and M. Cossi, J. Phys. Chem. A, 1998, 102, 1995.
33 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr.,
T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam,
S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi,
G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada,
M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida,
T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li,
J. E. Knox, H. P. Hratchian, J. B. Cross, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala,
K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain,
O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford,
J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham,
C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill,
B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople,
Gaussian 03, Revision C.02, Gaussian, Inc., Pittsburgh, PA, 2004.
V. C. Mora, J. A. Rosso, G. Carrillo Le Roux, D. O. Ma
´
rtire and
M. C. Gonzalez, Chemosphere, 2009, 75, 1405–1409.
M. L. Dell’Arciprete, C. J. Cobos, J. P. Furlong, D. O. Ma
M. C. Gonzalez, ChemPhysChem, 2007, 8, 2498–2505.
7
8
9
´
rtire and
P. Neta, R. E. Huie and A. B. Ross, J. Phys. Chem. Ref. Data,
988, 17, 1027–1284.
S. Padmaja, Z. Alfassi, P. Neta and R. Huei, Int. J. Chem. Kinet.,
993, 25, 193–198.
1
1
1
1
1
1
1
1
0 T. L. Luke, H. Mohan, V. M. Manoj, P. Manoj, J. P. Mittal and
C. T. Aravindakumar, Res. Chem. Intermed., 2003, 29(4), 379–391.
1 P. Manoj, R. Varghese, V. M. Manoj and C. T. Aravindakumar,
Chem. Lett., 2002, 74.
2 A. Vieira, J. Telo, H. Pereira, P. Patrocinio and R. Dias, J. Chim.
Phys. Phys.-Chim. Biol., 1999, 96, 116–123.
3 T. Ito, S. Morimoto, S. Fujita and S. Nishimoto, Radiat. Phys.
Chem., 2009, 78, 256–260.
4 G. Bosio, S. Criado, W. Massad, M. C. Gonzalez, N. A. Garcı
and D. O. Martire, Photochem. Photobiol. Sci., 2005, 4, 840–846.
5 P. Wardman, J. Phys. Chem. Ref. Data, 1989, 18, 1637–1755.
´
a
´
6
80 New J. Chem., 2011, 35, 672–680
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011