Sulfonylurea herbicides

Charles R. Powley

DuPont Crop Protection, Newark, DE, USA

1

Introduction

Sulfonylurea herbicides were first introduced in 1982 by DuPont Crop Protection.
They are typically applied at rates less than 100 g ha

−1

, have low mammalian tox-

icity, and degrade to innocuous compounds after application. They are used to control
a variety of broad-leafed weeds and grasses in cereals and other row crops and for
industrial weed control. The biological mode of action is via inhibition of acetolactate
synthase (ALS), an enzyme that is found in plants, but not in animals.

1

Approximately

25 sulfonylurea herbicides are currently registered for agricultural uses on a global
basis. A few examples of these molecules are shown in Figure 1. The compounds
are characterized by the presence of a sulfonylurea bridge between two heterocyclic
moieties. Sulfonylureas are both chemically and thermally unstable. Rapid hydrolytic
cleavage of the sulfonylurea bridge occurs in aqueous acidic solutions; most sulfony-
lureas demonstrate improved stability and solubility in aqueous neutral to slightly
alkaline solutions, where they exist in the anionic form through the loss of one of the
urea hydrogen atoms. These compounds, generally, are stable in organic solvents and
are soluble at levels greater than 1 mg per 100 mL in most common organic solvents,
with the exception of hydrocarbons.

Sulfonylureas are not directly amenable to gas chromatography (GC) because of

their extremely low volatility and thermal instability. GC has been used in conjunction
with diazomethane derivatization,

2

,3

pentafluorobenzyl bromide derivatization,

4

and

hydrolysis followed by analysis of the aryl sulfonamides.

5

These approaches have

not become widely accepted, owing to poor performance for the entire family of
sulfonylureas. Capillary electrophoresis (CE) has been evaluated for water analysis

6

–

8

and soil analysis.

9

The low injection volumes required in CE may not yield the

required sensitivity for certain applications. Enzyme immunoassay has been reported
for chlorsulfuron

10

and triasulfuron,

11

with a limit of detection (LOD) ranging from

20 to 100 ng kg

−1

(ppt) in soil and water.

The most common approaches to sulfonylurea determinations involve high-

performance liquid chromatography (HPLC). The earliest reported methods utilized
normal-phase liquid chromatography (LC) with photoconductivity detection;

12

,13

this

type of detector demonstrated undesirably long equilibration times and is no longer

Handbook of Residue Analytical Methods for Agrochemicals.

C

2003 John Wiley & Sons Ltd.

Sulfonylurea herbicides

401

N

O

O

S

N

H

N

H

N

O

N

O

O

O

Metsulfuron methyl

N

H

N

H

N

S

S

N

N

O

O

O

O

O O

O

Rimsulfuron

N

N

N

O

O

S

N

H

N

O

O

O

O

Tribenuron methyl

N

N

N

O

S

N

N

H

O

O

O

O

O

F

F

F

O

Na

+

Flupyrsulfuron methyl

N

S

N

N

H

N

H

N

N

O

O

O

O

O

O

Nicosulfuron

S

N

N

S

N

H

N

H

O

O

O

O

N

O

O

Thifensulfuron methyl

Figure 1

Structures of selected sulfonylurea herbicides

commercially available. More recent methods use reversed-phase HPLC with either
ultraviolet (UV) or mass spectrometry (MS) detection. HPLC/UV methods have been
reported for the determination of selected sulfonylureas in soil and water;

14

–

16

in many

cases, the sensitivity of these methods is not adequate, and the lack of specificity usu-
ally requires extensive cleanup and/or complicated column-switching arrangements
and mobile phase gradients. Early applications of HPLC with MS detection [liquid
chromatography/mass spectrometry (LC/MS)] involved thermospray ionization,

17

fast-atom bombardment,

18

,19

and direct liquid introduction.

20

However, quantitative

determination of sulfonylureas by LC/MS was not widespread until electrospray in-
terfaces were developed.

21

–

24

The current best practice for trace-level sulfonylurea

determination in biological and environmental matrices is HPLC with a positive-
ion electrospray interface and tandem mass spectrometry (MS/MS) detection of at
least one parent to daughter ion transition using either an ion trap or, preferably, a
triple-quadrupole mass spectrometer. The sensitivity and selectivity obtained by liquid
chromatography/tandem mass spectrometry (LC/MS/MS) meets the most strin-
gent regulatory criteria for detection, quantitation, and confirmation.

25

,26

All of the

procedures summarized in this article recommend LC/MS/MS as the means of de-
tection for these reasons.

