home de
wo kaufen?

Cyclic voltammetric analysis of antioxidant activity in cane sugars and palm sugars from Southeast Asia Jocelyn Siaa, Hong-Ben Yeea, José H. Santosb and M. Khairul-Anwar Abdurrahmana.

Mit Freude verwende ich seit einiger Zeit den Gula Java Kokosblüten-Zucker von Amanprana und ich bin sehr dankbar dafür, weil es der Einzige ist, den ich vertrage. Und nachdem er noch dazu gesund ist - sei....schmeckt mir der Kaffee gleich noch mehr...

Sarah G. Emmer, Austria

a) Institute of Medicine, Universiti Brunei Darussalam, Jalan Tungku Link, Gadong BE 1410, Brunei Darussalam b) Department of Chemistry, Universiti Brunei Darussalam, Jalan Tungku Link, Gadong BE 1410, Brunei Darussalam

Received 11 March 2009; 
revised 29 March 2009; 
accepted 11 May 2009. 
Available online 18 May 2009.

Abstract

Fourteen commonly available types of cane and palm sugar were analysed for antioxidant activity using cyclic voltammetry. Five of the sugars, dissolved in phosphate buffer, showed anodic current peaks which were indicative of antioxidant activity. The rank order of these sugars was: gula anau gula merah > China rock honey sugar > soft brown sugar > raw sugar. Gula anau is an unrefined palm sugar, while the other four are from sugarcane. Soft brown sugar and raw sugar are coated brown sugars. The China rock honey sugar contained chrysanthemum flowers, and its antioxidant properties appear to be due to these flowers and not to the sugar per se. Pure white sugar, other rock sugars and rock honey sugars (all refined from cane) as well as gula Melaka, an unrefined rock sugar from palm trees, had no observable antioxidant activity. It is concluded that, from a nutritional point of view, using gula anau as a sweetener or ingredient in foods or drinks has an added benefit owing to its antioxidant content. Keywords: Cyclic voltammetry; Antioxidant; Vitamin C; Palm sugar; Cane sugar; Brown sugar; White sugar; Chrysanthemum

Article Outline

1. Introduction
2. Methods
2.1. Sugars
2.2. Chrysanthemum flowers
2.3. Cyclic voltammetry
2.4. Diffusion coefficients
2.5. Redox potentials and peak potentials
3. Results
3.1. Vitamin C standards
3.2. Sugars
3.3. Chrysanthemum flowers
3.4. Peak anodic current potentials
3.5. Diffusion coefficients
4. Discussion
Acknowledgements
References

1. Introduction

Increasing awareness of the role of antioxidants in the diet has led to greater importance being placed on the consumption of foods containing antioxidants ([Vita, 2005], [Valdés, 2006] and [Li and Schellhorn, 2007]). At present, no claims are made by manufacturers or vendors about the presence of antioxidant compounds in sugars for sale in shops or markets. Furthermore, there appears to be extremely limited research into the presence of antioxidant compounds in sugars. One report includes data concerning the presence of antioxidants in refined sugar, raw cane sugar, turbinado sugar (Demerara sugar, a type of raw cane sugar) and dark molasses (Halvorsen et al., 2006).
In the current study, fourteen different types of sugar were examined using cyclic voltammetry. This is a well-established electrochemical technique for analysing the redox properties of molecules. Its application to biological molecules in foodstuffs is relatively new and previous studies have shown that there is a good correlation between results obtained using cyclic voltammetry and those obtained from more conventional, and cumbersome, methods for the determination of antioxidant activity ([Kilmartin and Hsu, 2003] and [Roginsky et al., 2003]).
Most of the sugars which were studied were derived from sugarcane (Saccharum spp.), two were palm sugars, one of which was derived from the coconut palm (Cocos nucifera) and the other from the nipah palm The aim of this work was to examine various types of sugar for the presence of antioxidants, and also to show that cyclic voltammetry is a simpler, cheaper and perhaps more practical alternative method of evaluating antioxidants in simple foodstuffs.

2. Methods

2.1. Sugars

Sugars were bought from local supermarkets and native markets. These included: refined white sugar (from cane, granulated pure white sugar); raw sugar (SIS), soft brown sugar and three types of red sugar (gula merah) all from cane; one white and two yellow rock sugars (gula batu) from cane; two palm sugars (gula anau and gula Melaka); and three fancy sugars, varieties of rock honey sugar, also from cane (see Table 1). One of these fancy sugars, china rock honey sugar, listed dried chrysanthemum flowers amongst its contents.
Table 1.

