This article was downloaded by: [University of Birmingham] On: 21 April 2015, At: 10:22 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Environmental Science and Health, Part B: Pesticides, Food Contaminants, and Agricultural Wastes Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/lesb20
Sorption, desorption and leaching potential of sulfonylurea herbicides in Argentinean soils a
a
b
Mariela P. Azcarate , Jorgelina C. Montoya & William C. Koskinen a
National Institute of Agricultural Technology (INTA), Anguil Agricultural Experiment Station, Anguil, La Pampa, Argentina b
Soil and Water Management Research Unit, Agricultural Research Service, U.S. Department of Agriculture, St. Paul, Minnesota, USA Published online: 25 Feb 2015.
Click for updates To cite this article: Mariela P. Azcarate, Jorgelina C. Montoya & William C. Koskinen (2015) Sorption, desorption and leaching potential of sulfonylurea herbicides in Argentinean soils, Journal of Environmental Science and Health, Part B: Pesticides, Food Contaminants, and Agricultural Wastes, 50:4, 229-237, DOI: 10.1080/03601234.2015.999583 To link to this article: http://dx.doi.org/10.1080/03601234.2015.999583
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Journal of Environmental Science and Health, Part B (2015) 50, 229–237 Copyright © Taylor & Francis Group, LLC ISSN: 0360-1234 (Print); 1532-4109 (Online) DOI: 10.1080/03601234.2015.999583
Sorption, desorption and leaching potential of sulfonylurea herbicides in Argentinean soils MARIELA P. AZCARATE1, JORGELINA C. MONTOYA1 and WILLIAM C. KOSKINEN2 1
National Institute of Agricultural Technology (INTA), Anguil Agricultural Experiment Station, Anguil, La Pampa, Argentina Soil and Water Management Research Unit, Agricultural Research Service, U.S. Department of Agriculture, St. Paul, Minnesota, USA
Downloaded by [University of Birmingham] at 10:22 21 April 2015
2
The sulfonylurea (SUs) herbicides are used to control broadleaf weeds and some grasses in a variety of crops. They have become popular because of their low application rates, low mammalian toxicity and an outstanding herbicidal activity. Sorption is a major process influencing the fate of pesticides in soil. The objective of this study was to characterize sorption–desorption of four sulfonylurea herbicides: metsulfuron-methyl (methyl 2-[(4-methoxy-6-methyl-1,3,5-triazin-2-yl)carbamoylsulfamoyl)] benzoate), sulfometuron-methyl (methyl 2-[(4,6-dimethylpyrimidin-2-yl)carbamoylsulfamoyl]benzoate), rimsulfuron (1-(4,6dimethoxypyrimidin-2-yl)-3-(3-ethylsulfonyl-2-pyridylsulfonyl)urea) and nicosulfuron (2-[(4,6-dimethoxypyrimidin-2-yl) carbamoylsulfamoyl]-N,N-dimethylnicotinamide) from different soil horizons of different landscape positions. Sorption was studied in the laboratory by batch equilibration method. Sorption coefficients (Kd-SE) showed that rimsulfuron (Kd-SE D 1.18 to 2.08 L kg¡1) and nicosulfuron (Kd-SE D 0.02 to 0.47 L kg¡1) were more highly sorbed than metsulfuron-methyl (Kd-SE D 0.00 to 0.05 L kg¡1) and sulfometuron-methyl (Kd-SE D 0.00 to 0.05 L kg¡1). Sorption coefficients (Kd-SE) were correlated with pH and organic carbon content. All four herbicides exhibited desorption hysteresis where the desorption coefficients (Kd-D) > Kd-SE. To estimate the leaching potential, Koc and ground-water ubiquity score (GUS) were used to calculate the half-life (t1/2) required to be classified as “leacher” or “nonleacher”. According to the results, rimsulfuron and nicosulfuron herbicides would be classified as leachers, but factors such as landscape position, soil depth and the rate of decomposition in surface and subsurface soils could change the classification. In contrast, these factors do not affect classification of sulfometuron-methyl and metsulfuron-methyl; they would rank as leachers. Keywords: Sorption, sulfonylurea herbicides, leaching potential, Argentinean soils, hysteresis.
Introduction Sulfonylurea (SUs) herbicides were discovered by DuPont Crop Protection in 1975 and first commercialized for wheat and barley crops in 1982.[1] They are widely used for control of annual and perennial broad-leaved weeds in many crops (i.e., wheat, barley, soy-bean, maize, oats and rice). They are the most important acetolactate synthase (ALS) inhibitors. ALS is a key enzyme in the biosynthesis of essential amino acids valine, leucine and isoleucine. SUs herbicides have been considered environmentally safe, in part because of their low application rates (4–50 g ha¡1).[2,3] However, there are several environmental concerns. Even at low rates, these herbicides can persist in the Address correspondence to Mariela P. Azcarate, National Institute of Agricultural Technology (INTA), Anguil Agricultural Experiment Station, Ruta Nacional No. 5, Km 580. C.C. 11 (6326) Anguil, La Pampa, Argentina; E-mail: azcarate. [email protected] Received June 25, 2014.
soil throughout more than one growing season and may injure rotational crops.[4] They also have a potential for off-site transport. Therefore, to better assess their impact on the environment, it is important to characterize their mobility in soil. The mobility and leaching of herbicides through the soil profile are influenced by the amount of water available for downward or lateral movement, the chemical nature of the herbicide, the persistence of the chemical and the sorptive capacity of the soil. The parameters most often used as indicators of sorption capacity of soil and pesticide mobility are the sorption–desorption coefficients, Kd or Kf and 1/n from the Freundlich isotherms. Kd or Kf is the magnitude of sorption; the greater the Kd or Kf value the greater the sorption. The slope of the isotherm (1/n) indicates whether sorption is independent of concentration (1/n D 1) or dependent on concentration (1/n < 1). For instance, Vicari et al.[5] found that rimsulfuron was adsorbed in amounts 3.7 to 7 times higher than primisulfuron, with Kd values ranging from 0.1 to 1.18 for primisulfuron and from 0.71 to
Downloaded by [University of Birmingham] at 10:22 21 April 2015
230 5.1 for rimsulfuron in six Colorado soils. Wu et al.[6] found Kf values for methiopyrsulfuron ranged from 0.75 to 2.46 L kg¡1. The isotherms were non-linear with 1/n < 1, indicating that sorption was greater for the lower initial solution concentrations. Sorption of SUs is affected by chemical nature of the SU and a number of soil properties. SUs herbicides are weak acids and exist predominantly in the anionic form at pH values greater than the pKa (3–5).[2,3] As the soil pH decreases, the proportion of the neutral form of the weak acid increases, which is sorbed to a greater extent in soils as compared to the anion form. Organic carbon content (OC) has been shown to influence the sorption of many pesticides, including SUs.[6–9] Mobility in soil is also affected by the ability of the sorbed herbicides to desorb from the soil. Various herbicides have exhibited desorption hysteresis, that is, once a herbicide is sorbed it is not readily desorbed.[6,10–12] Desorption was hysteretic on the soils with high organic matter content and low pH.[6,12] As a result of their mobility, the occurrence of SUs in surface and groundwater resources has been reported. Cessna et al.[13] determined leaching of thifensulfuronmethyl, tribenuron-methyl and rimsulfuron during sprinkler irrigation and found that thifensulfuron-methyl was the only one detected in groundwater, with concentrations ranging from 1.2 to 2.5 ng L¡1. Struger et al.[14] measured the occurrence of sulfonylurea and related herbicides in central Canadian surface waters for 3 years. In 2006, concentrations of >100 ng L¡1 were also observed for nicosulfuron, chlorimuron-ethyl, primisulfuron-methyl and rimsulfuron. Nicosulfuron was also detected in >20% of the total sample set for 2006, and chlorimuron-ethyl was detected in roughly 17% of the total sample set for 2008 in Ontario. Battaglin et al.[15] reported concentrations of SUs and other herbicides in 25 samples of Midwestern (USA) in groundwater. For instance, nicosulfuron was found in 8% of the samples, whereas flumetsulam was detected in 12% of the samples. The objective of this study was to characterize sorption– desorption of four SUs herbicides in soils from different soil horizons of different landscape positions in an Argentinean soil. The data obtained were used to determine potential risk of leaching to groundwater.
