Square Wave Adsorptive Stripping Voltammetry

ISSN 1061-9348, Journal of Analytical Chemistry, 2018, Vol. 73, No. 7, pp. 707–715. © Pleiades Publishing, Ltd., 2018.

ARTICLES

Square Wave Adsorptive Stripping Voltammetry Determination of Сhlorpyriphos in Irrigation Agricultural Water1 Luisa C. Meloa, Murilo S. S. Juliãoa, Maria A. L. Milhomeb, Ronaldo F. do Nascimentob, Djenaine De Souza c, *, **, Pedro de Lima-Netob, and Adriana N. Correiab aCurso

de Química, Universidade Estadual Vale do Acaraú, Campus Betânia, Sobral-CE, CEP 62.040-379 Brazil Departamento de Química Analítica e Físico-Química, Centro de Ciências, Universidade Federal do Ceará, Fortaleza-CE, CEP 60455-970 Brazil c Campus de Patos de Minas, Instituto de Química, Universidade Federal de Uberlândia, Patos de Minas-MG, CEP 38700-126 Brazil *e-mail: [email protected] **e-mail: [email protected]

b

Received April 11, 2016; in final form, December 4, 2017

Abstract⎯This work describes the electroanalytical determination of Chlorpyriphos pesticide in natural waters using hanging mercury drop electrode allied to square wave adsorptive cathodic stripping voltammetry. The best responses were obtained in Britton‒Robinson buffer solutions at pH 2.0, using a frequency of 100 s–1, a scan increment of 5 mV, a square wave amplitude of 25 mV and an accumulation potential of –0.4 V during 60 s. Therefore, voltammetric responses showed the presence of one well-defined and irreversible reduction peak, at –1.08 V vs. Ag/AgCl/KCl 1.0 M, which involves two electrons in the reduction of carbon‒nitrogen bond in the N-heterocyclic system with the participation of protonation equilibrium preceding the electron transfer reaction. Analytical curves were constructed and compared to similar curves performed by gas chromatograph coupled to a selective nitrogen‒phosphorus detector, which demonstrate that the proposed methodology is suitable for determining contamination by Chlorpyriphos in complex samples. Keywords: Chlorpyriphos, organophosphorous pesticide, natural water, adsorptive square wave voltammetry, hanging mercury drop electrode DOI: 10.1134/S1061934818070109

The Ceará State in Brazil is located in a semi-arid climate region that promotes irrigated agriculture to become the third largest producer and exporter of tropical fruits (melon, pineapple, papaya, banana, mango, watermelon and cashew) in Brazil [1]. Meanwhile, this high productivity with suitable quality in the produced tropical fruits is obtained only by use of chemical compounds to control pest and sickness in plantations. Therefore, the Jaguaribe-Apodi, the largest and most important irrigated area in Ceará State, has made the area the largest user of pesticides [2]. These pesticides present high potential to contaminate natural water and disperse pesticide residues in all irrigated areas in Ceará State. Among the pesticides frequently used in tropical fruit production in Jaguaribe-Apodi, Chlorpyriphos, O,O-diethyl-O-(3,5,6-trichloro-2-pyridyl)phosphorothionate (CPF), is an organophosphorous compound that is employed as acaricide and insecticide. Its toxicity is associated with the ability to inhibit ace1 The article is published in the original.

tyl cholinesterase, an important enzyme in the central nervous system in humans or in the pests that attack the tropical fruit plantations [3, 4]. Therefore, to maintain the quality in the tropical fruits produced in Ceará State it is very important to have rigorous control of pesticide residues in tropical fruits produced and in the natural water used in irrigation of these cultures, using precise and sensitive analytical methodologies. The usual analytical methods employed to determine the amount of CPF are based on gas chromatography techniques coupled to a nitrogen‒phosphorous detector (GC-NPD) [5, 6], which assures good precision and accuracy for the determination of CPF. However, this technique first needs the steps of extraction and clean-up of the samples, increasing the time and cost of analysis. Additionally, the use of GC-NPD for CPF determination can promote the chromatographic column contamination. Nowadays, electroanalytical methods have largely been employed as an alternative to chromatographic methods for pesticide identification and determination, because they allow for detection limits compati-

707

708

LUISA C. MELO et al.

