The influence of maize residues on the mobility and binding of benazolin: investigating physically extracted soil fractions.

Abstract

The amount of non-extractable residues and the distribution of benazolin and its metabolites were evaluated three months after herbicide application (14C-labelled) in physically extracted soil fractions of topsoil layers of undisturbed soil columns with and without incorporated maize straw (14C-labelled). In addition, a variety of wet-chemical and spectroscopic methods were used to characterise the structure of organic carbon within the different soil fractions. The addition of crop residues increased the amount of dissolved organic carbon, enhanced the aromaticity of the organic carbon structure and enforced the aggregation of organomineral complexes. After incorporation of crop residues, an increase in the formation of metabolic compounds of benazolin and of non-extractable residues was detected. These results indicate that the addition of crop residues leads to a decrease in mobility and bioaccessibility of benazolin and its metabolites.

Full text

Page 1

The influence of maize residues on the mobility and binding of
benazolin: Investigating physically extracted soil fractions
Frauke Schnitzlera,*, Arquimedes Lavorentib, Anne E. Bernsa, Norbert Drewesa,
Harry Vereeckena, Peter Burauela
aAgrosphere, ICG IV, Institute of Chemistry and Dynamics of the Geosphere, Forschungszentrum Ju ¨lich GmbH, D-52425 Ju ¨lich, Germany
bEscola Superior de Agricultura ‘‘Luiz de Queiroz’’, Departamento de Cie ˆncias Exatas, Universidade de S~ ao Paulo, Av. Pa ´dua Dias,
11, Caixa Postal 9, CEP 13418-900 Piracicaba (SP), Brazil
Received 15 May 2006; received in revised form 30 August 2006; accepted 2 September 2006
Addition of crop residues increases the amount of non-extractable
residues and decreases the mobility of benazolin and its metabolites.
Abstract
The amount of non-extractable residues and the distribution of benazolin and its metabolites were evaluated three months after herbicide
application (14C-labelled) in physically extracted soil fractions of topsoil layers of undisturbed soil columns with and without incorporated maize
straw (14C-labelled). In addition, a variety of wet-chemical and spectroscopic methods were used to characterise the structure of organic carbon
within the different soil fractions. The addition of crop residues increased the amount of dissolved organic carbon, enhanced the aromaticity of
the organic carbon structure and enforced the aggregation of organomineral complexes. After incorporation of crop residues, an increase in the
formation of metabolic compounds of benazolin and of non-extractable residues was detected. These results indicate that the addition of crop
residues leads to a decrease in mobility and bioaccessibility of benazolin and its metabolites.
? 2006 Elsevier Ltd. All rights reserved.
Keywords: Benazolin; Herbicide; Maize residues; Organic carbon; Soil fractionation
1. Introduction
The incorporation of crop residues into farmland soil may
influence the mobility and bioavailability of pesticides. Despite
many studies concerning the strong correlation between soil
organic matter content and the adsorption or sequestration of
xenobiotics, their binding mechanisms are still not understood
precisely. It is commonly reported that organic amendments
enrich soils of low organic matter content and consequently
promote adsorption of pesticides and reduce pesticide mobility
(Printz et al., 1995). However, only a few studies indicate an
increase in the adsorption of anionic herbicides when applied
to amended soils (Cox et al., 2000). There is also evidence in
the literature that organic matter may block adsorption sites
and thus inhibit the sorption of anionic pesticides (Celis
et al., 2005; Vereecken, 2005). Furthermore, the incorporation
of decomposable carbon may stimulate the biodegradation of
pesticides by increasing the soil microbial activity (Wanner
et al., 2005). On the other hand, soils amended with fresh
organic matter can protect xenobiotics from degradation
through enhanced sorption and preserve them in an extractable
form (Barriuso et al., 1997). Generally, it has been found that
the addition of organic amendment to soil increases the propor-
tion of non-extractable residues of pesticide (Barriuso et al.,
1997). The formation of non-extractable residues, also named
as ‘bound’ residues, may be due to physical sequestration
(Pignatello and Xing, 1996), covalent binding (Hatcher et al.,
1993) and also to incorporation into the soil microbial biomass
* Corresponding author. Tel.: þ49 2461 616 271; fax: þ49 2461 612 518.
E-mail address: f.schnitzler@fz-juelich.de (F. Schnitzler).
0269-7491/$ - see front matter ? 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.envpol.2006.09.020
Environmental Pollution 147 (2007) 4e13
www.elsevier.com/locate/envpol

