- Email: [email protected]
Assessment of olive cake as soil amendment for the controlled release of triazine herbicides
Science of the Total Environment 378 (2007) 119 – 123 www.elsevier.com/locate/scitotenv
Assessment of olive cake as soil amendment for the controlled release of triazine herbicides Laura Delgado-Moreno, Lourdes Sánchez-Moreno, Aránzazu Peña ⁎ Estación Experimental del Zaidín (CSIC), Profesor Albareda, 1, 18008-Granada, Spain Available online 6 February 2007
Abstract Organic matter-rich agricultural by-products are being produced in huge quantities and can be applied to soil as a disposal strategy. The application of two different rates (2 and 8% w/w) of olive cake to a Mediterranean calcareous soil resulted in an increased sorption of four triazine herbicides, which was higher for the more hydrophobic compounds (terbuthylazine and prometryn) and lower for the more polar ones (simazine and cyanazine). However, when the sorption coefficients were normalised to the total soil organic carbon (Koc), the results did not significantly differ between simazine and cyanazine which is an indication that the olive cake did not exert different sorption capacity for both compounds. On the contrary, Koc values for terbuthylazine and prometryn increased in the amended soils. Our results from experiments using mixtures of several pesticides suggest that competition for sorption sites resulted in a decrease of herbicide sorption. Desorption was hysteretical both for the amended and unamended soils, but the addition of olive cake at the highest dose diminished desorption of most of the herbicides. In conclusion, the addition of olive cake behaves as a promising method for reducing the risk of groundwater pollution by pesticides. © 2007 Elsevier B.V. All rights reserved. Keywords: Soil; Amendments; Triazine herbicides; Sorption; Desorption; Competition
1. Introduction The olive tree (Olea europaea L.), is an important and characteristic cultivation of the Mediterranean area. The extraction of olive oil, which in the last years has introduced a two-stage centrifugation process because of its high efficacy and low energy and water consumption, produces some by-products which could be potentially used in crop management. These byproducts are of increasing interest as soil amendments, because of the low organic matter (OM) content of the Mediterranean soils together with some problems of erosion. The olive culture in Spain is of high socio⁎ Corresponding author. Tel.: +34 958 181600; fax: +34 958 129600. E-mail address: [email protected] (A. Peña). 0048-9697/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.scitotenv.2007.01.023
economical importance, since it constitutes 11% of the national agricultural surface, representing about 2.42 · 106 ha (http://www.mapya.es). In Andalusia, with ca. 60% of the total national production, olive oil in the 2002–2003 campaign accounted for ca. 680,000 Mg. Agricultural by-products, along with other waste disposal amendments of high OM content, have been applied to soils in the last decades to enhance their fertility and improve their physicochemical properties, such as soil porosity and aggregation (Cox et al., 1997; Sluszny et al., 1999). Moreover, the addition of exogenous material to soil (peat, manure, sewage sludge, compost, etc.), has been extensively used in the last few years to modify the fate of pesticides in the environment (Senesi et al., 1997; Cox et al., 1997; Albarrán et al., 2003).
120
L. Delgado-Moreno et al. / Science of the Total Environment 378 (2007) 119–123
Triazine herbicides have been widely found in groundwaters (Burkart et al., 1999; Battaglin et al., 2000) due to their moderate sorption by soil and relatively long persistence. Therefore, modification of triazine retention by amending soil with agricultural byproducts of high OM content has been undertaken in order to reduce their residual effect in the environment. 2. Materials and methods 2.1. Soil Soil was sampled from the upper layer (0–25 cm) of an olive orchard located near Iznalloz (province of Granada, South of Spain), air-dried and passed through a 2-mm sieve. It was a calcaric Regosol, with silty clay loam texture, with 34% clay, 56% loam and 10% sand, containing a 44.1% CaCO3 and with pH of 7.9.
15 °C for 15 min (Eppendorf 5810R). A one-mL portion was measured from the supernatant and analysed as described below. Desorption experiments were performed with the highest concentration by replacing a 10-mL aliquot of the supernatant with 10 mL of water. At each desorption step the soil in the tube was resuspended in solution using a vortex mixer and then the tubes were mechanically shaken end-over-end in a thermostatic chamber at 15 ± 1 °C, for 24 h. This desorption step was three fold repeated. The sorption– desorption experiments were described by the empirical Freundlich equation (X = Kf × Ce1/n), being X (μg g− 1) the amount of herbicide sorbed at the equilibrium concentration, Ce (mg L− 1), Kf the Freundlich sorption– desorption coefficient (L kg− 1) and 1 / n a constant which depends on the sorbate, the sorbent and temperature properties. 2.5. Extraction and analytical procedure
2.2. Organic amendments
The selected herbicide was terbuthylazine, 99.3% purity (Dr. Ehrenstorfer, Germany). It is a pre- or postemergence broad-spectrum herbicide to control weeds in many crops (vines, fruit trees, olive trees, etc.) (Tomlin, 2003). Apart from terbuthylazine, other triazine herbicides no longer permitted (cyanazine, simazine and prometryn, all with a purity N99%) were also included as model compounds.