Azcarate et al.
Fig. 1. Chemistry structure for sulfonylurea herbicides: metsulfuron-methyl, sulfometuron-methyl, rimsulfuron and nicosulfuron.
(specific activity 62.2 mCi mg¡1; 98.8% pure), rimsulfuron (specific activity 51.8 mCi mg¡1; 98.9% pure) and sulfometuron-methyl (specific activity 48.9 mCi mg¡1; 99.0% pure) were kindly provided by DuPont Agricultural Products. All other chemical used were analytical grade reagents. Non-labeled herbicide solutions were prepared at different concentration (0.006; 0.02; 0.06; 0.2 mg L¡1) in 0.005 M CaCl2 solution. Radiolabeled chemical was added to nonradioactive solutions to give solution radioactivity of metsulfuron-methyl (229 kBq L¡1), nicosulfuron (199 kBq L¡1), rimsulfuron (160 kBq L¡1) and sulfometuronmethyl (34 kBq L¡1).
Soils The soils were collected from an agricultural field in Intendente Alvear, La Pampa, Argentina (35.40 S, 63.69 W), from four depth horizons (Ap, A12, AC and C) at three landscape positions: shoulder (SH), middle (MI) and footslope (FS). Samples were air-dried and passed through a 2 mm sieve. Subsamples of homogenized soils were
Materials and methods Table 1. Some properties of sulfonylureas herbicides studied.[2,9]
Chemicals The SUs herbicides used in this study were rimsulfuron, metsulfuron-methyl, nicosulfuron and sulfometuronmethyl. Some properties of these herbicides are listed in Table 1 and the chemical structures are in Figure 1. Pure analytical standards of 14C-labeled metsulfuron-methyl (specific activity 38.3 mCi mg¡1; 99.0% pure), nicosulfuron
Herbicides
pKa
Field Half-life (days)
Metsulfuron-methyl Sulfometuron-methyl Rimsulfuron Nicosulfuron
3.3 5.2 4.0 4.6
4–71 7–150 5–23 14–49
231
Sorption, desorption and leaching potential of sulfonylurea herbicides in Argentinean soils Table 2. Physical and chemical properties of the tested soils. Soil Depth (cm) Horizon SH
MI
FS
Downloaded by [University of Birmingham] at 10:22 21 April 2015
a
0–7 7–25 25–55 >55 0–7 7–23 23–40 >40 0–13 13–20 20–60 >60
Ap A12 AC >C Ap A12 AC >C Ap A12 AC >C
pH 5.75 § 6.10 § 6.41 § 6.61 § 6.00 § 6.12 § 6.39 § 6.64 § 7.05 § 7.81 § 8.57 § 9.14 §
0.12a 0.03 0.03 0.05 0.14 0.03 0.08 0.09 0.34 0.02 0.11 0.04
OC (g kg¡1) CEC (cmolc kg¡1) Clay (%) 4.6 § 3.0 § 2.1 § 0.5 § 9.6 § 3.1 § 2.6 § 1.0 § 11.3 § 3.7 § 0.6 § 0.2 §
1.1 0.3 0.1 0.5 1.5 0.4 0.2 0.7 1.4 1.7 0.4 0.1
5.05 § 7.26 § 6.65 § 6.05 § 8.45 § 10.31 § 9.25 § 8.17 § 11.25 § 6.98 § 9.56 § 8.00 §
0.65 0.59 2.12 1.58 0.77 1.43 1.08 1.43 0.51 1.75 1.49 1.97
2.3 § 1.6 § 1.6 § 1.6 § 1.6 § 3.6 § 2.6 § 1.6 § 3.6 § 3.3 § 4.1 § 2.3 §
1.2 0.0 0.0 0.0 0.0 1.0 1.0 0.0 0.0 0.6 0.7 1.2
Silt (%) 7.0 § 7.0 § 5.0 § 3.7 § 17.7 § 17.4 § 17.0 § 17.4 § 23.9 § 17.4 § 16.9 § 14.7 §
2.1 1.2 1.2 1.2 2.3 0.0 0.6 1.7 2.1 1.0 0.7 0.6
Sand (%)
Soil Texture
90.7 § 91.3 § 93.3 § 94.7 § 80.7 § 79.0 § 80.3 § 81.0 § 72.5 § 79.3 § 79.0 § 83.0 §
Sand Sand Sand Sand Loamy sand Loamy sand Loamy sand Loamy sand Loamy sand Loamy sand Loamy sand Loamy sand
2.1 1.2 1.2 1.2 2.3 1.0 0.6 1.7 2.1 1.2 1.4 1.0
§ Values are standard deviations for three replicates.
analyzed for texture, pH, OC and cation exchange capacity (CEC) (Table 2). Texture analysis was performed determined by the hydrometer method.[16] OC was determined by chromic acid oxidation.[17] The pH was measured with a glass electrode in a 1:1 soil/water suspension. CEC was determined by the method of ammonium acetate.[18]
equilibration. Kd ¡ SE D Cs =Cw
(1)
The coefficient of adsorption per unit OC (Koc) was calculated for adsorption by normalizing Kd-SE values to the percentage contents of OC according to Eq. (2): Koc D 100Kd ¡ SE =OC
(2)
Sorption–desorption experiment Sorption studies on soils were conducted by the batch equilibration method. Duplicate 8 g soil samples were equilibrated with 8 mL of 14C-labeled herbicides solutions (0.006, 0.02, 0.06 and 0.2 mg L¡1) using shaking mechanically at 20 C for 24 h. The highest concentration represents the usual rate used in the field diluted in the top 5 cm of soil. After equilibration, suspensions were centrifuged at 2,500 rpm for 10 min. Two mL of supernatant was removed for analysis, after which 1 mL aliquot was used to determine 14C by a liquid scintillation counter. Desorption was determined at the highest and lowest concentrations of the herbicide solutions (0.006 and 0.2 mg L¡1) immediately after the sorption experiments. The 2 mL aliquot of supernatant that was removed was replaced with the same volume of 0.005 M CaCl2 and the samples reequilibrated. Samples were shaken for 24 h and centrifuged. A total of three desorption equilibrations were carried out. The mass of chemical sorbed to soil was calculated from the difference between initial and final solution concentrations. Cs is the concentration of herbicide sorbed to soil (mg kg¡1) and Cw is the concentration of herbicide remaining in solution (mg L¡1) after equilibration. Sorption coefficients (Kd-SE) were determined for each soil and each concentration using Eq. (1). Desorption coefficients (Kd-D1, 2, 3) were also calculated after each desorption
Leaching potential To predict the leaching potential of the SUs herbicides, the Groundwater Ubiquity Score (GUS) was used (Eq. (3)). GUS D log t1=2 .4 ¡ log Koc /;
(3)
where t1/2 is herbicides half-life in soil. Herbicides with GUS index values 2.8 are ranked as leachers. t1/2 values were calculated using Koc values from the studies soils and appropriate GUS values (GUS D 1.8 and 2.8).[19] Statistical analysis Replicated soil sorption data were analyzed by Pearson correlation. The Pearson correlation coefficients (r) between sorption coefficients (Kd-SE) and soil properties (pH, OC and clay) were estimated for each herbicide. To determine the influence of landscape position and depth on the measured soil properties and Kd-SE values analysis of variance (ANOVA) was used. Equilibration steps (SE, D1, D2, D3) and sorption–desorption Kd were analyzed by linear and quadratic regression. The determination coefficient (r2) was used to check the best fit. All data analysis was run by InfoStat version 2013 for Windows.[20]
232 Results and discussion Sorption
pyridinecarboxamide SUs. Sorption Kd-SE values of rimsulfuron on the soils from the four horizons from the three landscape positions (SH, MI and FS) ranged from 1.18 to 2.08 L kg¡1. Nicosulfuron Kd-SE values were in the range from 0.02 to 0.47 L kg¡1. In contrast, metsulfuron-methyl and sulfometuron-methyl sorption Kd-SE values were very low (0.00–0.05 L kg¡1 and 0.02–0.05 L kg¡1, respectively). The sorption coefficients found in this work were lower than in the limited published studies. Oliveira et al.[21] found sorption Kd values ranging from 0.14 to 1.38 for nicosulfuron, 0.14 to 1.18 for sulfometuron-methyl, and 0.09 to 1.27 for metsulfuron-methyl in Brazilian soils. Gonzalez
Downloaded by [University of Birmingham] at 10:22 21 April 2015
There was no effect of initial solution concentration (0.006 to 0.2 mg L¡1) on sorption of the SUs; slope of the Freundlich equation »1.0. Therefore, the mean of the four Kd-SE values for each SU was calculated and the resultant mean Kd-SE values are shown in Figures 2 and 3. Overall, the sorption decreased in the order of rimsulfuron > nicosulfuron > metsulfuron-methyl D sulfometuron-methyl. The two least sorbed SUs were benzoate molecules, as opposed the more highly sorbed pyridinesulfonamide and
Azcarate et al.
Fig. 2. Rimsulfuron and nicosulfuron sorption (Kd-SE) and desorption (Kd-D) coefficients for SH, MI and FS soils at different depths.
233
Downloaded by [University of Birmingham] at 10:22 21 April 2015
Sorption, desorption and leaching potential of sulfonylurea herbicides in Argentinean soils
Fig. 3. Metsulfuron-methyl and sulfometuron-methyl sorption (Kd-SE) and desorption (Kd-D) coefficients for SH, MI and FS soils at different depths.
and Ukrainczyk[22] obtained Kf values for nicosulfuron in a range from 0.21 to 8.78. Walker et al.[23] found metsulfuron-methyl Kd values between 0.046 and 0.542, and decreased sorption with depth. Koskinen et al.[24] found for sulfometuron-methyl Kd range of 0.04–0.6 (0–20 cm depth) and sorption from depths 65–95 cm was less than in the surface soils. There is a close relationship between landscape position, depth and sorption. Rimsulfuron and nicosulfuron showed Kd-SE values £ landscape position £ depth significant interactions (P < 0.0001). Kd-SE values for rimsulfuron were the highest in FS (0–13 cm) and the lowest in SH (>55 cm).
Nicosulfuron showed the highest Kd-SE values in MI (7–23 cm) whereas the lowest were in FS (>60 cm). Metsulfuron-methyl showed significant correlation with landscape position (SH D MI > FS; P < 0.0001) and depth (Ap horizon > C horizon; P < 0.01). In contrast, no trend was observed for sulfometuron-methyl (Figs. 2 and 3). With regard to physicochemical properties were observed significant interactions among landscape position and depth: pH and OC (P < 0.0001), sand (P < 0.0001), silt (P < 0.001), CEC (P < 0.05) and clay (P < 0.05) (Table 2). OC decreased with depth (Ap, A12, AC, >C) while pH increased with depth, pH was higher in FS than MI and SH. OC
Downloaded by [University of Birmingham] at 10:22 21 April 2015
234 decreased in the order to FS > MI > SH. In relation with the texture, sand content was highest in SH at >55 cm and the lowest value was in FS in surface (0–13 cm). Silt content in Ap horizons decreased in following order FS > MI > SH whereas subsurface horizons (A12, AC and >C) decreased in order to FS D MI > SH. Clay content was higher in FS than MI and SH but decreased in depth. Soil physicochemical properties in the field can vary and can explain the measured variation in herbicides sorption within fields for different herbicides.[25–27] The soil properties that have shown the greatest effect on herbicide sorption are OC content, pH and clay content. Correlation analyses between Kd-SE, OC, soil pH and clay content showed that in SH profile there was a positive correlation between OC and sorption of rimsulfuron (r D 0.94; P < 0.05) and nicosulfuron (r D 0.98; P < 0.05), and a negative correlation with pH (rimsulfuron, r D ¡0.99; P < 0.05, nicosulfuron, r D ¡0.97; P < 0.05). In MI and FS soils, there was an increase in sorption with an increase in OC and a decrease in pH, however the correlations were not significant at P < 0.05. Due to the very low sorption and standard errors in Kd-SE values, there were no correlations between OC or pH and sorption of metsulfuronmethyl and sulfometuron-methyl for soils from the three landscape positions. Other studies have shown significant correlations between OC and sorption of SUs, i.e., metsulfuron-methyl (r D 0.941; P < 0.01) and sulfometuron-methyl (r D 0.835; P < 0.05),[21] while other authors did not find correlations for metsulfuron-methyl, sulfometuron-methyl[24,28] and rimsulfuron.[5] The contradiction is these results may be due to differences in pH levels of the soils used in the studies that may minimize the effects of OC. SUs are weak acids that exist predominantly in the anionic form when the soil pH values (Table 2) are higher than the pKa (Table 1). The neutral form is much more strongly sorbed in soils than the anion as reported for different SU herbicides sulfonylureas.[6,7,12,23,29,30] In contrast to the present work, Vicari et al.[5] found rimsulfuron sorption was not correlated with the pH (r D ¡0.37; P > 0.05) and suggested that it was probably because the soil pH (5.6–7.8) was above the pKa (4.0), and the proportion of neutral molecule was little or none. Walker et al.[23] found negative correlation with pH for metsulfuron-methyl, while other authors did not find correlations or they were not strongly correlated.[21,24,28,29] The pH range studied may sometimes be too narrow or too wide to underscore any influence of pH.[7,11] The difference between the pH at the surface of soil particles and in the soil solution might also differ according to the measurement technique used and the characteristics of the soil.[7] However, the present study shows that soil pH has a strong influence on sorption of some SUs herbicides. Although SH soil was the site of lowest pH and had the lowest Kd-SE for rimsulfuron and nicosulfuron, simultaneously this profile had the lowest OC content and coarsest
Azcarate et al. texture provided scarce sorption sites. The pH of soil can vary greatly with depth but generally increases with depth in the profile because the surface is more weathered[31] or like these soils studied where the FS profile showed the highest pH values because of the upward movement of saline groundwater. For SUs herbicides, Kd values decreased with depth as has been previously noted.[23] Sorption was not correlated with clay content for all soils tested. Similar results were reported by Oliveira et al.[21] and Cranmer et al.,[28] whereas Vicari et al.[5] found a correlation with rimsulfuron (r D 0.88; P < 0.05). Desorption All four herbicides exhibited desorption hysteresis for all SU/soil horizon/landscape position-depth combinations; the Kd desorption values (Kd-D1, 2, 3) increased with each consecutive desorption equilibration step (1/ndesorp < 1/ nsorption), even in soil samples where sorption was very low as with metsulfuron-methyl and sulfometuron-methyl (Figs. 2 and 3). For the two more extensively sorbed SUs, rimsulfuron and nicosulfuron, the increase in Kd-D with each desorption step was linear (Fig. 4). In contrast, for the more weakly sorbed metsulfuron-methyl and sulfometuron-methyl, the increase was curvilinear, with a greater increase in Kd-D with each desorption equilibration (Fig. 4). The observation that these weak acid herbicides exhibited hysteresis is surprising, since are often only slightly adsorbed by soils.[32] However, other authors also have reported this phenomenon for different SUs.[6,12,29] The specific nature of the processes responsible for the observed behavior remains unknown.[11,32,33] It could be an experimental artifact, a diffusion to inaccessible or stronger binding sites, change to stronger binding mechanism, higher degradation rate than the rate of desorption (particularly if the SUs’ half-lives are in the lower range of values reported in Table 1) or a combination of factors.[11] The desorption process is very important in the prediction of pesticide movement in soils.[31] Most transport models just use Kd-SE to predict depth of leaching and Kd-SE is inversely proportional to depth of leaching. These models do not incorporate hysteresis into the potential depth of leaching. If we used the desorption Kd-D3 value (Kd calculated after the third desorption step), as compared to the adsorption Kd-SE, the depth of leaching would be decreased by a factor 2-3X for rimsulfuron and nicosulfuron, and 4-15X for metsulfuron-methyl and sulfometuron-methyl.
Leaching potential Many researchers have proposed indices to predict the mobility of pesticides in the field. The most used index is the Groundwater Ubiquity Score (GUS), which was
235
Downloaded by [University of Birmingham] at 10:22 21 April 2015
Sorption, desorption and leaching potential of sulfonylurea herbicides in Argentinean soils
Fig. 4. Rimsulfuron and metsulfuron-methyl sorption (Kd-SE) and desorption (Kd-D) coefficients as a function of equilibration step.
developed by Gustafson[19] and is based on the values of Koc and half-life (t1/2). These important parameters can determine the mobility and persistence of the herbicides.[19,34] Using Koc values and appropriate GUS values, t1/2 values were calculated that would rank these herbicides as leachers or nonleachers (Table 3). In the Ap horizon, the Koc decreased in the order SH > MI > FS, which in turn decreased the t1/2 value below which the SU would be considered a nonleacher and above which it would be considered as leacher (Table 3). For instance, rimsulfuron would be considered a nonleacher for t1/2 values 78 days in SH soils and 41 days in FS soils. Smaller decreases in t1/2 values were also observed for the other three SUs. At each of the three landscape positions, rimsulfuron and nicosulfuron Koc values generally increased with soil depth thereby increasing the t1/2 value below which the SU would be considered a nonleacher and above which it would be considered a leacher (Table 3). For instance, for rimsulfuron to be considered a nonleacher, t1/2 would have to 184 days in the AC horizon soil.
236
Azcarate et al.
Table 3. Estimated half-lives (t1/2) for rimsulfuron, nicosulfuron, metsulfuron-methyl and sulfometuron-methyl that would classify the herbicide as a “leacher” or “nonleacher”. Rimsulfuron Site Horizon Koc (L kg¡1) t1/2 (days) SH
MI
Downloaded by [University of Birmingham] at 10:22 21 April 2015
FS
Ap A12 AC >C Ap A12 AC >C Ap A12 AC >C
333 456 581 2360 212 588 631 1308 184 457 2864 8210
Nicosulfuron
Metsulfuron-methyl
Sulfometuron-methyl
Koc (L kg¡1) t1/2 (days) Koc (L kg¡1) t1/2 (days) Koc (L kg¡1) t1/2 (days)
17a/78b 22/122 29/184 2 years/80 years 12/47 29/189 32/215 0.3 years/4 years 11/41 22/123 6 years/392 years >500 years
72 99 116 360 40 150 142 272 35 90 265 93
7/20 8/25 9/28 18/87 6/15 10/34 9/33 14/61 5/14 8/23 14/60 8/24
11 10 10 40 5 10 15 20 4 3 0 0
4/9 4/9 4/8 6/15 4/7 4/8 4/10 5/11 3/6 3/6 — —
7 10 15 40 3 13 8 30 4 8 66 93
4/8 4/9 4/10 6/15 3/6 4/9 4/8 5/13 3/7 4/8 7/19 8/24
t1/2 below which SUs would be considered a “nonleacher”. t1/2 above which would be considered as a “leacher”.
a
b
Reported t1/2 values for rimsulfuron, nicosulfuron, metsulfuron-methyl and sulfometuron-methyl were 5–23, 14– 49, 4–71 and 7–150 days, respectively.[2,9,35] Using the midpoint of these values of t1/2 for rimsulfuron (14 days), it would be considered a nonleacher in all soils from the three landscape position and from all soil horizons, which would indicate a low potential for leaching to groundwater. Rimsulfuron is very unstable, particularly at pH values above 8, where it is rapidly hydrolyzed. This is the reason why it has not been found in groundwater.[36] However, degradation products were detected in the water and groundwater study zones for several years following application of rimsulfuron. Thus, they are relatively stable and persist in the soil, from where they leach to the groundwater.[37] Using the midpoint of these values of t1/2 for nicosulfuron (31 days), it would be considered a leacher in the upper three horizons of the SH soil, the Ap horizon in the MI soil and in the Ap, A12, and lowest horizon in the FS soil. However, reported strong adsorption of nicosulfuron on clay minerals should decrease its mobility in soils and its potential to leach into groundwater. Sorption on mineral surfaces is especially important in the subsurface horizons and in soils with low organic matter content. The fast kinetics of the adsorption implies that the nicosulfuron moving down the soil profile should be readily and strongly retained on the surfaces of clay minerals.[38] The midpoints of t1/2 for metsulfuron-methyl and sulfometuron-methyl classify both of them as leachers. Hollaway et al.[39] found that the persistence of metsulfuronmethyl in alkaline soils is less than 1 year after herbicide application. It has been shown to range in mobility in soil from low mobility to relatively mobile.[40] Based on Koc values for a number of Brazilian soils, Oliveira et al.[21] calculated that a t1/2 value >16.5 days would also classify
metsulfuron-methyl and sulfometuron-methyl as leachers in all the Oxisol soils studied. In summary, knowledge of the fate and transport of herbicides in agricultural soils is of great importance to agronomic systems and to environmental protection. According to the results, rimsulfuron and nicosulfuron would rank as leachers depending on the landscape position and soil depth the samples were taken from. If rimsulfuron and nicosulfuron leach to groundwater, that would be dependent on the rate of decomposition in both surface and subsurface soils.[24] In contrast, metsulfuron-methyl and sulfometuron-methyl would be classified as leachers, regardless of landscape position and soil depth. It is also important to incorporate hysteresis when predicting potential mobility. Use of sorption Kd values may overpredict transport, when there is desorption hysteresis and sorption is concentration dependent (Freundlich 1/n values
Journal of Environmental Science and Health, Part B: Pesticides, Food Contaminants, and Agricultural Wastes Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/lesb20
Sorption, desorption and leaching potential of sulfonylurea herbicides in Argentinean soils a
a
b
Mariela P. Azcarate , Jorgelina C. Montoya & William C. Koskinen a
National Institute of Agricultural Technology (INTA), Anguil Agricultural Experiment Station, Anguil, La Pampa, Argentina b
Soil and Water Management Research Unit, Agricultural Research Service, U.S. Department of Agriculture, St. Paul, Minnesota, USA Published online: 25 Feb 2015.
