Study of Electrochemical Oxidation and Quantification of the Pesticide Pirimicarb Using a Boron-Doped Diamond Electrode

Accepted Manuscript Title: Study of Electrochemical Oxidation and Quantification of the Pesticide Pirimicarb Using a Boron-Doped Diamond Electrode Authors: Thiago Matheus Guimar˜aes Selva, William Reis de Ara´ujo, Raphael Prata Bacil, Thiago Regis Longo Cesar Paix˜ao PII: DOI: Reference:

S0013-4686(17)31289-6 http://dx.doi.org/doi:10.1016/j.electacta.2017.06.051 EA 29689

To appear in:

Electrochimica Acta

Received date: Revised date: Accepted date:

3-4-2017 7-6-2017 8-6-2017

Please cite this article as: Thiago Matheus Guimar˜aes Selva, William Reis de Ara´ujo, Raphael Prata Bacil, Thiago Regis Longo Cesar Paix˜ao, Study of Electrochemical Oxidation and Quantification of the Pesticide Pirimicarb Using a Boron-Doped Diamond Electrode, Electrochimica Actahttp://dx.doi.org/10.1016/j.electacta.2017.06.051 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

Study of Electrochemical Oxidation and Quantification of the Pesticide Pirimicarb Using a Boron-Doped Diamond Electrode

Thiago Matheus Guimarães Selva,a,b William Reis de Araújo,b Raphael Prata Bacil,b and Thiago Regis Longo Cesar Paixãob,*

a

Instituto Federal de Educação, Ciência e Tecnologia de Pernambuco, Av. Prof. Luiz

Freire, 500, Cidade Universitária, Recife – PE, Brazil, 50740-545 b

Universidade de São Paulo, Instituto de Química, Av. Prof. Lineu Prestes, 748, São

Paulo – SP, Brazil, 05508-000

*Corresponding author: Thiago Regis Longo Cesar Paixão E-mail: [email protected] Telephone: +55 11 30919150

Research highlights 

A complete electro-oxidation mechanism of the pesticide Pirimicarb was proposed;



The electrochemical mechanism was supported by voltammetry techniques and mass spectrometry data;



An electroanalytical method using boron-doped diamond electrode was proposed to quantify Pirimicarb in natural waters;

2



The proposed analytical method is simple, low-cost, accurate and portable.

ABSTRACT An electrochemical study of the carbamate pesticide pirimicarb (PMC), which acts on the central nervous system, was performed using a boron-doped diamond working electrode. Cyclic, differential pulse, and square-wave voltammetry experiments across a wide pH range (2.0 to 8.0) showed three irreversible oxidation processes in the voltammetric behavior of PMC. The two first processes were pH-dependent, while the third was not. The three oxidation process were independent of each other, and each involved the transfer of one electron. A reaction proposal for the electrochemical oxidation of PMC is shown, and it is supported by mass spectrometry experiments. After this, an analytical method of quantifies PMC in water samples by differential pulse (DP) voltammetry is proposed. The optimal DP voltammetric parameters (step potential, amplitude potential, and scan rate) were optimized using experimental design, and an analytical curve was obtained from 2.0 to 219 µmol L−1 with an estimated detection limit of 1.24 µmol L−1. The accuracy of the proposed method was evaluated by the addition and recovery method, with recoveries ranging from 88.6 to 96.3%. Some highlights of the proposed analytical method are its simplicity, reliability, and portability.

Keywords: Pirimicarb; Cyclic voltammetry; Differential pulse voltammetry; Squarewave voltammetry; Boron-doped diamond.

3

INTRODUCTION Although pesticides are used to enhance food quality by eliminating or preventing pest infestation in plant cultures [1], they can be harmful to the health of humans and animals. In addition, their use is increasing as a consequence of population growth, which increases the demand for food production [2]. Currently, the pesticide industry is one of the world’s largest, with expenditures reaching tens of billions of dollars annually [3]. Carbamate pesticides are one of the major classes of pesticides used around the world, and are largely used because of their broad spectrum of biological activity [4]. Pirimicarb ([2-(dimethylamino)-5,6-dimethylpyrimidin-4-yl] N,Ndimethylcarbamate; PMC) is a carbamate pesticide classified as “likely to be carcinogenic to humans” [5,6]. PMC shows a reasonable water solubility about 2,700 mg L−1 [7]. The Brazilian laws prescribe that the maximum residual level (MRL) of PMC varies from 0.05 to 1 mg kg−1, depending on the type of crop [8]. At USA, according to the Environmental Protection Agency, there are no products containing PMC [9] and in Europe the use of PMC is authorized in several countries [10]. The exposure to carbamates pesticides is associated with leukemia [11]. The biological mechanism of carbamate pesticides is associated with the inhibition of the acetylcholinesterase enzyme (AChE), which is responsible for the breakdown of excessive amounts of the neurotransmitter acetylcholine. Excess acetylcholine is called cholinergic crisis, and will result in an interruption of neural signals [12] and various health problems [13,14]. Thus, studies focusing on understanding the mechanism of action of this pesticide are important. Electrochemical techniques such as voltammetry are powerful tools that can provide insights into the mechanisms of action of a range of

4

compounds [15]. In addition, the development of sensitive, reliable, and accurate analytical methods for monitoring PMC in different matrices should be pursued. The most common method of detecting and quantifying PMC is the use of an extraction method coupled with high performance liquid chromatography with ultraviolet detection (HPLC-UV) [16–24]. Liquid chromatography with mass spectrometry (LC-MS) [25] and tandem mass spectrometry detectors (LC-MS/MS) [26,27], ultra-high performance liquid chromatography with tandem mass spectrometry detectors (UPLC-MS/MS) [28], gas chromatography with mass spectrometry (GC-MS) [29,30], tandem mass spectrometry (GC-MS/MS) [31], nitrogen-phosphorous (GCNPD) [17,30,32], and electron capture (GC-ECD) [32] detectors are also used. Methods based on capillary electrophoresis with UV [33] or mass spectrometry detection [34], electrochemiluminescence [35], voltammetric biosensors [4,36], and polarography [37] can also be found in the literature. Most of the previously mentioned methods are very reliable; however, they show some disadvantages based on the use of expensive instrumentation or biological materials that require specialized sensor storage or experimental conditions. In addition, they are not easily portable, and some of them use hazardous materials such as mercury, or are laborious and require well-trained staff, making their implementation in routine and field analyses difficult. However, electroanalytical techniques offer some advantages including easy portability, suitable sensitivity, and relatively inexpensive instrumentation [14,38], as the sensor material plays an important role in terms of both analytical parameters and price. Thus, the use of unmodified electrodes, or those requiring only experimentally simple modifications, can reduce the cost of the sensor, making it more attractive for implementation in routine or in-field analysis.

