1647

Full Paper

Amperometric Determination of Sugars at Activated Barrel Plating Nickel Electrodes Jun-Wei Sue, Chi-Jr Hung, Wei-Chung Chen, Jyh-Myng Zen* Department of Chemistry, National Chung Hsing University, Taichung 402, Taiwan *e-mail: [email protected] Received: February 21, 2008 Accepted: April 15, 2008 Abstract We report amperometric determination of sugars by using a disposable barrel plating nickel electrode (Ni-BPE) in this study. The activated Ni-BPE possesses good reproducibility in flow injection analysis of sugars with a relative standard deviation of 3.16% for 10 consecutive injections of glucose. The electrocatalytic mechanism for the detection of sugars as well as the use as a detector in high-performance liquid chromatography (HPLC) is investigated. We achieve a good separation of four sugars (glucose, fructose, sucrose, and maltose) in HPLC with favorable sensitivity at a detection potential of þ 0.55 V vs. Ag/AgCl. The results of wide linear calibration ranges and detection limits in the mM range meet the need of real sample analysis. This detection method is successfully used for quantitative determination of sugars in honey to identify its authentication. Keywords: Nickel, Amperometry, Sugars, Barrel plating electrode, Honey DOI: 10.1002/elan.200804228

1. Introduction The determination of sugars is considerably important in biological, environmental, clinical, and food analysis [1]. Electrochemical techniques allow for direct detection of sugars without the necessity of derivatization and are often characterized by lower detection limits at less equipment cost. Among these, pulsed amperometry (PAD) using noble metals as working electrodes to couple with high performance ionic chromatography (HPIC) are a powerful detection method for carbohydrates [2, 3]. The detector, however, requires the use of a more complicated pulse generation circuits to renew the electrode surface. Constant potential amperometric detection with working electrodes made of transition metals is relatively simple with a low baseline noise [4 – 9]. Notably, nickel electrode has been the most widely utilized electrode for determining carbohydrates in alkaline media [9 – 16]. Several nickel alloy electrodes were reported to further improve the sensitivity [17 – 22]; yet, it is difficult to control the composition and size distribution of each metal of the alloy. Chemically modified nickel electrodes were also developed by depositing nickel particles on a traditional electrode surface to improve the performance [23 – 27]. It is, however, not easy to maintain long-term stability because the electrochemical activity would gradually decrease due to the detachment and dissolution of the catalyst from the substrate. A radio frequency sputtering uniform film consisting of 0.8% highly dispersed nickel nanoparticles embedded in disordered graphite-like carbon was reported to improve the long-term stability with high sensitivity [28]. Overall, the use of a Electroanalysis 20, 2008, No. 15, 1647 – 1654

metallic nickel electrode as an amperometric detector for the detection of sugars after separating them with highperformance liquid chromatography (HPLC) is still a convincing method for practical sugar monitoring. In this paper, we report the determination of sugars based on the electrocatalytic behavior at a barrel plating nickel electrode (designated as Ni-BPE). The primary function of barrel plating is to provide an economical means to electroplate manufactured parts that meet specific finishing requirements [29, 30]. Since the fabrication cost of the NiBPE is low, it is thus disposable in nature. Our previous studies have successfully used this kind of disposable-type electrode for analytical applications [31, 32]. In this study, we further demonstrate that the activated Ni-BPE coupled with a specifically designed electrochemical cell allows us to obtain an electrocatalytic profile towards the oxidation of sugars. The user-friendly design of electrochemical cell is suitable for coupling with BPE for use in FIA and is especially attractive in practical applications. The proposed button and lock type of FIA-ECD device with special care to reusability and precision in electroanalysis is different from few attempts in the past. Both the mechanism of electrocatalytic oxidation in flow injection analysis (FIA) and the use as a detector in HPLC for the determination of sugars at the activated Ni-BPE are discussed and investigated. It was finally applied to the determination of sugars in real samples without the necessity of laborious sample preparation.

F 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1648

J.-W. Sue et al.

2. Experimental 2.1. Chemicals and Reagents Sodium hydroxide, glucose, galactose, fructose, ribose, and xylose were used as received from Sigma. All other compounds used in this work were ACS-certified reagents and used without any further purification. Aqueous solutions and real samples were prepared using double distilled deionized water. Unless otherwise stated the base electrolyte used in the study was made of 0.1 M NaOH solution.

