Photolysis pathway of imazapic in aqueous solution: Ultrahigh resolution mass spectrometry analysis of intermediates

Article abstract of DOI:10.1021/jf0720279

Direct degradation of imazapic, an herbicide of the imidazoline family, has been investigated in aqueous solution at different concentrations, pH values, and temperatures. The efficiency of the photodegradation process has been evaluated through degradation rate constants that could be fitted best with pseudo-first-order kinetics (C_t = C₀ e^-kt). Ultrahigh resolution mass spectrometry (FTICR/MS) was used in electrospray ionization mode as a tool to study the photolysis process on a molecular level, whereas UV-vis and high-performance liquid chromatography/mass spectrometry analysis were used to follow, by time, the evolution of the intermediates. Taking advantage of the high resolving power of FTICR/MS to perform precise formula assignments taking account of the natural abundance of isotopes, we herein propose and demonstrate an approach using 2D-derived van Krevelen visualization (O/C, H/C, m/z) to confirm the formation of imazapic intermediates. Such an approach allows a qualitative analysis of intermediates and elucidates the plausible reaction pathways of the photolysis process. More than eight photoproducts were separated and identified as a phototransformation of the imidazole ring. A mechanistical pathway was proposed.

Full text of DOI:10.1021/jf0720279

9936 J. Agric. Food Chem. 2007, 55, 9936–9943
Photolysis Pathway of Imazapic in Aqueous
Solution: Ultrahigh Resolution Mass Spectrometry
Analysis of Intermediates
M. HARIR,†,| A. GASPAR,† M. FROMMBERGER,† M. LUCIO,† M. EL AZZOUZI,|
D. MARTENS,‡ A. KETTRUP,§ AND PH. SCHMITT-KOPPLIN*,†
GSF - National Research Center for Environment and Health, Institute of Ecological Chemistry,
Ingoldstädter Landstraße 1, D-85764 Neuherberg, Germany,; LUFA, Obere Langgasse 40,
67346 Speyer, Germany,; Rumbecker Höhe 10, 59821 Arnsberg, Germany,; and University Mohamed
V-Agdal, Faculty of Sciences, Av. Ibn Batouta, BP 1014 Rabat, Morocco
Direct degradation of imazapic, an herbicide of the imidazoline family, has been investigated in
aqueous solution at different concentrations, pH values, and temperatures. The efficiency of the
photodegradation process has been evaluated through degradation rate constants that could be fitted
best with pseudo-first-order kinetics (Ct ) C0 e–kt). Ultrahigh resolution mass spectrometry (FTICR/
MS) was used in electrospray ionization mode as a tool to study the photolysis process on a molecular
level, whereas UV–vis and high-performance liquid chromatography/mass spectrometry analysis were
used to follow, by time, the evolution of the intermediates. Taking advantage of the high resolving
power of FTICR/MS to perform precise formula assignments taking account of the natural abundance
of isotopes, we herein propose and demonstrate an approach using 2D-derived van Krevelen
visualization (O/C, H/C, m/z) to confirm the formation of imazapic intermediates. Such an approach
allows a qualitative analysis of intermediates and elucidates the plausible reaction pathways of the
photolysis process. More than eight photoproducts were separated and identified as a phototrans-
formation of the imidazole ring. A mechanistical pathway was proposed.
KEYWORDS: Photolysis; imazapic; kinetic; HPLC/MS; FTICR/MS; intermediates
1. INTRODUCTION
organic matter and some of the most complex samples of
biogeochemical origins (2, 3). Using this approach, a resolving
power of more than 200000 with 200 ppb mass accuracy can
be achieved, which allows for an exact determination of the
elemental composition with only a few basic chemical constraints.
Over the past 30 years, owing to the extensive use of organic
compounds in natural areas, the amount of pesticides in aqueous
media has increased. Their elimination from natural waters is
due to biotic and abiotic degradation. Photodegradation is one
of the main pathways in the abiotic degradation of pesticides,
and there is a need to clarify their degradation behavior. Because
of the presence of some organic components at trace levels in
natural environments, studying these species at the molecular
level can be an analytical challenge, and laboratory setups are
usually used.
Liquid chromatography coupled to mass spectroscopy (LC/
MS) has been shown to be a powerful tool for the identification
of degradation products in environmental samples (1). However,
high-resolution techniques are usually needed to confirm mo-
lecular structure information. Fourier transform ioncyclotron
resonance mass spectrometry (FTICR/MS) with electrospray
ionization has been successfully applied to characterize natural
Imazapic (IUPAC name: (()-2-[4,5-dihydro-4-methyl-4-(1-
methylethyl)-5-oxo-1H-imidazol-2-yl]-5-methyl-3-pyridinecar-
boxylic acid) is a member of the imidazolinone family.
