Elsevier

Journal of Hazardous Materials

Volume 186, Issue 1, 15 February 2011, Pages 540-550
Journal of Hazardous Materials

Pesticide adsorption in relation to soil properties and soil type distribution in regional scale

https://doi.org/10.1016/j.jhazmat.2010.11.040Get rights and content

Abstract

Study was focused on the evaluation of pesticide adsorption in soils, as one of the parameters, which are necessary to know when assessing possible groundwater contamination caused by pesticides commonly used in agriculture. Batch sorption tests were performed for 11 selected pesticides and 13 representative soils. The Freundlich equations were used to describe adsorption isotherms. Multiple-linear regressions were used to predict the Freundlich adsorption coefficients from measured soil properties. Resulting functions and a soil map of the Czech Republic were used to generate maps of the coefficient distribution. The multiple linear regressions showed that the KF coefficient depended on: (a) combination of OM (organic matter content), pHKCl and CEC (cation exchange capacity), or OM, SCS (sorption complex saturation) and salinity (terbuthylazine), (b) combination of OM and pHKCl, or OM, SCS and salinity (prometryne), (c) combination of OM and pHKCl, or OM and ρz (metribuzin), (d) combination of OM, CEC and clay content, or clay content, CEC and salinity (hexazinone), (e) combination of OM and pHKCl, or OM and SCS (metolachlor), (f) OM or combination of OM and CaCO3 (chlorotoluron), (g) OM (azoxystrobin), (h) combination of OM and pHKCl (trifluralin), (i) combination of OM and clay content (fipronil), (j) combination of OM and pHKCl, or OM, pHKCl and CaCO3 (thiacloprid), (k) combination of OM, pHKCl and CEC, or sand content, pHKCl and salinity (chlormequat chloride).

Research highlights

Adsorption isotherms of 11 selected pesticides were measurement for 13 soils. ▶ Functions for estimating adsorption coefficients from soil properties were obtained. ▶ Adsorption coefficients of pesticides for soils of the Czech Republic were predicted.

Introduction

Groundwater contamination caused by pesticides used in agriculture is an environmental problem worldwide. Groundwater contamination depends on many factors and conditions. Groundwater vulnerability maps, which may be constructed using the DRASTIC methodology [1], assume soil texture, hydrological conditions, hydrogeological settings and climatic conditions and does not account for the properties of a contaminant. One factor is a soil cover and specific pesticide adsorption in soils. It was documented in many studies that sorption processes of organic compounds depend on the sorbent physicochemical properties as pH, cation exchange capacity, ionic strength, surface area, etc. [2]. The organic matter content is usually suggested to have a greatest effect on the pesticide adsorption in natural soils. Pesticide partition (distribution) coefficient (KD) is usually calculated (based on the positive correlation between the organic carbon content and the KD coefficient) assuming the organic carbon fraction in soil and the organic carbon partition coefficient (KOC) [3]. However, also other factors may play an important role. Richter et al. [4] summarized various equations relating the KD coefficient to organic carbon (OC) and pH. Either linear or nonlinear equation for the KD value and OC was presented. The multiple linear regression between the KD value and OC, pH, and OC multiplied by pH was also introduced. In addition, a nonlinear relationship between the KD value and OC with the pH impact (pH as the exponent in the exponential function) was shown. Kozák and Vacek [5], [6] assumed even more soil properties and proposed a pedotransfer rule for the prediction of the KD coefficients for atrazine. They used a multiple linear regression to evaluate relationship between the KD coefficients and the organic matter content (OM), pH, clay content and cation exchange capacity (CEC). Different sorption behavior of acidic and basic pesticides with respect to soil properties was documented by Kah and Brown [7]. pH impact depends on pesticide acidity (pKa) and base (pKb) constants [8]. The multiple linear regression between the log KD value and log OC and log D (the lipophilicity profile of the pesticides) was applied by Villaverde et al. [9]. Rodriguez-Rubio et al. [10] showed that the larger content of calcium carbonate increased 2,4-D sorption.

Eleven pesticides were examined in our study: terbuthylazine, prometryne, metribuzin, hexazinone, metolachlor, chlorotoluron, trifluralin, azoxystrobin, fipronil, thiacloprid and chlormequat chloride.