402

Compound class

2

Analytical methodology

LC/MS/MS is the preferred means of detection, quantitation, and confirmation of
sulfonylurea herbicides in biological and environmental matrices. Therefore, rec-
ommendations for establishing and optimizing LC/MS/MS analyses common to all
matrices are given first, followed by specific rationales for methods and sample prepa-
ration techniques for plant, soil, and water matrices.

2.1

LC/MS/MS analysis

A triple-quadrupole mass spectrometer with an electrospray interface is recommended
for achieving the best sensitivity and selectivity in the quantitative determination of
sulfonylurea herbicides. Ion trap mass spectrometers may also be used, but reduced
sensitivity may be observed, in addition to more severe matrix suppression due to
the increased need for sample concentration or to the space charge effect. Also, we
have observed that two parent to daughter transitions cannot be obtained for some
of the sulfonylurea compounds when ion traps are used in the MS/MS mode. Most
electrospray LC/MS and LC/MS/MS analyses of sulfonylureas have been done in the
positive ion mode with acidic HPLC mobile phases. The formation of (M

+ H)

+

ions

in solution and in the gas phase under these conditions is favorable, and fragmentation
or formation of undesirable adducts can easily be minimized. Owing to the acid–base
nature of these molecules, negative ionization can also be used, with the formation of
(M

− H)

−

ions at mobile phase pH values of approximately 5–7, but the sensitivity

is often reduced as compared with the positive ion mode.

Reversed-phase liquid chromatography with a C

8

, phenyl, or C

18

column (or equiv-

alent) is recommended. A binary pumping system capable of producing a linear
gradient is sufficient. Water and methanol, both acidified with equal amounts of acetic
acid, are used to form the gradient. Table 1 provides an example of HPLC conditions
used for the determination of 13 sulfonylureas with a wide range of polarity. In this
example, aqueous samples (100 µL) are injected at a weak mobile phase composition
to facilitate on-column focusing, followed by a steep gradient to facilitate removal of
as many of the matrix components as possible. The analytes are then eluted between
11 and 16 min, followed by cleaning and re-equilibration periods. Baseline resolu-
tion of the analytes is not obtained and is not necessary since they all have different
mass spectral transitions. A switching valve is used to divert the HPLC effluent to
waste before and after the 11–16-min time period, in order to reduce source con-
tamination and to enable more samples to be analyzed before the source needs to
be cleaned. Since the electrospray interface works optimally at low flow rates, the
HPLC flow is split post-column such that only 100 µL min

−1

actually passes through

the interface (approximately 10 : 1 split), while the remainder is diverted to a waste
container.

The MS/MS response for each analyte must first be optimized on the specific

instrument to be used. This is usually done by infusing a solution of the analyte into the
HPLC mobile phase without a column present. The composition of the mobile phase
should match that expected at the time of analyte elution within

±25%. The instrument

is first operated in the LC/MS mode, and the settings for the electrospray interface are

Sulfonylurea herbicides

403

Table 1

Example HPLC conditions for the determination of sulfonylurea herbicides

by LC/MS/MS

System

Agilent 1100 HPLC

Column

4.6-mm i.d.

×15 cm, Phenomenex C

8

analytical column

with 3-

µ

m-diameter packing

Column temperature

40

◦

C

Injection volume

0.100 mL

Autosampler temperature

4

◦

C

Flow rate

1.0 mL min

−1

Conditions

Time

A

a

(%)

B

a

(%)

0.0

75

25

2.0

75

25

12.0

30

70

15.0

20

80

16.0

10

90

18.0

10

90

18.5

75

25

23.0

75

25

Analyte

Retention time (min)

Nicosulfuron

12.0

Sulfometuron methyl

12.3

Thifensulfuron methyl

12.4

Metsulfuron methyl

12.9

Ethametsulfuron methyl

13.0

Rimsulfuron

13.2

Tribenuron methyl

13.7

Chlorsulfuron

14.2

Bensulfuron methyl

14.6

Azimsulfuron

14.6

Triflusulfuron methyl

15.6

Chlorimuron ethyl

15.8

Flupyrsulfuron methyl

15.9

Total run time

23.0

a

A

= 0.05% acetic acid; B = 0.05% acetic acid in methanol.

optimized to provide maximum response for the (M

+ H)

+

ion; this process is usually

automated. Then, the settings for the collision cell are optimized to produce maximum
response of one or two characteristic daughter ions. Most modern instruments allow
this to be done automatically. In general, sulfonylureas are very amenable to positive
ion electrospray MS and MS/MS analysis, with excellent sensitivity compared with
most other agrochemicals, and optimal responses can be easily obtained by proficient
operators.

