1597
Short Communication
Direct Determination of Vitamin C in Fruit Juices Using a Polyviologen-Modified Electrode Jyh-Myng Zen,* Dong-Mung Tsai, Hsueh-Hui Yang Department of Chemistry, National Chung-Hsing University, Taichung 402, Taiwan; e-mail:
[email protected] Received: March 12, 2002 Final version: April 23, 2002 Abstract A polyviologen-modified glassy carbon electrode (PV-GCE) was applied to directly determine vitamin C (i.e., ascorbic acid) in deeply colored, viscous, and turbid fruit juice samples. Compared to the results obtained at a bare GCE, the oxidation of vitamin C showed a large increase in current response at the PV-GCE in pH 4 buffer solutions. Preconcentration is the major factor for the signal enhancement. Square-wave voltammetry (SWV) was used for vitamin C detection at the PV-GCE. Under the optimized conditions, the SWV current signals showed a very wide linear range (up to 900 mM) with slope and regression coefficient of 0.051 mM mA 1 and 0.998, respectively. The detection limit (S/N 3) was 0.38 mM after 20 s of accumulation at open circuit. Since the content of vitamin C and pH value of most fruit juices are lower than 900 mM and close to pH 4, the determination can be done without dilution and other pretreatment of the samples. The results demonstrated a good precision and are in agreement with those obtained with other method. Keywords: Polyviologen, Vitamin C, Fruit juice
Vitamin C (i.e., ascorbic acid) is found in high concentration in some fruits and foods together with vasoactive amines [1]. It is an effective reducing agent and a powerful antioxidant in food and is known to take part in several biological reactions [2]. Rapid monitoring of vitamin C levels during production and quality control stages are an important step in industrial applications. The detection and determination of vitamin C is thus a continuing research interest. Several techniques are reported for vitamin C determination including spectroscopic, chromatographic, enzymatic, and electroanalytical methods, each with their advantages and disadvantages [3 ± 7]. Compared to other options, electroanalysis is relatively simple and sensitive. However, previous chemical modified electrodes still have a lot to improve regarding stability and sensitivity [8 ± 12]. Among available electrochemical methods, polymer-modified electrodes appear to have distinct advantages for their high catalytic ability, good stability, and broad potential window [12 ±16]. The conducting polymer films on electrode surfaces play an essential role in controlling the electrochemical reaction rate [17]. Electrodes modified with polyviologen (PV), which belongs to the class of ionic polymer named ionenes, were prepared by reductive electropolymerization of a trifunctional monomer with three 4-cyanopyridinium moieties [18]. Viologen derivatives were considered an effective reductant, especially for dioxygen [19, 20]. In addition, electrodes derivatized with the viologen reagents have been shown to be useful in hydrogen evolution catalysis [21, 22], redox reactions of biological molecules [23], binding of redox reactive metal complexes [24] and electrochromism [25]. However, the polyviologen-modified glassy carbon electrodes (PV-GCEs) have never been reported for the Electroanalysis 2002, 14, No. 22
determination of vitamin C. In the present study, we investigated the electrochemical behavior of vitamin C on the PVGCEs. Square-wave voltammetry (SWV) was used for the direct determination of vitamin C in fruit juices without either dilution or other pretreatment of the samples. Figure 1 shows cyclic voltammetric response of vitamin C on a bare GCE and the PV-GCE. An anodic oxidation peak was noticed at 0.40 V (vs. Ag/AgCl) on a bare GCE. For the PV-GCE, the peak potential (Epa) was shift positively to 0.22 V with ca. 9 times increase in the current magnitude. The increase in peak current was due to two factors. The first one is the electrocatalytic effect of the PV-GCE. However, the small decrease in overpotential (h) indicated this effect is modest. The other factor is the assistance of an electrotrostatic accumulation effect in the polyviologen matrix. Indeed, preconcentration, as shown below, is important to cause thek current and will be discussed later. GCE-(PV2 .2Br)n 2nA . $ GCE-(PV2 .2A . )n 2nBr (1)
Scheme 1.
¹ 2002 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 1040-0397/02/2211-1597 $ 17.50+.50/0
1598
Fig. 1. CV responses for a) the PV-GCE in pH 4.0 buffer solution and b) GCE and c) the PV-GCE in the presence of 1 mM vitamin C at a scan rate of 100 mV s 1.
Fig. 2. Effect of pH on the vitamin C oxidation potential (Epa) and its current (ipa) by SWV. [Vitamin C] 1 mM. Preconcentration time 20 s. SWV parameters are: fHz 100 Hz; Eamp 50 mV, and Estep 10 mV.
