Rearrangement products in aqueous photolysis of thifensulfuron methyl

Article abstract of DOI:10.1016/j.jphotochem.2017.06.028

Photo-degradation of [¹⁴C]-thifensulfuron methyl has been investigated in aqueous media using a light source which simulates sunlight. Degradation of thifensulfuron methyl proceeds predominantly via sulfonylurea bridge ipso-contraction, and via cleavage of the bridge structure, to yield products in which the thiophene and the triazine rings have disconnected. One significant degradation product, which accounts for nearly 10%, retained both rings with truncated bridge moiety. Surprisingly, this product had thiophene ring substituents rearranged from their original locations. Other laboratories have reported photodegradation of thifensulfuron-methyl, and identified similar degradation products as well. The structure of the rearrangement product has been misidentified in previous reports because the rearrangement of the thiophene ring is not widely recognized. An unambiguous identification of this product and potential rearrangement mechanisms are presented in this report.

Full text of DOI:10.1016/j.jphotochem.2017.06.028

Journal of Photochemistry and Photobiology A: Chemistry 346 (2017) 401–410
Contents lists available at ScienceDirect
Journal of Photochemistry and Photobiology A:
Chemistry
journal homepage: www.elsevier.com/locate/jphotochem
Invited paper
Rearrangement products in aqueous photolysis of thifensulfuron
methyl
*
Ashok K. Sharma , David L. Ryan, Nina L. Marr, Michael P. Wadsley, Steve F. Cheatham
Stine Haskell Research Center, E.I. DuPont de Nemours and Co., Newark, DE 19714, USA
A R T I C L E I N F O
A B S T R A C T
Article history:
Received 9 May 2017
Received in revised form 6 June 2017
Accepted 19 June 2017
Available online 20 June 2017
Photo-degradation of [14C]-thifensulfuron methyl has been investigated in aqueous media using a light
source which simulates sunlight. Degradation of thifensulfuron methyl proceeds predominantly via
sulfonylurea bridge ipso-contraction, and via cleavage of the bridge structure, to yield products in which
the thiophene and the triazine rings have disconnected. One significant degradation product, which
accounts for nearly 10%, retained both rings with truncated bridge moiety. Surprisingly, this product had
thiophene ring substituents rearranged from their original locations. Other laboratories have reported
photodegradation of thifensulfuron-methyl, and identified similar degradation products as well. The
structure of the rearrangement product has been misidentified in previous reports because the
rearrangement of the thiophene ring is not widely recognized. An unambiguous identification of this
product and potential rearrangement mechanisms are presented in this report.
Keywords:
Thifensulfuron methyl
Sulfonylurea
Thiophene rearrangement
Ipso-contraction
© 2017 Elsevier B.V. All rights reserved.
1. Introduction
unambiguous evidence of this rearranged structure, as well as
other photochemical reactions observed was warranted.
Thifensulfuron methyl, (CAS Number: 79277-27-3), CAS name
3-[[[[(4-methoxy-6-methyl-1,3,5-triazin-2-yl)amino]carbonyl]
amino]sulfonyl]-2-thiophenecarboxylic acid, methyl ester; is one
of the sulfonylurea herbicides introduced by E. I. DuPont de
Nemours and Company. Thifensulfuron methyl [abbreviated as
M6316], is used as an herbicide in wheat and barley cultivation.
High water solubility of M6316 [2,24 g/L at pH 7] creates the
potential for run-off water from the treated fields into larger water
bodies. Since photochemical degradation of M6316 is an important
pathway for dissipation in the environment, aqueous photolysis of
M6316 has been a subject of considerable interest, such as, by
Chovelon [1,2] and Headley [3,4]. The degradation pathway and
many of the degradates of Thifensulfuron methyl, have been
reported in these publications. These investigators have not
reported the rearrangement exhibited by thiophene ring during
photochemical degradation of this compound. Ipso contraction of
the sulfonylurea bridge [5–7], is a frequent occurrence for several
sulfonylureas, including thifensulfuron methyl. Degradation along
this route would be anticipated to yield Amine-1 as a degradate
(Fig.1). However, a careful examination of the product showed that
it was Amine-2. Due to unexpected nature of this observation,
Several studies with 14C-test substance were conducted to
insure mass accountability, structure determination and enable
kinetics analysis of the degradation process. We carried out the
aqueous photolysis study in pH 7 buffer, and natural water.