Peak anodic currents per unit concentration.


Sugar

Current Constanta (concentration ?A-1)

C.D0.5 (mol cm-2 s-0.5) × 10-6

Cane sugars

Nazri soft brown sugar peak1

8.1 ± 1.10 (0.931, 6)

1.6 ± 0.24 (0.884)

Nazri soft brown sugar peak2

3.7 ± 0.16 (0.992, 6)

4.4 ± 0.47 (0.936)

SIS raw sugar peak 1

18.2 ± 1.57 (0.978, 4)

0.7 ± 0.06 (0.956)

SIS raw sugar peak 2

8.2 ± 0.51 (0.989, 4)

1.5 ± 0.09 (0.979)

Sunflower gula merah

0.65 ± 0.039 (0.975, 8)

17.9 ± 0.24 (0.999)

Chek Hup white rock sugar

Granulated pure white sugar

Nabila gula merah

STM gula merah

Sunflower yellow lump sugar

Yellow gula batu

 

Fancy sugars

China rock honey sugar

6.9 ± 0.46 (0.982, 5)

4.9 ± 0.47 (0.947)

Luo Han Guo rock honey sugar

Luo Han Guo winter melon sugar

 

Palm sugars

Gula anau

0.41 ± 0.006 (0.998, 9)

29.3 ± 0.37 (0.999)

Gula Melaka

Values for C.D0.5 (±standard error) were calculated from the slopes of the regressions of Ipa on (scan rate)0.5, as shown in Fig. 10, constraining the line to pass through the origin; the correlation coefficient (r2) is given within parentheses. The values of C.D0.5 for vitamin C (mol.cm-2 s-0.5 × 10-6) were: 38.9 ± 0.58 (0.999), 21.3 ± 0.53 (0.996) and 3.3 ± 0.04 (0.999) at concentrations of 0.51, 0.26 mM and 51.1 ?M, respectively. For chrysanthemum flower infusion (10% v/v buffer) the slope was 10.8 ± 0.73 × 10-6 mol cm-2 s-0.5 (r2 = 0.973). All data are for 7 d.f.
a Current constants were obtained form the slope of the regression of View the MathML sourceon [analyte]-1, constraining the regression to pass through the origin. The regressions are drawn in Fig. 8; values are slope ± s.e. slope, with the correlation coefficient (r2) and degrees of freedom given within parentheses. The units of concentration for all the sugars is g% (w/v) phosphate buffer.

Sugars were dissolved in phosphate buffer (50 mM, pH 6.2). The raw sugar and the soft brown sugar were both light brown, granulated, sugars coloured and flavoured by coating with caramel or molasses. The red sugars were refined, granulated, sugars whose crystals were coloured artificially. All the rock sugars were refined, lightly coloured, sugars in large lumps or crystals. The palm sugars were unrefined: the gula Melaka was a single piece of solid sugar, and the gula anau was a viscous liquid. The lump sugars were crushed in a mortar with a pestle in order to aid dissolution. Each sugar was tested over a concentration range from 0.1% to 50% (w/v). Each individual sample concentration was tested twice, and each individual sugar was tested at least three times.

2.2. Chrysanthemum flowers

Dried chrysanthemum flowers are used in traditional Chinese medicines and were obtained from a Chinese apothecary. The dried petals were infused with hot, double-distilled, water, just off the boil at room temperature (25 °C) for 15 min at a concentration of 1% (w/v). This infusion was diluted further in phosphate buffer to produce solutions of 0.1–50% v/v stock/buffer.

2.3. Cyclic voltammetry

Cyclic voltammetry was performed using an eDAQ system (www.eDAQ.com), consisting of an E190 potentiostat connected to an e-corder which inputted to eChem software (running on a PC using Microsoft Windows Vista platform). The working electrode was a 3 mm diameter glassy carbon electrode; the reference electrode was Ag/AgCl in 3 M NaCl (+207 mV vs. standard hydrogen electrode); the auxiliary electrode was a 0.25 mm diameter Pt wire. The working electrode was polished for at least 4 min before individual cycles of voltammetry. Unless otherwise stated, applied potentials ranged from -200 to +800 mV, and the scan rate was 200 mV s-1. The volume of the voltammetric cell was approximately 15 ml.
Vitamin C (ascorbic acid) was used as a standard antioxidant for comparison of the sugars. Cyclic voltammograms were constructed for vitamin C at concentrations from 5.1 ?M to 2.6 mM. Vitamin C (Krüger GmbH & Co., Germany) was dissolved freshly at the start of the experiments. All solutions were prepared using double-distilled water.

