Electroanalytical determination of total phenolic compounds by square-wave voltammetry using a poly(vinylpyrrolidone)-modified carbon-paste electrode

Sensors and Actuators B 216 (2015) 192–197

Contents lists available at ScienceDirect

Sensors and Actuators B: Chemical journal homepage: www.elsevier.com/locate/snb

Electroanalytical determination of total phenolic compounds by square-wave voltammetry using a poly(vinylpyrrolidone)-modified carbon-paste electrode Jamille V. Piovesan ∗ , Cristiane L. Jost, Almir Spinelli Grupo de Estudos de Processos Eletroquímicos e Eletroanalíticos, Universidade Federal de Santa Catarina, Campus Universitário Reitor João David Ferreira Lima, Departamento de Química – CFM, 88040-900 Florianópolis, SC, Brazil

a r t i c l e

i n f o

Article history: Received 23 January 2015 Received in revised form 8 April 2015 Accepted 9 April 2015 Available online 20 April 2015 Keywords: Electrochemical sensor Carbon-paste Poly(vinylpyrrolidone) Kaempferol Total phenolic compounds

a b s t r a c t A carbon paste electrode (CPE) modified with poly(vinylpyrrolidone) (PVP) was employed as an electrochemical sensor for the determination of total phenolic compounds (TPCs) in vegetables by square-wave voltammetry (SWV). The calibration curve obtained using kaempferol as a model for TPCs showed two linear ranges from 0.05 to 0.50 ␮mol L−1 (R2 = 0.980) and from 0.50 to 6.0 ␮mol L−1 (R2 = 0.998). With the use of the most sensitive range, the limits of detection and quantification were 40 nmol L−1 and 160 nmol L−1 , respectively. The following sequence was determined for the content of TPCs in the vegetables analysed: spinach > cabbage > broccoli > chicory. The accuracy of the results provided by the proposed sensor was evaluated by comparison with the values obtained applying the Folin-Ciocalteau (FC) methodology. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Phenolic compounds are related to healthy dietary habits [1], since they are commonly found in fruits and vegetables. They are often present in grapes, apples and onions and in beverages such as fruit juice, soft drinks, wine and tea. As a consequence, the intake of foods rich in phenolic compounds has been recommended. The compounds reportedly have different beneficial effects on human health, since they can have anti-thrombotic, anti-bacterial, antiallergic and anti-inflammatory properties [2,3]. Since 1980, several studies have also shown that lower risk of chronic diseases is correlated with a diet containing fruits and vegetables rich in phenolic compounds [4,5]. In addition, many studies have indicated that the consumption of phenolic compounds reduces the occurrence of cancer [6]. Phenolic compounds are strong antioxidant reactants necessary for cell functioning. The mechanism of the antioxidant activity is mainly influenced by the number of hydroxyl groups and their position on the molecule ring [7]. Kaempferol

∗ Corresponding author at: Universidade Federal de Santa Catarina – UFSC, Campus Universitário Reitor João David Ferreira Lima, Centro de Ciências Físicas e Matemáticas – CFM, Departamento de Química, 88040-900 Florianópolis, SC, Brazil. Tel.: +55 48 3721 3606; fax: +55 48 3721 6850. E-mail address: [email protected] (J.V. Piovesan). http://dx.doi.org/10.1016/j.snb.2015.04.031 0925-4005/© 2015 Elsevier B.V. All rights reserved.

(5,7,4 trihydroxy-flavonol, Fig. 1) is an example of a phenolic compound with various natural sources, including apples, grapes, Ginkgo biloba, onions, leeks, citrus fruits and red wine [8]. Several clinical studies have shown that kaempferol and some glycosides of kaempferol also have a wide range of pharmacological application as an antioxidant, anti-inflammatory, anticancer, antimicrobial, cardioprotective, neuroprotective, anti-diabetic, anti-osteoporotic, anxiolytic, analgesic and anti-allergic agent [8]. For these reasons, kaempferol is an excellent model molecule for the determination of total phenolic compounds (TPCs). Concerning the relevance of phenolic compounds, several methods have been employed for their determination. Separation techniques such as chromatography [9,10] and capillary electrophoresis [11,12] have been used in order to study the properties of phenolic compounds and for their quantification. Optical methods [10,13] have also been widely used. In this context, the Folin-Ciocalteau (FC) methodology [13] is one of the most commonly used procedures for the determination of TPCs in food samples and plant extracts. However, low specificity is observed since the FC components react with other non-phenolic substances, resulting in an overestimation of the TPC content [14]. Various simpler and faster procedures employing electrochemical sensors have also been developed with appropriate selectivity and sensitivity for the determination of phenolic compounds [15–17]. In order to maximize the analytical response, the application of modified