A minimum number of transitions should be monitored at any given time during

the course of analysis. If only a few well-resolved peaks are to be monitored, the
groups of ions to be acquired may easily be changed so optimum sensitivity is ob-
tained. If many closely eluting or even overlapping chromatographic peaks are to be
monitored, acquiring too many signals at any given time will result in more poorly

404

Compound class

Table 2

Suggested ion transitions for the determination of sulfonylurea herbicides by LC/MS/MS

Analyte

Primary (quantitative) transition

Secondary (confirmatory) transition

Nicosulfuron

410

.9 → 213.0 ± 0.2

410

.9 → 182.0 ± 0.2

Sulfometuron methyl

365

.0 → 150.0 ± 0.2

365

.0 → 199.0 ± 0.2

Thifensulfuron methyl

387

.9 → 167.0 ± 0.2

387

.9 → 205.0 ± 0.2

Metsulfuron methyl

382

.0 → 167.0 ± 0.2

382

.0 → 199.0 ± 0.2

Ethametsulfuron methyl

410

.9 → 196.0 ± 0.2

410

.9 → 168.0 ± 0.2

Rimsulfuron

431

.9 → 182.0 ± 0.2

431

.9 → 325.0 ± 0.2

Chlorsulfuron

358

.0 → 167.0 ± 0.2

358

.0 → 141.0 ± 0.2

Azimsulfuron

425

.0 → 182.0 ± 0.2

425

.0 → 244.0 ± 0.2

Bensulfuron methyl

411

.0 → 149.0 ± 0.2

411

.0 → 182.0 ± 0.2

Flupyrsulfuron methyl

465

.9 → 182.0 ± 0.2

465

.9 → 139.0 ± 0.2

Chlorimuron ethyl

415

.0 → 186.0 ± 0.2

415

.0 → 83.0 ± 0.2

Triflusulfuron methyl

493

.0 → 264.0 ± 0.2

493

.0 → 96.0 ± 0.2

Triasulfuron

401

.8 → 167.1 ± 0.2

401

.8 → 141.1 ± 0.2

Cinosulfuron

413

.9 → 183.1 ± 0.2

413

.9 → 215.1 ± 0.2

Amidosulfuron

370

.1 → 261.0 ± 0.2

370

.1 → 218.0 ± 0.2

Oxasulfuron

407

.0 → 150.0 ± 0.2

407

.0 → 210.1 ± 0.2

Sulfosulfuron

471

.1 → 210.9 ± 0.2

471

.1 → 260.8 ± 0.2

Prosulfuron

419

.9 → 141.1 ± 0.2

419

.9 → 167.1 ± 0.2

Halosulfuron methyl

430

.8 → 182.1 ± 0.2

430

.8 → 222.1 ± 0.2

Primisulfuron methyl

468

.8 → 254.1 ± 0.2

468

.8 → 199.1 ± 0.2

defined peaks with noticeable decreases in resolution. If this is observed, adjustment
of chromatographic conditions to improve the resolution of some of the analytes is
recommended, so fewer signals have to be acquired at once. This is especially true if
two parent-to-daughter ion transitions per analyte are to be acquired.

If sulfonylurea herbicides can reasonably be expected to be present in an analytical

sample (based on prior knowledge), one parent-to-daughter ion transition is usually
considered sufficient to confirm its presence. In other cases where little is known about
the sample history, two parent-to-daughter ion transitions are generally considered
to be necessary for a definitive confirmation. Suggested ion transitions for most of
the registered sulfonylurea herbicides are listed in Table 2. Furthermore, the ratio of
the signals for the two transitions obtained for the sample should match that of an
authentic standard within

±30%, at most.

25

,26

At least four chromatographic standards prepared at concentrations equivalent to

50–70% of the limit of quantitation (LOQ) up to the maximum levels of analytes ex-
pected in the samples should be prepared and analyzed concurrently with the samples.
In LC/MS/MS analysis, the first injection should be that of a standard or reagent blank
and should be discarded. Then, the lowest standard should be injected, followed by
two to four blanks, control samples, fortifications or investigation samples, followed
by another chromatographic standard. This sequence is then repeated until all the
samples have been injected. The last injection should be that of a standard. In order to
permit unattended analysis of a normal analysis set, we recommend that samples and
standards be made up in aqueous solutions of ammonium acetate (ca 5 mM) with up
to 25% of an organic modifier such as acetonitrile or methanol if needed. In addition,
use of a chilled autosampler maintained at 4

◦

C provides additional prevention of

degradation during analysis.

