Effect of modified tapioca starch and xanthan gum on low temperature texture stability and dough

Journal of Food Engineering 119 (2013) 446–453 Contents lists available at SciVerse ScienceDirect Journal of Food Engineering journal homepage: www.elsevier.com/locate/jfoodeng Effect of modified tapioca starch and xanthan gum on low temperature texture stability and dough viscoelasticity of a starch-based food gel Nispa Seetapan ⇑, Asira Fuongfuchat, Chaiwut Gamonpilas, Pawadee Methacanon, Waranit Pongjaruwat, Nattawut Limparyoon National Metal and Materials Technology Center, 114 Thailand Science Park, Paholyothin Road, Pathumthani 12120, Thailand a r t i c l e i n f o Article history: Received 28 February 2013 Received in revised form 7 June 2013 Accepted 10 June 2013 Available online 20 June 2013 Keywords: Starch wrapper Texture stability Dough viscoelasticity Modified tapioca starch Xanthan gum a b s t r a c t This work investigated the effect of modified tapioca starch and xanthan gum on dough viscoelasticity and texture stability during storage at 4 °C of starch sheets for Chinese shrimp dumplings. Hydroxypropylated starch and hydroxypropylated-crosslinked starch were used to substitute for tapioca starch in the control formulation, and xanthan gum was added to adjust the formulation. During storage, texture of the control became firmer due to amylopectin retrogradation confirmed by differential scanning calorimetry and X-ray diffraction. Conversely, gel sheets containing modified starches showed less texture change. Dough viscoelasticity of the formulations substituted by hydroxypropylated starch were much softer and easier to deform than that of the control one. Dough with hydroxypropylated-crosslinked starch was, however, stiffer and more strain-resistant. Moreover, the formulation comprising the mixture of both types of modified starches and xanthan gum gave dough viscoelasticity similar to that of the control, and provided gel sheet the least texture change. Consequently, this modified formulation could be beneficial for the application of frozen/chilled dumpling wrappers. Ó 2013 Elsevier Ltd. All rights reserved. 1. Introduction Nowadays, consumers have changed their consumption behaviors to compensate for their hectic lifestyles. Therefore, demands of ready-meals have increased tremendously and, thus, competitions among food manufacturers have become very intense during the past decade. In order to compete with the market demands, most manufacturers have to continuously develop new food recipes with improved texture and extended shelf life. Freezing or chilling process is a crucial key in extension of food storage life while preserving the original food texture quality, particularly for starchbased ready meals. Frozen and refrigerated starch-based meals are commonly available in the South East Asia. Nevertheless, such foods are still less popularly consumed compared to those freshly cooked products since they typically have poor texture upon storage at low temperature as a result of starch retrogradation (Funami et al., 2005; Sae-kang and Suphantharika, 2006). Cooked gels of native starch tend to retrograde during cooling or freezing. Retrogradation generally leads to moisture loss and texture change, which significantly influences on quality of final products. The retrogradation is a reassociation process of starch molecules (amylose and amylopectin) after gelatinization (Lionetto ⇑ Corresponding author. Tel.: +66 2 564 6500; fax: +66 2 564 6445. E-mail address: nispam@mtec.or.th (N. Seetapan). 0260-8774/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jfoodeng.2013.06.010 et al., 2005; Zhou et al., 2011). To retard or inhibit the reassociation of starch molecules, chemical modification has been introduced to starch granules (Morikawa and Nishinari, 2000; Singh et al., 2007). In addition, hydrocolloids were reported to maintain food textural properties upon low temperature and long term storage as well (Arocas et al., 2009; Pongsawatmanit and Srijunthongsiri, 2008). Among hydrocolloids, xanthan gum has often been added in starch-based food since it provides excellent properties in terms of retarding retrogradation and syneresis, enhancing low-temperature storage and freeze–thaw stabilities, and modifying textural properties (Arocas et al., 2009; Muadklay and Charoenrein, 2008). The effect of low temperature storage on texture of starch pastes and gels has been investigated for several years (Fellows, 1997; Lee et al., 2002). Moreover, the study on starch gels which contains several ingredients has been established and reported in the literature (Arocas et al., 2009; Beak et al., 2004; BeMiller, 2011; Kindt et al., 2006; Pongsawatmanit et al., 2013). However, the fundamental knowledge on dough quality and texture stability of those semi-solid food gels upon storage was limited. Such information is considered crucial for food manufacturers such that appropriate raw material selection and process optimization can be readily followed to achieve the desirable final products. This study, therefore, explored the preparation of a semi-solid starch gel sheet model which is