2.4. Diffusion coefficients

According to the Randles–Ševcik equation (Randles, 1948):
Ipa=kn3/2AD1/2C.?1/2
where Ipa = peak anodic current; k = Randles–Ševcik constant (2.69 × 105 A.s.V-1/2 mol-1) at 25 °C; n = electron stoichiometry (i.e., the number of electrons transferred to the substrate molecule); A = exposed surface area of the electrode (cm2); D = diffusion coefficient of the substrate molecule (cm2 s-1); C = concentration of the substrate molecule (mol cm-3); ? = scan rate (V s-1).
Thus, if the concentration of the substrate is kept constant, a plot of Ipa vs. ?½ will yield a straight line whose slope can be used to determine the diffusion coefficient. Samples of sugars were scanned at rates from 50 to 3200 V s-1, in steps of log2 units.

2.5. Redox potentials and peak potentials

In cases where the oxidation of compounds was, at least partially, reversible, the redox potential was calculated form (Epa + Epc)/2 and the peak potential was calculated from (Epc - Epa), where Epa and Epc were the potentials at which the peak anodic and peak cathodic currents (Ipa and Ipc, respectively) were recorded.

3. Results

3.1. Vitamin C standards

Over a range of 5.1 ?M to 2.6 mM, vitamin C produced a concentration-dependent peak anodic current (Ipa) (Fig. 1 and Fig. 2). The potential (Epa) at which Ipa occurred was 168–218 mV for concentrations from 5.1 ?M to 2.6 mM (Fig. 2). There were no observable peak cathodic currents (Ipc).

 

 

Full-size image (21K) - Opens new window


Fig. 1. Vitamin C. Cyclic voltammograms for vitamin C in 0.1 M phosphate buffer, pH 6.2. The numbers 1 ? 3 correspond to concentrations of 0.010, 0.026 and 0.051 mM vitamin C in phosphate buffer. Inset are differentiated recordings, with the numbers 1 ? 4 corresponding to concentrations of 0, 0.010, 0.026 and 0.051 mM vitamin C in phosphate buffer.

 

 

Full-size image (37K) - Opens new window


Fig. 2. Characteristics of vitamin C by voltammetry. (A) Vitamin C produces a concentration-dependent linear increase in peak anodic current (Ipa). (B) The double reciprocal plot of View the MathML sourceon concentration-1 is linear. (C) The first order derivative of the Ipa also has a linear dependency on concentration. (D) The first order derivative has a linear relation with the Ipa. (E) From 5.11 ?M to 1.02 mM there was no significant change in the potential (Epa) at which the Ipa occurred, only at the highest concentration tested (*, 2.56 mM) does the Epa lie significantly more positive (two-way ANOVA and post hoc t-test). (F). For dIpa/dt, there was no significant shift in Epa over a concentration range from 5.11 ?M to 1.02 mM, but at 2.56 mM (*) the rightward shift is statistically significant (< 0.01) two-way ANOVA and post hoc t-test).

The relationship between Ipa and concentration of vitamin C appeared to be linear over the concentration range of 5.1 ?M to 2.6 mM (Fig. 2A). When both the ordinate and abscissa were plotted on reciprocal scales the regression was also linear (Fig. 2B). The slope of this transformation, 0.03 ± 0.007 mM ?A-1, gives the concentration of vitamin C per unit Ipa which was termed the "current constant".
The voltammograms were subjected to first order differentiation with respect to time (using eChem software). The resulting transformed voltammograms give d(Ia)/dt with a peak corresponding to the fastest rate of increase of the Ipa (Fig. 1 inset). In other words, the differentiated traces give the rate of oxidation of vitamin C. There was a linear correlation between the amplitude of the signal produced by vitamin C and the rate of increase of that signal (Fig. 2C).
When the concentration of vitamin C was kept fixed at either 0.05, 0.25 or 0.51 mM, the Ipa was linearly dependent upon the scan rate for a range from 50 to 3200 mV s-1 (Fig. 2D). Using the slope of this regression and the Randles–Ševcik equation, the diffusion coefficient of vitamin C was calculated and found to be 1.89 × 10-6 ± 2.67 × 10-7 (mean ± s.e. mean, of the three different concentrations of vitamin C, three different samples each measured in duplicate using two different recording electrodes). The negative logarithm of this, i.e. the pD, is 5.74 ± 0.064 (n = 3).

