- Email: [email protected]
Mobility of isoproturon from an alginate–bentonite controlled release formulation in layered soil
Chemosphere 41 (2000) 1495±1501
Mobility of isoproturon from an alginate±bentonite controlled release formulation in layered soil M. Fern andez-Perez *, E. Gonz alez-Pradas, M. Villafranca-S anchez, F. Flores-Cespedes Departamento de Quõmica Inorg anica, Universidad de Almerõa, 04120 Almerõa, Spain Received 8 July 1999; accepted 4 October 1999
Abstract The mobility of isoproturon [3-(4-isopropylphenyl)-1,1-dimethylurea] from an alginate-based controlled release (CR) formulation was investigated by using soil columns. A layered bed system simulating the typical arrangement under a plastic greenhouse, which is composed of sand, peat, amended soil and native soil was used. The CR formulation was based on sodium alginate (1.87%), isoproturon (1.19%), natural bentonite (3.28%), and water (93.66%), and was compared to technical grade isoproturon. The use of the alginate±bentonite CR formulation produced less vertical mobility of the active ingredient as compared to the technical product. There was no presence of herbicide in the leachate when the alginate±bentonite CR formulation was used. However, 0.90% of isoproturon appeared when the treatment was carried out with technical grade material. Isoproturon mobility was modelled using the programme CMLS, which showed the peat layer to retard pesticide leaching. Analysis of the soil columns showed the highest isoproturon concentration in the peat layer. Ó 2000 Elsevier Science Ltd. All rights reserved. Keywords: Isoproturon; Leaching; Controlled release; Bentonite; Soil
1. Introduction Pesticide pollution of water may come from runo, leaching and improper application. To reduce pesticide pollution, eorts have been made to improve agricultural pest management, such as biological control systems, integrated pest management and controlled release (CR) technology. CR formulations can regulate (and often reduce) the rate of availability of a pesticide, localising the chemical in the crop zone and reducing the amount accessible to leaching processes (Wilkins, 1990). Several research works have contributed to the development of formulations for the controlled release of chemicals in agriculture (Connick, 1982; Schreiber et al., 1993; Cotterill and Wilkins, 1996; Fern andez-Perez et al., *
Corresponding author. Tel.: +34-950-215-611; fax: +34950-215-008. E-mail address: [email protected] (M. FernaÂndez-PeÂrez).
1998). Alginates have been used as matrices for controlled release of pesticides by several workers. Connick et al. (1984) have described a series of alginate-kaolinbased 2,6-Dichlorobenzonitrile CR formulations. In an aquatic release experiment, the authors reported that the release of active ingredient becomes longer as kaolin concentration is increased. Pepperman and Kuan (1993) investigated the controlled release of metribuzin CR formulations. In a water-release study, signi®cant control of metribuzin release rates was obtained with the addition of linseed oil. Johnson and Pepperman (1995) investigated the leaching potential of atrazine alginate linseed oil CR formulations. Atrazine CR formulations with and without linseed oil leached signi®cantly less than a liquid atrazine formulation based on the technical material. Fern andez-Perez et al. (1999) showed that controlled release of diuron was obtained with alginate formulations that contained natural and acid-treated bentonite as modifying agents.
0045-6535/00/$ - see front matter Ó 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 5 - 6 5 3 5 ( 9 9 ) 0 0 5 1 6 - 0
1496
M. Fern andez-Perez et al. / Chemosphere 41 (2000) 1495±1501
The herbicide isoproturon [3-[4-isopropylphenyl)1,1-dimethylurea] is a relatively non-persistent, mobile compound log KOW 2:3, which has been found to leach (Johnson et al., 1994) and has been detected in groundwater (Fisher et al., 1991). CR formulations of isoproturon would potentially extend its eective lifetime, reduce risks of leaching or runo on application, and be safer to the user and non-target organisms. In a previous paper related with the study here presented, Villafranca-S anchez et al. (in press) described a series of alginate-based isoproturon CR formulations. In a water-release study, the authors reported that the use of a natural bentonite as modifying agent of the basic formulation (sodium alginate± isoproturon±water) reduces the release rate of the herbicide as compared to technical grade isoproturon and with alginate formulation without bentonite. The objective of this study is to evaluate the capacity of an alginate±bentonite CR formulation to reduce leaching of isoproturon. Thus, the mobility of isoproturon from an alginate-based CR formulation containing natural bentonite was compared to the technical grade material. In addition, the study included the modelling of the behaviour of technical grade isoproturon in soil columns simulating a typical layered soil structure in a plastic greenhouse. The mobility predicted by the model was compared to that obtained in a laboratory layered soil column.
2. Materials and methods 2.1. Soil characterization The soils used were Camborthid, a Luvic-Xerosol, a sea sand (all from an unused greenhouse in Almerõa) and a commercial peat (Hydro del B altico, Navasa). The soils (air dried and sieved with 2 mm sieve) were characterised using standard methods. The pH was determined in a 1:2.5 soil/water suspension using a glass electrode (Jackson, 1982); organic matter content was determined by the Walkley±Black method (Walkley and Black, 1934); clay content was determined by the hydrometer method (Black et al., 1982); cation exchange capacity was determined by the barium acetate method (Primo-Yufera and Carrasco-Dorrien, 1981); and water
saturation and ®eld capacity were determined following the guidelines by Hall et al. (1977). All these characteristics are shown in Table 1. 2.2. Sorption±desorption studies The sorption experiments were carried out using 0.01 M CaCl2 aqueous solution containing initial isoproturon concentrations (C0 ) between 1.74 and 31.23 mg lÿ1 . Aqueous suspensions of the samples were prepared by adding 25 ml of each isoproturon solution to 3.0 g of soil (0.5 g of peat) in stoppered conical ¯asks and shaken in a thermostated shaker bath at 25°C 0:1°C. Preliminary experiments showed that the time required for equilibrium to be reached between isoproturon sorbed and isoproturon in solution was 24 h. After equilibratium, the suspensions were centrifuged at 9250 g for 10 min., and the concentration of isoproturon in the supernatant liquid was determined by high performance liquid chromatography (HPLC). The amount of isoproturon sorbed (X) was calculated from the dierence between the initial (C0 ) and the equilibrium solution concentrations (C). Blanks containing no isoproturon and three replicates of each sorption point were used for each series of experiments. Desorption experiments were carried out by adding 25 ml of a 0.01 M CaCl2 solution to the stoppered conical ¯asks containing the highest initial pesticide concentration C0 31:23 mg lÿ1 , after removal of the sorption supernatant. This system was again shaken for a 24-h period to establish the new equilibrium. This treatment was followed by centrifugation and determination of the new equilibrium concentration in the supernatant. The amount of isoproturon desorbed in the ®rst equilibration was calculated. This process was repeated four times. Blanks containing no isoproturon were used for each case, and all desorption experiments were carried out in triplicate. The HPLC operating conditions were as follows: separation by isocratic elution on a 150 3:9 mm NovaPack LC-18 bonded-phase column (Waters, Millipore Corporation); sample volume, 20 ll; ¯ow rate, 1.0 ml minÿ1 ; mobile phase, an acetonitrile±water mix (80:20). Isoproturon was analysed at 239 nm, its wavelength of maximum absorption. External standard calibration was used.