traditionally used as a Chinese steamed shrimp dumpling (Har Gow). Such model food is of great interest since it is one of the most popular frozen Chinese dim N. Seetapan et al. / Journal of Food Engineering 119 (2013) 446–453 sums and, more scientifically, its formulation adjustment is not simple, i.e., both dough deformability and cooked gel strength are required. The dough of dumpling wrapped sheet should be easily processable, i.e. easy to roll into sheet, and its cooked sheet should be sufficiently strong and elastic to hold the filling weight without breaking upon cooking and storage at low temperature. The wrapped sheet of Har Gow was generally prepared by the mixture of wheat starch as a major component and tapioca starch with the addition of boiling water and vegetable oil. However, these starches, especially the tapioca, are susceptible to retrogradation. Therefore, in order to retain the texture quality of Har Gow storing at low temperature, modified tapioca starches, i.e., hydroxypropylated starch and hydroxypropylated-crosslinked starch, were employed to replace tapioca starch in the control formulation. In addition, the application of xanthan gum was investigated. Specifically, the effects of the modified starch substitution and xanthan gum addition on dough viscoelasticity, retrogradation and textural properties of starch gel sheets were examined so that better material selection can be made for application in Chinese shrimp dumplings. 447 properties of the gel sheets was investigated. The sheets were stored in a refrigerator at 4 ± 1 °C for 1–3 days. Prior to characterizations described in the following sections, the sheets were removed from the refrigerator at the different storage times, and then allowed to equilibrate at 25 °C for about 1 h in an airtight sealed container. Each formulation was prepared in duplicate. 2.3. Rapid Visco Analyzer (RVA) measurement Pasting characteristics of all starches used in this study were obtained using a Rapid Visco Analyzer (RVA: RV4 Newport Scientific Instruments & Engineering, Australia) according to the ICC Standard No. 162 (ICC, 2000). A starch sample of 2.5 g was mixed with 25 ml of water and stirred at 960 rpm to form a slurry. The idle temperature was set at 50 °C and tests were performed as follows: 50 °C for 1 min at 160 rpm, the temperature was ramped up to 95 °C at a constant rate of 0.2 °C/s, then held at 95 °C for 210 s, and finally cooled down to 50 °C at the constant rate of 0.18 °C/s. All tests were carried out in triplicate. 2.4. Viscoelastic measurement of starch dough 2. Materials and methods 2.1. Materials Low protein wheat starch (WS) and tapioca starch (TS) were purchased from a local supermarket. Wheat starch contains 22.93 ± 0.60% amylose (Amylose/Amylopectin assay kit, Megazyme, Ireland), 0.50% protein (Kjeldahl method, %N  5.83) and 7.55 ± 0.11% moisture while tapioca starch with 6.80 ± 0.07% moisture contains less amount of amylose (17.48 ± 0.55%) and protein (trace). Three hydroxypropylated tapioca starches with different degrees of substitution (DS), i.e., TAPFROZ-1 (DS = 0.019), TAPFROZ-2 (DS = 0.063), and TAPFROZ-3 (DS = 0.102), and one hydroxypropylated-crosslinked tapioca starch (TAPFIL-8 with DS = 0.061) were provided by Tapioca Development Corp., Ltd., Thailand. TAPFROZ-1, TAPFROZ-2 and TAPFROZ-3 were represented as H1, H2 and H3, respectively, and TAPFIL-8 was called here as HX. Xanthan gum (200 mesh) was received from Ultimate Product Co., Ltd., Thailand. Soy bean oil bought from a local supermarket was used in a preparation of gel sheet. 2.2. Preparation of starch dough and cooked gel Wheat starch (25 g) and tapioca starch (4 g) were thoroughly mixed. This composition was used as a control formulation. The dry mixture was then heated to 65–70 °C, while stirring. Boiling water (40 g) was immediately poured into the heated mixture, then vegetable oil (2.3 g) was added. All ingredients were thereafter combined and kneaded to form dough, which was then sheeted into 2-mm thickness using a dough sheeter. A rectangular dough sheet of 7 cm  10 cm was steamed to gelatinize the starch for 15 min. Subsequently, the cooked starch sheet was allowed to cool to the room temperature. In the preparation of modified formulations, modified tapioca starches (i.e., H and HX) were employed individually or in combination to substitute for the tapioca starch in the control formulation in order to investigate the influence of starch modification on the wrapper stability upon chilled (4 °C) storage, while keeping a fixed wheat starch content as in the control formulation. Furthermore, xanthan gum was added to adjust the formulation. The cooking procedure was the same as the control. After steaming, the gel sheets were kept in airtight sealed containers at 25 °C for 1 h prior to perform mechanical testing. The effect of chilled storage on the starch retrogradation and textural Viscoelastic behaviors of doughs were measured using a rotational rheometer (ARES, TA Inc., New Castle, USA), equipped with a 20-mm diameter serrated stainless steel plate fixture, and 2 mm gap size. After loading, sample was trimmed and a thin layer of low viscosity silicone oil was applied at the sample edge to prevent dehydration of the dough. Samples were allowed to relax for 5 min at the operating temperature of 25 °C prior to testing. Dynamic strain sweep experiment was performed from 0.1% to 300% strain at a fixed angular frequency of 1 rad/s in order to determine the strength of dough and the sensitivity of dough upon the application of strain. All measurements were performed on freshly prepared doughs and two replicates were used for each preparation. 