Page 2

(Stott et al., 1983). Thus, binding mechanisms can be part of
various physical and chemical interactions between com-
pounds and the soil structure. All these findings underline the
important role played by soil organic matter in the mobility
and bioaccessibility of pesticides.
To elucidate the interactions between organic matter and
pollutants in soil, physical methods based on particle-size
fractionation, initially developed to study soil organic matter
dynamics (Christensen, 2001), have recently been further
developed and applied (Se ´quaris et al., 2005). These physi-
cally based methods are expected to only slightly modify the
organic carbon compounds and the nature of their association
with the soil’s mineral components. The use of physical frac-
tionation methods in sorption studies of pesticides in soils has
increased steadily over the past few years (Bayard et al., 1998;
Taylor et al., 2004). However, only a few research groups have
combined physical and chemical methods of soil organic
matter fractionation in order to study pesticide-soil organic
matter interactions and localisation of pesticide ‘bound’ resi-
dues (Benoit et al., 2000; Doick et al., 2005). To our knowl-
edge, the use of14C-labelled crop residues and pesticides in
combination with a gentle particle-size fractionation and
chemical extraction linked to further studies of structural
changes of organic carbon pools has not yet been attempted.
In our work, the influence of maize straw on the mobility
and bioaccessibility of the auxin herbicide benazolin was in-
vestigated in physically extracted soil fractions. Benazolin
(4-chloro-2-oxobenzothiazolin-3-yl acetic acid) is a selec-
tive, systemic, growth-regulator, post-emergence herbicide
(pKa¼ 3.04, 20?C). Benazolin is the corresponding acidic
and first transformation product of benazolin-ethyl (ethyl 4-
chloro-2-oxobenzothiazolin-3-yl) that is degraded in soils by
de-esterification with a half-life of 1e2 days (Leake, 1989).
Further transformation of benazolin occurs via cleavage of
the acetic acid moiety (half-life 2e4 weeks) to thiazolin
(4-chloro-benzothiazolin-2-one) (Leake, 1989). After ring-
opening and the formation of several metabolites, thiazolin
finally mineralises to CO2. Previous studies have been devoted
to the fate and transport of benazolin in lysimeter and column
experiments (Burauel et al., 1995; Drewes, 2005; Jene, 1998;
Leake, 1991).
A variety of wet-chemical and spectroscopic methods can
be used to characterise the structure of the organic carbon
within the different particle-size fractions in the course of de-
composition processes. As shown by numerous studies, due to
their sensitivity and non-destructive nature, fluorescence tech-
niques are well suited for studies of the chemical and physical
properties of dissolved organic carbon (DOC) (Burauel and
Baßmann, 2005; Senesi et al., 1991). Their colloidal behaviour
in solution and thus the stability of the clay-sized particles can
be studied by using photon correlation spectroscopy (particle
hydrodynamic diameter) and electrophoretic mobility (zeta
potential) (Burauel and Baßmann, 2005). So far little attention
has been paid to the potential of photon correlation spectros-
copy (PCS) in determining particle size distributions in natural
waters and soil solutions (Se ´quaris and Lewandowski, 2003).
Zeta potential measurements have been used to characterise
the role of reactive surface sites of clay minerals and iron
oxide particles and their complexation by humic substances
or natural organic matter and of real soil systems (Se ´quaris
and Lewandowski, 2003; Tomba ´cz et al., 2004).
The specific objective of this study is to determine the mo-
bility and bioaccessibility of benazolin by coupling physical
soil fractionation, chemical extraction and physico-chemical
techniques (Fig. 1). This will enable us to evaluate the distri-
bution and retention properties of organic carbon and the
amount of ‘bound’ residues and mobile herbicide metabolites
in unamended and amended undisturbed soil column experi-
ments. These results will enable a better understanding of
binding and remobilisation processes of chemicals in soils.
Fig. 1. Scheme of physical soil fractionation.
5
F. Schnitzler et al. / Environmental Pollution 147 (2007) 4e13

Page 3

2. Materials and methods
2.1.14C-labelled maize straw
Maize plants were grown in a phytochamber CMP 4030 (Conviron) under
controlled conditions. For the labelling procedure,14CO2was produced in
a flask outside the chamber by continuously pumping (100 ml d?1) an aqueous
solution of NaH14CO3(14.2 mg of 50 mCi mmol?1, American Radiolabeled
Chemicals, Inc.) into a solution of hydrochloric acid (3 M). A constant stream
of CO2passing through the hydrochloric acid transported the14CO2into the
chamber. The plant shoots were exposed to the14CO2-enriched atmosphere
for 8 days. At the end of the labelling process, the chamber was flushed
with
and stems), freeze-dried and chopped into pieces of at most one to two
centimetres in length. The specific activity of the maize straw obtained
was equal to 31 ? 4 MBq kg?1, the C content amounted to 40.8 w% and the
C/N ratio was 16.
14CO2-free CO2for 2 days. The maize plants were harvested (leaves
2.2. Chemicals
[Benzene ring-U-14C] benazolin-ethyl (>99% radiochemical purity;
specific14C-activity: 5.846 MBq mg?1) was supplied by AgrEvo UK limited.
The non-labelled benazolin (99.7% purity) was purchased from Riedel-de
Hae ¨n.14C-sodium bicarbonate (50 mCi mmol?1) was obtained from American
Radiolabeled Chemicals, Inc.
All other chemicals and solvents were purchased from the companies
Sigma-Aldrich, Riedel-de Hae ¨n, Fluka and Merck and were of reagent grade
(analytically pure).
2.3. Soil columns
The undisturbed soil columns were removed by inserting 50 cm long steel