Imidazolinones are used as selective herbicides (4): alfalfa,
clover, peanuts, and certain trees and shrubs. According to the
American Cyanamid Company that first commercialized imid-
azolinone herbicides, these chemicals are “some of the most
potent herbicides on the market” (5). Upon application, they
are rapidly taken up by plant roots and shoots and are rapidly
transferred to other parts of the plant and accumulate in actively
growing plant tissues (6). Imazapic is soluble in water and is
not degraded hydrolytically in aqueous solution. There are few
published articles focused on the abiotic degradation of the
imidazolinone family and their intermediates.
* Author to whom correspondence should be addressed. Fax: (0049)
089 3187 3358. E-mail: schmitt-kopplin@gsf.de.
† GSF - National Research Center for Environment and Health.
‡ LUFA.
Exposition of the aqueous solution of imazapic to simulated
sunlight is a very useful tool for testing the susceptibility to
photolysis with the aim of understanding the reaction pathways
under natural conditions. In this work, a preliminary kinetic study
of the photodegradation of imazapic was performed. The effects
| University Mohamed V-Agdal.
§ Rumbecker Höhe.
10.1021/jf0720279 CCC: $37.00
2007 American Chemical Society
Published on Web 10/26/2007

Photolysis Pathway of Imazapic
J. Agric. Food Chem., Vol. 55, No. 24, 2007 9937
Figure 2. Effect of pH of the solution on the rate constant of degradation
of imazapic (30 °C; 0.14 mM).
Table 2. Variation of the Rate Constant of Degradation of Imazapic as a
Function of Temperature during the Photolysis Process (Natural pH, 0.14 mM)
Figure 1. Chemical structure and UV/vis spectrum of imazapic.
Table 1. Values of Rate Constant of Degradation as a Function of Imazapic
Concentrations during the Photolysis Process (30 °C, Natural pH)
temperature
(°K)
K (10-3 min-1
)
t1/2 (min)
r2
298
303
313
3.3
3.7
5.5
209.09
186.40
125.45
0.9886
0.9955
0.9894
concentration
(mM)
K (10-3 min-1
)
t1/2 (min)
r2
0.072
0.14
0.2
3.4
3.7
3.9
202.9
186.4
176.9
0.9967
0.9955
0.9984
The calculation of half-life (eq 1) was performed assuming the rate
of the reaction to be first-order in herbicide concentration, by means
of the kinetic equation
of factors affecting the photodegradation such as herbicide con-
centration, pH, and temperature were investigated. UV–vis spec-
trometry, high-performance liquid chromatography/mass spectrom-
etry (HPLC/MS), and FTICR mass spectrometry analysis were used
to characterize the main intermediate products. In parallel, a
possible photolytic degradation pathway of imazapic is proposed.
Ct ) C0 e-kt
(1)
where Ct represents the concentration at time t, C0 represents the initial
concentration, and k is the photolytic rate constant. The fitting of the
experimental data was satisfactory for all the samples. When the
concentration is reduced to 50% of its initial value, the half-life (t1/2
)
can be determined by
2. MATERIALS AND METHODS
2.1. Chemicals. Imazapic (Figure 1), purity > 99%, was purchased
from Dr. Ehrenstofer (Augsburg, Germany). All solvents used for HPLC
analysis were of chromatography grade, and all other chemicals were of
analytical grade and were used without further purification. Water was
purified using a Millipore-MilliQ system (Millipore, Billerica, MA).
2.2. Experimental Procedure. A stock solution of imazapic (0.36
mM) was prepared in pure water. For photodegradation studies,
phosphate buffers (KH2PO4 and K2HPO4) were used. The solutions
were buffed at pH 7.0, and then the pHs was adjusted by the addition
of NaOH or H3PO4. All irradiations of the solutions of imazapic were
performed using a Heraeus (Hanau, Germany) Suntest CPS sunlight
simulator (Xenon lamp; 290–800 nm; 765 Wm-2). The lamp was
positioned 20 cm apart from the cylindrical irradiation vessel. A Pyrex
flask containing 50 mL of imazapic solutions was exposed to the
simulated sunlight irradiation. Test controls were incubated in the dark
to be sure that the transformation of imazapic was only due to light
absorption. The solutions were homogenized by continuous stirring with
a magnetic stirrer. For kinetic studies, samples from irradiated solutions
were taken at regular time intervals and were analyzed directly by
HPLC/diode array detection (DAD). The pH and temperature (30 °C)
remained constant during photodegradation.