2.5. Mechanical testing of gel sheets Textural properties of cooked gel sheets were investigated using an indentation test. The test was performed with a 4-mm diameter spherical probe mounted on a Universal testing machine (Instron 5943, Instron, USA). A 100 N load cell and testing speed of 0.01 mm/s were used throughout the study. The test was performed up to a maximum indentation depth of 50% of the sample height. All measurements were conducted at 25 °C. From the indentation test, the modulus of the gel sheet can be evaluated by fitting the obtained indentation load with respect to the indentation depth using Eq. (1) (Gamonpilas et al., 2010) as follows: F¼ 16 1=2 2=3 ER h 9 ð1Þ where F is normal force (N), E is modulus (Pa), R is radius of indented probe (m) and h is an indentation displacement (m). Each sample was analyzed with 10 replicates. The mean value and standard deviation were reported. 2.6. X-ray diffractometry of retrograded starch powder Recrystallization of the retrograded starch powder was studied using an X-ray diffractometer (JDX-3530, JEOL, Japan) equipped with a copper tube operating at 40 kV and 40 mA, producing CuKa radiation of 0.154 nm wavelength. Diffractograms were obtained from scanning range of 4–40° (2h) at a rate of 1.2°/min, and a sampling interval of 0.02°. The samples were prepared by adapting the method of Hibi et al. (1990). Briefly, after storage at the desired temperature and time, the wrapper sheets were transformed into 448 N. Seetapan et al. / Journal of Food Engineering 119 (2013) 446–453 powder by consecutively dehydrating the sheets in a series of ethanol and water mixture at 30%, 50%, 70% and 100% ethanol. Total dehydration time was 24 h for each ethanol concentration. Finally the retrograded starch powder was precipitated and dried in a convection oven at 30 °C for 24 h, prior to XRD investigation. All retrograded powder had moisture content of 6–9%. 2.7. Differential scanning calorimetry of retrograded starch powder Thermal characteristics of the retrograded starch powder from stored gel sheets were studied using a differential scanning calorimetry (DSC 822e, Mettler Toledo, Schwerzenbach, Switzerland). Gel sheets were transformed into powder prior to testing using a similar procedure to the XRD analysis. Approximately 4 mg of dry retrograded powder was then mixed with deionized water (at a fixed ratio of starch: water of 1:2 by weight) and was placed in an aluminum pan. The pan was hermetically sealed and kept overnight prior to DSC analysis. A sealed empty pan was used as a reference. The sample was held isothermally at 20 °C for 1 min and subsequently heated from 20 to 90 °C at a heating rate of 5 °C min1 with a nitrogen flow rate of 60 cm3/min. Retrogradation melting enthalpy (DHr, J/g) relating to the retrograded starch, and transition temperatures; including onset (To), peak (Tp) and conclusion (Tc) temperatures, were recorded. Three replicates were used for each sample. apparent viscosity during pasting. Breakdown is the difference between a peak viscosity and a trough. Final viscosity is a viscosity after cooking and cooling. Setback is defined as the difference between a final viscosity and a trough. Pasting properties of all starches used in this work were summarized in Table 1. Wheat starch had the highest pasting temperature, the lowest values of peak viscosity, breakdown and final viscosity, indicating that this type of starch has high granule integrity. Tapioca starch and the modified tapioca starches used in this study, however, can swell more easily during gelatinization process as low pasting temperatures and high values of peak viscosity were observed. Among the hydroxypropylated tapioca starches (H1, H2, and H3), the significant decreased pasting temperature and the significant increased peak viscosity and breakdown (P < 0.05) were observed with increasing degree of substitution (DS) which indicated that the substituted groups facilitated the swelling process of starch granules. After cooling process of the experiment, tapioca starch had significantly higher final viscosity and setback than those of wheat starch. Larger retrogradation was thus anticipated in the former starch. More importantly, increasing DS (P0.06) of the hydroxypropylated starches led to a significant decrease in final viscosity and subsequently a significant decrease in setback because hydroxypropyl groups hindered the re-association of starch molecules (Raina et al., 2007). 