Instability of this compound towards hydrolysis at acidic as well as
basic pH is well known [8]. Influence of pH on photochemical
degradations has also been addressed previously [1]. Neutral pH,
which showed the least hydrolytic degradation for M6316 was
better suited for an investigation of photolysis in water.
2. Materials and methods
2.1. Test substance
Two radiolabeled versions of this [thiophene-2-14C] and
[triazine-2-14C]thifensulfuron methyl, each >95% purity, were
used in this study (Fig. 2). The test substance and stock solutions
of the test substance, were stored in a freezer at <ꢀ5 ꢁC.
2.2. Aqueous media
The 0.01 M pH 7 buffer solution was prepared by adding 195 mL
of 0.02 M sodium phosphate (monobasic) and 305 mL of 0.02 M
sodium phosphate (dibasic) to a 1000-mL volumetric flask. The
* Corresponding author.
E-mail address: ashok.k.sharma@dupont.com (A.K. Sharma).
http://dx.doi.org/10.1016/j.jphotochem.2017.06.028
1010-6030/© 2017 Elsevier B.V. All rights reserved.

402
A.K. Sharma et al. / Journal of Photochemistry and Photobiology A: Chemistry 346 (2017) 401–410
Fig. 1. Degradation Product from Thifensulfuron.
Fig. 2. 14C Isotope Label Location in Test Substance.
contents were diluted with distilled water to produce a 0.01 M pH
7.0 phosphate buffer and final pH adjustments were made with
1.0 N NaOH. The natural water sample was obtained from Lums
Pond located in New Castle County, Delaware, USA. The natural
water was stored at approximately 4 ꢁC, and used within 1 month
of collection. Just before test sample preparation the natural water
as well as buffer solution was filter sterilized by passing through a
(Atlas Electric Devices Co., Chicago, Illinois) for 15 days. The unit
was equipped with a 2200 W air cooled xenon lamp, a UV Special
Suprax1 Glass Filter (290 nm cut-off), a water cooled specimen
tray and an Isotemp Refrigerated Circulator (Fisher Scientific,
Pittsburgh, Pennsylvania). The temperature of each irradiated test
system was controlled by circulating water which was regulated to
maintain the temperature of the samples at 25 ꢂ1 ꢁC. A compari-
son of the spectral output and irradiance of natural sunlight and
the Suntest1 unit showed that this artificial light source is a good
approximation of natural sunlight. The spectral output and
irradiance (463 W-h/m2) of the Suntest1 unit xenon arc lamp
were determined with a model 742 spectroradiometer (Optronic
Laboratories, Inc., Orlando, Florida). One sample for each radiolabel
for the natural water and sterile pH 7 buffer was taken immediately
after the addition of the test substance and subsequent samples
were taken for analysis at 0.5,1, 2, 3, 5, 7, and 15 days. The quantum
yield for the test substance (ØTS) was determined using a chemical
actinometry method [9,10].
0.2 mm filter. Physicochemical Characteristics of the Natural Water
were as follows: pH (7.0), hardness (42 mg eq CaCO3/L),
conductivity (0.20 mmhos/cm), total dissolved solids (72 ppm),
phosphate (0.8 ppm), nitrate (0.8 ppm), sulfate (23 ppm), chloride
(17.5 ppm), and total organic carbon (13.3 ppm).
2.3. Test solutions
For natural water and sterile pH 7 buffer photolysis, the test
vessels contained approximately 100 mL of 5 m
g/mL [14C]thifen-
sulfuron methyl for each radiolabel. The test vessels were
continuously irradiated at 25 ꢂ1 ꢁC for 15 days. Volatiles traps
were connected to each of the irradiated test samples. Separate
dark control test vessels were prepared for each label and were
stored at 25 ꢂ1 ꢁC in a Forma Scientific Chamber (Thermo Forma,
Marietta, Ohio) in the dark until needed for analysis. No volatile
traps were used for the non-irradiated control samples.
An additional abbreviated photolysis experiment was con-
ducted with a test solution of thifensulfuron methyl, to focus on
the identification of the products formed and especially the
rearrangement reactions.