3.2. Sugars

Refined white granulated sugar, up to 50% w/v buffer, gave a signal no different from the phosphate buffer alone. No signal was found for two of the three red sugars (Nabila gula merah and STM gula merah), nor for the white or yellow rock sugars, nor for gula Melaka palm sugar, nor for the two Luo Han Guo mixed sugars. No peaks were observed in the anodic or cathodic directions of the cyclic voltammograms, nor were any peaks identified on the differentiated traces. Potentials were applied from -200 to +1200 mV.
In contrast, anodic current peaks were found in four refined cane sugars (SIS raw sugar, Nazri soft brown sugar, China rock honey sugar and gula merah) and one unrefined palm sugar (gula anau) (Fig. 3, Fig. 4, Fig. 5, Fig. 6 and Fig. 7, respectively). The cyclic voltammograms for SIS raw sugar, soft brown sugar and China rock honey sugar had more than one peak, however the smaller peaks tended to be obscured by the main peak; it was often easier to discern that there was a second or third compound present from the differentiated recording (Fig. 3, Fig. 4 and Fig. 5).

 

 

Full-size image (22K) - Opens new window


Fig. 3. SIS raw sugar. Cyclic voltammograms for SIS raw sugar in 0.1 M phosphate buffer, pH 6.2. The numbers 1 ? 4 correspond to concentrations of 3%, 6%, 12% and 25% (w/v) sugar in phosphate buffer. Inset are differentiated recordings, with the numbers 1 ? 5 corresponding to concentrations of 0%, 3%, 6%, 12% and 25% (w/v) sugar in phosphate buffer.

 

 

Full-size image (24K) - Opens new window


Fig. 4. Soft brown sugar. Cyclic voltammograms for soft brown sugar in 0.1 M phosphate buffer, pH 6.2. The numbers 1 ? 5 correspond to concentrations of 1%, 3%, 6%, 12% and 25% (w/v) sugar in phosphate buffer. Inset are differentiated recordings, with the numbers 1 ? 6 corresponding to concentrations of 0, 1, 3, 6, 12 and 25% (w/v) sugar in phosphate buffer.

 

 

Full-size image (24K) - Opens new window


Fig. 5. China rock honey sugar and chrysanthemum flowers infusion. Cyclic voltammograms for: (A) China rock honey sugar in 0.1 M phosphate buffer, pH 6.2. The numbers 1 ? 4 correspond to concentrations of 3%, 6%, 12% and 25% (w/v) sugar. Inset are differentiated recordings, with the numbers 1 ? 5 corresponding to concentrations of 0%, 3%, 6%, 12% and 25% (w/v) sugar in phosphate buffer; (B) chrysanthemum flower infusion in phosphate buffer, pH 6.2. The background (phosphate buffer alone) has been subtracted. The numbers 1 ? 4 correspond to concentrations of 0.5%, 1%, 2% and 5% (v/v) infusion. Inset are current recordings, with the numbers 1 ? 3 corresponding to concentrations of 6.25% (w/v) China rock honey sugar in phosphate buffer, chrysanthemum flowers infusion 2% (v/v) in phosphate buffer, and an equal volume mixture of these two.

 

 

Full-size image (19K) - Opens new window


Fig. 6. Gula merah. Cyclic voltammograms for gula merah in 0.1 M phosphate buffer, pH 6.2. The numbers 1 ? 4 correspond to concentrations of 1.5%, 3%, 6%, and 12% (w/v) sugar in phosphate buffer. Inset are differentiated recordings, with the numbers 1 ? 6 corresponding to concentrations of 0%, 1%, 3%, 6%, 12% and 25% (w/v) sugar in phosphate buffer.

 

 

Full-size image (24K) - Opens new window


Fig. 7. Gula anau. Cyclic voltammograms for gula anau in 0.1 M phosphate buffer, pH 6.2. The numbers 1 ? 5 correspond to concentrations of 1%, 2%, 5%, 10% and 20% (w/v) sugar in phosphate buffer. Inset are differentiated recordings, with the numbers 1 ? 7 corresponding to concentrations of 0%, 1%, 2%, 5%, 10%, 20% and 50% (w/v) sugar in phosphate buffer.