cylinders (Ø 20 cm) into the soil of a conventionally cropped farm plot in Mer-
zenhausen (Ju ¨lich, Germany). The soil type was an orthic luvisol with a soil
texture of 6.4% sand, 78.2% silt and 15.4% clay (pipette method after oxida-
tion of organic matter with H2O2). The lower edge of the cylinders was sharp-
ened to facilitate the drilling process and to avoid compaction of the soil core.
In order to collect the leachates the soil columns were placed on steel funnels
equipped with glass beads and a steel gauze (45 mm) to prevent a bottom
discharge from the soil. The cores were placed in a dark, temperature-con-
trolled room at 12?C and irrigated with 8 mm rainwater per week (divided
into two portions). Leachate was collected every week.
2.4. Column experiments and soil sampling
For the present study, one set of columns was treated with14C-maize straw
and non-labelled benazolin. Non-labelled maize straw and14C-benazolin were
applied to the second set. A third set contained columns without maize straw,
but with14C-labelled benazolin. Each set consisted of three columns.
The maize residues were incorporated into the top 0e5-cm soil column
layers. The amount of applied maize straw was equivalent to 10 t ha?1accord-
ing to good agricultural practice which in the case of the14C-labelled straw
corresponded to 0.97 MBq per column. The unamended soil columns were
treated in the same way to account for the disturbance in the columns with
maize derivates. The columns with14C-labelled maize residues were covered
with soda lime traps to absorb the14CO2. Prior to a trap change or watering,
the head space was flushed with nitrogen to evacuate the
the trap.
Benazolin-ethyl was applied 3 months after the maize straw incorporation
at a concentration of 460 g ha?1in accordance with the application rate of the
commercial product Galtak (AgrEvo UK limited). The non-labelled benazolin
application solutions were prepared in water. A separate application solution
was prepared for every column which was applied completely on the
corresponding column. The flasks were rinsed 5 times with water and the
washings were applied to the column. The14C-benazolin stock solution (spe-
cific radioactivity: 0.741 MBq mg?1) was prepared from14C-benazolin-ethyl
14CO2 through
and non-labelled benazolin in ethanol. A separate application solution for
every column was prepared (1.284 MBq per column) and the application
was conducted as described above. A soda lime trap was mounted on top of
the soil columns as described previously in case of14C-labelled maize straw.
The irrigation was continued one day after herbicide application.
The soil columns were sliced into 5 cm soil layers to determine the trans-
location of the crop residue and the herbicide after a total incubation time of
six months. The following results thus reflect the distribution of the chemicals
remaining at the time of sampling.
2.5. Particle-size fractionation
Beforehand, the soil moisture was determined and moist soil samples of
the topsoil layers were weighed to the desired dry amount of 100 g. Soil sam-
ples of the columns were then passed through a 2 mm sieve and transferred to
a bottle containing 200 ml of 1 mM CaCl2and shaken on a horizontal shaker
(150 rpm) for 6 h. 600 ml of 1 mM CaCl2was then added before starting the
sedimentation procedure (Se ´quaris and Lewandowski, 2003). Three particle-
size fractions were separated on the basis of sedimentation times (derived
from Stokes law) for a water height of 12 cm: 2000e20 mm after 6 min,
20e2 mm and <2 mm containing non-settling colloids after sedimentation
for 12 h (Fig. 1). The size fraction <2 mm was separated by centrifugation
at 10,000 ? g for 90 min. The aqueous upper phase, containing the DOC,
was <0.05 mm in size, and the settled phase was finally 2e0.05 mm in size.
The fraction samples were freeze-dried, weighed and analysed for OC and
14C content. Fractionation was performed in duplicate.
2.6. Sorption parameters of different soil fractions
A two-phase partitioning model describes the distribution of organic
contaminants between the mobile and the immobile soil fraction. Possible
binding of14C-benazolin compounds to the dissolved organic carbon could
be neglected (Se ´quaris et al., 2005). The partition coefficient Kd(V m?1) of
organic chemicals is thus given by the ratio of the adsorbed concentration
of organic chemicals in soil or soil fractions cS(m m?1) to the total concen-
tration of organic chemicals in the mobile phase cm(m V?1), which is equal
to the <0.05 mm fraction.
Kd¼cs
cm
ð1Þ
Since the organic carbon content in many soils can be considered to be the
dominant sorbent, especially shown for unpolar compounds, the Kdis often
related to the organic carbon fraction fOC(m m?1) in the soil and is defined
as KOC(V m?1).
KOC¼Kd
fOC
ð2Þ
2.7. Liquid and solid phase14C detection
The amount of14C in the liquid soil fraction <0.05 mm was detected using
a 2700 Tri-Carb liquid scintillation counter (Packard) and Instant Scint-Gel
Plus? (Packard) or Ultima Gold? XR (Packard) as a scintillation cocktail.
The soil samples were combusted in a Biological Oxidizer OX 500 (R.J. Har-
vey Instruments) to analyse the amount of14C. The14CO2/CO2was absorbed
in the scintillation cocktail Oxysolve C-400 (Zinser Analytik) and measured in
the same way as the liquid soil fractions. All samples were measured in
triplicate.
2.8. Soil extraction procedure
The extraction solution acetonitrile/water (4/1) was added to 10 g of dried
soil. Avolume of 10 ml for the sand-sized 2000e20 mm fraction and 20 ml for
the absorptive 20e2 mm fraction was added, respectively. The sample was
shaken with a horizontal shaker (220 rpm) for 60 min, afterwards an ultrasonic
system (Sonorex RK-510, Bandelin electronic radiofrequency: 35 kHz) was
6
F. Schnitzler et al. / Environmental Pollution 147 (2007) 4e13