Table 1 Characteristics of the four layers of the greenhouse soil Layer
pH
Organic matter (%)
Clay content (%)
Field capacity (% v/v)
Water saturation (% v/v)
C.E.C (meq/100g)
Sand Peat Amended soil Native soil
9.39 3.28 8.44 8.87
0.02 81.32 0.08 0.51
± ± 53.00 8.00
5.80 77.0 38.0 48.0
40 128 52 51
10.00 116.25 10.63 12.50
M. Fern andez-Perez et al. / Chemosphere 41 (2000) 1495±1501
2.3. Controlled release formulations Technical grade isoproturon (98.0%) was supplied by Rh^ one-Poulenc Agro. A natural bentonite (98% montmorillonite, containing sodium as exchange ion), previously described by Gonz alez-Pradas et al. (1983) was used as a modifying agent for the preparation of the alginate-based CR formulation. The CR formulation was obtained using a method similar to that proposed by Fern andez-Perez et al. (1999). That formulation contained sodium alginate (1.87%) from Sigma Chemical, St. Louis, MO, bentonite (3.28%), and technical grade isoproturon (1.19%). Technical grade isoproturon was ®rst dissolved in water. Sodium alginate and bentonite were gradually added, and the slurry was vigorously stirred for 1 h to obtain an homogeneous mixture. The alginate mixture was added dropwise to a 300 ml gellant bath of 0.25 M CaCl2 using the apparatus described by Connick (1982). The resulting beads were allowed to gel in the 0.25 M CaCl2 solution for a total of 5 min, then they were ®ltered and allowed to dry ®rst at room temperature and then in an oven (40°C) to constant weight. The product so obtained will be labelled in the text as IAB (isoproturon± alginate±bentonite). The alginate-based CR formulation IAB contained 137.7 g isoproturon kgÿ1 . 2.4. Layered soil column experiment Mobility of technical and formulated isoproturon was compared in soil columns, simulating the typical arrangement of the dierent layers in a greenhouse, which are composed of sand, peat, amended and native soil. 2.4.1. Column preparation Soil columns were prepared by splitting a poly(vinyl chloride) (PVC) pipe (7 cm i.d., 60 cm length) longitudinally and applying 2-mm thick silicone ridges around the inside of the column at 5-cm increments to minimise boundary ¯ow (Weber and Pepper, 1977). The two parts of each column were then put together and sealed with waterproof adhesive paste. Nylon mesh with an eective pore diameter of 60 lm and lined with a layer of ®berglass wool was sealed to the bottom of each column to prevent displacement of the soil from the columns and minimising the dead-end volume (Fleming et al., 1992). Each column contained from the bottom to the top, the native soil (20 cm), amended soil (20 cm), peat (2 cm) and sand (10 cm). The dierent layers of the soil, screened through a 2-mm sieve, were added to the column in small increments to minimise particle size layering, obtaining the following ®nal bulk densities: 1.17 (native soil), 1.43 (amended soil), 0.26 (peat), and 1.56 g
1497
cmÿ3 (sand). Prior to the application of the herbicide, the columns were saturated with a 0.01 M CaCl2 solution via capillarity and then left to drain for 24 h. 2.4.2. Application of herbicide to the soil columns The herbicide treatments were applied at a rate of 25 kg isoproturon haÿ1 . This rate is approximately 10 times higher than that used to prevent weed infestations and is enough to achieve adequate homogeneity of application over the surface. Both herbicide treatments (technical grade isoproturon, and IAB granules) were applied to duplicate soil columns. Technical grade isoproturon was applied according to the following method. A methanol solution containing 10 mg of isoproturon was added to 50 g (0.5 cm) of washed sand. The washed sand/herbicide mixture was left to dry overnight at room temperature before being added to the top of each column. After application of the herbicide, an additional 0.5 cm of washed sand was added to the top of the column. The alginate-based CR formulation IAB was evenly distributed on the sand layer, after that an additional 1 cm of washed sand was added to the top of the column. 2.4.3. Leaching and leachate collection The leaching solution used in all experiments was 0.01 M CaCl2 . This was done to simulate the soil solution and to prevent dispersion of the soil during the leaching procedure (Johnson and Pepperman, 1995). 1.5 l of solution (the amount of solution equivalent to that applied in greenhouses for one growth season and equivalent to 39 cm irrigation) was leached at a ¯ow rate of 8 ml hÿ1 (4.9 cm dayÿ1 ) using a Gilson Minipuls 3 Peristaltic Pump. The time and volume of each leachate fraction were recorded. Aliquots were taken from the leachate, passed through 0.5 lm PTFE ®lters, and injected directly in the HPLC system. At the end of the leaching procedure (8 days), the columns were allowed to drain for 48 h. 2.4.4. Column analysis The columns were split vertically, and the soil was removed following the established division for the different layers of the column, obtaining four fractions corresponding to the native soil, amended soil, peat, and sand. For the native soil and amended soil, a partition was made in two 10-cm portions labelled as (1) and (2) representing the upper and lower fractions, respectively. Each fraction was dried at room temperature and homogenised. Subsamples of soils, sand, and peat were extracted in conical ¯asks, placed in a shaker bath for 24 h with 25 ml of HPLC grade methanol, ®ltered through Whatman No. 42 paper, and analysed by HPLC, as described above. The extraction eciences of isoproturon were 93% for the sand, 88% for the peat and 85% ands 87% for the amended and native soils, respectively.
1498
M. Fern andez-Perez et al. / Chemosphere 41 (2000) 1495±1501
Table 2 Parameter Kd of the Henry equation and correlation coecients Sample Sand Peat Amended soil Native soil a *
Kd (l kgÿ1 )
r a
0.29 0.06 38.19 4.45 0.39 0.08 5.87 0.31
0.986 0.985 0.991 0.999
These values represent the standard errors. Signi®cant at the 0.001 probability level.
IAB granules were removed from the sand, and the isoproturon content was determined by dissolving 20 granules in a 0.03 M tripolyphosphate solution (20 ml), following by extraction into a water:methanol (80:20) mix (300 ml) and analysed by HPLC, as described above. 2.5. Leaching potential simulation The movement of isoproturon in lab columns that simulate the typical arrangement of the soil layers in a plastic greenhouse was modeled using the computer program `Chemical Movement in Layered Soils' (Nofziger and Hornsby, 1986). This program estimates the location of the peak concentration of non-polar organic chemicals as they move through a soil in response to downward movement of water. The model assumes that chemicals move only in the liquid phase in response to soil water movement. The change in depth of the chemical depends upon the quantity of water moving past the chemical, properties of the chemical, and selected properties of the soil (Nofziger and Hornsby, 1986). The inputs related to the characteristics of the four layers of the greenhouse soil and sorption parameters applied in the model are listed in Tables 1 and 2. The water input, as irrigation was taken from that used in normal agronomic practice for tomatoes in Almerõa greenhouses (Canovas-Martinez and Diaz-Alvarez, 1981). The total water applied was 39 cm over a period of 210 days.