A sterilized pH 7 buffer solution
containing 10 ppm test substance was placed in a glass vessel
covered with a quartz lid for the photo-degradation. The test vessel
was sealed and placed in the photolysis chamber of a Suntest
Model, equipped with a 1500 W xenon lamp and a light filter to
block irradiation below 290 nm. The temperature was controlled
with a continuous circulation water bath and the sample was
irradiated for 72 h. This experiment duration was chosen because
2.4. Photolysis experiments
The photolysis samples were continuously irradiated using a
Suntest XLS+ Enhanced Model bench top xenon exposure system

A.K. Sharma et al. / Journal of Photochemistry and Photobiology A: Chemistry 346 (2017) 401–410
403
the previously conducted studies had indicated peak concentration
of the rearranged product-amine around 24–48 h.
to a solid which was used directly in the following step. A sample
obtained from the previous step (Int-3, 0.0357 mol) was added to a
solution of potassium hydroxide (2.36 g, 0.036 mol) in 75 mL of
methanol, stirred and heated to reflux for 18 h. The mixture was
evaporated to dryness, the residue was triturated with 500 mL
dichloromethane and filtered. The filtrate was evaporated to
dryness and purified by chromatography (silica, 5% acetone/
CH2Cl2) to afford 0.39 g of product that was approximately 72%
pure by HPLC; mp 151–155 ꢁC. A 70-mg sample of this material was
purified further and combined pure fractions afforded Amine-1
(14 mg) that was >97% pure by HPLC; mp 173–174 ꢁC. 1H NMR
2.5. Sample analysis
Samples were analyzed on a HPLC system equipped with a
WatersTM 600E multisolvent Delivery System, WatersTM Model
600 Pump, WatersTM 712 Satellite WISP Autosampler, Applied
Biosciences Model 785A Absorbance Detector, a Radiomatic A-515
TR Flo-One /
1 beta Radio-Chromatography Detector using Radio-
matic FLO-ONE software for Windows. HPLC Conditions: Column
[4.6 mm ꢃ 250 mm MacMod Ace C-18AR column maintained at
40 ꢁC], solvent flow 1.0 mL/min, with UV Detection at 254 nm, HPLC
gradient program utilized solvent A [H2O with 0.1% Formic Acid,
pH = 2.2] and solvent B [ACN with 0.1% Formic Acid]. The gradient
program for separation: Time zero started with 90% A, changing to
55%A over 20 min, then 35%A by 30 min followed by 25%A by
45 min. The two amines displayed baseline separation.
(DMSO-d6)
d 2.41(s, 3H), 3.88 (s, 3H), 3.95 (s, 3H), 7.96 (d, 1H,
J = 5 Hz), 8.20 (d, 1H, J = 5 Hz), 10.2 (s br, NH); HRMS (ESI-QTOF) m/z
[M+H+] calcd for C11H13N4O3S, 281.0708; found, 281.0701; MS2 (ESI
+) m/z 249.04 (100%), 206.04 (31%), 165.01 (23%). A definitive
structural confirmation was based on high resolution 2D-NMR
data which showed the correlation of proton at
d8.2 with its
neighboring carbons at 132.89 and 144.02 as displayed in Fig. 5.
Additional confirmation by 13C NMR added to the identification of
the structure.
NMR data were acquired on
a Bruker 600 MHz Avance
spectrometer using a TCI cryoprobe. Data were acquired on
Amine-2 on a 5.9 mg/0.6 mL DMSO-d6 sample at 40 ꢁC and data on
Amine-1 were acquired on a 5.1 mg/0.6 mL DMSO-d6 sample at
27 ꢁC. Acquisition parameters were identical for each sample with a
D1 of 2 s, 128 scans, 256 ꢃ1024 data points, Fourier transformed to
a 1 K ꢃ 2 K matrix.
2.7. Synthesis of methyl-2-(4-methoxy-6-methyl-1,3,5-triazin-2-yl-
amino)-3-thiophene carboxylate [Amine-2, Fig. 4]
Int-6 was prepared by dissolving 1.57 g (10 mmol) of methyl 2-
aminohthiophene-3-carboxlyate (Int-5, Combi-Blocks, Inc. OS-
7483, CAS 4651-81-4) in dioxane (12 mL) and adding 2.0 g (Int-4,
11 mmol) 2,4-dichloro-6-methoxy-1,3,5-triazine [12] and 2.0 g
(I13.4 mmol) of N,N-diethylaniline. The entire mixture was stirred
and warmed to 80 ꢁC under nitrogen for 6 h resulting in a
suspension that was cooled and diluted with ethyl acetate. The
EtOAc solution was washed with 1N HCl. The organic layer was
washed with brine, dried with Na2SO4 and evaporated to obtain an
amber colored resin (3.1 g). This material was triturated with warm
diethyl ether and then with an equal volume of hexanes to obtain a
solid, which was filtered to afford 2.02 g (67%) of a tan powder.