J.V. Piovesan et al. / Sensors and Actuators B 216 (2015) 192–197

193

0.8

b

i / μA

0.6 0.4

a

0.2 0.0 -0.2

0.2

Fig. 1. Chemical structure of kaempferol.

electrodes with specific reactants has been reported [18]. For example, the well-known ability of the polymer poly(vinylpyrrolidone) (PVP) to extract phenolic compounds from fruit juices and plant extracts [19,20] has been exploited for the construction of electroanalytical sensors for single phenolic compounds [21–23]. The aim of this study was to apply a PVP-modified CPE for the quantification of total phenolic compounds (TPCs) instead of the determination of specific phenolic compounds. Kaempferol was used as the model molecule for the TPCs, and its content was determined in spinach, broccoli, cabbage and chicory samples by square-wave voltammetry. The accuracy of the analysis was evaluated by comparison with the results obtained applying the Folin-Ciocalteau methodology. 2. Experimental 2.1. Reagents, solutions and samples All reagents used in this study were purchased from Sigma–Aldrich. They were of analytical grade and used without further purification. The solutions were prepared with water purified using a Milli-Q system manufactured by Millipore (Bedford, MA, USA) with a resistivity of 18.2 M cm−1 . A stock solution of kaempferol was prepared at a concentration of 0.1 mmol L−1 in a mixture of water:ethanol (30:70 v/v). Britton–Robinson (B–R) and phosphate buffers were prepared at an initial concentration of 0.1 mol L−1 and tested as the supporting electrolyte. The pH was adjusted with 0.5 mol L−1 HCl or NaOH solution. Buffers and stock solutions of kaempferol were kept at 5 ◦ C for a maximum of 90 days. The PVP, with a molecular weight of 1,300,000 g mol−1 , was kindly provided by the Study Group on Polymeric Materials (POLIMAT-UFSC). For the determination of TPCs four different vegetables were used: spinach, broccoli, cabbage and chicory, all purchased in a supermarket in the city of Florianópolis, SC, Brazil. The vegetables were washed with distilled water and dried at a temperature of 25.0 ± 0.5 ◦ C. The extraction of TPCs from these matrices was carried out by soaking 5 g of fresh plant material (leaves, stems and flowers) with 100 mL of a solution of ethanol and water (70:30, v/v) for a period of 10 min. A further 100 mL of ethanol were then added and the mixture was sonicated for 10 min. Finally, the ethanolic extracts were filtered using a filter paper of medium porosity (25 ␮m) and stored at 5 ◦ C for 48 h. For the determination of TPCs, 100 ␮L of the ethanolic extracts were added to the electrochemical cell containing the supporting electrolyte. The TPC content was expressed as milligrams of kaempferol per gram of fresh plant material (mg g−1 ). Solutions of myricetin, luteolin, caffeic acid, ascorbic acid, rutin and quercetin, all at a concentration of 5.0 ␮mol L−1 , were electrochemically analysed to evaluate if these molecules are oxidized at the same potential as kaempferol. In this study, they cannot be considered interferents, but components of the plant material which can contribute to the content of total phenolic compounds.

0.4

0.6

0.8

E / V vs. Ag/AgCl Fig. 2. Cyclic voltammograms obtained using the CPE (a) and CPE/PVP sensor (b) for 10 ␮mol L−1 kaempferol in B–R buffer (pH 3.0), v = 100 mV s−1 .