A 1-mL aliquot of the herbicide mixture was vortexed (Reax 200, Heidolph, Kolheim, Germany) with 2 mL of hexane/toluene (1/1 v/v). Phase separation was achieved by sample freezing in a conventional freezer (− 18 °C) for 2 h. The organic phase was passed to a vial and analysed. Recoveries were of 89.2% for terbuthylazine and ranged from 73.2 to 92.1% for the other triazines. Samples were analysed by gas chromatography in a Varian Star 3400 CX equipped with a TSD detector and a Varian 8200 automatic injector. One μL of the sample was injected splitless, using He as carrier gas, on a HPUltra 2 column (cross-linked 5% phenyl methyl silicone) (25 m, 0.32 mm i.d., 0.17 mm film thickness). Injector and detector temperatures were 280 and 300 °C, respectively. Oven temperature was programmed starting at 45 °C (1 min), at 25 °C min− 1 until 160 °C and at 7 °C min − 1 until 210 °C (2 min). Under these experimental conditions retention time of terbuthylazine was 12.1 min and 11.6, 13.9 and 15.0 min, for simazine, prometryn and cyanazine, respectively.
2.4. Adsorption and desorption isotherms
3. Results and discussion
Isotherms were carried out using a batch equilibration method. Soil samples (5 g) were mechanically shaken end-over-end in a thermostatic chamber at 15 ± 1 °C with 20 mL of aqueous solutions of terbuthylazine and the other three herbicides each at 2, 5, 10, 15 and 20 mg L− 1, for 12 h. The experiments were run in duplicate with a control of the herbicides solution without soil. The slurry was centrifuged at 3000 rpm and
3.1. Sorption experiments
Olive cake (CA), from a two-stage centrifugation process, was air-dried and passed through a 2-mm sieve. It was kindly supplied by an olive oil industry (Romeroliva, Deifontes, Granada). Two dosages were applied to the soil, a low or agricultural dosage, equivalent to 50 Mg ha− 1 and a high dosage or dosage with remediation potential, at 200 Mg ha− 1. Amended soil had an OC content of 4.89 and 2.63%, and pH of 7.1 and 7.2 for the high and low dosage, respectively. 2.3. Herbicides
3.1.1. Non-amended soil Experimental data were fitted to the Freundlich model. Table 1 shows the values of Kf and 1 /n for the herbicides studied. Herbicide sorption on non-amended soil provided the lowest Freundlich sorption coefficient (Kf). However, since the sorption parameters are affected by
L. Delgado-Moreno et al. / Science of the Total Environment 378 (2007) 119–123
121
Table 1 Sorption parameters for the four triazine herbicides in non-amended and amended soil Adsorption parameters a
Desorption parameters a
2
c X10
Kf ± SDa
1 / n ± SDa
R2
Hd
De
Kf ± SD
1 / n ± SD
R
b Koc
1.8 ± 0.2 11.5 ± 0.7 2.2 ± 0.2
0.97 ± 0.05 0.93 ± 0.05 1.28 ± 0.06
0.95 0.95 0.97
82.3 247.8 137.5
17.0 98.6 41.2
15.4 ± 0.5 42.6 ± 1.0
0.22 ± 0.02 0.16 ± 0.02
0.91 0.86
0.23 0.17
36.0 23.7
3.2 ± 0.3 17.5 ± 0.3 6.2 ± 0.5
0.66 ± 0.04 0.64 ± 0.04 0.83 ± 0.05
0.94 0.97 0.96
71.7 275.4 197.4
14.5 76.4 42.1
10.4 ± 0.5 35.1 ± 0.6
0.29 ± 0.03 0.21 ± 0.01
0.84 0.93
0.44 0.33
45.0 20.1
0.7 ± 0.1 1.4 ± 0.2 0.8 ± 0.1
0.86 ± 0.07 0.95 ± 0.09 1.11 ± 0.05
0.92 0.87 0.98
22.4 26.6 42.1
5.2 12.8 10.9
4.1 ± 0.2 5.6 ± 0.3
0.36 ± 0.03 0.41 ± 0.03
0.87 0.90
0.41 0.43
57.4 59.4
0.7 ± 0.1 1.3 ± 0.1 0.8 ± 0.1
1.00 ± 0.07 1.05 ± 0.05 1.20 ± 0.04
0.94 0.98 0.99
30.1 39.8 41.4
7.3 14.0 12.2
3.9 ± 0.1 10.5 ± 0.2
0.38 ± 0.02 0.31 ± 0.01
0.96 0.98
0.38 0.30
55.7 56.8
TBTA S SCA (8%) SCA (2%) PROM S SCA (8%) SCA (2%) SIM S SCA (8%) SCA (2%) CYAN S SCA (8%) SCA (2%)
Standard error of two duplicates. bCalculated as Koc = Kd (Ce = 10 mg L− 1) ⁎ 100 / %OC. cObtained from Freundlich equation. dHysteresis coefficient (1 / ndes / 1 / nads). ePercentage of herbicide desorbed. TBTA: terbuthylazine, PROM: prometryn, SIM: simazine, CYAN: cyanazine. S: non-amended soil, SCA: amended soil with olive cake at two application dosages: 8% (200 Mg ha− 1) and 2% (50 Mg ha− 1).