Page 4

used for 6 min and then the sample was centrifuged for 30 min at 914 g
(Beckman, GPKR Centrifuge). This procedure was repeated four times.
During the fourth procedure the sample was shaken overnight and the ultra-
sonic and centrifugation step were performed the next day.
All supernatants were transferred into a round-bottomed flask and the ace-
tonitrile was removed with a rotary evaporator. The14C-activity of the remain-
ing extract was measured by using liquid
cleaned and concentrated by solid phase extraction (SPE). The amount of
14C-activity in the extracted solid soil samples was determined and this was
regarded as the non-extractable herbicide residue.
14C detection. The extract was
2.9. Solid phase extraction (SPE)
SPE was applied to the extracts after the soil extraction procedure and to
the leachates of the soil columns. The leachates were acidified with concen-
trated hydrochloric acid lowering the pH to a value of 2 before starting the
solid phase extraction.
The C18reversed phase columns (Supelco, DISCOVERY DSC-18LT) were
successively preconditioned with 6 ml methanol and 6 ml distilled water. After
adsorption, the column was rinsed with 10 ml distilled water and dried for half
an hour. The compounds were eluted with 6 ml acetonitrile. The extract was
measured for14C-activity and the proportion of metabolites was determined
by thin layer chromatography.
2.10. Thin layer chromatography (TLC)
For TLC the extracts obtained after SPE were applied to a silica gel plate
(Macherey Nagel, SIL G-25 UV254) with a Linomat IV (Camag). A toluene/
methanol/acetone/acetic acid (80:15:5:1) mixture was used as the mobile
phase. The developed plates were scanned with a Bioimager (BAS 1000 FU-
JIX) yielding retention factors (Rf) of 0.22 to 0.27 for benazolin, 0.42 to 0.46
for thiazolin and 0.59 to 0.64 for benazolin-ethyl. Each sample was measured
in duplicate.
2.11. Photon correlation spectroscopy
The photon correlation spectroscopy (PCS) provides the detection of the
spherical hydrodynamic diameter (z-average diameter dz) of particles. The
measurements were carried out with a Malvern Zetasizer 4 equipped with
a 5 mW He-Ne laser at a scattering angle of 90?. The correlation function
was analysed with the monomodal cumulant method of the Zetasizer 4
software. PCS measurements of <2 mm fractions were performed after dilu-
tion with the <0.05 mm fraction (1/20e1/100). The replication rate of each
sample was ten.
2.12. Zeta potential
A Laser Zee Meter (Pen Kem Model 501) was used for the electrokinetic
studies. The zeta potential (z) was detected in the <2 mm fraction at the same
dilutions as required for the PCS measurements. Each sample was measured
four times.
2.13. Fluorescence spectroscopy: emission spectra
Fluorescence spectra were obtained with a luminescence spectrometer
Cary Eclipse (Varian), which provides corrected spectra for the excitation
and the emission side. The samples of organic carbon were all diluted with
1 mM CaCl2 to a concentration below 5 mg C l?1before the fluorescence
was measured. The excitation wavelength was fixed at 254 nm, which gave
a usable emission range between 300 and 480 nm. The scan speed was
120 nm min?1and the bandwidths of the excitation and emission slits were
set at 5 nm. The humification index (HIXEM) was defined as the sum of fluo-
rescence intensity in the 435 / 480 nm region divided by the sum of intensity
in the 300 / 345 nm region (Zsolnay, 2003). Since this is an ‘‘internal’’ pa-
rameter, no further corrections were necessary (Cox et al., 2000). Each sample
was measured in duplicate.
2.14. Fluorescence spectroscopy: excitation-
emission-matrix (EEM) spectra
The fluorescence intensities were measured with a luminescence spectrom-
eter Cary Eclipse (Varian) at excitation wavelengths ranging from 250 to
450 nm with 5 nm increments and at emission wavelengths between 260
and 700 nm with 1 nm increments. The samples of organic carbon were all
diluted with 1 mM CaCl2 to a concentration below 5 mg C l?1before the
fluorescence was measured. The scan speed was 600 nm min?1and the band-
widths of the excitation and emission slits were set at 5 nm. The corrected
spectra were normalised to the water Raman scatter peak.
2.15. Organic carbon analyses
The amount of total organic carbon in the DOC solutions was detected
with a TOC analyser 5050A from Shimadzu. Each sample was measured three
times.
A Leco CHNS-932 analyser (LECO corporation) was used to determine
the OC content in the solid, freeze-dried soil samples. Three replicates of
each sample were measured.
3. Results and discussion
3.1. Leaching of maize straw and benazolin compounds
The results of the column experiments for
maize straw and benazolin are shown in Fig. 2. Six months
after incorporation of the
25% of the14C-activity was located in the top 0e5 cm soil
layer. Only traces of penetration below 10 cm were detected
(Fig. 2A). The amount of recovered14C-activity in the leach-
ate did not exceed 2% of the applied14C-activity (data not
shown). Most of the applied
14CO2, indicating a strong mineralisation of maize residues
especially during the first few weeks after incorporation.
This is in agreement with previous studies, which showed
a loss by 80% of14C-labelled crop residues due to mineralisa-
tion processes and14CO2developments within a few weeks
(Jenkinson, 1977; Muneer and Oades, 1989; Printz et al.,
1995). The rather small unmineralised part of the maize straw
still remaining was fixed or bound in the upper soil layers after
six months of incubation. It can be assumed that the organic
compounds were either transformed into more stable degrada-
tion products (Muneer and Oades, 1989), or incorporated into
the microbial biomass, or else became part of the soil humic
material (Stott et al., 1983). Hardly any14C-labelled maize
compounds and their metabolites were detected in deeper
soil layers. In addition to solid organic matter, incorporation
of organic amendments to soils leads to the formation of sol-
uble or dissolved organic matter (Vereecken et al., 2001). The
presence of DOM has proved to increase the aqueous solubil-
ity of organic pollutants, decreasing their sorption and facili-
tating transport through soil profiles (Williams et al., 2000).
However, our results indicate a negligible leaching of soluble
maize-derived compounds during the incubation period.
Fig. 2B shows that three months after the application of the
14C-labelled herbicide the majority of the
located in the top 10 cm of the unamended soil. The maximum
penetration ranged from 10 to 25 cm and about 1% of the
14C-labelled
14C-labelled maize straw, up to
14C-activity was collected as
14C-activity was
7
F. Schnitzler et al. / Environmental Pollution 147 (2007) 4e13

Page 5

applied14C-activity was still detected in the bottom layers at
a depth of 40 cm. The concentration of
leachate was determined to approximately 3% of the applied
14C-activity (data not shown). In the amended soil, most of
the14C-activity of the applied herbicide was detected in the
top 5 cm. Only traces of translocation below 20 cm were
measured. The14C-activity recovered from the drainage water
remained below 0.5% of the
shown). The incorporation of maize straw before applying
the herbicide led to a significant retention of
benazolin and its metabolites in the top soil layers. Different
processes have been assumed to explain this result. Cox
et al. (2000) reported that the addition of organic amendment
enriched soils of low organic matter content, modified surfaces
of soils and subsurface materials promoting the adsorption of
anionic herbicides and thus retarding their movement as a sec-
ondary effect. Furthermore, the application of crop residues
can enhance the microbial activity and consequently affect
14C-activity in the
14C-activity applied (data not
14C-labelled
the biodegradation of pesticides (Wanner et al., 2005). How-
ever, Barriuso et al. (1997) found that organic material
protected the pesticides from degradation through an increase
in adsorption thereby preserving them in an extractable form.
All these assumptions emphasise the major and sophisticated
role played by the soil organic constituents with respect to
the environmental fate of pesticides. Therefore, the fate of
the auxin herbicide benazolin and the dynamics of soil organic
carbon in amended and unamended soils were studied
simultaneously.
3.2. Distribution of organic carbon and benazolin
compounds across size fractions
Considering the organic carbon distribution among physi-
cally extracted soil fractions of the top 0e5 cm of unamended
and amended soil columns, Table 1 shows that the highest
amount (w70%) always occurred in the sand-sized fraction
Table 1
Distribution of mass yield and organic carbon as well as14C-activity and partition coefficients KOCof benazolin equivalents among soil fractions in unamended and
amended topsoil column layers
Particle-size fraction
Bulk soil 2000e20 mm 20e2 mm2e0.05 mm
<0.05 mm
Soil
Mass yield (%)
Organic carbon (%)
14C (% benazolin)
KOC(l kg?1)
100
100
100
(5.04 ? 0.02) 103
87.6 ? 0.3
70.11 ? 11.49
62.1 ? 0.1
(6.97 ? 0.24) 103
12.3 ? 0.4
29.67 ? 0.18
28.2 ? 0.8
(7.49 ? 0.45) 103
(6.4 ? 0.1) 10?2
0.13 ? 0.01
7.3 ? 1.2
n.d.
(9.3 ? 0.9) 10?4
0.08 ? 0.01
2.4 ? 0.3
n.d.
Soil þ maize
Mass yield (%)
Organic carbon (%)
14C (% benazolin)
KOC(l kg?1)
100
100
100
(1.33 ? 0.06) 104
89.3 ? 0.3
75.74 ? 4.82
68.4 ? 3.0
(1.40 ? 0.01) 104
10.6 ? 0.1
24.04 ? 0.14
24.8 ? 0.1
(1.60 ? 0.09) 104
(9.2 ? 0.5) 10?2
0.17 ? 0.01
5.7 ? 3.1
n.d.
(7.8 ? 0.5) 10?4
0.06 ? 0.01
1.1 ? 0.1
n.d.
n.d. ¼ not determined.
0 1020 3040 50 6070010 20
applied 14C-activity (%)
304050 6070
0-5
5-10
10-15
15-20
20-25
25-30
30-35
35-40
applied 14C-activity (%)
soil + maize
soil depth (cm)
A
0-5
5-10
10-15
15-20
20-25
25-30
30-35
35-40
soil
soil + maize
soil depth (cm)
B
Fig. 2. Translocation of14C-labelled compounds into soil; A: maize equivalents 6 months after incorporation; B: benazolin equivalents 3 months after application
and 6 months after incorporation of maize straw. Error bars represent the standard errors of the mean.
8
F. Schnitzler et al. / Environmental Pollution 147 (2007) 4e13