The mass spec data was acquired using an Agilent 6538 QTOF
mass spectrometer equipped with a dual electrospray ionization
source. The data was acquired in positive ion mode with a mass
range of 100–1000 amu for MS1, and 25–600 amu for MS2. The
isolation width and collision energy for the MS2 experiments were
1.3 amu and 20 V respectively. Nitrogen was used as the collision
cell gas with a pressure of 20 psi. Electrospray Source Parameters
were as follows: Gas temperature 300 ꢁC, drying gas 13 L/min,
Nebulizer: 45 psig, Fragmentor 93 V, and Skimmer 47.
2.6. Synthesis of methyl-3-(4-methoxy-6-methyl-1,3,5-triazin-2-yl-
amino)-2-thiophene carboxylate [Amine-1, Fig. 3]
HPLC analysis of this product [Int-6] showed
a purity of
approximately 70% (220 nm). LC/MS of the main component
A mixture of 5.85 g (0.0357 mol) 2,4-dichloro-6-methyl-1,3,5-
triazine [11] (CAS 1973-04-2, Int-1) and 5.61 g (0.0357 mol) of
methyl 2-aminothiophene-2-carboxylate (CAS 22888-78-4, Int-2)
was stirred at ambient temperature and 4.97 g (0.036 mol) of
potassium carbonate was added. The suspension was stirred for
1 day at ambient temperature; solids were removed by filtration,
rinsed well with dichloromethane and the filtrate was evaporated
exhibited a mass spectrum [M+H+ of 301.01] and its 1H NMR
[(DMSO-d6)
d 3.86 (s, 3H), 4.03 & 4.09 (singlets, 3H total, 2
rotamers), 7.18 (d, 1H), 7.26 (d,1H), 11.0 (NH)], which were
consistent with the structure.
A 0.50 g (1.67 mmol) sample of the product from the previous
step and 1.62 g (5.0 mmol) of cesium carbonate were stirred in dry
dioxane (30 mL) under nitrogen atmosphere for 10 min.
Fig. 3. Synthesis of Methyl-3-(4-methoxy-6-methyl-1,3,5-triazin-2-yl-amino)-2-thiophene carboxylate [Amine-1].

404
A.K. Sharma et al. / Journal of Photochemistry and Photobiology A: Chemistry 346 (2017) 401–410
Fig. 4. Synthesis of Methyl-2-(4-methoxy-6-methyl-1,3,5-triazin-2-yl-amino)-3-thiophene carboxylate [Amine-2].
Fig. 5. 1H and 13C NMR data for Amine-1 and Amine-2.
Trimethylboroxine (Aldrich 323136, CAS 823-96-1, 233 uL,
1.67 mmol) was added followed by 100 mg (0.11 mmol) of (1,10-
bis(diphenylphosphino)-ferrocene)dichloropalladium(II) complex
in dichloromethane (1:1, Aldrich 379670, CAS 72287-26-4) and the
mixture was stirred at 80 ꢁC. The reaction was continued for a total
of 16 h, at which time several by-products started to appear with
no additional increase in the desired product. The reaction mass
was evaporated to dryness, and the residue was partitioned
between EtOAc and aq. 0.5 M HCl. Dark insoluble materials were
removed by filtration. The organic layer was washed sequentially
with water and brine, dried over Na2SO4 and then evaporated to
obtain 280 mg of a resin which was about 38% desired product. The
mixture of products was purified by chromatography on 30 g of
silica gel, first eluting with 5% EtOAc/CH2Cl2 and then with 1:2
EtOAc/hexanes. Pure fractions containing Amine-2 were combined
to afford 67 mg (14%) of the desired compound, mp 157–158 ꢁC,
2.8. Structure confirmation of Amine-1 and Amine-2 by NMR
The 1H and 13C chemical shifts for both structures were
consistent with either of the two structures. In the 13C NMR of both
compounds [Fig. 5], the thiophene-carbons carrying the amine
substituent appeared at a significantly higher chemical shift of
144 ppm or 148.7 ppm as compared to the carbon attached to the
carbonyl group of methyl ester, appearing at 109 ppm or 112 ppm.