Click for updates To cite this article: Mariela P. Azcarate, Jorgelina C. Montoya & William C. Koskinen (2015) Sorption, desorption and leaching potential of sulfonylurea herbicides in Argentinean soils, Journal of Environmental Science and Health, Part B: Pesticides, Food Contaminants, and Agricultural Wastes, 50:4, 229-237, DOI: 10.1080/03601234.2015.999583 To link to this article: http://dx.doi.org/10.1080/03601234.2015.999583
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Journal of Environmental Science and Health, Part B (2015) 50, 229–237 Copyright © Taylor & Francis Group, LLC ISSN: 0360-1234 (Print); 1532-4109 (Online) DOI: 10.1080/03601234.2015.999583
Sorption, desorption and leaching potential of sulfonylurea herbicides in Argentinean soils MARIELA P. AZCARATE1, JORGELINA C. MONTOYA1 and WILLIAM C. KOSKINEN2 1
National Institute of Agricultural Technology (INTA), Anguil Agricultural Experiment Station, Anguil, La Pampa, Argentina Soil and Water Management Research Unit, Agricultural Research Service, U.S. Department of Agriculture, St. Paul, Minnesota, USA
Downloaded by [University of Birmingham] at 10:22 21 April 2015
2
The sulfonylurea (SUs) herbicides are used to control broadleaf weeds and some grasses in a variety of crops. They have become popular because of their low application rates, low mammalian toxicity and an outstanding herbicidal activity. Sorption is a major process influencing the fate of pesticides in soil. The objective of this study was to characterize sorption–desorption of four sulfonylurea herbicides: metsulfuron-methyl (methyl 2-[(4-methoxy-6-methyl-1,3,5-triazin-2-yl)carbamoylsulfamoyl)] benzoate), sulfometuron-methyl (methyl 2-[(4,6-dimethylpyrimidin-2-yl)carbamoylsulfamoyl]benzoate), rimsulfuron (1-(4,6dimethoxypyrimidin-2-yl)-3-(3-ethylsulfonyl-2-pyridylsulfonyl)urea) and nicosulfuron (2-[(4,6-dimethoxypyrimidin-2-yl) carbamoylsulfamoyl]-N,N-dimethylnicotinamide) from different soil horizons of different landscape positions. Sorption was studied in the laboratory by batch equilibration method. Sorption coefficients (Kd-SE) showed that rimsulfuron (Kd-SE D 1.18 to 2.08 L kg¡1) and nicosulfuron (Kd-SE D 0.02 to 0.47 L kg¡1) were more highly sorbed than metsulfuron-methyl (Kd-SE D 0.00 to 0.05 L kg¡1) and sulfometuron-methyl (Kd-SE D 0.00 to 0.05 L kg¡1). Sorption coefficients (Kd-SE) were correlated with pH and organic carbon content. All four herbicides exhibited desorption hysteresis where the desorption coefficients (Kd-D) > Kd-SE. To estimate the leaching potential, Koc and ground-water ubiquity score (GUS) were used to calculate the half-life (t1/2) required to be classified as “leacher” or “nonleacher”. According to the results, rimsulfuron and nicosulfuron herbicides would be classified as leachers, but factors such as landscape position, soil depth and the rate of decomposition in surface and subsurface soils could change the classification. In contrast, these factors do not affect classification of sulfometuron-methyl and metsulfuron-methyl; they would rank as leachers. Keywords: Sorption, sulfonylurea herbicides, leaching potential, Argentinean soils, hysteresis.
Introduction Sulfonylurea (SUs) herbicides were discovered by DuPont Crop Protection in 1975 and first commercialized for wheat and barley crops in 1982.[1] They are widely used for control of annual and perennial broad-leaved weeds in many crops (i.e., wheat, barley, soy-bean, maize, oats and rice). They are the most important acetolactate synthase (ALS) inhibitors. ALS is a key enzyme in the biosynthesis of essential amino acids valine, leucine and isoleucine. SUs herbicides have been considered environmentally safe, in part because of their low application rates (4–50 g ha¡1).[2,3] However, there are several environmental concerns. Even at low rates, these herbicides can persist in the Address correspondence to Mariela P. Azcarate, National Institute of Agricultural Technology (INTA), Anguil Agricultural Experiment Station, Ruta Nacional No. 5, Km 580. C.C. 11 (6326) Anguil, La Pampa, Argentina; E-mail: azcarate. [email protected] Received June 25, 2014.
soil throughout more than one growing season and may injure rotational crops.[4] They also have a potential for off-site transport. Therefore, to better assess their impact on the environment, it is important to characterize their mobility in soil. The mobility and leaching of herbicides through the soil profile are influenced by the amount of water available for downward or lateral movement, the chemical nature of the herbicide, the persistence of the chemical and the sorptive capacity of the soil. The parameters most often used as indicators of sorption capacity of soil and pesticide mobility are the sorption–desorption coefficients, Kd or Kf and 1/n from the Freundlich isotherms. Kd or Kf is the magnitude of sorption; the greater the Kd or Kf value the greater the sorption. The slope of the isotherm (1/n) indicates whether sorption is independent of concentration (1/n D 1) or dependent on concentration (1/n < 1). For instance, Vicari et al.[5] found that rimsulfuron was adsorbed in amounts 3.7 to 7 times higher than primisulfuron, with Kd values ranging from 0.1 to 1.18 for primisulfuron and from 0.71 to
Downloaded by [University of Birmingham] at 10:22 21 April 2015
230 5.1 for rimsulfuron in six Colorado soils. Wu et al.[6] found Kf values for methiopyrsulfuron ranged from 0.75 to 2.46 L kg¡1. The isotherms were non-linear with 1/n < 1, indicating that sorption was greater for the lower initial solution concentrations. Sorption of SUs is affected by chemical nature of the SU and a number of soil properties. SUs herbicides are weak acids and exist predominantly in the anionic form at pH values greater than the pKa (3–5).[2,3] As the soil pH decreases, the proportion of the neutral form of the weak acid increases, which is sorbed to a greater extent in soils as compared to the anion form. Organic carbon content (OC) has been shown to influence the sorption of many pesticides, including SUs.[6–9] Mobility in soil is also affected by the ability of the sorbed herbicides to desorb from the soil. Various herbicides have exhibited desorption hysteresis, that is, once a herbicide is sorbed it is not readily desorbed.[6,10–12] Desorption was hysteretic on the soils with high organic matter content and low pH.[6,12] As a result of their mobility, the occurrence of SUs in surface and groundwater resources has been reported. Cessna et al.[13] determined leaching of thifensulfuronmethyl, tribenuron-methyl and rimsulfuron during sprinkler irrigation and found that thifensulfuron-methyl was the only one detected in groundwater, with concentrations ranging from 1.2 to 2.5 ng L¡1. Struger et al.[14] measured the occurrence of sulfonylurea and related herbicides in central Canadian surface waters for 3 years. In 2006, concentrations of >100 ng L¡1 were also observed for nicosulfuron, chlorimuron-ethyl, primisulfuron-methyl and rimsulfuron. Nicosulfuron was also detected in >20% of the total sample set for 2006, and chlorimuron-ethyl was detected in roughly 17% of the total sample set for 2008 in Ontario. Battaglin et al.[15] reported concentrations of SUs and other herbicides in 25 samples of Midwestern (USA) in groundwater. For instance, nicosulfuron was found in 8% of the samples, whereas flumetsulam was detected in 12% of the samples. The objective of this study was to characterize sorption– desorption of four SUs herbicides in soils from different soil horizons of different landscape positions in an Argentinean soil. The data obtained were used to determine potential risk of leaching to groundwater.