2.2. Instruments and electrodes The voltammetric measurements were carried out on a potentiostat/galvanostat model PalmSens (Palm Instruments BV, The Netherlands) interfaced to a computer with the PSTrace software (version 2.5.2) for the acquisition and processing of data. A three-electrode electrochemical cell containing the CPE/PVP as the working electrode (A = 0.314 mm2 ), an Ag/AgCl electrode (3.0 mol L−1 KCl) as the reference and a Pt plate as the auxiliary electrode was used. All pH measurements were carried out using a pH meter model HI 2221 (HANNA Instruments Inc., Woonsocket, USA). The CPE/PVP was prepared by manual soaking of 10 mg (5% w/w) PVP and 160 mg (80% w/w) graphite powder for 10 min to obtain a uniform dispersion of the polymer in the graphite powder. Next, 30 mg (15% m/m) of mineral oil were added and macerated for 20 min to obtain the paste. The modified paste was introduced into a plastic syringe with a volume of 1.0 mL and a copper wire was inserted to obtain the electrical contact. To ensure an active and fresh surface, the modified CPE was gently manually abraded on a sheet of paper between experiments. The as prepared CPE/PVP sensor can be used until the complete depletion of the paste inserted in the syringe. For comparison purposes, electrodes with different amounts of PVP were prepared, as well as an unmodified CPE. For the determination of TPCs by the Folin-Ciocalteau method [24], a monochromatic Micronal spectrophotometer model B572 (Micronal SA, São Paulo, Brazil) and a glass cell with an optical path of 1.0 cm were used. 2.3. Electrochemical measurements Electrochemical measurements, namely cyclic voltammetry (CV) and square-wave voltammetry (SWV), were carried out in 10 mL of buffer solution (Britton–Robinson (B–R) or phosphate). The parameters of the SWV (Es – scan increment, a – amplitude and f – frequency) were optimized in the following ranges: Es = 1–10 mV, a = 10–100 mV and f = 10–100 Hz. For the construction of the calibration curve, successive additions of a stock solution of kaempferol were performed using a micropipette. After each addition, the solution was stirred in order to homogenize its composition. CV and SWV voltammograms were then recorded for the unstirred solution. 3. Results and discussion 3.1. Voltammetric behaviour of kaempferol using the CPE and CPE/PVP sensor Fig. 2 shows the CV results obtained for 10 ␮mol L−1 kaempferol in B–R buffer solution (pH 3.0) recorded at between +0.2 and +0.8 V. For the CPE (Fig. 2, curve a) a well-defined oxidation peak at +0.54 V was observed, while the reduction peak was detected at +0.44 V. For the CPE/PVP sensor (Fig. 2, curve b) the same peaks

194

J.V. Piovesan et al. / Sensors and Actuators B 216 (2015) 192–197

0.8

d

0.6 0.4

i / μA

were observed at +0.58 and +0.46 V, respectively. The separation between the reduction and the oxidation peaks was around 100 mV, characterizing the quasi-reversibility of the reaction. The oxidation peak is attributed to the reaction of the hydroxyl group located at the 4 position of the B ring in a process involving one proton and one electron, leading to the formation of the phenoxyl radical [25]. This oxidation peak was used as the analytical signal for the detection of kaempferol and the determination of TPCs, as will be shown later. The anodic current obtained using the CPE/PVP sensor is higher than that obtained using the CPE. This behaviour is attributed to the presence of hydrogen bonding between the imide group present in PVP and the hydroxyl groups present in kaempferol.

a

0.2 0.0 -0.2

0.2

0.4

0.6

0.8

E / V vs. Ag/AgCl Fig. 3. Cyclic voltammograms obtained with 10 ␮mol L−1 kaempferol in B–R buffer (pH 3.0) using the CPE/PVP sensor with (a) 30%, (b) 20%, (c) 10% and (d) 5% PVP (w/w), v = 100 mV s−1 .

3.2. Optimization of experimental conditions In order to optimize the CPE/PVP sensor response to study the electrochemical behaviour of kaempferol and the determination of TPCs, some experimental parameters were evaluated, such as percentage of PVP in the carbon paste, solution pH and composition of the supporting electrolyte. Fig. 3 shows the cyclic voltammograms obtained with 10 ␮mol L−1 kaempferol using the CPE with different amounts of PVP. It can be observed that an increase in the percentage of PVP did not lead to an increase in the analytical response. In fact, amounts of PVP in the carbon paste higher than 5% decreased the anodic and cathodic currents. This occurred because the PVP is not a conductive polymer. Thus, the higher the PVP content in the carbon paste, the lower the conductivity of the sensor and, as a consequence, the current decreases. Therefore, the amount of PVP chosen for the preparation of the sensor was 5% (w/w). The pH conditions investigated in this study were varied within the range of 2.0–12.0 using B–R buffer solution initially prepared at 0.1 mol L−1 . It was observed that the cyclic voltammograms with the best profile were obtained at acidic pH (Fig. 4A), as previously reported for quercetin and other phenolic compounds with