This observation was consistent with the location of these
functional groups and supported the two structures as drawn. In
addition, the 1,1-adequate [13,14] 1Hꢀꢀ13C two bond correlation
spectroscopy showed that in Amine-1, the proton appearing at
8.2 ppm was coupled to amine carrying carbon at 144 ppm as the
neighboring carbon and not the carbon carrying the ester group.
On the other hand, in Amine-2, the proton at 7.21 ppm was coupled
to the ester carrying carbon at 111.9 ppm as its neighbor, which
conclusively verified the locations of the functional groups on the
thiophene ring. Therefore 1,1-adequate [two bond specific
correlation NMR data] confirmed the structures of the two
isomeric compounds. The data were unambiguous in both cases
purity >97% by HPLC (220 nm). 1H NMR (DMSO-d6):
d 2.44 (s, 3H),
3.86 (2, 3H), 4.03 (s, 3H), 7.06 (d, J = 5 Hz, 1H), 7.21 (d, J = 5 Hz, 1H),
10.71 (s br, NH); HRMS (ESI-QTOF) m/z [M+H+] calcd for
C
11H13N4O3S, 281.0708; found, 281.0702; MS2 (ESI+) m/z 249.04
(100%), 206.04 (25%), 165.01 (15%).

A.K. Sharma et al. / Journal of Photochemistry and Photobiology A: Chemistry 346 (2017) 401–410
405
Table 1
which were neatly symmetric with the one thiophene proton of
Degradation rates for thifensulfuron-methyl.
Amine-2 correlating to the upfield carbon at 111.9 ppm and the
corresponding thiophene proton of Amine-1 correlating to the
downfield carbon at 144 ppm.
System
Buffera
Rate Constant [k]
DT50
Source
1.74 dꢀ1
0.40 d
0.56 d
0.56 d
3.2 h
2.2 d
0.18 d
with 12C-M6316
Bufferb, pH 7
Nat-Waterb
Buffer, pH 5.5
ACN
1.24 dꢀ1
1.23 dꢀ1
3. Results and discussion
0.6 ꢃ10ꢀ4 sꢀ1
2 ꢃ10ꢀ4 minꢀ1
4.33–3.85 dꢀ1
Ref. [1]
Ref. [2]
Ref. [3]
In the study conducted with radiolabeled M6316 over 15 days,
the material balance throughout the study was in 95.5-106.5%
range. The amount of M6316 showed a rapid rate of decline in
natural water, and the percentage of M6316 decreased from 96.9%
on day 0 to about 7% by day 2 in the samples treated with both the
thiophene-label [TH-label] and the triazine-label [TRZ-label]. The
levels of thifensulfuron methyl were below detection limit by Day
3. The amount of parent compound in the irradiated buffer samples
also decreased rapidly from 96.6% on day 0 to 10.2% at day 2 for the
TH-label samples and from 96.0% on day 0 to 3.7% at day 2 for the
samples treated with TRZ-label. Both radiolabels generated 4.6-
9.8% CO2 in natural water irradiation and in irradiated buffer. A
DT50 of 0.56 days was calculated based on the rate constants
calculated from optimized kinetics illustrated in Fig. 6. Three
studies in literature have reported similar DT50 ranging from 3.2 h
[1], to 2.2 days [2], as shown in Table 1. All studies reported in
literature utilized non-radiolabeled test substance and the experi-
ments were conducted at concentrations of 10 to 40 ppm [mg/L]. In
the main experiments reported here, the test substance concen-
tration was 5 ppm. The dominant degradation pathway is the
sulfonylurea bridge cleavage assisted by dissolved oxygen, as
demonstrated by Aziz et al. [1] and small differences in DT50 are
not surprising. Despite such differences in the DT50’s, the
degradation products reported in all studies were similar, with
the exception of Amine-2.
Lake Water, pH 6.8
a
Limited duration study over 72 h by DuPont.
b
Average of two labels in studies with 14C-thifensulfuron.