Azcarate et al.
Fig. 1. Chemistry structure for sulfonylurea herbicides: metsulfuron-methyl, sulfometuron-methyl, rimsulfuron and nicosulfuron.
(specific activity 62.2 mCi mg¡1; 98.8% pure), rimsulfuron (specific activity 51.8 mCi mg¡1; 98.9% pure) and sulfometuron-methyl (specific activity 48.9 mCi mg¡1; 99.0% pure) were kindly provided by DuPont Agricultural Products. All other chemical used were analytical grade reagents. Non-labeled herbicide solutions were prepared at different concentration (0.006; 0.02; 0.06; 0.2 mg L¡1) in 0.005 M CaCl2 solution. Radiolabeled chemical was added to nonradioactive solutions to give solution radioactivity of metsulfuron-methyl (229 kBq L¡1), nicosulfuron (199 kBq L¡1), rimsulfuron (160 kBq L¡1) and sulfometuronmethyl (34 kBq L¡1).
Soils The soils were collected from an agricultural field in Intendente Alvear, La Pampa, Argentina (35.40 S, 63.69 W), from four depth horizons (Ap, A12, AC and C) at three landscape positions: shoulder (SH), middle (MI) and footslope (FS). Samples were air-dried and passed through a 2 mm sieve. Subsamples of homogenized soils were
Materials and methods Table 1. Some properties of sulfonylureas herbicides studied.[2,9]
Chemicals The SUs herbicides used in this study were rimsulfuron, metsulfuron-methyl, nicosulfuron and sulfometuronmethyl. Some properties of these herbicides are listed in Table 1 and the chemical structures are in Figure 1. Pure analytical standards of 14C-labeled metsulfuron-methyl (specific activity 38.3 mCi mg¡1; 99.0% pure), nicosulfuron
Herbicides
pKa
Field Half-life (days)
Metsulfuron-methyl Sulfometuron-methyl Rimsulfuron Nicosulfuron
3.3 5.2 4.0 4.6
4–71 7–150 5–23 14–49
231
Sorption, desorption and leaching potential of sulfonylurea herbicides in Argentinean soils Table 2. Physical and chemical properties of the tested soils. Soil Depth (cm) Horizon SH
MI
FS
Downloaded by [University of Birmingham] at 10:22 21 April 2015
a
0–7 7–25 25–55 >55 0–7 7–23 23–40 >40 0–13 13–20 20–60 >60
Ap A12 AC >C Ap A12 AC >C Ap A12 AC >C
pH 5.75 § 6.10 § 6.41 § 6.61 § 6.00 § 6.12 § 6.39 § 6.64 § 7.05 § 7.81 § 8.57 § 9.14 §
0.12a 0.03 0.03 0.05 0.14 0.03 0.08 0.09 0.34 0.02 0.11 0.04
OC (g kg¡1) CEC (cmolc kg¡1) Clay (%) 4.6 § 3.0 § 2.1 § 0.5 § 9.6 § 3.1 § 2.6 § 1.0 § 11.3 § 3.7 § 0.6 § 0.2 §
1.1 0.3 0.1 0.5 1.5 0.4 0.2 0.7 1.4 1.7 0.4 0.1
5.05 § 7.26 § 6.65 § 6.05 § 8.45 § 10.31 § 9.25 § 8.17 § 11.25 § 6.98 § 9.56 § 8.00 §
0.65 0.59 2.12 1.58 0.77 1.43 1.08 1.43 0.51 1.75 1.49 1.97
2.3 § 1.6 § 1.6 § 1.6 § 1.6 § 3.6 § 2.6 § 1.6 § 3.6 § 3.3 § 4.1 § 2.3 §
1.2 0.0 0.0 0.0 0.0 1.0 1.0 0.0 0.0 0.6 0.7 1.2
Silt (%) 7.0 § 7.0 § 5.0 § 3.7 § 17.7 § 17.4 § 17.0 § 17.4 § 23.9 § 17.4 § 16.9 § 14.7 §
2.1 1.2 1.2 1.2 2.3 0.0 0.6 1.7 2.1 1.0 0.7 0.6
Sand (%)
Soil Texture
90.7 § 91.3 § 93.3 § 94.7 § 80.7 § 79.0 § 80.3 § 81.0 § 72.5 § 79.3 § 79.0 § 83.0 §
Sand Sand Sand Sand Loamy sand Loamy sand Loamy sand Loamy sand Loamy sand Loamy sand Loamy sand Loamy sand
2.1 1.2 1.2 1.2 2.3 1.0 0.6 1.7 2.1 1.2 1.4 1.0
§ Values are standard deviations for three replicates.
