Study of the degradation in aqueous solution of a refractory organic compound: avermectin type used as pesticide in agriculture

Mites, nematodes, fleas, ticks, lice, and flies are important vectors that transmit diseases to humans, animals, and agriculture worldwide. They also cause large economic losses in the quality and quantity of yields in the livestock industry and agricultural crops (Thanh et al. 2013; Khalil 2015). Due to serious and enormous economic damage caused by parasites species, their control is very important. Currently, the methods used for controlling these parasites are mainly based on the use of pesticides. Among these pesticides, the insecticide abamectin is a natural fermentation product of the soil bacterium Streptomyces avermitilis. Furthermore, abamectin is a blend of avermectins B1a and B1b (about 80% avermectin B1a and 20% avermectin B1b). These two components, B1a and B1b, have very similar biological and toxicological properties (Khalil 2015). Abamectin acts by stimulating the release of γ-aminobutyric acid thus causing paralysis by interrupting the nervous system of targeted insects (Kamel et al. 2007; Erramil et al. 2014). Avermectins have been used as a veterinary drug in livestock, to protect against a broad variety of parasites. In addition to veterinary applications, avermectin is also used to treat onchocerciasis in humans, a disease commonly known as river blindness (Raftery & Volz 2015). Abamectin was also registered as an agricultural insecticide, and it is the most active against the two-spotted spider mite, Tetranychus urticae Koch, which is one of the most destructive pests in ornamental, horticultural, and agricultural crops such as fruits and vegetables worldwide (Kwon et al. 2010). The mode of action of avermectins is not, however, specific to parasitic arthropods and nematodes, and consequently may affect other organisms in the environment (Shinawar Waseem et al. 2012). There are many reports on the persistence of abamectin in soil, soil-feces, and feces for an extended period of 14–70 days in concentrations high enough to exert toxic effects (Kolar et al. 2006). Photo-degradation kinetics of abamectin revealed very limited possibilities for photosensitized degradation of this compound induced by environmental light (Escalada et al. 2008). Furthermore, the excessive residue of abamectin can run off from the sites of application leading to the exposure of non-target aquatic organisms, showing adverse effects on the aquatic environment due to its high toxicity, even at very low concentrations (Shinawar Waseem et al. 2012). Nevertheless, all scientific papers show toxic effects of abamectin at very low concentrations. The LC₅₀ 96 h values for the different fish species of rainbow trout, bluegill sunfish, channel catfish, and carp are 3.2, 9.6, 24, and 40 mg L⁻¹, respectively. Likewise, in another report about toxicity of abamectin, it was found that values of lethal concentrations (LC₅₀ 96 h) of abamectin for different fish species, are near to those mentioned above, i.e. 3.0 mg L⁻¹ for rainbow trout and 4.8 mg L⁻¹ for bluegill sunfish (Vajargah et al. 2014).

In Brazil, the animal health market of avermectin-containing products generates annual sales of around one billion dollars (Raftery & Volz 2015). By 2011, the use of abamectin reached approximately 100,000 lb (c. 45359 kg) per year due to increased penetration of the cotton market and approval as a nematicide seed treatment on corn throughout the United States. The intensive use of pesticides, particularly abamectin in agriculture and in livestock, were a source of contamination of soil, ground water, rivers, lakes, rain water, and air worldwide (Erramil et al. 2014). This contamination has become a serious environmental problem. However, it has been shown that avermectin interferes with the growth of aquatic invertebrates even at concentrations below those expected to occur in this environment. Furthermore, as many organic compounds may resist conventional water treatment, their presence in water bodies used for water harvesting may also affect human health (Raftery & Volz 2015).