The effect of solution pH on the peak current (ipa) and Epa is shown in Figure 2. The Epa showed a shift toward less positive value as the pH increased. Interestingly, the intercept is around pH 4, which is close to the pKa1 of vitamin C. The ipa was found to increase sharply up to ca. pH 4 and decrease Electroanalysis 2002, 14, No. 22
J.-M. Zen et al.
afterwards. The ipa increased from pH 2 to 4 could be attributed the increase of the concentration of A . on the PVGCE surface as indicated in Equation 1. Thus, pH 4 was selected as an optimum condition in subsequent study. Note that the pH value of commercial fruit juices is close to pH 4. To optimize the SWV response of vitamin C, the corresponding instrumental SW parameters like frequency (fHz), amplitude (Eamp), and step potential (Estep) were studied. The optimized parameters were summarized as follows: fHz 100 Hz, Eamp 50 mV, and Estep 10 mV. As stated previously, preconcentration is important for the increase of the current. To achieve preconcentration, the electrodes were soaked in solution containing vitamin C without applying potential for certain accumulation time. For 100 mM vitamin C, the ipa was found to increase rapidly in the beginning accumulation time and eventually reached a saturated value at 20 s. It took even longer time for the ipa to level off for a lower concentration of vitamin C. This phenomenon is as expected and further confirms the ion-exchange process between PV-GCE and vitamin C anions as indicated in Equation 1. Therefore, in order to increase the sensitivity of detection, a longer time is needed for lower concentrations of vitamin C. Figure 3 shows the SW voltammograms for 0 to 900 mM vitamin C under optimized conditions. A very wide linear calibration curve was obtained up to 900 mM with a slope and correlation coefficient of 0.051 (mM mA 1) and 0.998, respectively. The electrode is stable for at least one week. Furthermore, the RSD value for the 12 successive measurements of 100 mM of vitamin C was < 2%. For 5 different electrodes, the RSD value was < 3% for the measurements of 100 mM of vitamin C. The detection limit (S/N 3) was 0.38 mM after 20 s of accumulation at open circuit. Note that the sensitivity was improved by 50 times more than a recent report based on the ruthenium(III) diphenyldithiocarbamate-modified carbon paste electrode for vitamin C detection [9]. The analytical utility of the PV-GCE was demonstrated by applying it to determine the vitamin C content in commercial orange and cherry juices. Since the content of vitamin C and pH value of most fruit juices are lower than 900 mM and close to pH 4, the determination can be done directly. In other words, neither dilution nor other pretreatment of the samples is required. The wide linear detection linear range together with the proper response in pH 4 solution is indeed the main advantage of the method in the determination of vitamin C in fruit juices. The results of good recoveries, as shown in Table 1, provide a sufficient evidence for the feasibility of using the PV-GCE for the determination of vitamin C in real samples. The result obtained was also compared with the values determined by iodimetry [26]. The iodine method is one of the often-used procedures for the determination of vitamin C in fruit juices. The good agreement with the iodimetry is again a promising feature for the applicability of the modified electrode for the direct determination of vitamin C in fruit juices. In summary, the PV-GCE showed excellent electrocatalytic and accumulation effects towards the vitamin C
1599
Direct Determination of Vitamin C in Fruit Juices
Fig. 3. SWV responses for increasing concentrations of vitamin C in pH 4 buffer solution. Other experimental conditions were the same as in Figure 2. Insert figure shows the calibration plot for vitamin C corresponding to its SWV responses. Table 1. Vitamin C assays for the two juice samples using the PVGCE. Number of samples assayed were three; experimental conditions were the same as in Figure 3. Parameters
Linear equation R Detected value (mM ) Spike (mM ) Detected value after spike (mM ) Recovery (%)
Fruit juices Orange
Cherry
ipa 3.85 0.059 [vitamin C] 0.9986 92.6 [a] 50 143.9
ipa 4.94 0.051 [vitamin C] 0.9993 131.0 50 180.2
100.9
98.4
[a] Detected by iodimetry: 95.7 mM, relative error: 3.2%.
oxidation reaction in acidic pH. This modified electrode showed good sensitivity and reproducibility for the vitamin C analytical assays. The accuracy of the method was demonstrated for vitamin C determination in various juice samples with high recoveries. Compared to the previous reported chemically modified electrodes, the PV-GCE has the advantages of direct detection, lower detection limit, wider linear range, and higher stability [8 ± 12]. Due to the electrotrostatic accumulation effect of PV, the PV-GCE also has the potential for the determination of other anionic analytes in various real samples. Experimental Ascorbic acid was obtained from Sigma (St. Louis, MO, USA) and used as received. All the other compounds used
Scheme 2.
in this work were prepared from ACS-certified regent grade chemicals without further purification and dissolved in doubly distilled deionized water. Unless otherwise noted, a 0.1 M, pH 4 acetate buffer solution was used for all electrochemical measurements. Water-soluble viologen oligomers (Scheme 2) were prepared from polybutylviologen dibromide (PBV-Br2) according to the procedure reported earlier [27, 28]. Briefly, equimolar amounts of 4,4'-bipyridyl and 1,4-dibromobutane were mixed in 20% DMF. After refluxing for 5 min, the reaction solution was allowed to stand at room temperature overnight to form a yellow precipitate in ca. 70% yield. Voltammetric measurements were carried out with a CHI Model 660 (Austin, TX, USA) electrochemical workstation. The three-electrode system consists of either GCE or PVGCE working electrode (geometric area 0.07 cm2), a Ag/ AgCl reference electrode (Model RE-5, BAS), and a platinum auxiliary electrode. The GCE, 3 mm diameter was polished with polishing kit (BAS), rinsed with deionized water, and then cleaned ultrasonically in (1 : 1) nitric acid and de-ionized water successively. The PV-GCEs were prepared by electropolyElectroanalysis 2002, 14, No. 22
1600 merization of 0.1% oligomeric unit at suitable reductive potentials in pH 4 BR buffer (I 0.1 M). The film deposition was controlled by the charge passed in the electropolymerization. During the electropolymerization, the resulting viologen moiety in the PV film could be reduced to its cation radical so that the growing film on the electrode appeared intensely blue [28, 29]. It was found that an applied polymerization potential of 1.0 V (vs. Ag/AgCl) for 30 s was the best condition for the determination of vitamin C by SWV. Polyviologen units prepared on GCE were mechanically stable. For real sample analysis, orange and cherry juices purchased from a local market were used.
Acknowledgements The authors gratefully acknowledge financial supports from the National Science Council of Republic of China. The helpful suggestion from Dr. H.-C. Chang is also greatly appreciated.
References [1] M. H. Pournaghi-Azhar, R. Ojani, Talanta 1995, 42, 1839. [2] D. Voet, J. G. Voet, Biochemistry, 2nd ed., Wiley, New York, 1995, p. 571. [3] R. Leubolt, H. Klein, J. Chromatogr. 1993, 640, 271. [4] E. Y. Backheet, K. M. Emara, H. F. Askal, G. A. Saleh, Analyst 1991, 116, 861. [5] K. Matsumoto, K. Yamada, Y. Osajima, Anal. Chem. 1981, 53, 1974. [6] R. Stevanato, L. Avigliano, A. Finazzi-Agro, A. Rigo, Anal. Biochem. 1985, 149, 537. [7] J.-M. Zen, D.-M. Tsai, A. Senthil Kumar, V. Dharuman, Electrochem. Commun. 2000, 2, 782.
Electroanalysis 2002, 14, No. 22
J.-M. Zen et al. [8] B. Nalini, S. S. Narayanan, Anal. Chim. Acta 2000, 405, 93. [9] V. S. Ijeri, P. V. Jaiswal, A. K. Srivastava, Anal. Chim. Acta 2001, 439, 291. [10] A. B. Florou, M. I. Prodromidis, S. M. Tzouwara-Karayanni, M. I. Karayannis, Anal. Chim. Acta 2000, 423, 107. [11] J.-J. Sun, D.-M. Zhou, H.-Q. Fang, H.-Y. Chen, Talanta 1998, 45, 851. [12] A.-M. Yu, H.-Y. Chen, Anal. Chim. Acta 1997, 344, 181. [13] E. W. Kristensen, W. G. Kuhr, R. M. Wightman, Anal. Chem. 1987, 59, 1752. [14] M. B. Gilbert, D. J. Curran, Anal. Chem. 1986, 58, 1028. [15] J. Wang, M.-S. Lin, Electroanalysis 1990, 2, 253. [16] L. A. Coury, E. W. Huber, E. M. Birch, W. R. Heineman, J. Electrochem. Soc. 1989, 136, 2603. [17] C. R. Martin, L. S. van Dyke, in Molecular Design of Electrode Surfaces (Ed: R. W. Murray), Wiley, New York 1992, p. 403. [18] K. Kamata, T. Suzuki, T. Kawai, T. Iyoda, J. Electroanal. Chem. 1999, 473, 145. [19] P. Martigny, F. C. Anson, J. Electroanal. Chem. 1982, 139, 383. [20] N. Oyama, N. Oki, N. Ohno, Y. Ohnuki, H. Matsuda, E. Tsuchida, J. Phys. Chem. 1983, 87, 3642. [21] J. A. Bruce, T. Murashi, M. S. Wrighton, J. Phys. Chem. 1982, 86, 1552. [22] R. N. Dominey, N. S. Lewis, J. A. Bruce, D. C. Bookbinder, M. S. Wrighton, J. Am. Chem. Soc. 1982, 104, 467. [23] D. C. Bookbinder, N. S. Lewis, M. S. Wrighton, J. Am. Chem. Soc. 1981, 103, 7656. [24] J. A. Bruce, M. S. Wrighton, J. Am. Chem. Soc. 1982, 104, 74. [25] T. Nagamura, K. Sakai, J. Chem. Soc. Faraday Trans I 1988, 84, 3529. [26] S. A. Wring, J. P. Hart, B. J. Birch, Anal. Chim. Acta 1990, 229, 63. [27] A. Factor, G. E. Heinsohn, Poly. Lett. 1971, 9, 289. [28] H.-C. Chang, M. Osawa, T. Matsue, I. Uchida, Chem. Commun. 1991, 9, 611. [29] K. Kamata, T. Kawai, T. Tyoda, Langmuir 2001, 17, 155.