2.2. Apparatus and Electrochemical Detector Setup Cyclic voltammetric and chronoamperometric experiments were carried out using a CHI 900 electrochemical workstation (CH Instruments, USA). The Ni-BPE three-electrode system is the same as reported in our previous study [32]. Scheme 1 shows typical pictures of the Ni-BPE work-

ing system used in this work. It consists of a Ni-BPE working electrode, a stainless tube counter electrode (outlet), and an Ag/AgCl reference electrode. The flow injection analysis (FIA) system was equilibrated in 0.1 M NaOH carrier solution at þ 0.5 V until the current became constant. The Ni-BPE surface was activated by cycling from 0.6 to  0.2 V in 0.1 M NaOH solution. The HPLC system consisted of a BAS-PM92E high-pressure microprocessor pump drive, a Rheodyne 7125 sample injection valve (20 mL loop), and the proposed electrochemical detector. The HPLC system with an anion-exchange column (Hamilton PRP-X100, 150  4.1 mm, 10 mm) was equilibrated in 0.1 M NaOH carrier solution at þ 0.55 V vs. Ag/AgCl until the current became constant. It usually takes 5 min at room temperature. X-ray photoelectron spectroscopy (XPS) (ULVAC-PHI Quantera) was performed by using an Al Ka X-ray radiation source (1486.6 eV) with 0.1 eV of resolution. The pressure inside the analyzer was maintained at about 1010 Torr during the measurement. The C1s peak at 284.6 eV was taken uniformly as an internal standard. Quantitative XPS

Scheme 1. Scheme and pictorial representation of the proposed flow injection electrochemical detector setup. Electroanalysis 20, 2008, No. 15, 1647 – 1654

www.electroanalysis.wiley-vch.de

F 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1649

Amperometric Determination of Sugars

analysis was carried out by using an XPS Peak (Version 4.1) program to pick up the intensity maximum and BE values. The peak area was used to calculate the active ingredient ratio. The electrochemical impedance spectra (EIS) was measured at 10 discrete frequencies per decade from 10 mHz to 100 kHz at amplitude of 10 mV (rms) using the AUTOLAB FRA frequency response analyzer with FRA2 module controlled by IBM compatible PC. The acquired data was analyzed based on equivalent electrical circuits by a weighted nonlinear least square method using the fitting software, EQCRT, elaborated by Boucamp. Fitting constraints were imposed that further iterations will stop when the chi-square (c2) change is less than 0.001% as compared to the previous iteration. The satisfaction of fit was assessed from minimum c2, correlation matrix, and relative error distribution plots. Less than 5% fluctuations between the experimental and fit data were assumed satisfactory in confirming the validity of the selected fitting circuit.

Fig. 1. Typical cyclic voltammograms of the activated Ni-BPE in 0.1 M NaOH before (a) and after (b) the addition of 1 mM glucose at a scan rate of 50 mV/s. Inset shows the proposed electrocatalytic mechanism.

3. Results and Discussion and can be simplified as As mentioned in our earlier work, the Ni-BPE surface needs to be activated via cycling from þ 0.6 to  0.2 V (vs. Ag/ AgCl) in 0.1 M NaOH solution before analytical applications [32]. The cyclic voltammograms show the growth of a characteristic oxidation peak at þ 0.47 V and reduction wave at þ 0.38 V during consecutive scans at a freshly prepared Ni-BPE. This phenomenon can be attributed to multilayer formation and reduction of nickel oxide at the NiBPE [33]. The electrochemical oxidation of glucose was first investigated at the activated Ni-BPE in 0.1 M NaOH solution (Fig. 1). As can been seen, upon glucose addition, there is an increase in the anodic peak current together with a decrease in the cathodic peak current. Note that the electrocatalytic function toward the oxidation of glucose can only be observed at the activated Ni-BPE. The electrocatalytic oxidation mechanism of glucose at the Ni-BPE surface represents a mediated catalytic oxidation as illustrated in Figure 1. Note that the electrooxidation of carbohydrates on the oxide metals is very complex and involves strong adsorption steps of reactants and/or reaction products. In addition, the reaction products are generally composed of a mixture of species. In this study, the interaction between NiOOH and the terminal OH functional groups of sugars was believed to be essential for the detection. The mechanism for the electrocatalytic oxidation of sugars can be expressed as [34 – 37] Ni þ 2OH ! Ni(OH)2 þ 2e Ni(OH)2 þ OH ! NiOOH þ H2O þ e NiOOH þ glucose ! Ni(OH)2 þ glucose radical glucose radical ! glucolactone Electroanalysis 20, 2008, No. 15, 1647 – 1654