The relationship between the Ipa amplitude and the concentration was similar for all the five positive sugars, in that for low level signals the relationship looked linear but at higher level signals (higher concentrations) the relationship deviated from linearity as if to plateau. Using double-reciprocal plots, the graphs were linearised (Fig. 8); as for vitamin C, the slope of this transformation gave the concentration of analyte per unit Ipa, or current constant (Table 1), and allows a direct comparison between the sugars and vitamin C. Assuming that the Ipa reflects antioxidant activity, comparing these values with vitamin C gives a rank order of antioxidant activity of the five sugars: gula anau gula merah > China rock honey sugar > soft brown sugar > SIS raw sugar.

 

 

Full-size image (38K) - Opens new window


Fig. 8. Sugars displaying peak anodic currents (Ipa). (A) SIS raw sugar peak1 (open upward triangle); SIS raw sugar peak2 (closed upward triangle). (B) Soft brown sugar peak1 (open downward triangle); soft brown sugar peak2 (closed downward triangle). (C) China rock honey sugar (closed diamond). (D) Gula merah (closed circle). (E) Gula anau (open circle). Concentrations are given as w/v phosphate buffer. Graphs show concentration dependency of Ipa, insets show double-reciprocal plots of View the MathML sourceversus concentration-1. Regression slopes and correlation coefficients are given in Table 1 (N.B. For SIS brown sugar, the points at a concentration of 50% were not included in the regression). Points show mean ± s.e. mean, unless occluded by symbol.

The red sugar, sunflower gula merah, was certainly red in colour. When it was being dissolved it was noticeable that the red colour came away from the sugar crystals so that in the early stages of dissolution there was a reddish coloured liquid with white crystals on the bottom of the beaker. The same was true for the two negative red sugars, Nabilah gula merah and STM gula merah.
The two refined brown sugars and the China rock honey sugar contained compounds whose oxidation appeared to be at least partially reversible, because the cathodic current traces in the cyclic voltamograms also displayed peaks (Ipc) (Fig. 3, Fig. 4 and Fig. 5).
For the China rock honey sugar, for the first peak, which was due to a substance with a redox potential of 258 ± 8.4 mV (n = 5), the Ipa/Ipc ratio was 1.1 ± 0.18 (n = 5) and the peak potential was 44.7 ± 5.28 mV (n = 5). This peak was only measurable at a concentration of 3.13% (w/v) buffer because otherwise it merged into the peak for a second substance. This second peak was due to a substance with a redox potential of 302 ± 6.7 mV (n = 6, at five concentrations from 3.13% to 50% w/v buffer); the Ipa/Ipc ratio was 2.3 ± 0.26 and the peak potential was 54.6 ± 3.80 mV.

3.3. Chrysanthemum flowers

The solutions of infused dried chrysanthemum flowers yielded multiple peaks in the cyclic voltammograms. Two prominent peaks were discernible below 400 mV and occurred close together (Fig. 5). The potentials at which these peaks occurred were similar to those observed in the China rock honey sugar, and when the China rock honey sugar (6.25%) was mixed with an equal volume of the chrysanthemum infusion (2%), there were still only two peaks in either the anodic or cathodic current traces (Fig. 5). Furthermore, on the differentiated traces, no additional peaks were seen. The first peak, was due to a substance with a redox potential of 238 ± 1.3 mV (n = 5); the Ipa/Ipc ratio was 1.0 ± 0.04 (n = 5) and the peak potential was 38.4 ± 17.7 mV (n = 5). This peak was measurable only at concentrations from 0.1% to 2% (w/v) buffer because otherwise it merged into the second peak. The second peak was due to a substance with a redox potential of 284 ± 5.3 mV (n = 7), at concentrations from 0.5% to 50% w/v buffer; the Ipa/Ipc ratio was 1.9 ± 0.37 and the peak potential was 52.9 ± 18.3 mV. The peak potentials for both the first and second peaks were not significantly different from those for the corresponding peaks in the China rock honey sugar (unpaired Student's t-tests). Likewise, the peak ratios did not significantly differ between the flower infusion and the sugar, nor did the redox potentials (unpaired t-tests, > 0.05).