Sulfonylurea herbicides

405

2.2

Crops, food and feed

2.2.1

Nature of the residue

Sulfonylurea herbicides are generally applied to crops as an early post-emergent
herbicide. Crops that are tolerant to these herbicides quickly metabolize them to
innocuous compounds. At maturity, residues of the parent compound in food and
feed commodities are nondetectable. Metabolites are not considered to be of concern,
and their levels are usually nondetectable also. For this reason, the residue definition
only includes the parent compound. Tolerances [or maximum residue limits (MRLs)]
are based on the LOQ of the method submitted for enforcement purposes and usually
range from 0.01 to 0.05 mg kg

−1

(ppm) for food items and up to 0.1 mg kg

−1

for

feed items. There is no practical need for residue methods for animal tissues or
animal-derived products such as milk, meat, and eggs. Sulfonylurea herbicides are not
found in animal feed items, as mentioned above. Furthermore, sulfonylurea herbicides
intentionally dosed to rats and goats are mostly excreted in the urine and feces, and
the traces that are absorbed are rapidly metabolized to nontoxic compounds. For this
reason, no descriptions of methods for animal-derived matrices are given here.

2.2.2

Rationale for methods

Sulfonylurea herbicides can be conveniently extracted from watery and dry plant
materials such as vegetables and cereal and corn grain, straw and forage using aque-
ous buffers adjusted to pH 6.0–7.0. In this pH range, the sulfonylureas exist in the
predominantly anionic form, where they exhibit maximum stability and solubility in
aqueous solutions. At lower pH values, there is an increased tendency for dissociation
by hydrolysis. The most acid-sensitive sulfonylurea is tribenuron methyl, which com-
pletely hydrolyzes in aqueous acidic solutions within 1 day. At pH

> 7.0, a few of the

sulfonylureas such as rimsulfuron and flupyrsulfuron methyl undergo an irreversible
rearrangement to bridge-condensed products. Also, the possibility of increased co-
extractives from the matrix is possible at alkaline pH. Purely aqueous buffers are
convenient for extraction purposes, and extraction efficiency studies conducted with
aged carbon-14 crop residues indicate that endogenous residues are completely recov-
ered from watery and dry plant material. In the case of dry plant material, the sample is
soaked in the aqueous buffer prior to homogenization. Addition of organic co-solvents
such as methanol or acetonitrile does not appear to be necessary for the extraction
of watery or dry plant samples and usually makes cleanup more difficult owing to
increased co-extractives and decreased retention of the analytes during solid-phase
extraction (SPE) cleanup using hydrophobic adsorbents.

Aqueous extracts are normally concentrated and purified using hydrophobic SPE

sorbents. We have obtained the best results with graphitized carbon sorbents; most
plant pigments and starches are strongly retained on these cartridges while the sul-
fonylurea analytes can be eluted with acidified methanol–dichloromethane. We have
observed that at least one sulfonylurea, triflusulfuron methyl, is degraded or irre-
versibly adsorbed on these cartridges, so alternatives with C

18

or polystyrene–divinyl

benzene cartridges are also available. In these cases, additional cleanup using strong
anion-exchange cartridges is often necessary.

406

Compound class

Oily crops such as soybeans and canola (oilseed rape) cannot be extracted with

aqueous buffers, because the extraction solvent cannot permeate the hydrophobic
plant tissue matrix. In these cases, homogenization in acetonitrile–hexane is recom-
mended. This solvent mixture is able to extract sulfonylureas from these samples with
a minimum of co-extracted oil. After extraction, the sulfonylureas partition into the
acetonitrile phase while most of the oil stays in the hexane phase. Further cleanup is
accomplished using a silica SPE cartridge and normal-phase conditions.

Analysis of the concentrated, purified sample extracts is effected by LC/MS or

LC/MS/MS, as described in Section 2.1.

2.2.3

Description of methods

Fortifications are made by pipetting 100–500 µL of the appropriate standards in ace-
tonitrile on to the sample (10 g) before any extraction solution is added and then
allowing the sample to air dry for 30 min.