Fig. 1. Sorption isotherms of isoproturon on the dierent layers of a greenhouse soil (error bars represent the standard deviation of three replicates): (a) sand, amended soil and native soil (b) peat.
To evaluate the sorption capacities of the dierent layers of the soil columns, the experimental data were analysed using the linear model (1) X Kd C;
3. Results and discussion 3.1. Sorption±desorption studies Fig. 1 shows the sorption isotherms of isoproturon on the dierent sorbents: sand, amended soil and native soil (Fig. 1(a)), and peat (Fig. 1 (b)). According to the slope of the initial portion of the curves, these isotherms may be classi®ed in general as C-type of the Giles classi®cation (Giles et al., 1960), which suggests a constant partition of the herbicide isoproturon between solution and the solid phase. The isotherm with the steepest initial gradient is that of peat: the sorption capacity of peat for isoproturon is higher than for the other soils.
1
where X is the pesticide adsorbed per kg of sorbent, (mg kgÿ1 ), Kd a parameter related to the solute partitioning between the sorbent and solvent (l kgÿ1 ), and C is the equilibrium solution concentration, (mg lÿ1 ). The Kd values are shown in Table 2. As can be seen from this table, the Kd values vary from 0.29 l kgÿ1 for the sand to 38.19 l kgÿ1 for the peat in the order sand < amended soil < native soil peat: Regression analysis showed that Kd values were correlated with the organic matter content and the cationic exchange capacity, the correlation coecients being higher than 0.99 in both cases.
M. Fern andez-Perez et al. / Chemosphere 41 (2000) 1495±1501
1499
Fig. 2. Cumulative isoproturon desorbed from the dierent layers of a greenhouse soil (error bars represent the standard deviation of three replicates).
The accumulated percentage of desorbed isoproturon from the dierent sorbents, after four desorption cycles, is presented in Fig. 2. These values range between 62.0% for the peat sample and 93.4% for the sand. For these experiments, the variation order is inversely related to that of the Kd values. 3.2. Layered soil column experiment Fig. 3 shows the isoproturon concentrations present in the soil layers and leachate. Slight dierences were observed in the distribution patterns of isoproturon between the technical grade product (Fig. 3(a)) and IAB granules (Fig. 3(b)) for the sand, peat and amended soil layers, although large dierences were found for the deeper layers of the columns. So, the amount of isoproturon found in the native soil (1) and native soil (2) layers was much less in the soil column where isoproturon was applied as IAB granules (1.62 and 0 mg kgÿ1 ) than that obtained in the column treated with technical grade isoproturon (8.73 and 6.51 mg kgÿ1 ). The total percentage of isoproturon recovered in the soil column, leachate and granules referred to the total amount of herbicide applied in the experiments are indicated in Table 3. As can be seen, 73.10% of the total amount was recovered from the soil column when isoproturon was applied as technical grade product, and 0.90% was present in the leachate. On the other hand, when the herbicide was applied as the IAB alginatebased formulation only 14.43% of the total amount was
Fig. 3. Concentration of isoproturon in soil layers of column experiments (error bars represent the standard deviation of two replicates). (a) Technical grade isoproturon (b) IAB formulation.
recovered from the soil column, and no isoproturon was found in the leachate. When the alginate-based formulation IAB was used, the remaining percentage of isoproturon in granules was 59.40%, that is, during the 8 days that experiment lasted 40.60% of isoproturon was released. If we compare the percentage of isoproturon released (40.60%) with that obtained in a previous paper (Villatranca-S anchez et al., in press), where the time necessary for the release of the same percentage of isoproturon from granules in water under static conditions was 4.08 days, we can conclude that the release rate of isoproturon from IAB granules in the soil is two times slower than that observed in water. This fact might be
Table 3 Isoproturon recovered from the granules, soil column and leachate Technical grade isoproturon
IAB
Total % recovered in the granules Total % recovered in the soil column Total % recovered in the leachate
± 73.10 0.90
59.40 14.43 ±
Total % recovery
74.00
73.83
1500
M. Fern andez-Perez et al. / Chemosphere 41 (2000) 1495±1501
explained if we consider that an occlusion of the formulation surface by soil particles takes place, as well as a slower diusion within the soil as compared to water (Ali and Wilkins, 1992). Soil water solutes may also retard movement of pesticides into the aqueous phase (Connick et al., 1984; Sharm et al., 1985). The reduced rate of release of active ingredient in soil compared with water, was also observed with imidacloprid±lignin matrix and diuron-alginate±bentonite granules (Fern andezPerez et al., 1998; 1999). The granules recovered from the soil did not show any signs of disintegration or major degradation at the end of the experiment. The integrity of the CR granule is an important feature of non-erodible controlled release formulations, as unpredictable fracture formation and breakage would increase the release surface of the matrix and thus, the amount of active ingredient released. The total recovery of isoproturon (soil column + leachate + granules) at 8 days was 73.83% of the total applied for IAB treatment and 74.00% for the technical grade isoproturon treatment (Table 3). The dierence to 100% of the total isoproturon applied is probably due to degradation metabolism. These percentages of degraded isoproturon are within the range reported in incubation experiments carried out with soils (Pieuchot et al., 1996; Benoit et al., 1999). 3.3. Leaching potential simulation The simulated mobility of the herbicide with depth (Fig. 4), showed that the slope of leaching velocity varies through the dierent layers. The highest value was obtained for the sand layer, indicating a fast movement of the herbicide, and this corresponds to the lowest value of the Kd parameter of this layer (0.29 l
kgÿ1 ). The slowest value of the slope was obtained for the peat layer, and this corresponds to the highest value of Kd parameter (38.19 l kgÿ1 ). Intermediate values were obtained for the amended and native soils, according to their Kd values. As can be seen from Fig. 4, isoproturon reaches a depth below 0.52 m (standard depth that appears in a typical layer arrangement in a greenhouse) at 120 days. Then, the presence of isoproturon below 0.52 m could lead to contamination of groundwater aquifers. The simulated mobility of the herbicide with depth, was compared to the variation of isoproturon concentration in dierent layers and leachate obtained in experimental conditions with the technical product (Fig. 3(a)). As can be seen, the herbicide concentration found in each layer varies in an opposite way as the slope does in the simulated mobility experiment. So, the highest concentration and the lowest slope correspond to the peat layer, retarding the movement of isoproturon. Once isoproturon breaks through peat, it is leached quickly. Thus, both the model and the soil column experiment, showed that isoproturon can reach the depth below 0.52 m. Prevention of pesticide leaching should therefore focus on reduction of pesticide breakthrough from the peat layer. According to the results obtained from the study with the IAB granules, it is possible that CR formulations could be of bene®t in this respect, because they reduce the rate of isoproturon released to the soil.
