Food Science and Technology Research
Online ISSN : 1881-3984
Print ISSN : 1344-6606
ISSN-L : 1344-6606
Original papers
Effect of Thermostability of Granulated Sugars on the Flavor of Cooked Food
Kaoru SakamotoShiro KishiharaNaoko Kataoka-Shirasugi
Author information
JOURNALS FREE ACCESS FULL-TEXT HTML

2018 Volume 24 Issue 1 Pages 111-118

Details
Abstract

This work addresses how granulated sugars with different melting points from two different refineries affect the taste of cooked food. Candied sweets and caramel sauces were prepared using the two types of granulated sugars. According to sensory evaluation, the candies and caramel sauces prepared by heating the wet sugar crystals, which remained undissolved in the solution, showed different tastes. This was confirmed by the color difference analysis and HPLC analysis of these two caramel sauces. However, the caramel sauces prepared from aqueous solutions of the sugars were not significantly different from each other in flavor. These results suggest that granulated sugars with different heating characteristics may have different crystal structures. During cooking, in which the sugar crystals are heated without being completely dissolved, the heating and melting characteristics of the sugars should be taken into consideration.

Introduction

Granulated sugar is widely used in cooking. Commercial granulated sugar in Japan is of very high purity (Takada et al., 2003) and can be regarded as sucrose crystals. The melting point of sucrose crystals is between 160°C and 190°C (Okuno et al., 2003a). The melting point of granulated sugar from different sugar factories and refineries has been reported to be between 168°C and 183°C (Kishihara et al., 2004). Profiles of differential scanning calorimetry (DSC)-endothermic curves for the sugars have been shown to vary, and the curve for the sugar with the lowest melting point has two peaks (at around 150°C and 170–180°C), whereas the curve for the sugar with the highest melting point has only one sharp peak at a higher temperature (around 190°C) (Kishihara et al., 2004; Sakamoto et al., 2005, 2006).

The degradation of sucrose upon heating was previously studied using three granulated sugars obtained from sugar beet and two granulated sugars obtained from sugar cane. Some sugars from both sugar beet and sugar cane had higher melting points, and others had lower melting points; therefore, difference in the melting points of granulated sugars had no relation to the origin of the sugars, i.e., from sugar beet or sugar cane (Sakamoto et al., 2005).

Commercial granulated sugars are regarded to be of equal quality because of their high purity. Sugar is known to turn brown during caramelization, following heating at a temperature higher than its melting point. Nevertheless, some commercial granulated sugars were observed to turn yellowish-brown at temperatures below its melting point, when the sugar was still being heated, while another granulated sugar was less tinted (Kishihara et al., 2004; Sakamoto et al., 2005, 2006).

Three reactions are likely relevant to the heat-induced browning of sugar; namely, caramelization, an amino-carbonyl reaction, and a reaction derived from polyphenols. Browning of the granulated sugar may have been caused mainly by caramelization of residues of reducing sugars, anhydrosugars, and/or melted sucrose derived by the breakdown or collapse of weaker regions within the crystal structure, rather than by such impurities as amino acids or polyphenols (Sakamoto et al., 2006).

There are some reports that suggest that differences in the melting points of sugars are caused by the impurities contained in sugar (Powers, 1958; Roos and Karel, 1991; Roos 1993; Roos et al., 2013). Conversely, some studies suggest that they are caused by differences in the crystal structure of the sugars (Brown and Levy, 1963, 1973; Hanson et al., 1973; Bock and Lemieux, 1982; McCain and Markley, 1986ab; Davies and Christofies, 1987). We theorized that the difference in behavior of the granulated sugars or sucrose crystals upon heating is due to the structure of the crystals, rather than the impurities included in the crystals (Okuno et al., 2003bc; Kishihara et al., 2004; Sakamoto et al., 2006, 2007, 2011, 2013; Sakuda et al., 2008).

