- Email: [email protected]
Voltammetric analysis of mancozeb and its degradation product ethylenethiourea
Journal of Electroanalytical Chemistry 758 (2015) 54–58
Contents lists available at ScienceDirect
Journal of Electroanalytical Chemistry journal homepage: www.elsevier.com/locate/jeac
Voltammetric analysis of mancozeb and its degradation product ethylenethiourea Olalla López-Fernández a,b, M. Fátima Barroso a,⁎, Diana M. Fernandes c, Raquel Rial-Otero b, Jesús Simal-Gándara b, Simone Morais a, Henri P.A. Nouws a, Cristina Freire c, Cristina Delerue-Matos a a b c
REQUIMTE/LAQV, Instituto Superior de Engenharia do Porto, Instituto Politécnico do Porto, Rua Dr. António Bernardino de Almeida 431, 4200-072 Porto, Portugal Nutrition and Bromatology Group, Analytical and Food Chemistry Department, Faculty of Food Science and Technology, University of Vigo, Ourense Campus, E-32004 Ourense, Spain REQUIMTE/LAQV, Dep. de Química e Bioquímica, Faculdade de Ciências, Universidade do Porto, Rua do Campo Alegre s/n, 4169-007 Porto, Portugal
a r t i c l e
i n f o
Article history: Received 4 June 2015 Received in revised form 14 August 2015 Accepted 20 August 2015 Available online 24 August 2015 Keywords: Mancozeb Ethylenethiourea Cyclic voltammetry Square-wave adsorptive stripping voltammetry
a b s t r a c t The purpose of this work was to develop a reliable alternative method for the determination of the dithiocarbamate pesticide mancozeb (MCZ) in formulations. Furthermore, a method for the analysis of MCZ's major degradation product, ethylenethiourea (ETU), was also proposed. Cyclic voltammetry was used to characterize the electrochemical behavior of MCZ and ETU, and square-wave adsorptive stripping voltammetry (SWAdSV) was employed for MCZ quantification in commercial formulations. It was found that both MCZ and ETU are irreversibly reduced (−0.6 V and −0.5 V vs Ag/AgCl, respectively) at the surface of a glassy carbon electrode in a mainly diffusion-controlled process, presenting maximum peak current intensities at pH 7.0 (in phosphate buffered saline electrolyte). Several parameters of the SWAdSV technique were optimized and linear relationships between concentration and peak current intensity were established between 10–90 μmol L−1 and 10–110 μmol L−1 for MCZ and ETU, respectively. The limits of detection were 7.0 μmol L−1 for MCZ and 7.8 μmol L−1 for ETU. The optimized method for MCZ was successfully applied to the quantification of this pesticide in two commercial formulations. The developed procedures provided accurate and precise results and could be interesting alternatives to the established methods for quality control of the studied products, as well as for analysis of MCZ and ETU in environmental samples. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Dithiocarbamates (DTCs) are important organosulfur compounds discovered in the 1930s, usually used as pesticides to treat a wide variety of fungal diseases in plants [1,2]. Due to their high activity and low production costs, they have been applied on a large scale throughout the world on several types of crops. Furthermore, due to the possibility of their combination with new generation systemic fungicides these compounds are quite successful for the management of disease resistance, especially in horticultural crops [3,4]. Based on their carbon skeleton, DTCs can be categorized into three subclasses: dimethyldithiocarbamates (DMDs), ethylenebis(dithiocarbamates) (EBDCs), and propylenebis(dithiocarbamates) (PBDs). They are mainly complexed with transition metals (e.g., manganese or zinc) [5]. DTCs are medium to highly toxic substances, depending on their structure [6]. Mancozeb (MCZ; manganese ethylenebis(dithiocarbamate) (polymeric) complex with zinc salt), a broad spectrum fungicide with multi-site contact activity, is one of the most widely used protective fungicides in the world and constitutes 50% of the market share of the ⁎ Corresponding author. E-mail address: [email protected] (M.F. Barroso).
http://dx.doi.org/10.1016/j.jelechem.2015.08.030 1572-6657/© 2015 Elsevier B.V. All rights reserved.
EBDC fungicides [5,7]. MCZ is used to protect fruits and vegetables from foliar diseases and is also employed in the protection of such products during their storage and transportation. MCZ strongly adsorbs to soil particles and usually does not move below the upper layer of the soil [6]. MCZ is easily degraded to ethylenethiourea (ETU) in the presence of moisture or oxygen and biological systems [8,9]. This degradation product is more toxic than the parent compound. ETU is suspected to cause thyroid and neurotoxic effects and induces carcinogenesis, mutagenesis, and teratogenesis [10,11]. The analysis of DTCs is complex because of their insolubility in most solvents, their ability to form stable complexes with a variety of metal ions and their instability, which is promoted by oxygen, moisture, pH, temperature, and especially plant components [8,9]. There are several analytical methods available for DTC determination based on the use of spectrophotometry [12], capillary electrophoresis [8,13], gas chromatography [4,14], high-performance liquid chromatography [10,14,15], DART-TOF mass spectrometry [16], (adsorptive-stripping) voltammetry [2,17], batch injection analysis with amperometric detection [18], enzyme-linked immunosorbent assay [19], and flow injection-Fourier transform infrared spectrometry [20]. Many of these methods are based on the acid hydrolysis of DTCs in the presence of tin(II) chloride and the subsequent analysis of CS2,
O. López-Fernández et al. / Journal of Electroanalytical Chemistry 758 (2015) 54–58