Page 6

and decreased with diminishing particle size. The distribution
of organic carbon is directly related to the mass yield charac-
teristics in each fraction (Table 1). In order to compare the
amount of organic carbon within each size fraction, the
specific organic carbon content of each fraction (gC kg?1frac-
tion) was calculated and is shown in Fig. 3A. An amount of
6.3 ? 0.4 g kg?1organic carbon was detected in the 2000e
20 mm fraction, 24.6 ? 0.2 g kg?1in the 20e2 mm fraction
and 16.2 ? 0.6 g kg?1in the <2 mm fraction of the un-
amended soil. This observation is in agreement with results
reviewed by Christensen (1996), who found the highest C con-
centration either in the fine silt (5e2 mm) or in the clay
(<2 mm). The incorporation of maize straw increased the
amount of organic carbon in each size fraction by approxi-
mately 10% after the incubation period of six months. Stem-
mer et al. (1999) assumed that this carbon enrichment in
each fraction was due to the physical breakdown of the fragile
straw and the formation of microbial biomass and metabolites.
The distribution of
(h14C-benazolin and its metabolites) across the size fractions
of topsoil layers without and with incorporated maize straw
decreased progressively from sand through silt to clay (Table
1). In order to scale the equilibrium partitioning of14C-bena-
zolin equivalents in each size fraction, partition coefficients Kd
were calculated (Eq. (1)). Fig. 3B shows the partition coeffi-
cients Kd of the
whole soil, the sand-sized and the silt-sized fraction of the
0e5 cm layers of the unamended and amended soil columns
(initial benazolin concentrations of 0.45 mg g?1). Kdvalues
of the clay-sized fractions were not shown due to the poor con-
tent of mass yield in this fraction (Table 1), causing large er-
rors of the partition coefficients. The Kd values of the
unamended soil samples can be directly related to the specific
organic carbon content in each size fraction (Fig. 3A). Thus, it
can be assumed that the binding of benazolin is dependent on
the organic carbon concentration. The given almost identical
14C-labelled benazolin equivalents
14C-labelled benazolin equivalents in the
KOCvalues support this assumption (Table 1). Furthermore,
the partition coefficients indicate a higher14C-benazolin im-
mobilisation strength for the silt-sized fraction than for the
sand-sized fraction. This increase in benazolin sorption capac-
ity with diminishing particle size can be in part related to the
increase of their corresponding specific surface areas. In addi-
tion, it must be noted that the Kdvalues obtained for14C-be-
nazolin equivalents were much higher than the Kdvalue of
about 0.36 l kg?1determined in a short-time adsorption batch
experiment with benazolin (initial benazolin concentrations of
0.04e5 mg g?1) and homogeneous soil samples of loamy sand
(Jene, 1998). Different soil texture, inner heterogeneity of the
soil and long-time adsorption processes like degradation of be-
nazolin, slow diffusion into soil aggregates and formation of
‘bound’ residues with the soil surface or organic matter can
be responsible for the higher Kdvalues detected.
The addition of maize straw before herbicide application
significantly increased the partition coefficients Kd(Fig. 3B),
especially in the silt-sized fraction. This increase was dispro-
portionately high in comparison to the low enhancement of or-
ganic carbon content by about 10% in all size fractions after
the incorporation of the crop residue (Fig. 3A). However, the
reported Kdvalues do agree with results from Se ´quaris et al.
(2005), who found values of about 125 l kg?1in the sand-sized
fractions and values which ranged from 350 up to 430 l kg?1
in the silt-sized fractions in topsoil samples of an orthic luvisol
amended with sugar beet or rape seed (initial benazolin con-
centrations of 0.02 mg g?1and 0.009 mg g?1, respectively).
Furthermore, Table 1 shows that the incorporation of maize
straw increased the KOCvalues by 100% in the soil and its
fractions. Thus, it can be assumed that the sorption of benazo-
lin compounds is not only related to the concentration of the
organic carbon, but also dependent on the carbon pool struc-
ture, which changed after the addition of maize straw. More-
over, the organic amendment can enhance the metabolism
of benazolin and in this way affect the binding processes.
0
5
10
15
20
25
30
soil
soil + maize
Corg (g kg-1)
A
0
100
200
300
400
500
soilsoil + maize
Kd (l kg-1)
B
soil
2000-20µm
20-2µm
<2µm
soil
2000-20µm
20-2µm
benazolin
equivalents
Fig. 3. Distribution of specific organic carbon Corg(A) and partition coefficients Kdof benazolin equivalents (h14C-benazolin and its metabolites) (B) among the
size fractions in unamended and amended top 0e5 cm soil column layers. Error bars represent the standard errors of the mean.
9
F. Schnitzler et al. / Environmental Pollution 147 (2007) 4e13