In cooking, when granulated solid sugar is heated or when the portion of solid that remains undissolved is heated, the melting characteristics of the granulated sugar may have an influence on the quality of the products. Such characteristics of sugars may influence the taste and flavor of cooked foods upon heating. To examine this effect, candy and caramel sauce were prepared and subjected to sensory evaluation, and the sugar composition was analyzed.

Materials and Methods

Sample sugars    Two granulated sugars from different refineries were used. The two sugar samples (W sugar and Z sugar), which are both commercially available products, were provided by two Japanese sugar-refining companies. The sugars were of high purity (> 99.9%). Their melting points were determined to be 168°C for W sugar and 183°C for Z sugar, by immersing glass capillaries packed with each sugar in a heated oil bath.

Preparation of candy    “Bekkoame”, a Japanese traditional candy, was prepared as the candy sample (Hidaka et al., 2009). One gram of granulated sugar was flattened and put into a silicone-coated paper cup. The cup was heated for 18 min in an oven pre-heated to 180°C. Then, the cup was allowed to stand on a stainless steel tray for 30 min. The candies prepared from W sugar and Z sugar were named W candy and Z candy.

Preparation of caramel sauce    Caramel sauce was prepared by two different methods. The heat stability of sucrose molecules when heated in the presence of remaining crystals is thought to be different from when sucrose is heated with all crystals dissolved. The two methods for preparing caramel sauce are as follows:

Method A: heating of wet sugar crystal (where the sugar crystal remains undissolved in the solution)

Method B: heating of aqueous solution of sugar (where the crystal is completely dissolved)

Method A: 100 g of the sugar sample and 50 mL of water were heated with stirring in a 1000 mL stainless steel vessel. Undissolved sugar crystal in the solution was heated on a hotplate stirrer. After the temperature of the solution reached 190°C, the solution was heated for a further 1 min. Then, 50 mL of water was added to the solution, and the mixture was transferred to a beaker for cooling. Heating of the sugar solution was controlled to take 9–10 min to reach 190°C from room temperature (25°C).

Method B: 100 g of the sugar sample was completely dissolved in 50 mL of water with stirring, but without heating, in a 1000 mL stainless steel vessel. The sugar solution was then heated on a hotplate stirrer and prepared using the same method as Method A.

The caramel sauces prepared from W sugar and Z sugar were named W sauce and Z sauce.

X-ray diffraction analysis    X-ray diffraction (XRD) analysis of the sugar and candy samples was carried out using an X-ray diffractometer (MiniFlexII; Rigaku Co., Ltd., Tokyo Japan). The samples were exposed to X-ray beams at 15 mA and 30 kV. Data were recorded over a diffraction angle (2θ) range of 10° to 70° with a step angle of 0.02°.

Differential scanning calorimetry measurements    Granulated sugars and candies were placed in a pan. The pan was maintained at 50°C. The samples were heated from 50 to 200°C at a heating rate of 10°C/min, and were examined using a DSC instrument (Pyris1 DSC; PerkinElmer, Inc., Waltham, MA, U.S.A.). An empty pan was used as the reference.

Sensory evaluation    Sensory evaluation of the two candies and two caramel sauces was carried out according to the paired comparisons method and Semantic differential method. Comparison of the two candies was made using a two-sided Wilcoxon's signed test. Participants for the evaluation were from a women's junior college (ages 18–19). Thirty-six participants evaluated the candies and the caramel sauces made from the aqueous sugar solution, and 82 evaluated the caramel sauces made from the wet sugar crystal.

Determination of pH and color difference analysis    The pH of 10-fold dilutions of the candies and caramel sauces (10% water solutions) was determined using a pH Meter D-52 (Horiba, Kyoto, Japan).

The color differences of 40% (w/w) water solutions of candies and undiluted caramel sauces were analyzed by a Color Meter NE2000 (Nippon Densyoku, Tokyo, Japan). The colors were represented by L*, a*, and b*, following the widely used system. L* is lightness (0 to 100), the positive or negative value of a* is redness (0 to +60) or greenness (0 to −60), respectively, and the positive or negative value of b* is yellowness (0 to +60) or blueness (0 to −60), respectively. Color differences were calculated using the following equation.