3.4. Peak anodic current potentials

Of all the compounds tested, vitamin C had the lowest Epa, at approximately 200 mV (Fig. 9). None of the sugars had compounds with an oxidation potential in this region. The main antioxidant in the China rock honey sugar, the raw sugar (peak1), the soft brown sugar (peak1) and the chrysanthemum infusion all had values for Epa in the region of 300–400 mV. The antioxidant compounds in gula anau and in the second peaks of the refined raw sugar and the soft brown sugar had potentials that were similar at 500–650 mV. The compound in gula merah, that yielded an anodic current peak, had a very high potential, greater than 700 mV. Furthermore, the peak potential of this compound became more positive, in a concentration-dependent fashion (Fig. 8), up to 900 mV at the highest concentration tested (50% w/v).

 

 

Full-size image (24K) - Opens new window


Fig. 9. Relationships between peak anodic currents (Ipa) and oxidation potentials (Epa). Vitamin C (closed star), chrysanthemum flowers infusion (open diamond), China rock honey sugar (closed diamond), SIS raw sugar peak1 (open upward triangle), soft brown sugar peak1 (open downward triangle, occluding open upward triangles), gula anau (open circle), SIS raw sugar peak2 (closed upward triangle), soft brown sugar peak2 (closed downward triangle) and gula merah (closed circle). The highest concentrations tested, and the highest Ipa for each analyte, were: vitamin C, 2.6 mM; chrysanthemum flower infusion 50% v/v buffer; all sugars, 50% w/v buffer.

3.5. Diffusion coefficients

For the five positive sugars, vitamin C and the chrysanthemum flower infusion, the plots of Ipa against the square root of the scan rate are shown in Fig. 10. In each case, the regression of Ipa on the square root of the scan rate was linear (least-squares regression); the regression lines were constrained to passing through the origin and their slopes are given in Table 1 together with their correlation coefficients. For the China honey rock sugar, and for the infusion of chrysanthemum flowers, there was not a clear enough separation of the two peaks to allow separate analysis, and data for the Ipa that occurred around 300 mV in both cases are shown. The slope of the regression is the value C.D0.5 (from the Randles–Ševcik equation), but the diffusion coefficients for compounds contained in the sugars could not be calculated, because their concentration is unknown.

 

 

Full-size image (23K) - Opens new window


Fig. 10. Regressions of peak anodic currents (Ipa) on the square root of the scan rate (?0.5). Vitamin C (10%, open square; 5% closed star; 1% open star), chrysanthemum flowers infusion (open diamond), China rock honey sugar (closed diamond), SIS raw sugar peak1 (open upward triangle), soft brown sugar peak1 (open downward triangle), gula anau (open circle), SIS raw sugar peak2 (closed upward triangle), soft brown sugar peak2 (closed downward triangle) and gula merah (closed circle). All the sugars were tested at concentrations of 10% w/v buffer; the chrysanthemum flower infusion was tested at 10% v/v buffer.

