Formation and transport of the sulfonic acid metabolites of alachlor and metolachlor in soil

Article abstract of DOI:10.1021/es991264s

Alachlor and metolachlor are dechlorinated and transformed into their corresponding ethane sulfonic acid (ESA) metabolites in soil. In a field-disappearance study, it was shown that alachlor ESA was formed at a faster rate and at concentrations 2-4 times

Full text of DOI:10.1021/es991264s

Environ. Sci. Technol. 2001, 35, 2455-2460
oxy-1-methylethyl)acetamide] was applied (6) and from
Formation and Transport of the
Sulfonic Acid Metabolites of
Alachlor and Metolachlor in Soil
groundwater samples in many agricultural areas (7). Alachlor
and metolachlor are detoxified rapidly in nonsensitive plants
via conjugation with glutathione and/ or homoglutathione
(8, 9). The glutathione conjugates degrade to the sulfonic
acid derivatives, which have been reported as major soil
metabolites of acetochlor (10), alachlor (11), and propachlor
(12).
‡
D . S . A G A * , † A N D E . M . T H U R M A N
University of Nebraska at Kearney, Kearney, Nebraska 68849,
and U.S. Geological Survey, 4821 Quail Crest Place,
Lawrence, Kansas 66049
The environmental significance of herbicide metabolites
is not well-known because of the difficulty in the analysis of
polar ionic compounds, such as the sulfonic acids. In many
pesticide metabolism studies, there has been some fraction
of water-soluble metabolic products that are unaccounted
for due to the difficulty of their extraction and analysis, thus
limiting the available information on pesticide conversion
products in many matrixes.
Alachlor and metolachlor are dechlorinated and transformed
into their corresponding ethane sulfonic acid (ESA)
metabolites in soil. In a field-disappearance study, it was
shown that alachlor ESA was formed at a faster rate
and at concentrations 2-4 times higher than metolachlor
ESA, conforming with the observed longer disappearance
half-life of metolachlor (15.5 d) in the field as compared to
alachlor (8 d). Runoff data also showed higher concentrations
of alachlor ESA as compared to metolachlor ESA, even
though they were applied at the same levels. Data from soil
cores showed transport of the ESA compounds in soil to
as far down as 75-90 cm below the surface, at concentrations
ranging from less than 0.5 µg/L to about 50 µg/L. In
contrast, no parent herbicide was detected at these depths.
To understand thoroughly the behavior of pesticides in
the environment, it is important to investigate the fate and
transport of their degradation products under actual agri-
cultural conditions. The specific objectives of this study were
to (i) determine the relative rates of disappearance of alachlor
and metolachlor in experimental field plots and (ii) determine
the persistence and mobility of alachlor, metolachlor, and
their sulfonic acid metabolites in soil.
Experimental Methods
Field-Disappearance Study. A field-disappearance study was
conducted in 1993 and 1994 at the Kansas State University,
Kansas River Valley Experimental Field, located on the Kansas
River alluvial flood plain, near Topeka, KS. In 1993, two
adjacent plots were used; plot 1 was 4.6 m × 30.5 m, whereas
plot 2 was 4.6 m × 18.3 m. In 1994, two new plots (east of
the 1993 field plots) were used; both plots were 3.1 m × 30.5
m. The field plots used in this study had less than 1% slope,
and the water table was at a depth of approximately 5.5 m.
The soil consisted of a Eudora silt-loam with a particle-size
distribution of 44-62% silt, 26-50% sand, and 5-21% clay
(13). The pH of the soil was in the range of 6.8-7.8. Soil
samples were submitted to Huffman Laboratories (Golden,
CO) for analysis of the organic carbon content. The organic
carbon contents of the soil were 0.45, 0.22, and 0.19%, at 15,
45, and 76 cm, respectively.
This observation correlates with the higher log K
oc
values for alachlor (3.33) and metolachlor (3.01) relative to
their corresponding ESA metabolites, alachlor ESA
(2.26), and metolachlor ESA (2.29).
Introduction
The chloroacetanilide herbicides are among the most
frequently used herbicides in the United States both for
agricultural and noncrop applications (1). Included in this
class of herbicides are acetochlor, alachlor, metolachlor, and
propachlor; all are selective herbicides used to control specific
annual grasses and broadleaf weeds. These herbicides and
their metabolites may leach to groundwater where their
degradation potential may be decreased substantially.