Terbuthylazine sorption was studied previously by Dousset et al. [11], Meyer-Windel et al. [12], Singh et al. [13], and Finocchiaro et al. [14]. Dousset et al. [11] found that the adsorption in clay soils was not related to clay or OC, but to the degree of humification of the organic matter. A non-linear equation was used by Meyer-Windel et al. [12] to describe relationship between KF and OC. The KF coefficient was higher for soils with higher OC [13]. Finocchiaro et al. [14] showed that the KD coefficient was significantly correlated with OC and amorphous iron oxides.

Prometryne sorption was measured by Yang et al. [15] and Oliver et al. [16]. Yang et al. [15] showed that the KF coefficient had a good correlation with OM for 6 soils. Oliver et al. [16] found that the relationship between KD and OC was not significant, but the relationship between derived KOC and pH was significant.

Metribuzin sorption was examined by Garcia-Valcarcel et al. [17], Daniel et al. [18], and Kah and Brown [7]. Metribuzin adsorption in 18 soil of Central Spain did not show any correlation with studied soil properties [17]. The KF was related positively to OM [18]. Adsorption was negatively correlated with soil pH and positively correlated with OC [7].

Hexazinone sorption was investigated by Donati et al. [19] and Calderon et al. [20]. Very low values of KD and KOC were measured for 6 soils [19]. Higher KF values were measured in soil with higher OM [20].

Metolachlor sorption was measured by Wang et al. [21], Singh et al. [13], Liu et al. [22], Weber et al. [23], Vryzas et al. [24] and Si et al. [25]. Wang et al. [21] showed good correlation between the KF coefficient and the OM. The KF coefficient was higher for soils with higher OC [13]. Liu et al. [22] presented that the KF coefficient increased with OC. The KD coefficient for soil humic acid was greater than on clay. Weber et al. [23] presented that the retention of metolachlor was positively influenced by OM and clay content. Principal component and multiple regression analyses showed KF dependence on OC in the plough layers and on the clay content and pH in the subsurface horizons [24]. Si et al. [25] showed the KD dependence on OM.

Chlorotoluron sorption was studied by Meyer-Windel et al. [12], Gao et al. [26], Hiller et al. [27], Kodešová et al. [28], [29]. A linear relationship between the Freundlich adsorption coefficient (KF) and organic carbon (OC) was proved by Meyer-Windel et al. [12]. Adsorption increases with increasing N content [26]. The KF values positively correlated with OC [27]. The KF values were higher in soil horizons with higher OC [28], [29].

Trifluralin sorption was examined by Tavares and Rezende [30] and Cooke et al. [31]. The herbicide adsorption increased with increasing OM [30]. Trifluralin exhibited strong partitioning to the soil solid phase and low desorption potential [31].

Azoxystrobin adsorption increased in the compost-amended soils [32].

Fipronil sorption was investigated by Bobe et al. [33], and Masutti and Mermut [34]. The KF coefficient increased with the increasing OM [33]. The KF values for two tropical soils showed that the presence of higher amount of poorly crystalline Fe-oxyhydroxides increased fipronil adsorption [34].Thiacloprid sorption was investigated using 22 soils [35]. Sorption was not significantly related with soil characteristics, namely OC, clay content, and pH. Correlation with OC was obtained for some separated land uses.

Chlormequat chloride adsorption decreased in the presence of heavy metals [36].

Ground water is an important source of drinking water in the Czech Republic. Our study is a part of a project, which is focused on the assessment of the risk of the selected pesticide leaching into the groundwater within the entire region of the Czech Republic. First of all, results of the project will be used to optimize a national groundwater quality monitoring system, which is operated by the Czech Hydrometeorological Institute [37]. In addition, they may help to improve pesticide management within the areas with an increased pesticide leaching potential. In order to construct the specific groundwater vulnerability maps for selected pesticides based on the modified DRASTIC methodology, the maps of the KF coefficients within the region of the Czech Republic were produced, which are presented here. The goals of this study were: (1) to select representative soils of the Czech Republic and determination of soil physical and chemical properties; (2) to measure adsorption isotherms of selected pesticides; (3) to determine pedotransfer rules for estimating sorption coefficients from the measured soil properties; (4) to apply the pedotransfer rules, the soil map of the Czech Republic [38] and the Czech soil information system PUGIS [39] for estimating the adsorption coefficients of pesticides for soils of the Czech Republic.