Watery and dry crops. Watery and dry crop samples (10 g) are extracted by homo-

genization in 2

× 90 mL of 20 mM, pH 6.0 potassium phosphate buffer. Dry crop

samples are allowed to soak (refrigerated) for 60 min in the first 90 mL of buffer
before homogenization in order to hydrate the matrix. After each homogenization
using a Tissumizer (Tekmar, Cincinnati, OH, USA) or similar equipment, the sample
is centrifuged, and the supernatants are combined. The final volume of the combined
supernatants is adjusted to 200 mL with water. An Envi-Carb cartridge (1-g

/12-mL;

Supelco, Bellefonte, PA, USA) is preconditioned with 10 mL of 0.1 N formic acid
in methanol–dichloromethane (1 : 9, v/v), 10 mL of methanol, 10 mL of 0.1 N HCl,
and finally 15 mL of water. The cartridge is not allowed to become dry during or
after preconditioning. A 10-mL aliquot of the sample extract is passed through the
cartridge, and the charge is discarded. The cartridge is washed with 10 mL of water
and 5 mL of methanol; both washes are discarded. Air is allowed to pass through
the cartridge briefly (several seconds) after charging and washing and then for an
additional 2 min under maximum vacuum after the last wash. The sulfonylureas are
eluted with 20 mL of 0.1 N formic acid in methanol–dichloromethane (1 : 9, v/v). The
eluate is evaporated to dryness under a gentle stream of dry nitrogen at 35

◦

C and

reconstituted in 2–5 mL of acetonitrile–5 mM ammonium acetate solution (1 : 9, v/v).

If the analytes of interest are not quantitatively recovered from a graphitized carbon

SPE cartridge (such is the case for triflusulfuron methyl), an alternative cleanup step
using strong anion-exchange (SAX) and polystyrene–divinylbenzene SPE cartridges
is available. In the case of triflusulfuron methyl, 2 g of sample (watery or dry crop) are
homogenized in 15 mL of acetonitrile–0.1 M ammonium carbonate (2 : 1, v/v). NaCl
(2.5 g) is added to induce phase separation, and the sample is again homogenized.
The sample is centrifuged, and the acetonitrile and aqueous phases are allowed to
separate. The sulfonylureas partition into the acetonitrile layer. An SAX SPE cartridge
(500-mg

/6-mL, available from various suppliers) is preconditioned with 2.5 mL of

methanol and 2.5 mL of acetonitrile–0.1 N acetic acid (3 : 7, v/v). An Oasis HLB
cartridge (500-mg

/12-mL; Waters, Bedford, MA, USA) is preconditioned with 5 mL

of methanol and 5 mL of acetonitrile–0.1 N acetic acid (3 : 7, v/v). The SAX cartridge
is attached to the top of the Oasis cartridge. A 25% aliquot of the acetonitrile layer from

Sulfonylurea herbicides

407

the sample extract is diluted threefold with 0.1 N acetic acid and passed through the
stacked cartridges. The cartridges are rinsed with 20 mL of acetonitrile–0.1 N acetic
acid (3 : 7, v/v). The SAX cartridge is allowed to dry at this point in the procedure
and is removed and discarded. The Oasis cartridge is washed with an additional 5 mL
of acetonitrile–0.1 N acetic acid (3 : 7, v/v) and is allowed to go to dryness and air
dry under vacuum for 10 min. The cartridge is eluted with 12 mL of acetonitrile. The
eluate is evaporated to dryness under a stream of nitrogen and reconstituted in 2.5 mL
of methanol–5 mM ammonium acetate solution (1 : 3, v/v). The sample is now ready
for analysis by LC/MS/MS.

Oily crops. Oily crop samples, such as canola seed and soybean, are extracted by

homogenization of 5 g of seed sample in 120 mL of hexane-saturated acetonitrile plus
40 mL of acetonitrile-saturated hexane. The resulting extract is centrifuged to separate
the two layers, and a 6-mL aliquot of the acetonitrile layer is evaporated to dryness
under a stream of nitrogen at ambient temperature. The sample is reconstituted in
4 mL of ethyl acetate. A silica SPE cartridge (5-g

/20-mL, Silica Mega-Bond Elut;

Varian Sample Preparation, Harbor City, CA, USA) is preconditioned with 20 mL
of ethyl acetate. The 4-mL sample is applied, followed by 2

× 1-mL rinses of the

sample tube with ethyl acetate; the charge and wash are discarded. The cartridge
is washed with an additional 15 mL of ethyl acetate and 20 mL of ethyl acetate–2-
propanol–methanol solvent mixture (15 : 4 : 1, v/v/v), which is discarded. The analytes
are eluted with 15 mL of ethyl acetate (containing 0.5% acetic acid), the eluate is
evaporated to dryness under a stream of nitrogen at ambient temperature, and the
residue is reconstituted in 3 mL of methanol–10 mM ammonium acetate (3 : 17, v/v)
for LC/MS/MS analysis. The silica SPE cleanup described above was optimized for
ethametsulfuron methyl; the ethyl acetate–2-propanol–methanol wash and the eluting
solvent may have to be optimized for sulfonylureas of different polarities.