H2S, or amines [8]. However, these methods present some disadvantages because they are time consuming, expensive and environmentally unfriendly. Although electrochemical determinations of DTC have been described [2,17], they are based on the use of mercury electrodes. Mercury is toxic and has negative effects on the environment and human health. Therefore, its use constitutes a major drawback. The aim of this work was to develop a simple, fast and cost-effective electrochemical method for MCZ and ETU detection and quantification in real samples. This proposed voltammetric method is environmentally friendly and is based on the study of the electrochemical reduction of MCZ and ETU on a glassy carbon electrode (GCE) surface. Performance characteristics such as linearity, limits of detection and quantification, precision, and accuracy were evaluated. This alternative methodology was applied to the analysis of MCZ in two commercial fungicide formulations available in Portugal and Spain (Micene WP® (Sipcam) and Mancozan® (Bayer CropScience)). 2. Experimental part 2.1. Reagents and solutions Mancozeb (96.8%), ethylenethiourea (99.9%) and potassium hexacyanoferrate(III) (99%) were purchased from Sigma-Aldrich (Germany). Ethylenediaminetetraacetic acid disodium salt 2-hydrate (EDTA, ≥99%) was obtained from Panreac (Spain) and used to dissolve MCZ. For the voltammetric analyses phosphate buffered saline (PBS) solutions (pH 5.0–9.0) were used. These buffers were prepared using potassium dihydrogenphosphate (99%, Riedel-de Haën, Germany), sodium chloride (99.5%, Panreac, Spain), potassium chloride (99.5%, Riedel-de Haën, Germany), and di-sodium hydrogenphosphate 7hydrate (Riedel-de Haën, Germany). Hydrochloric acid (37%, Scharlau, Spain) and sodium hydroxide (Pronalab, Portugal) solutions (both at 0.1 mol L−1) were used to adjust the pH of the solutions. The fungicide formulations Micene WP® (Sipcam, Spain) and Mancozan® (Bayer CropScience, Portugal), both containing 80% (w/w) of MCZ, were used to evaluate the accuracy of the voltammetric method. Because of the very low solubility of MCZ and ETU in water, EDTA was used to aid dissolution. Stock solutions of MCZ (1000 μmol L−1) and ETU (1000 μmol L−1) were prepared daily by dissolving an accurately weighed amount of the compound in an EDTA solution (3%) and ultrapure water, respectively. These solutions were stored in the dark at 4 °C until use. Solutions of the commercial products Micene WP® and Mancozan® were also prepared by dissolving an accurately weighed amount in a 3% EDTA solution. All solutions were prepared using ultra-pure water (resistivity = 18.2 MΩ cm) obtained from a Simplicity 185 water purification system (Millipore, France). 2.2. Equipment All voltammetric measurements were performed using a computercontrolled Autolab PGSTAT12 potentiostat/galvanostat (MetrohmAutolab, The Netherlands) and a Metrohm 663 VA Stand, containing a three-electrode cell composed of a glassy carbon electrode (GCE; working electrode), an Ag/AgCl (KCl 3 mol L− 1) reference electrode and a glassy carbon auxiliary electrode (Metrohm, Switzerland). The system was controlled by means of the General Purpose Electrochemical System (GPES) software package (v. 4.9, Metrohm-Autolab, The Netherlands). The GCE was manually cleaned before each measurement by polishing its surface with Micropolish Alumina (0.05 μm, Buehler, Germany) until a shining surface was obtained. The electrode was then rinsed with ultra-pure water before analysis. pH measurements were made with a Crison GLP-22 pH-meter and a combined glass electrode.
55
2.3. Voltammetric analysis The voltammetric studies were conducted in PBS buffers (pH 5.0– 9.0) using either cyclic voltammetry (CV) or SWAdSV. In these assays the GCE was immersed in a low-volume cell containing 1.0 × 10−3 L of PBS buffer. This cell was then inserted in a larger cell containing the reference and auxiliary electrodes and 40.0 × 10−3 L of the PBS buffer. Prior to the analysis, electrolytes were purged with nitrogen for 10 min to ensure complete oxygen removal. In the CV and SWAdSV assays, the potential was scanned between − 0.20 and −1.0 V for MCZ, and between 0.20 and − 1.2 V for ETU. In all the SWAdSV assays a step potential (ΔEs) of 0.0051 V and a pulse amplitude (ΔEp) of 0.01995 V were used. For the optimization of the SWAdSV procedure, the following parameters were studied: electrolyte pH, accumulation potential (Eacc), accumulation time (tacc) and square-wave frequency (f). In Table 1 the used conditions in the optimization studies are shown. In the optimized SWAdSV procedure accumulation potentials of − 0.50 V for MCZ and − 0.10 V for ETU, a tacc of 20 s, and an f of 100 Hz were used. 3. Results and discussion 3.1. Cyclic voltammetric studies Cyclic voltammograms of 200 μmol L−1 MCZ and ETU solutions in PBS buffer (pH 7.0) at different scan rates (ν) showed a single irreversible reduction process for both compounds at −0.6 V for MCZ and −0.5 V for ETU (Fig. 1). The peak potentials were found to be pH-independent (data not shown). The influence of ν on the peak current intensity (ip) was studied from 0.010 to 0.500 V s−1 (Fig. 1). Within this interval, linear relationships between ip and ν1/2 were established for MCZ: − ip (μ A) = 0.169 × ν1/2 (mV s−1)1/2 + 0.742 (R = 0.973) and for ETU: − ip (μ A) = 0.163 × ν1/2 (mV s−1)1/2 + 0.566 (R = 0.994). These equations indicate that the reduction processes of both compounds at the surface of the GCE are mainly controlled by diffusion [21]. 3.2. Square-wave adsorptive stripping voltammetric studies After the characterization of the electrochemical reduction behavior of MCZ and ETU, a more sensitive technique, SWAdSV, was used to develop an alternative procedure for their detection and quantification. For this purpose, the influence of several analytical features such as electrolyte pH, square-wave frequency ( f), Eacc and tacc on ip and the peak width and shape was evaluated and optimized (Table 1). Fig. 2 shows the effects of these variables (univariable study) on the ip values of 50 μmol L−1 MCZ and ETU solutions. For the optimization of the pH of the electrolyte (PBS buffer), several MCZ and ETU solutions with pH values ranging between 5.0 and 9.0 were analyzed (Fig. 2). It was found that the reduction peak potential was independent of pH and that the highest ip values were obtained at pH 7.0. Therefore, this pH was selected for subsequent experiments. The influence of f on ip and peak width and shape was evaluated from 10 to 150 Hz. The ip increased with frequency up to 100 Hz for MCZ and up to 150 Hz for ETU (Fig. 2). However for ETU the peak
Table 1 Parameters used in the optimization of the SWAdSV procedure. Magnitude optimized
MCZ Eacc (V)
Eacc tacc f pH a
Varied.