0.8 0.6

A

ed

chemical structures similar to that of kaempferol [21,23]. In alkaline solutions (Fig. 4B), the peaks are broad and the current decays quickly with increasing pH. At pH values above 10.0 no redox signal was observed. The potential of the oxidation peak – Epo shifted to less positive values with increasing pH (Fig. 5A). The slope of the straight line (−64.3 mV pH−1 ) is characteristic of a process that involves the same number of protons and electrons (−59.2 mV pH−1 ). The highest anodic peak current – ipa was obtained at pH 3.0, as shown in the graph of Fig. 5B. This solution pH was selected for further experiments. To evaluate the influence of the chemical composition of the supporting electrolyte on the sensor response, CV measurements were assessed in two buffers whose buffering capacity is valid at pH 3.0, that is, phosphate and B–R buffer solutions, both with an initial concentration of 0.1 mol L−1 . Fig. 6 shows the voltammograms obtained in the proposed two supporting electrolytes. The best response for the peak current and the best voltammetric profile were obtained in B–R buffer, and this solution was used for the development of the methodology. In this solution, an increase in

b c a

0.4 i / μA

i / μA

0.4 0.2 0.0

-0.2

a

B

b c

0.2

d 0.0

-0.2 0.0

0.2

0.4

0.6

0.8

E / V vs. Ag/AgCl

1.0

-0.2

0.0

0.2

0.4

0.6

E / V vs. Ag/AgCl

Fig. 4. Cyclic voltammograms obtained with 10 ␮mol L−1 kaempferol in B–R buffer using the CPE/PVP sensor; (A) (a) pH 2.0, (b) pH 3.0, (c) pH 4.0, (d) pH 5.0 and (c) pH 6.0; (B) (a) pH 7.0, (b) pH 8.0, (c) pH 9.0 and (d) pH 10.0; v = 100 mV s−1 .

0.7 0.5

0.4

0.4

0.3

0.3

0.2 0.1

0.2 0.1

B

0.5 i pa

E po

0.6

A

0.6

0.0 2

4

6 pH

8

10

2

4

6 pH

8

10

Fig. 5. Influence of the solution pH on the anodic peak (A) potential and (B) current for 10 ␮mol L−1 kaempferol in B–R buffer using a CPE/PVP sensor.

J.V. Piovesan et al. / Sensors and Actuators B 216 (2015) 192–197

0.8

i / μA

i / μA

4

b

0.2

3

1

-0.2

0 0.2 0.4 0.6 E / V vs. Ag/AgCl

0.8

1.0

b

2

0.0

0.0

a

5

a

0.6 0.4

195

c 0.3

0.4

0.5

0.6

0.7

0.8

E / V vs. Ag/AgCl

Fig. 6. Cyclic voltammograms obtained with 10 ␮mol L−1 kaempferol in (a) B–R and (b) phosphate buffers (pH 3.0) using the CPE/PVP sensor, v = 100 mV s−1 .

the reversibility of the reaction was also observed, expressed by a reduction peak not observed in phosphate buffer solution. This result also contributes to the choice of the supporting electrolyte, because it can subsequently be exploited to achieve a higher analytical signal using SWV.