The quantum yield (f) of the test substance thifensulfuron
methyl was estimated using the equation:
ꢀ
ꢁ
kTS S l
I
el
A
fTS
¼
fA
ꢃ
kA
S l
I
el
TS
The irradiance values at latitude 40ꢁN were used for the
calculations. Measured actinometer was 5037 while eM6316 was
70.1, summed for 297 to 800 nm wavelengths. Rate constants for
photodecay of thifensulfuron and PNAP in pH 7 buffer were 1.24
and 0.507, respectively. Therefore, quantum Yield of thifensulfuron
methyl in sterile pH 7 Buffer was:
e
fTS = 0.000186[(1.24/0.507) ꢃ (5037/70.1)]
fTS = 0.033
3.1. Degradation products
Rapid degradation of M6316 in all studies (Table 1) supported
that similar products should be anticipated in all experiments. We
did observe most of the degradation products (Figs. 7 and 12)
reported previously [1]. The analytical method gave a good
separation of the anticipated components, especially the signifi-
cant degradates like Deg-3 and Deg-7, as well as the ipso-
contracted amine targeted for identification in this study. The two
possible isomeric amines [Amine-2 and Amine-1] were well
separated. HPLC analysis of the two Amines [1,2] reference
standards (Fig. 8), and their UV spectra showed that the UV
The amount of thifensulfuron methyl in the dark control sterile
buffer pH 7 samples decreased slowly from 97 to 99% parent
compound on Day 0 to 82–88% at Day 15, and producing Deg-1 and
Deg-2 only; indicating only minor contribution from hydrolysis at
pH 7.
The quantum yield of the PNAP [p-nitroacetophenone] acti-
nometer was calculated using the molar concentration of pyridine
[PYR] in the following equations [9,10]:
fA = 0.0169[PYR] = 0.0169 * 0.011 = 0.000186
Fig. 6. Kinetics Analysis of Photodegradation.

406
A.K. Sharma et al. / Journal of Photochemistry and Photobiology A: Chemistry 346 (2017) 401–410
Fig. 7. Representative Chromatogram for Photodegradation.
Fig. 8. HPLC, UV and Mass Spectra of the Amine Isomers.
maxima observed in 225–250 nm range would be useful for
identification of the product formed via photolysis. While the mass
spectra of the two amines, including their MS/MS spectra (Fig. 8)
support the identification, they could elucidate the structures of
the two isomers.
Identification of many products has been detailed in previous
reports, and is included here only to illustrate that our experiment
was not uniquely suited to favor the rearrangement observed. Our
study revealed mostly the same degradation products, except for
the rearrangement products. A mass spectral identification of

A.K. Sharma et al. / Journal of Photochemistry and Photobiology A: Chemistry 346 (2017) 401–410
407
Fig. 9. Mass Spectra of Other Photolysis Products in 72 h Sample.
several products is included in Fig. 9. A significant portion of the
product mix in studies with 14C-M6316 appeared as a number of
peaks, which were minor individually, but collectively amounted
to a substantial portion of the mixture. The reason for such a large
number of products, especially those containing the thiophene
ring, becomes obvious when the photochemical rearrangement of
thiophenes [discussed later] is considered.
Pathway kinetics modelling suggested that the complex
mixture of products observed was likely a result of multiple
degradation processes occurring simultaneously. It further sug-
gested that a degradation pathway leading to amine-2 was an
independent reaction of M6316. A portion of the degraded to polar
compounds, which formed a cluster of peaks near 5 min (Fig. 7),
were modelled as Sink-2 in the kinetics evaluation. Only 10–15% of
the molecule degraded via sulfonylurea bridge ipso-contraction
accompanied by thiophene ring rearrangement, resulting in the
formation of amine-2. This amine was detectable in the chromato-
grams within 8 h after irradiation, at a retention time of 29.3 min.
The product was a single peak and appeared to match the retention
time of amine-2. Proportion of this component increased until 48 h
of photolysis followed by a slow decline. We initially presumed this
product structure to be Amine-1, and only during structure
confirmation, it was noted that the product showed subtle
differences relative to its reference standard, which triggered a
more careful investigation. Literature reports on the rearrange-
ment of thiophene [16,17] ring resulting in scrambling of
substituents during photolysis alerted us to the possibility of
isomers of Amine-1 as a degradate.