Section snippets

Soil chemical and physical properties

Thirteen representative soils (11 samples from humic horizons of various soil types and 2 substrates) of the Czech Republic (Table 1) were selected to study a pesticide adsorption in soils. Soils were selected based on the soil map of the Czech Republic [38], and the Czech soil information system PUGIS [39] to cover a large variability of soil properties, which may influence pesticides adsorption in soils. 20 kg of dry soil was collected from each location. The soil samples were dried, ground

Pesticide adsorption isotherms

Resulting parameters KF and n of the Freundlich adsorption equations are not shown here. The KF coefficient is commonly used to assess pesticide sorption in various soils (the large KF value indicates large pesticide sorption). However, fitting the same experimental data the KF value depends on the n coefficient (the lower KF values are obtained for lower n values, e.g. higher 1/n values). The relatively similar n values for various soils were obtained only for metolachlor. Therefore the

Conclusions

Analysis in some degree verified previous findings, which were summarized in the introduction part. While relationships of high significance were obtained for basic pesticides of moderate solubility and moderate adsorption ability (terbuthylazine and prometryne), relationships of the lowest significance were obtained for basic pesticides of large solubility and low adsorption ability (metribuzin and hexazinone). Terbuthylazine adsorption proved to correlate with OM. In addition, the multiple

Acknowledgements

Authors acknowledge the financial support of the Ministry of Education, Youth and Sports of the Czech Republic (grants no. 2B06095 and MSM 6046070901). Authors would like to thank colleagues Vít Penížek and Marcela Mühlhanselová for helping with the soil selection, to Roman Grabic for valuable consultancy in chemistry and to students Helena Pijálková, Tomáš Tykal, Roman Božek and Táňa Jonášková for helping with some of the sorption experiments.

References (56)

  • O. Richter et al.

    Environmental Fate Modelling of Pesticides

    (1996)
  • J. Kozák et al.

    The mathematical model (BPS) for prediction of pesticide behavior in soil

    Plant. Product.

    (1996)
  • J. Kozák et al.

    Pedotransfer functions as a tool for estimation of pesticides behavior in soils

    Plant. Product.

    (2000)
  • M. Kah et al.

    Prediction of adsorption of ionizable pesticides in soils

    J. Agric. Food Chem.

    (2007)
  • A.G. Hornsby et al.

    Pesticide Properties in the Environment

    (1996)
  • J. Villaverde et al.

    Adsorption and degradation of four acidic herbicides in soils from southern Spain

    Pest Manag. Sci.

    (2008)
  • P. Rodriguez-Rubio et al.

    Sorption of 2,4-D on natural and organic amended soils of different characteristics

    J. Environ. Sci. Health Part B

    (2006)
  • S. Meyer-Windel et al.

    On the relation of herbicide adsorption and soil organic fraction

    Z. Pflanz. Bodenkunde

    (1997)
  • N. Singh et al.

    Sorption behavior of metolachlor, isoproturon, and terbuthylazine in soils

    J. Environ. Sci. Health Part B

    (2001)
  • R. Finocchiaro et al.

    Adsorption of molinate, terbuthylazine, bensulfuron-methyl, and cinosulfuron on different Italian soils

    Fresen. Environ. Bull.

    (2005)
  • W.C. Yang et al.

    Adsorption and correlation with their thermodynamic properties of triazine herbicides on soils

    J. Environ. Sci. China

    (2003)
  • D.P. Oliver et al.

    Land use effects on sorption of pesticides and their metabolites in sandy soils. II. Atrazine and two metabolites, deethylatrazine and deisopropylatrazine, and prometryne

    Aust. J. Soil. Res.

    (2003)
  • A.I. Garcia-Valcarcel et al.

    Adsorption of triazines in soils with low organic matter content

    Fresen. Environ. Bull.

    (1998)
  • P.E. Daniel et al.

    Atrazine and metribuzin sorption in soils of the Argentinean humid pampas

    Environ. Toxicol. Chem.

    (2002)
  • L. Donati et al.

    KOC Estimation of deethylatrazine, deisopropylatrazine, hexazinone and terbuthylazine by reversed-phase chromatography and sorption isotherms

    Toxicol. Environ. Chem.

    (1994)
  • Q.Q. Wang et al.

    Adsorption of acetanilide herbicides on soils and its correlation with soil properties

    Pest. Sci.

    (1999)
  • W.P. Liu et al.

    Adsorption of chloroacetanilide herbicides on soil (I)—structural influence of chloroacetanilide herbicide for their adsorption on soils and its components

    J. Environ. Sci. China

    (2001)
  • J.B. Weber et al.

    Sorption and mobility of C-14-labeled imazaquin and metolachlor in four soils as influenced by soil properties

    J. Agric. Food Chem.

    (2003)
  • Cited by (110)

    View all citing articles on Scopus
    View full text