a
the slope (1 / n), for comparison purposes we have preferred to compare the sorbed quantity at 10 mg L− 1 (X10). According to these values, sorption increased in the order simazine ≈ cyanazine b terbuthylazine ≈ prometryn (Fig. 1), higher for the most hydrophobic herbicides terbuthylazine and prometryn (Log Kow 3.2 and 3.1), and lower for the less hydrophobic ones, simazine and cyanazine (Log Kow 2.1) (Tomlin, 2003). The values of organic C-normalised sorption coefficients (Koc values in Table 1) obtained for the different triazine herbicides were 22–30 for simazine and cyanazine, and lower than 100 for terbuthylazine and prometryn. Other reports in the literature have cited Koc b 100 or around 100 for simazine and cyanazine (Liu and Qian, 1995; Barriuso et al., 1997; Oliveira et al., 2001) and values of 200 for prometryn and 230–300 for terbuthylazine (Businelli, 1997; Sluszny et al., 1999). The lower values encountered here could be explained by competition for sorption sites, because of the simultaneous sorption of the four herbicides. The phenomenon of competition has been described for triazine herbicides (Xing et al., 1996; Turin and Bowman, 1997; Businelli, 1997) and for other contaminants (Martins and Mermoud, 1998) leading to non-linear isotherms and to the lowering of the sorption constants (Kf or Koc). This is especially the case when chemicals with similar structure compete for the same sorption sites (Xing et al., 1996). Special care should be taken to avoid the simultaneous application of herbicides because a decrease in soil retention can occur (Table 1).
3.1.2. Addition of olive wastes to soil The addition of olive cake to soil resulted in increased sorption, which was higher for the highest dose (8%) (Fig. 1). In amended soils the highest sorption corresponded again to terbuthylazine and prometryn, revealing the affinity of these herbicides for the organic amendment (Table 1). In accordance with Barriuso et al. (1997) differences in sorption between amended and non-amended soils were found to be negligible for the most polar compounds with Koc values lower than 50 L kg− 1 C, such as simazine and cyanazine (Table 1). For more hydrophobic compounds, such as terbuthylazine and prometryn, Koc values increased when the soil was amended. This fact is an indication that the OM derived from the olive cake has a greater sorption capacity for both compounds. This could be due to the different nature of the organic fractions present in olive cake when compared to that from natural soil (Chefetz et al., 2004). The addition of amendments increased significantly Koc values which approached values reported in the literature, especially for prometryn and terbuthylazine. Competition for sorption sites in amended soil may have diminished since additional sites were probably available. Therefore, the compounds more tightly bound to the soil matrix would displace those less retained, such as cyanazine and simazine, for which no increase in Koc values was noticed.
122
L. Delgado-Moreno et al. / Science of the Total Environment 378 (2007) 119–123
Fig. 1. Sorption–desorption isotherms (amount of herbicide sorbed (X) vs. concentration of herbicide in solution (Ce)) of simazine, cyanazine, prometryn and terbuthylazine in non-amended (♦) and amended soil (8%) (■) (solid symbols — sorption points, open symbols — desorption points).
Since the triazines are weak bases (Table 1), they could be in one of two forms depending on the pH: the ion or the neutral molecule, and the sorbability of these two forms can differ to a large extent. Concerning the pKa of the herbicides, the three Cl-triazines (pKa 0.6–2) could be considered neutral herbicides in soil since their pKa is ≪ pH − 1 (Green and Karickhoff, 1990). Prometryn, like other 2-methylthiotriazines, such as terbutryn or ametryn, is a less weak base (pKa 4.1) and could be partly protonated when soil was added with olive cake and the pH was lowered to 7.05. This would explain the highest increase in Koc values due to the existence of additional mechanisms, such as cation exchange. 3.2. Desorption experiments The desorption process from natural soil was hysteretical for all herbicides studied (Table 1). This was an indication that desorption was not reversible, as can be seen in the percentage of herbicide desorbed (D) after four desorption steps (Table 1) and in the coefficient of hysteresis (H ). This fact contrasts with the results obtained by Albarrán et al. (2003) who showed negative hysteresis of simazine in unamended soils and amended with olive cake, which pointed to a high reversibility of the sorption–desorption process.