analyzed for texture, pH, OC and cation exchange capacity (CEC) (Table 2). Texture analysis was performed determined by the hydrometer method.[16] OC was determined by chromic acid oxidation.[17] The pH was measured with a glass electrode in a 1:1 soil/water suspension. CEC was determined by the method of ammonium acetate.[18]
equilibration. Kd ¡ SE D Cs =Cw
(1)
The coefficient of adsorption per unit OC (Koc) was calculated for adsorption by normalizing Kd-SE values to the percentage contents of OC according to Eq. (2): Koc D 100Kd ¡ SE =OC
(2)
Sorption–desorption experiment Sorption studies on soils were conducted by the batch equilibration method. Duplicate 8 g soil samples were equilibrated with 8 mL of 14C-labeled herbicides solutions (0.006, 0.02, 0.06 and 0.2 mg L¡1) using shaking mechanically at 20 C for 24 h. The highest concentration represents the usual rate used in the field diluted in the top 5 cm of soil. After equilibration, suspensions were centrifuged at 2,500 rpm for 10 min. Two mL of supernatant was removed for analysis, after which 1 mL aliquot was used to determine 14C by a liquid scintillation counter. Desorption was determined at the highest and lowest concentrations of the herbicide solutions (0.006 and 0.2 mg L¡1) immediately after the sorption experiments. The 2 mL aliquot of supernatant that was removed was replaced with the same volume of 0.005 M CaCl2 and the samples reequilibrated. Samples were shaken for 24 h and centrifuged. A total of three desorption equilibrations were carried out. The mass of chemical sorbed to soil was calculated from the difference between initial and final solution concentrations. Cs is the concentration of herbicide sorbed to soil (mg kg¡1) and Cw is the concentration of herbicide remaining in solution (mg L¡1) after equilibration. Sorption coefficients (Kd-SE) were determined for each soil and each concentration using Eq. (1). Desorption coefficients (Kd-D1, 2, 3) were also calculated after each desorption
Leaching potential To predict the leaching potential of the SUs herbicides, the Groundwater Ubiquity Score (GUS) was used (Eq. (3)). GUS D log t1=2 .4 ¡ log Koc /;
(3)
where t1/2 is herbicides half-life in soil. Herbicides with GUS index values 2.8 are ranked as leachers. t1/2 values were calculated using Koc values from the studies soils and appropriate GUS values (GUS D 1.8 and 2.8).[19] Statistical analysis Replicated soil sorption data were analyzed by Pearson correlation. The Pearson correlation coefficients (r) between sorption coefficients (Kd-SE) and soil properties (pH, OC and clay) were estimated for each herbicide. To determine the influence of landscape position and depth on the measured soil properties and Kd-SE values analysis of variance (ANOVA) was used. Equilibration steps (SE, D1, D2, D3) and sorption–desorption Kd were analyzed by linear and quadratic regression. The determination coefficient (r2) was used to check the best fit. All data analysis was run by InfoStat version 2013 for Windows.[20]
232 Results and discussion Sorption
pyridinecarboxamide SUs. Sorption Kd-SE values of rimsulfuron on the soils from the four horizons from the three landscape positions (SH, MI and FS) ranged from 1.18 to 2.08 L kg¡1. Nicosulfuron Kd-SE values were in the range from 0.02 to 0.47 L kg¡1. In contrast, metsulfuron-methyl and sulfometuron-methyl sorption Kd-SE values were very low (0.00–0.05 L kg¡1 and 0.02–0.05 L kg¡1, respectively). The sorption coefficients found in this work were lower than in the limited published studies. Oliveira et al.[21] found sorption Kd values ranging from 0.14 to 1.38 for nicosulfuron, 0.14 to 1.18 for sulfometuron-methyl, and 0.09 to 1.27 for metsulfuron-methyl in Brazilian soils. Gonzalez
Downloaded by [University of Birmingham] at 10:22 21 April 2015
There was no effect of initial solution concentration (0.006 to 0.2 mg L¡1) on sorption of the SUs; slope of the Freundlich equation »1.0. Therefore, the mean of the four Kd-SE values for each SU was calculated and the resultant mean Kd-SE values are shown in Figures 2 and 3. Overall, the sorption decreased in the order of rimsulfuron > nicosulfuron > metsulfuron-methyl D sulfometuron-methyl. The two least sorbed SUs were benzoate molecules, as opposed the more highly sorbed pyridinesulfonamide and
Azcarate et al.
Fig. 2. Rimsulfuron and nicosulfuron sorption (Kd-SE) and desorption (Kd-D) coefficients for SH, MI and FS soils at different depths.
233
Downloaded by [University of Birmingham] at 10:22 21 April 2015
Sorption, desorption and leaching potential of sulfonylurea herbicides in Argentinean soils
Fig. 3. Metsulfuron-methyl and sulfometuron-methyl sorption (Kd-SE) and desorption (Kd-D) coefficients for SH, MI and FS soils at different depths.
and Ukrainczyk[22] obtained Kf values for nicosulfuron in a range from 0.21 to 8.78. Walker et al.[23] found metsulfuron-methyl Kd values between 0.046 and 0.542, and decreased sorption with depth. Koskinen et al.[24] found for sulfometuron-methyl Kd range of 0.04–0.6 (0–20 cm depth) and sorption from depths 65–95 cm was less than in the surface soils. There is a close relationship between landscape position, depth and sorption. Rimsulfuron and nicosulfuron showed Kd-SE values £ landscape position £ depth significant interactions (P < 0.0001). Kd-SE values for rimsulfuron were the highest in FS (0–13 cm) and the lowest in SH (>55 cm).
Nicosulfuron showed the highest Kd-SE values in MI (7–23 cm) whereas the lowest were in FS (>60 cm). Metsulfuron-methyl showed significant correlation with landscape position (SH D MI > FS; P < 0.0001) and depth (Ap horizon > C horizon; P < 0.01). In contrast, no trend was observed for sulfometuron-methyl (Figs. 2 and 3). With regard to physicochemical properties were observed significant interactions among landscape position and depth: pH and OC (P < 0.0001), sand (P < 0.0001), silt (P < 0.001), CEC (P < 0.05) and clay (P < 0.05) (Table 2). OC decreased with depth (Ap, A12, AC, >C) while pH increased with depth, pH was higher in FS than MI and SH. OC
Downloaded by [University of Birmingham] at 10:22 21 April 2015
234 decreased in the order to FS > MI > SH. In relation with the texture, sand content was highest in SH at >55 cm and the lowest value was in FS in surface (0–13 cm). Silt content in Ap horizons decreased in following order FS > MI > SH whereas subsurface horizons (A12, AC and >C) decreased in order to FS D MI > SH. Clay content was higher in FS than MI and SH but decreased in depth. Soil physicochemical properties in the field can vary and can explain the measured variation in herbicides sorption within fields for different herbicides.[25–27] The soil properties that have shown the greatest effect on herbicide sorption are OC content, pH and clay content. Correlation analyses between Kd-SE, OC, soil pH and clay content showed that in SH profile there was a positive correlation between OC and sorption of rimsulfuron (r D 0.94; P < 0.05) and nicosulfuron (r D 0.98; P < 0.05), and a negative correlation with pH (rimsulfuron, r D ¡0.99; P < 0.05, nicosulfuron, r D ¡0.97; P < 0.05). In MI and FS soils, there was an increase in sorption with an increase in OC and a decrease in pH, however the correlations were not significant at P < 0.05. Due to the very low sorption and standard errors in Kd-SE values, there were no correlations between OC or pH and sorption of metsulfuronmethyl and sulfometuron-methyl for soils from the three landscape positions. Other studies have shown significant correlations between OC and sorption of SUs, i.e., metsulfuron-methyl (r D 0.941; P < 0.01) and sulfometuron-methyl (r D 0.835; P < 0.05),[21] while other authors did not find correlations for metsulfuron-methyl, sulfometuron-methyl[24,28] and rimsulfuron.[5] The contradiction is these results may be due to differences in pH levels of the soils used in the studies that may minimize the effects of OC. SUs are weak acids that exist predominantly in the anionic form when the soil pH values (Table 2) are higher than the pKa (Table 1). The neutral form is much more strongly sorbed in soils than the anion as reported for different SU herbicides sulfonylureas.[6,7,12,23,29,30] In contrast to the present work, Vicari et al.[5] found rimsulfuron sorption was not correlated with the pH (r D ¡0.37; P > 0.05) and suggested that it was probably because the soil pH (5.6–7.8) was above the pKa (4.0), and the proportion of neutral molecule was little or none. Walker et al.[23] found negative correlation with pH for metsulfuron-methyl, while other authors did not find correlations or they were not strongly correlated.[21,24,28,29] The pH range studied may sometimes be too narrow or too wide to underscore any influence of pH.[7,11] The difference between the pH at the surface of soil particles and in the soil solution might also differ according to the measurement technique used and the characteristics of the soil.[7] However, the present study shows that soil pH has a strong influence on sorption of some SUs herbicides. Although SH soil was the site of lowest pH and had the lowest Kd-SE for rimsulfuron and nicosulfuron, simultaneously this profile had the lowest OC content and coarsest
Azcarate et al. texture provided scarce sorption sites. The pH of soil can vary greatly with depth but generally increases with depth in the profile because the surface is more weathered[31] or like these soils studied where the FS profile showed the highest pH values because of the upward movement of saline groundwater. For SUs herbicides, Kd values decreased with depth as has been previously noted.[23] Sorption was not correlated with clay content for all soils tested. Similar results were reported by Oliveira et al.[21] and Cranmer et al.,[28] whereas Vicari et al.[5] found a correlation with rimsulfuron (r D 0.88; P < 0.05). Desorption All four herbicides exhibited desorption hysteresis for all SU/soil horizon/landscape position-depth combinations; the Kd desorption values (Kd-D1, 2, 3) increased with each consecutive desorption equilibration step (1/ndesorp < 1/ nsorption), even in soil samples where sorption was very low as with metsulfuron-methyl and sulfometuron-methyl (Figs. 2 and 3). For the two more extensively sorbed SUs, rimsulfuron and nicosulfuron, the increase in Kd-D with each desorption step was linear (Fig. 4). In contrast, for the more weakly sorbed metsulfuron-methyl and sulfometuron-methyl, the increase was curvilinear, with a greater increase in Kd-D with each desorption equilibration (Fig. 4). The observation that these weak acid herbicides exhibited hysteresis is surprising, since are often only slightly adsorbed by soils.[32] However, other authors also have reported this phenomenon for different SUs.[6,12,29] The specific nature of the processes responsible for the observed behavior remains unknown.[11,32,33] It could be an experimental artifact, a diffusion to inaccessible or stronger binding sites, change to stronger binding mechanism, higher degradation rate than the rate of desorption (particularly if the SUs’ half-lives are in the lower range of values reported in Table 1) or a combination of factors.[11] The desorption process is very important in the prediction of pesticide movement in soils.[31] Most transport models just use Kd-SE to predict depth of leaching and Kd-SE is inversely proportional to depth of leaching. These models do not incorporate hysteresis into the potential depth of leaching. If we used the desorption Kd-D3 value (Kd calculated after the third desorption step), as compared to the adsorption Kd-SE, the depth of leaching would be decreased by a factor 2-3X for rimsulfuron and nicosulfuron, and 4-15X for metsulfuron-methyl and sulfometuron-methyl.