3.2. Gel sheet incorporated with modified starch 2.8. Statistical analysis Experimental results were subjected to statistical analyses using the commercial SPSS 11.5 (SPSS Inc., Chicago, IL) computer program. Data were averaged and mean comparisons were evaluated using the least significant difference (LSD) technique at 95% confidence. A statistical difference at P < 0.05 was considered to be significant. 3. Results and discussion 3.1. RVA pasting characteristics RVA pasting profiles of wheat and tapioca starch used in this study were shown in Fig. 1. Similar profiles were obtained for modified tapioca starches. Pasting parameters were determined from the profile as follows. Pasting temperature is the temperature where the onset of viscosity increases. Peak viscosity is the highest Peak viscosity Final viscosity TS Trough WS Pasting temperature Fig. 1. RVA profiles of low protein wheat and tapioca starches. 3.2.1. Viscoelasticity of dough sheets Viscoelasticity is a crucial property that related to dough sheeting characteristics since the prepared dough must be rolled into a sheet. The results obtained were summarized in Table 2. Critical strain (cc) is defined as the smallest strain at which the material starts to deform under an applied strain. G0 is the storage modulus. Tan d is the ratio of G00 /G0 , where d is a phase angle, and G00 is the loss modulus. From Table 2, viscoelastic properties of the control dough was observed to be cc = 3.68%, G0 = 18 kPa, and tan d = 0.11. On the contrary, the pure wheat dough had the values of cc = 7.10%, G0 = 29 kPa, and tan d = 0.07 (data not shown in Table 2), suggesting a much stiffer and less susceptible to flow. The obtained data indicated that the presence of tapioca starch facilitated the deformation and flow of the control dough, resulting in a continuous dough sheet. This might be from the fact that during dough making process, tapioca starch could easily gelatinize (due to its low pasting temperature of about 69 °C), and thus acting as a binder for the partially gelatinized wheat granules of wheat starch (pasting temperature of 92 °C). When replacing tapioca starch by hydroxypropylated tapioca starch (H1, H2 and H3, individually), all modified doughs after equilibrating at room temperature had lower G0 and higher tan d, especially that incorporated with H3, than the control dough. Such results were in agreement with the final viscosity values from the aforementioned RVA test (Table 1) which showed that increasing DS led to the decrease in the final viscosity of the starch gel. In addition, cc of the H-modified doughs decreased inversely proportional to the DS, implying that the modified dough of high DS was vulnerable to flow under applied deformation. The obtained viscoelastic results indicated that modified doughs were softer and less strain-resistant than the control, especially the one substituted with H3. However, the extremely soft dough led to a problem in sheeting process, whereby a continuously smooth dough sheet could be difficult to achieve. 3.2.2. Mechanical properties of gel sheets at different chilled storage times All dough sheets were cooked and their gel textures before and after chilled storage at 4 °C for 1–3 days were determined by an 449 N. Seetapan et al. / Journal of Food Engineering 119 (2013) 446–453 Table 1 Pasting characteristics of starches. Starch Wheat starch (WS) Tapioca starch (TS) H1 (DS = 0.019) H2 (DS = 0.063) H3 (DS = 0.102) Pasting temperature (°C) a Peak viscosity (mPa s) Breakdown (mPa s) d 92.20 (0.05) 69.07b (0.49) 68.97b (0.43) 63.25c (0.43) 60.68d (0.67) d 1102 (24) 2615a (57) 2391c (30) 2481b (33) 2634a (5) 347 (8) 1575a (51) 1262c (23) 1343b (18) 1566a (3) Final viscosity (mPa s) d 1405 (58) 1799a (32) 1843a (30) 1717b (27) 1563c (1) Setback (mPa s) 650b (61) 759a (24) 713a (26) 579c (11) 495d (8) abcd Means values followed by the same superscript letters within the same column are not significantly different (P P 0.05). Values in parentheses are the standard deviations. Table 2 Critical strain (cc), G0 and tan d determined at the LVE region from viscoelastic experiment of doughs. Sample* cc (%) G0 (Pa) tan d WS-TS (control) WS-H1 WS-H2 WS-H3 3.68a (0.00) 2.66b (0.00) 0.99c (0.00) 0.29d (0.00) 18,000a (300) 11,700b (700) 10,200c (200) 8900c (170) 0.11c (0.00) 0.17b (0.00) 0.18b (0.00) 0.20a (0.00) Wheat starch (WS):tapioca starch (TS) or modified starch (H) = 86:14 w/w. Values in parentheses are the standard deviations. abcd Means values followed by the same superscript letters within the same column are not significantly different (P P 0.05). * Starches:water:oil = 41:56:3 w/w/w. indentation technique. The firmness was given in terms of the indentation modulus as defined in Eq. (1). The moduli with respect to storage time are shown in Table 3. The modulus obtained after 1 h storage was referred to as the modulus at day 0. It was clearly seen that the modulus of all gel sheets was found to significantly increase during 3 days of storage (P < 0.05). Particularly, the gel modulus of the control formulation became much firmer with increasing storage time (P < 0.05). Moduli of all gel sheets were linearly correlated with storage time (r2 > 0.99, fitting curves were not shown). Table 3 also shows the change in gel moduli of the H-substituted gel sheets. In accordance