2.3

Soil

2.3.1

Nature of the residue

The degradation rate of sulfonylureas in soil is dependent on many factors, including
soil properties, temperature, and the chemical stability of the compound itself. In gen-
eral, sulfonylureas do not have a tendency to accumulate in soil from season to season.
Both chemical and microbial degradation can occur, and the degradation products do
not show any significant herbicidal activity or have toxicological concerns. For this
reason, the intact sulfonylureas are the only compounds that are monitored in normal
practice, and the methodology description will exclude metabolites. A recent study
was conducted by DuPont that evaluated the sensitivity of various nontarget plant
species to a variety of sulfonylurea herbicides.

27

The results of this study indicate that

application rates of less than 0.1 g of active ingredient (ai) per hectare (ha) do not mea-
surably affect yield and/or quality of sensitive species; only 20% of the trials produced
measurable effects when application rates of 0.1–0.5 g ai ha

−1

were used. Using the

approximation that 0.1 g a.i. ha

−1

results in soil concentrations of 0.1 µg kg

−1

(ppb),

an LOQ of 0.05 µg kg

−1

would be sufficiently sensitive for the determination of all

sulfonylureas in soil.

408

Compound class

2.3.2

Rationale for methods

Ammonium carbonate solution containing a small amount (ca 10%) of methanol are
optimal for extracting aged residues of sulfonylurea herbicides. Some of the sulfony-
lureas have a tendency to bind to clay in certain soils, and the counter ions provided
by ammonium carbonate are necessary to release these residues. A small amount
of organic co-solvent is necessary to help release residues of the more hydrophobic
sulfonylureas from soils which contain high levels of organic matter. The resulting
extract is concentrated and purified using a polymer resin-based SPE cartridge, which
has been found to be very effective in removing most soil-related co-extractives from
the sample. An additional liquid–liquid partitioning into ethyl acetate removes polar
compounds that may interfere with the LC/MS/MS analysis.

2.3.3

Description of method

Soil (25 g) is extracted with 2

× 100 mL of methanol–0.1 M ammonium carbonate

(1 : 9, v/v) solution using a wrist-action shaker. A refrigerated centrifuge set at 4

◦

C

and 10 000 rpm is recommended to separate the supernatant from the solids after each
extraction. The combined supernatants are charged onto an Oasis HLB SPE cartridge
(1-g

/20-mL, Waters) which has been preconditioned with 25 mL of methanol fol-

lowed by 25 mL of the methanol–0.1 M ammonium carbonate (2 : 1, v/v) extracting
solution. The sample container is rinsed with 10 mL of water, and the rinsate is passed
through the cartridge. The cartridge is allowed to go dry only after the water wash has
passed through and is then washed with 10 mL of hexane. The cartridge is air-dried un-
der vacuum for 5 min and eluted with 10 mL of 1 M ammonium hydroxide–acetonitrile
(1 : 19, v/v) followed by 5 mL of ethyl acetate. The cartridge is only allowed to go dry
after all the ethyl acetate has passed through. The combined eluates are evaporated
to approximately 1 mL under a stream of nitrogen at 25–30

◦

C. Then, 2 mL of water

are added, and evaporation is continued until the volume is reduced to 1.8 mL, after
which 10 µL of acetic acid are added. [Note: if tribenuron methyl or other extremely
acid-sensitive sulfonylureas are to be analyzed, use 2 mL of pH 5.5, 5 mM ammonium
acetate solution instead of water, and do not add any acetic acid. This alternative will
work for all sulfonylureas except the most polar sulfonylurea tested (nicosulfuron).]
In either case, the 1.8 mL of aqueous sample is extracted with 2

× 5 mL and 1 × 2 mL

of ethyl acetate through vortex mixing. Centrifugation is used to help separate the
layers, and additional acetic acid (if originally used) is not added after each extrac-
tion. The combined ethyl acetate layers are evaporated to dryness under a stream of
nitrogen at 25–30

◦

C. Just prior to LC/MS/MS analysis, the sample is reconstituted

in 0.3 mL of methanol and diluted to 3.0 mL with pH 6.5, 5 mM ammonium acetate.