Fig. 8. Voltammograms obtained applying (a) SWV, (b) DPV and (c) LSV with 10 ␮mol L−1 kaempferol in B–R buffer (pH 3.0) using the CPE/PVP sensor, v = 100 mV s−1 .

with the oxidation of kaempferol on the surface of the CPE/PVP sensor can be used for its electroanalytical determination without any accumulation step. 3.4. Choice of electroanalytical technique

3.3. Influence of the potential scan rate The cyclic voltammograms in Fig. 7A show the electrochemical behaviour of kaempferol in B–R buffer (pH 3.0) using the CPE/PVP sensor. The voltammograms were obtained using different scan rates (10–500 mV s−1 ). It was observed that on increasing the scan rate there was an increase in the peak current for both the oxidation and reduction reactions. The plot of Fig. 7B shows the currents as a function of the square root of the scan rate. As can be seen, a linear dependence is observed for lower scan rates. A deviation from linearity is observed for higher scan rates, indicating that the reactions are no longer diffusion-controlled. The value for the slope of the plot of log ip vs. log v for the oxidation reaction (Fig. 7C) was 0.58, indicating that this reaction is diffusion-controlled with a minimum contribution from adsorption. For the reduction reaction, a slope of 0.16 suggests that the mechanism is not purely diffusional. Thus, the results reported above indicate that the reaction associated

Three different electroanalytical techniques were evaluated to test their sensitivity in the detection of kaempferol: linear scan voltammetry – LSV, differential-pulse voltammetry – DPV and square-wave voltammetry – SWV. The parameters of the three techniques were adjusted to obtain the same scan rate of 100 mV s−1 . Fig. 8 shows the voltammograms obtained. As can be observed, the highest sensitivity was obtained using SWV, which is due to the suppression of the capacitive current in the SWV technique. Additionally, for reversible or quasi-reversible reactions, as is the case for kaempferol, the difference between the current produced by the cathodic and anodic pulses (ip = ipc − ipa ) gave a voltammogram with a peak current that was around twice that of the individual currents. Hence, SWV was applied in the potential interval of +0.25 to +0.80 V in the subsequent experiments. The optimized values for the SWV were Es = 3 mV, a = 100 mV and f = 100 Hz.

2.0

A

o

1.5

B ipa

1.0 i / μA

i / μA

2.0 1.5 1.0 0.5 0.0 -0.5 -1.0

a

0.5 0.0

-0.5

ipc

-1.0 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 E / V vs. Ag/AgCl

log i / μA

0.3

3

6

9 12 15 18 21 24 ν 1/2 / (mV s -1) 1/2

C ipa

0.0 -0.3

ipc

-0.6 -0.9 1.0

1.5 2.0 2.5 log ν / mVs -1

3.0

Fig. 7. (A) Cyclic voltammograms obtained with 10 ␮mol L−1 kaempferol in B-R buffer (pH 3.0) using the CPE/PVP sensor, (a–o) v = 10–500 mV s−1 ; (B) plot ip vs. v1/2 , and (C) plot log i vs. log v.

196

J.V. Piovesan et al. / Sensors and Actuators B 216 (2015) 192–197

25 20

A

p

Δi / μA

Δi / μA

15 10

a

5 0

0.2 0.3 0.4 0.5 0.6 0.7 0.8

E / V vs. Ag/AgCl

21 18 15 12 9 6 3 0

B II I 0

1

2

3

4

5

6

[Kaempferol] / μmol L-1

Fig. 9. (A) Square-wave voltammograms for (a) blank, (b–p) successive additions of 0.1 mmol L−1 kaempferol in B-R buffer (pH 3.0) using the CPE/PVP sensor, Es = 3 mV, a = 100 mV and f = 100 Hz; (B) the calibration curve for kaempferol. Table 1 Analytical parameters for the determination of kaempferol using the CPE/PVP sensor. Analytical parameter

Range I

Peak potential (V) Linear range (␮mol L−1 ) Correlation coefficient (R2 ) Slope (␮A L ␮mol−1 ) Standard deviation of slope (␮A L ␮mol−1 ) Intercept (␮A) Standard deviation of intercept (␮A) LOD (␮mol L−1 ) LOQ (␮mol L−1 ) Repeatability of i (intra-day) (%)a,b Repeatability of i (inter-day) (%)a,b

0.05–0.50 0.980 10.75 0.50 0.24 0.17 0.04 0.16 3.32 3.68

a b

Range II +0.52 0.50–6.0 0.998 2.97 0.04 4.29 0.16 0.16 0.54 – –

Relative standard deviation. n = 5.