t1/2 ) 0.693/k
(2)
3. ANALYSES
3.1. High-Performance Liquid Chromatography. HPLC/
DAD was performed using a Perkin-Elmer Series 200 quaternary
pump equipped with a 235C diode array detector, and a
LiChrospher 100 RP-18 column (250 × 4 mm id, 5 µm) was
used. The mobile phase was a mixture of water (pH ) 3.0;
H3PO4) and acetonitrile (80/20, v/v); the flow rate was 0.9 mL/
min, and the injection volume was 20 µL. For UV–vis spectra,
the DAD output from the HPLC experiments was used.
3.2. Liquid Chromatography/Mass Spectroscopy. The HPLC
unit (Agilent Technologies, Palo Alto, CA) included a G1316A
column oven, a G1329A autosampler with a thermostat
(G1330A), a G1311A quaternary pump, and a G1322A degasser
and was coupled with an API 2000 tandem mass spectrometer
(Applied Biosystems, Toronto, Canada) equipped with an
electrospray ionization (ESI) source. The chromatographic
separation of the compounds was performed on a 75 × 2.0 mm
(4 µm particle size) Synergy MAX-RP C12 column, provided
with a 4 × 2.0 mm precolumn, both from Phenomenex
(Aschaffenburg, Germany). The column oven temperature was
20 °C, and the flow rate was 300 µL/min. As eluents, methanol
and ammonium carbonate (pH 5.0) were employed. The gradient
2.3. Degradation Rate Constant Determination and Calculation
of Half-Life. The degradation of imazapic in the aqueous phase was
modelled with a first-order kinetic. First-order rate constants “k” were
determined from the slope of the linear plot of the natural logarithm of
the imazapic concentration at various sampling intervals against the
irradiation time.

9938 J. Agric. Food Chem., Vol. 55, No. 24, 2007
Harir et al.
Figure 3. LC/MS chromatogram of degradation products of imazapic subjected to xenon lamp (see Figure 6). Pyrex reactor; column conditions: MeOH/
amonium carbonate “ph5.0”, 130 µL/min, 20 °C.
was increased from 10% to 100% methanol in 8 min, held for
1.40 min, then returned to the initial conditions and held until
a total time of 14 min was reached, in order to equilibrate the
column for the next run.
For system control and data acquisition, the Analyst software,
version 1.4, from Applied Biosystems was used. The following
source parameters were employed: TEM, 350 °C; CUR, 45 psi;
IS, 1500 V; GS1, 50 psi; GS2, 60 psi. The focusing potential
was optimized and fixed at 350 V. The compounds were fully
scanned in the first quadrupole between 65 and 320 m/z, at a
dwell time of 1 s. The entrance potential was fixed at 10 eV,
whereas the declustering potential was varied between 10 and
30 eV. The focusing potential was set at 350 eV.
analyzer with a quadrupole and fragmented by varying the
collision voltage of the collision cell (CID). A total of 180 scans
were accumulated for one spectrum.
3.4. Elemental Compositions. FTICR spectra were exported
to peak lists. From those lists, possible elemental formulas were
calculated for each peak in the batch mode by a software tool
written in-house. The generated formulas were validated by
setting sensible chemical constraints (N rule; O/C ratio e 1;
H/C ratio e 2n + 2; element counts, C e 20, H e 30, O e 6,
N e 4), and the calculated formulas were considered valid only
when the corresponding 13C peak was detected in the spectrum.