with the viscoelastic results, gel sheets from the substituted modified starch formulations had lower modulus, i.e. with softer texture. Moreover, the higher the DS of the substituted modified starch, the lower the gel modulus was. Upon chilled storage, the gel moduli of all modified formulations were significantly increased, but at a slower rate than the control. The least change in the gel modulus upon the storage was observed in the system incorporated with H3, i.e. having high degree of substitution. This finding also corresponded well with the lowest value of setback (see RVA results in Table 1). Such results indicated that the substitution of native tapioca starch with the hydroxypropylated tapioca starch could potentially minimize the increase of gel modulus during a short term storage at 4 °C by restricting starch retrogradation process. Our observed data on the firmer texture were similar to the previous result of Keetels and co-workers (1996). They reported the increased firmness of 30% w/w potato starch gel during 7 °C storage in relation to the increase of melting enthalpy of retrograded samples, which was caused mainly from amylopectin recrystallization. Moreover, Zhou et al. (2010) reported that when water content was less than 60%, majority of recrystallization process was the association of amylopectin chains, while the amyloseamylopectin co-crystallization was believed to be less significant because of the restricted chain mobility. From the formulation point of view, our studied semi-solid systems comprised a blend of 41% starch, 56% water and 3% oil. With such level of water content, it was hypothesized that retrogradation, hence the texture change, was likely dominated by amylopectin recrystallization. Such hypothesis was proved using XRD and DSC analyses which will be discussed in the next section. 3.2.3. Wide-angle X-ray scattering of retrograded starch powders Recrystallization of starch molecules were investigated by wide-angle X-ray scattering experiments. In this study, the surface of gel sheets after storage was not sufficiently smooth to enable the measurement. Therefore, the specimens for XRD experiments were prepared in powder form (as described in Section 2.6). Fig. 2a and b show XRD patterns of wheat starch and tapioca starch raw materials, respectively. Both starch samples demonstrated the A-type crystalline pattern with peaks at 2h  15°, 17°, 18° and 23° (Cheetham and Tao, 1998; Liu, 2005). However, after cooking into the studied gel sheets, distinct peaks at 2h  12° and 20° appeared (Fig. 2c), corresponding to the V-type crystalline pattern which is often found in amylose–lipid complexes (Hibi et al., 1990; Lionetto et al., 2005). XRD patterns of the gel sheets stored at 4 °C however showed a major peak at 2h  17° (Fig. 2d–f) which corresponds to a B-type crystalline structure, i.e. a characteristic of recrystallized amylopectin (Farhat et al., 2000; Park et al., 2009). With increasing storage time, peak intensity at 2h  17° increased, indicating an enhanced recrystalline part, while the helical complexes of amylose and lipid at 2h  20° weakened (Hibi et al., 1990; Wu and Sarko, 1978). The modified gel sheets demonstrated the same XRD patterns as the control (Fig. 3), and intensity of the peak Table 3 Modulus of cooked gel sheets during storage time and temperature. Sample* WS-TS (control) WS-H1 WS-H2 WS-H3 Modulus (kPa) Day 0/25 °C Day 1/4 °C Day 2/4 °C Day 3/4 °C 4.17a,D (0.20) 3.36b,D (0.27) 2.74c,D (0.33) 2.63c,D (0.23) 19.00a,C (2.71) 10.03b,C (1.39) 6.28c,C (0.96) 5.07d,C (0.50) 28.16a,B (3.62) 16.94b,B (1.41) 8.89c,B (1.38) 7.78c,B (1.13) 43.66a,A (4.30) 22.64b,A (2.58) 11.88c,A (1.57) 9.93d,A (2.16) Wheat starch (WS):tapioca starch (TS) or modified starch (H) = 86:14 w/w. abcd Means values followed by the same superscript letters within the same column are not significantly different (P P 0.05). ABCD means values followed by the same superscript letters within the same row are not significantly different (P P 0.05). Values in parentheses are the standard deviations. * Starches:water:oil = 41:56:3 w/w/w. 450 N. Seetapan et al. / Journal of Food Engineering 119 (2013) 446–453 pattern, it was found that there was no significant difference in degree of structure recrystallization among formulations. WS-TS (f) (e) (d) (c) (b) (a) Fig. 2. X-ray diffraction patterns of wheat starch and native tapioca starch (a and b), the fresh gel sheet (day 0/25 °C) (c), and the retrograded gel sheet under 4 °C storage from 1 to 3 days (d–f, respectively) of WS-TS (control). corresponding to amylopectin recrystallization (2h  17°) increased with storage time as well. An increase of the 17° peak intensity, associated to degree of recrystallinity, was consistent with an increase in the gel modulus detected upon storage. However, when considering the degree of amylopectin recrystallization from the ratio of area under the 17° peak and total area of XRD WS-H1 3.2.4. Differential scanning calorimetry analysis of retrograded starch powders DSC was used to detect retrogradation behavior of gels stored at 4 °C by monitoring the melting endothermic peaks of recrystallized