ETU tacc (s)
a
5
−0.5 −0.5 −0.5
a
20 20
f (Hz)
Eacc (V)
tacc (s)
f (Hz)
10 10
a
5
−0.1 −0.1 −0.1
a
10 10
a
100
20 20
a
100
56
O. López-Fernández et al. / Journal of Electroanalytical Chemistry 758 (2015) 54–58
Fig. 1. Cyclic voltammograms of MCZ (A) and ETU (B) solutions (200 μmol L−1; pH = 7). Scan rates (v s−1) A: 0.010, B: 0.020, C: 0.050, D: 0.100, E: 0.150, F: 0.200, G: 0.300, and H: 0.500.
became larger and less accentuated using this latter frequency. Therefore, and to make a compromise between sensitivity and selectivity, 100 Hz frequencies were used for both compounds in the subsequent studies. The influence of Eacc on ip was studied in the range from − 0.60 to 0.00 V for MCZ and − 0.50 to 0.00 V for ETU (Fig. 2). Although ip was not maximum for MCZ an Eacc of −0.50 V was chosen for further assays. For ETU the Eacc selected was −0.10 V. The effect of tacc (between 0 and 30 s) on ip was also investigated (Fig. 2). The ip values of the reduction of MCZ increased with tacc, observing a maximum at 30 s. Because no significant differences were seen between 20 s and 30 s, the shortest time was chosen. For ETU, a maximum at tacc = 20 s was obtained. Therefore, all the analyses were carried out with a 20 s accumulation time. With the optimized experimental conditions, several figures of merit [22,23] of the developed voltammetric methods were calculated (Table 2). As can be seen in this table linear relationships and good correlations between ip and concentration were obtained between 10.0–90.0 μmol L−1 and 10.0–110.0 μmol L−1 for MCZ and ETU, respectively. Typical voltammograms and the corresponding calibration lines for both compounds in the linear range are shown in Fig. 3. Each point of the calibration curve is the average ip value of three independent measurements. From the data shown in Table 2 it can also be concluded that (i) the method for MCZ analysis is more sensitive than the one for
ETU, (ii) both methods are precise (Vx0 b 5%) and (iii) the limits of detection (LODs) for both compounds are in the low μmol L−1 range. The precision of the results was studied by intra- and inter-day assays at two different concentration levels (50 μmol L− 1 and 70 μmol L−1) and was determined by calculating relative standard deviations (RSDs). For the intra-day studies, each concentration was analyzed in triplicate on three occasions along each working day. For inter-day studies the analyses were performed over a period of 1 week. No significant differences were found between intra-day and inter-day experiments with RSD values ranging between 2.2 and 10.1%, indicating the high precision of the voltammetric methods. For MCZ analysis in the selected commercial products, matrixinduced suppression or enhancement effects were evaluated by comparing the signals of standard MCZ solutions (concentrations between 10 and 90 μmol L−1) with those obtained for the same (final) concentrations of MCZ added to 20 μmol L−1 solutions of the commercial products (Micene WP® or Mancozan®). No significant differences were observed between the slopes of the calibration curves and the standard addition curves (p N 0.05) (ratios between the slope of the calibration curve and the standard addition curve were 1.03 and 1.07 for Micene WP® and Mancozan®, respectively) which indicate the absence of matrix effects for both commercial products. Average recoveries within the studied range were 91.6 ± 4.9% (n = 5) for Micene WP® and 100 ± 13%
Fig. 2. Univariable optimization of pH, frequency ( f), accumulation potential (Eacc), and accumulation time (tacc) using 50.0 μmol L−1 MCZ (A) and ETU (B) solutions. The different conditions studied for each variable are indicated on the X-axis.
O. López-Fernández et al. / Journal of Electroanalytical Chemistry 758 (2015) 54–58 Table 2 Figures of merit of the developed SWAdSV methods for the analysis of MCZ and ETU. Figure of merit
MCZ
ETU
Linear range (μ mol L−1) n Correlation coefficient (r) Slope (m) (μ A/μ mol L−1) Standard deviation of the slope (Sm) (μ A/μ mol L−1) Intercept (a) (μ A) Standard deviation of the intercept (Sa) (μ A) Standard deviation of the linear regression (Sy/x) Standard deviation of the method (Sx0) Coefficient of variation of the method (Vx0) (%) Limit of detection (LOD) (μ mol L−1) Limit of quantification (LOQ ) (μ mol L−1)
10.0–90.0 9 0.997 0.0186 0.0006 0.091 0.032 0.044 2.3 4.7 7.0 23
10.0–110 7 0.998 0.0146 0.0004 0.017 0.027 0.038 2.6 4.8 7.8 26
(n = 5) for Mancozan®, indicating that the method provides accurate results. Furthermore, due to the absence of matrix effects the calibration curve can be used for the quantification of MCZ in the two commercial products instead of the standard addition method. This allows the analysis of more samples in a single run. Studies about the determination of DTCs by voltammetry are scarce in the literature. Amorello and Orecchio [2] proposed a method for the determination of nine DTCs (mancozeb, maneb, propineb, nabam, Na(CH3)2DTC, zineb, ziram, ferbam and thiram) by adsorptivestripping voltammetry in simulated pesticide formulations using a hanging mercury drop working electrode. The recovery and precision obtained for mancozeb were similar to those reported in this work (N 95% and RSD = 8.1%, respectively). In the previous work an LOD of 14 nmol L−1 was achieved, which is much lower than the one obtained in this study. This is mainly because of the use of much longer tacc (120 s). Carvalho and coworkers [24] determined ETU in water samples by cathodic stripping voltammetry, also using a hanging mercury drop electrode. The recoveries obtained varied between 93 and 110% and the RSD was 1.9%. The LOD obtained, after a tacc of 300 s, was also much lower (14 nmol L− 1) than the one obtained in the present work. In any case, the proposed method is more environmentally friendly than the other published methods due to the use of a GCE instead of a mercury electrode. 3.3. Analytical application In order to assess the applicability of the developed method, two different batches of the commercially available products were analyzed.
