Organic Compounds in the Environment

5 downloads 0 Views 111KB Size Report
Organic Compounds in the Environment. Effects of Dissolved Organic Carbon on Sorption and Mobility of Imidacloprid in Soil. F. Flores-Céspedes, E.
Organic Compounds in the Environment Effects of Dissolved Organic Carbon on Sorption and Mobility of Imidacloprid in Soil F. Flores-Ce´spedes, E. Gonza´lez-Pradas,* M. Ferna´ndez-Pe´rez, M. Villafranca-Sa´nchez, M. Socı´as-Viciana, and M. D. Uren˜a-Amate ABSTRACT

On the other hand, dissolved organic matter is an issue of interest because it has been reported to interact with organic contaminants, so affecting the fate of these contaminants in the soil or aquatic systems. Data reveal that the interaction of pesticides with dissolved organic matter increases the apparent solubility of these chemicals, and thereby results in an increase in the mobility of the chemicals in soils (Chiou et al., 1986; Graber et al., 1995; Nelson et al., 1998). In contrast to the above, it also has been shown that application to soil of endogenous or exogenous organic matter increases its organic carbon content, resulting in increased pesticide sorption and decreased leaching (Johnson et al., 1997; Guo et al., 1993). Imidacloprid is a systemic insecticide with a novel mode of action. This insecticide is effective for controlling aphids, whiteflies, thrips, scales, psyllids, plant bugs, and various other harmful pest species, including resistant strains. It is used as seed-dressing, soil treatment, and foliar treatment in different crops (Tomlin, 1999). This insecticide is widely applied in southeastern Spain where intensive horticulture is practiced. In addition, organic matter amendment of soil is a very common practice in these areas, where the native soils are of poor quality. We have considered it useful to study the sorption and leaching processes of imidacloprid in a calcareous soil, representative of this type of agricultural practice. We will analyze the effect of dissolved organic matter on the adsorption processes by using two different types of organic matter: a commercial peat and an organic chemical such as tannic acid. This might be of great interest because it would allow us to study in a detailed way the influence of the soluble organic fraction present in the organic matter used for the soil amendment. This soluble organic fraction could be a key factor to understanding the processes that govern the fate of pesticides and other organic contaminants in the soil environment. Taking into account the above, and given that data in the literature indicate that several pesticides have been detected in ground water from southeastern Spain (Chiron et al., 1993, 1995; Ferna´ndez-Alba et al., 1998), the main objective of this paper is to measure the effect of dissolved organic carbon on the sorption and mobility of imidacloprid in a calcareous soil from southeastern Spain.

To evaluate the effects of dissolved organic carbon on sorption and mobility of the insecticide imidacloprid [1-(6-chloro-3-pyridinyl) methyl-N-nitro-2-imidazolidinimine] in soils, adsorption and column experiments were performed by using a typical calcareous soil from southeastern Spain and two different types of dissolved organic carbon, that is, dissolved organic carbon extracts from a commercial peat (DOC-PE) and high-purity tannic acid (DOC-TA). The experiments were carried out from a 0.01 M CaCl2 aqueous medium at 25ⴗC. The results obtained from the sorption experiments show that the presence of both DOC-PE and DOC-TA, over a concentration range of 15 to 100 mg L⫺1, produces in all cases a decreasing amount of imidacloprid adsorbed in the soil studied. From the column experiments the retardation coefficients (RC) were calculated for imidacloprid by using either 0.01 M CaCl2 aqueous solution (RC ⫽ 2.10), 0.01 M CaCl2 DOC-PE solution (RC ⫽ 1.65), or 0.01 M CaCl2 DOC-TA solution (RC ⫽ 1.87). The results indicate that mobility of imidacloprid is increased 21.4 and 11.0% in the presence of DOC-PE and DOC-TA solutions, respectively. Dissolved organic carbon reduces imidacloprid sorption by competing with the pesticide molecules for sorption sites on the soil surface, allowing enhanced leaching of imidacloprid and potentially increasing ground water contamination.

E

ntrance of a pollutant into water from use on agricultural land is regulated by the factors controlling the fate of pesticides in the soil, that is, transport, retention, and transformation processes. Transport of the pesticide from the soil might occur through leaching, volatilization, or runoff, while retention of pesticide tends to retard or prevent transport; in addition, pesticides may undergo transformations in the soil as the result of chemical, biological, or photochemical processes. The ultimate fate of a pesticide depends on a combination of these parameters; however, retention may be said to exert the most profound influence of the several processes operating to determine the fate of pesticides in soil systems (Arnold and Briggs, 1990; Cheng, 1990; Mansour, 1993; Roberts and Kearny, 1995; Zebarth and Szeto, 1999; Baer and Calvet, 1999). The structure and properties of the chemical, such as its solubility in water, vapor pressure, and polarity also affect its behavior in the environment. Likewise, soil properties such as organic matter and clay content, pH, and ion exchange capacity affect the behavior of the chemical in the environment.