Simazine and cyanazine reached desorption percentages of more than 50% of their initial concentration, but for prometryn and terbuthylazine only 35–45% was desorbed. The addition of olive cake increased the retention of these herbicides to the soil since the final desorbed quantity was around 25% for terbuthylazine and prometryn. Therefore their leaching and consequently their ability to reach ground water were greatly diminished. In contrast, no differences, between soil and soil amended with olive cake, were observed in percentage values of simazine and cyanazine desorbed. This fact follows the same trend observed for Koc values and indicates that less hydrophobic compounds, such as simazine and cyanacine, were more weakly bound to soil than the most hydrophobic ones, terbuthylazine and prometryn. 4. Conclusions Olive cake addition to agricultural soil increases sorption of all triazine herbicides tested and reduces desorption of the most hydrophobic ones, terbuthylazine and prometryn. Both processes favour the reduction of their environmental impact and may attenuate the leaching of triazine herbicides to groundwater. Treating soils with olive cake could also be considered as a
L. Delgado-Moreno et al. / Science of the Total Environment 378 (2007) 119–123
disposal strategy for these residues, which are produced in huge quantities in the Mediterranean region, both to enhance the fertility and improve the physicochemical properties of the soil. The sorption behaviour is also related with the physicochemical characteristics of the compounds, and increase with the hydrophobic character of the herbicides. Further research is required to establish the organic fraction responsible for the pesticide–substrate interaction. Care should be taken with the simultaneous application of different herbicides, since competition among them can reduce their retention by soil. Acknowledgements LDM thanks the Spanish Ministry of Education for a FPU grant received. The Project CAO001-007 from Junta de Andalucía has supported this study. References Albarrán A, Celis R, Hermosín MC, López-Piñero A, Ortega-Calvo JJ, Cornejo J. Effects of solid-mill waste addition to soil on sorption, degradation and leaching of the herbicide simazine. Soil Use Manage 2003;19:150–6. Barriuso E, Houot S, Serra-Wittling C. Influence of compost addition to soil on the behaviour of herbicides. Pestic Sci 1997;49:65–75. Battaglin WA, Furlong ET, Burkhardt MR, Peter CJ. Occurrence of sulfonylurea, sulfonamide, imidazolinone, and other herbicides in rivers, reservoirs, and ground water in the Midwestern United States, 1998. Sci Total Environ 2000;248:123–33. Burkart MR, Kolpin DW, Jaquis RJ, Cole KJ. Agrichemicals in ground water of the Midwestern USA: relations to soil characteristics. J Environ Qual 1999;28:1908–15.
123
Businelli D. Pig slurry amendment and herbicide coapplication effects on s-triazine mobility in soil: an adsorption–desorption study. J Environ Qual 1997;26:102–8. Chefetz B, Bilkis YI, Polubesova T. Sorption–desorption behavior of triazine and phenylurea herbicides in Kishon river sediments. Water Res 2004;38:4383–94. Cox L, Celis R, Hermosin MC, Becker A, Cornejo J. Porosity and herbicide leaching in soils amended with olive-mill wastewater. Agric Ecosyst Environ 1997;65:151–61. Green RE, Karickhoff SN. Sorption estimates for modelling. In: Cheng HH, editor. Pesticides in the soil environment: processes, impacts and modelling, vol. 2. Soil Sci Soc America, WI, Book Series; 1990. p. 79–101. Chapter 4. Liu J, Qian C. Hydrophobic coefficients of s-triazine and phenylurea herbicides. Chemosphere 1995;31:3951–9. Martins JM, Mermoud A. Sorption and degradation of four nitroaromatic herbicides in mono and multi-solute saturated/unsaturated soil batch systems. J Contam Hydrol 1998;33:187–210. Oliveira Jr RS, Koskinen WC, Ferreira FA. Sorption and leaching potential of herbicides on Brazilian soils. Weed Res 2001;41:97–119. Senesi N, La Cava P, Miano TM. Adsorption of imazethapyr to amended and nonamended soils and humic acids. J Environ Qual 1997;26:1264–70. Sluszny C, Graber ER, Gerstl Z. Sorption of s-triazine herbicides in organic matter amended soils: fresh and incubated systems. Water Air Soil Pollut 1999;115:395–410. Tomlin CDS. The pesticide manual. 13rd edn. Alton, Hampshire, UK: British Crop Protection Council; 2003. 1344 pp. Turin HJ, Bowman RS. Sorption behavior and competition of bromacil, napropamide, and prometryn. J Environ Qual 1997;26:1282–7. Xing B, Pignatello JJ, Gigliotti B. Competitive sorption between atrazine and other organic compounds in soils and model sorbents. Environ Sci Technol 1996;30:2432–40.