57
Different solutions of the products Micene WP® and Mancozan® were prepared to obtain concentrations of 50 and 70 μmol L−1 MCZ. These solutions were analyzed by the developed method using a (direct) calibration curve. The results are summarized in Table 3. The data show a good agreement between the labeled and the experimental values for each product for both concentration levels, demonstrating the accuracy of the procedure. Also, the RSDs were low (b4%), proving the precision of the developed method for the analysis of MCZ in the commercial products. There are hardly any studies on the determination of MCZ by voltammetry in formulations. The method proposed by Amorello and Orecchio [2] for the determination of different dithiocarbamates by adsorptive-stripping voltammetry was not applied to the quality control of pesticide formulations. Silva et al. [18] determined the concentration of MCZ in insecticide samples using batch injection analysis with pulsed amperometric detection on boron-doped diamond electrodes. Recoveries and the RSD were similar to those found in our method (95%, RSD b 2%). 4. Conclusions As a response to the rising demand of analytical methods, which should be able to detect and quantify agrochemicals, and to the growing environmental interest, a validated approach to regulatory laboratories for routine analysis of mancozeb or ethylenethiourea is offered. Furthermore, the studies indicate that MCZ can be quantified accurately, precisely and rapidly in pesticide formulations by SWAdSV with a GCE in PBS (pH 7.0) buffer. The proposed methods provide economic and environmental advantages, when compared to other previously published methods. The toxicity and/or the waste volume are much lower than that generated by the other methods such as electroanalysis (mainly based on mercury electrodes) or HPLC (large volumes of mobile phases are used). Acknowledgments O. López-Fernández would like to thank her predoctoral fellowship and her grants for mobility research staff from the University of Vigo. M.F. Barroso and D.M. Fernandes are grateful to Fundação para a Ciência e a Tecnologia (FCT) for their post-doc fellowships (SFRH/BPD/78845/ 2011 and SFRH/BPD/74877/2010), financed by POPH-QREN-Tipologia 4.1-Formação Avançada, subsidized by Fundo Social Europeu and Ministério da Ciência, Tecnologia e Ensino Superior. This work received financial support from the European Union (FEDER funds through COMPETE) and national funds (FCT) through project UID/QUI/50006/2013.
Fig. 3. Typical SWAdSV signals in the linear range obtained for (A) MCZ (10, 30, 50, 70, and 90 μmol L−1) and (B) ETU (20, 50, 70, 90 and 110 μmol L−1) solutions (at pH 7.0 in PBS buffer) using the optimized conditions (increase of −ip with increasing concentration). Inset: calibration curves.
58
O. López-Fernández et al. / Journal of Electroanalytical Chemistry 758 (2015) 54–58
Table 3 Results obtained in the analysis of MCZ in commercial pesticide formulations. Product
Analyzed solution (μmol L−1)
MCZ (% w/w) SWAdSV⁎
Labeled
Micene WP®
50 70 50 70
78.0 ± 2.6 77.5 ± 3.1 77.0 ± 3.5 75.2 ± 1.5
80 80 80 80
Mancozan®
⁎ Mean ± standard deviation (n = 3).
References [1] [2] [3] [4] [5] [6] [7] [8] [9]
K.W. Weissmahr, C. Houghton, D.L. Sedlak, Anal. Chem. 70 (1998) 4800–4804. D. Amorello, S. Orecchio, Microchem. J. 110 (2013) 334–339. J.K. Rai, J. Keshavayya, S.K. Rai, Int. J. ChemTech Res. 4 (2012) 867–869. S. Mujawar, S.C. Utture, E. Fonseca, J. Matarrita, K. Banerjee, Food Chem. 150 (2014) 175–181. G. Crnogorac, W. Schwack, Trends Anal. Chem. 28 (2009) 40–50. R.J. Cassella, V.A. Salim, S. Garrigues, R.E. Santelli, M. de la Guardia, Anal. Sci. 18 (2002) 1253–1256. E.D. Caldas, J. Tresson, P.E. Boon, Food Chem. Toxicol. 44 (2006) 1562–1571. Z. Vryzas, E.N. Papadakis, E. Papadopoulou-Mourkidou, J. Agric. Food Chem. 50 (2002) 2220–2226. H. Van Lishaut, W. Schwack, J. AOAC Int. 83 (2000) 720–727.
[10] C. Blasco, G. Font, Y. Picó, J. Chromatogr. A 1028 (2004) 267–276. [11] S. Bogialli, A. Di Corcia, J. Biochem. Biophys. Methods 70 (2007) 163–179. [12] E.D. Caldas, M.H. Conceição, M.C.C. Miranda, L.C.K.R. de Souza, J.F. Lima, J. Agric. Food Chem. 49 (2001) 4521–4525. [13] A.K. Malik, W. Faubel, Fresenius J. Anal. Chem. 367 (2000) 211–214. [14] O. López-Fernández, R. Rial-Otero, A. Cid, J. Simal-Gándara, Food Chem. 145 (2014) 1002–1010. [15] O. López-Fernández, R. Rial-Otero, C. González-Barreiro, J. Simal-Gándara, Food Chem. 134 (2012) 366–374. [16] T. Cajka, K. Riddellova, P. Zomer, H. Mol, J. Hajslova, Food Addit. Contam. 28 (2011) 1372–1382. [17] M.S. Lin, B.I. Jan, H.-J. Leu, J.S. Lin, Anal. Chim. Acta 388 (1999) 111–117. [18] R.A.G. Silva, L.A.J. Silva, R.A.A. Munoz, E.M. Richter, A.C. Oliveira, J. Electroanal. Chem. 733 (2014) 85–90. [19] F. Gueguen, F. Boisde, A.-L. Queffele, J.-P. Haelters, D. Thouvenot, B. Corbel, P. Nodet, J. Agric. Food Chem. 48 (2000) 4492–4499. [20] A.R. Cassella, R.J. Cassella, S. Garrigues, R.E. Santelli, R.C. de Campos, M. de la Guardia, Analyst 125 (2000) 1829–1833. [21] A.M.O. Brett, C.M.A. Brett, Electrochemistry: principles, methods, and applications, Oxford Science Publications, Oxford University Press, UK, 1993. [22] International Organization for Standardization, Water Quality—Calibration and Evaluation of Analytical Methods and Estimation of Performance Characteristics [ISO 8466-1], 1990 (Geneva). [23] J.N. Miller, J.C. Miller, Statistics and Chemometrics for Analytical Chemistry, sixth ed. Pearson Education Limited, Harlow, 2000. [24] L.M. de Carvalho, P.C. do Nascimento, D. Bohrer, L. Del-Fabro, Electroanalysis 16 (2004) 1508–1513.