  

High performance liquid chromatographic analysis of sugars    Twenty-fold dilutions of the candies and caramel sauces (5% water solutions) were analyzed. A high performance liquid chromatography (HPLC) system (Hitachi Model ELITE LaChrom; Tokyo Japan) consisting of pump L-2130, RI detector L-2490, column oven L-2300, and cosmosil packed column 5NH2-MS (4.5 mm ϕ×150 mm, Nacalai Tesque, Inc. Tokyo, Japan) was used for the analysis. The eluent composition was 70% acetonitrile/30% water. The flow rate was 1 mL/min.

Results and Discussion

X-ray diffraction pattern and differential scanning calorimetry measurements    The results of XRD are shown in Figure 1. XRD patterns of crystals were observed for granulated sugars, whereas the candies presented amorphous patterns.

Fig. 1.

XRD patterns of the sample sugars and candies

A: W sugar, B: Z sugar, C: W candy, D: Z candy

W candy and Z candy were prepared from W sugar and Z sugar.

The results of DSC are shown in Figure 2. The DCS-endothermic curve for W sugar (with the lowest melting point) shows a small peak at 150.7°C and has a broad rise around 150–190°C. The curve for Z sugar (with the highest melting point) contains one sharp peak at 190.5°C.

Fig. 2.

DSC measurements of the sample sugars and candies

A: W sugar, B: Z sugar, C: W candy, D: Z candy

W candy and Z candy were prepared from W sugar and Z sugar.

Sensory evaluation of candy    Results of sensory evaluation using the paired test are shown in Table 1. W candy was evaluated to have “higher sourness” (p < 0.05) and “higher bitterness in the first impression” (p < 0.05), whereas Z candy was evaluated to have “higher sweetness”, “more preferred aftertaste”, and “more preferred whole taste” (p < 0.01).

Table 1. Sensory evaluation of the two candies by paired test
W candy Z candy Significant difference
More preferred color 18 18 n.s.
More preferred aroma 18 18 n.s.
Higher sweetness 7 29 **
Higher sourness 25 11 *
Higher bitterness in the first impression 24 12 *
Higher bitterness in the aftertaste 22 14 n.s.
More preferred aftertaste 10 26 **
More preferred whole taste 8 28 **

The number of participants was 36.

Asterisks represent significantly different values (*: P < 0.05, **: P < 0.01, n.s.: P > 0.05).

W candy and Z candy were prepared from W sugar and Z sugar.

Results of sensory evaluation using the Semantic differential method are shown in Figure 3. The results were similar to those of the paired test; W candy had “higher bitterness” (p < 0.01), whereas Z candy had “higher sweetness” (p < 0.01), “more accustomed taste” (p < 0.05), and “more overall preference” (p < 0.05). These results revealed that W and Z candies have a different flavor (eating quality).

Fig. 3.

Sensory evaluation of the two candies using two-sided Wilcoxon's signed test according to Semantic differential method

The number of participants was 36.

Asterisks represent significantly different values (*: P < 0.05, **: P < 0.01).

W candy and Z candy were prepared from W sugar and Z sugar.

pH and color of candy    When granulated sugar is decomposed by heating, organic acid is formed, and it is thought that the pH may decrease. The pH of 10% aqueous solution of W and Z candies was compared. W candy showed lower pH (Table 2), which is consistent with the result that W candy had “higher sourness” in the sensory evaluation by the paired test.

Table 2. pH and color difference analysis of the two candies
Color b
pH a L* a* b*
W candy 4.0 47.2 −1.1 34.4
Z candy 4.2 48.3 −3.8 28.7
a  10% aqueous solution of candy,

b  40% aqueous solution of candy

L* is lightness, the positive or negative value of a* is redness or greenness, respectively, and the positive or negative value of b* is yellowness or blueness, respectively.

W candy and Z candy were prepared from W sugar and Z sugar.