Assessment of olive cake as soil amendment for the controlled release of triazine herbicides Laura Delgado-Moreno, Lourdes Sánchez-Moreno, Aránzazu Peña ⁎ Estación Experimental del Zaidín (CSIC), Profesor Albareda, 1, 18008-Granada, Spain Available online 6 February 2007
Abstract Organic matter-rich agricultural by-products are being produced in huge quantities and can be applied to soil as a disposal strategy. The application of two different rates (2 and 8% w/w) of olive cake to a Mediterranean calcareous soil resulted in an increased sorption of four triazine herbicides, which was higher for the more hydrophobic compounds (terbuthylazine and prometryn) and lower for the more polar ones (simazine and cyanazine). However, when the sorption coefficients were normalised to the total soil organic carbon (Koc), the results did not significantly differ between simazine and cyanazine which is an indication that the olive cake did not exert different sorption capacity for both compounds. On the contrary, Koc values for terbuthylazine and prometryn increased in the amended soils. Our results from experiments using mixtures of several pesticides suggest that competition for sorption sites resulted in a decrease of herbicide sorption. Desorption was hysteretical both for the amended and unamended soils, but the addition of olive cake at the highest dose diminished desorption of most of the herbicides. In conclusion, the addition of olive cake behaves as a promising method for reducing the risk of groundwater pollution by pesticides. © 2007 Elsevier B.V. All rights reserved. Keywords: Soil; Amendments; Triazine herbicides; Sorption; Desorption; Competition
1. Introduction The olive tree (Olea europaea L.), is an important and characteristic cultivation of the Mediterranean area. The extraction of olive oil, which in the last years has introduced a two-stage centrifugation process because of its high efficacy and low energy and water consumption, produces some by-products which could be potentially used in crop management. These byproducts are of increasing interest as soil amendments, because of the low organic matter (OM) content of the Mediterranean soils together with some problems of erosion. The olive culture in Spain is of high socio⁎ Corresponding author. Tel.: +34 958 181600; fax: +34 958 129600. E-mail address: [email protected] (A. Peña). 0048-9697/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.scitotenv.2007.01.023
economical importance, since it constitutes 11% of the national agricultural surface, representing about 2.42 · 106 ha (http://www.mapya.es). In Andalusia, with ca. 60% of the total national production, olive oil in the 2002–2003 campaign accounted for ca. 680,000 Mg. Agricultural by-products, along with other waste disposal amendments of high OM content, have been applied to soils in the last decades to enhance their fertility and improve their physicochemical properties, such as soil porosity and aggregation (Cox et al., 1997; Sluszny et al., 1999). Moreover, the addition of exogenous material to soil (peat, manure, sewage sludge, compost, etc.), has been extensively used in the last few years to modify the fate of pesticides in the environment (Senesi et al., 1997; Cox et al., 1997; Albarrán et al., 2003).
120
L. Delgado-Moreno et al. / Science of the Total Environment 378 (2007) 119–123
Triazine herbicides have been widely found in groundwaters (Burkart et al., 1999; Battaglin et al., 2000) due to their moderate sorption by soil and relatively long persistence. Therefore, modification of triazine retention by amending soil with agricultural byproducts of high OM content has been undertaken in order to reduce their residual effect in the environment. 2. Materials and methods 2.1. Soil Soil was sampled from the upper layer (0–25 cm) of an olive orchard located near Iznalloz (province of Granada, South of Spain), air-dried and passed through a 2-mm sieve. It was a calcaric Regosol, with silty clay loam texture, with 34% clay, 56% loam and 10% sand, containing a 44.1% CaCO3 and with pH of 7.9.
15 °C for 15 min (Eppendorf 5810R). A one-mL portion was measured from the supernatant and analysed as described below. Desorption experiments were performed with the highest concentration by replacing a 10-mL aliquot of the supernatant with 10 mL of water. At each desorption step the soil in the tube was resuspended in solution using a vortex mixer and then the tubes were mechanically shaken end-over-end in a thermostatic chamber at 15 ± 1 °C, for 24 h. This desorption step was three fold repeated. The sorption– desorption experiments were described by the empirical Freundlich equation (X = Kf × Ce1/n), being X (μg g− 1) the amount of herbicide sorbed at the equilibrium concentration, Ce (mg L− 1), Kf the Freundlich sorption– desorption coefficient (L kg− 1) and 1 / n a constant which depends on the sorbate, the sorbent and temperature properties. 2.5. Extraction and analytical procedure
2.2. Organic amendments
The selected herbicide was terbuthylazine, 99.3% purity (Dr. Ehrenstorfer, Germany). It is a pre- or postemergence broad-spectrum herbicide to control weeds in many crops (vines, fruit trees, olive trees, etc.) (Tomlin, 2003). Apart from terbuthylazine, other triazine herbicides no longer permitted (cyanazine, simazine and prometryn, all with a purity N99%) were also included as model compounds.