3.2. Identification of Amine-2
Several laboratories have reported the aqueous photolysis of
thifensulfuron-methyl. While they reported Deg-3, 4, 6, and 7 as
the degradation products, only Aziz and coworkers [1] reported
amine-1 as an additional photoproduct. This structure was
presumed to have arisen via direct ipso contraction. We suspected
Fig. 10. Proposed Mechanisms for rearrangement.

408
A.K. Sharma et al. / Journal of Photochemistry and Photobiology A: Chemistry 346 (2017) 401–410
amine-2 as the correct structure for this product, due to subtle
differences between the degradate formed and amine-1, and
carried out identification work to be certain of this proposal
including unambiguous synthesis and spectra of the isomers.
Chromatographic separation of the isomeric amines and their
comparison with the photodegradate generated are illustrated in
Figs. 7 and 8. HPLC analysis showed amine-1 and amine-2 as
distinctly separated peaks at 28.8 and 29.3 min, respectively, of
which the latter matched the observed photodegradate by
retention time comparison. Reference standards of these amines
showed molecular ions at 281 (Fig. 8). The mass spectra of the two
isomers displayed minor differences in the fragmentation. A
method which clearly distinguished the two isomers was the UV
spectrum. In particular, the UV maxima around 225–245 nm for
the two isomers were different. The UV spectra coupled with HPLC
retention times enabled an unequivocal determination of the
isomer formed. A comparison of the UV spectra of the reference
standard of Amine-2 and the peak at 29.3 min verified that this
peak was indeed Amine-2 and not Amine-1, due to a nearly perfect
matching of the UV spectrum (Fig. 8).
3.3. Mechanistic considerations for Amine-2
While investigating photochemistry of substituted thiophenes,
Wynberg [16,17] found that thiophenes undergo rearrangement in
a way that the substituents on the ring undergo scrambling their
positions. Wynberg’s work and many other laboratories confirmed
such rearrangements and it is the subject of a review by Rendell
[18]. Two possible mechanisms have been advanced for such a
rearrangement, as illustrated in Fig.10. Intermediate A1 formed via
photolysis [path A], reverts back to the thiophene ring structure
preferentially via sulfur bonding with the carbon carrying amine
portion [and not the carbon carrying ester]. Such an occurrence
would lead to amine-2, a product in which the locations of the ester
and amine have interchanged on the thiophene ring. An alternative
mechanism has also been proposed in which the rearrangement
may proceed via the intermediacy of structure B1. With such a
mechanism, the double bond within cyclobutene keeps migrating
via photochemically induced 1,3-migrations of sulfur, until the
double bond is conjugated to the ester [structure B3].
Thereafter, ring opening of the bicyclic intermediate B3 would
lead to amine-2.
Fig. 11. Thifensulfuron-methyl isomers in Photolysis Mixture.

A.K. Sharma et al. / Journal of Photochemistry and Photobiology A: Chemistry 346 (2017) 401–410
409
Fig. 12. Route of Photodegradation for thifensulfuron-methyl.
Even though the rearrangement mechanism has been
Acknowledgement
explained using thiophene-amine [amine-1] structure, it is not
certain that this rearrangement in fact does occur after the amine
has been formed. The rearrangement may have occurred at an
earlier stage including the parent compound structure. In the
mass spectra recorded on the samples after 72 h of irradiation, we
observed small amounts of several isomers of thifensulfuron-
methyl, all matching the exact mass of the parent compound, but
showed different fragmentation pattern (Fig. 11). These
isomers most likely represent scrambling of the thiophene
substituents. The amounts of parent compound isomers never
exceeded a few percent of the total residue, probably because
further degradation was rapid. Rearrangement of the parent
structure and scrambling of thiophene substituents, helps explain
the complex mixture of products formed during photodegrada-
tion, especially those containing the thiophene ring. Detection of
the rearranged M6316 isomers demonstrated that rearrangement
of thiophene ring played a significant role during the degradation
process.
Contributions from William Zimmerman, John Daub and Rafael
Shapiro are gratefully acknowledged.
References
[1] S. Aziz, S. Dumas, M. Azzouzi, M. Sarakha, J.M. Chovelon, Photophysical and
photochemical studies of thifensulfuron-methyl herbicide in aqueous
solution, J. Photochem. Photobiol. A: Chem. 209 (2010) 210–218.