ble with environmental regulations. Among the electroanalytical methods available, square wave voltammetry (SWV) is a suitable method for pesticide detection, supplying detection limits that can be compared to those of chromatographic techniques [7, 8]. The use of SWV allied to adsorptive steps, like those that consist in an accumulation step to pre-concentrate the analyte into, or onto, the working electrode that is then electrochemically oxidized or reduced in the current measurement step [9], can promote an expressive increase in the analytical sensitivity. Furthermore, by employing the appropriate theoretical model developed for SWV it is possible to obtain information concerning the kinetics and mechanisms of electron transfer, identifying the intermediate reactive species, elucidating the mechanism of action of pesticides and supplying information about its redox properties in the environment [10]. However, very few studies have reported on the electroanalytical determinations or have studied the electrochemical behavior of CPF using solid surface electrodes, either bare or modified by adsorption of molecules, which has the drawback of an intense pesticide adsorption onto the electrodic surface, which damages the response reproducibility [11]. Some enzymatic electrodes prepared by the use of acetyl cholinesterase enzyme [12, 13] or DNA [14] have been reported by CPF determinations. However, in the enzymatic electrodes the detection efficiency was unsatisfactory because they were selective only for the chemical class, since all organophosphorous compounds can inhibit the enzyme responses and the use of DNA biosensor is vulnerable to organic solvents. Therefore, the aim of this work was to develop an electroanalytical methodology for evaluation of the electrochemical reduction reaction of CPF by square wave adsorptive cathodic stripping voltammetry (AS-SWV) coupled to hanging mercury drop electrode (HMDE). In addition, we employed the procedure for CPF determination in irrigated agricultural water used in the production of tropical fruits in Ceará State. EXPERIMENTAL Reagents and apparatus. All voltammetric measurements were performed on a potentiostat/galvanostat Autolab PGSTAT 30 from Metrohm/Eco Chemie coupled to GPES version 4.9 software (General Purpose Electrochemical System, Metrohm Autolab). A polarographic station model 663 Stand from Metrohm/Eco Chemie was used as the working electrode, with an Ag/AgCl/KCl 1.0 M as reference electrode and a graphite rod as auxiliary electrode. The working electrode system was operated in HMDE mode, and the surface, with an area of 0.52 mm2, was renewed after each measurement. A new drop was automatically formed in the system after dislodging

the old drop and extruding more triple-distilled mercury. A Micronal B474 pH meter equipped with a 3.0 M Ag/AgCl/KCl–glass combined electrode was used to adjust the pH values. All the solutions were prepared with water purified by a Milli-Q system (Millipore Corp.). CPF pesticide was provided by the Riedel-de Haën, with a purity of 99.9%. Stock solutions of CPF (1.0 × 10–3 and 1.0 × 10−5 M) were prepared daily by dissolving a suitable quantity of it in pure ethanol, which was then stored in a dark flask in a refrigerator to prevent degradation. All chemicals were of analytical-reagent grade. Britton‒Robinson (BR) buffer (0.04 M) was used as the supporting electrolyte, and the pH was adjusted to the desired value by adding suitable amounts of 0.2 M NaOH stock solution. Due to the lower solubility of CPF, the voltammetric experiments were performed using the mixture of BR buffer and ethanol in the proportion of 3 : 1 (v/v). A gas chromatography coupled to a nitrogen phosphorous detector system from Thermo Finnigan, model GC-Trace, in conjunction with an OV-5 capillary column (5% diphenyl and 95% dimethylpolysiloxane) with 30 m × 0.25 mm i.d. × 0.25 μm thickness was employed in chromatography data acquisition. Helium gas was used as carrier at a flow rate of 1.0 mL/min, as described by Munch and colleagues [15]. Instrument operation and data processing was carried out with Chrom Quest 4.1 software. All voltammetric and chromatographic measurements were performed in triplicate. Optimization of the procedure. All measurements were carried out under ambient conditions. The appropriate solutions were transferred into the electrochemical cell and the optimization of the analytical procedure for SWV was carried out. For this, a systematic study of the experimental parameters that affect the responses, such as the pH of the medium, the pulse potential frequency (f) related to total pulse duration, the amplitude of the pulse (a) and the height of the potential step (ΔEs) or scan increment were evaluated. All parameters were optimized in relation to the maximum value of the peak current and the maximum selectivity (half-peak width). Before each experiment, a stream of N2 was passed through the solution for 10 min to remove all oxygen, which can interfere with voltammetric measurements on an HMDE. The working electrode was then placed in the measuring cell, which contained 7.5 mL of an electrolyte support solution and 2.5 mL of ethanol. The organic solvent was used to ensure the solubility of CPF in the aqueous medium. A known concentration of CPF was added to this cell, after which the experimental and voltammetric parameters were studied. For measure-