3.5. Calibration curve Applying the optimized experimental conditions discussed above, the calibration curve was obtained by performing successive additions of a standard solution of kaempferol to the supporting electrolyte solution. The square-wave voltammograms obtained are shown in Fig. 9A. A well-defined peak was observed at +0.52 V. It is evident that the current increased and the peak potential did not shift with increasing kaempferol concentrations. The calibration curve (Fig. 9B) presented two linear ranges: the first (higher slope, range I) of 0.05–0.50 ␮mol L−1 and the second (lower slope, range II) of 0.50–6.0 ␮mol L−1 . Some analytical parameters are shown in Table 1. The calibration curve related to the first linear range can be expressed by the equation i/␮A = 0.24 + 10.75 [kaempferol]/␮mol L−1 . The limit of detection (LOD) and limit of quantification (LOQ) were calculated according to the equations: LOD = 3.0Sb /B and LOQ = 10.0Sb /B, where Sb is the standard deviation of the linear coefficient and B is the slope of the curve. The limits of detection (LOD) and quantification (LOQ) were 0.04 and 0.16 ␮mol L−1 , respectively. The two linear ranges of the calibration curve obtained in this study were somewhat higher than those obtained in previous studies [23] using the same sensor but for different phenolic compounds. In addition, the LOD and the LOQ achieved in this study were significantly lower than the previously published values of 0.17 and 0.52 ␮mol L−1 , respectively. The relative standard deviations for measurements of the anodic current carried out in intra- and inter-day studies were lower than 4%, indicating the excellent repeatability furnished by the proposed sensor. 3.6. Electrochemical evaluation of other phenolic compounds using the CPE/PVP sensor In plant extracts, as is well-known, many phenolic compounds are present besides kaempferol. Hence, for the determination of the

TPC content of a sample, it is necessary to verify that the proposed methodology is sensitive to all (or most) of these compounds. In this regard, solutions of myricetin, luteolin, caffeic acid, ascorbic acid, rutin and quercetin, all at a concentration of 5.0 ␮mol L−1 , were electrochemically analysed using the CPE/PVP sensor developed, in order to evaluate if these molecules are oxidized at a potential the same as (or close to) that of kaempferol. These compounds were evaluated because they are normally present in high concentrations in plant extracts [10,14]. With the exception of ascorbic acid, all compounds analysed were oxidized at the CPE/PVP surface. Myricetin was oxidized at +0.37 V, quercetin at +0.44 V, caffeic acid at +0.48 V, rutin at +0.51 V, kaempferol at +0.52 V and luteolin at +0.54 V in B-R buffer at pH 3.0. The relative response of the anodic current was of the same order of magnitude for all phenolic compounds. Thus, instead of interferents, the analysed phenolic compounds should be considered as contributors to the current measured. These experiments confirmed that the proposed CPE/PVP sensor is suitable for the determination of total phenolic compounds. 3.7. Determination of total phenolic compounds (TPC) in samples of plant extracts The main components of the analysed vegetables spinach, cabbage, broccoli and chicory are gallic acid, quercetin and kaempferol. As is well-known, gallic acid is the most used component to quantify total phenolic compounds (TPC) because it is a simpler molecule. On the other hand, kaempferol is very little used as model molecule to quantify TPC. Hence, we investigated the electrochemical behaviour of kaempferol and its applicability as standard compound to determine the level of TPC in some vegetables using the proposed CPE/PVP sensor. For the voltammetric determination of TPCs in plant extracts, 100 ␮L of the ethanolic extract of chicory, broccoli, cabbage and spinach were added to the electrochemical cell. Square-wave voltammograms were recorded in the range of +0.35 to +0.7 V, as shown in Fig. 10. The analysis was carried out in triplicate and the TPC was determined by the method of external calibration using the calibration curve for kaempferol. The final concentration was expressed as milligrams of kaempferol per gram of fresh plant material (mg g−1 ). The values for TPC content obtained for each sample using the proposed sensor and the Folin-Ciocalteau (FC) method are shown in Table 2, together with the relative error between the results obtained applying the two methods. The vegetable with the highest content of TPCs determined with the proposed sensor was spinach (3.20 mg g−1 ), followed by cabbage (2.90 mg g−1 ), broccoli (2.10 mg g−1 ) and finally chicory (0.80 mg g−1 ). The values for TPC content determined by the FC method for all plant extracts analysed were higher than the corresponding values obtained using the CPE/PVP sensor. As reported in previous