The color of the two candies is shown in Table 2. When W and Z candies were compared, there were no significant differences in the values of L* and a*, but a significant difference was seen in the value of b*. While a color difference was not perceived by sensory evaluation, the result of color difference analysis showed variation between samples. The W candy was more yellowish than the Z candy, and the color difference (Δ E) between W candy and Z candy was 6.3. The relationship between the color difference and visibility is as follows: 0–0.5 = trace, 0.5–1.5 = slight, 1.5–3.0 = noticeable, 3.0–6.0 = appreciable, 6.0–12.0 = much, over 12.0 = very much (Stearns, E. I. 1951). Therefore, the difference between W and Z candy was “much”.

HPLC analysis of candy    The result of HPLC analysis of 5% aqueous solutions of the candy is shown in Figure 4. The amount of sucrose decreased compared to the initial amount. Peaks of fructose and glucose appeared in the HPLC chromatograph. There were quantitative differences in sucrose, glucose and fructose of the two candies. There was less sucrose in W candy than in Z candy, but more fructose and glucose were formed in W candy. In sensory evaluation, Z candy had not only “sweet taste” but also “good aftertaste”, “accustomed taste”, and “good taste”. This suggests that Z candy is sweeter than W candy because Z candy contains more sucrose than W candy.

Fig. 4.

HPLC analysis of the two candies

W candy and Z candy were prepared from W sugar and Z sugar.

Other smaller peaks could be seen in the HPLC chromatograph. This suggests that many kinds of thermolysis products were formed. When sucrose crystals are heated, caramelization occurs. It has been reported that during caramelization, anomerization and isomerization occur, which produce, as by-products, disaccharides and oligosaccharides when sugar is condensed, anhydrosugar by intramolecular rearrangement, and many other kinds of volatile products such as 5-hydroxymethylfurfural (Takada et al., 2003). In W candy, which showed “higher bitterness in the first impression” in the sensory test, most of the sucrose was decomposed and more fructose and glucose were formed. This implies that some of this decomposed matter formed bitter products.

Difference in caramel sauces prepared by the two methods    Caramelization represents the transformation of sugar upon heating and is an important phenomenon that produces a very characteristic flavor and color in cooking. This transformation is a very complicated reaction, whereby dehydration, decomposition, condensation, and polymerization happen simultaneously and numerous product materials are formed. The mechanism of caramelization has not been fully elucidated, but it is thought that caramels of different quality can be produced according to the reaction conditions. In cooking that exploits the characteristics of the heated sugar, melting properties should be taken into consideration. In this work, the two sugars from different refineries had different coloring and they partially melted when held at a lower temperature than their melting points (Kishihara et al., 2004; Sakamoto et al., 2005, 2006). The two kinds of granulated sugars used in this study had significantly different DSC-endothermic curve profiles (Kishihara et al., 2004; Sakamoto et al., 2005). Profiles of the DCS-endothermic curves for the sugars varied, with the curve for the sugar with the lowest melting point having two peaks (at around 150°C and at 170–180°C) and the curve for the sugar with the highest melting point having only one sharp peak at a higher temperature (around 190°C). It is thought that when crystals are directly heated, the differences in DSC-endothermic curves can influence the caramelization.

Caramel sauce is prepared by adding water to sugar crystals and heating. Typically, sugar is heated when it is not fully dissolved (Method A). In Method A, W sauce was evaluated to have “more preferred color” (p < 0.001), “more preferred aroma” (p < 0.05), and “higher bitterness in the first impression” (p < 0.05), whereas Z sauce was evaluated to have “higher sweetness” (p < 0.05), “more preferred aftertaste”, and “more preferred whole taste” (p < 0.01). Using Method B, no difference was found between W and Z caramel sauces. We assume that this is because in Method B, the caramel sauce was prepared by heating the completely dissolved sugar crystals.