Alachlor [2-chloro-N-(2,6-diethylphenyl)-N-(methoxy-
methyl)acetamide] degrades to a more polar compound,
alachlor ESA [2-(2,6-diethylphenyl)methoxymethylamino-
2-oxo-ethanesulfonic acid], which has been found more
frequently and at higher concentrations than the parent
herbicide in groundwater from wells in the Midwestern
United States (2, 3). Although recent toxicological studies on
alachlor ESA indicate that this metabolite is not mutagenic
and does not bioaccumulate at the levels commonly found
in surface and groundwater (4), it is important to investigate
the fate of alachlor ESA in the environment because it is
relatively persistent and mobile (5).
Suction lysimeters (Soil-Moisture Equipment Corp., Santa
Barbara, CA) 5 cm in diameter were installed in duplicate
near the center of each plot at depths of 0.3, 0.6, 0.9, 1.2, and
1.5 m, following the procedure described by the manufac-
turer. Samples were collected whenever there was rain or at
a 2-week interval during dry periods for about 3 months
after application.
Each plot was equipped with a surface water runoff
collection system. A 19-L plastic bucket was installed at the
down-slope end of each plot so that the rim was even with
the top of the soil. A float-activated bilge pump attached to
a battery/ solar panel system was placed about 0.6 cm from
the bottom of each bucket. During surface water runoff, water
was continuously pumped from the buckets into 1875-L
galvanized metal troughs. The water level of the trough from
each field plot was recorded.
The formation of the metolachlor ESA (2-[(2-ethyl-6-
m ethylphenyl)(2-m ethoxy-1-m ethylethyl)am ino]-2-oxo-
ethanesulfonic acid) metabolite in soil has been recently
reported in samples collected from a field plot where
metolachlor [2-chloro-N-(2-ethyl-6-methylphenyl)-N-(meth-
Herbicide was applied in late May of each year using a
boom sprayer at an application rate of 2.8 kg/ ha. In the 1993
study, alachlor (Monsanto, St. Louis, MO) and metolachlor
(Ciba Geigy, Greensboro, NC) were applied together in
duplicate plots (plots 1 and 2). In the 1994 study, the two
herbicides were applied to plot 1, whereas only metolachlor
was applied to plot 2. The herbicides were incorporated by
* Corresponding author phone: (308)865-8802; fax: (308)865-8157;
e-mail: agad@unk.edu.
† University of Nebraska at Kearney.
‡ U.S. Geological Survey.
9
10.1021/es991264s CCC: $20.00
Published on Web 04/27/2001
2001 Am erican Chem ical Society
VOL. 35, NO. 12, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 2 4 5 5

disking to a depth of 10 cm. After the herbicides were
incorporated, corn (Zea mays L.) was planted in each plot.
Soil cores of 30.5-152.4 cm in depth from the surface
were obtained using a 5 cm diameter, split-tube sampler
(CME Co., St. Louis, MO) before and after herbicide ap-
plication. These soil cores were collected near the lysimeters.
Sampling of soil cores was continued at approximately 2-week
intervals. Two soil cores were obtained per plot and were
mixed as composites after they were divided into 15-cm
intervals. The soil composites were placed in polypropylene
bags and frozen at -10 °C until they were thawed for herbicide
analysis.
120 A pore size, C18 Hypersil ODS (Keystone Scientific,
Bellefonte, PA). The mobile phase was 5/ 35/ 60 acetonitrile/
methanol/ phosphate buffer (pH 7.0, 25 mM) at a flow rate
of 1.2 mL/ min. The ESA metabolites have rotational isomers
at room temperature that make separation difficult because
of the presence of two unresolved peaks. In this HPLC
method, it was critical to maintain the oven temperature at
60 °C to obtain baseline separation of the ESA compounds.
Since it was not relevant in this study to determine individual
isomeric forms of each metabolites, the combined concen-
trations of the isomers were measured by raising the HPLC
column temperature to 60 °C. The sample injection volume
was 80 µL. ESA was monitored at a wavelength of 200 nm
with a 4-nm bandwidth. The reference wavelength was set
at 450 nm with a 80-nm bandwidth. The detection limit of
this method is 0.5 µg/ L using a 100-mL water sample for SPE.
To confirm the identity of the ESA compounds, the ultraviolet
spectra were scanned from 190 to 400 nm, and the spectra
were matched to standard spectra in a customized automated
library search. The concentration of alachlor ESA was further
confirmed by SPE-ELISA (14). Immunoassay for metolachor
ESA is not available; therefore, it was only analyzed by HPLC.
Gas Chrom atography/Mass Spectrom etry (GC/MS)
Analysis. For the determination of herbicide concentrations
in water samples by GC/ MS, 100 mL of water was used for
SPE as described previously (15). For soil analysis by GC/
MS, the soil-extraction procedure just described was followed
except that 250 ng of a surrogate compound (deuterated
atrazine-d5) was added to the soil prior to extraction. In
addition, the ethyl acetate fraction of the soil extract was
passed through an ion-exchange SPE resin to reduce the
amount of natural organic acids in the extract, which cause
rapid degradation of the GC/ MS column. The ethyl acetate
fractions from the SPE elution, which contained the parent
herbicides, were evaporated to about 50-75 µL for analysis
by GC/ MS under selected-ion monitoring (SIM) mode.
Synthesis and Characterization of Alachlor and Metola-
chlor ESA. Alachlor ESA and metolachlor ESA were synthe-
sized using the procedure described by Feng (10). Either
alachlor or metolachlor was refluxed with excess (10 times
more than the starting moles of the herbicide) sodium sulfite
in 100 mL of 10% ethanol/ water (10/ 90, v/ v) for 3-6 h or
until the mixture became homogeneous. Following acidi-
fication with sulfuric acid, the product was extracted into
methylene chloride. The methylene chloride was evaporated,
and the reaction products were dissolved in hot ethanol.
The hot ethanol mixture was filtered and allowed to stand
undisturbed for recrystallization of the ESA. The white crystals
that formed were collected and washed several times with
cold ethanol. The chemical structures of the white crystals
were confirmed as alachlor ESA or metolachlor ESA by
negative-ion, fast-atom bombardment (FAB) mass spec-
trometry (6) and by nuclear magnetic resonance spectroscopy
(16).
Extraction Procedure for Soil. Approximately 15-20 g of
soil was extracted in duplicate with 20 mL of a 75/ 25 (v/ v)
methanol/ water mixture in a Teflon-lined, screw-capped test
tube. This mixture was shaken to a slurry using a Vortex
mixer (Daigger and Co. Inc., Wheeling, IL) and heated at 75
°C for 30 min. Then the soil mixture was allowed to equilibrate
and cool to room temperature in a mechanical shaker for at
least 1 h. Each sample then was centrifuged, and the clear
supernatant was poured directly into a 40-mL vial. The
extraction procedure was repeated on the same soil sample,
and the second supernatant was combined with the first.
The combined extracts were evaporated at 50 °C using a
Turbovap (Zymark, Palo Alto, CA) until only 10 mL of water
remained. The concentrate was transferred to a test tube for
automated solid-phase extraction (SPE) using a C18 Sep-Pak
cartridge (Waters, Milford, MA). The C18 cartridges were
preconditioned sequentially with methanol (1 mL), ethyl
acetate (1 mL), methanol (1 mL), and water (3 mL). Then the
soil extracts were passed through the cartridge and eluted
first with ethyl acetate (2.5 mL) followed by methanol (2.5
mL). This sequential elution separated the parent herbicides
(eluted in ethyl acetate) from their more polar ESA metabo-
lites (eluted in methanol) as described previously (14). An
aliquot (about 5 g) of each soil sample was weighed and
dried to correct for the percent moisture content of the soil.
Blank soil samples spiked with 5 µg/ kg alachlor or metolachlor
were extracted to determine whether ESA or OXA (oxanilic
acid derivative) can be formed during the extraction pro-
cedure at elevated temperature. Neither ESA nor OXA were
detected in the spiked blank samples. The recovery of this
extraction procedure ranged from 80 to 90% of the spiked
concentrations.