2.4

Water

2.4.1

Nature of the residue

Hydrolysis rates of sulfonylurea herbicides in water are heavily dependent upon pH.
In general, acidic conditions promote faster hydrolysis, usually by cleavage of the
sulfonylurea bridge. Neutral to alkaline conditions favor the compounds existing in

Sulfonylurea herbicides

409

their anionic forms, where they are generally more stable. However, a small number
of sulfonylureas undergo ipso-rearrangement and bridge contraction at alkaline pH as
mentioned above. Hydrolysis (and photolysis) products of sulfonylurea herbicides do
not exhibit any herbicidal activity and have been shown to have no other toxicological
or ecotoxicological risks. Therefore, the definition of the residue in drinking and
surface waters only includes the parent compounds. The LOQ of methods for the
determination of sulfonylurea herbicides in drinking waters was established based on
European Union (EU) drinking water guidelines,

28

which require analytical methods

that can measure levels down to 0.1 µg L

−1

(ppb). The method described below has

an LOQ of 0.05 µg L

−1

. In the case of surface waters, the methods must be sensitive

enough to determine the no observable effect level (NOEL) derived from the most
relevant ecotoxicological studies. In the case of herbicides, the appropriate studies to
evaluate would be those conducted on nontarget aquatic plant species, usually algae
and Lemna gibba. The results of these studies using various sulfonylureas indicate that
0.05 µg L

−1

is a sufficiently low LOQ to meet surface water method requirements.

2.4.2

Rationale for method

The method for water is a simplified version of the soil method described above. The
water sample is adjusted to pH of 6

.5, concentrated, and purified using a similar SPE

procedure to that employed in the soil method. The additional cleanup with the ethyl
acetate partitioning is not necessary.

2.4.3

Description of method

A water sample (50 mL) is adjusted to pH of 6

.5 by addition of 0.5 mL of 1.0 M

ammonium acetate. (Note: for brackish water, the pH is adjusted to 6.0–6.2 by addition
of dilute acetic acid, and the sample is diluted with an equal volume of purified water.)
A polymer-based ENV SPE cartridge (0.5-g

/6-mL, Part No. 952493; Varian, Harbor

City, CA, USA) is preconditioned with 10 mL of methanol followed by 10 mL of
10 mM ammonium acetate. The water sample is loaded onto the SPE cartridge at a
flow rate of 2–5 mL min

−1

. Just before the entire water sample has passed through the

cartridge, the sample container is rinsed with 10 mL of purified water, and the rinsate
is added to the cartridge. The cartridge is air-dried under vacuum for 5 min and rinsed
with 10 mL of hexane. The cartridge is air-dried under vacuum for an additional 5 min.
The analytes are eluted with 15 mL of 25 mM ammonium hydroxide in methanol. The
eluate is evaporated to dryness under a stream of nitrogen at 30–40

◦

C. Finally, the

sample is reconstituted in 3 mL of 5 mM ammonium acetate for LC/MS/MS analysis.

3

Conclusions and future directions

The methods described above generally produce recoveries in the 80–110% range
with relative standard deviations of 10% or less, at the stated LOQ and higher levels.
The LC/MS/MS traces are generally free of interference, especially for soil and
water analyses. On rare occasions, an interfering peak may be observed at one of the
transitions monitored for plant-based samples, but we have never seen interference on

410

Compound class

both of the channels. Matrix enhancement or suppression was not observed using the
extractions and cleanups described above. Calibration curves are linear with negligible
intercepts; therefore, either linear regression or response factors may be used for
calculations.

Since sulfonylurea determinations are generally carried out at trace levels, the pos-

sibility of contamination must always be kept in mind. Samples should be kept isolated
from solid analytical standards and concentrated stock solutions of standards. Only
dilute standards, which are used for fortifications and chromatographic standards,
should be located near the analytical samples and only for minimum amounts of
time. If a confirmed response for one or more of the analytes being determined is
obtained in an investigative sample (other than a deliberately fortified sample or a
sample that was known to have been recently treated with the analyte), the possibil-
ity of contamination must be ruled out before a positive result can be reported. If a
reliable control sample is not available, a reagent blank (prepared by taking the spec-
ified amount of extracting solution or purified water through the entire procedure)
should be run. If analyte responses are present in the reagent blank, the equipment
used to prepare the sample should be thoroughly cleaned (dilute bleach solutions are
excellent for this purpose) and checked by preparation of additional reagent blanks
before repeating the analysis. The previous data should be discarded as false-positive
results.