[2] L. Scrano, S.A. Bufo, M. D'Auria, J.M. Chovelon, Photochemical properties and
degradation by-products of triasulphuron and thifensulphuron-methyl, Int. J.
Environ. Anal. Chem. 86 (2006) 253–264.
[3] J. Headley, J. Du, X. Xu, K. Peru, Kinetics of photodegradation of thifensulfuron-
methyl in aqueous solutions, Commun. Soil Sci. Plant Anal. 33 (15/18) (2002)
3287–3302.
[4] J. Headley, J.L. Du, K. Peru, D.W. McMartin, Mass spectrometry of the photolysis
of sulfonylurea herbicides in prairie waters, Mass Spectrom. Rev. 29 (2010)
593–605.
[5] M. Bottaro, P. Frascarolo, F. Gosetti, E. Mazucco, V. Gianotti, E. Pollico, L.
Piacentini, G. Pavese, M.C. Gennaro, Hydrolytic and photoinduced degradation
of tribenuron methyl studied by HPLC-DAD-MS/MS, J. Am. Soc. Mass Spectrom.
19 (2008) 1221–1229.
[6] J. Rouchaud, O. Neus, C. Moulard, The cyclization transformation of the
sulfonylurea herbicide flupyrsulfuron in the soil of winter wheat crops, Pest
Manage. Sci. 59 (2003) 940.
[7] M. Caselli, Light-induced degradation of metsulfuron-methyl in water,
Chemosphere 59 (8) (2005) 1137–1143.
[8] A.K. Sarmah, J. Sabadie, Hydrolysis of sulfonylurea herbicides in soils and
aqueous solutions: a review, J. Agric. Food Chem. 50 (2002) 6253–6265.
[9] D. Dulin, T. Mill, Development and evaluation of sunlight actinometers,
Environ. Sci. Technol. 16 (11) (1982) 815–820.
[10] Y. Samiullah, Prediction of the Environmental Fate of Chemicals, vol. 1,
Springer Science & Business Media., New York, NY 10010, USA, 2012.
[11] A.G. Arvavenitis, P.J. Gilligan, R.J. Chorvat, R.S. Cheeseman, H. Shen, Non-
peptide corticotropin-releasing hormone antagonists: syntheses and
In conclusion, photodegradation of thifensulfuron-methyl
observed in our laboratory clearly demonstrated thiophene
rearrangement products not reported by other laboratories, even
though many other products observed were identical to those
found by other investigators. Degradation of M6316 is fast with a
half-life of less than 1 day. Degradation products identified are
summarized in Fig. 12, and most of them result from molecular
cleavage into two portions across of the sulfonylurea bridge.
Importantly, amine-2 is substantiated as the correct structure of
the ipso degradation product.

410
A.K. Sharma et al. / Journal of Photochemistry and Photobiology A: Chemistry 346 (2017) 401–410
structure-activity relationships of 2-anilinopyrimidines and triazines, J. Med.
Chem. 42 (1999) 805.
[12] J.R. Dudley, J.T. Thurston, F.C. Schaefer, D.H. Hansen, C.J. Hull, P. Adams,
Cyanuric chloride derivatives, III. Alkoxy-s-triazines, J. Am. Chem. Soc. 73
(1951) 2986–2990.
[16] H. Wynberg, R. Kellog, H. van Driel, G. Beekhuis, The photochemistry of
thiophenes, VII. Observations on the mechanism of arylthiophene
rearrangements, J. Am. Chem. Soc. 89 (1967) 3501–3505.
[17] H. Wynberg, Some observations on the chemical, photochemical, and spectral
properties of thiophenes, Acc. Chem. Res. 4 (1971) 65–73.
[13] B. Reif, M. Köck, R. Kerssebaum, K. Heonjoong, W. Fenical, C. Griesinger,
ADEQUATE, a new set of experiments to determine the constitution of small
molecules at natural abundance, J. Magn. Reson. Ser A 118 (1996) 282–285.
[14] M. Köck, R. Kerssebaum, W. Bermel, A broadband ADEQUATE pulse sequence
using chirp pulses, J. Magn. Reson. Chem. 41 (2003) 65–69.
[18] W. Rendall, M. Torres, E. Lown, O. Strausz, Photoisomerization of furan, pyrrole
and thiophene: the intermediate formation of their dewar and ring contracted
forms, Rev. Chem. Intermed. 6 (1986) 335–364.