4. Conclusions The use of alginate±bentonite CR formulations clearly reduces the vertical mobility of isoproturon into the greenhouse soil layer columns in comparison with the technical grade product. The peat shows the highest sorption capacity, and according to both the column experiments and the model simulation, this layer retards isoproturon movement through the soil. Modelling of this process shows that leaching of isoproturon is rapid once it has moved beyond the peat layer. The use of alginate±bentonite CR formulations is an ecient system that will reduce the amount of pesticide that breaks through the peat layer, and thus reduce the risk of groundwater pollution.
Acknowledgements
Fig. 4. Depth of peak isoproturon concentration as function of time (simulated).
We thank Rh^ one-Poulenc Agro for samples of isoproturon. This research was supported by the CICYT Project AMB93-0600.
M. Fern andez-Perez et al. / Chemosphere 41 (2000) 1495±1501
References Ali, M.A., Wilkins, R.M., 1992. Factors in¯uencing release and distribution of carbofuran from granules into soils. Sci. Total Environ. 123/124, 469±480. Benoit, P., Barriuso, E., Vidon, Ph., Real, B., 1999. Isoproturon sorption and degradation in a soil from grassed buer strip. J. Environ.Qual. 28, 121±129. Black, C.A., Evans, D.D., White, J.L., Ensminger, L.E., Clark, F.E., 1982. Methods of soil analysis. In: Agronomy Monograph 9, second ed., ASA-SSSA, Madison, Wi. Canovas-Martinez, F., Dõaz-Alvarez, J.R., 1981. Cultivos sin Suelo. Instituto de Estudios almerienses, Almerõa. Connick Jr, W.J., 1982. Controlled release of the herbicides 2, 4-D and dichlobenil from alginate gels. J. Appl. Polym. Sci. 27, 3341±3348. Connick, W.J., Bradow, J.H., Wells, W., Steward, K.K., Van, T.K., 1984. Preparation and evaluation of controlled release formulations of 2, 6-Dichlorobenzonitrile. J. Agric. Food Chem. 32, 1199±1205. Cotterill, J.V., Wilkins, R.M., 1996. Controlled release of phenylurea herbicides from a lignin matrix: release kinetics and modi®cation with urea. J. Agric. Food Chem. 44, 2908± 2912. Fern andez-Perez, M., Gonz alez-Pradas, E., Ure~ na-Amate, M.D., Wilkins, R.M., Lindup, I., 1998. Controlled release of imidacloprid from a lignin matrix: water release and soil mobility study. J. Agric. Food Chem. 46, 3826±3834. Fernandez-Perez, M., Villafranca-S anchez, M., Gonz alez-Pradas, E., Flores-Cespedes, F., 1999. Controlled release of diuron from an alginate±bentonite formulation: water release kinetics and soil mobility study. J. Agric. Food Chem. 47, 791±798. Fisher, G.C., Clark, L., Ramsay, P.M., 1991. Pesticides in a chalk catchment: inputs and aquatic residues. In: Proceedings of the British Crop Protection Symposium ± Pesticides in Soil and Water: Current Perspectives, pp. 193±200. Fleming, G.F., Simmons, F., Wax, L.M., Wing, R.E., Carr, M.E., 1992. Atrazine movement in soil columns as in¯uenced by starch-encapsulation and acrylic additives. Weed Sci. 40, 465±470. Giles, C.H., Evan, T.H., Nakhwa, S.N., Smith, D., 1960. Studies in adsorption. Part XI. J. Chem. Soc. 111, 3973±3993. Gonzalez-Pradas, E., L opez-Gonz alez, J.D., Del F., C., ReyBueno, Valenzuela-Calahorro, 1983. Estudio de la super®cie y porosidad de bentonitas homoi onicas, I. Super®cie especõ®ca y porosidad. An. Edaf. y Agrobiol. 3/4, 507±522.
1501
Hall, D.G.M., Reeve, M.J., Thomansson, A.J., Wright, V.F., 1977. Water retention porosity and density of ®eld soils. Soil Survey; Technical Monograph 9, Soil Survey of England and Wales. Harpenden, pp. 6±11. Jackson, M.L., 1982. An alisis quõmico de suelos. Editorial Omega, Barcelona. Johnson, A.C., Haria, A.H., Bhardwaj, C.L., V olkner, C., Batchelor, C.H., Walker, A., 1994. Water movement and isoproturon behaviour in a drained heavy clay soil: 2. Persistence and transport. J. Hydrol. 163, 217±231. Johnson, R.M., Pepperman, A.B., 1995. Mobility of atrazine from alginate controlled release formulations. J. Environ. Sci. Health. B 30, 27±47. Nofziger, D.L., Hornsby, A.G., 1986. A microcomputer-based management tool for chemical movement in soil. Appl. Agric. Resear. 1, 50±56. Pepperman, A.B., Kuan, J.W., 1993. Slow release formulations of metribuzin based on alginate±kaolin±linseed oil. J. Control. Release 26, 21±30. Pieuchot, M., Perrin-Ganier, C., Portal, J.M., Schiavon, M., 1996. Study on the mineralization and degradation of isoproturon in three soils. Chemosphere 33, 467±478. Primo-Y ufera, E., Carrasco-Dorrien, J.M., 1981. Quõmica Agricola I, Suelos y fertilizantes. Editorial Alhambra, Barcelona. Schreiber, M.M., Hickman, M.V., Vail, G.D., 1993. Starchencapsulated atrazine: ecacy and transport. J. Environ. Qual. 22, 443±453. Sharm, S.R., Singh, R.P., Saxena, S.K., Ahmed, S.R., 1985. Eect of dierent saline alkaline salts fertilisers and surfactants on the movement of some carbamoyl group containing pesticides in soil. Anal. Lett. 18, 2322±2343. Villafranca-S anchez, M., Gonz alez-Pradas, E., Fern andezPerez, M., Martinez-L opez, F., Flores-Cespedes, F., Ure~ na-Amate, M.D. Controlled release of isoproturon from an alginate±bentonite formulation: water release kinetics and soil mobility. Pesticide Science (in press). Walkley, A., Black, I.A., 1934. An examination of the Degtjare method for determining soil organic matter and a proposed modi®cation of the chromic acid titration method. Soil Sci. 37, 29±38. Weber, J.B., Peeper, T.F., 1977. Herbicide mobility in soils. In: Truelove, B. (Ed.), Research Methods in Weed Science. Southern Weed Science Society, Auburn, Alabama. Wilkins, R.M., 1990. Controlled Delivery of Crop-Protection Agents. Taylor and Francis, Bristol, PA, p. 320.