Department of Inorganic Chemistry, Univ. of Almerı´a, La Can˜ada de San Urbano s/n, 04120, Almerı´a, Spain. Received 31 May 2001. *Corresponding author ([email protected]).

Abbreviations: DOC-PE, dissolved organic carbon extracts from peat; DOC-TA, dissolved organic carbon extracts from tannic acid; HPLC, high performance liquid cromatography; RC, retardation coefficient.

Published in J. Environ. Qual. 31:880–888 (2002).

880

881

FLORES-CE´SPEDES ET AL.: SORPTION AND MOBILITY OF IMIDACLOPRID

MATERIALS AND METHODS Imidacloprid Analytically pure imidacloprid (99.9%) was obtained from Bayer Hispania Industrial (Barcelona, Spain), and used as sorbate in this study. Its molecular formula and selected properties are shown in Fig. 1 (Tomlin, 1999).

Soil Characterization The soil used was a calcareous soil (Calcic Regosol), which is the most representative soil in the area studied. This soil was sampled in the surface layer (0–20 cm) of an 8000-m2 area from the Almerı´a region (southeastern Spain), defined by the Universal Transverse Mercator (UTM) system coordinates x ⫽ 526.100, y ⫽ 4071.700 (36⬚45⬘ N, 2⬚40⬘ W). Six soil samples randomly distributed in the area were mixed and stored until analysis. Triplicate air-dried ⬍2-mm particle size samples were analyzed according to standard methods. Soil pH was determined in a 1:2.5 soil to water suspension using a glass electrode (Jackson, 1982), organic matter content by the Walkley–Black method (Walkley and Black, 1934), clay content by the hydrometer method (Black et al., 1982), cation exchange capacity by the barium acetate method (Mehlich, 1948), and bulk density by using intact soil cores (Uhland, 1949). Water retention characteristics were determined by following the guidelines of Hall et al. (1977) and corrected to take into account the stoniness of the soil. Total nitrogen content was determined by the Kjeldahl method (Jackson, 1982). Specific surface area was calculated at 77.4 K from N2 adsorption isotherms using the Brunauer–Emmett–Teller (BET) equation, and soil porosity was determined by Hg porosimetry by applying the Washburn equation (Washburn, 1921). All these soil properties are given in Table 1.

Dissolved Organic Carbon Two different types of dissolved organic carbon were used. The first was a dissolved organic carbon extract from a commercial peat (Vapo Peat A; Projar S.A., Almerı´a, Spain) containing 36.4% organic carbon and labeled in text as DOC-PE. The soluble fraction solution was prepared by mixing peat (25 g) with demineralized MilliQ water (1000 mL; Millipore, Bedford, MA), followed by 24 h of shaking and filtration through a 0.45-␮m Millipore HA filter. The concentration of the dissolved organic carbon solution obtained by this process was 150 mg C L⫺1 and the resulting pH of the solution was 3.9. The second type of dissolved organic carbon, a high-purity tannic acid (99.9%) from Fluka Chemie (Stockholm, Sweden) containing 62.5% organic carbon, was used as a pure chemical source of dissolved organic carbon. It is labeled in text as DOC-TA. The total surface acidity (R-COOH, Ar-COOH, and ArOH groups) of DOC-PE and DOC-TA was determined using the Schnitzer and Gupta method (Schnitzer and Gupta, 1965). The total amount of carboxylic groups was determined using the method proposed by Schnitzer and Wright (Schnitzer and Gupta, 1965; Schitzer, 1982). The total amount of phenolic hydroxyl groups was determined following the method proposed by Schnitzer (Schnitzer, 1982), that is:

Fig. 1. Chemical structure and physical and chemical properties of imidacloprid.

(Total surface acidity) ⫺ (Carboxylic groups) ⫽ Phenolic hydroxyl groups (mol kg⫺1) These data are given in Table 2.

Sorption Isotherms The sorption experiments were carried out as follows: 0.01 M CaCl2 aqueous solutions containing initial pesticide or dissolved organic carbon concentrations (Co ) were prepared, the ranges being between 2 and 25 mg L⫺1 (imidacloprid), 5 and 69 mg C L⫺1 (DOC-PE), and 3 and 66 mg C L⫺1 (DOCTA). An amount of 3 g of each soil sample and 0.025 L of imidacloprid or dissolved organic carbon solution were placed in several stoppered conical flasks and shaken for 24 h in a thermostatted shaker bath at 25.0 ⫾ 0.1⬚C. As can be seen from Fig. 2, a time of 24 h is long enough to make quite sure that the adsorption equilibrium was reached. After shaking, the solutions were centrifuged and the concentration of imidacloprid in the supernatant liquid was determined by high performance liquid chromatography (HPLC) using a Beckman (Fullerton, CA) liquid chromatographic system equipped with a diode-array detector and data station. The HPLC operating conditions were as follows: separation by isocratic elution was performed on a 150- ⫻ 3.9-mm Nova-Pack LC-18 bonded-phase column (Millipore); sample volume, 20 ␮L; flow rate, 1.0 mL min⫺1; and a mobile phase consisting of an acetonitrile–water mix (35:65). Imidacloprid was analyzed at 270 nm, its wavelength of maximum absorption. The concentration of dissolved organic carbon in the supernatant was determined by using a TOC Carbon Analyzer 5050A (Shimadzu, Kyoto, Japan) with IR detection of CO2 following thermal oxidation. The difference in pesticide or dissolved organic carbon concentrations between the initial and final equilibrium solutions was assumed to be due to sorption and the amount of imidacloprid or dissolved organic carbon retained per kilogram of adsorbent was calculated. Two replicates were analyzed for each imidacloprid or dis-