Page 7

The non-constancy of KOCvalues for a given pesticide is in
agreement with studies by Benoit et al. (1996), who reported
that both the relative affinity of different organic solids, i.e.
modified with composted straw, and the degradation of the an-
ionic 2,4-D provoked an increase of KOC.
3.3. Distribution of benazolin and its metabolites after
extraction of soil fractions
Physical soil fractionation followed by chemical extraction
allows a detailed characterisation of the retention properties of
organic carbon in the different size fractions. Furthermore, the
proportion of ‘bound’ residues and potentially mobile herbi-
cide metabolites can be evaluated. Therefore the sand- and
the silt-sized fractions of the unamended and amended topsoil
layers were extracted with acetonitrile. Fig. 4 shows that more
than half of the residual
a finding also presented by Burauel et al. (1995). The amount
of ‘bound’ residues increased with diminishing particle size
and with the addition of maize straw. Radiochromatographic
thin-layer analysis of the extracts showed that the major com-
ponent was the relatively non-mobile thiazolin, together with
distinct quantities of an unidentified more hydrophobic metab-
olite and significant amounts of the parent benazolin-ethyl.
Only small quantities of the potential mobile, acidic, metabo-
lite benazolin were present. The amount of thiazolin and the
quantity of the unidentified metabolite decreased with increas-
ing occurrence of non-extractable residues.
The differences in the contents of herbicide ‘bound’ residue
among size fractions indicate different binding types and
capacities of the organic carbon within the fractions. Usually,
the sand-sized fraction mostly consists of non-humified, un-
protected and non-associated particulate organic carbon
(POC) (Six et al., 2002). Several authors reported that with de-
creasing particle size an enhanced hydrophobicity of organic
matter (Capriel et al., 1995), an increased density of reactive
sites (Benoit et al., 2000) and a formation of microaggregates
14C-activity was not extractable,
(<250 mm) (Christensen, 2001) was found. On the basis of
these findings it can be hypothesised that benazolin and espe-
cially its more hydrophobic metabolites like thiazolin were
more easily trapped in silt-sized microaggregates (20e2 mm)
than bound to the non-humified particulate organic matter in
the sand-sized fraction (2000e20 mm).
The occurrence of small quantities of benazolin in the
acetonitrile extracts may be due to the high leaching potential
and the relatively fast degradation of the acidic herbicide. The
relatively high amounts of the parent benazolin-ethyl in the
chemical extracts three months after herbicide application
indicate that the parent was not bioavailable possibly due to
a sequestration in soil aggregates during the incubation period.
This finding is in agreement with studies by Leake (1991),
who after a solvent extraction detected up to 7% of the applied
14C-activity as the parent benazolin-ethyl in the topsoil layers
of lysimeter studies after an incubation of six months.
The addition of maize derivatives led to the same pattern of
metabolites, and to a significant enhancement of non-extract-
able residues of approximately 30% (Fig. 4). However, the
increase of the organic carbon content in each size fraction
was comparatively low at 10% after the incorporation of the
crop residue (Fig. 3A). Thus, the increase of ‘bound’ residues
after the addition of maize straw cannot only be due to the
higher quantity of organic carbon in the size fractions, but
the changing structure of organic carbon pools after the incor-
poration of decomposing organic material and the formation
of metabolites probably affected the various binding processes
of the herbicide in soil fractions.
3.4. Physico-chemical investigations of the
<0.05 mm and <2 mm fraction
A variety of physico-chemical methods were applied to in-
vestigate the structural differences of soil organic carbon in the
finest fractions of unamended and amended topsoil layers.
Chemical and spectroscopic analyses of the <0.05 mm fraction
0%
50%
60%
70%
80%
90%
100%
remainder
n.i. metabolite
thiazolin
benazolin
benazolin-ethyl
´bound´ residue
extracted 14C-activity
2000-20µm20-2µm 2000-20µm
+ maize
20-2µm
+ maize
n.i. = not identified
Fig. 4. Distribution of benazolin and its metabolites among chemically extracted soil fractions in unamended and amended topsoil column layers.
10
F. Schnitzler et al. / Environmental Pollution 147 (2007) 4e13

Page 8

using DOC and HIXEM determination indicate a general
increase in organic carbon concentration, and in the extent
of humification and aromaticity due to progressive decompo-
sition processes of the residual maize straw after incorporation
of crop residues (Fig. 5A). A three-dimensional fingerprint of
the DOC can be obtained by excitation-emission matrix
(EEM) spectroscopy (Coble, 1996; Burauel and Baßmann,
2005). In Fig. 6, EEM spectra of the <0.05 mm fraction with-
out (A) and with applied maize straw (B) are shown after an
incubation period of six months. Two classes of fluorophores
can be generally assigned to the DOM spectra: the humic-
like (emission range: 380e480 nm) and the protein-like fluo-
rophores (emission range: 300e350 nm) (Coble, 1996). In
both DOM fractions only humic-like fluorophores were
detected, but the addition of maize residues enhanced the in-
tensity of the signals and consequently indicate an increase
in the degree of humification of the dissolved organic carbon.
Low-molecular-weight carbohydrates usually serve as the
main substrate for microorganisms, so that their biodegrada-
tion is much faster than that of highly stable aromatic com-
pounds (Gregorich et al., 2003).
In addition, the results of the particle hydrodynamic
diameter, dz, and the zeta potential, z, within the <2 mm frac-
tion (Fig. 5B) indicate an increase of the stability of the soil
aggregates after the application of residual maize straw. The
negative charge of the organic-matter-coated clay and oxide
particles is due to the fact that adsorbed macro ions have an
excess of acidic groups, which cannot be bound to the surface
for steric reasons (Tomba ´cz et al., 2004). A ‘masking’ of
mineral surfaces by adsorbed organic matter and clustering
of organic matter patches at highly adsorbed mineral sites pro-
vides steric and electrostatic stabilisation of the particles and
0
1
2
3
4
5
6
7
8
9
0
2
4
6
8
10
12
14
16
18
DOC (mg l-1)
HIXEM
soilsoil + maize
A
540
560
580
600
620
640
660
680-42
-40
-38
-36
-34
-32
-30
-28
-26
dz (nm)
zeta potential (mV)
soil soil + maize
B
Fig. 5. A: Amount of dissolved organic carbon DOC ( ) and humification indices HIXEM(A) of the <0.05 mm fraction in unamended and amended topsoil
column layers. B: Hydrodynamic particle diameter dz( ) and zeta potential z (A) of the <2 mm fraction in unamended and amended topsoil column layers. Error
bars represent the standard errors of the mean.
300400500600700300400500600700
emission wavelength (nm)
250
300
350
400
450
excitation wavelengh (nm)
A
B
emission wavelength (nm)
250
300
350
400
450
excitation wavelength (nm)
Fig. 6. Excitation-emission-matrix spectra of the <0.05 mm fraction in unamended (A) and amended (B) topsoil column layers.
11
F. Schnitzler et al. / Environmental Pollution 147 (2007) 4e13