ELISA Procedure for Alachlor and Metolachlor. The ethyl
acetate fractions from the SPE of the soil samples were
analyzed for alachlor and metolachlor using enzyme-linked
immunosorbent assay (ELISA) obtained from Idetek/ Quantix
Systems (Sunnyvale, CA). First, the ethyl acetate was evapo-
rated to dryness and then reconstituted with 5 mL of water.
The ELISA procedure described in the kit insert was followed.
Optical densities were read at 650 nm on a Vmax microplate
reader with Softmax software (Molecular Devices, Menlo Park,
CA). Concentrations of the analytes were calculated using
four-parameter-fit data reduction. Samples with herbicide
concentrations exceeding the linear working range were
diluted and reanalyzed.
Results and Discussion
Disappearance of Herbicides and Their Metabolites in
Surface Runoff. The disappearance half-lives of the applied
herbicides were determined from their concentrations in the
surface water runoff. The chloroacetanilide herbicides were
observed to decay exponentially, and the disappearance may
be interpreted using first-order kinetics. The equation for a
first-order reaction is
Analysis of alachlor ESA from the methanol fractions was
performed by the SPE-ELISA method described by Aga et al.
(14) using RaPID Alachlor ELISA (Ohmicron Corp, Newtown,
PA).
High-Perform ance Liquid Chrom atography (HPLC)
Analysis. The HPLC analysis of the methanol extracts for the
confirmation of ESA was performed in an HP model 1090
series II liquid chromatograph with a photodiode-array (PDA)
detector (Hewlett-Packard, Palo Alto, CA). An HPLC method
was developed and optimized for the separation of alachlor
ESA and metolachlor ESA in the methanol fraction of the
SPE eluant. The HPLC was equipped with a 4.6 mm × 250
mm reversed-phase column packed with 3-µm particle size,
C ) Coe-kt
(1)
where C is the concentration after time t, Co is the initial
concentration, and k is the rate constant (17). Thus, a plot
of the logarithm of concentration against time gives a straight
line with a slope proportional to the rate constant. The value
9
2 4 5 6 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 12, 2001

TABLE 1. Concentrations (µg/L) of Alachlor, Alachlor ESA, and
Metolachlor in Surface Water Runoff during the 1993
Field-Disappearance Study
event
date
days after
application
alachlor
ESA
alachlor
metolachlor
Plot 1, 1993
35.14
8.88
6/8/93
6/18/93
6/19/93
7/1/93
7/2/93
7/5/93
7/9/93
7/11/93
11
21
22
34
35
39
43
45
24.20
6.30
7.15
48.84
5.95
5.95
5.95
3.73
38.74
11.95
10.62
4.68
3.93
4.31
7.90
2.37
1.74
2.58
0.73
0.94
2.17
2.23
Plot 2, 1993
43.63
12.71
7.90
6/8/93
6/18/93
6/19/93
7/1/93
7/2/93
7/5/93
7/9/93
11
21
22
34
35
39
43
18.10
6.30
5.65
48.25
59.66
2.28
46.93
12.43
10.62
5.83
8.69
2.29
4.54
6.15
0.97
2.05
FIGURE 1. Cumulative rainfall in 1993 (herbicide application date,
May 27) and 1994 (herbicide application date, May 25) at the Kansas
River Valley Experimental Field during the field-disappearance study.
4.25
4.22
rate of herbicide degradation generally increases with
increasing soil temperature within the range of mesophilic
microbial activity (19). Water is required for microbial activity
and affects the oxygen level in soil. Because the average
monthly temperature was comparable during 1993 and 1994,
ranging from 22 to 26 °C, the moisture content of the soil
probably played a more important role in causing the
differences in herbicide half-lives calculated for the two years.
This hypothesis is consistent with previous reports showing
a faster degradation of alachlor in soil with higher moisture
content (19, 20).
The runoff data for the ESA metabolites indicate that
alachlor ESA (Tables 1 and 2) is present in the runoff water
at relatively higher concentrations than metolachlor ESA
(Table 2) even though the application rates for the parent
herbicides were the same. This difference may be due to the
relatively longer half-life of metolachlor as compared to
alachlor; thus, the formation of metolachlor ESA could be
slower. Also, it is possible that the degradation of metolachlor
to several other polar metabolites is more significant than
the metolachlor ESA formation (21).