A 1-mL aliquot of the herbicide mixture was vortexed (Reax 200, Heidolph, Kolheim, Germany) with 2 mL of hexane/toluene (1/1 v/v). Phase separation was achieved by sample freezing in a conventional freezer (− 18 °C) for 2 h. The organic phase was passed to a vial and analysed. Recoveries were of 89.2% for terbuthylazine and ranged from 73.2 to 92.1% for the other triazines. Samples were analysed by gas chromatography in a Varian Star 3400 CX equipped with a TSD detector and a Varian 8200 automatic injector. One μL of the sample was injected splitless, using He as carrier gas, on a HPUltra 2 column (cross-linked 5% phenyl methyl silicone) (25 m, 0.32 mm i.d., 0.17 mm film thickness). Injector and detector temperatures were 280 and 300 °C, respectively. Oven temperature was programmed starting at 45 °C (1 min), at 25 °C min− 1 until 160 °C and at 7 °C min − 1 until 210 °C (2 min). Under these experimental conditions retention time of terbuthylazine was 12.1 min and 11.6, 13.9 and 15.0 min, for simazine, prometryn and cyanazine, respectively.
2.4. Adsorption and desorption isotherms
3. Results and discussion
Isotherms were carried out using a batch equilibration method. Soil samples (5 g) were mechanically shaken end-over-end in a thermostatic chamber at 15 ± 1 °C with 20 mL of aqueous solutions of terbuthylazine and the other three herbicides each at 2, 5, 10, 15 and 20 mg L− 1, for 12 h. The experiments were run in duplicate with a control of the herbicides solution without soil. The slurry was centrifuged at 3000 rpm and
3.1. Sorption experiments
Olive cake (CA), from a two-stage centrifugation process, was air-dried and passed through a 2-mm sieve. It was kindly supplied by an olive oil industry (Romeroliva, Deifontes, Granada). Two dosages were applied to the soil, a low or agricultural dosage, equivalent to 50 Mg ha− 1 and a high dosage or dosage with remediation potential, at 200 Mg ha− 1. Amended soil had an OC content of 4.89 and 2.63%, and pH of 7.1 and 7.2 for the high and low dosage, respectively. 2.3. Herbicides
3.1.1. Non-amended soil Experimental data were fitted to the Freundlich model. Table 1 shows the values of Kf and 1 /n for the herbicides studied. Herbicide sorption on non-amended soil provided the lowest Freundlich sorption coefficient (Kf). However, since the sorption parameters are affected by
L. Delgado-Moreno et al. / Science of the Total Environment 378 (2007) 119–123
121
Table 1 Sorption parameters for the four triazine herbicides in non-amended and amended soil Adsorption parameters a
Desorption parameters a
2
c X10
Kf ± SDa
1 / n ± SDa
R2
Hd
De
Kf ± SD
1 / n ± SD
R
b Koc
1.8 ± 0.2 11.5 ± 0.7 2.2 ± 0.2
0.97 ± 0.05 0.93 ± 0.05 1.28 ± 0.06
0.95 0.95 0.97
82.3 247.8 137.5
17.0 98.6 41.2
15.4 ± 0.5 42.6 ± 1.0
0.22 ± 0.02 0.16 ± 0.02
0.91 0.86
0.23 0.17
36.0 23.7
3.2 ± 0.3 17.5 ± 0.3 6.2 ± 0.5
0.66 ± 0.04 0.64 ± 0.04 0.83 ± 0.05
0.94 0.97 0.96
71.7 275.4 197.4
14.5 76.4 42.1
10.4 ± 0.5 35.1 ± 0.6
0.29 ± 0.03 0.21 ± 0.01
0.84 0.93
0.44 0.33
45.0 20.1
0.7 ± 0.1 1.4 ± 0.2 0.8 ± 0.1
0.86 ± 0.07 0.95 ± 0.09 1.11 ± 0.05
0.92 0.87 0.98
22.4 26.6 42.1
5.2 12.8 10.9
4.1 ± 0.2 5.6 ± 0.3
0.36 ± 0.03 0.41 ± 0.03
0.87 0.90
0.41 0.43
57.4 59.4
0.7 ± 0.1 1.3 ± 0.1 0.8 ± 0.1
1.00 ± 0.07 1.05 ± 0.05 1.20 ± 0.04
0.94 0.98 0.99
30.1 39.8 41.4
7.3 14.0 12.2
3.9 ± 0.1 10.5 ± 0.2
0.38 ± 0.02 0.31 ± 0.01
0.96 0.98
0.38 0.30
55.7 56.8
TBTA S SCA (8%) SCA (2%) PROM S SCA (8%) SCA (2%) SIM S SCA (8%) SCA (2%) CYAN S SCA (8%) SCA (2%)
Standard error of two duplicates. bCalculated as Koc = Kd (Ce = 10 mg L− 1) ⁎ 100 / %OC. cObtained from Freundlich equation. dHysteresis coefficient (1 / ndes / 1 / nads). ePercentage of herbicide desorbed. TBTA: terbuthylazine, PROM: prometryn, SIM: simazine, CYAN: cyanazine. S: non-amended soil, SCA: amended soil with olive cake at two application dosages: 8% (200 Mg ha− 1) and 2% (50 Mg ha− 1).