At the present time, LC/MS/MS with triple-quadrupole instruments is the analytical

method of choice for the determination of residues of sulfonylurea herbicides. We can
expect to see improved triple-quadrupole instrumentation become more available and
affordable as time passes, so that more analytical laboratories will have this capability.
Time-of-flight (TOF) instrumentation may also play an increasingly important role in
sulfonylurea analysis. Even though the metabolites are innocuous, stricter regulatory
requirements may mandate that they be monitored, and LC/MS/MS is the method of
choice for these compounds also.

Acknowledgements

The author thanks the following scientists at DuPont Crop Protection for developing
the methods summarized in this chapter: Jennifer S. Amoo and William Jones for the
watery/dry crop method; M. Elena Cabusas, Brock Peterson and Dennis Walker for
the alternative watery/dry crop method; James J. Stry, Michael Gagnon and Sidney
Hill for the soil method and the oily crop method; and Lei Jin and Timothy Devine
for the water method.

References

1. E.M. Beyer, M.J. Duffy, J.V. Hay, and D.D. Schlueter, ‘Sulfonylureas,’ in “Herbicides: Chemistry,

Degradation and Mode of Action,” ed. P.C. Kearney and D.D. Kaufmann, Marcel Dekker,
New York, pp. 117–189 (1987).

2. I. Ahmad and G. Crawford, J. Agric. Food Chem., 38, 138 (1990).
3. P. Klaffenbach and P.T. Holland, Biol. Mass Spectrom., 22, 565 (1993).

Sulfonylurea herbicides

411

4. E.G. Cotterill, Pestic. Sci., 34, 291 (1992).
5. D.G. Thompson and L.M. MacDonald, J. AOAC Int., 75, 1084 (1992).
6. G. Vinelli, A. Vicari, and P. Catizone, J. Agric. Food Chem., 41, 742 (1993).
7. A.J. Krynitsky, J. AOAC Int., 80, 392 (1997).
8. B.M. Berger and N.L. Wolfe, Fresenius’ J. Anal. Chem., 356, 508 (1996).
9. G. Dinelli, A. Vicari, and V. Brandolini, J. Chromatogr. A, 700, 201 (1995).

10. M.M. Kelley, E.W. Zahnow, W.C. Peterson, and S.T. Toy, J. Agric. Food Chem., 33, 962 (1985).
11. J.F. Brady, J. Turner, and D.H. Skinner, J. Agric. Food Chem., 43, 2542 (1995).
12. E.W. Zahnow, J. Agric. Food Chem., 30, 854 (1982).
13. E.W. Zahnow, J. Agric. Food Chem., 33, 479 (1985).
14. G.C. Galletti, A. Bonetti, and G. Dinelli, J. Chromatogr. A, 692, 27 (1995).
15. C.R. Powley and P.A. de Bernard, J. Agric. Food Chem., 46, 514 (1998).
16. A.K. Sarmah and R.S. Kookana, J. Environ. Sci. Health, B34, 363 (1999).
17. L.M. Shalaby, F.Q. Bramble, and P.W. Lee, J. Agric. Food Chem., 40, 513 (1992).
18. R.W. Reiser, A.C. Barefoot, R.F. Dietrich, A.J. Fogiel, W.R. Johnson, and M.T. Scott, J. Chro-

matogr., 554, 91 (1991).

19. R.W. Reiser and A.J. Fogiel, Rapid Commun. Mass Spectrom., 8, 252 (1994).
20. L.M. Shalaby, Biomed. Mass Spectrom., 12, 261 (1985).
21. D. Volmer, J.G. Wilkes, and K. Levsen, Rapid Commun. Mass Spectrom., 9, 767 (1995).
22. L.J. Marek and W.C. Koskinen, J. Agric. Food Chem., 44, 3878 (1996).
23. M. Rodriguez and D.B. Orescan, Anal. Chem., 70, 2710 (1998).
24. R. Bossi, B. Koppen, N.H. Spliid, and J.C. Streibeg, J. AOAC Int., 81, 775 (1998).
25. T. Cairns, ‘Development of Guidelines for Detection, Confirmation and Quantitation by GC/MS,

LC/MS and MS/MS,’ Presented at ASMS Workshop, November 1996.

26. L.Y.T. Li, D.A. Campbell, P.K. Bennett, and J. Henion, Anal. Chem., 68, 3397 (1996).
27. T.T. Obrigawitch, G. Cook, and J. Wetherington, Pestic. Sci., 52, 199 (1998).
28. European Commission, Directorate General Health and Consumer Protection, ‘Guidance Doc-

ument on Residue Analytical Methods,’ SANCO/825/00 rev. 6, June 20, 2000, European Com-
mission, Brussels (2000).