JOURNAL OF ANALYTICAL CHEMISTRY

Vol. 73

No. 7

2018

SQUARE WAVE ADSORPTIVE STRIPPING VOLTAMMETRY DETERMINATION

ments, the supporting electrolytes Na2CO3, KCI and BR buffer were initially tested at the same concentrations. The best results were obtained in 0.04 M BR buffer. After optimizing the experimental and voltammetric parameters, analytical curves were obtained in pure electrolyte solutions by the standard addition method. The standard deviation of the mean current, measured at the reduction potential of CPF for 10 voltammograms of the blank solution in pure electrolytes (sb), together with the slope of the straight line of the analytical curves (S) were used in the determination of the quantification and detection limits (LOQ and LOD, respectively), according to guidelines recommended by the IUPAC [16, 17]. The recovery experiments were done in order to confirm the methodology’s efficiency. These experiments were carried out by adding a known amount of CPF to the supporting electrolytes followed by standard additions from the CPF stock solutions and plotting the resulting analytical curves. All measurements were performed in triplicate. The recovery efficiencies (R, %) were calculated considering the ratio between the value of the concentration obtained by extrapolating the analytical curves of the corresponding spiked samples and the concentration previously added [18]. The precision and accuracy of the methodology were tested with different standard solutions of CPF, and the relative standard deviations (RSD) were calculated, considering the standard deviation of the mean current values obtained and the mean peak current values. Application to natural water from irrigated areas. The suitability of the electroanalytical methodology in measuring CPF was tested by spiked water samples from two irrigated areas in Ceará, Brazil. The first samples were obtained in the Jaguaribe/Apodi region, from which natural water samples were collected at the main pumping station, the Barrage of Pedrinhas located in Limoeiro do Norte. The second sample was obtained in the Acaraú River lower region, from which water samples were collected from the Jaibaras weir in Sobral. The samples were stored under refrigeration until use. To perform the analysis, the samples were filtered and added to the electrochemical cell without any other pretreatment. For voltammetric experiments, 9.0 mL BR buffer solution was added in electrochemical cell next to 1.0 mL of each sample of water and 50 μL of stock solution of CPF (1.0 × 10–5 M). The recovery curves were constructed by the standard addition method, and the recovery percentage was obtained by graphical method, in which the abscissa axis refers to the concentration of CPF in the electrochemical cell. When the curve obtained is extrapolated to this axis, the sample concentration is obtained and the recovery values can be calculated according to the previously described method. All JOURNAL OF ANALYTICAL CHEMISTRY

Vol. 73

709

recovery curves were constructed using data from two different samples. Application to orange juice samples. The applicability of the proposed methodology was also evaluated using samples of orange juice, a crop that is common to both areas of irrigation. The orange juice was obtained using a manual juicer and filtered using filter paper with a porosity of 28 μm and permeability of 55 L/s m2 Afterward, 30 mL of the juice sample was centrifuged for 10 min at 4000 rpm. From the supernatant, 4 mL aliquots were spiked with 134 μL of 1.0 × 10–3 M CPF, shaken for 30 min and 17 μL of the spiked juice were transferred to the electrochemical cell containing 10 mL of supporting electrolyte (to give CPF concentration of 5.7 × 10–8 M) and analyzed by the proposed method. The recovery curves were constructed in triplicate and the recovery percentages were calculated as described above. RESULTS AND DISCUSSION Basic electrochemical investigation. Preliminary AS-SWV experiments, using 1.0 × 10–5 M CPF with 0.04 M BR buffer (pH from 2.0 to 10.0) and ethanol in the proportion of 3 : 1 (v/v) were conducted to determine the voltammetric responses of CPF on the HMDE. For this, the voltammetric parameters were kept constant, with a pulse frequency (f) of 100 s–1, a pulse amplitude (a) of 50 mV, and step potential (ΔEs) of 2.0 mV. Figure 1a shows the voltammetric responses obtained in different pH values, indicating that the CPF presents one well-defined reduction process, which was dislodging to more cathodic values when the pH was increased. Additionally, the increase in pH values promoted an intense decrease in Ip values, indicating that the participation of protonation equilibrium preceded the reaction of electron transfer, which indicates a general mechanism called chemical-electrochemical (CE) with protonation steps followed by an electron transfer. Therefore, when the CPF molecule is protonated, the redox process in HMDE is fast and promotes the best conditions under which the CPF can be determined. For this, pH 2.0 was chosen for further experiments. Relationships between peak current (Ip) and peak potential (Ep) with different pH values are shown in Fig. 1b, where it can be observed that the peak potential (Ep) shifts toward more negative values as the pH increases, which is represented by two linear equations, where Equation (1) represents the responses for pH values from 2.0 to 5.0 and Equation (2) represents similar responses in pH values higher than 5.0: Ep = ‒0.975 ‒ 0.058 pH, (1) Ep = ‒1.280 ‒ 0.004 pH, where values of the potential are given in volts.

No. 7

2018

(2)

710


I, µA (a)

pH –0.6

–0.4

2 3 4 5 6 7 8 9 10

–0.2

0 –0.8

–1.0

–1.2

–1.4

–1.6 E, V –Ip, mA

Ep, V 1.38

(b) 0.6

1.32 1.26

0.4 1.20 0.2

1.14 1.08 2

4

6

8

0

10 pH

Fig. 1. (a) ‒ Square wave adsorptive stripping voltammograms for 1.0 × 10–5 M clorpyriphos solution in BR buffer 0.04 M‒ethanol (3 : 1; v/v) on HMDE recorded with ΔEs = 2 mV, a = 50 mV, ƒ = 100 s–1, Ea = ‒0.4 V, ta = 60 s. (b) ‒ Relationship between pH values, peak potential (right on y axis) and peak currents (y axis) for 1.0 × 10–5 M CPF.