In the 1993 study, only alachlor ESA was determined
because the metolachlor ESA standard was not available at
that time. The concentration of alachlor ESA reached a
maximum of about 60 µg/ L in runoff water from plot 2 at
approximately 4 weeks after application (Table 1). During
this time, the concentration of the parent herbicide was about
6 µg/ L in the runoff water from plot 2. In the 1994 study, the
concentrations of alachlor ESA in the runoff were significantly
lower; the peak concentration was only 2.7 µg/ L at about 9
weeks after application (Table 2). Again, this can be attributed
to the more frequent rainfall that occurred during 1993. In
contrast, the rain was less frequent and less in volume during
the 1994 study. At lower intensity and frequency of precipi-
tation, soluble metabolites are more likely to be moved into
the subsurface soil by the infiltrating water before runoff
begins. Analysis of post-harvest runoff samples collected at
least 120 d after application showed no detectable concen-
trations of either parent herbicides (alachlor and metolachlor)
or ESA metabolites in both years of field study.
TABLE 2. Concentrations (µg/L) of Alachlor, Metolachlor and
Their ESA Metabolites in Surface-Water Runoff during the
a
1994 Field-Disappearance Study
event
date
days after
application alachlor
alachlor
ESA
metolachlor
ESA
metolachlor
Plot 1, 1994
5/28/94
6/8/94
3
14
21
29
36
58
65
86
39.11
nd
0.42
nd
2.51
2.47
1.44
2.70
1.59
18.04
8.33
7.69
4.5
7.45
3.24
0.69
nd
nd
0.05
nd
0.56
0.22
0.05
0.38
0.34
9.94
2.07
3.89
2.70
0.09
0.12
nd
6/15/94
6/23/94
6/30/94
7/22/94
7/29/94
8/19/94
Plot 2, 1994
5/28/94
6/8/94
3
14
21
28
29
36
58
68
86
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
20.89
12.08
6.34
11.95
6.63
1.58
0.26
1.00
0.29
0.26
0.52
0.26
0.94
0.60
0.52
0.38
1.26
0.22
6/15/94
6/22/94
6/23/94
6/30/94
7/22/94
8/1/94
8/19/94
a
nd, not detected; na, not applicable.
of the slope is taken as k and is used to calculate the half-life,
1/ 2, which is the time needed for 50% disappearance.
Therefore, eq 1 becomes
t
0.6932
t1/ 2
)
(2)
k
In the field-disappearance study conducted in 1993, the
average half-lives were 6.5 d for alachlor and 9 d for
metolachlor (Table 1). Results of the 1994 study (Table 2)
showed slightly longer half-lives for alachlor (8 d) and
metolachlor (15.5 d). The differences in the observed half-
lives between 1993 and 1994 may be attributed to the
difference in the moisture conditions in the field during the
two periods of study. The year 1993 was extremely wet, with
an average total monthly precipitation of 6.86 cm between
the months of June and September as compared to 4.22 cm
in 1994. The cumulative rain during the period of study are
shown in Figure 1.
Transport and Transform ation of Chloroacetanilides in
Soil. Lysimeter Samples. In the 1993 study, alachlor was
detected in the 0.3-m lysimeter from day 1 (at 8.23 µg/ L) to
day 21 (at 0.07 µg/ L) (Figure 2). However, no alachlor was
observed after day 26 in the lysimeters deeper than 0.6 m.
Unlike the parent herbicide, alachlor ESA was transported
at least as deep as 1.5 m below ground surface, where the
Soil temperature and moisture content affect herbicide
biodegradation by controlling microbial activity (18). The
9
VOL. 35, NO. 12, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 2 4 5 7

concentrations ranging from less than 0.05 to 4.84 µg/ L,
suggesting greater transport of metolachlor as compared to
alachlor. Again, metolachlor ESA was not included in the
analysis of the 1993 samples due to unavailability of the
standard.
The unusually high concentration of alachlor ESA in the
1.5-m lysimeter on day 20 can be attributed from previous
application of alachlor at least 2 yr prior to this study. A large
storm occurred few days prior to sampling and resulted in
a flush of stored alachlor ESA metabolites that were readily
transported down to the deepest lysimeter by infiltrating
stormwater. It is possible that the stored alachlor ESA in soil
were quickly flushed downward due to its very high water
solubility, such that the observed concentration in the 1.5-m
lysimeter was much higher than those found in the shallow
lysimeters.