4. RESULTS AND DISCUSSION
3.3. Fourier Transform Ion Cyclotron Resonance/Mass
Spectrometry. High-resolution mass spectra for molecular
formula assignment were acquired on a Bruker (Bremen,
Germany) APEX Qe Fourier transform ion cyclotron resonance
mass spectrometer (FTICR/MS) equipped with a 12 Tesla
superconducting magnet and an APOLLO I ESI source. The
samples were diluted in methanol to reach a concentration of 1
mg/L. For positive electrospray, water/methanol (50/50, v/v)
was used. Samples were introduced into the microelectrospray
source at a flow rate of 120 µL/h with a nebulizer gas pressure
of 20 psi and a drying gas pressure of 15 psi (250 °C). Spectra
were externally calibrated on clusters of arginine (10 mg/L in
methanol) with a mass accuracy greater than 0.1 ppm. Spectra
were acquired with a time domain of 1 megaword over a mass
range of 100–600 m/z. For fragmentation experiments, ions were
mass-selected between the ion source and the FTICR/MS
4.1. Preliminary Kinetic Study of Imazapic. 4.1.1. Effect
of Concentration. The photolysis of imazapic in aqueous
solution at three different concentrations (0.072, 0.14, and 0.2
mM) was studied. Degradation of the herbicide under simulated
sunlight irradiation showed the same tendency as a function of
the remaining concentration and time of irradiation. The
concentration of imazapic in the dark did not change during
theexperimentaltime.Thisisinagreementwiththeliterature(7,8),
where no hydrolysis reaction under similar conditions was
observed. However, by increasing the herbicide concentration,
slight variations in the rate constants of degradation were
observed (Table 1). These changes were however too low to
be considered as significant. These results suggest that there is
no concentration effect in imazapic photodegradation as was
previously observed for imazapyr (9).
Figure 4. Absorbance spectral changes of imazapic and the byproducts (P2) and (P3) (see Figure 3) that occur during the photolysis process of an
aqueous solution using HPLC/DAD.

Photolysis Pathway of Imazapic
J. Agric. Food Chem., Vol. 55, No. 24, 2007 9939
Figure 5. Evolution curve of imazapic degradation and its byproducts “xenon lamp, Pyrex reactor” (see Table 3). The insert shows the first-order linear transforms
ln(C/C0) ) f(t) of imazapic degradation. Conditions: C0 ) 0.14 mM, T ) 30 °C.
4.1.2. pH Effect. The effect of pH on the photolysis of
imazapic using a solar UV simulator and a Pyrex reactor (λ
g 290 nm) was studied in aqueous solution at pH values
ranging from 2 to 9. The rate constant of degradation of
imazapic increases as the pH of the reaction solution rises
from 2 to 5; beyond 5, the rate appears to plateau, as shown
in Figure 2. The disappearance of imazapic follows first-
order kinetics, and the rate constants of degradation (0.8,
1.26, 4.1, 4.2, 4.2, 3.81, and 3.695 × 10-3 min-1) were
reached when the solution pH was varied between 2 and 9.
As shown in Figure 2, the rates of degradation increase by
increasing the pH and appear to plateau at pH g 5. This
could be explained probably by the fact that the molar
absorption coefficients of the molecule are slightly higher at
higher pH compared to the acidic pH (> pH 5). Similar
behavior was reported for the degradation of imazaquin (10)
and imazapyr (8).
4.1.3. Temperature Effect. Increased temperature is known
to enhance the degradation of pesticides by accelerating both
abiotic chemical reactions and microbial activity (11–13). The
effect of temperature on imazapic photodegradation was studied
at 25, 30, and 40 °C with a temperature accuracy of (0.5 °C.
As expected, the degradation of imazapic was significantly
longer at 25 and 30 °C than at 40 °C (Table 2). The rates of
degradation obtained showed results with slightly differing
values at 25 (3.3 × 10-3 min-1) and 30 °C (3.7 × 10-3 min-1
)
Figure 6. Extracted ion chromatogram of peaks P1, P2, P3, and imazapic obtained from LC-MS analysis (Figure 3).

9940 J. Agric. Food Chem., Vol. 55, No. 24, 2007
Table 3. FTICR/MS and LC/MS Analysis Data of Imazapic and Its Byproducts (in Positive Mode)a
in FTICR-MS system
Harir et al.
in LC/MS system
intermediates
experimental mass
theoritical mass
error ((ppm)
13C peaks
formula structure
extracted ion mass
tr (min)
P1a
P1b
P2a
P2b
P2c
P3
P3b
P4a
P4b
P5
imazapic
163.05027
181.06079
148.03930
166.04989
180.06554
233.12844
191.08136
164.03424
182.04482
248.13935
276.13426
163.05020
181.06077
148.07930
166.04987
180.06552
233.12845
191.08150
164.03422
182.04479
248.13935
276.13427
0.404
0.115
0.400
0.120
0.110
0.061
0.755
0.121
0.192
0.013
0.030
164.05356a
182.06412a
b
167.05326
181.0689
234.13181
192.08486a
165.04235a
183.04817
249.14265
277.13762
C8H7N2O2
C8H9N2O3
C8H6NO2
162.9
180.9
148.0
166.0
180.1
233.0
noc
no
no
no
276.2
1.11
1.12
1.51
1.50
1.42
4.62
C8H8NO3
C9H10NO3
C13H17N2O2
C10H11N2O2
C8H6NO3
C8H8NO4
C13H18N3O2
C14H18N3O3
5.46
a Small peaks. b Too small peak. c No: not extracted.
as compared to that at 40 °C (5.5 × 10-3 min-1). However,
temperature dependence implies that the degradation of pesti-
cides, as in the case of imazapic, would increase in the hot
seasons, volatilizing them from the water interface. For the same
reasons, their degradation would decrease in cold seasons, where
the risk of pesticide accumulation may increase. To simulate
better laboratory conditions, a temperature of 30 °C was selected
for further experiments.