Leaching potential Many researchers have proposed indices to predict the mobility of pesticides in the field. The most used index is the Groundwater Ubiquity Score (GUS), which was
235
Downloaded by [University of Birmingham] at 10:22 21 April 2015
Sorption, desorption and leaching potential of sulfonylurea herbicides in Argentinean soils
Fig. 4. Rimsulfuron and metsulfuron-methyl sorption (Kd-SE) and desorption (Kd-D) coefficients as a function of equilibration step.
developed by Gustafson[19] and is based on the values of Koc and half-life (t1/2). These important parameters can determine the mobility and persistence of the herbicides.[19,34] Using Koc values and appropriate GUS values, t1/2 values were calculated that would rank these herbicides as leachers or nonleachers (Table 3). In the Ap horizon, the Koc decreased in the order SH > MI > FS, which in turn decreased the t1/2 value below which the SU would be considered a nonleacher and above which it would be considered as leacher (Table 3). For instance, rimsulfuron would be considered a nonleacher for t1/2 values 78 days in SH soils and 41 days in FS soils. Smaller decreases in t1/2 values were also observed for the other three SUs. At each of the three landscape positions, rimsulfuron and nicosulfuron Koc values generally increased with soil depth thereby increasing the t1/2 value below which the SU would be considered a nonleacher and above which it would be considered a leacher (Table 3). For instance, for rimsulfuron to be considered a nonleacher, t1/2 would have to 184 days in the AC horizon soil.
236
Azcarate et al.
Table 3. Estimated half-lives (t1/2) for rimsulfuron, nicosulfuron, metsulfuron-methyl and sulfometuron-methyl that would classify the herbicide as a “leacher” or “nonleacher”. Rimsulfuron Site Horizon Koc (L kg¡1) t1/2 (days) SH
MI
Downloaded by [University of Birmingham] at 10:22 21 April 2015
FS
Ap A12 AC >C Ap A12 AC >C Ap A12 AC >C
333 456 581 2360 212 588 631 1308 184 457 2864 8210
Nicosulfuron
Metsulfuron-methyl
Sulfometuron-methyl
Koc (L kg¡1) t1/2 (days) Koc (L kg¡1) t1/2 (days) Koc (L kg¡1) t1/2 (days)
17a/78b 22/122 29/184 2 years/80 years 12/47 29/189 32/215 0.3 years/4 years 11/41 22/123 6 years/392 years >500 years
72 99 116 360 40 150 142 272 35 90 265 93
7/20 8/25 9/28 18/87 6/15 10/34 9/33 14/61 5/14 8/23 14/60 8/24
11 10 10 40 5 10 15 20 4 3 0 0
4/9 4/9 4/8 6/15 4/7 4/8 4/10 5/11 3/6 3/6 — —
7 10 15 40 3 13 8 30 4 8 66 93
4/8 4/9 4/10 6/15 3/6 4/9 4/8 5/13 3/7 4/8 7/19 8/24
t1/2 below which SUs would be considered a “nonleacher”. t1/2 above which would be considered as a “leacher”.
a
b
Reported t1/2 values for rimsulfuron, nicosulfuron, metsulfuron-methyl and sulfometuron-methyl were 5–23, 14– 49, 4–71 and 7–150 days, respectively.[2,9,35] Using the midpoint of these values of t1/2 for rimsulfuron (14 days), it would be considered a nonleacher in all soils from the three landscape position and from all soil horizons, which would indicate a low potential for leaching to groundwater. Rimsulfuron is very unstable, particularly at pH values above 8, where it is rapidly hydrolyzed. This is the reason why it has not been found in groundwater.[36] However, degradation products were detected in the water and groundwater study zones for several years following application of rimsulfuron. Thus, they are relatively stable and persist in the soil, from where they leach to the groundwater.[37] Using the midpoint of these values of t1/2 for nicosulfuron (31 days), it would be considered a leacher in the upper three horizons of the SH soil, the Ap horizon in the MI soil and in the Ap, A12, and lowest horizon in the FS soil. However, reported strong adsorption of nicosulfuron on clay minerals should decrease its mobility in soils and its potential to leach into groundwater. Sorption on mineral surfaces is especially important in the subsurface horizons and in soils with low organic matter content. The fast kinetics of the adsorption implies that the nicosulfuron moving down the soil profile should be readily and strongly retained on the surfaces of clay minerals.[38] The midpoints of t1/2 for metsulfuron-methyl and sulfometuron-methyl classify both of them as leachers. Hollaway et al.[39] found that the persistence of metsulfuronmethyl in alkaline soils is less than 1 year after herbicide application. It has been shown to range in mobility in soil from low mobility to relatively mobile.[40] Based on Koc values for a number of Brazilian soils, Oliveira et al.[21] calculated that a t1/2 value >16.5 days would also classify
metsulfuron-methyl and sulfometuron-methyl as leachers in all the Oxisol soils studied. In summary, knowledge of the fate and transport of herbicides in agricultural soils is of great importance to agronomic systems and to environmental protection. According to the results, rimsulfuron and nicosulfuron would rank as leachers depending on the landscape position and soil depth the samples were taken from. If rimsulfuron and nicosulfuron leach to groundwater, that would be dependent on the rate of decomposition in both surface and subsurface soils.[24] In contrast, metsulfuron-methyl and sulfometuron-methyl would be classified as leachers, regardless of landscape position and soil depth. It is also important to incorporate hysteresis when predicting potential mobility. Use of sorption Kd values may overpredict transport, when there is desorption hysteresis and sorption is concentration dependent (Freundlich 1/n values