Due to the relatively dry season in the 1994 study, limited
water volumes were obtained from the lysimeters. Most of
the samples obtained were from the 0.3-m-deep lysimeters,
and the sample volumes were not enough for GC/ MS and
HPLC analysis. Therefore, no data from the 1994 lysimeters
are presented here.
Soil Samples. Analysis of herbicides in soil cores comple-
ments the information provided by lysimeter samples.
Whereas the data from lysimeter samples can be used to
infer the rate of transport of compounds in solution, soil
cores can account for the extractable herbicides adsorbed in
soil particles in addition to those dissolved in the soil
porewater. Therefore, for the study of alachlor and metola-
chlor transport, soil cores from several depths of the
experimental plots also were analyzed. Unfortunately, the
soil core samples collected during the 1993 study were not
analyzed right away due to the lack of a reliable analytical
method for soil during the earlier stage of the study.
Significant microbial degradation during storage was sus-
pected, and thus, analysis of these samples was not pursued
further.
Results from the soil core samples collected from the 1994
field study suggest that alachlor ESA is formed at a faster rate
in surface soil than metolachlor ESA. For example, on day
2, the concentration of alachlor ESA in the top 15 cm of soil
from plot 1 was 43.5 µg/ kg (Figure 3A), whereas metolachlor
ESA was only 11.91 µg/ kg (Figure 3B), despite the same
application rate used for the parent herbicides. It can be
calculated that on day 2 alachlor ESA was 3% of the parent
compound and then increased to 10% and 20% on days 8
and 9, respectively. Metolachlor ESA started at 0.4% of the
parent compound on day 2 and increased to 3 and 5% of the
parent compounds on days 8 and 14, respectively. Apparently,
metolachlor ESA forms more slowly than alachlor ESA, which
is consistent with a previous study (22) of enzymatic and
nonenzymatic glutathione conjugation of 2-chloroacetanilide
herbicides. In the said study, the enzymatic activation of the
conjugation reaction by the glutathione transferase was more
significant for alachlor than for metolachlor.
FIGURE 2. Concentration of alachlor and its metabolite, alachlor
ESA, in the unsaturated zone; collected with lysimeters at 0.3-, 0.6-,
0.9-, 1.2-, and 1.5-m depths in 1993.
concentrations ranged from 3 to 73 µg/ L. Alachlor ESA was
observed at considerably higher concentrations than alachlor
in the soil porewater collected from the lysimeters at various
depths. A peak in the concentration of alachlor ESA was
observed between 14 and 20 d after application in the 0.6-,
0.9-, 1.2-, and 1.5-m lysimeters. This period coincides with
several rains of at least 1.25 cm/ d.
It is interesting to note that the concentration of alachlor
ESA increased substantially in the samples collected from
the shallow lysimeters (0.3-0.9 m) between day 82 and day
136. Between these two sampling dates, corn was harvested
from the field plots. It is possible that corn plants took up
the ESA metabolite and then released it back to the soil after
harvest, resulting in a considerable increase in ESA con-
centration in the soil. A previous study (12) reported that
propachlor ESA metabolite produced in soil is taken up by
soybean plants and that this metabolite is not present in
plants grown and treated under hydroponic culture. This
suggests that the final step in the formation of ESA derivatives
occurs in soil and not in plants. On the other hand, this
observed increase in alachlor ESA concentration may have
been caused by the change in hydrology or soil disturbance
that occurred after harvest. Metolachlor was measured in
the 0.3-, 0.6-, and 0.9-m lysimeters up to day 30 at
Alachlor ESA and metolachlor ESA concentrations both
increased progressively until they reached a peak concentra-
tion 9-10 weeks after application. The highest concentration
of alachlor ESA detected in surface soil (0-15 cm) was 210
µg/ kg, which is about 60% of the concentration of alachlor
present in the same soil sample. At 9-10 weeks, the
concentration of alachlor was less than 15% of the initial
concentration applied. On the other hand, the highest
concentration observed for metolachlor ESA in surface soil
was 93 and 128 µg/ kg in plots 1 and 2, respectively. This
corresponds to about 26 and 34% of the metolachlor present
in the same soil sample. Like alachlor, the concentration of
metolachlor during this period was less than 15% of the initial
concentration applied. To compare, the peak concentration
9
2 4 5 8 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 12, 2001

apparent statistical difference between the alachlor and
metolachlor soil half-lives due to the significant spatial
variability in the herbicide concentrations in soil. In addition,
the experimentally determined values of log Koc for alachlor
and metolachlor were 3.33 and 3.01, respectively (23). The
difference in these log Koc values is small, which may also
explain the very similar disappearance half-lives of these
two herbicides in surface soil.