5

It is possible to find some electrochemical studies of PMC in the literature. Pingarrón and coworkers [39] studied the anodic behavior of PMC using a glassy carbon electrode in a aqueous (Britton-Robinson buffer) and organic media. Based on the experiments performed, the authors observed two oxidation processes in aqueous media, while in acetonitrile it was possible to observe three oxidation processes by differential pulse voltammetry. After studying the electrochemical properties of PMC, the authors reported an analytical method for detecting PMC in soil samples using differential pulse voltammetry using an acetonitrile medium with an estimated detection limit of 6.1 x 10-7 mol L-1. In an earlier study, Batley and Afgan [40] examined a range of pesticides in aqueous media, including PMC. The study was performed by cyclic voltammetry using a glassy carbon electrode, and the authors observed an oxidation process at +1.15 V vs. a saturated calomel reference electrode (SCE). The authors also commented that the voltammetric signal appeared as a shoulder in the region of the oxygen evolution wave, and that the resolution of the signal varied with the electrode age. This was considered a drawback to the use of this electroanalytical method to measure PMC. A problem associated with carbamate pesticides is related to the potential required to oxidize them. This class of pesticide requires a very large oxidation potential, which is not always easy to achieve using conventional electrodes materials such as glassy carbon, platinum, or gold. For this reason, most of the electrochemical methods found in the literature make use of an alkaline derivatization process [38,41,42] when reporting analytical methods. In addition, oxidation products may adsorb onto the electrode surface, increasing the difficulty of monitoring such compounds. Boron-doped diamond (BDD) electrodes have emerged as a promising material for use in voltammetric analysis due to its beneficial characteristics, which include large anodic and cathodic

6

potential ranges (> ±2 V), low background currents, and minimal adsorption. The electrochemical features of the BDD electrodes are directly linked to the boron doping level necessary to become the material conductive, but in higher doping levels the BDD electrode exhibits high amount of sp2 sites enhanced the adsorption of organic materials and the background current, as well as, narrowing the potential window [43–46]. Thus, BDD electrodes are an excellent alternative for overcoming the limitations of more conventional electrodes [41]. In addition, BDD electrodes can be considered ecofriendly as they can be used to replace mercury as a working electrode in numerous analyses. This work presents an electrochemical study of the carbamate pesticide PMC using a boron-doped diamond electrode in cyclic voltammetry, differential pulse voltammetry, and square wave voltammetry, as well as mass spectrometry. Additionally, a sensitive and accurate electroanalytical method for quantifying the pesticide in natural water by differential pulse voltammetry is proposed.

1. EXPERIMENTAL 2.1 Reagents and solutions All chemicals were used as received. Pirimicarb (98.7%) was acquired from SigmaAldrich® (St. Louis, MO, USA, catalogue number 45627). All others chemicals were of analytical grade and were used without any further purification. Aqueous solutions were prepared using deionized water obtained from a water purification system (Direct-Q® 5 Ultrapure Water Systems, Millipore, MA, USA) with a resistivity >18.1 MΩ cm. A PMC stock solution (10 mmol L−1) in acetonitrile (Merck®, Darmstadt, Germany) was prepared for cyclic voltammetry (CV), whereas a

7

1.00 mmol L−1 solution in a mixture of acetonitrile (20%) and phosphate buffer solution (pH 7, 80%) was prepared for the differential pulse (DP) voltammetry and square wave (SW) voltammetry experiments. A 0.5 mol L−1 acid solution of sulfuric was used for cathodic and anodic pretreatment of the boron-doped diamond electrode surface. Table S1 (Supplementary Information) lists the compositions and pH values of the buffer solutions used as supporting electrolytes for the pH study. The ionic strength (I) of all buffers was controlled to be at least 0.1 mol L−1 using sodium sulfate. Tetrabutylammonium tetrafluoroborate (TBATFB) and anhydrous acetonitrile (Merck®, Darmstadt, Germany) were used to prepare the supporting electrolyte for the voltammetric experiments in organic media.

2.2 Electrochemical instrumentation and measurements The pH values of the buffer solutions were measured using a Metrohm pH meter (Herisau, Switzerland, model 827 pH lab) coupled to a glass electrode (model Unitrode PT1000). The electrochemical measurements were carried out using a potentiostat/galvanostat Autolab PGSTAT128N (Eco Chemie, Utrecht, The Netherlands) controlled by a laptop running NOVA 1.11.2. A standard three-electrode configuration was used in the voltammetric measurements. A BDD film with a doping level of approximately 8000 ppm chemical vapor deposited (CVD) on a polycrystalline silicon wafer (8 × 8 mm2) was obtained from NeoCoat SA (La Chauxde-Fonds, Switzerland). The BDD film was used as the working electrode coupled to an electrochemical cell, as reported in previous work [47]. The counter and reference electrodes were a platinum coiled wire and a home-made Ag|AgCl|KCl (3 mol L−1) electrode [48], respectively.

8

2.2.1 CV experiments Exploratory CV experiments were carried out at different pH values (2, 3, 4, 5, 7, and 8). The BDD electrode was cathodically pre-treated before each measurement in 0.5 mol L-1 sulfuric acid by applying potentials of +3 V for 5 s and −3 V for 60 s to ensure consistent conditions at the electrode surface. Exploratory CV experiments were carried out from -0.5 to 2 V at a scan rate (v) of 100 mV s−1. The nature of the limiting step of the electrochemical process was evaluated using an acetate buffer solution (pH 4) and scan rates ranging from 20 to 1000 mV s−1. In these studies, the concentration of PMC was 500 μmol L-1.

2.2.2 Study of BDD pre-treatment Anodic and cathodic pre-treatments were evaluated for the oxidation of PMC by DP voltammetry using the following parameters: step potential (ΔEs) = 10 mV, potential amplitude (ΔEa) = 50 mV, and scan rate (v) = 50 mV s−1. Both pre-treatments were performed in 0.5 mol L−1 sulfuric acid solution. The anodic pre-treatment was performed by applying +3 V for 60 s, while the cathodic pre-treatment was carried out by applying +3 V for 5 s followed by -3 V for 60 s. After each pre-treatment, a set of 5 cyclic voltammetry scans at 100 mV s−1 from 0 to 2 V were carried out in phosphate buffer solution (pH 7), followed by 10 DP voltammetry scans. Next, an aliquot of 1 mmol L−1 PMC stock solution was added to yield a final concentration of ~50 µmol L−1.

2.2.3 Study of the effect of pH on voltammetric response

9

The pH effect was evaluated by DP voltammetry using a BDD electrode that had been cathodically pre-treated using same procedure and parameters as described in Section 2.2.2. The potential range was from 0.5 to 1.8 V. The pH values studied ranged from 2 to 8 using the buffers solutions described in Table S1, and the PMC concentration was 100 µmol L−1.

2.2.4 Study of the voltammetric response of PMC in organic media The voltammetric behavior of PMC in organic media was investigated by DP voltammetry using a 0.1 mol L−1 solution of TBATFB in acetonitrile using the same parameters as mentioned in Section 2.2.2. The maximum potential range was from 0 to 1.8 V. The PMC concentration used in this study was 1 mmol L−1.