Page 9

has a significant effect on particle interaction in a mineral as-
semblage. Thus, within the clay-sized fraction the addition of
maize straw resulted in an enforced formation of aggregates
and an enhanced association of organic carbon on mineral
surfaces.
These physico-chemical investigations of the finest frac-
tions support the hypothesis that the incorporation of maize
residues led to structural changes in soil organic carbon and
organominerals.
3.5. Partition of benazolin and its
metabolites within the leachate
The results of the radiochromatographic thin-layer analyses
of the leachates during the incubation period after the applica-
tion of the herbicide are shown in Fig. 7. The partition of
benazolin and its metabolites were determined relative to the
total14C-activity sampled at weekly intervals within the drain-
age water of the unamended and amended soil columns. Bena-
zolin and thiazolin were directly detected after the application
of the herbicide in the leachate of the unamended soil columns
(Fig. 7A). Small quantities of metabolite A were measured
three weeks after the addition of the herbicide. This observa-
tion is in agreement with lysimeter studies by Jene et al.
(1998), who found that benazolin was the dominant fraction
in the drainage water with values of about 80% in the first
few months accompanied by small amounts of thiazolin and
unidentified metabolites. Leake (1991) reported that degrada-
tion of benazolin-ethyl and benazolin occurred simultaneously
with movement. Furthermore, a two-week retardation of the
herbicide was found in the amended soil columns (Fig. 7B),
before unidentified, more hydrophobic metabolites finally
leached. The
water during the incubation period showed that the incorpora-
tion of maize straw led to a higher metabolism of benazolin
14C-chromatographic analyses of the drainage
possibly due to an enhanced microbial activity in the topsoil.
In addition, previous results showed that the addition of maize
residues increased the amount of organic carbon, enhanced the
aromaticity of the organic compounds and enforced aggrega-
tion of organominerals in the soil (Table 1, Figs. 5 and 6).
Thus, it can be assumed that stronger interaction processes
occur between benazolin metabolites and the soil matrix after
incorporation of crop residues resulting in a decrease of mobil-
ity of the auxin herbicide.
4. Conclusions
The use of
combination with a gentle particle-size fractionation followed
by a chemical extraction opened up the possibility of charac-
terising the distribution of organic carbon in different size
fractions and evaluating the amount of ‘bound’ residues and
mobile herbicide metabolites in them. Furthermore, the chang-
ing structure of soil organic carbon, and thus its possible
retention property, was examined by using wet-chemical and
spectroscopic methods in unamended and amended undis-
turbed column experiments. Finally, our results show that
the addition of crop residues decreases the mobility and bioac-
cessibility of benazolin and its metabolites.
Particle-size fractionation coupled with chemical extraction
and physico-chemical studies is a promising approach for
investigating pesticide availability in relation to interactions
with soil organic constituents with respect to different soil
management practices.
14C-labelled maize straw and benazolin in
Acknowledgements
The authors gratefully acknowledge the assistance of
A. Steffen, J. Noe ¨l and M. Lie `vre.
0
10
20
30
40
50
60
70
80
90
100
222528 31
week
benazolin
n.i. metabolite A

CO2
collected 14C-activity (%)
collected 14C-activity (%)
thiazolin
BA
0
10
20
30
40
50
60
70
80
90
100
22252831
week
n.i. metabolite A
CO2
n.i.
metabolite C
n.i.
metabolite B
retardation
Fig. 7. Partitioning of benazolin and its metabolites relative to the total14C-activity (¼100%) sampled at any given time within the leachates of unamended (A) and
amended (B) soil columns (n.i. ¼ not identified).
12
F. Schnitzler et al. / Environmental Pollution 147 (2007) 4e13