Mobility of isoproturon from an alginate±bentonite controlled release formulation in layered soil M. Fern andez-Perez *, E. Gonz alez-Pradas, M. Villafranca-S anchez, F. Flores-Cespedes Departamento de Quõmica Inorg anica, Universidad de Almerõa, 04120 Almerõa, Spain Received 8 July 1999; accepted 4 October 1999
Abstract The mobility of isoproturon [3-(4-isopropylphenyl)-1,1-dimethylurea] from an alginate-based controlled release (CR) formulation was investigated by using soil columns. A layered bed system simulating the typical arrangement under a plastic greenhouse, which is composed of sand, peat, amended soil and native soil was used. The CR formulation was based on sodium alginate (1.87%), isoproturon (1.19%), natural bentonite (3.28%), and water (93.66%), and was compared to technical grade isoproturon. The use of the alginate±bentonite CR formulation produced less vertical mobility of the active ingredient as compared to the technical product. There was no presence of herbicide in the leachate when the alginate±bentonite CR formulation was used. However, 0.90% of isoproturon appeared when the treatment was carried out with technical grade material. Isoproturon mobility was modelled using the programme CMLS, which showed the peat layer to retard pesticide leaching. Analysis of the soil columns showed the highest isoproturon concentration in the peat layer. Ó 2000 Elsevier Science Ltd. All rights reserved. Keywords: Isoproturon; Leaching; Controlled release; Bentonite; Soil
1. Introduction Pesticide pollution of water may come from runo, leaching and improper application. To reduce pesticide pollution, eorts have been made to improve agricultural pest management, such as biological control systems, integrated pest management and controlled release (CR) technology. CR formulations can regulate (and often reduce) the rate of availability of a pesticide, localising the chemical in the crop zone and reducing the amount accessible to leaching processes (Wilkins, 1990). Several research works have contributed to the development of formulations for the controlled release of chemicals in agriculture (Connick, 1982; Schreiber et al., 1993; Cotterill and Wilkins, 1996; Fern andez-Perez et al., *
Corresponding author. Tel.: +34-950-215-611; fax: +34950-215-008. E-mail address: [email protected] (M. FernaÂndez-PeÂrez).
1998). Alginates have been used as matrices for controlled release of pesticides by several workers. Connick et al. (1984) have described a series of alginate-kaolinbased 2,6-Dichlorobenzonitrile CR formulations. In an aquatic release experiment, the authors reported that the release of active ingredient becomes longer as kaolin concentration is increased. Pepperman and Kuan (1993) investigated the controlled release of metribuzin CR formulations. In a water-release study, signi®cant control of metribuzin release rates was obtained with the addition of linseed oil. Johnson and Pepperman (1995) investigated the leaching potential of atrazine alginate linseed oil CR formulations. Atrazine CR formulations with and without linseed oil leached signi®cantly less than a liquid atrazine formulation based on the technical material. Fern andez-Perez et al. (1999) showed that controlled release of diuron was obtained with alginate formulations that contained natural and acid-treated bentonite as modifying agents.
0045-6535/00/$ - see front matter Ó 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 5 - 6 5 3 5 ( 9 9 ) 0 0 5 1 6 - 0
1496
M. Fern andez-Perez et al. / Chemosphere 41 (2000) 1495±1501
The herbicide isoproturon [3-[4-isopropylphenyl)1,1-dimethylurea] is a relatively non-persistent, mobile compound log KOW 2:3, which has been found to leach (Johnson et al., 1994) and has been detected in groundwater (Fisher et al., 1991). CR formulations of isoproturon would potentially extend its eective lifetime, reduce risks of leaching or runo on application, and be safer to the user and non-target organisms. In a previous paper related with the study here presented, Villafranca-S anchez et al. (in press) described a series of alginate-based isoproturon CR formulations. In a water-release study, the authors reported that the use of a natural bentonite as modifying agent of the basic formulation (sodium alginate± isoproturon±water) reduces the release rate of the herbicide as compared to technical grade isoproturon and with alginate formulation without bentonite. The objective of this study is to evaluate the capacity of an alginate±bentonite CR formulation to reduce leaching of isoproturon. Thus, the mobility of isoproturon from an alginate-based CR formulation containing natural bentonite was compared to the technical grade material. In addition, the study included the modelling of the behaviour of technical grade isoproturon in soil columns simulating a typical layered soil structure in a plastic greenhouse. The mobility predicted by the model was compared to that obtained in a laboratory layered soil column.
2. Materials and methods 2.1. Soil characterization The soils used were Camborthid, a Luvic-Xerosol, a sea sand (all from an unused greenhouse in Almerõa) and a commercial peat (Hydro del B altico, Navasa). The soils (air dried and sieved with 2 mm sieve) were characterised using standard methods. The pH was determined in a 1:2.5 soil/water suspension using a glass electrode (Jackson, 1982); organic matter content was determined by the Walkley±Black method (Walkley and Black, 1934); clay content was determined by the hydrometer method (Black et al., 1982); cation exchange capacity was determined by the barium acetate method (Primo-Yufera and Carrasco-Dorrien, 1981); and water
saturation and ®eld capacity were determined following the guidelines by Hall et al. (1977). All these characteristics are shown in Table 1. 2.2. Sorption±desorption studies The sorption experiments were carried out using 0.01 M CaCl2 aqueous solution containing initial isoproturon concentrations (C0 ) between 1.74 and 31.23 mg lÿ1 . Aqueous suspensions of the samples were prepared by adding 25 ml of each isoproturon solution to 3.0 g of soil (0.5 g of peat) in stoppered conical ¯asks and shaken in a thermostated shaker bath at 25°C 0:1°C. Preliminary experiments showed that the time required for equilibrium to be reached between isoproturon sorbed and isoproturon in solution was 24 h. After equilibratium, the suspensions were centrifuged at 9250 g for 10 min., and the concentration of isoproturon in the supernatant liquid was determined by high performance liquid chromatography (HPLC). The amount of isoproturon sorbed (X) was calculated from the dierence between the initial (C0 ) and the equilibrium solution concentrations (C). Blanks containing no isoproturon and three replicates of each sorption point were used for each series of experiments. Desorption experiments were carried out by adding 25 ml of a 0.01 M CaCl2 solution to the stoppered conical ¯asks containing the highest initial pesticide concentration C0 31:23 mg lÿ1 , after removal of the sorption supernatant. This system was again shaken for a 24-h period to establish the new equilibrium. This treatment was followed by centrifugation and determination of the new equilibrium concentration in the supernatant. The amount of isoproturon desorbed in the ®rst equilibration was calculated. This process was repeated four times. Blanks containing no isoproturon were used for each case, and all desorption experiments were carried out in triplicate. The HPLC operating conditions were as follows: separation by isocratic elution on a 150 3:9 mm NovaPack LC-18 bonded-phase column (Waters, Millipore Corporation); sample volume, 20 ll; ¯ow rate, 1.0 ml minÿ1 ; mobile phase, an acetonitrile±water mix (80:20). Isoproturon was analysed at 239 nm, its wavelength of maximum absorption. External standard calibration was used.