J.V. Piovesan et al. / Sensors and Actuators B 216 (2015) 192–197

3.0 2.5

surface is easily renewable and in association with square-wave voltammetry it provided low limits of detection and quantification. The results obtained support the conclusion that the proposed CPE/PVP sensor associated with square-wave voltammetry is a potential methodology for the determination of the TPC content in plant matrices without the need for elaborate sample preparation and pre-concentration steps, which significantly reduce the analysis time.

d

Δi / μA

2.0 1.5

a

1.0 0.5 0.0 0.3

0.4

0.5

0.6

197

0.7

Acknowledgements

E / V vs. Ag/AgCl Fig. 10. Square-wave voltammograms for ethanolic extracts of (a) chicory, (b) broccoli, (c) cabbage and (d) spinach in BR buffer 0.1 mol L−1 (pH 3.0) obtained using the CPE/PVP sensor, Es = 3 mV, a = 100 mV and f = 100 Hz. Table 2 Content of total phenolic compounds in samples of plant extracts (mg kaempferol g−1 of fresh plant material). Sample

CPE/PVP

Spinach Broccoli Cabbage Chicory

3.20 2.10 2.90 0.80

± ± ± ±

0.03 0.03 0.02 0.01

FC 3.37 2.44 3.08 0.88

Relative error ± ± ± ±

0.03 0.01 0.02 0.01

−5.04% −13.9% −5.84% −9.09%

publications [26–28], the values obtained applying the FC method are generally higher than those obtained using electrochemical sensors. This behaviour is attributed to the different oxidizing agents used in each case, i.e., the chemical reagents used in the FC method (phosphotungstic–phosphomolybdic acids) and the potential applied for the oxidation reaction on the electrode surface in the electrochemical sensors [27]. In the first case, the FC reagents oxidize not only phenolic compounds but also other non-phenolic species in the sample (aromatic amines, glucose, ascorbic acid and sulphur dioxide) [24,27,29]. In the second case, the applied potential oxidizes only the phenolic compounds. Despite this difference, the accuracy provided by the CPE/PVP sensor is comparable to that of the FC method, taking into consideration that the maximal relative error between the determinations was around 15%. The precision of measurements obtained with the two procedures was also comparable, indicating that the CPE/PVP sensor permits the determination of TPCs with the same reliability as the FC method.

The authors are grateful to the Brazilian government agencies CNPq and CAPES for financial support and the scholarship awarded to JP and to POLIMAT-UFSC for providing the PVP. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24]

[25] [26]

4. Conclusions

[27]

In this study, electrochemical experiments using a CPE/PVP sensor showed that kaempferol is oxidized at around +0.58 V in B-R buffer (pH 3.0). The resulting product is reduced at around +0.46 V. Under favourable conditions, this reaction is quasi-reversible and the process is diffusion-controlled. The calibration curve for the kaempferol showed two linear ranges at concentrations between 0.05 and 0.50 ␮mol L−1 (R2 = 0.980) and 0.50 and 6.0 ␮mol L−1 (R2 = 0.998). Using the most sensitive range, the limits of detection and quantification were 40 nmol L−1 and 160 nmol L−1 , respectively. The sensor was applied to the electroanalytical determination of total phenolic compounds by square-wave voltammetry in samples of plant extracts. The content of TPCs was expressed as mg of kaempferol per g of fresh plant material. The plant that showed the highest content of phenolics was spinach (3.20 mg g−1 ), followed by cabbage (2.90 mg g−1 ), broccoli (2.10 mg g−1 ) and chicory (0.80 mg g−1 ). The precision and accuracy of the results provided by the proposed sensor were evaluated by comparison with the results obtained using the FolinCiocalteau method. The proposed CPE/PVP sensor is of low cost, the

[28] [29]

A. Wojdyło, J. Oszmianski, R. Czemerys, Food Chem. 105 (2007) 940. C. Bianchini, A. Curulli, M. Pasquali, D. Zane, Food Chem. 156 (2014) 81. N. Balasundram, K. Sundram, S. Samman, Food Chem. 99 (2006) 191. A. Cifuentes, Electrophoresis 27 (2006) 283. P. Kris-Etherton, C.L. Keen, Curr. Opin. Lipidol. 13 (2002) 41. P.L. Teissedre, A.L. Waterhouse, J. Agric. Food Chem. 48 (2000) 3801. S. Hogan, L. Zhang, J. Li, H. Wanga, K. Zhou, Food Chem. 117 (2009) 438. ˜ E. Burgos-Morón, C. Pérez-Guerrero, M. López-Lázaro, J.M. Calderón-Montano, Mini-Rev. Med. Chem. 11 (2011) 298. Z. Spáˇcil, L. Nováková, P. Solich, Talanta 76 (2008) 189. M. Naczk, F. Shahidi, J. Chromatogr. A 1054 (2004) 95. E. Hurtado-Fernandez, M. Gomez-Romero, A. Carrasco-Pancorbo, A. FernandezGutierrez, J. Pharm. Biomed. 53 (2010) 1130. R.G. Peres, G.A. Micke, M.F.M. Tavares, D.B. Rodriguez-Amaya, J. Sep. Sci. 32 (2009) 3822. V.L. Singleton, J.A. Rossi, Am. J. Enol. Vitic. 16 (1965) 144. A. Escarpa, M.C. González, Anal. Chim. Acta 427 (2001) 119. C. Jianrong, M. Yuqing, H. Nongyue, W. Xiaohua, L. Sijiao, Biotechnol. Adv. 22 (2004) 505. A. Merkoc¸i, Biosens. Bioelectron. 26 (2010) 1164. B. Pérez-López, A. Merkoc¸i, Trends Food Sci. Technol. 22 (2011) 625. M. Buleandra, A.A. Rabinca, C. Mihailciuc, A. Balan, C. Nichita, I. Stamatin, A.A. Ciucu, Sens. Actuator B: Chem. 203 (2014) 824. Z. Borneman, V. Gokmen, H.H. Nijhuis, Sep. Purif. Technol. 53 (2001) 22. W.D. Loomis, J. Battaile, Phytochemistry 5 (1966) 423. A.C. Franzoi, A. Spinelli, I.C. Vieira, J. Pharm. Biomed. 47 (2008) 973. X. Yang, F. Wang, S. Hu, Colloids Surf. B 52 (2006) 8. J. Piovesan, A. Spinelli, J. Braz. Chem. Soc. 25 (2014) 517. A.L. Waterhouse, Determination of total phenolics, in: R.E. Wrolstad (Ed.), Current Protocols in Food Analytical Chemistry, John Wiley & Sons, Inc., New York, 2002. J. He, S. Yuan, J. Du, X. Hu, Y. Wang, Bioelectrochemistry 75 (2009) 110. ˜ ˜ J.M. Pingarrón, Anal. V.C. Sanz, M.L. Mena, A. Gonzélez-Cortés, P. Yénez-Sede no, Chim. Acta 528 (2005) 1. P. Ibarra-Escutia, J. Gómez, C. Calas-Blanchard, J.L. Marty, M.T. Ramírez-Silva, Talanta 81 (2010) 1636. M.R. Montereali, L. Della Seta, W. Vastarella, R. Pilloton, J. Mol. Catal. B: Enzym. 64 (2010) 189. A.M. Granero, H. Fernández, E. Agostini, M.A. Zón, Talanta 83 (2010) 249.

Biographies Jamille Valéria Piovesan received her master’s degree in 2014 from the Federal University of Santa Catarina, Florianópolis, Brazil. Currently, she is studying for a doctorate at the same university. Her research work is concentrated on the development of new sensors for the electroanalytical determination of organic compounds of biological and environmental interest. Cristiane L. Jost received her doctorate in Analytical Chemistry in 2010 from the Federal University of Santa Maria, Santa Maria, RS, Brazil. Currently, she is a full-time professor and researcher at the Department of Chemistry of the Federal University of Santa Catarina, Florianópolis, Brazil. Her research focuses on the electroanalytical determination of heavy metals and organics in environmental samples. Almir Spinelli received his doctorate in Applied Electrochemistry from the University of Poitiers, France, in 1992. Currently, he is a full-time professor and researcher at Department of Chemistry of the Federal University of Santa Catarina, Florianópolis, Brazil. His research focuses on electrochemical processes, as well as electroanalytical determinations and corrosion studies.