4. Discussion

The results show that some refined and unrefined sugars do contain antioxidants. Previous studies relating the redox potential of a compound to a lipid peroxidation assay showed that compounds with a redox potential below 450 mV were antioxidant, and that compounds with a redox potential of 490 mV or above were pro-oxidant (Simic, Manojlovic, Šegan, & Todorovic, 2007). The highest levels of antioxidant activity appeared to be found in the palm sugar gula anau. This sugar had an antioxidant activity equivalent to 1.7 mg of vitamin C per 1 g of gula anau. The nature of the antioxidant has not been described, but it is unlikely to be vitamin C itself because, even though the diffusion coefficient of this compound was close to that of authentic vitamin C, the potential at which the peak anodic currents occurred was approximately 100 mV more positive than that of authentic vitamin C.
A red sugar, gula merah, also contained a substance in appreciable amount that was oxidisable, however, the potential at which the peak anodic current arose were very high, around 700 mV. This means that this substance is not a powerful antioxidant (Simic et al., 2007). Because the colour readily came away from the sugar crystals during dissolution, and also because the red colour of this product did not look like that of any other sugar, there is some doubt as to the origin of this antioxidant. No further information has been supplied by the manufacturer.
The mixed sugar, China rock honey sugar, had a comparatively low level of antioxidant activity. This is almost certainly entirely due to its chrysanthemum flower content. The two peaks of anodic current, that were present in this mixed sugar, were also present in aqueous extracts of dried chrysanthemum petals alone. When samples of the sugar and flowers were mixed, there were still only the two anodic peaks. Furthermore, the Ipa/Ipc ratios, the redox potentials and the peak potentials for the two substances did not significantly differ between the two samples. Finally, the predominant peaks in these two substances also had similar diffusion coefficients. Two other mixed sugars with similar or identical contents apart from the chrysanthemum flowers had no discernible antioxidant activity. Therefore, it is probable that any antioxidant activity present in the China rock honey sugar is due to the dried chrysanthemum flowers.
The refined SIS raw sugar and the Nazri soft brown sugar both had low levels of antioxidant activity from two compounds. From the current constants, assuming that the compounds in the two sugars are the same, the concentrations of both compounds in the soft brown sugar were just over double (2.25 and 2.22 times great for peaks 1 and 2, respectively) those in the raw sugar; the oxidation potentials for the compounds were the same. When this difference in concentration is taken into account, the diffusion coefficients for the compound of peak1 become almost identical, while those for peak2 become very close. The main difference between these two products appeared to be particle size; the raw sugar was composed of larger crystals.
From the potentials at which the anodic current peaks arose, it would appear that the antioxidant in the refined raw and soft brown sugars could be the same as that in the unrefined palm sugar. From the values of the current constants, the amount present in the gula anau was 20 times greater than that in the raw sugar. Correcting the value of C.D0.5 to reflect this 20-fold difference, gives a compound with the same diffusion coefficient in the two sugars (the corrected C.D0.5 for gula anau becomes 1.47 ?mol cm-2.s-0.5 c.f. 1.5 ?mol cm-2 s-0.5 for the raw sugar).
The antioxidants in the chrysanthemum petals and the China rock honey sugar were the most powerful amongst the sugars, arising at potentials approximately 100 mV lower than any other. As noted above, it was not always possible to resolve the peaks for these compounds because they merged into each other. Antioxidant activity has been reported in chrysanthemum flower buds from three different species (Woo et al., 2008), and at least 17 polyphenol compounds which may have antioxidant activity can be extracted from Chrysanthemum morifolium flowers (Lai et al., 2007).
The calculated diffusion coefficient for vitamin C (1.89 × 10-6 cm s-1) is in agreement with values in the literature obtained using modified carbon paste electrodes (5.7 × 10-6 cm s-1, Marian, Sandulescu, & Bonciocat, 2000) or glassy carbon electrodes (3.2 × 10-6 cm s-1, Sabzi & Pournaghi-Azar, 2005).
The cyclic voltammograms for gula anau show a remarkable similarity to those for vanillic acid. Simic et al. (2007) showed that vanillic acid (1 mM) has a single broad-based peak at a potential of 0.73 mV, with a linear relationship between Ipa and the square root of the scan rate. The voltammograms also showed no peak on the reduction curve. These authors also used a glassy carbon electrode with a Ag/AgCl reference electrode. Vanillic acid is among several antioxidant compounds found in refined or coated brown sugars from sugarcane (Payet, Sing, & Smadja, 2005).
Using a colourimetric assay which determined the total amount of antioxidants in a sample, Halvorsen et al. (2006) showed that raw cane sugar had nearly twice the antioxidant capacity of turbinado sugar, and that the level of antioxidant activity in refined granulated sugar was approximately 80 times less than that in the raw cane sugar. Using cyclic voltammetry, antioxidant activity was not detected in any of the white or yellow sugars that were tested.
In conclusion, it is evident that most types of refined sugars on the market do not contain any antioxidants. Very low levels of antioxidant activity were found in the refined raw and brown cane sugars. Higher levels were found in a rock honey sugar (also refined) which also contained chrysanthemum flowers, and the activity was due to the flowers.
An oxidisable substance was found at appreciable levels in the unrefined palm sugar, gula anau, however whether or not this functions as an antioxidant remains to be determined. If it is, then gula anau, should be promoted as the sweetener of choice because of its antioxidant content.

Acknowledgements

This work is funded by Kementarian Bangunan Negara Brunei Darussalam, under grant STIC-3 RKN9. The authors gratefully acknowledge this support. Dr. Khor Vi An, Kedai Ubat Beserah, Kuantan, Malaysia, is thanked for the gift of dried chrysanthemum flowers.