Table 1 Characteristics of the four layers of the greenhouse soil Layer
pH
Organic matter (%)
Clay content (%)
Field capacity (% v/v)
Water saturation (% v/v)
C.E.C (meq/100g)
Sand Peat Amended soil Native soil
9.39 3.28 8.44 8.87
0.02 81.32 0.08 0.51
± ± 53.00 8.00
5.80 77.0 38.0 48.0
40 128 52 51
10.00 116.25 10.63 12.50
M. Fern andez-Perez et al. / Chemosphere 41 (2000) 1495±1501
2.3. Controlled release formulations Technical grade isoproturon (98.0%) was supplied by Rh^ one-Poulenc Agro. A natural bentonite (98% montmorillonite, containing sodium as exchange ion), previously described by Gonz alez-Pradas et al. (1983) was used as a modifying agent for the preparation of the alginate-based CR formulation. The CR formulation was obtained using a method similar to that proposed by Fern andez-Perez et al. (1999). That formulation contained sodium alginate (1.87%) from Sigma Chemical, St. Louis, MO, bentonite (3.28%), and technical grade isoproturon (1.19%). Technical grade isoproturon was ®rst dissolved in water. Sodium alginate and bentonite were gradually added, and the slurry was vigorously stirred for 1 h to obtain an homogeneous mixture. The alginate mixture was added dropwise to a 300 ml gellant bath of 0.25 M CaCl2 using the apparatus described by Connick (1982). The resulting beads were allowed to gel in the 0.25 M CaCl2 solution for a total of 5 min, then they were ®ltered and allowed to dry ®rst at room temperature and then in an oven (40°C) to constant weight. The product so obtained will be labelled in the text as IAB (isoproturon± alginate±bentonite). The alginate-based CR formulation IAB contained 137.7 g isoproturon kgÿ1 . 2.4. Layered soil column experiment Mobility of technical and formulated isoproturon was compared in soil columns, simulating the typical arrangement of the dierent layers in a greenhouse, which are composed of sand, peat, amended and native soil. 2.4.1. Column preparation Soil columns were prepared by splitting a poly(vinyl chloride) (PVC) pipe (7 cm i.d., 60 cm length) longitudinally and applying 2-mm thick silicone ridges around the inside of the column at 5-cm increments to minimise boundary ¯ow (Weber and Pepper, 1977). The two parts of each column were then put together and sealed with waterproof adhesive paste. Nylon mesh with an eective pore diameter of 60 lm and lined with a layer of ®berglass wool was sealed to the bottom of each column to prevent displacement of the soil from the columns and minimising the dead-end volume (Fleming et al., 1992). Each column contained from the bottom to the top, the native soil (20 cm), amended soil (20 cm), peat (2 cm) and sand (10 cm). The dierent layers of the soil, screened through a 2-mm sieve, were added to the column in small increments to minimise particle size layering, obtaining the following ®nal bulk densities: 1.17 (native soil), 1.43 (amended soil), 0.26 (peat), and 1.56 g
1497
cmÿ3 (sand). Prior to the application of the herbicide, the columns were saturated with a 0.01 M CaCl2 solution via capillarity and then left to drain for 24 h. 2.4.2. Application of herbicide to the soil columns The herbicide treatments were applied at a rate of 25 kg isoproturon haÿ1 . This rate is approximately 10 times higher than that used to prevent weed infestations and is enough to achieve adequate homogeneity of application over the surface. Both herbicide treatments (technical grade isoproturon, and IAB granules) were applied to duplicate soil columns. Technical grade isoproturon was applied according to the following method. A methanol solution containing 10 mg of isoproturon was added to 50 g (0.5 cm) of washed sand. The washed sand/herbicide mixture was left to dry overnight at room temperature before being added to the top of each column. After application of the herbicide, an additional 0.5 cm of washed sand was added to the top of the column. The alginate-based CR formulation IAB was evenly distributed on the sand layer, after that an additional 1 cm of washed sand was added to the top of the column. 2.4.3. Leaching and leachate collection The leaching solution used in all experiments was 0.01 M CaCl2 . This was done to simulate the soil solution and to prevent dispersion of the soil during the leaching procedure (Johnson and Pepperman, 1995). 1.5 l of solution (the amount of solution equivalent to that applied in greenhouses for one growth season and equivalent to 39 cm irrigation) was leached at a ¯ow rate of 8 ml hÿ1 (4.9 cm dayÿ1 ) using a Gilson Minipuls 3 Peristaltic Pump. The time and volume of each leachate fraction were recorded. Aliquots were taken from the leachate, passed through 0.5 lm PTFE ®lters, and injected directly in the HPLC system. At the end of the leaching procedure (8 days), the columns were allowed to drain for 48 h. 2.4.4. Column analysis The columns were split vertically, and the soil was removed following the established division for the different layers of the column, obtaining four fractions corresponding to the native soil, amended soil, peat, and sand. For the native soil and amended soil, a partition was made in two 10-cm portions labelled as (1) and (2) representing the upper and lower fractions, respectively. Each fraction was dried at room temperature and homogenised. Subsamples of soils, sand, and peat were extracted in conical ¯asks, placed in a shaker bath for 24 h with 25 ml of HPLC grade methanol, ®ltered through Whatman No. 42 paper, and analysed by HPLC, as described above. The extraction eciences of isoproturon were 93% for the sand, 88% for the peat and 85% ands 87% for the amended and native soils, respectively.
1498
M. Fern andez-Perez et al. / Chemosphere 41 (2000) 1495±1501
Table 2 Parameter Kd of the Henry equation and correlation coecients Sample Sand Peat Amended soil Native soil a *
Kd (l kgÿ1 )
r a
0.29 0.06 38.19 4.45 0.39 0.08 5.87 0.31
0.986 0.985 0.991 0.999
These values represent the standard errors. Signi®cant at the 0.001 probability level.
IAB granules were removed from the sand, and the isoproturon content was determined by dissolving 20 granules in a 0.03 M tripolyphosphate solution (20 ml), following by extraction into a water:methanol (80:20) mix (300 ml) and analysed by HPLC, as described above. 2.5. Leaching potential simulation The movement of isoproturon in lab columns that simulate the typical arrangement of the soil layers in a plastic greenhouse was modeled using the computer program `Chemical Movement in Layered Soils' (Nofziger and Hornsby, 1986). This program estimates the location of the peak concentration of non-polar organic chemicals as they move through a soil in response to downward movement of water. The model assumes that chemicals move only in the liquid phase in response to soil water movement. The change in depth of the chemical depends upon the quantity of water moving past the chemical, properties of the chemical, and selected properties of the soil (Nofziger and Hornsby, 1986). The inputs related to the characteristics of the four layers of the greenhouse soil and sorption parameters applied in the model are listed in Tables 1 and 2. The water input, as irrigation was taken from that used in normal agronomic practice for tomatoes in Almerõa greenhouses (Canovas-Martinez and Diaz-Alvarez, 1981). The total water applied was 39 cm over a period of 210 days.