4.2. Analysis of Photoproducts. 4.2.1. Photolysis. HPLC/
MS analyses were carried out in order to follow the main
intermediates resulted from the photolytic process of imazapic.
ESI in the positive mode was used to detect them. As described
in Figure 3, the HPLC chromatogram shows the presence of
three peaks corresponding to the intermediates P1, P2, and P3 at
retention times 1.11, 1.50, and 4.62 min, respectively. The
imazapic peak appears at 5.62 min with a calculated half-life
of 11.03 h-1
.
The gradual disappearance of imazapic and the appearance
of the intermediates were monitored during the photolysis
process by recording the UV spectra of samples in the 200–280,
320, and 330 nm ranges. Typical spectra are shown in Figure
4. The characteristic absorption almost begins to disappear after
20 h of irradiation time for imazapic (Figure 4), and begins to
appear after 2 h of irradiation time for the intermediates (Figure
4; P2 and P3) under the experimental conditions.
When irradiated at λ > 290 nm, a 10 ppm imazapic aqueous
solution was degraded quite slowly related to its kinetic of
disappearance. After 2 h of irradiation, the remaining imazapic was
∼98%. By increasing the time of irradiation, the amount decreased
from ∼77% in 6 h to ∼48% after 14 h of irradiation. After one
day, the amount had decreased to 22%, as shown in Figure 5.
The evolution and apparition of imazapic and its intermedi-
ates, respectively, as a function of the irradiation time is shown
in Figure 5. We can clearly observe that the intermediates P3
and P2b are not stable under sunlight irradiation since they are
in turn phototransformed. A maximum concentration was
observed within 18 h of irradiation. For both intermediates P1b
and P2c, a longer irradiation time is required to be sure of their
evolution.
4.2.2. Extract Mass Ions Analysis. Figure 6 shows extracted
mass spectra of each peak present in LC/MS chromatogram
(Figure 4). Five peaks at different retention times and with
different m/z ratios were observed as compared with the blank
sample at time zero. At retention time 4.62 min (Figure 3, P3),
a peak at m/z ) 233.0 was extracted. By considering the
retrosynthesis way, we suggest that this mass corresponds to
the loss of one molecule of isocyanic acid (HNCO) due to the
transposition of the R-proton of the imidazole ring (as shown
before with imazamox (14)).
Figure 7. Visualization of 2D-derived Van Krevelen’s diagram between
theoretical (∆, O/C; ), H/C) and experimental (2, O/C; (, H/C) mass
ratios of intermediates of imazapic in correlation with the FTICR mass
spectrum of imazapic degradation (the signal-to-noise ratio of the basic
peak is 8271:1).
The extracted ion chromatograms of m/z ) 148.0, P2a; m/z
)166.0, P2b; and m/z ) 180.1, P3c (Figure 6) showed peaks
with slightly different retention times, all contributing to the
peak at around 1.5 min (Figure 3; P2). For m/z ) 162.9, P1a,
and 180.9, P1b (Figure 6), the findings were similar (Figure 3;
P1). On the basis of the data obtained by LC/MS analysis (Table
3) and in the absence of availably standard compounds, we were
unable to have a clear view of the intermediates produced.
Consequently, a new investigation was used to clarify and to
characterize the main intermediates of imazapic resulting from
the photolysis process (see the FTICR/MS section).