starch molecules. Results were summarized in Table 4. There was no melting peak of amylopectin (at 50–80 °C) observed in the freshly-prepared gel sheets which indicated that all gel sheets completely gelatinized under cooking process. In contrast, upon 4 °C storage the melting endotherm between 40 and 70 °C was detected similarly to the previous observation (Yu et al., 2009). It was also observed that there were no significant differences in To, Tp and Tc over storage time of most samples. Furthermore, the values of Tc among different formulations at the same storage time were not significantly different. However, To and Tp were found to be affected by different dough formulations. In particular, the formulation with highest DS hydroxypropylated starch (WS-H3) tended to have high values of To and Tp. Fast retrogradation during 1-day storage was detected as a large increase in retrogradation melting enthalpy (DHr) as a comparison to Day 0 (DHr = 0 J/g). From Day 1 to Day 2, DHr increased significantly (P < 0.05); however, further storage induced only small increase in DHr. This observed result was in agreement with previous investigation of Lionetto et al. (2005) where fast retrogradation of wheat starch gel was detected within the first 15 h of storage at 25 °C. Unfortunately, the melting of amylose–lipid complex generally appeared at temperature P100 °C (Zhang et al., 2012) cannot be WS-H2 (d) (c) (b) (a) (d) (c) (b) (a) WS-H3 (d) (c) (b) (a) Fig. 3. X-ray diffraction patterns of the fresh gel sheet (day 0/25 °C) (a), and the retrograded gel sheet under 4 °C storage from 1 to 3 days (b–d, respectively) of WS-H1, WSH2 and WS-H3 cooked gels. N/D N/D N/D 2.98a,A (0.04) 2.42b,AB (0.25) 2.43b,A (0.16) 2.21b,A (0.26) 1.94ns,B (0.02) 1.83B (0.26) 1.74B (0.14) 1.49B (0.14) N/D 61.03ns (0.06) 61.40 (0.57) 60.97 (0.75) 60.40 (0.30) 60.73ns (0.64) 60.23 (0.61) 60.93 (0.70) 61.13 (0.41) 3.3. Gel sheet incorporated with mixed modified starches and xanthan gum N/D N/D N/D N/D N/D N/D WS-H2 WS-H3 N/D = non-detectable. abcd means values followed by the same superscript letters within the same column are not significantly different (P P 0.05). ABCD means values followed by the same superscript letters within the same row are not significantly different (P P 0.05). Values in parentheses are the standard deviations. * Starches:water:oil = 41:56:3 w/w/w, wheat starch (WS):tapioca starch (TS) or modified starch (H) = 86:14 w/w. N/D N/D N/D 61.83ns,NS (0.47) 60.60NS (1.01) 60.70NS (0.44) 61.63NS (0.67) N/D 51.23c (0.21) 51.96b (0.06) 52.50a,A (0.17) 52.11b,B (0.17) 51.37c (0.40) 51.37bc (0.34) 52.44a,A (0.10) 51.94ab,B (0.10) 51.53b,NS (0.42) 51.97ab,NS (0.38) 51.59b,B (0.14) 52.44a,A (0.20) N/D 41.57ns (0.40) 43.10 (1.85) 43.27 (0.46) 43.63 (0.65) WS-TS (control) WS-H1 N/D 42.40b,NS (0.53) 42.70b,NS (0.66) 42.63b,NS (0.57) 44.83a,NS (0.38) 41.73c (0.64) 42.95bc (1.34) 43.60ab (0.72) 44.57a (0.31) Day 0 25 °C Day 2 4 °C Day 1 4 °C 451 detected in this study since poor baselines were obtained as a result of water boiling. The results of XRD and DSC confirmed that amylopectin recrystallization was responsible for an increase of gel modulus during 4 °C storage. However, no differences were observed between degree of amylopectin recrystallization of the different formulations since the recrystallization peaks were dominantly from wheat starch. The determination of gel modulus appeared to be the most suitable of the techniques used to monitoring retrogradation of the studied starch gels. Therefore, it could be said that the substitution of tapioca starch with modified starch, especially H2 and H3, in the formulation could delay the texture change upon storage. 3.04ns,A (0.12) 2.99A (0.33) 2.56A (0.43) 2.45A (0.29) Day 2 4 °C Day 1 4 °C DHr (J/g) Day 0 25 °C Day 0 25 °C Tp (°C) Day 0 25 °C Day 3 4 °C To (°C) Amylopectin retrogradation Sample* Table 4 Thermal properties of cooked starch powders during storage. Day 1 4 °C Day 2 4 °C Day 3 4 °C Tc (°C) Day 1 4 °C Day 2 4 °C Day 3 4 °C Day 3 4 °C N. Seetapan et al. / Journal of Food Engineering 119 (2013) 446–453 The aforementioned results showed that by replacing tapioca starch with hydroxypropylated starch of high DS, especially for the WS-H2 and WS-H3 formulations, provided gel sheet with high texture stability upon 4 °C storage, but their dough viscoelasticity and their fresh gel modulus were relatively low that their freshly cooked wrappers were incapable of holding the filling weight. In order to strengthen the dough and the freshly cooked gel, HX (hydroxypropylated-crosslinked tapioca starch with the similar DS to H2) was introduced as another additive into the dough formulation of WS-H2 since HX had lower breakdown (19 ± 4 mPa s) and higher final viscosity (3152 ± 27 mPa s). Therefore, a mix of WS-H2 with HX in the preparation could provide strength to the dough and cooked gel. The proportion between H2 and HX was varied while the content of wheat starch was kept constant as in the control formulation. Table 5 shows the viscoelastic results of doughs prepared from the mixture of wheat starch, H2 and HX (WS-H2-HX) along with the doughs from wheat starch