Fig. 1. Sorption isotherms of isoproturon on the dierent layers of a greenhouse soil (error bars represent the standard deviation of three replicates): (a) sand, amended soil and native soil (b) peat.
To evaluate the sorption capacities of the dierent layers of the soil columns, the experimental data were analysed using the linear model (1) X Kd C;
3. Results and discussion 3.1. Sorption±desorption studies Fig. 1 shows the sorption isotherms of isoproturon on the dierent sorbents: sand, amended soil and native soil (Fig. 1(a)), and peat (Fig. 1 (b)). According to the slope of the initial portion of the curves, these isotherms may be classi®ed in general as C-type of the Giles classi®cation (Giles et al., 1960), which suggests a constant partition of the herbicide isoproturon between solution and the solid phase. The isotherm with the steepest initial gradient is that of peat: the sorption capacity of peat for isoproturon is higher than for the other soils.
1
where X is the pesticide adsorbed per kg of sorbent, (mg kgÿ1 ), Kd a parameter related to the solute partitioning between the sorbent and solvent (l kgÿ1 ), and C is the equilibrium solution concentration, (mg lÿ1 ). The Kd values are shown in Table 2. As can be seen from this table, the Kd values vary from 0.29 l kgÿ1 for the sand to 38.19 l kgÿ1 for the peat in the order sand < amended soil < native soil peat: Regression analysis showed that Kd values were correlated with the organic matter content and the cationic exchange capacity, the correlation coecients being higher than 0.99 in both cases.
M. Fern andez-Perez et al. / Chemosphere 41 (2000) 1495±1501
1499
Fig. 2. Cumulative isoproturon desorbed from the dierent layers of a greenhouse soil (error bars represent the standard deviation of three replicates).
The accumulated percentage of desorbed isoproturon from the dierent sorbents, after four desorption cycles, is presented in Fig. 2. These values range between 62.0% for the peat sample and 93.4% for the sand. For these experiments, the variation order is inversely related to that of the Kd values. 3.2. Layered soil column experiment Fig. 3 shows the isoproturon concentrations present in the soil layers and leachate. Slight dierences were observed in the distribution patterns of isoproturon between the technical grade product (Fig. 3(a)) and IAB granules (Fig. 3(b)) for the sand, peat and amended soil layers, although large dierences were found for the deeper layers of the columns. So, the amount of isoproturon found in the native soil (1) and native soil (2) layers was much less in the soil column where isoproturon was applied as IAB granules (1.62 and 0 mg kgÿ1 ) than that obtained in the column treated with technical grade isoproturon (8.73 and 6.51 mg kgÿ1 ). The total percentage of isoproturon recovered in the soil column, leachate and granules referred to the total amount of herbicide applied in the experiments are indicated in Table 3. As can be seen, 73.10% of the total amount was recovered from the soil column when isoproturon was applied as technical grade product, and 0.90% was present in the leachate. On the other hand, when the herbicide was applied as the IAB alginatebased formulation only 14.43% of the total amount was
Fig. 3. Concentration of isoproturon in soil layers of column experiments (error bars represent the standard deviation of two replicates). (a) Technical grade isoproturon (b) IAB formulation.
recovered from the soil column, and no isoproturon was found in the leachate. When the alginate-based formulation IAB was used, the remaining percentage of isoproturon in granules was 59.40%, that is, during the 8 days that experiment lasted 40.60% of isoproturon was released. If we compare the percentage of isoproturon released (40.60%) with that obtained in a previous paper (Villatranca-S anchez et al., in press), where the time necessary for the release of the same percentage of isoproturon from granules in water under static conditions was 4.08 days, we can conclude that the release rate of isoproturon from IAB granules in the soil is two times slower than that observed in water. This fact might be
Table 3 Isoproturon recovered from the granules, soil column and leachate Technical grade isoproturon
IAB
Total % recovered in the granules Total % recovered in the soil column Total % recovered in the leachate
± 73.10 0.90
59.40 14.43 ±
Total % recovery
74.00
73.83
1500
M. Fern andez-Perez et al. / Chemosphere 41 (2000) 1495±1501
explained if we consider that an occlusion of the formulation surface by soil particles takes place, as well as a slower diusion within the soil as compared to water (Ali and Wilkins, 1992). Soil water solutes may also retard movement of pesticides into the aqueous phase (Connick et al., 1984; Sharm et al., 1985). The reduced rate of release of active ingredient in soil compared with water, was also observed with imidacloprid±lignin matrix and diuron-alginate±bentonite granules (Fern andezPerez et al., 1998; 1999). The granules recovered from the soil did not show any signs of disintegration or major degradation at the end of the experiment. The integrity of the CR granule is an important feature of non-erodible controlled release formulations, as unpredictable fracture formation and breakage would increase the release surface of the matrix and thus, the amount of active ingredient released. The total recovery of isoproturon (soil column + leachate + granules) at 8 days was 73.83% of the total applied for IAB treatment and 74.00% for the technical grade isoproturon treatment (Table 3). The dierence to 100% of the total isoproturon applied is probably due to degradation metabolism. These percentages of degraded isoproturon are within the range reported in incubation experiments carried out with soils (Pieuchot et al., 1996; Benoit et al., 1999). 3.3. Leaching potential simulation The simulated mobility of the herbicide with depth (Fig. 4), showed that the slope of leaching velocity varies through the dierent layers. The highest value was obtained for the sand layer, indicating a fast movement of the herbicide, and this corresponds to the lowest value of the Kd parameter of this layer (0.29 l
kgÿ1 ). The slowest value of the slope was obtained for the peat layer, and this corresponds to the highest value of Kd parameter (38.19 l kgÿ1 ). Intermediate values were obtained for the amended and native soils, according to their Kd values. As can be seen from Fig. 4, isoproturon reaches a depth below 0.52 m (standard depth that appears in a typical layer arrangement in a greenhouse) at 120 days. Then, the presence of isoproturon below 0.52 m could lead to contamination of groundwater aquifers. The simulated mobility of the herbicide with depth, was compared to the variation of isoproturon concentration in dierent layers and leachate obtained in experimental conditions with the technical product (Fig. 3(a)). As can be seen, the herbicide concentration found in each layer varies in an opposite way as the slope does in the simulated mobility experiment. So, the highest concentration and the lowest slope correspond to the peat layer, retarding the movement of isoproturon. Once isoproturon breaks through peat, it is leached quickly. Thus, both the model and the soil column experiment, showed that isoproturon can reach the depth below 0.52 m. Prevention of pesticide leaching should therefore focus on reduction of pesticide breakthrough from the peat layer. According to the results obtained from the study with the IAB granules, it is possible that CR formulations could be of bene®t in this respect, because they reduce the rate of isoproturon released to the soil.
