Study of cyantraniliprole sorption in soils of Russia

. The sorption of the new insecticide cyantraniliprole was studied by the batch equilibrium method. The study of quantitative characteristics of cyantraniliprole sorption by the soils of different types showed that this process is well described by linear isotherms. According to obtained values of K oc , cyantraniliprole belongs to medium-mobility active substances of pesticides. The sorption value of cyantraniliprole was closely correlated with the content of organic carbon and clay. The regression equation of the dependence of K d on soil organic carbon content was obtained.


Introduction
Cyantraniliprole is a new broad-spectrum insecticide. It is approved for use in Argentina, Canada, the USA, Great Britain and several EU countries. In the Russian Federation, cyantraniliprole was registered in 2019 as the active substance of insecticides recommended for pest control in plantings of cabbage, tomatoes, cucumbers and onions [1]. Laboratory filtration experiments showed high mobility of cyantraniliprole [2,3]. In the lysimeter experiment cyantraniliprole was found in the leachate 2 weeks after its application (after heavy showers), and also was found in most samples during the experiment [4].
Pesticide sorption by soil is considered the process that has the greatest impact on pesticide behavior and fate in the environment [5,6]. Sorption is а result of the interaction of pesticides with soil colloids, non-colloidal organic matter of the soil and soil solution, and it can be reversible and irreversible. The extent of pesticide sorption to soil is expressed by the distribution coefficients between the aqueous and soil phases (Kd). The distribution coefficient value varies widely with soil type and soil organic matter content, for this reason the sorption coefficient normalized to soil organic carbon content (Koc) is more often used [7].
Sorption indexes are necessary to assess the potential pesticide mobility in soil, and they are also widely used as input parameters to mathematical models of the pesticide behavior in soils. The accuracy of the prediction of pesticide migration in soil largely depends on the quality of input sorption data. Since in further studies it is planned to use the data of the sorption experiments for mathematical modeling of the behavior of cyantraniliprole in soils of Russia, the main goal of this work was to study the cyantraniliprole sorption in soils of the different climatic zones of the Russian Federation -Luvic Chernozem (Kursk region), Halpic Kastanozem (Saratov region) and Albic Glossic Retisol (Moscow, large lysimeter of Moscow State University) [9] and obtain quantitative characteristics of cyantraniliprole sorption.

Pesticide
Cyantraniliprole is a second-generation insecticide from class of anthranilamides, which is used on a wide range of crops and is effective against sucking pests and thrips. In terms of resistance it refers to low-persistent -persistent active substances: DT50 = 8.7 -91.9 days (DT50geomean lab. = 34.4 days), in terms of migration ability it refers to medium-mobility active substances: Koc = 157-367 (Koс average = 241 ml/g) [8].

Soils
The properties of soils are shown in Table 1.

Study of sorption kinetics for one concentration of сyantraniliprole in one soil
The sorption study was carried out in accordance with the OECD methodology [10]. A preliminary experiment was conducted to study the sorption kinetics of сyantraniliprole in order to determine the time period for establishing equilibrium in the soil solution system. At the first, 10 g of air-dried soil samples (Retisol, arable horizon) were sifted through a 1 mm sieve and placed in flasks, then 20 ml of 0.01M solution of CaCl2 were added and flasks were shaken for 4 hours before the experiment. After that, 5 ml of the test compound solution was added into the flasks to bring the final volume to 25 cm 3 and to obtain сyantraniliprole concentration of 0.1 μg/ml. Flasks were shaken for 16 hours.
Sampling was conducted after 0.1, 2, 4, 8, 12, 16 hours. The study was carried out in 2 replicates. The selected solutions were centrifuged at a speed of 4000 rpm for 15 min, equilibrium content of сyantraniliprole in the liquid phase was determined by high performance liquid chromatography (HPLC).
The solutions were prepared from an analytical standard containing 99.8% сyantraniliprole.