References

Halvorsen et al., 2006 B.L. Halvorsen, M.H. Carlsen, K.M. Phillips, S.K. Bøhn, K. Holte and D.R. Jacobs et al., Content of redox-active compounds (i.e. antioxidants) in foods consumed in the United States, American Journal of Clinical Nutrition 84 (2006), pp. 95–135. View Record in Scopus | Cited By in Scopus (47)
Kilmartin and Hsu, 2003 P.A. Kilmartin and C.F. Hsu, Characterisation of polyphenols in green, oolong, and black teas, and in coffee, using cyclic voltammetry, Food Chemistry 82 (2003), pp. 501–512. Abstract | Article | View Record in Scopus | Cited By in Scopus (33) Lai et al., 2007 J-P. Lai, Y.H. Lima, J. Sua, H-M. Shen and C.M. Ong, Identification and characterization of major flavonoids and caffeoylquinic acids in three Compositae plants by LC/DAD-APCI/MS, Journal of Chromatography B 848 (2007), pp. 215–225. Abstract | Article | PDF (1400 K) | View Record in Scopus | Cited By in Scopus (11) Li and Schellhorn, 2007 Y. Li and H.E. Schellhorn, New developments and novel therapeutic perspectives for vitamin C, Journal of Nutrition 137 (2007), pp. 2171–2184. View Record in Scopus | Cited By in Scopus (29)
Marian et al., 2000 I.O. Marian, R. Sandulescu and N. Bonciocat, Diffusion coefficient (or concentration) determination of ascorbic acid using carbon paste electrodes in Fredholm alternative, Journal of Pharmaceutical and Biochemical Analysis 23 (2000), pp. 227–230. Abstract | Article | View Record in Scopus | Cited By in Scopus (7)
Payet et al., 2005 B. Payet, A.S.C. Sing and J. Smadja, Assessment of antioxidant activity of cane brown sugars by ABTS and DPPH radical scavenging assays: determination of their polyphenolic and volatile constituents, Journal of Agricultural Food Chemistry 53 (2005), pp. 10074–10079. View Record in Scopus | Cited By in Scopus (10)
Randles, 1948 J.E.B. Randles, A cathode ray polarograph. Part II. The current–voltage curves, Transaction of the Faraday Society 44 (1948), pp. 327–338.
Roginsky et al., 2003 V. Roginsky, T. Barsukova, C.F. Hsu and P.A. Kilmartin, Chain-breaking antioxidant activity and cyclic voltammetry characterization of polyphenols in a range of green, Oolong, and black teas, Journal of Agricultural Food Chemistry 51 (2003), pp. 5798–5802. View Record in Scopus Cited By in Scopus (14)
Sabzi and Pournaghi-Azar, 2005 R.E. Sabzi and M.H. Pournaghi-Azar, Electrocatalytic determination of ascorbic acid on a glassy carbon electrode chemically modified with cobalt pentacyanonitrosylferrate, Analytical Sciences 21 (2005), pp. 689–692. View Record in Scopus | Cited By in Scopus (9)
Simic et al., 2007 A. Simic, D. Manojlovic, D. Šegan and M. Todorovic, Electrochemical behavior and antioxidant and prooxidant activity of natural phenolics, Molecules 12 (2007), pp. 2327–2340. View Record in Scopus | Cited By in Scopus (6)
Valdés, 2006 F. Valdés, Vitamina C, Actas Dermo-sifiliográficas 97 (2006), pp. 557–568. Abstract | View Record in Scopus | Cited By in Scopus (5)
Vita, 2005 J.A. Vita, Polyphenols and cardiovascular disease: effects on endothelial and platelet function, American Journal of Clinical Nutrition 81 (2005), pp. 292S–297S.
Woo et al., 2008 K.S. Woo, J.S. Yu, I.G. Hwang, Y.R. Lee, C.H. Lee and H-S. Yoon et al., Antioxidative activity of volatile compounds in flower of Chrysanthemum indicum, C. morifolium, and C. zawadskii, J. Korean Society Food Science Nutrition 37 (2008), pp. 805–809. View Record in Scopus | Cited By in Scopus (1)

------------------------------------------------

Gula Java Fin Kokosblüten ZuckerAmanprana Gula Java Fin Kokosblüten ZuckerVollkorncrepes mit Kokosblüten ZuckerGula Java Kakao Kokosblüten ZuckerAmanprana Gula Java Blocs Kokosblüten ZuckerAmanprana Gula Java Brut Kokosblüten Zucker