4. Conclusions The use of alginate±bentonite CR formulations clearly reduces the vertical mobility of isoproturon into the greenhouse soil layer columns in comparison with the technical grade product. The peat shows the highest sorption capacity, and according to both the column experiments and the model simulation, this layer retards isoproturon movement through the soil. Modelling of this process shows that leaching of isoproturon is rapid once it has moved beyond the peat layer. The use of alginate±bentonite CR formulations is an ecient system that will reduce the amount of pesticide that breaks through the peat layer, and thus reduce the risk of groundwater pollution.
Acknowledgements
Fig. 4. Depth of peak isoproturon concentration as function of time (simulated).
We thank Rh^ one-Poulenc Agro for samples of isoproturon. This research was supported by the CICYT Project AMB93-0600.
M. Fern andez-Perez et al. / Chemosphere 41 (2000) 1495±1501
References Ali, M.A., Wilkins, R.M., 1992. Factors in¯uencing release and distribution of carbofuran from granules into soils. Sci. Total Environ. 123/124, 469±480. Benoit, P., Barriuso, E., Vidon, Ph., Real, B., 1999. Isoproturon sorption and degradation in a soil from grassed buer strip. J. Environ.Qual. 28, 121±129. Black, C.A., Evans, D.D., White, J.L., Ensminger, L.E., Clark, F.E., 1982. Methods of soil analysis. In: Agronomy Monograph 9, second ed., ASA-SSSA, Madison, Wi. Canovas-Martinez, F., Dõaz-Alvarez, J.R., 1981. Cultivos sin Suelo. Instituto de Estudios almerienses, Almerõa. Connick Jr, W.J., 1982. Controlled release of the herbicides 2, 4-D and dichlobenil from alginate gels. J. Appl. Polym. Sci. 27, 3341±3348. Connick, W.J., Bradow, J.H., Wells, W., Steward, K.K., Van, T.K., 1984. Preparation and evaluation of controlled release formulations of 2, 6-Dichlorobenzonitrile. J. Agric. Food Chem. 32, 1199±1205. Cotterill, J.V., Wilkins, R.M., 1996. Controlled release of phenylurea herbicides from a lignin matrix: release kinetics and modi®cation with urea. J. Agric. Food Chem. 44, 2908± 2912. Fern andez-Perez, M., Gonz alez-Pradas, E., Ure~ na-Amate, M.D., Wilkins, R.M., Lindup, I., 1998. Controlled release of imidacloprid from a lignin matrix: water release and soil mobility study. J. Agric. Food Chem. 46, 3826±3834. Fernandez-Perez, M., Villafranca-S anchez, M., Gonz alez-Pradas, E., Flores-Cespedes, F., 1999. Controlled release of diuron from an alginate±bentonite formulation: water release kinetics and soil mobility study. J. Agric. Food Chem. 47, 791±798. Fisher, G.C., Clark, L., Ramsay, P.M., 1991. Pesticides in a chalk catchment: inputs and aquatic residues. In: Proceedings of the British Crop Protection Symposium ± Pesticides in Soil and Water: Current Perspectives, pp. 193±200. Fleming, G.F., Simmons, F., Wax, L.M., Wing, R.E., Carr, M.E., 1992. Atrazine movement in soil columns as in¯uenced by starch-encapsulation and acrylic additives. Weed Sci. 40, 465±470. Giles, C.H., Evan, T.H., Nakhwa, S.N., Smith, D., 1960. Studies in adsorption. Part XI. J. Chem. Soc. 111, 3973±3993. Gonzalez-Pradas, E., L opez-Gonz alez, J.D., Del F., C., ReyBueno, Valenzuela-Calahorro, 1983. Estudio de la super®cie y porosidad de bentonitas homoi onicas, I. Super®cie especõ®ca y porosidad. An. Edaf. y Agrobiol. 3/4, 507±522.
1501
Hall, D.G.M., Reeve, M.J., Thomansson, A.J., Wright, V.F., 1977. Water retention porosity and density of ®eld soils. Soil Survey; Technical Monograph 9, Soil Survey of England and Wales. Harpenden, pp. 6±11. Jackson, M.L., 1982. An alisis quõmico de suelos. Editorial Omega, Barcelona. Johnson, A.C., Haria, A.H., Bhardwaj, C.L., V olkner, C., Batchelor, C.H., Walker, A., 1994. Water movement and isoproturon behaviour in a drained heavy clay soil: 2. Persistence and transport. J. Hydrol. 163, 217±231. Johnson, R.M., Pepperman, A.B., 1995. Mobility of atrazine from alginate controlled release formulations. J. Environ. Sci. Health. B 30, 27±47. Nofziger, D.L., Hornsby, A.G., 1986. A microcomputer-based management tool for chemical movement in soil. Appl. Agric. Resear. 1, 50±56. Pepperman, A.B., Kuan, J.W., 1993. Slow release formulations of metribuzin based on alginate±kaolin±linseed oil. J. Control. Release 26, 21±30. Pieuchot, M., Perrin-Ganier, C., Portal, J.M., Schiavon, M., 1996. Study on the mineralization and degradation of isoproturon in three soils. Chemosphere 33, 467±478. Primo-Y ufera, E., Carrasco-Dorrien, J.M., 1981. Quõmica Agricola I, Suelos y fertilizantes. Editorial Alhambra, Barcelona. Schreiber, M.M., Hickman, M.V., Vail, G.D., 1993. Starchencapsulated atrazine: ecacy and transport. J. Environ. Qual. 22, 443±453. Sharm, S.R., Singh, R.P., Saxena, S.K., Ahmed, S.R., 1985. Eect of dierent saline alkaline salts fertilisers and surfactants on the movement of some carbamoyl group containing pesticides in soil. Anal. Lett. 18, 2322±2343. Villafranca-S anchez, M., Gonz alez-Pradas, E., Fern andezPerez, M., Martinez-L opez, F., Flores-Cespedes, F., Ure~ na-Amate, M.D. Controlled release of isoproturon from an alginate±bentonite formulation: water release kinetics and soil mobility. Pesticide Science (in press). Walkley, A., Black, I.A., 1934. An examination of the Degtjare method for determining soil organic matter and a proposed modi®cation of the chromic acid titration method. Soil Sci. 37, 29±38. Weber, J.B., Peeper, T.F., 1977. Herbicide mobility in soils. In: Truelove, B. (Ed.), Research Methods in Weed Science. Southern Weed Science Society, Auburn, Alabama. Wilkins, R.M., 1990. Controlled Delivery of Crop-Protection Agents. Taylor and Francis, Bristol, PA, p. 320.