Study of сyantraniliprole sorption
At the first, 10 g of air-dried soil samples were sifted through a 1 mm sieve and placed in flasks, then 20 ml of 0.01M solution of CaCl2 were added and flasks were shaken for 4 hours before the experiment. After that, 5 ml of the test compound solution was added into the flasks to bring the final volume to 25 cm 3 and to obtain сyantraniliprole concentration of 0.5; 0.2; 0.1; 0.05 and 0.02 μg/ml. The flasks were placed on a mechanical shaker; samples were taken for analysis after 16 hours. The study was carried out in 2 replicates.

Determination of сyantraniliprole in soil
Quantitative determination of сyantraniliprole content in water and soil was carried out in accordance with the methodological guidelines [11]. Detection limits of сyantraniliprole were 0.5 μg/L and 2.5 μg/kg for water and soil respectively.

Construction of isotherms and calculation of sorption coefficients
The dependence of the sorption value of сyantraniliprole on its equilibrium concentration in solution was approximated by the equation of linear sorption isotherm: where Kdsorption coefficients, qconcentration of pesticide adsorbed on soil (μg/g), а Ce equilibrium concentration of pesticide in solution (μg/ml). Since the soil organic matter plays a dominant role in the pesticide sorption, the sorption coefficient is often normalized by the content of soil organic carbon and Koc coefficient is used for a comparative assessment of the sorption ability of soils, regardless of the content of organic matter in them [7]. Koc value was calculated as:

Koc = (Kd x 100) / Corg
( 2) where Сorgcontent of soil organic carbon (%). Moreover, the relationship between the concentration of adsorbed сyantraniliprole and its concentration in an aqueous solution was approximated by the Freundlich equation: where Kf and 1/n are the Freundlich coefficient and the Freundlich exponent, which indicate the affinity of the soil for the pesticide and the degree of linearity between the amounts adsorbed and the solution concentration, respectively.

Results and discussion
The preliminary experiment on determining the time to reach equilibrium in the soil-solution system for сyantraniliprole showed (Fig. 1) that after 8 hours of the experiment, the sorption rate decreases significantly, and 16 hours are enough to achieve the equilibrium of сyantraniliprole between solid and liquid phases. Concentration, μg/g

Time, h
The sorption of сyantraniliprole is well described by linear isotherms (Fig. 2), which allows to use a simple and universal index, the distribution coefficient (Kd). The values of Kd, as well as Freundlich coefficient (Кf) and Freundlich exponent (1/n) are shown in Table  2. The values of determination coefficient R 2 indicate that the accuracy of the description of the obtained data is higher when using the linear equation. For linear equation all values of R 2 are above 0.99.  The values of the sorption coefficient varied from 1.49 in the mixed horizon E-EB of Albic Glossic Retisol to 11.37 in the arable layer of Luvic Chernozem. Sorption coefficients normalized to soil organic carbon content (Koc) ranged from 185 (in the arable horizon of Halpic Kastanozem) to 387 ml/g (in the arable layer of Albic Glossic Retisol), which allowed сyantraniliprole to be classified as a medium-mobile pesticides [12]. These data generally correspond to the results of сyantraniliprole registration research conducted in EU (Kd = 2.51 -7.14, Koc = 157 -367 ml/g) [13], (Koс = 251 ml/g) [2]. All obtained Kd values are significant at a 0.01% significance level. The highest value of Koc was obtained for the arable horizon of Albic Glossic Retisol, this suggests that, in addition to organic matter, mineral part of soil is also involved in the sorption of the pesticide.
To determine the factors affecting the cyantraniliprole sorption, the regression equations were obtained (determination coefficients R 2 are given in Table 3). According to the Table 3, the sorption coefficient most depends on the content of organic matter (R 2 = 0.94) and clay (R 2 = 0.73) in the soil. The regression equation of the dependence of the sorption coefficient on the soil organic carbon content has the following form (R 2 = 0,94):

Conclusions
1. The study of quantitative characteristics of cyantraniliprole sorption by the different soils showed that this process is well described by linear isotherms (R 2 =0.99 for all soils). 2. According to obtained values of Koc, cyantraniliprole belongs to medium-mobile pesticides.
3. The sorption coefficient of cyantraniliprole was closely correlated with the content of organic carbon and clay.