Table 1. Characteristics of the soil. Soil sample

pH

Organic carbon

Calcic Regosol

8.7

% 1.01

Texture

Bulk density

Water content at 33 KPa

Sand

Silt

Clay

Cation exchange capacity

Total N content

Total pore volume

Surface area

kg L⫺1 1.6

L L⫺ 1 0.13

50.4

% 31.3

16.4

cmolc kg⫺1 17.8

% 0.11

mL g⫺1 0.44

m 2 g⫺1 13.3

882

J. ENVIRON. QUAL., VOL. 31, MAY–JUNE 2002

Table 2. Total surface acidity and carboxylic and phenolic hydroxyl groups of the dissolved organic carbon (DOC). DOC sample† DOC-PE DOC-TA

Total surface acidity 3.76 11.4

Carboxylic groups mol kg⫺1 0.05 1.88

Phenolic hydroxyl groups 3.71 9.55

† DOC-PE, dissolved organic carbon extracted from peat; DOC-TA, dissolved organic carbon extracted from tannic acid.

solved organic carbon concentration. The differences between the analysis of the two replicates were always less than 10%. Blanks containing no imidacloprid and no added dissolved organic carbon (only soil plus 0.01 M CaCl2 ) were used for each series of experiments.

Column Experiments Adsorption dynamics using disturbed soil columns were carried out as follows. Polyvinyl chloride tubes (250 mm long, 70.0 mm i.d.) were used. Silicone ridges were applied around the inside of the column at 5-cm increments to minimize boundary flow. Nylon mesh with an effective pore diameter of 60 ␮m was sealed to the bottom of each column to prevent displacement of the soil from the column and to minimize the dead-end volume. An amount (1150 g) of calcareous soil was added in small increments to the tubes to minimize particle size segregation. The columns were packed to an appropriate bulk density (␳ ⫽ 1.50 kg L⫺1 ). Prior to the application of the insecticide, the columns were saturated with distilled water via capillarity and then left to drain for 24 h. The liquid-filled pore volume was determined as the difference between the mass of the soil in the column, after excess water had been added and drainage had effectively ceased, and the oven-dry mass of soil (Vp ⫽ 0.320 L). The type of column experiment carried out was that involving a pulse experiment, in which a pulse of imidacloprid was added to the top of the column and leached with a known amount of water, and its concentration in the leachate measured. The objective of this experiment is to determine the leaching potential of the pesticide imidacloprid in the soil used. Prior to conducting the experiments, the columns were

saturated with 0.01 M CaCl2 aqueous solution from the bottom to rising capillarity (Achik et al., 1988). The pulse experiments were carried out as follows. Fifty grams of soil were contaminated with 0.025 L of a methanol solution containing 5 mg of imidacloprid (dose equivalent to 13 kg ha⫺1 ). The soil was then thoroughly mixed and air-dried (Thomas and Banks, 1991). The contaminated soil sample was added to the top of each column, which already contained 1100 g of uncontaminated soil. The upper part of the column was covered with acid-washed sand, as evenly as possible, over the whole surface of the column to facilitate the distribution of the insecticide solution as well as to minimize surface disturbance (Bolan et al., 1986; Gamerdinger et al., 1991; Goicolea et al., 1991). Next, the columns were eluted with 5.0 L of 0.01 M CaCl2 solution (approximately 15 times the pore volume) by using a peristaltic pump (flow equal to 3.3 mL h⫺1 ) and the leachate was periodically collected in different fractions of about 0.030 and 0.080 L, corresponding to periods of 10 and 24 h, respectively. Bromide ions were used as tracers in this study, applying an amount of 10 mg together with the pesticide and using the same experimental conditions as those above for imidacloprid. The concentration of bromide ions in the eluate was determined by capillary electrophoresis (CE) with a Beckman P/ACE 5000 system. All the experiments were carried out by using duplicate columns.

Imidacloprid–Dissolved Organic Carbon Interaction Studies Sorption Experiments The batch experiments were performed to study the effect of dissolved organic carbon on imidacloprid sorption. Both the peat-extracted and the tannic acid studies were carried out as follows. Three grams of soil were pre-equilibrated in a thermostatted shaker bath (25.0 ⫾ 0.1⬚C for 24 h) with 0.025 L of a 0.01 M CaCl2 aqueous solution containing a mix of imidacloprid (Co ⫽ 20 mg L⫺1 ) and dissolved organic carbon of varied concentration (Co ⫽ 0, 15, 35, 70, and 100 mg L⫺1 ); these mixed solutions will be named in the text as DOC-0, DOC-15, DOC-35, DOC-70, and DOC-100, respectively. After equilibration, the solutions were centrifuged and the

Fig. 2. Sorption kinetic of imidacloprid, dissolved organic carbon extracts from peat (DOC-PE), and dissolved organic carbon extracts from tannic acid (DOC-TA) in batch studies. The initial concentration was 20, 49, and 50 mg L⫺1 for imidacloprid, DOC-PE, and DOC-TA, respectively.

FLORES-CE´SPEDES ET AL.: SORPTION AND MOBILITY OF IMIDACLOPRID

concentration of imidacloprid in the supernatant liquid was determined by HPLC as described above. The difference in pesticide concentration between the initial and final equilibrium solutions was assumed to be due to sorption; therefore, it was possible to determine the variation of the amount of imidacloprid retained per kilogram of adsorbent as a function of the concentration of dissolved organic carbon present in solution. Blanks containing no soil were used to check that no previous association occured between imidacloprid and dissolved organic carbon during the pre-equilibration time of the experiments. Column Experiments The effect of dissolved organic carbon on imidacloprid mobility on soil was evaluated by using the same experimental conditions as those corresponding to the columns experiments above, but using as the eluting agent a 0.01 M CaCl2 solution containing 100 mg of DOC-PE or DOC-TA per liter.

RESULTS AND DISCUSSION Sorption Experiments Figure 3 shows the sorption isotherm of imidacloprid on the calcareous soil selected. According to the initial portion of the curves, this isotherm may be classified as an L type of the Giles classification (Giles et al., 1960), which suggests both that this soil has an average affinity for the insecticide imidacloprid and that there is no strong competition from the solvent for sorption sites. The sorption capacity was calculated from the Kf parameter of Freundlich’s adsorption equation (Kipling, 1980). The linear form of this equation is: log X ⫽ log Kf ⫹ n log C

[1]

where X is milligrams of imidacloprid sorbed per kilogram of soil, C the equilibrium solution concentration (mg L⫺1 ), and Kf and n are constants that characterize the sorption capacity for the pesticide. The constant Kf

Fig. 3. Adsorption isotherm of imidacloprid on the calcareous soil.

883

is the amount of pesticide sorbed per kilogram of soil for a C value of 1 mg L⫺1 and n is a measurement of the intensity of adsorption and reflects the degree to which adsorption is a function of concentration (Calvet, 1980; Sa´nchez and Sa´nchez, 1984). The Kf and n values were calculated from the least square method applied to the linear form of the Freundlich equation, the resulting values being Kf ⫽ 1.05 mg1⫺n kg⫺1 Ln and n ⫽ 0.81. The correlation coefficient was 0.997, being significant at the 0.001 probability level. The Kf value for the soil used is similar to that obtained for adsorption of imidacloprid on a calcic-cambisol soil from Valencia (eastern Spain) (Kf ⫽ 0.98 mg1⫺n kg⫺1 Ln ), higher than that obtained for adsorption on a luvic-xerosol (southeastern Spain) (Kf ⫽ 0.33 mg1⫺n kg⫺1 Ln ), and lower than that corresponding to adsorption of this insecticide on a vertisol-cambisol soil from Bologna (northern Italy) (Kf ⫽ 1.99 mg1⫺n kg⫺1 Ln ). Taking into account the characteristics of these soils, this behavior could be related to the fact that the adsorption of imidacloprid is fundamentally conditioned by the organic carbon content of the soil. The higher the organic carbon content, the higher the Kf value (Capri et al., 2001). The Kd value, which denotes the distribution ratio of imidacloprid in the sorbed phase and solution, was also calculated from the fit of the experimental adsorption isotherm (X ⫽ KdC); the resulting value was 0.61 L kg⫺1. From Kd and the organic matter content of the soil sample, the KOC constant (adsorption constant per kilogram of the organic carbon in soil) was calculated for the insecticide imidacloprid and the soil studied [KOC ⫽ (Kd/OC) ⫻ 100]. The value obtained was 60.3 L kg⫺1, which suggests that imidacloprid has a very low potential for adsorption on this type of soil (Madhun et al., 1986). Concerning the sorption of DOC-PE and DOC-TA on the soil studied, initial mass isotherms were used to

884

J. ENVIRON. QUAL., VOL. 31, MAY–JUNE 2002

Fig. 4. Initial mass isotherm corresponding to dissolved organic carbon extracts from peat (DOC-PE) and dissolved organic carbon extracts from tannic acid (DOC-TA) on the calcareous soil.