Page 10

References
Barriuso, E., Houot, S., Serra-Wittling, C., 1997. Influence of compost addi-
tion to soil on the behaviour of herbicides. Pesticide Science 49, 65e75.
Bayard, R., Barna, L., Mahjoub, B., Gourdon, R., 1998. Investigation of naph-
thalene sorption in soils and soil fractions using batch and column assays.
Environmental Toxicology and Chemistry 17, 2383e2390.
Benoit, P., Barriuso, E., Houot, S., Calvet, R., 1996. Influence of the nature of
soil organic matter on the sorption-desorption of 4-chlorophenol, 2,4-di-
chlorophenol and the herbicide 2,4-dichlorophenoxyacetic acid (2,4-D).
European Journal of Soil Science 47, 567e578.
Benoit, P., Barriuso, E., Bergheaud, V., Etie ´vant, V., 2000. Binding capacities
of different soil size fractions in the formation of herbicide-bound residues.
Agronomie 20, 505e512.
Burauel, P., Baßmann, F., 2005. Soils as filter and buffer for pesticides e
experimental concepts to understand soil functions. Environmental Pollu-
tion 133, 11e16.
Burauel, P., Hu ¨cker, K., Dust, M., Reinken, G., Fu ¨hr, F., 1995. The fate of
[benzene-U-14C-]benazolin-ethyl in a lysimeter study supported by de-
tailed laboratory experiments and model calculations. In: Walker, A.,
Allen, R., Bailey, S.W., Blair, A.M., Brown, C.D., Gu ¨nther, P.,
Leake, C.R., Nicholls, P.H. (Eds.), Pesticide Movement to Water, BCPC
Monograph no. 62. British Crop Protection Council, pp. 211e216.
Capriel, P., Beck, T., Borchert, H., Gronholz, J., Zachmann, G., 1995. Hydro-
phobicity of the organic matter in arable soils. Soil Biology and Biochem-
istry 27, 1453e1458.
Celis, R., Real, M., Hermosı ´n, M.C., Cornejo, J., 2005. Sorption and leaching
behaviour of polar aromatic acids in agricultural soils by batch and column
leaching tests. European Journal of Soil Science 56, 287e297.
Christensen, B.T., 1996. Carbon in primary and secondary organomineral
complexes. In: Carter, M.R., Stewart, B.A. (Eds.), Advances in Soil Sci-
ence e Structure and Organic Matter Storage in Agricultural Soils. CRC
Lewis Publishers, Boca Raton, pp. 97e165.
Christensen, B.T., 2001. Physical fractionation of soil and structural and func-
tional complexity in organic matter turnover. European Journal of Soil
Science 52, 345e353.
Coble, P.G., 1996. Characterization of marine and terrestrial DOM in seawater
using excitation-emission matrix spectroscopy. Marine Chemistry 51,
325e346.
Cox, L., Celis, R., Hermosı ´n, M.C., Cornejo, J., Zsolnay, A., Zeller, K., 2000.
Effect of organic amendments on herbicide sorption as related to the nature
of the dissolved organic matter. Environmental Science and Technology
34, 4600e4605.
Doick, K.J., Burauel, P., Jones, K.C., Semple, K.T., 2005. Distribution of aged
14C-PCB and14C-PAH residues in particle-size and humic fractions of an
agricultural soil. Environmental Science and Technology 39, 6575e6583.
Drewes, N., 2005. Umsatz verschiedener Ernteru ¨cksta ¨nde in einem Bodensa ¨u-
lenversuchssystem e Einfluss auf die organische Bodensubstanz und den
Transport zweier Xenobiotika. Schriften des Forschungszentrum Ju ¨lich,
Reihe Umwelt/Environment, Volume 59, PhD, RWTH Aachen University,
Germany.
Gregorich, E.G., Beare, M.H., Stoklas, U., St-Georges, P., 2003. Biodegrad-
ability of soluble organic matter in maize-cropped soils. Geoderma 113,
237e252.
Hatcher, P.G., Bortiatynski, J.M., Minard, R.D., December, J., Bollag, J.M.,
1993. Use of high-resolution13C NMR to examine the enzymatic covalent
binding of13C-labelled 2,4-dichlorophenol to humic substances. Environ-
mental Science and Technology 27, 2098e2103.
Jene, B., 1998. Transport of bromide and benazolin in lysimeters and a field
plot with grid suction bases in a sandy soil. PhD thesis, University of
Hohenheim, Germany.
Jene, B., Fent, G., Kubiak, R., 1998. The movement of14C-benazolin and
bromide in large zero-tension outdoor lysimeters and the undisturbed field.
ACS Symposium Series 699, 136e151.
Jenkinson, D.S., 1977. Studies on the decomposition of plant material in soil v.
the effect of plant cover and soil type on the loss of carbon from14C
labelled ryegrass decomposing under field conditions. Journal of Soil
Science 28, 424e434.
Leake, C.R., 1989. Benazolin-ethyl e degradation and mobility in the labora-
tory and outdoors. Aspects of Applied Biology 21, 129e130.
Leake, C.R., 1991. Lysimeter studies. Pesticide Science 31, 363e373.
Muneer, M., Oades, J.M., 1989. The role of Ca-organic interactions in soil
aggregate stability. II. Field studies with14C-labelled straw, CaCO3and
CaSO4$ H2O. Australian Journal of Soil Research 27, 401e409.
Pignatello, J.J., Xing, B., 1996. Mechanisms of slow sorption of organic
chemicals to natural particles. Environmental Science and Technology
30, 1e11.
Printz, H., Mittelstaedt, W., Fu ¨hr, F., 1995. Effect of maize straw amendment
on formation and translocation of residues of methabenzthiazuron in an
orthic luvisol. Journal of Environmental Science Health, Part BdPesti-
cides, Food Contaminants, and Agricultural Wastes 30, 269e288.
Senesi, N., Miano, T.M., Provenzano, M.R., Brunetti, G., 1991. Characteriza-
tion, differentiation, and classification of humic substances by fluorescence
spectroscopy. Soil Science 152, 259e271.
Se ´quaris, J.-M., Lewandowski, H., 2003. Physicochemical characterization of
potential colloids from agricultural topsoils. Colloids and Surfaces A:
Physicochemical and Engineering Aspects 217, 93e99.
Se ´quaris, J.-M., Lavorenti, A., Burauel, P., 2005. Equilibrium partitioning of
14C-benzo(a)pyrene and14C-benazolin between fractionated phases from
an arable topsoil. Environmental Pollution 135, 491e500.
Six, J., Contant, R.T., Paul, E.A., Paustin, K., 2002. Stabilization mechanisms
of soil organic matter: implications for C-saturation of soil. Plant and Soil
241, 155e176.
Stemmer, M., von Lu ¨tzow, M., Kandeler, E., Pichlmayer, F., Gerzabek, M.H.,
1999. The effect of maize straw placement on mineralization of C and N in
soil particle size fractions. European Journal of Soil Science 50, 73e85.
Stott, D.E., Kasim, G., Martin, J.P., Haider, K., 1983. Stabilization and incor-
poration into biomass of specific plant carbons during biodegradation in
soils. Plant and Soil 70, 15e26.
Taylor, J.P., Mills, M.S., Burns, R.G., 2004. Sorption and desorption behaviour
of acetochlor in surface, subsurface and size-fractionated soil. European
Journal of Soil Science 55, 671e679.
Tomba ´cz, E., Libor, Z., Ille ´s, E., Majzik, A., Klumpp, E., 2004. The role of
reactive surface sites and complexation by humic acids in the interaction
of clay mineral and iron oxide particles. Organic Geochemistry 35,
257e267.
Vereecken, H., 2005. Mobility and leaching of glyphosate: a review. Pest
Management Science 61, 1139e1151.
Vereecken, H., Nitzsche, O., Schulze, M., 2001. Analysis of the transport of
hydrophobic organic xenobiotics in the presence of dissolved organic
carbon using soil column experiments. In: Humic Substances and Chemi-
cal Contaminants. Soil Science Society of America, Madison, USA, pp.
449e470.
Wanner, U., Fu ¨hr, F., Burauel, P., 2005. Influence of the amendment of corn
straw on the degradation behaviour of the fungicide dithianon in soil. En-
vironmental Pollution 133, 63e70.
Williams, C.F., Agassi, M., Letey, J., Farmer, W.J., Nelson, S.D., Ben-
Hur, M., 2000. Facilitated transport of napropamide by dissolved organic
matter through soil columns. Soil Science Society of America Journal 64,
590e594.
Zsolnay, A., 2003. Dissolved organic matter: artefacts, definitions, and func-
tions. Geoderma 113, 187e209.
13
F. Schnitzler et al. / Environmental Pollution 147 (2007) 4e13