Preparation of activated rough electrode in KMnO₄ solutions and its application for the electrocatalytic oxidation of ascorbic acid

Sun Hanwen, Lian Kaoqi, Liang Shuxuan, Liu zhanfeng
(Key Laboratory of Analytical Science and Technology of Hebei Province, College of Chemistry and Environmental Science, Hebei University, Baoding  071002, China)

Abstract The preparation of activated rough glass carbon electrode (ARE) in KMnO₄ solution and its electrocatalytic oxidation for Ascorbic acid (AA) were studied. The ARE showed higher electrochemical activity of ascorbic acid and its mechanism of catalysis was studied. In the buffer solution of pH 5.0, a sensitive oxidation peak of AA was observed at 0.17V at ARE, which is 500mV more negative than on flat glassy carbon electrode. The oxidation peak of AA is linear with its concentration from 4×10^-6mol/L to 8×10^-4mol/L in the pH 5.0 buffer solution (R=0.9993), Detectian limit of it reached 1×10^-6mol/L. ARE had been successfully applied to the determination of AA in medicament sample.
Keywords Activated rough electrode, KMnO₄ solution, ascorbic acid, electrocatalytic

Ascorbic acid (vitamin C，AA) is used clinically in the treatment and prevention of scurvy. Ascorbic acid/ascorbate is a vital component in the diet of human being. Ascorbate is known to take part in several biological reactions and is present in mammalian brain. Ascorbate is possibly the primary antioxidant in human blood plasma. In addition, AA is found in high concentration in some fruits and foods. A variety of methods have been used for its determination, including spectrophotometry ^[1], titrimetry ^[2], flow inject analysis^[3] , high performance liquid chromatography^[4] and electrochemistry. Electrochemical detectors have been proven as highly sensitive means for determining electroactive species. Although AA is electroactive, its oxidation at the carbon electrode requires undesirably high working potentials. The high overpotential causes a loss of selectivity and sensitivity. One possibility for alleviating this problem is the use of chemically modified electrodes (CMEs). Poly (neutral red) film^[5], mimetic biomembrane^[6], poly-3,4-ethlenedioxy thiophene^[7], ferrocyanide film^[8], zeolite^[9] and polyalizarin red film^[10] were used to modify the different electrode for the detection of AA, but the CMEs often have short-lived shortcomings. The other possibility is activating electrode by the physical pretreating method, we describe herein a simple and rapid method to prepare activated rough glassy carbon electrode (ARE) in KMnO₄ solution and its electrocatalytic oxidation for AA. This electrode was not only prepared simply but also had very good stability and reproducibility through reactivating , it overcame the short-lived disadvantage of CMEs. Furthermore, it had higher sensitivity and alternative in detecting AA , The result of experiment is that the oxidation peak of AA is linear with its concentration from 4×10^-6 mol/L to 8×10^-3mol/L in the pH 5.0 buffer solution. The proposed method has been applied to the determination of AA in real samples with satisfactory results.

1 EXPERIMENTAL
1.1 Apparatus and reagents
Cyclic voltammetry measurements were carried out using a MEC-12B multi-function electrochemical analyzer (Jiangsu Jiangfen Instrument Int., China). All the cyclic voltammograms were made using a three-electrode system with a glassy carbon as a working electrode, a saturated calomel electrode as a reference electrode and a platinum wire electrode as a counter electrode. The KQ218 ultrasonic instrument (Kunshan Ultrasonic Instrument Factory, China) was used.
Ascorbic acid purchased from Tianjin Chemicals (China) was used without further purification. All other chemicals used in this investigation were of analytical grade. Clark-Lubs buffer solution of various pHs were prepared,All solutions were prepared with double-distilled water.
1.2 Electrodes preparation
1.2.1 Preparation of the activated plat electrode
The glassy carbon electrodes were polished with an ultrafine sand paper and wettish alumina in sequence to obtain a mirror-like surface, and treated by ultrasonic in 95% ethanol for 10 min, followed by doubly distilled water for 10 min. The electrode activated in aqueous solution of 4×10^-4 mol/L KMnO₄ by using cyclic scanning for 10 circles_, the scanning rate was 0.1V/s, the potential range was from -1.5 to 1.5V. Then the activated flat electrode(AFE) was obtained. The electrode was cleaned with distilled water before use.
1.2.2 Preparation of the activated rough electrode
The glassy carbon electrodes were gently polished with SiC sand paper (CW 800) to obtain the electrode with more rough surface. Then removing the crumb by ultrasonic vibrating in double-distilled water for 5 min. The activated rough electrode (ARE) was prepared by using the method described above. The electrode had very good stability and reproducibility through reactivating every time before using electrodes.
1.3 Experiment method
The cyclic voltammetry and the stripping voltammetry was used in the determination of AA. potential scanning was performed in the range of -0.4 to 1.0V, the accumulation potential was -0.6V, the accumulation time was 0.5 min and the scan rate was 100mV/s. At the same time wrote down cyclic voltammograms for investigating the cyclic voltammetric behaviors of AA on ARE.

2. RESULTS AND DISCUSSION
2.1 Activation condition of the electrode
2.1.1 Selection of the activation reagent the effect of different activation reagent on the activity of electrode had been investigated by cyclic voltammetry, 0.2 mol/L H₂SO₄, 0.2 mol/L HClO₄, 4×10^-5 mol/L KMnO₄, 4×10^-5 mol/L Cr⁶⁺ was used as activation reagent respectively. The result showed : compared with that on the flat electrode, the oxidation peak current of AA on the activated electrode increased when the electrode was activated in KMnO_4, Cr⁶⁺ aqueous solution, especially in 4×10^-5 mol/L KMnO₄. The activation of other reagent on peak current was slight, and some made the electrode passivate instead, so we selected KMnO₄ as the activation reagent.
2.1.2 Selection of the activation reagent concentration In addition, the effect of activation reagent concentration on the activity of electrode was investigated though detection of AA. the rough electrode had been activated in different concentration of KMnO₄. It was found that the oxidation peak current of AA increased with the concentration of KMnO₄ solution increased, and it tended to stable when the concentration of KMnO₄ solution was over 5×10^-5 mol/L, consequently we used 4×10^-5 mol/L KMnO₄ as the activation reagent.
2.1.3 Selection of the activation scan cycle the effect of activation scan cycle on the activity of electrode was investigated by the same method. the oxidation peak current of AA increased with the scan cycle increased, and it tended to smooth when scan cycle was over 10. Then the electrode activated by using cyclic scanning for 10 circles.
2.2 Electrochemical behaviour of AA at ARE

Figure 1 Cyclic voltammograms of AA at ARE in pH 5.0 buffer solution
    a: in the blank solution, b: a+4×10^-4 mol/L AA

2.2.1 Cyclic voltammograms Figure 1a showed the Cyclic voltammograms of the ARE in the blank solution (pH 5.0 Clark-Lubs buffer solution), Upon the addition of 4×10^-4 mol/L AA, an enhancement in the oxidation peak current (Figre 1b) was observed at 0.17V, which is 500mV more negative than on flat glassy carbon electrode. Compared with that on the flat electrode, it indicated the strong electrocatalytic activity of the ARE to AA. Only an oxidation peak of AA was observed at ARE, it showed the redox process of AA on the ARE was irreversible.
2.2.2 Effects of pH on the voltammetric responses The influences of the pH value of supporting electrolyte on the peak currents of AA were investigated, as shown in Figure 2.

Figure 2 Relationship between pH and the peek current of AA

    The result showed that for pH= 2 the oxidation peak current of AA was greater with lower reproducibility. No obvious change of the peak currents of AA with the increasing pH was observed in the pH range of 3-7, and the oxidation peak current decreased rapidly with increasing pH in alkalescent solution. The oxidation peak potential was inversely proportional to the pH value in the range of 3-7, indicating that protons take part in the electrode process. as shown in Figure 3.

Figure 3 Relationship between pH and the peak potential of AA

Considered as a whole, Clark-Lubs buffer solution (pH=5.0) was choosed as the supporting electrolyte.

2.2.3 Effects of scan rate on the voltammetric responses Furthermore, the effect of the scan rate on the oxidation peak of AA was investigated (Figure 4). the oxidation peak current (i_Pa)was proportional to the scan rate at lower scan rate(10-200mV/s). the linear regression equation was i_Pa(mA)=2.305+0.020n（mV/s），with a correlation coefficient of R=0.9996. It followed Langmuir adsorption isotherm:
    i_P=n²F²AΓ_TV /4RT
    Which indicated that electrode process was controlled by adsorption.

Figure 4 The relationship between peak current and scan rate

2.2.4 Effects of accumulation time and accumulation potential The effect of accumulation times on the peak currents of AA were investigated for the determination of 4×10^-4 mol/L AA by stripping scanning. The accumulation time has a certain influence on the electrocatalytic oxidations of AA . It could be seen that the peak currents increased with increasing accumulation time in the range of 6-30s, and the peak currents remained almost constant when the accumulation time increasing further. Under a negative accumulation potential, the peak current was stronger with better reproducibility. For the determination of 4×10^–4mol/L AA the peak currents and the peak potential remained almost constant at the accumulation potential from -0.1V to -0.9V. Considered as a whole, 30s accumulation time and -0.6V accumulation potential were used.
2.3 Interference tests
Under the selection condition, the interference tests for some materials that probably coexisted with AA have been conducted in pH 5.0 buffer solution containning 4×10^-4 mol/L AA. The experiment showed: the oxidizable materials Fe³⁺,H₂O₂, etc. had important influence on the determination; the other interferences were as follows: 30- fold of uric acid, cysteine, cystine, 100-fold of tartaric acid,citric acid, glutamic acid,aminoacetic acid, phenylalanine, glucose did not interfere with the determination.
2.4 The relationship between peak current and Concentration

Figure 5 The relationship between peak current and concentration

    Determination of AA by the stripping voltammetry under the optimum conditions, as it is shown in Figure 5. the catalytic peak current (Di_P = peak current - blank current) and the concentration of AA from 4×10^–6 mol/L to 8×10^–4 mol/L is linear with correlation coefficient of 0.9993, the linear formula is expressed as Di_P = 0.2039 + 0.0118c. the detection limit is 1.0×10^-6 mol/L. The relative standard deviations (RSD) are 1.2% and 2.0%, respectively, for uninterrupted and interrupted repetitive determinations for 7 times at 1×10^–5 mol/L AA in Clark-Lubs buffer solution (pH 5.0). And after the ARE had been used more than 200 times, the peak currents of AA remain almost constant . These results suggested that the ARE had a good reproducibility and stability for the determination of AA.
2.5 Sample analysis
Two vitamin C tablets (0.1g/ tablet) were ground into fine powder,then dissolved in water and filtered before making a volume of 100 mL. and diluted by 20 times( sample 1) and 40 times( sample 2) with Clark-Lubs buffer solution(pH 5) for determining AA by the stripping voltammograms. The results were 104.2 and 104.6 mg/ tablet（the reference content is 100 mg/ tablet）with RSD(n=7) of 1.2% and 1.0%.
2. 6 Electrode catalysis mechanism discussions    
Figure 6 demonstrates the cyclic voltammograms of 4×10^–4mol/L AA in phosphate buffer solution (pH5.0) on different electrodes. As it is shown above, compared with flat glassy carbon electrode (Figure 6a), the oxidation peak of AA could be observed at rough electrode (Figure 6c), the background current increased obviously at the same time. The oxidation peak current of AA increased notably at flat electrode activated in KMnO₄ solution (Figure 6b). It reached the greatest at ARE (Figure 6d), and the peak shape was more perfect than the others, and the background current was lower than it on rough electrode. Obviously, the activating in KMnO₄ solution and roughening leaded to the electrocatalytic effect of AA at ARE.

Figure 6 Cyclic voltammograms of 4×10^–4　mol　/L AA at different electrodes
a: Flat electrode, b: Activated flat electrode, c: Rough electrode, d: Activated rough electrode

    Many microholes and ditches which made the molecules or ions easier to penetrate and shift out came into being after the electrode was roughened and immerged into the solutions. In addition, it could also increase the surface area of the electrode, and absorb, expose or produce some active sites, therefore increased the peak current and sensitivity properly, but it was not the main factor of electrocatalysis, and the peak current did not increase remarkably. After activating in the KMnO₄ solution, the surface impurity of electrodes and some inert layers were oxidized and removed firstly, then the fresh surface was exposed, the charge electric current and redox current which was produced by some impurity were dropped, so that, compared with rough electrode, the background current reduced on ARE. Secondly, during the activation of electrode in KMnO₄ solution, a great deal of surface oxide were produced, these oxide may act as active centre and play a role of  "catalyst" in electrochemistry, thus improved the sensitivity and reversible. which was the main factors of electrocatalysis. The ARE with the active sites spread on its surface was similar to the porous electrode ^[12]and powder microelectrode which were made of electroactive material, besides the electrochemical behavior of AA on ARE and on porous electrode is extremely similar. We can draw conclusion that ARE was on the basis of obtaining larger surface area through roughening, activated further in KMnO₄ solution, therefore achieved the best results of catalysis.

ACKNOWLEDGEMENTS  We express thanks to the Natural Science Foundation of Hebei Province (China), for much support to the studied subject (203110).

REFERENCES
[1] Pereira , Airton Vicente, Fatibell-Filho et al. Analytic Chimica Acta. 1998, 19 (1-3):55-62.
[2] Zheng X H, Zheng Y, Qiu J B et al. Journal of Fujian Teachers University (Natural Science) 2000, 16 (1): 56-59.
[3] Yebra-Biurrun M C. Talanta, 2000, 52 (3): 367-383.
[4] Wang S, Schram I M, Sund R B. European Journal of Pharmaceutical Sciences.1995, 3 (4): 231-239.
[5] Jiao K, Luo X L, Sun W et al. Chinese Analytical Laboratory, 2002, 21 (2): 89.
[6] Wu Z Y, Tang J L, Wang E K. Chinese Analytical Chemistry, 2001, 29 (8): 881.
[7] Hou S F, Fang H Q, Chen H Y. Chemical Journal of Chinese Universities 1995, 16 (1): 39.
[8] Pournaghi-Azar M H, Osani R. Journal of Solid State Elecrochemistry, 2000, 4: 75.
[9] Jiang Y X, Song W B, Liu Y et al. Chemical Journal of Chinese Universities, 1999, 20 (5): 717.
[10] Wan Q J, Yu J H, Wang G et al. Chemical Journal of Chinese Universities, 2000, 21 (11): 1651.
[11] Zhang H C, Zuo X B, Wang Z R. Analytical Instrument. 1997, (1): 2-7.
[12] Quan C X et al. Dynamics Guide of Electrode Process, Science Publishing Company, Beijing, 2001.