Ni(II) ! Ni(III) þ e 2Ni(III) þ glucose ! 2Ni(II) þ glucolactone Overall, when applying a potential around the oxidation peak (i.e., þ 0.47 V) to the Ni-BPE, the increase in the oxidation current can be used for the amperometric determination of glucose.

3.1. Characterization of the Activated Ni-BPE Since the activation process is necessary for further analytical application, the XPS measurements were performed to obtain precise information about the surface composition of the Ni-BPE. Figure 2 shows the survey scan XPS responses of the activated Ni-BPE. The high-resolution XPS spectrum of the Ni2p region is shown in the inset of Figure 2. The Ni2p spectrum shows shakeup satellite lines at a higher binding energy in addition to the expected Ni2p1/2 and Ni2p3/2 lines. As to the Ni2p3/2 lines, the peak around a binding energy of 853.0 eV is attributed to metallic nickel, while the peaks at a higher energy ( 855.0 eV) are associated with nickel hydroxide (Ni(OH)2)/oxide film (NiO and Ni2O3). The peaks around binding energies of 855.57 and 860.97 eV can be assigned to the Ni2p3/2 photoelectrons from NiO and Ni(OH)2, respectively. In general, the XPS results match very well with that reported previously [28]. Two marked peaks corresponding to the energy levels of NiO (855.57 eV) and Ni(OH)2 (860.97 eV), respectively, representing the key components on the interface. Upon electrochemical activation, the NiO peak was found to decrease by ca. 32%, accompanied with an increase in the Ni(OH)2 level. The Ni(OH)2/NiO ratio was found to

www.electroanalysis.wiley-vch.de

F 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1650

J.-W. Sue et al.

Fig. 2. XPS spectra of the activated Ni-BPE. Inset shows detailed XPS spectra of NiO and Ni(OH)2.

increase from 0.69 to 1.01 after electrochemical activation. This is crucial to the success of the proposed system as such detection and separation require the use of a strong alkaline medium in connection to the Ni(OH)2/NiOOH redox pair [34 – 37]. As stated earlier, we emphasized on the interaction between NiOOH and the terminal OH functional groups of sugars. If this is the case, the interaction should reflect in the measured charge transfer resistance (RCT). The EIS was therefore utilized to characterize the oxidation mechanism of the activated Ni-BPE. Figure 3 shows typical EIS responses of the activated Ni-BPE in the presence of five different sugars (i.e., three hexoses and two pentoses) in 0.1 M NaOH at an applied potential of þ 0.55 V. As can be seen, under identical experimental conditions, all sugars yield similar semicircle characteristics in the presence of sugars except those with a different radius. The EIS responses were analyzed by an in-built curve fitting program based on an input Boukamp equivalent (Randles) circuit of Rs(RCTCdl), where Rs ¼ solution resistance, RCT ¼ charge transfer resistance, and Cdl ¼ double layer capacitance value. Table 1 summarizes the data obtained regarding the catalytic oxidation at the activated Ni-BPE. It is obvious that both the Cdl and Rs values are very similar for all five sugars studied. The major difference is in RCT representing the effective faradic electron-transfer behavior on the activated Ni-BPE. It is thus concluded that the probability of interaction between the NiOOH and the terminal OH functional groups of sugars (i.e., steric effect) plays a major role in the variation of RCT. Scheme 2 illustrates the possible orientation of the active sites of sugars with the activated NiBPE. As can be seen, the ability of terminal OH functional groups to contact with NiOOH is directly related to the measured RCT (Table 1). Among these sugars, glucose possesses the highest probability in the interaction with NiOOH and hence results in the lowest RCT. Electroanalysis 20, 2008, No. 15, 1647 – 1654

Fig. 3. Complex plan EIS plots for A) three hexoses, B) two pentoses oxidation at the activated Ni-BPE at an applied potential of 0.55 V vs. Ag/AgCl in 0.1 M NaOH.

www.electroanalysis.wiley-vch.de

F 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1651

Amperometric Determination of Sugars

Scheme 2. Schematic illustration of the interaction between active sites of sugars and activated Ni-BPE.