incorporated with only one type of modified starch (WS-H2 and WS-HX). It was shown that WS-HX dough was distinctively stronger or more elastic (i.e., with larger cc and G0 , and lower tan d) than WS-H2 due to the crosslink structure in the HX granules. For WS-H2-HX doughs, cc and G0 tended to decrease, whereas tan d increased with increasing content of H2. Therefore, dough with high content of H2 was soft and much susceptible to flow. Moduli of the fresh and the stored gels were presented in Table 6. The freshly cooked WS-HX gel was the strongest (the largest modulus, 5.12 kPa); and modulus of the fresh WS-H2 gel was lowest (2.74 kPa). Upon storage, all of the stored gel moduli continuously increased. In case of WS-H2-HX doughs, it was found that the presence of HX at 10% of total starch weight helped to significantly strengthen the texture of WS-H2HX gel after cooking. However, the increased firmness was significantly observed (P < 0.05) in all dough formulations and was the most pronounced in the system with the highest HX content. Among three WS-H2-HX formulations, WS-H2-HX (86:4:10) gave a dough viscoelastic property (Table 5) and provided a fresh gel modulus (Table 6) similar to those of the control (see Tables 2 and 3), but the WS-H2-HX (86:4:10) gel became the firmest during storage. Hence, xanthan gum (XG) was added in the dough preparation of WS-H2-HX (86:4:10) in order to enhance water holding capacity of the stored gel, thus retarding the firming of the gel texture. In addition, the presence of XG in the mixed starch formulation resulted in softer dough with lower cc and G0 than the system without XG addition (Table 5). As shown in Table 6, both samples had similar texture, i.e. similar gel moduli (P P 0.05) after cooking. Nevertheless, the gel formulation containing XG showed the slowly increase in the modulus upon chilled storage for 3 days (Table 6). This observation suggested that a combination of modified starches and XG appeared to delay 452 N. Seetapan et al. / Journal of Food Engineering 119 (2013) 446–453 Table 5 Critical strain (cc), G0 and tan d determined at the LVE region from viscoelastic experiment of doughs (starches:water:oil = 41:56:3 w/w/w). Sample WS-H2 WS-H2-HX WS-HX WS-H2-HX-XG Compositions cc (%) G0 (Pa) tan d 0.99f (0.00) 1.37e (0.00) 1.91d (0.00) 3.68b (0.00) 5.12a (0.00) 2.65c (0.00) 10,200d (200) 10,900d (310) 13,200c (240) 18,200a (670) 18,600a (550) 16,000b (590) 0.18b (0.00) 0.19a (0.00) 0.16c (0.00) 0.12d (0.00) 0.09e (0.00) 0.12d (0.00) * WS:H2:HX (w/w/w) XG 86:14:0 86:10:4 86:7:7 86:4:10 86:0:14 86:4:10 – – – – 0.07 Values in parentheses are the standard deviations. abcd Means values followed by the same superscript letters within the same column are not significantly different (P P 0.05). * XG content was represented as the percentage by starch weight. Table 6 Modulus of gel sheets during storage time and temperature Sample WS-H2 WS-H2-HX WS-HX WS-H2-HX-XG Compositions Modulus (kPa) WS:H2:HX (w/w/w) XG* Day 0/25 °C Day 1/4 °C Day 2/4 °C Day 3/4 °C 86:14:0 86:10:4 86:7:7 86:4:10 86:0:14 86:4:10 – – – – – 0.07 2.74d,D (0.33) 3.17c,C (0.18) 2.95c,d,D (0.68) 4.07b,D (0.50) 5.12a,D (0.35) 4.47b,D (0.13) 6.28d,C (0.96) 10.78b,B (1.21) 7.19d,C (1.20) 11.35b,C (1.76) 16.53a,C (0.69) 8.87c,C (0.84) 8.89d,B (1.38) 11.03c,B (2.18) 8.37d,B (0.91) 14.14b,B (1.35) 22.52a,B (3.05) 10.68c,B (2.07) 11.88d,A (1.57) 13.84c,A (2.57) 14.14c,A (2.62) 20.84b,A (4.33) 29.11a,A (1.34) 14.34c,A (1.42) abcd means values followed by the same superscript letters within the same column are not significantly different (P P 0.05). ABCD means values followed by the same superscript letters within the same row are not significantly different (P P 0.05). Values in parentheses are the standard deviations. * XG content was represented as the percentage by starch weight. the firmer texture of gel much more effectively than the formulation without XG. The improved low-temperature storage stability of XG-starch gel was extensively reported in literatures (Arocas et al., 2009; Muadklay and Charoenrein, 2008; Sae-kang and Suphantharika, 2006). For the studied model systems, less textural changes under low temperature storage might be a result of the retarded retrogradation process in the WS-H2-HX-XG gel which could have arisen from the enhanced water retention capacity of XG addition. 4. Conclusions This study investigated texture stability under 4 °C storage up to 3 days and dough viscoelasticity of a model starch gel, i.e. the wrapper sheets for a Chinese shrimp dumpling (Har Gow). The studied model was semi-solid consisting of 41% starch, 56% water and 3% oil. Starches in the control formulation were a combination of 86% wheat starch and 14% tapioca starch. Upon 4 °C storage, modulus of the control gel sheet was progressively increased which was believed to be mainly contributed by amylopectin retrogradation. This was confirmed by DSC and XRD analyses. By replacing tapioca starch in the control formulation with a mixture of modified starches (i.e., hydroxypropylated starch and hydroxypropylated-crosslinked starch) and xanthan gum (0.07% of total starch weight), it was found that its dough viscoelasticity was similar to that of the control. More importantly, the firmer texture of this modified gel formulation was considerably delayed upon the chilled storage. Acknowledgements This study was granted by the National Metal and Materials Technology Center (MT-B-53-POL-07-483-I). Authors are grateful to Tapioca Development Corp. Ltd. for the provision of modified tapioca starches. References Arocas, A., Sanz, T., Fiszman, S.M., 2009. Improving effect of xanthan and locust bean gums on the freeze-thaw stability of white sauces made with different native starches. Food Hydrocolloids 23, 2478–2484. Beak, M.H., Yoo, B., Lim, S.