2.2.5 Monitoring of PMC electrolysis by mass spectrometry A 2.5 mL aliquot of 50 µmol L−1 PMC in 10 mmol L−1 ammonium carbonate buffer solution (pH 10) was electrolyzed for 4 h at a potential of 1.2 V using the BDD electrode. The solution was stirred throughout the electrolysis. Before starting the electrolysis, a 100 µL aliquot of solution was taken, diluted five times, and analyzed on a mass spectrometer (6430 triple-quad mass spectrometer, Agilent Technologies, Santa Clara, CA) with an electron spray ionization (ESI) source. The ESI parameters used were a N2 nebulizer pressure of 5 psi, the N2 drying gas was 6 L min−1 at 140 °C, and the inlet capillary voltage was 4.5 kV. The mass spectrometer was operated in positive multiple reaction monitoring mode. The liquid sheath composition was ammonium propionate (5 mmol L−1) in a 1:1 mixture of methanol:H2O.

10

After each hour of electrolysis, a new 100 µL aliquot was taken for immediate analysis on the mass spectrometer. Another PMC solution at the same concentration was electrolyzed for 4 h at a potential of 1.8 V. Before the starting and after each hour of electrolysis, a 100 µL aliquot was taken for analysis using the mass spectrometer as described previously.

2.3.6 DP voltammetric parameter optimization A 23 full-factorial central composite design (CCD) was used to optimize the DP voltammetry parameters in order to obtain better sensitivity for PMC detection. The variables optimized were (a) the step potential (ΔEs), (b) the amplitude potential (ΔEa), and (c) the scan rate (v) and three center point replications were performed in order to estimate the experimental error. The superior (+) and inferior (−) levels for each variable studied are presented in Table S2, and these were chosen in order obtain the degree of influence of each factor on the peak current obtained in the DPV. The experimental design was performed and analyzed using the Statistica 13.0 software (Dell Inc., Aliso Viejo, CA, USA). The potential range used in all the experiments was from 0.7 to 1.15 V. An analytical curve was obtained under the optimized DP voltammetric parameters provided by the experimental design. The electrochemical cell was filled with 3.00 mL of the phosphate buffer solution (pH 7) and ten DP voltammetry scans were recorded. Then, successive aliquots of the stock solution (1.00 mmol L−1 PMC) were added to the cell and, after each addition, three DP voltammetry scans were recorded. The solution was stirred for 15 s before each scan.

11

2.4 Sample preparation and analysis Tap and weir water were collected, filtered through a syringe filter with a 0.45 µm porous membrane (Whatman®), and analyzed. The analysis was performed using the external calibration approach. The DPV parameters were the same as those used for the analytical curve. In this procedure, 500 µL of 0.5 mol L−1 phosphate buffer solution (pH 7) was placed in the electrochemical cell, followed by the addition of 2.00 mL of the sample under analysis. Three DP voltammetry scans were then recorded. The system was first checked to determine if there was an oxidation signal in the region of the oxidation signal of PMC at approximately 1 V. After this, the spiked samples were then analyzed following the same protocol.

2.5 Interference study Experiments were performed to verify if some of the species commonly present in tap and natural water could interfere with the proposed method. Interference from two pesticides from different classes was also evaluated, the first being an organophosphate (methyl parathion) and the second being carbaryl, the most used carbamate pesticide. , Ca 2

,

,

,

, and

ions were used in 100-fold excesses

compared to PMC. The following ions were in 10-fold excesses compared to PMC: , carbaryl.

,

, ,

and , and

, as well as the pesticides methyl parathion and were also examined at the same concentration as that of

PMC. The PMC concentration used in this study was 10.0 µmol L−1.

2. RESULTS AND DISCUSSION

12

3.1 Voltammetric study of the pirimicarb First, the electrochemical pre-treatment of the BDD surface as described in the Section 2.2.2 was evaluated. It was observed that the anodically pre-treated BDD provided higher background current values than those obtained after cathodic pretreatment, which could compromise the signal/noise ratio. Thus, the cathodic treatment of the BDD surface was chosen for use in this work. Cyclic voltammograms for a 100 µmol L−1 solution of PMC using a BDD electrode at different scan rates are shown in Fig. 1. For comparison, cyclic voltammograms of 100 µmol L−1 solution of PMC in 0.1 mol L−1 phosphate buffer solution (pH 7) were recorded using a glassy carbon electrode, Fig S1.

INSERT FIGURE 1

The voltammograms in Fig. 1 show four oxidation processes between 1.0 and 1.6 V, and their irreversibility was confirmed using SW voltammetry and criteria for the estimation of the irreversibility (data not shown). The electrochemical performance of PMC has been documented in the literature by Pingarrón et al. [39], who described only two irreversible anodic peaks when using a glassy carbon electrode, which is consistent with the results obtained, according the Figure S1. This is probably due to the limited useful potential window of glassy carbon before electrolyte oxidation and due to the larger background current. On the other hand, BDD shows a wider potential window [45].

13

The oxidation of PMC at the BDD surface was then evaluated at several pH values between 2.0 and 8.0 using DP voltammetry. The DP voltammograms obtained at each pH value are shown in Fig. 2.

INSERT FIGURE 2

The inset of Fig. 2 illustrates the peak potential of each electrochemical oxidation process (Ep1, Ep2, Ep3, and Ep4) as recorded at different pH values. The first electrochemical process, Ep1, shows linearity between pH 2.0 and 5.0 with a slope of 64 mV, which is close to the theoretical value for an electron transfer involving the same number of electrons and protons (59.16 mV per unit of pH). The second process, Ep2, show a restricted linearity between pH 2.0 and 4.0 with a slope of 55 mV, which in the same way indicates the transfer of the same number of electrons and protons. The average measured W1/2 (half-peak width) in the DPV indicated that all the electrochemical processes of PMC involved one electron, as the W1/2 values were close to 90 mV. Thus, it can be concluded that Ep1 and Ep2 involved the transfer of one electron and one proton. The distributions of PMC species at different pH values as simulated using Marvin software (Fig S2) shows that the pKa of PMC is 5.0, which matches the observed loss of linearity in the Ep1 process shown in the inset of Fig. 2 (Ep vs. pH voltammetric data), indicating that the first electrochemical process is related to the PMC molecule.

14

3.2. Mechanism evaluation The cyclic voltammograms shown in Fig. 1 show that the current values increased with increasing scan rate (v), while the potential peaks shifted to more positive potentials, suggesting a chemical coupling reaction. In order to evaluate these processes systematically, we constructed a log peak current (Ip) versus log(v) plot and a peak potential (Ep) versus log(v) plot, as shown in Fig. 3.

INSERT FIGURE 3

The log(Ip) vs. log(ν) plots allow diffusion and adsorption processes to be distinguished by the slopes of the curves. In a diffusion-controlled process, Ip is proportional to ν1/2, resulting in a slope of 0.5 for this plot, while in an adsorptioncontrolled process Ip α ν, and therefore a slope of 1.0 is obtained [49]. In the mechanistic evaluation, each plot for a diffusion-controlled process could present up to three regions: a “kinetics region”, where a coupled chemical reaction was predominant; a “diffusion region”, where an electrochemical process is predominant; and a “mixed region”, where both acted together [50]. In Fig. 3a, the Ep1 process presents a slope of 0.41 across most of the scan rate range, suggesting a diffusion controlled process. In addition, in the kinetic region the Ep x log(ν) plot presents an inconclusive slope of approximately 84 mV dec‒1, when the expected value would be around 30 mV dec‒1 [50] for an EC mechanism. In Fig. 3b, process Ep2 presents a slope of 0.81 until a rate of 100 mV s‒1, and then a slope of 0.29 for further values. Therefore, Ep2 is adsorption and diffusion controlled in the first and

15

second sections, respectively. As this is an adsorptive process, the Ep x log(ν) slope cannot be used to determine the electrochemical mechanism involved, although the plot profile suggests a chemical coupled reaction occurs until 100 mV s‒1. In Fig. 3c, process Ep3 presents a slope of 73 mV dec‒1, probably due to the fouling of the electrode surface caused by Ep2. The value itself suggests that Ep3 is an adsorptive process. As for Ep2, the adsorptive character forbids an electrochemical mechanism, and in addition, the plot suggests a coupled chemical reaction occurs until 130 mV s‒1. Fig. 3d shows process Ep4, with the Ep x log(ν) plot suggesting a chemical coupled reaction until 130 mV s‒1, although the slope is 154 mV dec‒1 due to the previous adsorption processes. The log(Ip) x log(ν) plot presents an anomalous constant curve with a slope of 5 mV dec‒1, which suggests a diffusionless process and so the possibility that Ep4 is not a PMC oxidation process. Two experiments were performed to test this hypothesis: differential pulse voltammograms in acetonitrile (ACN), and differential pulse voltammograms in buffer solutions with various pH values, as shown in Fig. S3 and S4, respectively. The DP voltammograms performed in dry acetonitrile show that PMC exhibits only one irreversible anodic peak. Different proportions of water were then added to the dry acetonitrile in order to verify the effect of water on the voltammetric profile of PMC, as shown in Fig. S5. It was verified that as the percentage of water increased, the other electrochemical processes related to PMC gradually appeared, so that all four anodic processes could be seen in a mixed medium (50:50), as shown in Fig. S5a. However, it is observed that process Ep4 also appears in the analytical blank (Fig. S5b), reinforcing the hypothesis that it is not related to electro-oxidation of PMC.

16

In order to elucidate the origin of Ep4, DP voltammograms were recorded in different buffer solutions (Fig. S4), and it was verified that the BDD electrode generated hydroxyl radicals due to water oxidation that can then interact and enhance the electrooxidation of organic compounds, as reported by Brett et al. [51]. Thus, Fig. 3d can be explained by the PMC oxidation products adsorbed on the BDD electrode causing a trapping effect, which then undergo oxidation at high concentrations in a diffusionless manner. Aiming to improve our understanding of the PMC mechanism, we performed electrolysis of PMC for 4 h under two different conditions, applying voltages of 1.2 and 1.8 V. Aliquots of electrolyzed PMC solution were collected every hour for injection into the mass spectrometer, as shown in Fig. S6. It was observed that only one molecular peak was initially present (M+, m/z = 239.2), and as the electrolysis time increased, it lost intensity and new peaks appeared, indicating consumption. Table 1 highlights the major peaks identified during electrolysis.

INSERT TABLE 1

It is important to note that carbamate hydrolysis (168.2, M+ - 71) was not observed in the mass spectra under either natural conditions or after electrolysis, which indicates that the any oxidation processes observed cannot be attributed to the presence of this group in the molecule. Another fact that corroborates this hypothesis is that if the hydrolysis were occurring, it would be possible to see an oxidation process at a lower potential related to the formation of a phenolic group, which was not observed.

17

The peaks with m/z ratios of 225.2 and 253.1 are described in the literature as products of degradation of pirimicarb by advanced oxidation processes that occur via routes such as photocatalysis or even under environmental conditions in soils [52–55]. These oxidation processes and the degradation pathway of PMC generally occur by radical attacks, primarily by hydroxyl radicals. Thus, the formation of these degradation compounds was due to the fourth oxidation process, which requires the formation of hydroxyl radicals. This fact is corroborated by many works in the literature [51][56][57] regarding the versatility of the BDD electrode for wastewater treatment and remediation of organic compounds by degradation via the formation of hydroxyl radicals. Based on the data and discussion above, we propose the mechanism of electrooxidation of PMC on the BDD electrode material shown in Scheme 1.

INSERT SCHEME 1

The first process, Ep1, can be deduced by pH dependence, which is in agreement with the pKa value of the molecule predicted by Marvin Software®. The second process, Ep2, is attributed to the oxidation of the tertiary amine of the molecule to form a cation radical followed by deprotonation to give a radical, as referenced in the literature regarding the electro-oxidation of tertiary amines [58,59]. The third process, Ep3, we attribute to the oxidation of the other amine of the pyrimidine group with the formation of a cation radical, which is stabilized by the aromaticity of the ring. In the literature, the proposed electro-oxidation mechanism for carbamate pesticides, attributes the oxidation process to the carbamate group, since the

18

electrochemical mechanism occurs in two steps, the first one independent of protons and the second one pH dependent, both steps occurs in an irreversible way [60–62]. Comparing their mechanism to our proposition, the PMC reveals four oxidation processes, Ep1 and Ep2, which shows pH dependence. The processes Ep3 and Ep4 do no exhibit proton dependence, according to Fig. 2. It is important to note that Ep4 is a process that occurs independently of diffusion, Fig. 3, being ascribed to the formation of hydroxyl radical in the BDD surface, as previously reported by Enache et al [63]. Therefore, our proposed mechanism is based on the observed data from the PMC oxidation, which does not fit in the proposed mechanism for classical carbamate eletrooxidation reported in the literature [60–62]. Another result that supports this fact, generally the carbamate group undergoes a hydrolysis reaction, which was not observed in the mass spectra under either natural conditions or after electrolysis, indicating a certain stability of the molecule.

3.2 Optimization of the DP voltammetry parameters by experimental design DP voltammetry parameter response surface diagram was plotted (Fig. S7) based on the parameters as evaluated using 23 full-factorial CCD in order to obtain the critical values for the parameters. The significance of the step, amplitude, and scan rate and their interactions were evaluated by ANOVA. Only the step, amplitude, and square of the amplitude were considered statically relevant (p < 0.05), as shown in Table 2.

INSERT TABLE 2

19

A quadratic model was used to represent the peak current obtained by DP voltammetry, and the 85.9% variance (96.2/112) can be explained by the proposed model/equation: Ip = 8.20 + 1.04  Step + 1.96  Amplitude – 1.54  Amplitude2

(Eq. 1)

Additionally, the quadratic model did not show a lack of fit since the ratio between the mean square of the lack of fit and the mean square of the pure error (MSlack of fit

/ MSPure Error = 7.08) was smaller than the tabulated F-value (F8,2 / 95% = 19.37),

implying that the quadratic model adjusted very well to the peak current data extracted for the optimized values. Hence, the critical values were estimated by the 23 fullfactorial CCD experiment, and the step was 13 mV and the amplitude was 132 mV in the proposed model. Since the scan rate was not statically relevant for the data evaluated, we chose a value of 21 mV s–1 by considering a combination of time and resolution.

3.3 Analytical performance After the optimization of the DP voltammetry parameters and the experimental conditions, an analytical curve was constructed. The curve presented linearity from 2.00 to 219 µmol L−1 (R2 = 0.998). The limits of detection (LOD = 3σ/S) and quantification (LOQ = 10σ/S) were estimated statistically, where σ is the standard deviation of at least 10 measurements in the absence of the pesticide and S is the sensitivity [64]. The estimated LOD and LOQ values were 1.24 and 4.14 µmol L−1, respectively. Fig. 4 shows the DP voltammetry scans registered for the successive addition of aliquots of a 1.00 mmol L−1 PMC stock solution. The analytical curve obtained is shown in the inset of Fig. 4.

20

INSERT FIGURE 4

The main analytical parameters for the proposed method are summarized in Table 3. The reproducibility was tested by taking 10 successive DP voltammograms of a 50.0 µmol L−1 solution of PMC in a phosphate buffer solution (pH 7). The relative standard deviation (RSD) obtained was 1.05%.

INSERT TABLE 3

3.4 Sample analysis and interference study The proposed electrochemical method was applied to determine the concentration of PMC in spiked water samples. The results of the PMC recovery tests for the water samples can be seen in Table 4. The recovery values achieved for PMC are in accordance with the levels of PMC in the original samples [65]. This demonstrates that the proposed method is both accurate and reliable. In addition, it is suitable for routine analysis and can be easily adapted for in-field analysis.

INSERT TABLE 4

Table 5 compares the analytical performance of the proposed method for quantifying PMC with those of electrochemical methods presented in the literature. As

21

can be seen, some of the methods are more sensitive than that proposed in this paper; however, the proposed method does not require the use of high cost materials such as biomolecules [4,36] or toxic materials such as mercury [37] or excess acetonitrile [39]. Thus, it can be considered simpler and more eco-friendly when compared to other electrochemical methods.

INSERT TABLE 5

The interference study demonstrated that the only species studied that caused interference in the proposed method were

,

, and

. These three cations

showed interference when their concentrations were 10 times higher than that of PMC, and only

did not show interference at the same concentration as PMC. However,

the proposed electrochemical method can be considered selective when compared to others methods of quantifying carbamate pesticides, since they show typical oxidation potentials higher than 1.3 V vs. Ag|AgCl|KCl (3 mol L−1), while the first oxidation signal for PMC is around 300 mV before this.

3. CONCLUSIONS A systematic electrochemical study of PMC using a boron-doped diamond (BDD) electrode was performed for the first time, showing the existence of three irreversible and independent anodic processes. Based on the voltammetric data, a pH study, and results obtained by mass spectrometry (MS) of the compound under natural settings and different electrolysis conditions, we proposed an electro-oxidation mechanism for PMC.

22

It was also possible to observe two degradation products of PMC in MS data after electrolysis that indicated the ability of the BDD electrode to form hydroxyl radicals in aqueous media. An analytical procedure for quantifying PMC using a cathodically pretreated BDD surface was also explored, obtained a linear range from 2.0 to 219 µmol L−1 with an estimated detection limit of 1.24 µmol L−1. The method was tested in tap and weir water samples, and suitable recoveries were reached, from 88.6 to 96.3%. The proposed method is low-cost and eco-friendly, since it does not make use of biomolecules or mercury. Moreover, it showed a suitable reproducibility (RSD = 1.05%) and better selectivity when compared to other carbamates, which makes it reliable for routine and in-field analyses.

ACKNOWLEDGEMNTS We would like to acknowledge to the Brazilian agencies CAPES (Grant number: 3359/2014 PRÓ-FORENSES Edital 25/2014) and CNPq (Grant number: 444498/20141) for supporting this work. The authors are very grateful to the technicians Fernando Silva Lopes and Daniel Rossado Oliveira for their help with the MS analysis (FAPESP, Grant number: 2012/06642-1).

REFERENCES [1]

J. Cooper, H. Dobson, The benefits of pesticides to mankind and the environment, Crop Prot. 26 (2007) 1337–1348. doi:10.1016/j.cropro.2007.03.022.

[2]

Y. Ni, P. Qiu, S. Kokot, Simultaneous determination of three organophosphorus pesticides by differential pulse stripping voltammetry and chemometrics, Anal. Chim. Acta. 516 (2004) 7–17. doi:10.1016/j.aca.2004.04.007.

[3]

J. Moros, S. Armenta, S. Garrigues, M. de la Guardia, Univariate near infrared

23

methods for determination of pesticides in agrochemicals, Anal. Chim. Acta. 579 (2006) 17–24. doi:10.1016/j.aca.2006.07.009. [4]

T.M.B.F. Oliveira, M. Fátima Barroso, S. Morais, P. de Lima-Neto, A.N. Correia, M.B.P.P. Oliveira, et al., Biosensor based on multi-walled carbon nanotubes paste electrode modified with laccase for pirimicarb pesticide quantification., Talanta. 106 (2013) 137–43. doi:10.1016/j.talanta.2012.12.017.

[5]

Pirimicarb - toxicity, ecological toxicity and regulatory information, (n.d.). http://www.pesticideinfo.org/Detail_Chemical.jsp?Rec_Id=PC33366#ChemID (accessed October 7, 2016).

[6]

https://toxnet.nlm.nih.gov/cgibin/sis/search2/r?dbs+hsdb:@term+@rn+@rel+23103-98-2 (accessed October 21, 2016).

[7]

https://pubchem.ncbi.nlm.nih.gov/compound/pirimicarb#section=Solubility (accessed June 1, 2017).

[8]

http://portal.anvisa.gov.br/documents/111215/117782/p10_.pdf/79fad5a61cd6-4d77-82d9-89e6460957ad (accessed June 1, 2017).

[9] https://iaspub.epa.gov/apex/pesticides/f?p=CHEMICALSEARCH:31:::NO:1,3,31,7 ,12,25:P3_XCHEMICAL_ID:3435# (accessed June 1, 2017). [10]

http://ec.europa.eu/food/plant/pesticides/eu-pesticidesdatabase/public/?event=activesubstance.detail&language=EN&selectedID=1713 (accessed June 1, 2017).

[11]

K.M. Lafiura, D.M. Bielawski, N.C. Posecion, E.M. Ostrea, L.H. Matherly, J.W. Taub, et al., Association between prenatal pesticide exposures and the generation of leukemia-associated T(8;21)., Pediatr. Blood Cancer. 49 (2007) 624–8. doi:10.1002/pbc.21283.

[12]

D.M. Quinn, Acetylcholinesterase: enzyme structure, reaction dynamics, and virtual transition states, Chem. Rev. 87 (1987) 955–979. doi:10.1021/cr00081a005.

[13]

K.D. Khalaf, A. Morales-Rubio, M. de la Guardia, Simple and rapid flow-injection spectrophotometric determination of carbaryl after liquid-liquid extraction, Anal. Chim. Acta. 280 (1993) 231–238. doi:10.1016/0003-2670(93)85126-5.

[14]

F.C. Moraes, L.H. Mascaro, S.A.S. Machado, C.M.A. Brett, Direct electrochemical determination of carbaryl using a multi-walled carbon nanotube/cobalt phthalocyanine modified electrode., Talanta. 79 (2009) 1406–11. doi:10.1016/j.talanta.2009.06.013.

[15]

P. Kovacic, Mechanism of Organophosphates (Nerve Gases and Pesticides) and Antidotes: Electron Transfer and Oxidative Stress, Curr. Med. Chem. 10 (2003)

24

2705–2709. doi:10.2174/0929867033456314. [16]

R. Carabias-Martı ́nez, E. Rodrı ́guez-Gonzalo, P.H. Paniagua-Marcos, J. Hernández-Méndez, Analysis of pesticide residues in matrices with high lipid contents by membrane separation coupled on-line to a high-performance liquid chromatography system, J. Chromatogr. A. 869 (2000) 427–439. doi:10.1016/S0021-9673(99)01218-2.

[17]

E. Bolygó, N.C. Atreya, Solid-phase extraction for multi-residue analysis of some triazole and pyrimidine pesticides in water, Fresenius. J. Anal. Chem. 339 (1991) 423–430. doi:10.1007/BF00322364.

[18]

I.A. Stuart, R.O. Ansell, J. MacLachlan, P.A. Bather, S. Yoshida, H. Murata, et al., Surface partitioning studies of N-methylcarbamate-treated post-harvest crops using SFE-HPLC-postcolumn reaction-fluorescence, Analyst. 124 (1999) 275–280. doi:10.1039/a808013e.

[19]

J. Zhou, C. Ma, S. Zhou, P. Ma, F. Chen, Y. Qi, et al., Preparation, evaluation and application of molecularly imprinted solid-phase microextraction monolith for selective extraction of pirimicarb in tomato and pear, J. Chromatogr. A. 1217 (2010) 7478–7483. doi:10.1016/j.chroma.2010.09.079.

[20]

Q. Wu, X. Zhou, Y. Li, X. Zang, C. Wang, Z. Wang, Application of dispersive liquid– liquid microextraction combined with high-performance liquid chromatography to the determination of carbamate pesticides in water samples, Anal. Bioanal. Chem. 393 (2009) 1755–1761. doi:10.1007/s00216-009-2625-z.

[21]

Q. Wu, Q. Chang, C. Wu, H. Rao, X. Zeng, C. Wang, et al., Ultrasound-assisted surfactant-enhanced emulsification microextraction for the determination of carbamate pesticides in water samples by high performance liquid chromatography, J. Chromatogr. A. 1217 (2010) 1773–1778. doi:10.1016/j.chroma.2010.01.060.

[22]

G. Zhao, S. Song, C. Wang, Q. Wu, Z. Wang, Q.H. Wu, et al., Solid-phase microextraction with a novel graphene-coated fiber coupled with highperformance liquid chromatography for the determination of some carbamates in water samples, Anal. Methods. 3 (2011) 2929. doi:10.1039/c1ay05358b.

[23]

M. Sun, X. Ma, J. Wang, W. Wang, Q. Wu, C. Wang, et al., Graphene grafted silica-coated Fe3O4 nanocomposite as absorbent for enrichment of carbamates from cucumbers and pears prior to HPLC., J. Sep. Sci. 36 (2013) 1478–85. doi:10.1002/jssc.201201036.

[24]

S. Zhou, H. Chen, B. Wu, C. Ma, Y. Ye, Sensitive determination of carbamates in fruit and vegetables by a combination of solid-phase extraction and dispersive liquid-liquid microextraction prior to HPLC, Microchim. Acta. 176 (2012) 419– 427. doi:10.1007/s00604-011-0735-8.

[25]

E. Pose-Juan, E. Herrero-Hernández, A. Álvarez-Martín, M.J. Sánchez-Martín,

25

M.S. Rodríguez-Cruz, Development of a procedure for the multiresidue analysis of pesticides in vineyard soils and its application to real samples, J. Sep. Sci. 37 (2014) 2215–2224. doi:10.1002/jssc.201400298. [26]

D. Perret, A. Gentili, S. Marchese, M. Sergi, G. D’Ascenzo, Validation of a Method for the Determination of Multiclass Pesticide Residues in Fruit Juices by Liquid Chromatography/Tandem Mass Spectrometry after Extraction by Matrix SolidPhase Dispersion, J. AOAC Int. 85 (2002) 724–730.

[27]

H.R. Cho, J.S. Park, J. Kim, S.B. Han, Y.S. Choi, Multiresidue method for the quantitation of 20 pesticides in aquatic products, Anal. Bioanal. Chem. 407 (2015) 9043–9052. doi:10.1007/s00216-015-9071-x.

[28]

Z. Shi, J. Hu, Q. Li, S. Zhang, Y. Liang, H. Zhang, Graphene based solid phase extraction combined with ultra high performance liquid chromatography– tandem mass spectrometry for carbamate pesticides analysis in environmental water samples, J. Chromatogr. A. 1355 (2014) 219–227. doi:10.1016/j.chroma.2014.05.085.

[29]

S. Shoeibi, I. Moazzami Goudarzi, B. Rastegar, Hossein; Janat, M. Sadeghi, Nafiseh; Hajimahmoodi, M. Amirahmadi, Spiked Calibration Curve : A Valid Method for Simultaneous Analysis of Pesticides in Melon Using Gas Chromatography Mass Spectrometry ( GC / MS ), Iran. J. Chem. Chem. Eng. 33 (2014) 21–27.

[30]

B. Morzycka, Simple method for the determination of trace levels of pesticides in honeybees using matrix solid-phase dispersion and gas chromatography, J. Chromatogr. A. 982 (2002) 267–273. doi:10.1016/S0021-9673(02)01505-4.

[31]

H. Chen, R. Chen, S. Li, Low-density extraction solvent-based solvent terminated dispersive liquid–liquid microextraction combined with gas chromatographytandem mass spectrometry for the determination of carbamate pesticides in water samples, J. Chromatogr. A. 1217 (2010) 1244–1248. doi:10.1016/j.chroma.2009.12.062.

[32]

B. Lozowicka, M. Jankowska, I. Hrynko, P. Kaczynski, Removal of 16 pesticide residues from strawberries by washing with tap and ozone water, ultrasonic cleaning and boiling, Environ. Monit. Assess. 188 (2016) 51. doi:10.1007/s10661-015-4850-6.

[33]

F. Moffatt, P. Chamberlain, P.A. Cooper, K.M. Jessop, Capillary electrochromatography-applicability to the analysis of pesticides and compounds of environmental concern and the theory of electroosmosis, Chromatographia. 48 (1998) 481–490. doi:10.1007/BF02466638.

[34]

J. Hernández-Borges, M. Rodríguez-Delgado, F.J. García-Montelongo, A. Cifuentes, Highly sensitive analysis of multiple pesticides in foods combining solid-phase microextraction, capillary electrophoresis-mass spectrometry, and

26

chemometrics, Electrophoresis. 25 (2004) 2065–2076. doi:10.1002/elps.200405938. [35]

Z. Lin, G. Chen, Determination of carbamates in nature water based on the enhancement of electrochemiluminescent of Ru(bpy)32+ at the multi-wall carbon nanotube-modified electrode, Talanta. 70 (2006) 111–115. doi:10.1016/j.talanta.2005.12.026.

[36]

T.M.B.F. Oliveira, M. Fátima Barroso, S. Morais, M. Araújo, C. Freire, P. de LimaNeto, et al., Laccase–Prussian blue film–graphene doped carbon paste modified electrode for carbamate pesticides quantification, Biosens. Bioelectron. 47 (2013) 292–299. doi:10.1016/j.bios.2013.03.026.

[37]

P. Yáñez-Sedeño, J. Manuel de Villena, J.M. Pingarrón, L.M. Polo, Polarographic determination of pirimicarb, Anal. Chim. Acta. 234 (1990) 309–313. doi:10.1016/S0003-2670(00)83571-0.

[38]

Y. Ni, P. Qiu, S. Kokot, Simultaneous voltammetric determination of four carbamate pesticides with the use of chemometrics, Anal. Chim. Acta. 537 (2005) 321–330. doi:10.1016/j.aca.2004.12.080.

[39]

J.M. Pingarron, A. Gonzalez, L.M. Polo, Electroanalytical Study of Pirimicarb by Anodic Voltammetry at a Glassy-Carbon Electrode in Aqueous and Acetonitrile Media, Electroanalysis. 2 (1990) 493–497.

[40]

G.E. Batley, B.K. Afgan, Voltammetric analysis of some carbamate pesticides, J. Electroanal. Chem. Interfacial Electrochem. 125 (1981) 437–445. doi:10.1016/S0022-0728(81)80360-9.

[41]

T.N. Rao, B.H. Loo, B. V Sarada, C. Terashima, a Fujishima, Electrochemical detection of carbamate pesticides at conductive diamond electrodes., Anal. Chem. 74 (2002) 1578–83. http://www.ncbi.nlm.nih.gov/pubmed/12033247.

[42]

A. Santalad, L. Zhou, F. Shang, D. Fitzpatrick, R. Burakham, S. Srijaranai, et al., Micellar electrokinetic chromatography with amperometric detection and offline solid-phase extraction for analysis of carbamate insecticides., J. Chromatogr. A. 1217 (2010) 5288–97. doi:10.1016/j.chroma.2010.06.024.

[43]

A. Kraft, Doped diamond: A compact review on a new, versatile electrode material, Int. J. Electrochem. Sci. 2 (2007) 355–385. doi:10.1242/jeb.00516.

[44]

N. Yang, J.S. Foord, X. Jiang, Diamond electrochemistry at the nanoscale: A review, Carbon N. Y. 99 (2016) 90–110. doi:10.1016/j.carbon.2015.11.061.

[45]

J. V. Macpherson, A practical guide to using boron doped diamond in electrochemical research, Phys. Chem. Chem. Phys. 17 (2015) 2935–2949. doi:10.1039/C4CP04022H.

[46]

R. Bogdanowicz, A. Fabiańska, L. Golunski, M. Sobaszek, M. Gnyba, J. Ryl, et al., Influence of the boron doping level on the electrochemical oxidation of the azo

27

dyes at Si/BDD thin film electrodes, Diam. Relat. Mater. 39 (2013) 82–88. doi:10.1016/j.diamond.2013.08.004. [47]

T.M.G. Selva, W.R. de Araujo, T.R.L.C. da Paixão, Non-Invasive Salivary Electrochemical Quantification of Paraquat Poisoning Using Boron Doped Diamond Electrode, Electroanalysis. 27 (2015) 1642–1648. doi:10.1002/elan.201400745.

[48]

J.J. Pedrotti, L. Angnes, I.G.R. Gutz, Miniaturized Reference Electrodes with Microporous Polymer Junctions, Electroanalysis. 8 (1996) 673–675.

[49]

C.M.A. Brett, A.M.O. Brett, Electrochemistry: Principles, Methods, and Applications, 1st ed., Oxford University Press, New York, 1993.

[50]

D. Pletcher, R. Greff, R. Peat, L.M. Peter, J. Robinson, Instrumental Methods in Electrochemistry, 1st ed., Woodhead Publishing Limited, Cambridge, 2002.

[51]

T.A. Enache, A.-M. Chiorcea-Paquim, O. Fatibello-Filho, A.M. Oliveira-Brett, Hydroxyl radicals electrochemically generated in situ on a boron-doped diamond electrode, 2009. doi:10.1016/j.elecom.2009.04.017.

[52]

J. Fenoll, I. Garrido, P. Hellín, P. Flores, N. Vela, S. Navarro, Photocatalytic oxidation of pirimicarb in aqueous slurries containing binary and ternary oxides of zinc and titanium, J. Photochem. Photobiol. A Chem. 298 (2015) 24–32. doi:10.1016/j.jphotochem.2014.10.014.

[53]

T. Chen, F. Fu, Z. Chen, D. Li, L. Zhang, G. Chen, Study on the photodegradation and microbiological degradation of pirimicarb insecticide by using liquid chromatography coupled with ion-trap mass spectrometry, J. Chromatogr. A. 1216 (2009) 3217–3222. doi:10.1016/j.chroma.2009.02.022.

[54]

Y. Wu, C. Chen, Y. Huang, W. Lin, Y. Yen, C. Lu, Pirimicarb degradation by BiVO 4 photocatalysis: Parameter and reaction pathway investigations, Sep. Sci. Technol. 51 (2016) 2284–2296. doi:10.1080/01496395.2016.1202279.

[55]

E. Romero, P. Schmitt, M. Mansour, Photolysis of pirimicarb in water under natural and simulated sunlight conditions, Pestic. Sci. 41 (1994) 21–26. doi:10.1002/ps.2780410105.

[56]

A. Kapałka, B. Lanova, H. Baltruschat, G. Fóti, C. Comninellis, Electrochemically induced mineralization of organics by molecular oxygen on boron-doped diamond electrode, 2008. doi:10.1016/j.elecom.2008.06.005.

[57]

H. Wu, J. Sun, L. Zheng, X. Li, C. Gao, Detection of hydroxyl radical in the electro catalytic system of BDD electrode, in: 2016 IEEE Int. Conf. Mechatronics Autom., IEEE, 2016: pp. 558–562. doi:10.1109/ICMA.2016.7558624.

[58]

J.R.L. Smith, D. Masheder, Amine oxidation. Part IX. The electrochemical oxidation of some tertiary amines: the effect of structure on reactivity, J. Chem. Soc. Perkin Trans. 2. (1976) 47. doi:10.1039/p29760000047.

28

[59]

A. Adenier, M.M. Chehimi, I. Gallardo, J. Pinson, N. Vilà, Electrochemical Oxidation of Aliphatic Amines and Their Attachment to Carbon and Metal Surfaces, Langmuir. 20 (2004) 8243–8253. doi:10.1021/LA049194C.

[60]

T. Shono, H. Hamaguchi, Y. Matsumura, Electroorganic Chemistry. XX. Anodic Oxidation of Carbamates, J. Am. Chem. Soc. 97 (1975) 4264–4268. doi:10.1021/ja00848a020.

[61]

L. Haya, J.I. Pardo, A.M. Mainar, E. Fatás, J.S. Urieta, Regioselectivity of Electrochemical C-H Functionalization Via Iminium Ion, Electrochim. Acta. 142 (2014) 299–306. doi:10.1016/j.electacta.2014.07.092.

[62]

L. Haya, F.J. Sayago, A.M. Mainar, C. Cativiela, J.S. Urieta, A.G. Beck-Sickinger, Quantum-chemical predictions of redox potentials of carbamates in methanol, Phys. Chem. Chem. Phys. 13 (2011) 17696. doi:10.1039/c1cp21576k.

[63]

T.A. Enache, A.-M. Chiorcea-Paquim, O. Fatibello-Filho, A.M. Oliveira-Brett, Hydroxyl radicals electrochemically generated in situ on a boron-doped diamond electrode, Electrochem. Commun. 11 (2009) 1342–1345. doi:10.1016/j.elecom.2009.04.017.

[64]

J.N. Miller, J.C. Miller, Statistics and Chemometrics for Analytical Chemistry, 6th ed., Pearson, Gosport, 2010. doi:10.1198/tech.2004.s248.

[65]

A. Gustavo González, M. Ángeles Herrador, A practical guide to analytical method validation, including measurement uncertainty and accuracy profiles, TrAC Trends Anal. Chem. 26 (2007) 227–238. doi:10.1016/j.trac.2007.01.009.

29

Figure capitions Fig. 1. Cyclic voltammograms recorded using a BDD electrode in a 0.1 mol L−1 acetate buffer solution (pH 5, I = 0.1 mol L−1) containing 100 µmol L−1 of PMC at different scan rates (ν): a) 25, b) 100, c) 200, and d) 300 mV s−1. The black curve was recorded in the absence of PMC at 25 mV s-1.

30

Fig. 2. DP voltammograms recorded using a BDD electrode in presence of 100 µmol L−1 PMC in different electrolytes (as shown in Table S1). DP voltammetry parameters: step potential = 10 mV, amplitude potential = 50 mV, and scan rate = 50 mV s−1. Inset: Ep vs. pH plots for the four PMC oxidation process. Slopes: Ep1 = 64 mV (■), Ep2 = 55 mV (●), Ep3 = 14 mV (▲), Ep4 = 10 mV (▼).

31

Fig. 3. Plots of logarithm peak current and peak potential versus the logarithm of the scan rate for each oxidation process: a) Ep1, b) Ep2, c) Ep3, d) Ep4. These data were obtained from cyclic voltammograms recorded using a BDD electrode in solutions containing 500 µmol L-1 PMC in 0.1 mol L-1 acetate buffer (pH 5, I = 0.1 mol L-1).

32

Fig. 4. DP voltammograms for the consecutive addition of aliquots of a 1.00 mmol L−1 PMC stock solution to an 80:20 mixture of H2O and acetonitrile, recorded using a BDD electrode and normalized to the baseline. DP voltammetric parameters: step (ΔEs) = 13 mV, amplitude (ΔEa) = 132 mV, and scan rate (v) = 21 mV s−1. (a) Blank, (b) 2.00, (c) 3.98, (d) 5.96, (e) 7.94, (f) 11.9, (g) 19.6, (h) 38.5, (i) 56.6, (j) 74.1, (l) 107, (m) 138, (n) 167, and (o) 219 µmol L−1. Inset: analytical curve.

33

Scheme 1. Schematic view of the proposed mechanism for PMC electro-oxidation using a BDD electrode.

Table 1. Peaks identified using mass spectrometry during electrolysis of PMC. Mass peaks (m/z)

Comparison between electrolysis condition (Potential applied)

193.1 [M+ - 46]

Appears only at 1.2 V

225.2 [M+ - 14]

Appears and enhanced at 1.8 V

239.2 [M+]

Decreases in both

253.1 [M+ + 14]

Appears and enhanced at 1.8 V

34

Table 2. Analysis of variance (ANOVA) table for the quadratic model adjusted to the peak current of the PMC based on the variation of step, amplitude, and scan rate. Factor

Sum of

Degrees of

Mean

Calculated

p

Square (SS)

freedom

Square

F value

Step

14.7

1

14.7

52.8

0.0184

Step2

4.96

1

4.96

17.8

0.0518

Amplitude

52.6

1

52.6

189

0.00525

Amplitude2

28.9

1

28.9

96.4

0.0102

Scan rate

0.0212

1

0.0212

0.0760

0.809

Scan rate2

0.553

1

0.553

1.99

0.294

Lack of Fit

15.8

8

1.98

7.08

0.129

Pure Error

0.557

2

0.279

Total

112

16

Table 3. Analytical performance parameters of the electrochemical method for quantifying PMC. Parameter

Value

Sensitivity

90.2 mA mol−1 L

Range

2.00–219 µmol L−1

R2

0.9982

35

LOD

1.24 µmol L−1

LOQ

4.14 µmol L−1

RSDa

1.05%

a

For n = 10 at 50 µmol L−1 of PMC.

Table 4. Recovery tests for real samples. (n = 3) CPMC / µmol L−1 Sample

Tap water

Weir water

Recovery (%) Added

Found

29.13

27.43 ± 0.27

94.2

50.15

48.29 ± 0.27

96.3

50.15

44.43 ± 0.64

88.6

36

Table 5. Comparison of the PMC quantification method proposed in this paper with other electrochemical methods found in the literature. Method

Electrode material

Sample(s)

Limit of detection

Linear range (mol L−1)

(mol L−1)

Electrochemiluminescence [31]

MWCNT/Nafion/GC immobilized Natural water

2.0 × 10−9

8.0 × 10−9 to 1.0 × 10−6

with Square-wave voltammetry [4]

Biosensor LACC/ MWNT/GCPE

Tomato and lettuce

1.8 × 10−7

9.90 × 10−7 to 1.15 × 10−5

Square-wave voltammetry [32]

Biosensor LACC/PB/GPE

Tomato and potato

2.94 × 10−8 a

2.99 × 10−7 to 2.99 × 10−6

Differential pulse voltammetry [35]

Glassy carbon

Soil

6.1 × 10−7

1.0 × 10−5 to 10 × 10−5

Differential pulse polarography [33]

Mercury

Drinking, spring,

2.8 × 10−7

3.0 × 10−7 to 8.0 × 10−5

1.14 × 10−6

2.00 × 10−6 to 219 × 10−6

river and sea water Differential pulse voltammetry (this work) a

mol kg−1

Boron-doped diamond

Tap and weir water

37

MWCNT, multi-walled carbon nanotube; GC, glassy carbon; , tris(2,2’-bipyridyl)ruthenium; GCPE, graphite carbon paste electrode; LACC, laccase enzyme; PB, Prussian blue; GPE, graphene doped carbon paste electrode;