Removing abamectin by biological processes, the most cost-effective solution for wastewater treatment, does not always appear to be relevant for the toxicity of abamectin. By contrast, physical and physicochemical techniques have proved their efficiency for this purpose (Fourcade et al. 2012). Among the physical processes such as adsorption, coagulation, flocculation, electro-flocculation, reverse osmosis, ultra-filtration, these most conventional methods are non-destructive and merely transfer pollutants from one phase to another, which always results in secondary pollution. Physicochemical processes are destructive and are widely used to remove recalcitrant compounds (Fourcade et al. 2012). Among them, we can cite advanced oxidation processes (AOPs). Even though AOPs have been shown to be highly efficient, their operation is still quite expensive. An attractive option is a short AOP pretreatment which leads to the formation of biodegradable intermediates of the recalcitrant pollutants, which can be subsequently degraded in a biological process, with the aim of reducing energy costs (Ferrag-Siagh et al. 2014). Therefore, before the examination of the combination of an AOP with a conventional biological treatment for pollutant removal, the relevance of an AOP pretreatment has to be checked by monitoring some specific parameters (compound concentration or chemical oxygen demand (COD) and biodegradability of the pretreated pollutant solution through biological oxygen demand (BOD₅) measurements) (Ferrag-Siagh et al. 2014). The evolution of global parameters like COD, also provides useful information on oxidation, while the BOD₅ on COD ratio approximates effluent biodegradability, since a value of 0.4 is considered by several authors as the boundary of biodegradability (Pulgarin et al. 1999).

The degradation of ivermectin (IVM) in aqueous solution by photocatalysis with TiO₂ in suspension was evaluated. Photocatalysis with UVC (at 253.7 nm) and TiO₂ in suspension resulted in the degradation of 98% of IVM (500 μg L⁻¹) in water in 600 s.

EF treatment using a carbon-felt as cathode was considered in this study. The effects of some parameters such as the concentration of catalyst, temperature, current intensity, and initial abamectin concentration were considered.

To evaluate the main interactive effects of the studied variables on the oxidation of abamectin and to determine their optimal values in order to increase the yield of abamectin oxidation, a central composite design (CCD) (2⁴), was developed to predict the removal yield of this pollutant.

The aim of this study was, therefore, to examine the relevance of the EF process for the removal of abamectin and its possible combination with a biological treatment.

Commercial abamectin (Vertimec 1.8 Emulsifiable Concentrate (EC)) was obtained from Syngenta (Agadir, Morocco). The pH of the solutions was adjusted by the addition of sulfuric acid solution (H₂SO₄; 98% purity) or sodium hydroxide (NaOH; 99% purity) obtained from Biochem Chemopharma (Georgia, USA). Methanol (MeOH) of high performance liquid chromatography (HPLC) analysis grade (99.9% purity) was obtained from Sigma-Aldrich (Steinheim, Germany). Ultrapure water (18 MΩ cm of resistivity) was used for HPLC analysis in this study. Catalyst (FeSO₄, 7H₂O) and anhydrous sodium (Na₂SO₄) were of analytical grade and supplied by Panreac-Quimica-SA (Barcelona, Spain) and Biochem Chemopharma, respectively. Distilled water was used to prepare all solutions.

Electrolysis was carried out with a carbon-felt electrode of 25 cm² (5 × 5 cm) surface area and 1.27 cm thickness used as a cathode. The anode electrode was in a platinized titanium grid Ti/Pt of 12.5 cm² (5 × 2.5 cm) and 1 mm thick. The oxygen supply in the solutions was ensured by a diffuser of very fine bubbles of air using a sintered glass (4–8 μm of porosity).

All abamectin solutions were adjusted with pH = 2.9 using sulfuric acid (H₂SO₄) at 1 mol L⁻¹. The electrolyte (Na₂SO₄) (0.05 mol L⁻¹) was added for the solutions of abamectin to ensure good conductivity in the medium (6.5 mS cm⁻²) (Oturan 2013).

Samples were taken at intervals of regular time over a period of electrolysis of 5 h; the hydroxide of iron was precipitated using NaOH at 6 mol L⁻¹. The supernatant was recovered by centrifugation and was used for different chemical analyses. To consider the choice of the pH value, the electrolyte concentration and the stirring velocity were based on earlier studies which used the EF process for degradation of other insecticides and pesticides (Djoudi et al. 2007).

All samples were filtered through a 0.45 μm membrane filter for various chemical analyses (COD, total organic carbon (TOC), BOD₅, and HPLC).

To measure the COD we followed the ‘closed reflux, colorimetric method’ for the examination of water and wastewater. The oxidizable matter contained in a sample was oxidized by heating with backward flow in a strongly acid medium with a quantity of potassium bichromate in a closed test-tube. The oxygen uptake by the sample causes a change of color whose absorbance is proportional to the quantity of the reduced potassium bichromate and is measured in equivalents of oxygen.