In general, the infiltration rate of metolachlor in soil was
greater than alachlor. For example, metolachlor was detected
in soil at a depth of 30-45 cm at a concentration of 80 µg/ kg
on day 8, which increased to 142 µg/ kg on day 14. On the
other hand, the concentration of alachlor was only 3.6 µg/ kg
on day 8 and 13.3 µg/ kg on day 14 at the same depth. This
observation is consistent with the higher water solubility of
metolachlor (530 ppm at 20 °C) relative to that of alachlor
(242 ppm at 25 °C) and the slightly higher log Koc of alachlor
as compared to metolachlor.
The concentrations of alachlor ESA in soil were almost
always higher than metolachlor ESA by about 2-3 times at
all depths. Results of this study showed that the two ESA
metabolites were transported at higher concentrations and
to greater depths as compared to their parent herbicides.
For example, as high as 140 µg/ kg of alachlor ESA and 122
µg/ kg of metolachlor ESA were observed in soils at the 60-
75-cm depth on day 43. Neither alachlor nor metolachlor
exceeded 3.0 µg/ kg at depths lower than 45 cm. Moreover,
alachlor ESA (13.3 µg/ kg) and metolachlor ESA (18.7 µg/ kg)
were observed at a depth of 75-90 cm on day 297 but not
the parent herbicides. These results are not surprising
considering that the ESA derivatives have higher water
solubility and much lower Koc than the parent herbicides.
The experimentally determined values of log Koc are 2.26 and
2.29 for alachlor ESA and metolachlor ESA (23), respectively.
This field study correlates with a recent study showing that
the concentrations of ESA and OXA metabolites of alachlor
and metolachlor are often much higher than the parent
herbicides in groundwater, tile drain, and stream runoff (24).
Acknowledgments
The authors thank Idetek/ Quantix Systems, Sunnyvale, CA,
for donating the metolachlor ELISA kits. The use of brand or
trade names in this paper is for identification purposes only
and does not constitute endorsement by the U.S. Geological
Survey. The authors also thank Michelle Yockel and Lisa
Zimmerman for their assistance in the field study and data
analysis.
FIGURE 3. (A) Alachlor disappearance and alachlor ESA formation
in the upper 15 cm of soil (plot 1, 1994). (B) Metolachlor disappearance
and metolachlor ESA formation in the upper 15 cm of soil (plot 1,
1994). Note: The scale for metolachlor concentration is on the left
y-axis, but the scale for metolachlor ESA is on the right y-axis. (C)
Metolachlor disappearance and metolachlor ESA formation in the
upper 15 cm of soil (plot 2, 1994)). Note: The scale for metolachlor
concentration is on the left y-axis, but the scale for metolachlor
ESA is on the right y-axis.
Literature Cited
(1) Gianessi, L. P.; Puffer, C. M. Use of Selected Pesticides for
Agricultural Crop Production in the United States; NTIS:
Springfield, VA, 1982-1985; p 490.
(2) Baker, D. B.; Bushway, R. J.; Adams, S. A.; Macomber, C. S.
Environ. Sci. Technol. 1993, 27, 562-564.
(3) Kolpin, D. W.; Goolsby, D. A.; Aga, D. S.; Iverson, J. L.; Thurman,
E. M. Pesticides in Near-Surface Aquifers: Results of the
Midcontinental United States Ground Water Reconnaissance.
In U.S. Geological Survey Open-File Report 93-418; Goolsby, D.
A., Boyer, L. L., Mallard, G. E., Eds.; U.S. Geological Survey:
Denver, 1993; pp 64-74.
(4) Heydens, W. F.; Wilson, A. G. E.; Kraus, L. J.; Hopkins, W. E.;
Hotz, K. J. Toxicol. Sci. 2000, 55, 36-43.
