Metabolites of alachlor in water: Identification by mass spectrometry and chemical synthesis

Article abstract of DOI:10.1021/es970380x

This biodegradation study of alachlor in water was designed to assess the presence and the identity of metabolites. Alachlor was incubated in river water for 28 days. The pesticide and its transformation products were extracted by solid phase extraction a

Full text of DOI:10.1021/es970380x

Environ. Sci. Technol. 1997, 31, 3637-3646
has classified alachlor among the “high-priority pesticides”,
Metabolites of Alachlor in Water:
Identification by Mass Spectrometry
and Chemical Synthesis
including those products used in amounts over 50 tonnes
per annum and with some potential to leach (11).
As alachlor degradation products are generally of lower
molecular weight and more oxidized than the parent com-
pound, they may be consequently more water soluble, more
mobile, and have a greater potential to leach (5). Therefore,
concern about alachlor degradation is mainly focused on the
possibility of detecting its TP in surface and ground water,
which they can contaminate by run-off or leaching, or where
they can originate from the pesticide. Indeed, in some cases
TP can be as least as toxic as the parent compound (12). A
series of TP of alachlor has been found in rivers and wells and
in treated soils (5, 13, 14).
The maximum permitted concentration, according to a
EC directive for pesticides in drinking water, i.e., 0.1 µg/ L,
also extends to “related products” (15). Therefore, it is
fundamental to identify the TP of the most widely used
pesticides in order to optimize analytical methods and plan
monitoring programs.
S . M A N G I A P A N , * , E . B E N F E N A T I ,
†
†
†
‡
P . G R A S S O , M . T E R R E N I ,
M . P R E G N O L A T O , G . P A G A N I , ‡ A N D
‡
§
D . B A R C E L O´
Istituto di Ricerche Farmacologiche “Mario Negri”,
Milan, Italy, Dipartimento di Chimica Farmaceutica,
Universit a` degli Studi, Pavia, Italy, and Centro de
Investigacion y Desarrollo-Centro Superior de Investigaciones
Cientificas (CID-CSIC), Barcelona, Spain
This biodegradation study of alachlor in water was designed
to assess the presence and the identity of metabolites.
Alachlor was incubated in river water for 28 days. The
pesticide and its transformation products were extracted
by solid phase extraction and analyzed by gas chromatography/
mass spectrometry. Selected samples were also analyzed
by liquid chromatography/mass spectrometry. The
Because most environmental studies on alachlor are about
mineralization in soil or degradation by isolated microorgan-
isms, we undertook a laboratory study of alachlor biotic
degradation in an aquatic system in order to detect more
metabolites or confirm already detected ones.
disappearance of the pesticide was not significative, but
several compounds were identified as alachlor metabo-
lites. Nine compounds were confirmed by comparison with
synthetic standards. One metabolite has never been re-
ported. For seven molecules, formulae were presumed on
the basis of spectrum interpretation and literature data.
Experimental Section
Degradation Studies. We made several tests with different
inocula and media over 2 years. A detailed description of all
biodegradation experiments and the results about degradation
kinetics have been previously reported (16). In short, in the
metabolism tests, alachlor (Alltech, Deerfield, IL; purity 98.5%)
was incubated at a concentration of 1 mg/ L in dark bottles
filled with 2-5 L of river water, stirred at 20 °C for 28 days.
Dark bottles were used to avoid photodegradation, since the
study was focused on biodegradation in water. The water
samples were collected from the Olona river (Italy) in May
Introduction
Alachlor, 2-chloro-2′,6′-diethyl-N-(methoxymethyl)acetanil-
ide, is a pre-emergence herbicide of the chloroacetanilide
family, widely used on different crops since 1969. In Italy,
its use has grown since the ban on atrazine (1).
Although the persistence of alachlor in the environment
2 2 3
is limited, its complete mineralization to CO , H O, and NH
1
993 and May and September 1994.
In order to follow the disappearance of alachlor every 7
days, a 10 mL subsample was extracted with a C18 cartridge
and analyzed in GC with a NPD detector (16). At the end of
the biodegradation experiment, the remaining water was
filtered (0.45 µm) and extracted on a C18 cartridge and
subsequently on a Carbopack-B column to ensure better
recovery of polar compounds.
Three samples with relative blanks were obtained from
the incubations and analyzed in the metabolism study.
Sam ple Extraction. C18 cartridges employed to extract
weekly subsamples (1 mL cartridge) and the final whole
samples (3 mL cartridge) were from J. T. Baker (Phillipsburg,
NJ). The phase was washed with ethyl acetate and then
activated with methanol. After passage of the sample, the
column was eluted with ethyl acetate.
has never been reported (2), and in fact, a lot of transformation
products (TP) are generated by degradation.
Several studies of the environmental biotic and abiotic
degradation have been reported, and different compounds
have been found, depending on the experimental conditions.