a
the slope (1 / n), for comparison purposes we have preferred to compare the sorbed quantity at 10 mg L− 1 (X10). According to these values, sorption increased in the order simazine ≈ cyanazine b terbuthylazine ≈ prometryn (Fig. 1), higher for the most hydrophobic herbicides terbuthylazine and prometryn (Log Kow 3.2 and 3.1), and lower for the less hydrophobic ones, simazine and cyanazine (Log Kow 2.1) (Tomlin, 2003). The values of organic C-normalised sorption coefficients (Koc values in Table 1) obtained for the different triazine herbicides were 22–30 for simazine and cyanazine, and lower than 100 for terbuthylazine and prometryn. Other reports in the literature have cited Koc b 100 or around 100 for simazine and cyanazine (Liu and Qian, 1995; Barriuso et al., 1997; Oliveira et al., 2001) and values of 200 for prometryn and 230–300 for terbuthylazine (Businelli, 1997; Sluszny et al., 1999). The lower values encountered here could be explained by competition for sorption sites, because of the simultaneous sorption of the four herbicides. The phenomenon of competition has been described for triazine herbicides (Xing et al., 1996; Turin and Bowman, 1997; Businelli, 1997) and for other contaminants (Martins and Mermoud, 1998) leading to non-linear isotherms and to the lowering of the sorption constants (Kf or Koc). This is especially the case when chemicals with similar structure compete for the same sorption sites (Xing et al., 1996). Special care should be taken to avoid the simultaneous application of herbicides because a decrease in soil retention can occur (Table 1).
3.1.2. Addition of olive wastes to soil The addition of olive cake to soil resulted in increased sorption, which was higher for the highest dose (8%) (Fig. 1). In amended soils the highest sorption corresponded again to terbuthylazine and prometryn, revealing the affinity of these herbicides for the organic amendment (Table 1). In accordance with Barriuso et al. (1997) differences in sorption between amended and non-amended soils were found to be negligible for the most polar compounds with Koc values lower than 50 L kg− 1 C, such as simazine and cyanazine (Table 1). For more hydrophobic compounds, such as terbuthylazine and prometryn, Koc values increased when the soil was amended. This fact is an indication that the OM derived from the olive cake has a greater sorption capacity for both compounds. This could be due to the different nature of the organic fractions present in olive cake when compared to that from natural soil (Chefetz et al., 2004). The addition of amendments increased significantly Koc values which approached values reported in the literature, especially for prometryn and terbuthylazine. Competition for sorption sites in amended soil may have diminished since additional sites were probably available. Therefore, the compounds more tightly bound to the soil matrix would displace those less retained, such as cyanazine and simazine, for which no increase in Koc values was noticed.
122
L. Delgado-Moreno et al. / Science of the Total Environment 378 (2007) 119–123
Fig. 1. Sorption–desorption isotherms (amount of herbicide sorbed (X) vs. concentration of herbicide in solution (Ce)) of simazine, cyanazine, prometryn and terbuthylazine in non-amended (♦) and amended soil (8%) (■) (solid symbols — sorption points, open symbols — desorption points).
Since the triazines are weak bases (Table 1), they could be in one of two forms depending on the pH: the ion or the neutral molecule, and the sorbability of these two forms can differ to a large extent. Concerning the pKa of the herbicides, the three Cl-triazines (pKa 0.6–2) could be considered neutral herbicides in soil since their pKa is ≪ pH − 1 (Green and Karickhoff, 1990). Prometryn, like other 2-methylthiotriazines, such as terbutryn or ametryn, is a less weak base (pKa 4.1) and could be partly protonated when soil was added with olive cake and the pH was lowered to 7.05. This would explain the highest increase in Koc values due to the existence of additional mechanisms, such as cation exchange. 3.2. Desorption experiments The desorption process from natural soil was hysteretical for all herbicides studied (Table 1). This was an indication that desorption was not reversible, as can be seen in the percentage of herbicide desorbed (D) after four desorption steps (Table 1) and in the coefficient of hysteresis (H ). This fact contrasts with the results obtained by Albarrán et al. (2003) who showed negative hysteresis of simazine in unamended soils and amended with olive cake, which pointed to a high reversibility of the sorption–desorption process.