Contents lists available at ScienceDirect
Journal of Electroanalytical Chemistry journal homepage: www.elsevier.com/locate/jeac
Voltammetric analysis of mancozeb and its degradation product ethylenethiourea Olalla López-Fernández a,b, M. Fátima Barroso a,⁎, Diana M. Fernandes c, Raquel Rial-Otero b, Jesús Simal-Gándara b, Simone Morais a, Henri P.A. Nouws a, Cristina Freire c, Cristina Delerue-Matos a a b c
REQUIMTE/LAQV, Instituto Superior de Engenharia do Porto, Instituto Politécnico do Porto, Rua Dr. António Bernardino de Almeida 431, 4200-072 Porto, Portugal Nutrition and Bromatology Group, Analytical and Food Chemistry Department, Faculty of Food Science and Technology, University of Vigo, Ourense Campus, E-32004 Ourense, Spain REQUIMTE/LAQV, Dep. de Química e Bioquímica, Faculdade de Ciências, Universidade do Porto, Rua do Campo Alegre s/n, 4169-007 Porto, Portugal
a r t i c l e
i n f o
Article history: Received 4 June 2015 Received in revised form 14 August 2015 Accepted 20 August 2015 Available online 24 August 2015 Keywords: Mancozeb Ethylenethiourea Cyclic voltammetry Square-wave adsorptive stripping voltammetry
a b s t r a c t The purpose of this work was to develop a reliable alternative method for the determination of the dithiocarbamate pesticide mancozeb (MCZ) in formulations. Furthermore, a method for the analysis of MCZ's major degradation product, ethylenethiourea (ETU), was also proposed. Cyclic voltammetry was used to characterize the electrochemical behavior of MCZ and ETU, and square-wave adsorptive stripping voltammetry (SWAdSV) was employed for MCZ quantification in commercial formulations. It was found that both MCZ and ETU are irreversibly reduced (−0.6 V and −0.5 V vs Ag/AgCl, respectively) at the surface of a glassy carbon electrode in a mainly diffusion-controlled process, presenting maximum peak current intensities at pH 7.0 (in phosphate buffered saline electrolyte). Several parameters of the SWAdSV technique were optimized and linear relationships between concentration and peak current intensity were established between 10–90 μmol L−1 and 10–110 μmol L−1 for MCZ and ETU, respectively. The limits of detection were 7.0 μmol L−1 for MCZ and 7.8 μmol L−1 for ETU. The optimized method for MCZ was successfully applied to the quantification of this pesticide in two commercial formulations. The developed procedures provided accurate and precise results and could be interesting alternatives to the established methods for quality control of the studied products, as well as for analysis of MCZ and ETU in environmental samples. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Dithiocarbamates (DTCs) are important organosulfur compounds discovered in the 1930s, usually used as pesticides to treat a wide variety of fungal diseases in plants [1,2]. Due to their high activity and low production costs, they have been applied on a large scale throughout the world on several types of crops. Furthermore, due to the possibility of their combination with new generation systemic fungicides these compounds are quite successful for the management of disease resistance, especially in horticultural crops [3,4]. Based on their carbon skeleton, DTCs can be categorized into three subclasses: dimethyldithiocarbamates (DMDs), ethylenebis(dithiocarbamates) (EBDCs), and propylenebis(dithiocarbamates) (PBDs). They are mainly complexed with transition metals (e.g., manganese or zinc) [5]. DTCs are medium to highly toxic substances, depending on their structure [6]. Mancozeb (MCZ; manganese ethylenebis(dithiocarbamate) (polymeric) complex with zinc salt), a broad spectrum fungicide with multi-site contact activity, is one of the most widely used protective fungicides in the world and constitutes 50% of the market share of the ⁎ Corresponding author. E-mail address: [email protected] (M.F. Barroso).
http://dx.doi.org/10.1016/j.jelechem.2015.08.030 1572-6657/© 2015 Elsevier B.V. All rights reserved.
EBDC fungicides [5,7]. MCZ is used to protect fruits and vegetables from foliar diseases and is also employed in the protection of such products during their storage and transportation. MCZ strongly adsorbs to soil particles and usually does not move below the upper layer of the soil [6]. MCZ is easily degraded to ethylenethiourea (ETU) in the presence of moisture or oxygen and biological systems [8,9]. This degradation product is more toxic than the parent compound. ETU is suspected to cause thyroid and neurotoxic effects and induces carcinogenesis, mutagenesis, and teratogenesis [10,11]. The analysis of DTCs is complex because of their insolubility in most solvents, their ability to form stable complexes with a variety of metal ions and their instability, which is promoted by oxygen, moisture, pH, temperature, and especially plant components [8,9]. There are several analytical methods available for DTC determination based on the use of spectrophotometry [12], capillary electrophoresis [8,13], gas chromatography [4,14], high-performance liquid chromatography [10,14,15], DART-TOF mass spectrometry [16], (adsorptive-stripping) voltammetry [2,17], batch injection analysis with amperometric detection [18], enzyme-linked immunosorbent assay [19], and flow injection-Fourier transform infrared spectrometry [20]. Many of these methods are based on the acid hydrolysis of DTCs in the presence of tin(II) chloride and the subsequent analysis of CS2,
O. López-Fernández et al. / Journal of Electroanalytical Chemistry 758 (2015) 54–58
H2S, or amines [8]. However, these methods present some disadvantages because they are time consuming, expensive and environmentally unfriendly. Although electrochemical determinations of DTC have been described [2,17], they are based on the use of mercury electrodes. Mercury is toxic and has negative effects on the environment and human health. Therefore, its use constitutes a major drawback. The aim of this work was to develop a simple, fast and cost-effective electrochemical method for MCZ and ETU detection and quantification in real samples. This proposed voltammetric method is environmentally friendly and is based on the study of the electrochemical reduction of MCZ and ETU on a glassy carbon electrode (GCE) surface. Performance characteristics such as linearity, limits of detection and quantification, precision, and accuracy were evaluated. This alternative methodology was applied to the analysis of MCZ in two commercial fungicide formulations available in Portugal and Spain (Micene WP® (Sipcam) and Mancozan® (Bayer CropScience)). 2. Experimental part 2.1. Reagents and solutions Mancozeb (96.8%), ethylenethiourea (99.9%) and potassium hexacyanoferrate(III) (99%) were purchased from Sigma-Aldrich (Germany). Ethylenediaminetetraacetic acid disodium salt 2-hydrate (EDTA, ≥99%) was obtained from Panreac (Spain) and used to dissolve MCZ. For the voltammetric analyses phosphate buffered saline (PBS) solutions (pH 5.0–9.0) were used. These buffers were prepared using potassium dihydrogenphosphate (99%, Riedel-de Haën, Germany), sodium chloride (99.5%, Panreac, Spain), potassium chloride (99.5%, Riedel-de Haën, Germany), and di-sodium hydrogenphosphate 7hydrate (Riedel-de Haën, Germany). Hydrochloric acid (37%, Scharlau, Spain) and sodium hydroxide (Pronalab, Portugal) solutions (both at 0.1 mol L−1) were used to adjust the pH of the solutions. The fungicide formulations Micene WP® (Sipcam, Spain) and Mancozan® (Bayer CropScience, Portugal), both containing 80% (w/w) of MCZ, were used to evaluate the accuracy of the voltammetric method. Because of the very low solubility of MCZ and ETU in water, EDTA was used to aid dissolution. Stock solutions of MCZ (1000 μmol L−1) and ETU (1000 μmol L−1) were prepared daily by dissolving an accurately weighed amount of the compound in an EDTA solution (3%) and ultrapure water, respectively. These solutions were stored in the dark at 4 °C until use. Solutions of the commercial products Micene WP® and Mancozan® were also prepared by dissolving an accurately weighed amount in a 3% EDTA solution. All solutions were prepared using ultra-pure water (resistivity = 18.2 MΩ cm) obtained from a Simplicity 185 water purification system (Millipore, France). 2.2. Equipment All voltammetric measurements were performed using a computercontrolled Autolab PGSTAT12 potentiostat/galvanostat (MetrohmAutolab, The Netherlands) and a Metrohm 663 VA Stand, containing a three-electrode cell composed of a glassy carbon electrode (GCE; working electrode), an Ag/AgCl (KCl 3 mol L− 1) reference electrode and a glassy carbon auxiliary electrode (Metrohm, Switzerland). The system was controlled by means of the General Purpose Electrochemical System (GPES) software package (v. 4.9, Metrohm-Autolab, The Netherlands). The GCE was manually cleaned before each measurement by polishing its surface with Micropolish Alumina (0.05 μm, Buehler, Germany) until a shining surface was obtained. The electrode was then rinsed with ultra-pure water before analysis. pH measurements were made with a Crison GLP-22 pH-meter and a combined glass electrode.
55
2.3. Voltammetric analysis The voltammetric studies were conducted in PBS buffers (pH 5.0– 9.0) using either cyclic voltammetry (CV) or SWAdSV. In these assays the GCE was immersed in a low-volume cell containing 1.0 × 10−3 L of PBS buffer. This cell was then inserted in a larger cell containing the reference and auxiliary electrodes and 40.0 × 10−3 L of the PBS buffer. Prior to the analysis, electrolytes were purged with nitrogen for 10 min to ensure complete oxygen removal. In the CV and SWAdSV assays, the potential was scanned between − 0.20 and −1.0 V for MCZ, and between 0.20 and − 1.2 V for ETU. In all the SWAdSV assays a step potential (ΔEs) of 0.0051 V and a pulse amplitude (ΔEp) of 0.01995 V were used. For the optimization of the SWAdSV procedure, the following parameters were studied: electrolyte pH, accumulation potential (Eacc), accumulation time (tacc) and square-wave frequency (f). In Table 1 the used conditions in the optimization studies are shown. In the optimized SWAdSV procedure accumulation potentials of − 0.50 V for MCZ and − 0.10 V for ETU, a tacc of 20 s, and an f of 100 Hz were used. 3. Results and discussion 3.1. Cyclic voltammetric studies Cyclic voltammograms of 200 μmol L−1 MCZ and ETU solutions in PBS buffer (pH 7.0) at different scan rates (ν) showed a single irreversible reduction process for both compounds at −0.6 V for MCZ and −0.5 V for ETU (Fig. 1). The peak potentials were found to be pH-independent (data not shown). The influence of ν on the peak current intensity (ip) was studied from 0.010 to 0.500 V s−1 (Fig. 1). Within this interval, linear relationships between ip and ν1/2 were established for MCZ: − ip (μ A) = 0.169 × ν1/2 (mV s−1)1/2 + 0.742 (R = 0.973) and for ETU: − ip (μ A) = 0.163 × ν1/2 (mV s−1)1/2 + 0.566 (R = 0.994). These equations indicate that the reduction processes of both compounds at the surface of the GCE are mainly controlled by diffusion [21]. 3.2. Square-wave adsorptive stripping voltammetric studies After the characterization of the electrochemical reduction behavior of MCZ and ETU, a more sensitive technique, SWAdSV, was used to develop an alternative procedure for their detection and quantification. For this purpose, the influence of several analytical features such as electrolyte pH, square-wave frequency ( f), Eacc and tacc on ip and the peak width and shape was evaluated and optimized (Table 1). Fig. 2 shows the effects of these variables (univariable study) on the ip values of 50 μmol L−1 MCZ and ETU solutions. For the optimization of the pH of the electrolyte (PBS buffer), several MCZ and ETU solutions with pH values ranging between 5.0 and 9.0 were analyzed (Fig. 2). It was found that the reduction peak potential was independent of pH and that the highest ip values were obtained at pH 7.0. Therefore, this pH was selected for subsequent experiments. The influence of f on ip and peak width and shape was evaluated from 10 to 150 Hz. The ip increased with frequency up to 100 Hz for MCZ and up to 150 Hz for ETU (Fig. 2). However for ETU the peak
Table 1 Parameters used in the optimization of the SWAdSV procedure. Magnitude optimized
MCZ Eacc (V)
Eacc tacc f pH a
Varied.
ETU tacc (s)
a
5
−0.5 −0.5 −0.5
a
20 20
f (Hz)
Eacc (V)
tacc (s)
f (Hz)
10 10
a
5
−0.1 −0.1 −0.1
a
10 10
a
100
20 20
a
100
56
O. López-Fernández et al. / Journal of Electroanalytical Chemistry 758 (2015) 54–58
Fig. 1. Cyclic voltammograms of MCZ (A) and ETU (B) solutions (200 μmol L−1; pH = 7). Scan rates (v s−1) A: 0.010, B: 0.020, C: 0.050, D: 0.100, E: 0.150, F: 0.200, G: 0.300, and H: 0.500.
became larger and less accentuated using this latter frequency. Therefore, and to make a compromise between sensitivity and selectivity, 100 Hz frequencies were used for both compounds in the subsequent studies. The influence of Eacc on ip was studied in the range from − 0.60 to 0.00 V for MCZ and − 0.50 to 0.00 V for ETU (Fig. 2). Although ip was not maximum for MCZ an Eacc of −0.50 V was chosen for further assays. For ETU the Eacc selected was −0.10 V. The effect of tacc (between 0 and 30 s) on ip was also investigated (Fig. 2). The ip values of the reduction of MCZ increased with tacc, observing a maximum at 30 s. Because no significant differences were seen between 20 s and 30 s, the shortest time was chosen. For ETU, a maximum at tacc = 20 s was obtained. Therefore, all the analyses were carried out with a 20 s accumulation time. With the optimized experimental conditions, several figures of merit [22,23] of the developed voltammetric methods were calculated (Table 2). As can be seen in this table linear relationships and good correlations between ip and concentration were obtained between 10.0–90.0 μmol L−1 and 10.0–110.0 μmol L−1 for MCZ and ETU, respectively. Typical voltammograms and the corresponding calibration lines for both compounds in the linear range are shown in Fig. 3. Each point of the calibration curve is the average ip value of three independent measurements. From the data shown in Table 2 it can also be concluded that (i) the method for MCZ analysis is more sensitive than the one for
ETU, (ii) both methods are precise (Vx0 b 5%) and (iii) the limits of detection (LODs) for both compounds are in the low μmol L−1 range. The precision of the results was studied by intra- and inter-day assays at two different concentration levels (50 μmol L− 1 and 70 μmol L−1) and was determined by calculating relative standard deviations (RSDs). For the intra-day studies, each concentration was analyzed in triplicate on three occasions along each working day. For inter-day studies the analyses were performed over a period of 1 week. No significant differences were found between intra-day and inter-day experiments with RSD values ranging between 2.2 and 10.1%, indicating the high precision of the voltammetric methods. For MCZ analysis in the selected commercial products, matrixinduced suppression or enhancement effects were evaluated by comparing the signals of standard MCZ solutions (concentrations between 10 and 90 μmol L−1) with those obtained for the same (final) concentrations of MCZ added to 20 μmol L−1 solutions of the commercial products (Micene WP® or Mancozan®). No significant differences were observed between the slopes of the calibration curves and the standard addition curves (p N 0.05) (ratios between the slope of the calibration curve and the standard addition curve were 1.03 and 1.07 for Micene WP® and Mancozan®, respectively) which indicate the absence of matrix effects for both commercial products. Average recoveries within the studied range were 91.6 ± 4.9% (n = 5) for Micene WP® and 100 ± 13%
Fig. 2. Univariable optimization of pH, frequency ( f), accumulation potential (Eacc), and accumulation time (tacc) using 50.0 μmol L−1 MCZ (A) and ETU (B) solutions. The different conditions studied for each variable are indicated on the X-axis.
O. López-Fernández et al. / Journal of Electroanalytical Chemistry 758 (2015) 54–58 Table 2 Figures of merit of the developed SWAdSV methods for the analysis of MCZ and ETU. Figure of merit
MCZ
ETU
Linear range (μ mol L−1) n Correlation coefficient (r) Slope (m) (μ A/μ mol L−1) Standard deviation of the slope (Sm) (μ A/μ mol L−1) Intercept (a) (μ A) Standard deviation of the intercept (Sa) (μ A) Standard deviation of the linear regression (Sy/x) Standard deviation of the method (Sx0) Coefficient of variation of the method (Vx0) (%) Limit of detection (LOD) (μ mol L−1) Limit of quantification (LOQ ) (μ mol L−1)
10.0–90.0 9 0.997 0.0186 0.0006 0.091 0.032 0.044 2.3 4.7 7.0 23
10.0–110 7 0.998 0.0146 0.0004 0.017 0.027 0.038 2.6 4.8 7.8 26
(n = 5) for Mancozan®, indicating that the method provides accurate results. Furthermore, due to the absence of matrix effects the calibration curve can be used for the quantification of MCZ in the two commercial products instead of the standard addition method. This allows the analysis of more samples in a single run. Studies about the determination of DTCs by voltammetry are scarce in the literature. Amorello and Orecchio [2] proposed a method for the determination of nine DTCs (mancozeb, maneb, propineb, nabam, Na(CH3)2DTC, zineb, ziram, ferbam and thiram) by adsorptivestripping voltammetry in simulated pesticide formulations using a hanging mercury drop working electrode. The recovery and precision obtained for mancozeb were similar to those reported in this work (N 95% and RSD = 8.1%, respectively). In the previous work an LOD of 14 nmol L−1 was achieved, which is much lower than the one obtained in this study. This is mainly because of the use of much longer tacc (120 s). Carvalho and coworkers [24] determined ETU in water samples by cathodic stripping voltammetry, also using a hanging mercury drop electrode. The recoveries obtained varied between 93 and 110% and the RSD was 1.9%. The LOD obtained, after a tacc of 300 s, was also much lower (14 nmol L− 1) than the one obtained in the present work. In any case, the proposed method is more environmentally friendly than the other published methods due to the use of a GCE instead of a mercury electrode. 3.3. Analytical application In order to assess the applicability of the developed method, two different batches of the commercially available products were analyzed.