The range of alachlor metabolites produced during metabolic
studies in vitro or in vivo in plants and animals is likewise
very wide. The large series of alachlor TP, detected in various
studies, is presented in reviews by Chesters et al. (3) and
Sharp (4). Moreover, the number of alachlor metabolites is
continually rising, and recent studies report new compounds
(
5, 6).
The final whole water sample was extracted according to
Di Corcia and Marchetti (17). Briefly, a glass column
containing 400 mg of Carbopack-B phase (Supelco, Bellefonte,
PA) was employed. The phase was washed with 10 mL of a
The picture of alachlor TP is therefore complex, and we
only know the environmental and toxicological properties of
a few compounds (7-9).
The U.S. Environmental Protection Agency classifies
alachlor as a probable carcinogen for humans (group B2)
mixture of CH
2 2 3 3
Cl :CH OH (4:1) followed by 5 mL of CH OH
and then activated with 20 mL of 10 mg/ mL ascorbic acid
solution in HCl 0.01 M. The water sample passed through
the column at a flux rate of 20 mL/ min. Compounds adsorbed
(
10) and most of its TP are structurally similar (aniline
derivates). The Commission of the European Community
on the phase were eluted by 10 mL of a mixture of CH
2 2
Cl :
*
Corresponding author address: Istituto di Ricerche Farmaco-
CH OH (4:1) for neutral fraction recovery and by the same
3
logiche Mario Negri, via Eritrea 62, 20157 Milano, Italy. Fax: +39 2
mixture acidified with trifluoroacetic acid (0.2% v/ v) for the
acidic fraction. Finally, the extracts were concentrated under
a nitrogen stream to 0.05-0.5 mL. All solvents (Carlo Erba,
Milan, Italy) were for the analysis of pesticide residues.
3
9001916. E-mail: mangiapan@irfmn.mnegri.it.
†
Istituto di Ricerche Farmacologiche Mario Negri.
Dipartimento di Chimica Farmaceutica.
Centro de Investigacion y Desarrollo.
‡
§
S0013-936X(97)00380-5 CCC: $14.00

1997 Am erican Chem ical Society
VOL. 31, NO. 12, 1997 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 6 3 7

SCHEME 1. Synthesis Scheme of Standards
Before injection on the GC, samples from Carbopack-B
extracts were derivatized with diazomethane or silylating
agent (BSTFA). Diazomethane was prepared in the laboratory
by distillation from Diazald ethereal solution (Aldrich-Chemie,
Steinheim, Germany) added dropwise to a potassium hy-
droxide ethanol mixture. BSTFA was from Fluka Chemie
instrument; a SE 52 (25 m long; 0.25 mm i.d.; 0.25 µm film
thickness) and an Easy 1701 (25 m long; 0.25 mm i.d.; 0.25
µm film thickness), the latter both from Analytical Technology
(Cernusco s/ N, Milan, Italy). The inlet and transfer line were
held at 260 and 280 °C, respectively. The carrier gas (He)
head pressure was 70 kPa. For the CP-Sil 8 CB and SE 52
columns, the oven program went from 80 °C (1 min) to 300
°C (5 min) at a rate of 6 °C/ min; for the Easy 1701, operating
temperatures were raised from 60 to 275 °C at a rate of 6
°C/ min.
(
Buchs, Switzerland).
We used these two derivatization reagents to make
possible, or simply to improve, the detectability in GC analysis
of any TP-containing functional groups with active hydrogens
such as COOH, OH, NH, and SH, typical of some alachlor TP.
Diazomethane is used to convert carboxylic acids into methyl
esters, while BSTFA substitutes a trimethylsilyl group for the
acidic hydrogen of OH, NH, and SH groups.
The ion trap mass spectrometer was employed in electron
ionization (EI) and in positive chemical ionization (PCI)
modes, with CH
4
as ionizing gas, to obtain more information
on molecular weight.
Derivatization was done by adding to subsamples, dried
under N
2
, 0.5 mL of the ethereal solution of diazomethane
On both instruments, the extracts were injected (2 µL)
with a splitless system.
or 30 mL of BSTFA. The reactions with diazomethane and
BSTFA lasted 10 min and were maintained at room temper-
ature and 60 °C, respectively.
Instrum ental Analysis. The extracts were analyzed in GC/
MS on two different instruments: a Varian 3400 gas chro-
matograph coupled with a Varian Saturn II ion trap mass
spectrometer and a HP 5890 series II gas chromatograph
equipped with a HP MSD 5971 series mass spectrometer.
For the first instrument, a CP-Sil 8 CB Chrompack
Selected extracts were also analyzed in LC/ MS with a high-
flow pneumatically assisted electrospray interface (LC-ESP-
MS), using a VG Platform ESP from Fisons Instruments
(Manchester, U.K.). The analyses were conducted in the
positive ionization mode at isocratic condition (60:40 CH
3
-
CN) for 35 min on a C18 column (2.1 mm i.d., 30 cm, packed
with 10 µm particles) from Waters-Millipore (Milford, MA).
The operational ESP parameters were flow, 0.3 mL/ min;
source temperature, 150 °C; extraction voltage, 20 V.
Reference Standards and Their Synthesis. We used
several standards to verify the identity of the presumed
alachlor metabolites detected in our samples.
Two standards were bought: 2,6-diethylaniline (DEA) (I)
from Alltech (Deerfield, IL) and chloroacetic acid from Aldrich-
Chemie (Steinheim, Germany). Another compound, N-(2,6-
diethylphenyl)-N-(methoxymethyl)acetamide (MW, 235), was
synthetized at CID-CSIC, Barcelona. The other 12 reference
(
(
Middleburg, The Netherlands) fused silica capillary column
25 m long; 0.25 mm i.d.; 0.25 µm film thickness) was used.
The carrier gas (He) head pressure was 30 kPa. Inlet and
transfer line temperatures were set at 240 and 280 °C. The
oven program was raised from 80 °C (1 min) to 300 °C (4 min)
at a rate of 10 °C/ min.
On the HP instrument, the following fused silica capillary
columns were used: the same CP-Sil 8 CB as on the ion trap
3
6 3 8
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 31, NO. 12, 1997

SCHEME 2. Synthesis Schemes of Standards
compounds were synthetized in our laboratories. The
synthesis schemes are shown in Schemes 1 and 2.
Identities of the new synthetized products were established
2-Oxo-N-(2,6-diethylphenyl)-N-(m ethoxym ethyl)acet-
am ide (IX), 2-Oxo-N-(2,6-diethylphenyl)acetam ide (X), and
2-Oxo-N-(2,6-diethylphenyl)-N-(methyl)acetamide (XI). Neat
1
by MS and also by H NMR for some of them. The 300 MHz
DMSO (4 mmol) was added dropwise, at -60 °C under N
to a stirred solution of oxalylchloride (1.44 mmol) in dry CH
Cl (6 mL). After 5 min, the yellow solution was forced by N
2
,
1
H NMR spectra were recorded on a Bruker ACE-300
2
-
2
spectrometer, in deuterochloroform using deuterated tetra-
2
methylsilane as internal standard; chemical shifts are in d
pressure through a double-tipped needle, into a solution of
(
parts per million) and coupling constants (J) in hertz. DIS-
the appropriate alcohol (VI-VIII) (0.48 mmol) in dry CH
2
Cl
2
MS spectra were obtained on a VG TS 250 instrument by EI
at 70 eV. The TLC analyses were run on silica gel 60 F254
Merck and flash column chromatography on silica gel 60
(10 mL), at -60 °C with stirring. After 15 min, 500 mL of Et
3
N
was injected into the solution that was warmed to room
temperature and poured into 20 mL of H O. Organic products
were extracted with CH Cl and washed once with 2 M HCl
and with NaHCO3, with re-extraction of each aqueous layer
with CH Cl . The collected organic layers were dried (an-
2
(
60-200 µm, Merck).
The formulae and the mass spectra of the synthetized
2
2
compounds and of the two commercial standards, analyzed
by HP gas chromatography mass spectrometry, are reported
in Figures 1 and 2.
N-(2,6-Diethylphenyl)m ethyleneam ine (II), 2,6-Diethyl-
form anilide (XII), and 2-Chloro-N-(2,6-diethylphenyl)acet-
am ide (XVI). These compounds were synthetized as reported
in ref 5, and analytical data are in agreement with those
2
2
hydrous sodium sulphate) and evaporated under reduced
pressure to give the pure products (IX-XI) (see GC/ MS
analysis).
2-Methylthio-N-(2,6-diethylphenyl)acetam ide (XIV).
2-(Methylthio)acetylchloride (4.3 mmol) was added to a
solution of 2,6-diethylaniline (3.3 mmol) and NaHCO (10
3
reported.
mmol) in CH Cl (50 mL), and the mixture was stirred at
2
2
1
(
XII) H NMR (CDC1
3
) d: 1.20, (t, CH
3
CH
2
Ar, 6H), 2.65 (m,
reflux until reaction was completed. Water (20 mL) was
added, and organic products were extracted with CH Cl . The
CH
3
CH
2
Ar, 4H), 6.85 (bs, NH, 1H), 7.10-7.35 (m, Ar, 3H), 8.05
2
2
(
d, J ) 11.5 Hz, CHO, 0.55 from 1H), 8.45 (s, CHO, 0.45 from
combined organic phases were dried (anhydrous sodium
sulfate) and evaporated under reduced pressure. The pure
product (see GC/ MS analysis) was used without further
purification.
1
H).
2
-Acetoxy-N-(2,6-diethylphenyl)-N-(m ethoxym ethyl)-
acetam ide (III). To a solution of II (6 mmol) in dry
cyclohexane (20 mL) 2-acetoxyacetyl chloride (0.645 mL), dry
methanol (1.5 mL), and triethylamine (1.5 mL) were added.
The reaction mixture was stirred for 1 h at room temperature,
then water (20 mL) was added, and organic products were
extracted with ethyl acetate (4 × 50 mL). The combined
organic phases were dried (anhydrous sodium sulfate) and
evaporated under reduced pressure. The pure product was
used in the next reaction without further purification.
Analytical data are in agreement with those reported (6).
N-Methyl-2,6-diethylform anilide (XV). Methyl iodide (1.2
mmol) was added to a stirred mixture of 2,6-diethylforma-
nilide (XII) (0.5 mmol), 18-crown-6 (0.05 mmol) and 50%
KOH/ NaOH (5 mL) in toluene (5 mL) at room temperature.
After 15 min, the organic layer was washed once with H
2
O,
dried (anhydrous sodium sulfate), evaporated under reduced
pressure, and the residue was purified by flash chromatog-
raphy (n-hexane/ ethyl acetate (6:4)) to give the pure product
(
see GC/ MS analysis).
2
-Acetoxy-N-(2,6-diethylphenyl)acetam ide (IV), 2-Acet-
oxy-N-(2,6-diethylphenyl)-N-(m ethyl)acetam ide (V), 2-Hy-
droxy-N-(2,6-diethylphenyl)-N-(m ethoxym ethyl)acetam -
ide (VI), 2-Hydroxy-N-(2,6-diethylphenyl)acetam ide (VII),
Results and Discussion
The disappearance of alachlor was not significative, taking
into account the analytical variability (10%), in any incubation
performed to study alachlor metabolism, during the 4 weeks
of the experiments (16).
2
-Hydroxy-N-(2,6-dieth ylph en yl)-N-(m eth yl)acetam ide
(
VIII), and 2,6-diethylacetanilide (XIII). These compounds
were synthetized as reported (6) and analytical data are in
agreement with those reported.
However, GC/ MS analysis of the final extracts, representing
the incubated samples concentrated from 4000 to 40 000
times, found about 20 alachlor degradation products. Several
chromatographic peaks not present in the blanks and, in some
cases, characterized by typical ions of alachlor fragmentation
appeared in the chromatograms. The areas of these peaks
were generally less than 1% of the alachlor peak. These results
1
(
VII) H NMR (CDC1
3
) d: 1.18, (t, CH
3
CH
2
Ar, 6H), 2.55 (q,
OH, 2H),
CH
3
CH
2
Ar, 4H), 3.70 (bs, OH, 1H), 4.15 (s, COCH
2
7
.12-7.28 (m, Ar, 3H), 7.95 (bs, NH, 1H).
1
(
VIII) H NMR (CDC1
3
) d: 1.24, (t, CH
N, 3H), 3.38 (t, J ) 5 Hz, OH, 1H),
OH, 2H), 7.14-7.38 (m, Ar, 3H).
3 2
CH Ar, 6H), 2.51 (q,
3 2 3
CH CH Ar, 4H), 3.24 (s, CH
3
.58 (d, J ) 5 Hz, COCH
2
VOL. 31, NO. 12, 1997 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 6 3 9

FIGURE 1. GC-MS spectra and formulae of synthetic standards used to confirm the metabolites. (a) Commercial standard. (b) Standard
synthetized at CID-CSIC, Barcelona, Spain. On the upper left-hand side of the spectra, the Roman numeral is the synthesis scheme number
(see Schemes 1 and 2), and the Arabic number is the one assigned to identify the compound (see Table 2).
3
6 4 0
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 31, NO. 12, 1997

FIGURE 2. GC-MS spectra and formulae of synthetic standards not found as metabolites in our samples. The Roman numeral, on the upper
left-hand side of the spectra, is the synthesis scheme number (see Schemes 1 and 2).
agreed with findings of other authors operating in similar
conditions (18).
spectrum mean ions and relative abundance, chemical name,
extraction solid phases used, samples in which it was found,
reference number of the standard used to confirm, and
literature references. Formulae and spectra of the standards
used to confirm these compounds are shown in Figure 1.
The metabolites identified have a molecular weight lower
than alachlor, from which they originate mainly by hydrolytic
and oxidation reactions operated by microorganisms.
The two compounds 6 and 7 were already present as
impurities in the alachlor standard used in the incubation
experiments (standard purity 98.5%). We analyzed this
standard in GC/ MS (in triplicate) and compared the levels of
each impurity with the alachlor peak, measuring the area
ratios. In the same way, we measured the TP/ alachlor ratio
in the samples. The ratios of 6 and 7 to alachlor were,
respectively, 0.0066 and 0.000 28 in the standard and 0.045
and 0.000 27, as average, in the samples. Therefore, we could
establish for sure that 7 was formed during incubation, but
nothing can be said about the other product.
First, we tried to assign a structural formula to the
molecules detected on the basis of literature data, hypothetical
fragmentation and comparison with corresponding PCI
spectra. This enabled us to verify the molecular weight. The
mass spectra libraries NIST 92 and Wiley 5 helped us identify
DEA and chloroacetic acid.
Because different alachlor degradation pathways, caused
by physical, chemical or biological agents, can give rise to the
same final TP, we compared our metabolites with those
reported in the literature, resulting from different degradation
processes.
Then, we synthetized the presumed metabolites. The
synthetic compounds and the commercial standards were
analyzed in GC/ MS in the same conditions as the water
samples. Retention time and mass spectrum of standards
and presumed metabolites were compared.
The synthetic standards VI, VII, and VIII, characterized by
OH groups, were also derivatized with BSTFA and mass
spectrum mean ions are reported in Table 1.
Metabolites 6 and 7 derive from loss of the methoxymethyl
group and chlorine, respectively.
Derivatization of the extracts with diazomethane and
BSTFA did not give good results; a lot of new peaks, difficult
to resolve, appeared in the chromatogram, and we could not
identify new TP with reactive groups.
Confirm ed Metabolites. We identified 9 metabolites, for
which Table 2 reports the following parameters: number
assigned to identify the compound, molecular weight, mass
To our knowledge, 5 has not been previously reported; we
only found it in the C18 extract of the May 94 sample. All
other metabolites in Table 2 have been reported. This
compound might derive from metabolite 6 by loss of the
chlorine atom and subsequent oxidations to the hydroxylic
(the correspondent metabolite was not found) and then the
carbonylic group.
VOL. 31, NO. 12, 1997 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 6 4 1

b
a
c
b
N
N
a
N
N
N
N
N
c
N
N
b
N
N
b
N
N
N
a
a
3
6 4 2
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 31, NO. 12, 1997

TABLE 3. Presumed Transformation Products
mean ions and relative abundance
no. MW
presumed compound
77 (20) 7-ethylindoline
extraction phase samples
ref
1
0
147 147 (50) 132 (100) 117 (39)
C18
Carbopack-B/n
C18
C18
Carbopack-B/n
May 93
May 94
May 94
May 93
May 94
Sep. 94
May 94
Sept. 94
5
1
1
1
2
175 (60) 160 (100) 146 (61) 134 (46) alachlor m etabolite
191 191 (32) 160 (100) 144 (20) 132 (37) alachlor m etabolite
1
1
1
3
4
5
219 (100) 189 (36) 188 (27) 174 (55) alachlor m etabolite
C18
221 221 (63) 189 (38) 160 (100) 146 (69) alachlor m etabolite
Carbopack-B/n,a May 93
Sept. 94
223 (49) 208 (43) 174 (42) 146 (100) 1-chloroacetyl-2,3-dihydro-7-ethylindole C18
May 94 3, 4, 19
Sept. 94
1
1
6
7
235 (43) 189 (20) 160 (100) 146 (43) alachlor m etabolite
285 285 (12) 253 (48) 204 (88) 176 (100) 2′-(1-hydroxyethyl)-6′-ethyl-2-chloro-
N-(m ethoxym ethyl)acetanilide
Carbopack-B/n
C18
Carbopack-B/n
C18
C18
Sept. 94
May 93 21, 24
Sept. 94
May 94
May 93
1
1
8
9
301 (7)
252 (15) 224 (19) 176 100) alachlor m etabolite
313 (22) 267 (25) 234 (35) 188 (100) alachlor m etabolite
Carbopack-B/n
a
MW is reported when it was possible to confirm the m olecule by PCI.
In two alachlor degradation studies, 2 and 6 were produced
Presum ed TP. Spectra of the remaining presumed
metabolites were compared with literature data, when
available. Table 3 shows the same parameters as Table 2 for
every presumed metabolite. Spectra and hypotheses about
the formulae of some of these compounds are presented in
Figure 3. For metabolites 12, 14, and 17, the molecular weights
of 191, 221, and 285 were confirmed by PCI spectrum.
7-Ethylindoline (MW, 147) was tentatively identified by
Potter and Carpenter (5), on the basis of the mass spectrum.
We could not confirm the identity of this metabolite because
attempts to synthetize it gave two peaks both with a parent
ion at m/ z 147 and with very similar MS spectra.
by the soil fungus Chaetomium globosum (19) and by
Chironomid larvae (20), and in both these experiments, the
metabolites’ identities were confirmed by authentic standards.
Compound 6 was also identified during an incubation of
alachlor with male rat liver microsomes (21).
DEA (metabolite 2) originates from loss of both functional
groups on N of alachlor. A mutagenic potential is documented
for this compound, probably linked to its oxidation to
nitrosobenzene (7, 8).
In a recent study, groundwater samples were collected
beneath a Massachusetts cornfield and analyzed in GC/ MS
for alachlor metabolites; metabolites 2, 3, 7, and 9 were
detected and confirmed by standards (5).