We measured BOD₅ by following the manometric method, which is based on the principle of the Warburg respirometer, during which the breathing of the biomass is directly measured by ‘IS 602’ type apparatus. A sample was placed in an incubation bottle, connected to a pressure gauge with mercury which was closed with a stopper provided with a pressure pick-up (Oxytop). The selected volume is a function of the desired range of measurements. The measuring apparatus was placed in incubator at 20 °C. The oxygen being consumed by the bacteria for 5 days resulted in the decrease of pressure in the bottles. However, the oxidation of organic matter caused formation of carbon dioxide (CO₂), which was trapped by a pellet of KOH and created a vacuum in the bottle (Rodier 1996).

The derived organic matter after treatment of the solution was also quantified by TOC, using a carbon analyzer (LCK 385 TOC, Lange Hach). The two phases during the process are: total inorganic carbon, initially expelled using the agitator Fake-x5, then TOC, oxidized out of CO₂. The carbon dioxide passes through a membrane in the indicating tank, where it causes a change of color, which is evaluated using a photometer (Hach DR2800).

During electrolysis, high performance liquid chromatography (HPLC YL9100) followed the abamectin concentrations. The HPLC was equipped with a standard degasser (YL 9101), a manual injector (20 μL), two pumps (Model Analytical SS-1) and a detector with visible ultraviolet ray (UV/Vis detector YL9120). Analytical separation was performed using a column Teknokroma ODSA120-C18 (250 mm × 4.6 mm, particle size 5 μm) operated at 50 °C. The mobile phase consisted of MeOH–water (90:10, V/V) with a flow of 0.5 mL min⁻¹. For a linear baseline, the system was left to stabilize for 2–3 h. Abamectin was detected at 245 nm wavelength. To perform data processing and acquisition, we used YL Clarity-software.

An experiment was realized on the carbon-felt electrode (cathode) by a solution of abamectin (20 mg L⁻¹) under magnetic agitation with 1,000 rpm for 3 h. The samples were taken at two intervals of regular time in a period of 3 h for COD and Aba concentration analysis, to determine the carbon felt adsorption capacity on the Aba molecule.

In order to compare the EF process with simple electrochemical oxidation, a few experiments were realized to confirm the efficiency of the EF process on Aba degradation, and the oxidizing power of the OH^• on the organic matter present in the aqueous solutions. A first experiment was realized under the following conditions: [Aba]₀ = 9 mg L⁻¹, [Fe²⁺] = 0.15 mmol L⁻¹, I = 150 mA, and T = 50 °C, using a carbon-felt cathode and a platinized titanium grid anode. A second experiment was realized with an inert cathode of stainless steel and under the same conditions as the first experiment. A third experiment was also realized but without Fe²⁺ and under the same conditions as the first one. There was a final experiment without bubbling air but under the same conditions as the first experiment. Samples were taken every hour for 5 h of electrolysis and COD was measured in order to follow the evolution of the organic matter.

The use of a design experiment approach, and in particular the response surface methodology and the CCD, allowed us to calculate the optimal values of the operating parameters affecting the treatment of solutions of Aba by electrochemical process with a minimum number of experiment trials. This type of design comprises a two level factorial design (−1, +1), superimposed by center points (coded 0) (Abdel-Ghani et al. 2009; Benredouane et al. 2016).

The interactions are the driving force of many optimizations of the processes. Without the use of factorial experiments, some significant interactions can pass unperceived, and total optimization cannot be reached (Abdel-Ghani et al. 2009).

[COD]₀ and [COD]_t were, respectively, the initial COD of Aba and its concentration at a given time t (mg L⁻¹). The studied parameters were the initial abamectin concentration [Aba]_0, the catalyst concentration [FeSO₄, 7H₂O], the current intensity (I), and temperature (°C).

The original values of each factor and their corresponding levels are shown in Table 1; the selection of the levels of various factors was carried out because of a preliminary test and the results of previous research (Abdel-Ghani et al. 2009). The pH and time of electrolysis were fixed at 2.9 and 300 min, respectively.

The results of the second-order model obtained according to CCD are collected in Table 2. The model coefficients were estimated by standard least-square regression method using JMP 8.0.2 software (Chaudhary & Balomajumder 2014). This software was used to analyze the results and evaluate the adequacy of the model using Student's t-test, R² test, and Fisher's test.