(5) Goolsby, D. A.; Battaglin, W. A.; Fallon, J. D.; Aga, D. S.; Kolpin,
D. W.; Thurman, E. M. Abstracts of the Technical Meeting, U.S.
Geological Survey Toxic Substances Hydrology Program, Sep-
tember 1993; 83 pp.
of alachlor ESA was approximately 15% and that of metola-
chlor ESA was 8% of the initial concentration applied.
In addition to water solubility, herbicide adsorption in
soil plays an important role in determining the susceptibility
to loss by leaching or runoff. Thus, it is helpful to compare
the soil organic carbon partition coefficients (Koc) of these
herbicides and their ESA metabolites. As a matter of fact, the
disappearance half-lives of alachlor (26-40 d) and metola-
chlor (27-30 d) in surface soil are longer than in the surface
runoff, suggesting that soil adsorption may play a more
important role than water solubility in the rate of disap-
pearance of alachlor and metolachlor in soil. There is no
(6) Aga, D. S.; Thurman, E. M.; Yockel, M. E.; Zimmerman, L. R.;
Williams, T. D. Environ. Sci. Technol. 1996, 30, 592-597.
(7) Kalkhoff, S. J.; Kolpin, D. W.; Thurman, E. M.; Ferrer, I.; Barcelo,
D. Environ. Sci. Technol. 1998, 32, 1738-1740.
(8) Field, J. A.; Thurman, E. M. Environ. Sci. Technol. 1996, 30,
1413-1418.
9
VOL. 35, NO. 12, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 2 4 5 9

(9) Lamoureux, G. L.; Stafford, L. E.; Tanaka, F. S. J. Agric. Food
Chem. 1971, 19, 346-350.
(10) Feng, P. C. C. Pestic. Biochem. Physiol. 1991, 40, 136-142.
(11) Sharp, D. B. Alachlor. In Herbicides: Chemistry, Degradation
and Mode of Action; Kearney, P. C., Kaufman, D. D., Eds.;
Dekker: New York, 1988; pp 301-333.
(19) Hamaker, J. W. In Organic Chemicals in the Soil Environment;
Goring, C. A. I., Hamaker, J. W., Eds.; Marcel Dekker: New York,
1972.
(20) Ne`gre, M.; Gennari, M.; Raimondo, E.; Celi, L.; Trevisan, M.;
Capril, E. J. Agric. Food Chem. 1992, 40, 1071-1075.
(12) Lamoureux, G. L.; Rusness, D. G. Pestic. Biochem. Physiol. 1989,
34, 187-204.
(13) Meyer, M. T. Ph.D. Thesis, University of Kansas, 1994.
(14) Aga, D. S.; Thurman, E. M.; Pomes, M. L. Anal. Chem. 1994, 66,
1495-1499.
(15) Thurman, E. M.; Meyer, M. T.; Pomes, M. L.; Perry, C. A.; Schwab,
A. P. Anal. Chem. 1990, 62, 2043-2048.
(16) Morton, M. D.; Walters, F. H.; Aga, D. S.; Thurman, E. M.; Larive,
C. K. J. Agric. Food Chem. 1997, 45, 1240-1243.
(17) Hurle, K.; Walker, A. In Interactions Between Herbicides and the
Soil; Hance, R. J., Ed.; 1988; pp 83-122.
(21) Walker, A.; Moon, Y.; Welch, S. J. Pestic. Sci. 1992, 35, 109-116.
(22) Liu, D.; Maguire, R. J.; Pacepavicius, G. J. Environ. Toxicol. Water
Qual. 1995, 10, 249-258.
(23) Aga, D. S. Ph.D. Thesis, University of Kansas, 1995.
(24) Phillips, P. J.; Wall. G. R.; Thurman, E. M.; Eckhardt, D. A.;
Vanhoesen, J. Environ. Sci. Technol. 1999, 33, 3531-3537.
Received for review November 9, 1999. Revised manuscript
received March 7, 2001. Accepted March 16, 2001.
(18) Gerstl, Z. In Chemistry, Agriculture and the Environment;
Richardson, M. L., Ed.; The Royal Society of Chemistry: London,
1991; pp 332-369.
ES991264S
9
2 4 6 0 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 12, 2001