Metabolite 3 might originate from metabolite 4 through
a reduction followed by dehydration.
Metabolite 9 appeared silylated after reaction with BSTFA,
giving ions at m/ z 323, 308, 188, and 45, in agreement with
the standard behavior (see Table 1).
Suba and Pearson (18), in several studies of alachlor
degradation, identified many degradation products, some of
which correspond to our confirmed metabolites. Compounds
, 8, and 9 were found as major photoproducts of the sensitized
photolysis (in presence of acetone) of alachlor, and compound
together with 6 was found in soil photolysis studies. The
From spectral analysis, we conclude that metabolites 11,
1
2, and 13 do not present chlorine and the methoxymethyl
group; the latter, when present, gives a strong ion at m/ z 45
in metabolite 12 there is a loss of 45, but the ion at m/ z 45
is absent).
The metabolite 11 may have a MW of 175, and the losses
of 15 and 29 may correspond to CH
and CHO groups. On
(
3
the basis of these considerations, a hypothetical structural
formula is presented in Figure 3. A metabolite with the same
formula, formed in flooded soil, is reported by Chesters et al.
7
(
3).
By comparison with authentic standards, compound 12
of MW 191 did not correspond to either standard XIII or XV
Figure 2). The first, 2,6-diethylacetanilide, was found in
Massachusetts’ groundwater samples by Potter and Carpenter
5) and in sensitized photolysis and aerobic soil metabolism
9
same authors (18) also found 8 and 9 in an aerobic soil
metabolism study and 7 in an anaerobic aquatic metabolism
study. Compounds 9 and 7 were identified as alachlor
photodegradation products by Somich et al. (22) and 9 by
Chiron et al. (6) and by Pe n˜ uela and Barcel o´ (23).
Metabolites 8 and 9 originate from the progressive
oxidation of the carbon bound to chlorine in the alachlor
molecule.
(
(
experiments by Suba and Pearson (18). A hypothetical
formula of 12 is reported in Figure 3.
Metabolite 13 does not correspond to standard XI (Figure
2
b). A hypothetical formula is reported in Figure 3.
The Microtox test showed that hydroxyalachlor (metabolite
Metabolite 14, with MW 221, differed from the standard
9
) had toxicities higher than or not significantly different from
VIII, which was found by some authors (6, 22, 23). It may
therefore correspond to 2′,6′-diethyl-N-(methoxymethyl)-
formanilide (Figure 3). Observing the spectrum, parent ion
the parent compound (9).
A very small peak (metabolite 4) with a spectrum and
retention time corresponding to the standard XII (2′,6′-
diethylformanilide) was seen by us after derivatization of the
C18 extract with BSTFA. The BSTFA did not react with 4, but
probably reacted with a coeluting polar component, changing
its retention time. This compound was found in Mas-
sachusetts groundwater samples (5). It might originate from
+
•
M
3
(221) could lose CH O to give the ion at m/ z 190 and
CH
CH
3
3
OH to give the ion at m/ z 189. The combined losses of
OH and CHO can give the ion at m/ z 160, which is typical
•
of alachlor’s fragmentation pattern. Further evidence is that
the metabolite did not react with BSTFA. We have not found
this presumed metabolite in the literature.
metabolite 5 through oxidation followed by loss of CO
2
.
1-Chloroacetyl-2,3-dihydro-7-ethylindole (MW, 223) was
observed as a major component in the alachlor degradation
study with C. globosum, and its structural formula was
assigned on the basis of MS and IR spectra but was not
confirmed by comparison with an authentic standard (19).
The formation of TP, which can reach groundwater and
can be in some cases at least as toxic as the parent compound,
represents a potential threat to the environment and human
health.
VOL. 31, NO. 12, 1997 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 6 4 3

FIGURE 3. GC-MS spectra and presumed formulae of some transformation products not confirmed by comparison with standards. The Arabic
numeral, on the upper left-hand side of the spectra, is the number assigned to identify the compound (see Table 3).
3
6 4 4
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 31, NO. 12, 1997

TABLE 4. TPs Detected by LC-MS
other
no. MW (M + H)+ (M + Na)+ ions
name
extraction phase
ref
6a 225
226
248
243 2-chloro-N-(2,6-diethylphenyl)acetam ide
Carbopack-B/a 3, 4, 18, 19,
2
0, 21
a
7
8
0
1
235
249
233
281
236
250
234
282
258
272
256
304
162 N-(2,6-diethylphenyl)-N-(m ethoxym ethyl)acetam ide
2-oxo-N-(2,6-diethylphenyl)-N-(m ethoxym ethyl)acetam ide
202 8-ethyl-N-m ethoxym ethyl-4-m ethyl-2-oxytetrahydroquinoline
Carbopack-B/n 5, 18, 22
Carbopack-B/n 18
Carbopack-B/a 18, 22
a
2
2
2-m ethylthio-N-(m ethoxym ethyl)-N-(2,6-diethylphenyl)acetam ide Carbopack-B/n 4, 18
a
Already seen and confirm ed by GC-MS (see Table 2).
TABLE 5. Standards Synthetized but Not Found in the Incubated Samples
mean ions
MW
name
standardsa
ref
1
1
2
2
2
2
91
91
07
19
21
37
191
191
207
219
221
237
176
174
192
204
205
176
148
162
176
190
190
160
134
147
160
162
175
148
2′,6′-diethylacetanilide
2′,6′-diethyl-N-m ethylform anilide
2-hydroxy-N-(2,6-diethylphenyl)acetam ide
2-oxo-N-(2,6-diethylphenyl)-N-(m ethyl)acetam ide
2-hydroxy-N-(2,6-diethylphenyl)-N-(m ethyl)acetam ide
2-m ethylthio-N-(2,6-diethylphenyl)acetam ide
XIII
XV
VII
XI
VIII
XIV
3, 5, 6, 18, 22
3
6, 22
a
Num bers of the synthesis schem e.
The spectrum of this compound is consistent with that of
compound 15, found among our samples.
GC/ MS; two others (MW, 233 and 281) were assumed on the
basis of their spectrum and comparison with literature (4, 18,
22). All the metabolites detected are reported in Table 4.
The metabolite with MW 233 was already found by Chiron
et al. (6) and confirmed as a photodegradation product.
Feng et al. identified 2′-(1-hydroxyethyl)-6′-ethyl-2-chloro-
N-(methoxymethyl)acetanilide (MW, 285) as the most abun-
dant metabolite produced during incubation of alachlor with
male rat liver microsomes. The metabolite was confirmed
by identification of its acid hydrolysis product, 2-ethylaniline
+
The molecular ion plus hydrogen (M + H) and molecular
+
ion plus sodium (M + Na) can generally be seen in the LC-
(
21).
The PCI-MS spectrum described in this work is very similar
ESP-MS spectrum. In one case (metabolite 6), the molecular
ion plus ammonium was present. The voltage used (20 V)
does not generally enable us to see other compound fragments
but this voltage improves sensitivity, and we can therefore
detect more TP, even if the spectrum is less informative.
However, for metabolite 7, the ion at m/ z 162, representing
the loss of OCH3 and COCH3 groups, and the ion at m/ z 202
(M-OCH3) for metabolite 20 are present. By LC-ESP-MS, we
did not find [(methoxymethyl)(2,6-diethylphenyl)amino]-
oxoacetic acid and 2-[(methoxymethyl)(2,6-diethylphenyl)-
amino]-2-oxoethanesulfonic acid, the two major metabolites
identified in previous studies of alachlor in soil (18) and water
(25), deriving from its microbial degradation.
to the PCI-MS spectrum of metabolite 17 for the presence of
a protonated molecular ion at m/ z 286, a base ion at m/ z 254
+
+
(
(
MH -CH
3
OH), and ions at m/ z 268 (MH -H
OH-HCl).
2
O) and m/ z 218
+
MH -CH
3
The EI spectrum of 17 shows a fragmentation pattern very
similar to that of the PCI spectrum: presence of the molecular
ion and loss of CH OH and of H O from the molecular ion
3
2
to give ions at m/ z 253 and 267. There is also the loss of
•
CH
3
OCH
2
from the molecular ion to give an ion at m/ z 240
+
and the ion itself CH
3
OCH
2
at m/ z 45. Finally, the ion at
m/ z 204 corresponds to the ion at m/ z 188 in the alachlor EI
spectrum. In fact, both are produced from the correspoding
Table 5 summarizes the standards identified by other
authors as corresponding to alachlor metabolites (see the ref
column in Table 5), that we synthetized and analyzed but did
not find in our samples.
•
molecular ions by loss of ClCH
2
3
and CH OH with H
abstraction from the ortho ethyl on the ring followed by
cyclization (20).
A probable isomer of this compound appeared in the
chromatogram of an extract analyzed by GC/ MS. It was
characterized by a slightly longer retention time and an almost
identical MS spectrum. An isomer of 2′-(1-hydroxyethyl)-
Figure 2 shows spectra and formulae of these compounds.
Because compounds VII and VIII present an OH group,
we searched for the corresponding trimethylsilyl ethers in
derivatized extracts, but found none.
6
′-ethyl-2-chloro-N-(methoxymethyl)acetanilide was identi-
fied by Feng and Patanella (21), too. They suggested that the
two isomers originate because the rotation about the aromatic
nitrogen bond is much more restricted in the presence of an
adjacent benzylic hydroxyl group.
Acknowledgments
We thank Dr. A. Messeguer, CID-CSIC, Barcelona, for his help
with metabolite synthesis. This work was supported by the
Commission of the European Community (EC Project EV5V-
CT92-0061).