To study the effect of temperature on the Aba degradation by an EF process, a kinetic study was performed under the following operating conditions during 1 h of electrolysis: 20 to 50 °C temperature, 9 and 20 mg L⁻¹ of initial Aba concentration, 150 mA of current intensity, 0.15 mmol L ⁻¹ catalyst concentration [Fe²⁺], 1,000 rpm stirring speed, and pH = 2.9.

At 50 °C temperature, Aba degradation of 94 and 79% is observed for an initial concentration of 9 and 20 mg L⁻¹, after 20 and 60 min of treatment by EF process (Figure 8(a) and 8(b)). This result proves that the reaction of Aba oxidation by the hydroxyl radicals is favored by the increase of temperature. Indeed, the temperature of solution influences the electron transfer rate and the mass transfer rate; Qiang et al. (2003) have shown that the regeneration of Fe²⁺ is promoted at high temperature. Yahiaoui et al. (2013) also explained in their paper ‘Combination of an electrochemical pre-treatment with a biological oxidation for the mineralization of non-biodegradable organic dyes: Basic yellow dye’ that an increase in temperature reduces the viscosity of the solution, which implies an increase of the apparent rate constant (K_app) and the apparent mass-transfer coefficient (K_m).

The high value of R² (Table 3) of the straight lines obtained by plotting Ln([Aba]₀/[Aba]_t) versus time at different temperatures, confirms the first-order kinetics of Aba degradation.

The removal efficiency increased from 20 to 45% in the first 20 minutes of treatment time when the current intensities decreased from 400 to 150 mA for 20 mg L⁻¹ of initial Aba concentration (Figure 12(b)).

The model equation obtained by CCD was used to determine the optimal values of the operating parameters so that we could obtain the highest COD abatement and reach the biodegradability threshold in a short processing time.

The biodegradability tests and the degradation of abamectin were realized on a real surface water solution (Table 4), after electrochemical treatment according to the EF process. In order to carry out these tests, we used the previously optimized operating conditions. For this purpose, a kinetic study of electrochemical degradation was realized on abamectin solution at 9 mg L⁻¹ (this is the concentration used by farmers) during 5 h of electrolysis. Other operating parameters were as follows: [Fe^{2 +}] = 0.15 mmol L⁻¹, I = 150 mA, and T = 47 °C.

Aqueous solutions with a concentration of 5.77 × 10⁻³ mM of abamectin, i.e., 10 mg L⁻¹, were treated in order to characterize as much as possible the reaction intermediates during degradation by the EF process under the optimal conditions.

The liquid chromatography-tandem mass spectrometry (LC-MS/MS) method was used in positive mode in order to identify the organic by-products resulting from degradation. Table 5 gives the main degradation by-products identified.

An examination of the results, presented in this study, reveals the following conclusions:

• The EF process compared to simple electrochemical oxidation is effective for degrading refractory molecules, but only after optimization.

• The use of an experimental design methodology allows the determination of the optimum conditions for degradation, while reducing the maximum number of experiments to be realized. The optimization of degradation conditions gives the following optimal conditions: T = 47 °C, I = 150 mA, catalyst = 0.15 mmol L⁻¹, which leads to the elimination of 58.14, 47.85, and 37.4% of total organic carbon for concentrations of 9, 20, and 40 mg L⁻¹, respectively.

• An increase in temperature leads to an increase in the speed of degradation of abamectin.

• A biodegradability test was realized on a solution electrolyzed with T = 47 °C, I = 150 mA, catalyst = 0.15 mmol L⁻¹, [Aba]₀ = 9 mg L⁻¹; the ratio of BOD₅/COD increased appreciably from 0.0584 to 0.64 after 5 h of degradation. Thus, in order to maintain a significant content of residual organic matter for the bacteria, 4 h of electrolysis proves more relevant, leading to a ratio BOD₅/COD of 0.433. This improvement of biodegradability was confirmed by the elimination of 78 and 80% of total organic carbon (COT) after 4 and 5 h of electrolysis, respectively.

• A biodegradability test on a real solution leads to similar results as the aqueous solution.

• A biological treatment can be combined with the EF process to ensure total mineralization.