measure the sorption capacity (Nodvin et al., 1986). For those systems in which the solute to be adsorbed is present in the adsorbent, such as for the adsorption of organic carbon on soils, it is necessary to apply the initial mass isotherm (proposed by Nodvin) in order to correct the amount of organic carbon released from soil and to have reliable data about the amount of organic carbon adsorbed. The amount of DOC-PE and DOC-TA adsorbed on the soil during the experiment was plotted as a function of the concentration of DOC-PE and DOC-TA added to the soil system (Fig. 4). These linear isotherms are described by the relationship: RE ⫽ mXi ⫺ b

[2]

where RE is the amount of DOC-PE and DOC-TA released or adsorbed per kg of soil (mg kg⫺1 ); m is the slope of the straight lines obtained, equal to the fraction of DOC-PE and DOC-TA adsorbed by the soil; Xi is the initial concentration of DOC in solution, expressed as mg kg⫺1 soil; and b is the dissolved organic carbon released from the soil when Xi is zero (mg kg⫺1 ). The values of m (0.23 for DOC-PE and 0.92 for DOCTA) were used to calculate the distribution coefficient (Cd ), which was used to compare the adsorption of DOC-PE and DOC-TA on the soil studied: Cd ⫽ [m/(1 ⫺ m)] ⫻ (V/M)

[3]

where V is the volume of solution (0.025 L) and M is the mass of soil (0.003 kg). The resulting values for Cd were 2.49 L kg⫺1 for adsorption of DOC-PE and 95.8 L kg⫺1 for adsorption of DOC-TA. The fact that DOC-TA has a distribution coefficient 38 times higher than DOC-PE could be justified taking into account that the adsorption of the dissolved organic carbon on soils could proceed mainly by hydrophobic and Van der Waals interactions through the soil organic matter and the carboxilic and phenolic

groups present in those molecules (Jardine et al., 1989). The greater content of phenolic and carboxilic groups of DOC-TA relating to DOC-PE would explain the higher adsorption capacity of DOC-TA than DOC-PE on the soil studied. Figure 5 shows, in terms of percentage, the variation of the amount of imidacloprid adsorbed in the presence of DOC-PE and DOC-TA relating to the amount adsorbed when no dissolved organic carbon is present (⌬X). As can be seen from Fig. 5, the presence of DOCTA produces a continuosly decreasing amount of imidacloprid adsorbed as the DOC-TA concentration increases. On the other hand, as can be also seen in Fig. 5, in the presence of DOC-PE, despite the general decrease of adsorbed imidacloprid observed, the decreasing amount of imidacloprid adsorbed is mitigated as the DOC-PE concentration increases. Taking into account that results obtained from the blank experiments above indicate that there is not a previous association between the dissolved organic carbon and imidacloprid molecules during the adsorption process (Fig. 6), the general fact that adsorption of imidacloprid decreases in the presence of dissolved organic carbon could be justified considering that imidacloprid, DOC-PE, and DOC-TA adsorption on the soil mainly proceeds by hydrophobic interaction. The competition between the dissolved organic carbon and imidacloprid molecules for the adsorption sites of the soil surface would explain the decreasing adsorption of imidacloprid (which does not have a very marked hydrophobic character) in presence of DOC-PE and DOC-TA. Also, the existence of dissolved organic carbon in the aqueous phase would enhance the desorption of imidacloprid, which results in a lower adsorption rate (Lee et al., 1990; Celis et al., 1998). The particular behavior of the DOC-PE molecules relating to the relative increase of adsorbed imidacloprid as the DOC-PE concentration increases might be

FLORES-CE´SPEDES ET AL.: SORPTION AND MOBILITY OF IMIDACLOPRID

885

Fig. 5. Percentage variation of the amount of imidacloprid adsorbed in presence of dissolved organic carbon extracts from peat (DOC-PE) and dissolved organic carbon extracts from tannic acid (DOC-TA) relating to the amount adsorbed when no dissolved organic carbon is present.

explained taking into account that as the DOC-PE concentration increases a higher amount of this type of unspecific chemical is adsorbed on the soil surface, which creates available sites for the imidacloprid molecules (Celis et al., 1998).

Column Experiments Bromide ions, used as tracers, and imidacloprid curves obtained from the pulse experiments after the columns were eluted either with 0.01 M CaCl2 solution,

0.01 M CaCl2 DOC-PE solution, or 0.01 M CaCl2 DOCTA solution, are shown in Fig. 7. As can be seen from Fig. 7, and as expected, the appearance of imidacloprid in the leachate is delayed with respect to the water front, which is determined by the bromide ion movement. This fact may be quantified by using the retardation coefficient (RC); this coefficient is equal to the number of pore volumes that is necessary when eluting the column before a chemical added in the top of the column appears in the leachate. The higher the RC value, the

Fig. 6. Evolution of high performance liquid chromatography (HPLC) chromatogram corresponding to the mix solution of imidacloprid and dissolved organic carbon extracts from peat (DOC) from the beginning (—) to the end (- - -) of the adsorption process.

886

J. ENVIRON. QUAL., VOL. 31, MAY–JUNE 2002

Fig. 7. Curves obtained from the pulse experiment for bromide ions (䉫), imidacloprid eluted with 0.01 M CaCl2 (䊐), 0.01 M CaCl2 dissolved organic carbon extracts from peat (DOC-PE) (䊊), and 0.01 M CaCl2 dissolved organic carbon extracts from tannic acid (DOC-TA) solutions (䉭).

greater the adsorption of the chemical on the soil and the lower mobility, and vice versa. Table 3 shows the RC values obtained for imidacloprid by eluting the column with either 0.01 M CaCl2 aqueous solution, 0.01 M CaCl2 DOC-PE solution, or 0.01 M CaCl2 DOC-TA solution. As can be seen from Table 3, RC values range from 1.65 for the elution of imidacloprid with 0.01 M CaCl2 DOC-PE solution to 2.10 for the elution of imidacloprid with 0.01 M CaCl2 solution. These data have been compared with those obtained from the equation (Kan and Tomson, 1990): R ⫽ 1 ⫹ (␸/␪)Kd/(1 ⫹ KDOCCDOC)

[4]

where ␸ is the soil bulk density, ␪ is the soil volumetric water content, Kd is the distribution coefficient of the pesticide, KDOC is the distribution coefficient of pesticide in the presence of dissolved organic carbon, and CDOC is the concentration of dissolved organic carbon in solution. The KDOC value of the pesticide was calculated from the octanol–water partition coefficient (KOW ) value of the compound using the following equation (Kan and Tomson, 1990): KDOC ⫽ 0.63KOW

[5]

The values obtained for the retardation coefficient (RC) using a CDOC equal to 100 mg L⫺1 are 3.23 when Table 3. The retardation coefficient (RC) for imidacloprid derived by using different eluting agents. RC 0.01 M CaCl2 solution 2.10

0.01 M CaCl2 DOC-PE† solution

0.01 M CaCl2 DOC-TA† solution

1.65

1.87

† DOC-PE, dissolved organic carbon extracted from peat; DOC-TA, dissolved organic carbon extracted from tannic acid.

no dissolved organic carbon is present and 3.01 when dissolved organic carbon is present. The results are in agreement with those obtained with the column experiments, that is, the presence of dissolved organic carbon decreases the retardation factor and enhances the mobility of the pesticide. From Table 3, and taking into account that the RC value obtained for imidacloprid when the column is eluted only with 0.01 M CaCl2 solution is 2.10, we can deduce in terms of percentage the variation of the mobility of imidacloprid when the column is eluted with dissolved organic carbon solution. Considering that the retarding coefficient for imidacloprid is 1.65 when the column is eluted with 0.01 M CaCl2 DOC-PE solution and 1.87 when the column is eluted with 0.01 M CaCl2 DOC-TA solution, it can be easily deduced that mobility of imidacloprid is increased respectively by 21.4 and 11.0% in the presence of DOC-PE and DOC-TA solution by comparing the corresponding RC values with those obtained when the column is eluted with 0.01 M CaCl2 solution. The increased mobility might be explained by considering the low adsorption capacity of this insecticide on the soil surface and taking into account that the adsorption process is additionally diminished in the presence of dissolved organic carbon. The fact that elution with DOC-PE solution causes a higher mobility of imidacloprid than the elution with DOCTA solution, despite of the lower adsorption of imidacloprid in the presence of DOC-TA than in the presence of DOC-PE, might be justified considering that for the concentration of DOC-PE used in this experiment (100 mg C L⫺1 ), it would be possible the binding of imidacloprid molecules to new active sites originated as a consequence of the previous adsorption of the peat on the soil surface. This situation would involve an easier movement of the adsorbed complex (DOC-PE/imidacloprid) through the soil column, owing to its associa-

FLORES-CE´SPEDES ET AL.: SORPTION AND MOBILITY OF IMIDACLOPRID

tion with the dissolved organic carbon used as eluting agent, allowing a higher mobility of the insecticide imidacloprid. Similar results have also been reported by other authors (Lee et al., 1990; Graber et al., 1995). This fact is in accordance with previous experimental results obtained by these same authors by studying the mobility of diuron and atrazine in soil columns filled with peatamended soils (Gonza´lez-Pradas et al., 1998; Socı´asViciana et al., 1999). In these papers, an unexpected behavior is shown of diuron and atrazine relating to their mobility (higher than that expected taking account the lipophilic character of these herbicides) in those soils that were amended with the higher percentages of organic carbon. The explanation given in those papers was that a process would take place for which diuron and atrazine bind to the soil dissolved or ganic carbon, then move through the column and reach the deeper portions of the soil column. Thus, it is possible that the dissolved organic carbon may reduce pesticide sorption through stable soluble organic fraction–pesticide interactions in solution or by competing with the pesticide molecules for sorption sites on the soil surface, allowing an enhanced leaching process.

CONCLUSIONS Considering the above and taking into account that the lipophilic–lipophobic character is generally used for predicting the fate of pesticides in the environment, the results obtained in this paper could add to the knowledge about the fate of imidacloprid and other similar insecticides in areas (such as southeastern Spain) where, due to the poor organic matter content, organic matter amendment of soils is a very common practice. Thus, it is possible that the dissolved organic carbon may reduce imidacloprid sorption, either by competing with the pesticide molecules for sorption sites on the soil surface or by the binding of imidacloprid molecules to new active sites created as a consequence of the previous adsorption of the peat on the soil surface, allowing an enhanced leaching process and an increased ground water potential contaminant for imidacloprid. ACKNOWLEDGMENTS We thank Bayer Hispania Indusrial for samples of imidacloprid. This research was supported by the CICYT Project AMB97-0507.

REFERENCES Achik, J., M. Schiavon, and P. Jamet. 1988. Study of carbofuran movement in soils. Environ. Prot. Eng. 14:143–153. Arnold, D.J., and G.G. Briggs. 1990. Fate of pesticides in soil; predictive and practical aspects. p. 101–122. In D.H. Hutson and T.R. Roberts (ed.) Environmental fate of pesticides. John Wiley & Sons, Chichester, UK. Baer, U., and R. Calvet. 1999. Fate of soil applied herbicides: Experimental data and prediction of dissipation kinetics. J. Environ. Qual. 28:1765–1777. Black, C.A., D.D. Evans, J.L. White, L.E. Ensminger, and F.E. Clark. 1982. Methods of soil analysis. 2nd ed. Agron. Monogr. 9. ASA and SSSA, Madison, WI. Bolan, N.S., D.R. Scotter, J.K. Syers, and R.W. Tillman. 1986. The

887

effect of adsorption on sulfate leaching. Soil. Sci. Soc. Am. J. 50: 1419–1424. Calvet, R. 1980. Adsorption–desorption phenomena. p. 1–30. In R.J. Hance (ed.) Interactions between herbicides and the soil. Academic Press, London. Capri, E., M.G. Camisa, F. Flores-Ce´spedes, C.R. Glass, E. Gonza´lezPradas, and M. Trevisan. 2001. Imidacloprid and pyrimethanil soil sorption. Agronomie 21:57–64. Celis, R., E. Barriuso, and S. Houot. 1998. Sorption and desorption of atrazine by sludge-amended soil: Dissolved organic matter effects. J. Environ. Qual. 27:1348–1356. Cheng, H.H. 1990. Pesticides in soil environment: An overview. p. 1–7. In H.H. Cheng (ed.) Pesticides in the soil environment: Processes, impacts, and modelling. SSSA, Madison, WI. Chiou, C.T., R.L. Malcom, T.I. Brinton, and D.E. Kile. 1986. Water solubility enhancement of some organic pollutants and pesticides by dissolved humic and fulvic acids. Environ. Sci. Technol. 20:502– 508. Chiron, S., A. Ferna´dez-Alba, and D. Barcelo´. 1993. Comparison of on-line solid phase disk extraction to liquid–liquid extraction for monitoring selected pesticides in environmental waters. Environ. Sci. Technol. 27:2352–2359. Chiron, S., A. Valverde, A. Ferna´dez-Alba, and D. Barcelo´. 1995. Automated sample preparation for monitoring groundwater pollution by carbamate insecticides and their transformation products. J. Assoc. Off. Anal. Chem. Int. 78:1346–1352. Ferna´ndez-Alba, A.R., A. Agu¨era, M. Contreras, G. Pen˜uela, I. Ferrer, and D. Barcelo´. 1998. Comparison of various sample handling and analytical procedures for the monitoring of pesticides and metabolites in ground waters. J. Chromatogr. A 823:35–47. Gamerdinger, A.P., A.T. Lemley, and R.J. Wagenet. 1991. Nonequilibrium sorption and degradation of three 2-chloro-s-triazine herbicides in soil–water systems. J. Environ. Qual. 20:815–822. Giles, C.H., T.H. MacEvan, S.N. Nakhwa, and D. Smith. 1960. Studies in adsorption. Part XI. J. Chem. Soc. 111:3973–3993. Goicolea, M.A., J.F. Arranzl, R.J. Barrio, and Z. Go´mez de Balugera. 1991. Adsorption–leaching study of the herbicides metamitron and chloridazon. Pestic. Sci. 32:259–264. Gonza´lez-Pradas, E., M. Villafranca-Sa´nchez, M. Ferna´ndez-Pe´rez, M. Socı´as-Viciana, and M.D. Uren˜a-Amate. 1998. Sorption and leaching of diuron on natural and peat-amended calcareous soil from Spain. Water Res. 9:2814–2820. Graber, E.R., Z. Gerstl, E. Fischer, and U. Mingelgrin. 1995. Enhanced transport of atrazine under irrigation with effluent. Soil Sci. Soc. Am. J. 59:1513–1519. Guo, L., T.J. Bicki, A.S. Felsot, and T.D. Hinesly. 1993. Sorption and movement of alachlor in soil modified by carbon rich wastes. J. Environ. Qual. 22:186–194. Hall, D.G.M., M.J. Reeve, A.J. Thomansson, and V.F. Wright. 1977. Water retention, porosity and density of field soils. Tech. Monogr. 9. Soil Survey of England and Wales, Harpenden, UK. Jackson, M.L. 1982. Ana´lisis quı´mico de suelos. Editorial Omega, Barcelona, Spain. Jardine, P.M., N.L. Weber, and J.F. McCarthy. 1989. Mechanisms of dissolved organic carbon adsorption on soil. Soil Sci. Soc. Am. J. 53:1378–1385. Johnson, A., F. Worral, C. White, A. Walker, T.J. Besieu, and R.J. Williams. 1997. The potential of incorporated organic matter to reduce pesticide leaching. Toxicol. Environ. Chem. 58:47–61. Kan, A.T., and M.B. Tomson. 1990. Ground water transport of hydrophobic compounds in the presence of dissolved organic matter. Environ. Toxicol. Chem. 9:253–263. Kipling, J.J. 1980. Adsorption from solutions of non-electrolytes. Academic Press, London. Lee, D.Y., W.J. Farmer, and Y. Aochi. 1990. Sorption of napropamide on clay and soil in the presence of dissolved organic matter. J. Environ. Qual. 19:567–573. Madhun, Y.A., J.L. Young, and S.C. Fang. 1986. Sorption of bromacil, chlortoluron and diuron by soils. Soil Sci. Soc. Am. J. 50:1467–1471. Mansour, M. 1993. Fate and prediction of environmental chemicals in soils, plants, and aquatic systems. Lewis Publ., Boca Raton, FL. Mehlich, A. 1948. Determination of cation- and anion-exchange properties of soils. Soil Sci. 66:429–445. Nelson, S.D., J. Letey, W.J. Farmer, C.F. Williams, and M. Ben-Hur.

888

J. ENVIRON. QUAL., VOL. 31, MAY–JUNE 2002

1998. Facilitated transport of napropamide by dissolved organic matter in sewage sludge-amended soil. J. Environ. Qual. 27:1194– 1200. Nodvin, S.C., C.T. Driscoll, and G.E. Likens. 1986. Simple partitioning of anions and dissolved organic carbon in a forest soil. Soil Sci. 142:27–35. Roberts, T.R., and P.C. Kearny. 1995. Environmental behaviour of agrochemicals. Vol. 9. Progress in pesticide biochemistry and toxicology. John Wiley & Sons, Chichester, UK. Sa´nchez, M., and M. Sa´nchez. 1984. Aspects of the adsorption of azinphos-methyl by smectites. J. Agric. Food Chem. 32:720–725. Schnitzer, M. 1982. Organic matter characterization. p. 581–594. In A.L. Page et al. (ed.) Methods of soil analysis. Part 2. 2nd ed. Agron. Monogr. 9. ASA and SSSA, Madison, WI. Schnitzer, M., and U.C. Gupta. 1965. Determination of acidity in soil organic matter. Soil Sci. Soc. Am. Proc. 29:274–277. Socı´as-Viciana, M.M., M. Ferna´ndez-Pe´rez, M. Villafranca-Sa´nchez, E. Gonza´lez-Pradas, and F. Flores-Ce´spedes. 1999. Sorption and

leaching of atrazine and MCPA in natural and peat-amended calcareous soils from Spain. J. Agric. Food Chem. 47:1236–1241. Thomas, C.M., and P.A. Banks. 1991. Flurtamore adsorption and mobility in three Georgia soils. Weed Sci. 39:275–279. Tomlin, C. 1999. The pesticide manual. British Crop Protection Council, Surrey, UK. Uhland, R.E. 1949. Physical properties of soil as modified by crops and management. Soil Sci. Soc. Am. Proc. 14:361–366. Walkley, A., and I. Black. 1934. An examination of the Degtjareff method for determining soil organic matter, and a proposed modification of the organic acid tritation method. Soil Sci. 37:29–38. Washburn, E.W. 1921. Note on a method of determining the distribution of pore sizes in a porous material. Proc. Natl. Acad. Sci. USA 7:115–116. Zebarth, B.J., and S.Y. Szeto. 1999. Leaching behavior of chlorinated hydrocarbon soil fumigants in repached soil columns. J. Environ. Qual. 28:564–575.