The results of sensory evaluation using the Semantic differential method is shown in Figure 5. In Method A, W sauce had “more preferred color” (p < 0.05) and “higher bitterness” (p < 0.05), whereas Z sauce had “higher sweetness” (p < 0.01), “more preferred aftertaste”, “higher refreshness”, “more accustomed taste”, and “more overall preference” (p < 0.001). For Method B, the results were the same as for the paired test (Table 3), indicating that there is no difference between W sauce and Z sauce. This contrasts with Method A, in which a difference was found.

Fig. 5.

Sensory evaluation of the two caramel sauces using two-sided Wilcoxon's signed test according to Semantic differential method

Method A: heating of wet sugar crystal, Method B: heating of aqueous solution of sugar The number of participants was 82 in Method A, 36 in Method B.

Asterisks represent significantly different values (*: P < 0.05, ***: P < 0.001).

W sauce and Z sauce were prepared from W sugar and Z sugar.

Table 3. Sensory evaluation of the two caramel sauces by paired test
Method A Method B
W sauce Z sauce Significant difference W sauce Z sauce Significant difference
More preferred color 58 24 *** 19 16 n.s.
More preferred aroma 52 30 * 17 18 n.s.
Higher sweetness 31 50 * 17 18 n.s.
Higher sourness 39 42 n.s. 19 14 n.s.
Higher bitterness in the first impression 52 30 * 21 15 n.s.
Higher bitterness in the aftertaste 37 45 n.s. 21 15 n.s.
More preferred aftertaste 28 54 ** 13 23 n.s.
More preferred whole taste 29 53 ** 17 19 n.s.

Method A: heating of wet sugar crystal, Method B: heating of aqueous solution of sugar

The number of participants was 82 in Method A, 36 in Method B.

Asterisks represent significantly different values (*: P < 0.05, **: P < 0.01, ***: P < 0.01, n.s.: P < 0.05).

W sauce and Z sauce were prepared from W sugar and Z sugar.

pH measurement, color difference analysis, and HPLC analysis were also carried out. In Method B, as shown in Table 4 and Figure 6, there was no difference between the pH measurement and HPLC analysis of W and Z sugars, though there was slight difference in color. This result is clearly different from the results of Method A, which were obtained following heating of undissolved crystals.

Table 4. pH and color difference analysis of the two caramel sauces (method B)
Color b
pH a L* a* b*
W sauce 3.8 45.1 27.8 77.8
Z sauce 3.8 46.9 24.7 81.0
a  10% aqueous solution of sauce,

b  40% aqueous solution of sauce

L* is lightness, the positive or negative value of a* is redness or greenness, respectively, and the positive or negative value of b* is yellowness or blueness, respectively.

W sauce and Z sauce were prepared from W sugar and Z sugar.

Fig. 6.

HPLC analysis of the two caramel sauces (Method B)

Method B: heating of aqueous solution of sugar

Candies prepared from the two sugars were shown to differ in color by the color difference analysis and in flavor by the sensory evaluation. Caramel sauces prepared from the wet crystal (Method A) of the two sugars were also different from each other in color and flavor, while those from the aqueous solution of the sugars (Method B) were not significantly different from each other in flavor.

The above observations suggest that sucrose crystals may have some structural differences.

Conclusion

In our experiment, candies were prepared by directly heating granulated sugar crystals, with our results showing variations in taste, color and HPLC chromatogram between two kinds of sugars from different sugar refineries. This suggests that the granulated sugars have different heating characteristics, such as melting point, despite their high purity, and that these characteristics can influence the color and flavor of foods prepared by heating the sugar crystals. In contrast, when the caramel sauce was prepared using completely dissolved crystals, no differences were apparent.

These results suggest that granulated sugars with different heating characteristics may have different crystal structures.

Acknowledgments    The two sugar samples were provided by Itochu Sugar Co., Ltd. and Mitsui Sugar Co., Ltd. Parts of this study were supported by A-STEP (Adaptable and Seamless Technology transfer Program through target driven R&D), a program to support the research activities of female researchers (Collaborations) funded by MEXT and JSPS KAKENHI Grant Numbers JP22500733, JP17K00823.

References
 
© 2018 by Japanese Society for Food Science and Technology
feedback
Top