Equation (1) represents a straight line with a slope ΔEp/ΔpH that is very close to the theoretical value predicted by the Nernst equation for an electrochemical reduction process involving an equal number of protons and electrons. The slope in Equation (2) presents the value of –4 mV, which is a very small value compared to the 60 mV that corresponds to the same number of protons and electrons in the redox process. Additionally, a close analysis of both equations shows an intersection at pH 5.0; this pH is close to the value of the ionization constant (pKa) of CPF reported in the literature (pKa = 5.1 [19]), which may indicate the change in the redox mechanism for pH values above 5.0 [20].

Using the BR buffer with pH 2.0 and the voltammetric conditions described previously, AS-SWV was carried out, and the forward current (related to the reduction process), backward current (related to the oxidation process) and resultant components are shown in Fig. 2. The forward and backward currents are related to a reduction process since in both components the peak currents presented values in the same direction; consequently, the resultant current is slightly smaller than that of the corresponding resultant component. According to SWV diagnostic criteria for the electrochemical redox process, this behavior is a characteristic of irreversible processes where the electrode kinetics are slow and the chemical kinetics are fast. Furthermore, the results showed that the


Vol. 73

No. 7

2018


resultant peak current is located at approximately ‒1.12 V vs. Ag/AgCl 1.00 M. Additionally, some experiments using continuous scan potential from 0.0 to –1.25 V, without renovation of electrode, were performed, and the Ip was increasing with successive cycles. Therefore, an evaluation of the influence of the potential of the accumulation (Ea) and time of the accumulation (ta) of the species in the current values were evaluated. For this, the Ep values changed from 0.0 to –1.0 V, during times from 15 to 120 s by square wave voltammetry, similar to previous experiments. For Ep, there are no changes in its values with the changes in Ea and ta. However, for Ip the variation in Ea and ta revealed that the peak heights increased for Ea higher than ‒0.4 V and stayed practically constant for Ea lower than ‒0.4 V, Additionally, the increase in peak heights only was observed until 60 s of the ta; above this time, the peak heights stayed constant, considering all Ea. Therefore, the maximum of the Ip and analytical reproducibility in the responses was obtained employing ‒0.4 V as Ea for 60 s. Some cyclic experiments were developed in order to evaluate the electrochemical behavior of CPF at HMDE and compare to obtained information of ASSWV experiments. Therefore, when the potential was scanned from ‒0.8 to ‒1.2 V, the CPF responses revealed a well-defined peak in cathodic scan around ‒1.08 V and the absence of redox process in the anodic scan, confirming the irreversibility in the reduction reaction. Successive scan potentials were performed using scan rates (v) from 0.01 to 0.4 V/s, and the results presented a linear relationship between the Ip and the v, which, according to the theoretical criteria for cyclic voltammetry [21], may indicate processes featuring the adsorption of electroactive species on the electrode. This possibility was also confirmed by analysis of the logIp vs. logv, where the slope presented a value around 1.04, which is very close to 1.0, a value expected for a redox process controlled by surface adsorption. Optimization of the voltammetric parameters. As is well known, the voltammetric parameters in SWV are f, a and ΔEs, and these values can influence the height (related to sensibility) and width (related to selectivity) of the signal. Therefore, these parameters were individually studied and values of the peak current (Ip) and peak potential (Ep) were used to evaluate the voltammetric conditions to obtain the best analytical signal and to confirm information about electron-transfer and electrode kinetic mechanisms. The variation in the frequency of application of pulse potential usually exerts a marked effect on the response SWV. This effect thus provides a good criterion for diagnosis used to indicate any process of adsorption or reaction in solution, reversibility or irreJOURNAL OF ANALYTICAL CHEMISTRY

Vol. 73

711

I, µA Resulting Forward Backward

–0.8 –0.6 –0.4 –0.2 0 –0.9

–1.0

–1.1

–1.2 E, V

Fig. 2. Square wave adsorptive stripping voltammograms for 1.0 × 10–5 M clorpyriphos solution in BR buffer (pH 2.0) on the HMDE, recorded with ƒ = 100 s–1, a = 50 mV, ΔEs = 2 mV, Ea = ‒0.4 V and ta = 60 s and showing forward, backward and resultant currents.

versibility of the electrochemical process. Based on this, the f values were varied from 10 to 400 s–1. The results showed that an increase in f promoted an increase in Ip values and a shift toward more negative Ep values, accompanied by an increase in the halfheight width of the voltammetric peak with a loss in analytical selectivity. For this, for analytical purposes, the 100 s–1 was selected as suitable values for f value. Additionally, a linear relationship between the Ip and the f value was observed, which, according to the theoretical model proposed by Lovric et. al. for SWV [7], may indicate an irreversible redox process that is accompanied by a reactant adsorption process. An increase in the f values was accompanied by a shift toward more negative values in peak potentials, a behavior indicating that both peaks involve reactant and/or product adsorption as the rate-determining step in electron transfer. This behavior is in close agreement with the use of accumulation steps, as shown previously. The analysis of the influence of the f on the Ep values can be used to estimate the number of electrons transferred during the electrode process, by use of the relation developed by SWV parameters considering an irreversible process with adsorbed reactant, known as Lovric’s Equation [7]:

ΔE p = − 2.3RT , Δ log( f ) αnF

(3)

where R is the gas constant, T ‒ the temperature, α ‒ the electron transfer coefficient, n ‒ the number of electrons and F is the Faraday constant

No. 7

2018

712


Therefore, the relationship between Ep vs. f logarithm resulted in a straight line with slopes of 30.1 V. Considering αn = 1.0 (a typical value for accumulations steps), and substituting the known values of R, F and T, the room temperature, into the right-hand side of Equation (3) and the known value of the slope, n was determined to be equal to 2. Based on these findings, it can be ascertained that the most probable mechanism for the reduction of

Cl

Cl

Cl

S O P OC2H5 + H+ OC2H5

N

Cl

Cl

Cl

Cl

S O P OC2H5 + H+ + 2e− OC2H5

N H

+

CPF at HMDE involves the reduction of carbon‒nitrogen bond in the N-heterocyclic system with participation of protonation equilibrium preceding the electron transfer reaction, as shown in Scheme. This proposed mechanism is similar to the well-established general mechanism for N-heterocyclic compounds that in acid media undergoes a totally irreversible electrode reaction with a total uptake of the 2e–/2H+ [21].

Cl

Cl

H

S O P OC2H5 OC2H5

Cl

Cl

Cl

S O P OC2H5 OC2H5

N+

N H

Step 1

Step 2

Proposed mechanism for CPF reduction reaction.

The influence of a on the intensities of Ip was considered for values of a ranging from 5 to 50 mV. The results demonstrated that the increase in a values promoted an increase in Ip until 50 mV. However, Ip vs. a only presented linear relationships until 25 mV; above these values a suitable linearity was not observed. Additionally, the increase in a values promoted a displacement in Ep to more cathodic values, similar to what occurs in the irreversible redox process [7, 22]. Therefore, for the analytical proposal 25 mV was adopted. Finally, the voltammetric responses were evaluated using ∆Es from 2 to 10 mV, and the results indicated that an increase in ΔEs promotes an increase in Ip values and a change in Ep to the cathode region, without considerable variations in half-height width. As is well known, an increase in ∆Es promotes an increase in the scan rates of the voltammetric experiments, and consequently, for fast electron transfer process, it will also increase the signal and the sensitivity of the technique, as observed for CPF, so for subsequent experiments the ΔEs adopted was 5 mV. Analytical performance using AS-SWV. Using the optimized experimental (BR buffer pH 2.0, Ea = ‒0.4 V and ta = 60 s) and voltammetric (ƒ = 100 s–1, a = 25 mV and ΔEs = 5 mV) parameters, analytical curves were constructed for the range of concentration values from 2.5 × 10–8 to 2.5 × 10–7 M. The voltammograms obtained and the linear relationships between the Ip values and the added concentrations are shown in Fig. 3, where it can be observed that the voltammetric responses are linearly proportionally to the CPF concentration in the range evaluated. The

insert represents the middle values between the three constructed analytical curves. The valid analytical figures for the use of the proposed method was individually evaluated. Analytical curves were constructed in triplicate, and the data presenting middle values obtained. Initially, an examination of the relationship between values of Ip and added concentrations (insert in Fig. 3) showed that the analytical curves exhibit a resultant straight line, which was characterized by equation below: Ip (A) = 2.0 × 10–9 + 0.441[CPF].

(4)

The linearity information was evaluated by analysis of the determination coefficient (r 2) and the y intercept of the linear regression [23]. The r 2 presented a value of the 0.9993, which is considered evidence of acceptable fit of the data to the regression line (r2 > 0.998). Furthermore, the y intercept presented a value of 1.22% representing a decrease of 2% of the percentage of the analyte concentration confirming the linearity in the voltammetric data. The precision was evaluated from data of the reproducibility and repeatability using multiple voltammetric measurements with 7.44 × 10–8 M CPF and previously optimized parameters. The reproducibility experiments included five different measurements in different solutions at various days (inter-day precision), and the repeatability experiments included ten subsequent experiments performed in the same solution and at the same time (intra-day precision). In both, the RSD values were calculated as described in the “Experimental” section and presented values of 1.3 and 4.0% for repeatability and reproducibility,


Vol. 73

No. 7

2018


713

I, nA –Ip, nA –120 120 –90

80 40

–60 0

60

120 180 240 300 [CPF] × 10–9, M

–30

0 –0.9

–1.0

–1.1

–1.2 E, V

Fig. 3. Square wave adsorptive stripping voltammograms for clorpyriphos in 0.04 M BR buffer of pH 2.0 on the HMDE with f = 100 s–1, a = 25 mV, ΔEs = 5 mV, Ea = ‒0.4 V and ta = 60 s and concentrations in the interval from 2.5 × 10–8 to 2.5 × 10–7 M of CPF. The insert is the analytical curve was obtained from the voltammograms presented.

respectively. These are indicative of the good precision of the proposed methodology. The LOD and LOQ values were calculated to evaluate the sensitivity of the proposed methodology. For this, ten replicate blanks were analyzed to calculate the mean value of the peak current related to the peak potential value to reduce the CPF and the standard deviation of its values. The LOD and LOQ were individually calculated as described in the “Experimental” section, and the obtained values were 4.4 × 10–10 M (0.15 μg/L) and 1.5 × 10–9 M (0.52 μg/L), respectively. These LOD and LOQ values are very close to those previously published from other works that employed electroanalytical methodologies by use of the modified electrodes [11‒14]. However, the use of HMDE proposed in this work improves the reproducibility in the experiments and the effects of the other compounds present in complex samples, as will be shown in the end of this work. Furthermore, the LOD and LOQ values observed in the present work are satisfactory, as they are lower than the maximum values allowed by CONAMA (the National Counsil for the Environment – Brazil) for wastewaters, i.e., 20 μg/L [24, 25], and lower than the maximum value recommended by ANVISA (the National Agency for Sanitary Vigilance – Brazil) for its residues in food samples, i.e., from 50 to 100 μg/L [26]. The accuracy also was determined to evaluate the efficiency of the proposed method. For this, recovery curves were performed by spiking a known concentration of CPF (7.44 × 10–8 M, [CPF]added), either supporting electrolyte, and evaluating the voltammetric JOURNAL OF ANALYTICAL CHEMISTRY

Vol. 73

responses by the standard addition method [18]. The recovery percentages were used to evaluate and quantify the CPF that was added. In this way, the recovery concentration ([CPF]found) was identified graphically, with the abscissa axis denoting the concentration of CPF in the electrochemical cell. Extrapolating the curve along this axis yields the spiked concentration, allowing for the calculation of the recovery values (R, %), as shown in Equation (5) [18]:

%R =

[CPE]found × 100. [CPE]added

(5)

All the curves were plotted using three different samples. The average concentration recovered was 6.8 × 10–8 M, and the R was 91.7% with an RSD of 1.8%. These results are shown to be in a suitable range for analytical applications, which are acceptable values from 70 to 130% for recovery percentages [27], indicating that the present methodology can be successfully applied to the analytical determination of CPF with good efficiency and accuracy in the analytical results. Table 1 presents an abstract with all valid figures for the analytical determination of CPF using AS-SWV, as previously discussed. All data indicate that the proposed methodology provides an alternative method for the electrochemical determination of CPF, substituting modified electrodes that present the inconvenience of low reproducibility of the results, high cost of a modifying agent, or even the loss of signal by inactivation of the enzymes employed.

No. 7

2018

714


Comparison with a chromatographic method. To evaluate the applicability of the proposed methodology to CPF determination, the analytical results were compared to the official method for the quantification of the CPF in water samples established by the EPA, which is based on gas chromatography with a nitrogen-phosphorus detector [28]. Thus, analytical curves for the quantification of CPF were constructed using an OV-5 column (5% diphenyl, 95% dimethylpolysiloxane), and the temperature ramp was started at 100°C, then increased to 150°C at a rate of 15 grad/min and then to 230°C at 6 grad/min until 290°C at 30 grad/min. The temperatures of the injector and detector were 250 and 300°C, respectively. Under these conditions, 2.0 μL of CPF sample (solubilized in ethyl acetate) was injected in splitless mode, using hydrogen as the carrier gas at a flow rate of 1.0 mL/min; the retention time of CPF was 13.3 min. The analytical curves were constructed for the CPF concentration range of 1.4 × 10–7 to 2.8 × 10–6 M, resulting in a linear relation between the area of the chromatographic peak and concentration. These results were used to calculate the LOD and LOQ values similarly to previously calculated values to HMDE allied to AS-SWV. The calculated values using GC-NPD were 8.9 × 10–10 M (0.31 μg/L) and 2.7 × 10–9 M (0.95 μg/L), respectively, which confirms that the use of AS-SWV promoted the acquisition of better analytical sensitivity for the determination of CPF, showing the efficiency of the method proposed in this work. The recovery was also calculated by the standard addition method at three levels of fortification, obtaining a value of 100 ± 7%, indicating an excellent accuracy in the use of GC-NPD. The precision was expressed as repeatability and reproducibility of 2.6 and 6.9%, respectively. All valid parameters calculated by use of GC-NPD shown in Table 1 presented values similar to the recommended limits for analytical methods. However, these data were higher than those previously calculated using the AS-SWV, indicating that the proposed method shown in this paper presents suitable sensitivity, precision and accuracy and thus can be successfully used to identify and quantify low concentrations of CPF. Application to natural water from irrigated areas and orange juices. The robustness or reliability of the proposed method was evaluated in two natural water samples collected from irrigated areas, which contained different levels of organic matter, and in samples of the orange juice. Voltammetric experiments were performed using these samples to evaluate if the compounds present in complex samples, such as natural water and orange juice, will contribute to variability in the voltammetric responses. The water samples were used as collected, and 1 mL was added in electrochemical cells containing

9 mL of the supporting electrolyte. This procedure was aimed at identifying any influence of the matrix components on the quantification of CPF. It is important to highlight that, prior to the introduction of the first aliquot of CPF, the electrolyte was free from any signs of CPF. Consequently, the level of any possible contamination by this pesticide was below the detection limit of the proposed methodology. Thus, analytical curves were again obtained by ASSWV experiments. The analytical curves obtained under these new experimental conditions (containing 10% of the possible interferences) showed that the peak current values displayed some influence due to organic matter present in the water samples, as demonstrated in the analytical sensitivity, defined by the slope of the analytical curve, shown in Equation (6): Ip (A) = 2.0 × 10–9 + 0.380 [CPF].

(6)

The comparison with Equation (4) showed a decrease around 13% in the slope of the analytical curves in the natural water, which is related to the organic matter amount present in this sample, constituted by humic and fulvic acids being large molecules. In electrochemical cell, these compounds are capable of trapping the CPF molecule and hence hindering it from being adsorbed on the electrode surface. Since only an adsorbed molecule can block the reaction pathway, this causes a reduction in the amount of current observed in the recovery curve. Even so, the proposed method is suitable for the determination of CPF in natural waters, with low interference from the components of the samples. Besides, to evaluate the interference responses and consequently the selectivity of the proposed method, the recovery experiments were performed in triplicate for each sample using the standard addition method. Initially, the samples were artificially contaminated with 5.0 × 10–8 M CPF, and the standard addition method was employed to construct the recovery curves; the median values of R calculated are presented in Table 2. The R values demonstrate that these samples had very little effect on the peak current values of CPF. In addition, any observed differences between the recovery values obtained from pure electrolyte and the natural water samples were expected because these samples were never pre-treated to remove the organic and inorganic compounds that could possibly interfere with the analytical response. Still, the R figures obtained fall in a range suitable for analytical applications (an interval between 70 and 130% [27]). This indicates that the proposed method is suitable for the quantification of CPF in complex samples, such as natural water. CPF is one of the most widely used pesticides in citric fruit cultures produced in irrigated areas of Ceará. For this, the methodology developed in the


Vol. 73

No. 7

2018


preceding sections was hence applied to “in natura” orange juice, which was artificially contaminated, as previously described, resulting in a final CPF concentration of 8.0 × 10–8 M in the electrochemical cell. Successive additions of standard solutions of CPF were added (concentrations from 2.5 × 10–8 to 9.9 × 10–8 M) and the voltammetric responses were evaluated. These experiments were performed in triplicate without any removal of components from the samples. Only dilutions of the interferences were effected. All data obtained in orange juice showed that in the voltammetric responses an overlay of the background current occurs, maybe related to other matrix components. The R and RSD values were calculated and their values were 77.5 and 4.0%, respectively, which is considered acceptable for analytical methods, thereby demonstrating the efficiency of the proposed method. CONCLUSIONS The present study shows how the use of AS-SWV allied to HMDE with pre-concentration steps makes it possible to determine CPF at levels similar to those published in the literature using modified electrodes and lower than the maximum residue values permitted for natural water and orange juice, as stipulated by environmental agencies. All the results obtained, using the optimized experimental and voltammetric parameters, demonstrate that frequent and routine determination of CPF residues in natural water and orange juice can be safely carried out using simple and inexpensive electroanalytical methods without losing reliability or precision. Furthermore, the data presented in this work confirmed the practicality and viability of the proposed method, providing an important tool for evaluating the contamination of natural water used in irrigated areas, which can be contaminated by intensive use of pesticides. ACKNOWLEDGMENTS The authors wish to thank the Brazilian research funding institutions CNPq, FAPEMIG, CAPES and FINEP for their financial support. Luisa Célia Melo also wishes to thank CNPq and FUNCAP for the PhD grant. REFERENCES 1. Instituto de Pesquisa e Estratégia Econômica do Ceará. http://www2.ipece.ce.gov.br/publicacoes/ceara_em_ numeros/2011/economico/04_05_Comercio_Exterior.pdf. Accessed April, 2016. 2. Milhome, M.A.L., Sousa, D.O.B., Lima, F.A.F., and Nascimento, R.F., Eng. Sanit. Ambiental, 2009, vol. 14, p. 363. 3. Timchalk, C., Campbell, J.A., Liu, G., Lin, Y., and Kousba, A.A., Toxicol. Appl. Pharmacol., 2007, vol. 219, p. 217. JOURNAL OF ANALYTICAL CHEMISTRY

SPELL: 1. OK

Vol. 73

715

4. Agência de Vigilância Sanitária (ANVISA). http://portal.anvisa.gov.br/wps/portal/anvisa/home/agrotoxicotoxicologia. Accessed April, 2016. 5. Sánchez-Brunete, C., Martínez, L., and Tadeo, J.L., J. Agric. Food Chem., 1994, vol. 42, p. 2210. 6. Guardino, X., Obilis, J., Rosell, M.G., Farran, A., and Serra, C., J. Chromatogr. A, 1998, vol. 823, p. 91. 7. Mirceski, V., Komorsky-Lovrić, Š., and Lovrić, M., Square Wave Voltammetry: Theory and Applications, Berlin: Springer, 2007. 8. De Souza, D., Codognoto, L., Toledo, R.A., Pedrosa, V.A., Oliveira, R.T.S., Mazo, L.H., Avaca, L.A., and Machado, S.A.S., Quim. Nova, 2004, vol. 27, no. 5, p. 790. 9. Wang, J., Stripping Analysis: Principles, Instrumentation and Applications, Deerfield Beach: VCH, 1985. 10. Comninelles, C. and Chen, G., Electrochemistry for the Environment, New York: Springer, 2010. 11. Yan, F., He, Y., Ding, L., and Su, B., Anal. Chem., 2015, vol. 87, p. 4436. 12. Hildebrandt, A., Bragós, R., Lacorte, S., and Marty, J.L., Sens. Actuators, B, 2008, vol. 133, p. 195. 13. Viswanathan, S., Radecka, H., and Radecki, J., Biosens. Bioelectron., 2009, vol. 24, p. 2772. 14. Prabhakar, N., Sumana, G., Arora, K., Singh, H., and Malhotra, B.D., Electrochim. Acta, 2008, vol. 53, p. 4344. 15. Munch, J.W., Method 507: Determination of Nitrogenand Phosphorus-Containing Pesticides in Water by Gas Chromatography with a Nitrogen–Phosphorus Detector, Revision 2.1, 1995, 01-31.http://www.caslab.com/ EPA-Methods/PDF/507.pdf. Accessed April, 2016. 16. Mocak, J., Bond, A.M., Mitchel, S., and Scollary, G., Pure Appl. Chem., 1997, vol. 69, p. 297. 17. Analytical Methods Committee, Analyst, 1987, vol. 112, p. 199. 18. Skoog, D.A., West, D.M., and Holler, F.J., Fundamentals of Analytical Chemistry, Philadelphia: Saunders College, 1996. 19. Al-Meqbali, A.S.R., El-Shawi, M.S., and Kamal, M.M., Electroanalysis, 1998, vol. 10, p. 784. 20. Compton, R.G. and Banks, C.E., Understanding Voltammetry, London: World Sci., 2007. 21. Fry, A.J., Synthetic Organic Electrochemistry, New York: Wiley, 1989, 2nd ed. 22. O’dea, J.J., Ribes, A., and Osteryoung, J.G., J. Electroanal. Chem., 1993, vol. 345, p. 287. 23. Christian, G.D., Analytical Chemistry, New York: Wiley, 2004, 6th ed, 24. Conselho Nacional do Meio Ambiente (CONAMA, Environmental National Council in Brazil). http://www.aga-ambiental.com.br/leis/resolucaoconama20-86.doc. Accessed April, 2016. 25. Ministério da Saúde, Portaria no. 2.914, December 12, 2011. 26. Agência Nacional de Vigilância Sanitária (ANVISA). http://www.anvisa.gov.br/toxicologia/monografias/index.htm. Accessed April, 2016. 27. ICH-Q2Bn. Validations of Analytical Procedures: Methodology, International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use, Geneva, November 1996. 28. Other Clean Water Act Test Methods: Microbiological, Environmental Protection Agency (EPA). http:// water. epa.gov/scitech/methods/cwa/bioindicators/upload/ 2007_ 11_06_meth ods_method_507.pdf. Accessed April, 2016.

No. 7

2018