3.2. Amperometric Determination of Sugars The FIA parameters were first optimized for the purpose of analytical application. Since optimum flow rate is necessary to deliver the reactant in FIA, the effect of flow rate was studied over the range 100 – 500 mL/min. The current responses of glucose were found to decrease with the increase in flow rate up to 300 mL/min; and after that, the current intensity did not increase appreciably. A flow rate of 300 mL/min was therefore selected for subsequent experiments. The effect of applied potential (Eapp) on the detection of glucose was evaluated next. Typical hydrodynamic voltammograms for the oxidation of glucose at the activated Ni-BPE were detected over the þ 0.2 to þ 0.6 V range at a flow rate of 300 mL/min. An increase of the anodic response starting at þ 0.3 V was found to rise rapidly up to þ 0.55 V. Since a relatively higher deviation of FIA signals were

observed when Eapp > þ 0.55 V, these potentials are not suitable for analytical application. Subsequent amperometric detection work employed a potential of 0.55 V that offered the most favorable signal-to-noise characteristics. Note that the profiles of hydrodynamic voltammograms are in agreement with the cyclic voltammograms of glucose (Fig. 1). The detection of glucose at Eapp ¼ þ 0.55 Vand Hf ¼ 300 mL/min was chosen as instrument setting in the following investigation. The detection of three hexoses (glucose, galactose, and fructose) and two pentoses (xylose and ribose) by the proposed system were then evaluated and compared. The peak height of the current response for all five sugars is proportional to the respective concentration for up to 0.5 mM. The linear range for each sugar was over 3 orders of magnitude. The calibration curves and detection limits obtained for these five sugars are summarized in Table 2.

Table 2. Results for the calibration of three hextoses and two pentoses at the activated Ni-BPE. Sugar Hextose

Pentose

Glucose Galactose Fructose Ribose Xylose

Calibration curve [a]

R [c]

LOD [b] (mM )

Y ¼ 0.0068 X þ 0.0069 Y ¼ 0.0055 X þ 0.0377 Y ¼ 0.0044 X þ 0.0635 Y ¼ 0.0046 X þ 0.0564 Y ¼ 0.0036 X þ 0.0140

0.9995 0.9986 0.9979 0.9990 0.9987

0.48 0.56 1.02 1.20 2.39

[a] Conditions are the same as those in Figure 5. [b] LOD value was calculated based on S/N ¼ 3. [c] Correlation coefficient ( R ¼ r2) was calculated in a linear range from 5 experimental data points.

Electroanalysis 20, 2008, No. 15, 1647 – 1654

www.electroanalysis.wiley-vch.de

F 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1652

J.-W. Sue et al.

The activated Ni-BPE exhibits good sensitivity for the determination of sugars with detection limits (S/N ¼ 3) of 0.48, 0.56, 1.02, 1.20, and 2.39 mM for glucose, galactose, fructose, ribose, and xylose, respectively. Note that the detection limit of glucose is better than that of 0.9 mM obtained earlier under either PAD conditions or cyclic chronopotentiometry with Au electrodes as well as at a Cu electrode [38 – 41]. These values are also comparable to those obtained at Ni (0.4 mM for glucose) [9], Ni alloy (0.1 mM) [18], and Ni-based modified (between 0.33 and 0.55 mM) [25] electrodes.

3.3. Application of the Ni-BPE in Monitoring of Natural Honey Although the sensitivity is sufficient to be applied to real sample applications, the reproducibility of the proposed system is further evaluated by consecutive injections (n ¼ 10) of all five sugars. As can be seen in Figure 4A, no significant loss was observed in the peak current signal with a coefficient of variation of < 5% for each sugar at the activated Ni-BPE. Virtually the same peak responses were also observed for 3-daysM trial (Figure 4B). The applicability of the proposed detector is thus confirmed in this study. By