4.3. FTICR-MS. 4.3.1. 2D-DeriVed Van KreVelen Diagrams
Visualization. Generally, Van Krevelen diagrams illustrate a plot
of each elemental composition present in the FTICR mass

Photolysis Pathway of Imazapic
J. Agric. Food Chem., Vol. 55, No. 24, 2007 9941
Figure 8. Mass scale of A, B, C, and D segments of intermediates at molecular masses of m/z 163.05027, 164.03424, 166.04989, 180.06554, 181.06079,
182.04482, and 248.13935 in the FTICR mass spectrum of Figure 7 within the mass window m/z 5.0.
nine intermediates are observed with different elemental dis-
tributions and all of which were singly charged as proven by
examination of the 12C/13C peak distance. By superimposing
the FTICR mass spectrum (Figure 7) to a 2D-derived Van
Krevelen plot, we can clearly observe the correlation existing
between each separated group of intermediates related to the
mass spectrum. On the basis of this applicable analysis, we are
able, first, to attribute for each contribution of H/C and O/C
ratio represented in a 2D-derived van Krevelen diagram an exact
mass in the FTICR spectrum and, second, to separate all
intermediate products from each other in silico without further
classical separation (by, e.g., chromatography).
Additional interpretation requires mass scale expansion to
resolve different elemental compositions at a given mass range
(Figure 8). The mass accuracy available from FTICR/MS for ions’
molecular mass ([M + H]+ ) 163.05027, 164.03424, 166.04989,
Figure 9. Further mass scale-expanded segment of the imazapic mass
180.06554, 181.06079, 182.04482, and 248.13935) makes possible
a unique assignment of elemental compositions for all seven of
the resolved mass spectral peaks. Those elemental compositions
in turn reveal the formula structure and the nature of the
rupture.
spectrum of Figure 7, allowing for the visual resolution and elemental
composition assignment (on the basis of accurate mass measurement)
of the natural abundance of isotopes of (E)-5-methyl-2-((3-methylbut-2-
en-2-ylimino)methyl)nicotinic acid (its signal-to-noise ratio is 8271:1). The
tabulated data for this byproduct show an average mass error of 0.062
ppm and a mass segment scale of m/z 3.0.
4.3.2. Natural Abundance of 13 Carbon Isotopes of Imazapic.
The relative natural abundance of stable isotopes associated with
a given ionic species in a mass spectrum can be used as a
diagnostic tool to distinguish between the various atomic
combinations. Because of the natural abundance of carbon-13
(1.1%); nitrogen-15 (0.36%); and oxygen-18 (0.020%), several
software packages are available for the analysis of mass
spectrometry data. The dominance of carbon-13 isotopes was
previously used to explain the even m/z peaks in high-resolution
mass spectra of complex mixtures (16). Then, an investigation
based on isotopes approach can offer a much greater under-
standing of the photolysis process of imazapic. As shown in
Figure 7, all peaks observed in the FTICR mass spectrum of
spectrum onto two or three axes according to its H/C, O/C, and/
or N/C atomic ratios. The H/C ratio separates compounds
according to the degree of saturation, whereas O/C or N/C ratios
separate according to O and N classes (15). According to the
results obtained in the case of imazamox (14), we herein propose
and demonstrate a new approach using 2D-derived van Krevelen
visualization to characterize the main intermediates of the
photolysis process of imazapic. In the 2D-derived van Krevelen
visualization, a plot is generated with the atomic ratio of
hydrogen to carbon on the y axis, left; the ratio of oxygen to
carbon on the y axis, right; and the nominal mass (m/z) of the
intermediates on the x axis (Figure 7). At a given mass range,

9942 J. Agric. Food Chem., Vol. 55, No. 24, 2007
Harir et al.
Figure 10. Degradation pathway proposed for the photolysis of imazapic in pure water.
imazapic are singly charged, and the principal equivalent
isotopic masses of each intermediate are shown in Table 3.
Pathway A. The reduction of the acidic group by the loss of
one carbon monoxide (C)O) fragment is the most probable
route to obtain the 4-tert-butyl-2-(3-hydroxy-5-methylpyridin-
2-yl)-4-methyl-1H-imidazol-5(4H)-one (P5, C13H18N3O2; [M +
H]+ ) 248.13935) intermediate as a decarbonylation process
of imazapic.
As an example, the display of the abundant intermediate at
[M + H]+ ) 233.12844 is shown in Figure 9. Every isotope
has a unique, characteristic “mass defect”; so the mass of the
ion, which shows the total mass defect, identifies its isotopic
and elemental composition. This figure reveals the presence of
Pathway B. The transposition of the R proton of the imidazole
cycle leads to the formation of the (E)-5-methyl-2-((3-methylbut-
2-en-2-ylimino)methyl) nicotinic acid (P3, C13H17N2O2; [M +
H]+ ) 233.12844) intermediate as shown in Figure 10.
However, the reduction of the acidic group and the cleavage of
ethylene bond (C)C) of the P3 (C13H17N2O2) intermediate leads
to the formation of (E)-2-((acetylimino)methyl)-5-(metho-
12
12
12
12
C ,
13
C
12
13C1,
C
12
13C2, and
C
13
18O16O species corre-
sponding to the natural abundance of isotopes of this intermedi-
ate (C13H17N2O2).
4.3.3. Proposed Degradation Pathways. Essentially, as de-
scribed before with imazamox (14), the main intermediates of
imazapic were identified and characterized by using ESI/FTICR
mass spectrometry in the positive mode. It is known that high-
resolution mass spectra permitted a determination of exact ions’
mass values (m/z) and the elemental composition of each
compound. On the basis of the fragmentation pathways analysis,
a general schema of the photolysis process of the main
intermediates is proposed in Figure 10. As a summary, Table
3 shows the formula structure for the proposed degradation
intermediates resulting from the photolysis of imazapic in pure
water.
xymethyl)nicotinic acid (P3a, C10H11N2O2; [M + H]+
)
191.08136).
Pathway C. Initially, the cleavage of both amine and imine
groups ()N–C and NH–C)O) may led to the formation of
2-(1H-diazirin-3-yl)-5-methylnicotinic acid (an unstable inter-
mediary was not detected), as shown in Figure 10. Then, the
opening of the 1H-diazirin-3-yl cycle is probably the only way
to obtain the 2-carbamimidoyl-5-methylnicotinic acid (P2c,
C9H10NO3; [M + H]+ ) 180.06554) intermediate.
Pathway F. The formation of 2-carbamoyl-5-methylnicotinic
acid (P1b, C8H9N2O3; [M + H]+ ) 181.06079) can be explained
by the cleavage of both C)N and C–N bonds of the imidazole
cycle. Whereas, the formation of the 3-methyl-5H-pyrrolo[3,4-
However, due to the high degree of photodecomposition of
the imidazole ring under sunlight irradiation (10–17) and on
the basis of a previous study (14), the main photoproducts
obtained from the photolysis process of imazapic are most
probably resultant from the pathways cited below.

Photolysis Pathway of Imazapic
J. Agric. Food Chem., Vol. 55, No. 24, 2007 9943
b]pyridine-5,7(6H)-dione (P1a, [M + H]+ ) 163.05027;
C8H7N2O2) intermediate can be explained by the loss of one
water molecule, as excepted by the Nourish reaction (18).
Pathway E. The plausible structure of the 5-methyl-2,3py-
ridinecarboxylic acid (P4b, [M + H]+ ) 182.04482; C8H8NO4)
intermediate is shown in Figure 10. In the absence of a
hydrolysis reaction, the presence of this intermediate could be
explained by the fact that light may modify (by a successive
oxidation) the structure of the imidazole ring, and an enolic
equilibrium in the ground state as well as in the first excited
state could be envisaged (10). A similar degradation pathway
has already been proposed for the photolysis of imazapyr (17)
and imazamox (14). On the other hand, the formation of the
2-formyl-5-methylnicotinic acid (P2b, [M + H]+ ) 166.04989;
C8H8NO3) intermediate could be interpreted by a reduction of
the 5-methyl-2,3-pyridinecarboxylic acid compound.
ionization and electrospray ionization. Energy Fuels 1997, 11,
554–560.
(4) Goatz, A.; Lavy, T.; Gbur, E. Degradation and field persistence
of Imazethapyr. Weed Sci. 1990, 38, 421–489.
(5) Loux, M.; Liebl, R.; Slife, F. Avaibility and persistence of
imazaquin, imazethapyr, and clomazone in soil. Weed Sci. 1989,
37, 259–267.
(6) Loux, M.; Reese, K. Effect of soil type and pH on persistence
and carryover of imidazolinone herbicides. Weed Technol. 1993,
7, 259–267.
(7) Laifer, A. The kinetics of enVironmental Aquatic Photochemistry;
American Chemical Society: Washington, DC, 1988; p 49.
(8) Mallipudi, N. M.; Stout, S. J.; da Cunda, A. R.; Lee, A. H.
Photolysis of Imazapyr (AC 243997) Herbicide in Aqueous Media.
J. Agric. Food Chem. 1998, 39, 412–417.
(9) El Azzouzi, M.; Mountacer, H.; Mansour, M. Kinetics of
photochemical degradation of imazapyr in aqueous solution.
Fresenius EnViron. Bull. 1999, 8, 709–717.
(10) Barkani, H.; Catastini, C.; Emmelin, C.; Sarakha, M.; El Azzouzi,
M.; Chovelon, J. M. Study of the phototransformation of
imazaquin in aqueous solution: a kinetic approach. J. Photochem.
Photobiol., A. 2004, 170, 27–35.
(11) Getzin, L. W. Factors influencing the persistence and effectiveness
of chlorpyrifos in soil. J. Econ. Entomol. 1985, 78, 412–418.
(12) Racke, K. D. Environmental fate of chlorpyrifos. ReV. EnViron.
Contam. Toxicol. 1993, 131, 1–150.
(13) Bondarenko, S.; Gan, J. Degradation and sorption of selected
organophosphate and carbamate insecticides in urban stream
sediments. EnViron. Toxicol. Chem. 2004, 23, 1809–1814.
(14) Harir, M.; Frommberger, M.; Gaspar, A.; Martens, D.; Kettrup,
A.; El Azzouzi, M.; Schmitt-Kopplin, Ph. Characterization of
Imazamox degradation by-products by using Liquid Chromatog-
raphy Mass Spectrometry and High Resolution Fourier Transform
Ion Cyclotron Resonance Mass Spectrometry. Anal. Bioanal.
Chem. 2007. [Online] DOI: 10.1007/s00216-007-1343-7.
(15) Wu, Z.; Rodgers, R. P.; Marshall, A. G. Two- and three-
dimensional van krevelen diagrams: a graphical analysis comple-
mentary to the Kendrick mass plot for sorting elemental compo-
sitions of complex organic mixtures based on ultrahigh-resolution
broadband fourier transform ion cyclotron resonance mass mea-
surements. Anal. Chem. 2004, 76, 2511–6.
Pathways E′ and F′. The loss of one water molecule from
5-methyl-2,3-pyridinecarboxylic acid (P4b, [M + H]+
)
182.04482; C8H8NO4) and 2-carbamoyl-5-methylnicotinic acid
(P1b, [M + H]+ ) 181.06079; C8H9N2O3) intermediates leads
to the formation of 3-methylfuro[3,4-b]pyridine-5,7-dione (P4a,
[M + H]+ ) 164.03424; C8H6NO3) and 3-methyl-5H-pyr-
rolo[3,4-b]pyridine-5,7(6H)-dione (P1a, [M + H]+ ) 163.05027;
C8H7N2O2), respectively, as was the case for imazamox (14)
and imazapyr (17). The presence of such intermediates could
be explained by equilibrium between the cyclized and the
opened forms in aqueous solution.
In summary, it was found that the photolysis efficiency of
imazapic was more effective at higher pH and temperatures,
but no significant concentration effect was observed. However,
by combining the data from ultrahigh resolution mass spec-
trometry and liquid chromatography/mass spectrometry, nine
intermediates were identified and characterized. The 2D-derived
van Krevelen visualization is shown to be an effective and
informative graphical method of analysis. This approach was
usefully applied in the cases of imazamox (14), previously, and
imazapic and could be taken as a general applied method to
analyze similar degraded pollutants.
(16) Jan, S. P. Fourier Transform-Ion Cyclotron Resonance-Mass
Spectrometer as a New Tool for Organic Chemists. Synlett 1999,
2, 249–266.
LITERATURE CITED
(17) Quivet, R.; Faure, J.; Georges, J. O. P.; Herbreteau, B. Kinetic
study of imazapyr photolysis and characterization of the main
photoproducts. Toxicol. EnViron. Chem. 2004, 86, 197–206.
(18) Norrish, R. G. W.; Bamford, C. H. Photodecomposition of
aldehydes and ketones. Nature 1937, 140, 195.
(1) Ferrer, I.; Thurman, E. M. Liquid Chromatography Mass Spec-
troscopy/Mass Spectroscopy, MS/MS and Time-of-Flight MS:
Analysis of emerging Contaminants. American Chemical Society
Symposium Series 850; Oxford University Press: Washington, DC,
2003.
(2) Brown, T. L.; Rice, J. A. Effect of Experimental Parameters on
the ESI FT-ICR Mass Spectrum of Fulvic Acid. J. Anal. Chem.
2000, 72, 384–390.
Received for review July 6, 2007. Revised manuscript received August
30, 2007. Accepted August 30, 2007. The authors thank the German
Academic Exchange Service (DAAD) for a grant to M.H. and the
German–Israeli Foundation for Scientific Research and Development
(GIF) for financial support of M.H. and M.F.
(3) Fievre, A.; Solouki, T.; Marshall, A. G.; Cooper, W. T. High-
resolution Fourier transform ion cyclotron resonance mass
spectrometry of humic and fulvic acids by laser desorption/
JF0720279