57
Different solutions of the products Micene WP® and Mancozan® were prepared to obtain concentrations of 50 and 70 μmol L−1 MCZ. These solutions were analyzed by the developed method using a (direct) calibration curve. The results are summarized in Table 3. The data show a good agreement between the labeled and the experimental values for each product for both concentration levels, demonstrating the accuracy of the procedure. Also, the RSDs were low (b4%), proving the precision of the developed method for the analysis of MCZ in the commercial products. There are hardly any studies on the determination of MCZ by voltammetry in formulations. The method proposed by Amorello and Orecchio [2] for the determination of different dithiocarbamates by adsorptive-stripping voltammetry was not applied to the quality control of pesticide formulations. Silva et al. [18] determined the concentration of MCZ in insecticide samples using batch injection analysis with pulsed amperometric detection on boron-doped diamond electrodes. Recoveries and the RSD were similar to those found in our method (95%, RSD b 2%). 4. Conclusions As a response to the rising demand of analytical methods, which should be able to detect and quantify agrochemicals, and to the growing environmental interest, a validated approach to regulatory laboratories for routine analysis of mancozeb or ethylenethiourea is offered. Furthermore, the studies indicate that MCZ can be quantified accurately, precisely and rapidly in pesticide formulations by SWAdSV with a GCE in PBS (pH 7.0) buffer. The proposed methods provide economic and environmental advantages, when compared to other previously published methods. The toxicity and/or the waste volume are much lower than that generated by the other methods such as electroanalysis (mainly based on mercury electrodes) or HPLC (large volumes of mobile phases are used). Acknowledgments O. López-Fernández would like to thank her predoctoral fellowship and her grants for mobility research staff from the University of Vigo. M.F. Barroso and D.M. Fernandes are grateful to Fundação para a Ciência e a Tecnologia (FCT) for their post-doc fellowships (SFRH/BPD/78845/ 2011 and SFRH/BPD/74877/2010), financed by POPH-QREN-Tipologia 4.1-Formação Avançada, subsidized by Fundo Social Europeu and Ministério da Ciência, Tecnologia e Ensino Superior. This work received financial support from the European Union (FEDER funds through COMPETE) and national funds (FCT) through project UID/QUI/50006/2013.
Fig. 3. Typical SWAdSV signals in the linear range obtained for (A) MCZ (10, 30, 50, 70, and 90 μmol L−1) and (B) ETU (20, 50, 70, 90 and 110 μmol L−1) solutions (at pH 7.0 in PBS buffer) using the optimized conditions (increase of −ip with increasing concentration). Inset: calibration curves.
58
O. López-Fernández et al. / Journal of Electroanalytical Chemistry 758 (2015) 54–58
Table 3 Results obtained in the analysis of MCZ in commercial pesticide formulations. Product
Analyzed solution (μmol L−1)
MCZ (% w/w) SWAdSV⁎
Labeled
Micene WP®
50 70 50 70
78.0 ± 2.6 77.5 ± 3.1 77.0 ± 3.5 75.2 ± 1.5
80 80 80 80
Mancozan®
⁎ Mean ± standard deviation (n = 3).
References [1] [2] [3] [4] [5] [6] [7] [8] [9]
K.W. Weissmahr, C. Houghton, D.L. Sedlak, Anal. Chem. 70 (1998) 4800–4804. D. Amorello, S. Orecchio, Microchem. J. 110 (2013) 334–339. J.K. Rai, J. Keshavayya, S.K. Rai, Int. J. ChemTech Res. 4 (2012) 867–869. S. Mujawar, S.C. Utture, E. Fonseca, J. Matarrita, K. Banerjee, Food Chem. 150 (2014) 175–181. G. Crnogorac, W. Schwack, Trends Anal. Chem. 28 (2009) 40–50. R.J. Cassella, V.A. Salim, S. Garrigues, R.E. Santelli, M. de la Guardia, Anal. Sci. 18 (2002) 1253–1256. E.D. Caldas, J. Tresson, P.E. Boon, Food Chem. Toxicol. 44 (2006) 1562–1571. Z. Vryzas, E.N. Papadakis, E. Papadopoulou-Mourkidou, J. Agric. Food Chem. 50 (2002) 2220–2226. H. Van Lishaut, W. Schwack, J. AOAC Int. 83 (2000) 720–727.
[10] C. Blasco, G. Font, Y. Picó, J. Chromatogr. A 1028 (2004) 267–276. [11] S. Bogialli, A. Di Corcia, J. Biochem. Biophys. Methods 70 (2007) 163–179. [12] E.D. Caldas, M.H. Conceição, M.C.C. Miranda, L.C.K.R. de Souza, J.F. Lima, J. Agric. Food Chem. 49 (2001) 4521–4525. [13] A.K. Malik, W. Faubel, Fresenius J. Anal. Chem. 367 (2000) 211–214. [14] O. López-Fernández, R. Rial-Otero, A. Cid, J. Simal-Gándara, Food Chem. 145 (2014) 1002–1010. [15] O. López-Fernández, R. Rial-Otero, C. González-Barreiro, J. Simal-Gándara, Food Chem. 134 (2012) 366–374. [16] T. Cajka, K. Riddellova, P. Zomer, H. Mol, J. Hajslova, Food Addit. Contam. 28 (2011) 1372–1382. [17] M.S. Lin, B.I. Jan, H.-J. Leu, J.S. Lin, Anal. Chim. Acta 388 (1999) 111–117. [18] R.A.G. Silva, L.A.J. Silva, R.A.A. Munoz, E.M. Richter, A.C. Oliveira, J. Electroanal. Chem. 733 (2014) 85–90. [19] F. Gueguen, F. Boisde, A.-L. Queffele, J.-P. Haelters, D. Thouvenot, B. Corbel, P. Nodet, J. Agric. Food Chem. 48 (2000) 4492–4499. [20] A.R. Cassella, R.J. Cassella, S. Garrigues, R.E. Santelli, R.C. de Campos, M. de la Guardia, Analyst 125 (2000) 1829–1833. [21] A.M.O. Brett, C.M.A. Brett, Electrochemistry: principles, methods, and applications, Oxford Science Publications, Oxford University Press, UK, 1993. [22] International Organization for Standardization, Water Quality—Calibration and Evaluation of Analytical Methods and Estimation of Performance Characteristics [ISO 8466-1], 1990 (Geneva). [23] J.N. Miller, J.C. Miller, Statistics and Chemometrics for Analytical Chemistry, sixth ed. Pearson Education Limited, Harlow, 2000. [24] L.M. de Carvalho, P.C. do Nascimento, D. Bohrer, L. Del-Fabro, Electroanalysis 16 (2004) 1508–1513.