-T., 2004. Effects of sugars and sugar alcohols on thermal transition and cold stability of corn starch gel. Food Hydrocolloids 18, 133–142. BeMiller, J.N., 2011. Pasting, paste, and gel properties of starch-hydrocolloid combinations. Carbohydrate Polymers 86, 386–423. Cheetham, N.W.H., Tao, L., 1998. Variation in crystalline type with amylose content in maize starch granules: an X-ray powder diffraction study. Carbohydrate Polymers 36, 277–284. Farhat, I.A., Blanshard, J.M.V., Mitchell, J.R., 2000. The retrogradation of waxy maize starch extrudates: Effects of storage temperature and water content. Biopolymers 53, 411–422. Fellows, P., 1997. Food Processing Technology-Principles and Practice. Woodhead Publishing Ltd., London. Funami, T., Kataoka, Y., Omoto, T., Goto, Y., Asai, I., Nishinari, K., 2005. Effects of nonionic polysaccharides on the gelatinization and retrogradation behavior of wheat starch. Food Hydrocolloids 19, 1–13. Gamonpilas, C., Charalambides, M.N., Williams, J.G., Dooling, P.J., Gibbons, S.R., 2010. On the characterization of the mechanical behaviour of starch gels using indentation techniques. Applied Rheology 20, 33283–33291. Hibi, Y., Kitamura, S., Kuge, T., 1990. Effect of lipids on the retrogradation of cooked rice. Cereal Chemistry 67, 7–10. Keetels, C.J.A.M., Oostergetel, G.T., van Vliet, T., 1996. Recrystallization of amylopectin in concentrated starch gels. Carbohydrate Polymers 30, 61–64. Kindt, M., Lercker, G., Mazzaracchio, P., Barbiroli, G., 2006. Effects of lipids on the quality of commercial frozen ready pasta meals. Food Control 17, 847–855. Lee, M.H., Baek, M.H., Cha, D.S., Park, H.J., Lim, S.T., 2002. Freeze-thaw stabilization of sweet potato starch gel by polysaccharide gums. Food Hydrocolloids 16, 345– 352. Lionetto, F., Maffezzoli, A., Ottenhof, M.-A., Farhat, I.A., Mitchell, J.R., 2005. The retrogradation of concentrated wheat starch systems. Starch/Stärke 57, 16–24. Liu, Q., 2005. Understanding starches and their role in foods. In: Cui, S.W. (Ed.), Food Carbohydrates: Chemistry, Physical Properties, and Applications. Taylor & Francis Group, LLC., Boca Raton, pp. 309–355. Morikawa, K., Nishinari, K., 2000. Effects of concentration dependence of retrogradation behaviour of dispersions for native and chemically modified potato starch. Food Hydrocolloids 14, 395–401. N. Seetapan et al. / Journal of Food Engineering 119 (2013) 446–453 Muadklay, J., Charoenrein, S., 2008. Effects of hydrocolloids and freezing rates on freeze-thaw stability of tapioca starch gels. Food Hydrocolloids 22, 1268–1272. Park, E.Y., Baik, B., Lim, S., 2009. Influences of temperature-cycled storage on retrogradation and in vitro digestibility of waxy maize starch gel. Journal of Cereal Science 50, 43–48. Pongsawatmanit, R., Srijunthongsiri, S., 2008. Influence of xanthan gum on rheological properties and freeze-thaw stability of tapioca starch. Journal of Food Engineering 88, 137–143. Pongsawatmanit, R., Chantaro, P., Nishinari, K., 2013. Thermal and rheological properties of tapioca starch gels with and without xanthan gum under cold storage. Journal of Food Engineering 117, 333–341. Raina, C.S., Singh, S., Bawa, A.S., Saxena, D.C., 2007. A comparative study of Indian rice starches using different modification model solutions. LWT 40, 885–892. Sae-kang, V., Suphantharika, M., 2006. Influence of pH and xanthan gum addition on freeze-thaw stability of tapioca starch pastes. Carbohydrate Polymers 65, 371– 380. 453 Singh, J., Kaur, L., McCarthy, O.J., 2007. Factors influencing the physic-chemical, morphological, thermal and rheological properties of some chemically modified starches for food applications-A review. Food Hydrocolloids 21, 1–22. Wu, H.C.H., Sarko, A., 1978. The double helical molecular structure of amylose. Carbohydrate Research 61, 27–31. Yu, S., Ma, Y., Sun, D.-W., 2009. Impact of amylose content on starch retrogradation and texture of cooked milled rice during storage. Journal of Cereal Science 50, 139–144. Zhou, X., Baik, B.-K., Wang, R., Lim, S.-T., 2010. Retrogradation of waxy and normal corn starch gels by temperature cycling. Journal of Cereal Science 51, 57–65. Zhou, X., Wang, R., Yoo, S.-H., Lim, S.-T., 2011. Water effect on the interaction between amylose and amylopectin during retrogradation. Carbohydrate Polymers 86, 1671–1674. Zhang, B., Huang, Q., Luo, F.-X., Fu, X., 2012. Structural characterizations and digestibility of debranched high-amylose maize starch complexed with lauric acid. Food Hydrocolloids 28, 174–181.