Fig. 4. Long-term stability of the Ni-BPE detector in FIA for the detection of 100 mM sugars for ten continuous injections (A) and 100 mM glucose for a period of three days (B). Eapp ¼ 0.55 V (vs. Ag/AgCl); Hf ¼ 300 mL/min; carrier solution ¼ 0.1 M NaOH. Electroanalysis 20, 2008, No. 15, 1647 – 1654

www.electroanalysis.wiley-vch.de

F 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1653

Amperometric Determination of Sugars

glucose and fructose (95% confidence interval), respectively, giving an acceptable accuracy and precision. The method was subsequently used to analyze honey purchased from supermarket. Honey contains up to 27 sugars which are mostly monosaccharides: glucose (about 30 – 35%) and fructose (36 – 42%) [42]. Sucrose is the most common additive in a fake honey (the most commonly available honeys on the market include 0.5 – 1.5% sucrose and most of EU countries norms allow for up to 5% of sucrose in honey). In other words, the detection of fructose, glucose, and especially sucrose can consequently be used to justify the quality of honey [39]. Figure 5C shows the analytical result of the honey sample. Our analysis indicated that the sample of honey contained 266.39 mM sucrose, 842.39 mM glucose, and 766.59 mM fructose. The fact that the sample contains ca. 8% of sucrose clearly indicates adulterated honey. The method is thus satisfactory for real sample applications.

4. Conclusions We characterized the structure of the activated Ni-BPE and investigated its electrochemical behavior. During potential cycling in alkaline solution a stable activated nickel oxide film was confirmed to form at the electrode surface. The activated Ni-BPE was demonstrated to determine micromolar concentration of sugars at an applied detection potential of þ 0.55 V vs. Ag/AgCl with high sensitivity by hydrodynamic amperometry technique. It was also proved to be useful for sugar detection by HPLC with good reproducibility. The detection method works very well in the analysis of honey and expected to be equally convenient in the determination of sugars in other natural products as well as food and beverages.

5. Acknowledgements Fig. 5. HPLC-ECD responses for A) the standard separation of glucose, fructose, sucrose, and maltose in 0.1 M NaOH carrier solution under optimized conditions, B) the assay of a fructose product, and (C) an adulterated with sucrose honey sample.

using a Hamilton RCX-10 anion-exchange column for sugar separation, Figure 5A shows the chromatogram obtained with 10 mM each of glucose, fructose, sucrose, and maltose. As can be seen, the fact that glucose gets the highest peak response is again in good agreement with the discussion in previous section. Good chromatographic separation can then allow a direct determination of the sugar contents in real samples. A local commercially available famous brand fructose product was first used to verify the proposed method. To avoid the deterioration of sample, it was freshly diluted appropriately with 0.1 M NaOH solution before measurement. As shown in Figure 5B, the product contains glucose and fructose only and the content of fructose is indeed much higher than glucose. The mean values for 5 measurements were 0.49  0.01 and 5.44  0.07 mM for Electroanalysis 20, 2008, No. 15, 1647 – 1654

The authors gratefully acknowledge financial support from the National Science Council of Taiwan. This work is supported in part by the Ministry of Education, Taiwan under the ATU plan.

6. References [1] X. Cheng, S. Zhang, H. Zhang, Q. Wang, P. He, Y. Fang, Food Chem. 2008, 106, 830. [2] D. C. Johnson, W. R. LaCourse, Anal. Chem. 1990, 62, 589A. [3] W. R. LaCourse, D. C. Johnson, Carbohydr. Res. 1991, 215, 159. [4] J. B. Kafil, C. O. Huber, Anal. Chim. Acta 1985, 175, 275. [5] L. M. Santos, R. P. Baldwin, Anal. Chem. 1987, 59, 1766. [6] J. M. Zadeii, J. Marioli, T. Kuwana, Anal. Chem. 1991, 63, 649. [7] L. A. Colon, R. Dadoo, R. N. Zare, Anal. Chem. 1993, 65, 476. [8] J. M. Marioli, T. Kuwana, Electroanalysis 1993, 5, 11. [9] P. Luo, F. Zhang, R. P. Baldwin, Anal. Chim. Acta 1991, 244, 169.

www.electroanalysis.wiley-vch.de

F 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1654

J.-W. Sue et al.