Simazine and cyanazine reached desorption percentages of more than 50% of their initial concentration, but for prometryn and terbuthylazine only 35–45% was desorbed. The addition of olive cake increased the retention of these herbicides to the soil since the final desorbed quantity was around 25% for terbuthylazine and prometryn. Therefore their leaching and consequently their ability to reach ground water were greatly diminished. In contrast, no differences, between soil and soil amended with olive cake, were observed in percentage values of simazine and cyanazine desorbed. This fact follows the same trend observed for Koc values and indicates that less hydrophobic compounds, such as simazine and cyanacine, were more weakly bound to soil than the most hydrophobic ones, terbuthylazine and prometryn. 4. Conclusions Olive cake addition to agricultural soil increases sorption of all triazine herbicides tested and reduces desorption of the most hydrophobic ones, terbuthylazine and prometryn. Both processes favour the reduction of their environmental impact and may attenuate the leaching of triazine herbicides to groundwater. Treating soils with olive cake could also be considered as a
L. Delgado-Moreno et al. / Science of the Total Environment 378 (2007) 119–123
disposal strategy for these residues, which are produced in huge quantities in the Mediterranean region, both to enhance the fertility and improve the physicochemical properties of the soil. The sorption behaviour is also related with the physicochemical characteristics of the compounds, and increase with the hydrophobic character of the herbicides. Further research is required to establish the organic fraction responsible for the pesticide–substrate interaction. Care should be taken with the simultaneous application of different herbicides, since competition among them can reduce their retention by soil. Acknowledgements LDM thanks the Spanish Ministry of Education for a FPU grant received. The Project CAO001-007 from Junta de Andalucía has supported this study. References Albarrán A, Celis R, Hermosín MC, López-Piñero A, Ortega-Calvo JJ, Cornejo J. Effects of solid-mill waste addition to soil on sorption, degradation and leaching of the herbicide simazine. Soil Use Manage 2003;19:150–6. Barriuso E, Houot S, Serra-Wittling C. Influence of compost addition to soil on the behaviour of herbicides. Pestic Sci 1997;49:65–75. Battaglin WA, Furlong ET, Burkhardt MR, Peter CJ. Occurrence of sulfonylurea, sulfonamide, imidazolinone, and other herbicides in rivers, reservoirs, and ground water in the Midwestern United States, 1998. Sci Total Environ 2000;248:123–33. Burkart MR, Kolpin DW, Jaquis RJ, Cole KJ. Agrichemicals in ground water of the Midwestern USA: relations to soil characteristics. J Environ Qual 1999;28:1908–15.
123
Businelli D. Pig slurry amendment and herbicide coapplication effects on s-triazine mobility in soil: an adsorption–desorption study. J Environ Qual 1997;26:102–8. Chefetz B, Bilkis YI, Polubesova T. Sorption–desorption behavior of triazine and phenylurea herbicides in Kishon river sediments. Water Res 2004;38:4383–94. Cox L, Celis R, Hermosin MC, Becker A, Cornejo J. Porosity and herbicide leaching in soils amended with olive-mill wastewater. Agric Ecosyst Environ 1997;65:151–61. Green RE, Karickhoff SN. Sorption estimates for modelling. In: Cheng HH, editor. Pesticides in the soil environment: processes, impacts and modelling, vol. 2. Soil Sci Soc America, WI, Book Series; 1990. p. 79–101. Chapter 4. Liu J, Qian C. Hydrophobic coefficients of s-triazine and phenylurea herbicides. Chemosphere 1995;31:3951–9. Martins JM, Mermoud A. Sorption and degradation of four nitroaromatic herbicides in mono and multi-solute saturated/unsaturated soil batch systems. J Contam Hydrol 1998;33:187–210. Oliveira Jr RS, Koskinen WC, Ferreira FA. Sorption and leaching potential of herbicides on Brazilian soils. Weed Res 2001;41:97–119. Senesi N, La Cava P, Miano TM. Adsorption of imazethapyr to amended and nonamended soils and humic acids. J Environ Qual 1997;26:1264–70. Sluszny C, Graber ER, Gerstl Z. Sorption of s-triazine herbicides in organic matter amended soils: fresh and incubated systems. Water Air Soil Pollut 1999;115:395–410. Tomlin CDS. The pesticide manual. 13rd edn. Alton, Hampshire, UK: British Crop Protection Council; 2003. 1344 pp. Turin HJ, Bowman RS. Sorption behavior and competition of bromacil, napropamide, and prometryn. J Environ Qual 1997;26:1282–7. Xing B, Pignatello JJ, Gigliotti B. Competitive sorption between atrazine and other organic compounds in soils and model sorbents. Environ Sci Technol 1996;30:2432–40.