A compound tentatively identified as 2′-(1-hydroxyethyl)-
6
′-ethyl-2-chloro-N-(methoxymethyl)acetanilide was found
by Bonfanti et al. (24) in an in vitro experiment, incubating
alachlor with rat hepatocytes.
Metabolites 16, 18, and 19 present typical ions of the
alachlor spectrum (m/ z at 132, 146, 160, 176, 188, etc.). Since
chromatogram peaks were very low, probably MS spectra
present interfering ions. We have not formulated any
hypothesis about their structures.
HPLC-MS Analysis. Neutral and acidic fractions of
Carbopack-B extracts of the May and September 1994 samples
were also analyzed in LC-ESP-MS. This indicated 5 me-
tabolites, three (6, 7, and 8) already seen and confirmed by
Literature Cited
(1) Galassi, S.; Vailati, G.; Genesi, M. Inquinamento 1994, 8, 70-72.
(
(
2) Trevisan, M. Acqua-Aria 1992, 9, 853-868.
3) Chesters, G.; Simsiman, G. V.; Levy, J.; Alhajjar, B. J.; Fathulla,
R. N.; Harkin, J. M. Rev. Environ. Contam. Toxicol. 1989, 110,
1
-74.
(
4) Sharp, D. B. In Herbicides. Chemistry, degradation and mode of
action; Kearney, P. C.; Kaufman, D. D., Eds; Marcel Dekker: New
York, 988; Vol. 3, p. 301.
VOL. 31, NO. 12, 1997 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 6 4 5

(
(
(
(
(
5) Potter, T. L.; Carpenter, T. L. Environ. Sci. Technol. 1995, 29,
557-1563.
6) Chiron, S; Abian, J.; Ferrer, M.; Sanchez-Baeza, F.; Messeguer,
A.; Barcel o´ D. Environ. Toxicol. Chem. 1995, 14, 1287-1298.
7) Brown, M. A.; Kimmel, E. C.; Casida, J. E. Life Sci. 1988, 43,
(18) Suba, L. A.; Pearson, D. A. The environmental studies of alachlor;
Monsanto Final Report MSL-0860, Monsanto Agricultural Com-
pany, 1979.
1
(19) Tiedje, J. M.; Hagedorn, M. L. J. Agric. Food Chem. 1975, 23,
77-81.
2
087-2094.
8) Lyons, C. D.; Katz, S. E.; Bartha, R. Bull. Environ. Contam. Toxicol.
985, 35, 696-703.
9) Kross, B. C.; Vergara, A.; Raue, L. E. J. Toxicol. Environ. Health.
992, 37, 149-159.
(20) Wei, L. Y.; Vossbrinck, C. R. J. Agric. Food Chem. 1992, 40, 1695-
1699.
1
(21) Feng, P. C. C.; Patanella, J. E. Pestic. Biochem. Physiol. 1989, 33,
16-25.
(22) Somich, C. J.; Kearney, P. C.; Muldoon, M. T.; Elsasser, S. J. Agric.
Food Chem. 1988, 36, 1322-1326.
(23) Pe n˜ uela, G. A.; Barcel o´ , D. J. Chromatogr. A 1996, 754, 187-
195.
(24) Bonfanti, M.; Taverna, P.; Chiappetta, L.; Villa P.; D’Incalci, M.;
Bagnati, R.; Fanelli, R. Toxicology 1992, 72, 207-219.
(25) Hoobler, M. A.; Letendre L. J. The aquatic metabolism studies of
alachlor; Monsanto Final Report MSL-5307, Monsanto Agri-
cultural Company, 1986.
1
(
(
10) Regulating Pesticides in Food. National Academy Press. Wash-
ington, 1987.
11) Fielding, M. (Ed.) Pesticides in ground and drinking water;
Commission of the European Communities: E. Guyot SA,
Brussels (Water Pollution Research Report 27), 1992.
12) Galassi, S.; Provini, A.; Halfon, E. Intern. J. Environ. Anal. Chem.
(
(
(
1
996, 65, 331-344.
13) Kolpin, D. W.; Thurman, E. M.; Goolsby, D. A. Environ. Sci.
Technol. 1996, 30, 335-340.
14) Thurman, E. M.; Goolsby, D. A.; Aga, D. S.; Pomes, M. L.; Meyer,
M. T. Environ. Sci. Technol. 1996, 30, 569-574.
15) CEE Directive no. 80/ 778.
Received for review April 29, 1997. Revised manuscript re-
ceived October 10, 1997.
X
(
(
16) Galassi, S.; Provini, A.; Mangiapan, S.; Benfenati, E. Chemosphere
ES970380X
1
996, 32, 229-237.
17) Di Corcia, A.; Marchetti, M. Environ. Sci. Technol. 1992, 26, 66-
4.
(
X
7
Abstract published in Advance ACS Abstracts, October 15, 1997.
3
6 4 6
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 31, NO. 12, 1997