[10] M. Fleischmann, K. Korinek, D. Pletcher, J. Electroanal. Chem. 1971, 31, 39. [11] K. G. Schick, V. G. Magearu, C. O. Huber, Clin. Chem. 1978, 24, 448. [12] W. Buchberger, K. Winsauer, C. Breitwieser, Fresenius J. Anal. Chem. 1983, 315, 518. [13] I. B. Casella, E. Desimoni, T. R. I. Cataldi, Anal. Chim. Acta 1991, 248, 117. [14] A. M. Fermier, L. A. Colon, J. High Resol. Chromatogr. 1996, 19, 613. [15] R. Jelinek, S. Kolusheva, Chem. Rev. 2004, 104, 5987. [16] C. W. Welch, R. G. Compton, Anal. Bioanal. Chem. 2006, 384, 601. [17] J. M. Marioli, P. F. Luo, T. Kuwana, Anal. Chim. Acta 1993, 282, 571. [18] P. F. Luo, T. Kuwana, Anal. Chem. 1994, 66, 2775. [19] M. Morita, O. Niwa, S. Tou, N. Watanabe, J. Chromatogr. A 1999, 837, 17. [20] I. H. Yeo, D. C. Johnson, J. Electroanal. Chem. 2001, 495, 110. [21] M. I. Mora, J. M. Marioli, J. Liquid Chromatogr. Rel. Technol. 2001, 24, 711. [22] J. H. Sinfelt, Bimetallic Catalysts, Wiley, New York 1983, p. 20. [23] E. Wang, A. Liu, J. Electroanal. Chem. 1991, 319, 217. [24] A. Stitz, W. Buchberger, Electroanalysis 1994, 6, 251. [25] I. G. Casella, M. Gatta, Electroanalysis 2001, 13, 549. [26] K. Ohnishi, Y. Einaga, H. Notsu, C. Terashima, T. N. Rao, S. Park, A. Fujishima, Electrochem. Solid-State Lett. 2002, 5, D1. [27] I. G. Casella, M. Gatta, Anal. Chem. 2000, 72, 2969.

Electroanalysis 20, 2008, No. 15, 1647 – 1654

[28] T. You, O. Niwa, Z. Chen, K. Hayashi, M. Tomita, S. Hirono, Anal. Chem. 2003, 75, 5191. [29] E. P. Dagarmo, J. T. Black, R. A. Kosher, Materials and Processes in Manufacturing, 9th ed., Wiley, New York 2003. [30] R. K. Mobley, Plant Engineering, Butterworth-Heinemann, Woburn, MA 2001. [31] J.-W. Sue, A. S. Kumar, H.-H. Chung, J.-M. Zen, Electroanalysis 2005, 17, 1245. [32] J.-W. Sue, C.-Y. Tai, W.-L. Cheng, J.-M. Zen, Electrochem. Commun. 2008, 10, 277. [33] C. Zhao, C. Shao, M. Li, K. Jiao, Talanta 2007, 71, 1769. [34] J. Wang, G. Chen, M. P. Chatrathi, Electroanalysis 2004, 16, 19. [35] M.-C. Chen, H.-J. Huang, Anal. Chim. Acta 1997, 341, 83. [36] P. B. Allen, I. Rodriguez, C. L. Kuyper, R. M. Lorenz, P. Spicar-Mihalic, J. S. Kuo, D. T. Chiu, Anal. Chem. 2003, 75, 1578. [37] Y.-Y. Liao, T.-C. Chou, Electroanalysis 2000, 12, 55. [38] W. R. LaCourse, in Carbohydrate Analysis by Modern Chromatography and Electrophoresis (Ed: Z. El Rassi), Journal of Chromatography Library, Vol. 66, Elsevier, Amsterdam 2002, p. 905. [39] A. Basa, J. Magnuszewskaa, T. Krogulec, A. S. Baranski, J. Chromatogr. A 2007, 1150, 312. [40] G. Chen, L. Zhang, X. Wu, J. Ye, Anal. Chim. Acta 2005, 530, 15. [41] G. Chen, L. Zhang, Y. Zhu, J. Pharm. Biomed. Anal. 2006, 41, 129. [42] H. Rybak-Chmielewska, in Chemical and Functional Properties of Food Saccharides (Ed: P. Tomasik), CRC Press, Boca Raton, FL 2003, p. 75.

www.electroanalysis.wiley-vch.de

F 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim