The effect of temperature on the physicochemical properties and lamellar structure of canna starch subjected to enzymatic degradation

Abstract

Canna starch was degraded thermally and enzymatically with or without α-amylase (Bacillus subtilis) at increasing temperatures (50–60 °C). Considerable disruption of the starch granules was found in enzymatically degraded canna starch (EDCS) compared to native canna starch (NCS), while only some swollen starch granules were observed in thermally degraded canna starch (TDCS) even at the critical gelatinization temperature (60 °C). With respect to physicochemical properties, both thermal and enzymatic degradation reduced the swelling power (SP) due to the internal rearrangements within the starch granules. Increased temperature narrows the gelatinization range of canna starch subjected to thermal and enzymatic degradation, however these treatments have different effects on the enthalpy. Small angle X-ray scattering (SAXS) investigation reveals that the lamellar order becomes less pronounced with enhanced temperature, both in TDCS and EDCS, however, a reduced ΔL from the correlation function indicates that the lamellar thickness distribution of TDCS and EDCS are narrower than their native counterpart.

---

1. Introduction

Starch is composed of amylose and amylopectin joined through α-D-(1 → 4) and α-D-(1 → 6) glycosidic bonds. Amylose is an almost linear polymer composed of α-(1,4)-glucose chains about 1000 residues long, and is water accessible.¹ Amylopectin is more highly branched, consisting of linear α-(1,4)-glucose chains with α-(1,6)-glycosidic bonds at branch points. This polymer is the major component of densely packed crystalline regions within the starch macromolecule and is generally accepted as being water insoluble. Starch granules consist of amorphous and ordered areas, with the ordered areas formed from clusters of short amylopectin chains. Starch samples show different physicochemical properties because they contain different proportions of amorphous and ordered regions within the large starch granules. Degradation and separation of starch into amylose and amylopectin usually occurs during food processing and storage,^2,3 which affects the food quality.

The behavior of aqueous starch solutions is of great interest to scientists because many industrial applications involve the heating of foods containing starch in water. Thermal and enzymatic degradation are two of important reactions that are encountered industrially. The thermal degradation process is carried out at a high moisture level (>60%, w/w), and is normally conducted with wider temperature ranges than annealing and at lower temperature than thermal decomposition.⁴ In contrast, enzymatic degradation, via α-amylase, is a process that improves the crystalline perfection of starch by preferentially eroding amorphous regions over ordered crystallites.^5,6 Both thermal and enzymatic degradation can impart important characteristics starches such as the waxy appearance achieved by decreasing the amount of soluble amylose.⁷ Therefore, these two degradation pathways occur in the industrial production of syrup,⁸ fat mimetics,⁹ and also during the baking of products containing starch. Different food varieties have different requirements, a high degree of conversion to fermentable sugars is needed in the production of syrup, while in production of fat mimetics and shortbread biscuits, it is important to control the degradation of starches to get the desired texture of the final product. The degree of degradation depends on many factors, such as temperature, starch concentration, origin of the α-amylase enzyme, and degradation time. Amongst these factors, temperature is one of the most important parameters and affects reaction rates and starch characteristics.

Characterization of α-amylase action on commercial starch has been the subject of numerous investigations and reports,^10–12 however there is a dearth of information on canna starch, which is an underutilized starch source.^13,14 Cannas, a crop that can adapt successfully to a wide range of habitats, including marginal regions, are nutrient-rich and contain over 66% starch in their rhizomes (dry weight). Canna starch granules are spherical or oval with smooth surfaces and they contain a large amount of amylose.¹⁵ The aim of this study was to investigate the effect of temperature on the physicochemical properties and internal structural changes that occur during both thermal and enzymatic degradation of canna starch.

2. Materials and methods

2.1 Materials

The NCS (30.9% amylose, 6.1% water on a dry basis) was obtained from Ziyun Jiahe Chemical Co. Ltd. (Guizhou, China). The α-amylase (3700 U g⁻¹) from Bacillus subtilis was purchased from Beijing Aoboxing Bio-tech Co. Ltd. (China). Doubly distilled water was used in all experiments. Other reagents were of analytical grade and were purchased from China National Medicine Group (China).

2.2 Methods

2.2.1 Preparation of thermally degraded canna starch (TDCS) and enzymatically degraded canna starch (EDCS). The starch slurry (10%, w/v) was prepared in phosphate buffer (pH 6.9) with α-amylase added (15 U g⁻¹ starch). Then the samples were incubated in a water bath with continuous stirring at different temperatures (50, 55, and 60 °C) for three hours. Samples of EDCS were isolated by centrifugation (8000 × g, 20 min) and washed twice with doubly distilled water and then lyophilized (the water content of the final degraded starch was 5.8% on a dry basis). The supernatants were kept together for a subsequent dextrose equivalent (DE) determination. As a control, TDCS was prepared as above without addition of α-amylase. Starches treated at 50, 55, and 60 °C were denoted with treating method and respective treating temperature (e.g. 50-TDCS indicates thermal degradation at 50 °C, similarly, 50-EDCS indicates enzymatic degradation at 50 °C).

2.2.2 Surface morphology. The morphological attributes of the starch dispersions were assessed by a polarized light microscopy (Nikon Eclipse E200, Japan) equipped with a 50× objective lens.

2.2.3 Swelling power (SP) and water solubility index (WSI). The SP and WSI were determined following the method of Schoch (1964),¹⁶ and were calculated as follows:

2.2.4 Amylose leaching (AML). The value of AML was determined according to a previous literature,¹⁷ the amylose content was determined using the dual-wavelength iodine binding technique,¹⁸ and the total amylose is the sum of soluble and insoluble amylose. These three indexes were calculated using the following equations:

2.2.5 Differential scanning calorimetry (DSC) analysis. Starch was dispersed in distilled water (1 : 2; w/v) in an aluminum pan, which was then hermetically sealed and equilibrated for 24 h at room temperature before testing. The NETZSCH Phoenix DSC (NETZSCH 204 F1) was calibrated for temperature and for enthalpy measurements.

2.2.6 X-ray diffraction (XRD) analysis. XRD experiments were performed with a Bruker-AXS D8 ADVANCE powder diffractometer using Cu-Kα radiation (wavelength λ = 1.54 nm), operating at 40 kV and 40 mA. Data were collected by step-scanning at 0.02° intervals over the 2θ range of 3–40°. Degree of crystallinity was estimated using the nonlinear peak fitting method.¹⁹

2.2.7 Small-angle X-ray scattering (SAXS) analysis. Samples for SAXS measurements were prepared as aqueous starch pastes (∼45% (w/v) starch), centrifuging at 8000 × g for 10 min after an overnight equilibration. The SAXS experiments were conducted on beamline BL16B1 of the Shanghai Synchrotron Radiation Facility. An incident wavelength of 1.24 nm was used, and the sample-to-detector distance was set to 2150 mm. Scattering was detected in the q ranges of 0.06–1.88 nm⁻¹ (where q = (4π sin θ)/λ, the scattering vector). The isotropic scattering patterns were radially averaged, empty scattering was subtracted, and the resulting SAXS intensity was analyzed as a function of the scattering vector q. Data processing was performed using the FIT2D software package.

2.2.8 Analysis of SAXS data. The obtained SAXS data were analyzed via the one dimensional correlation function γ₁(x) (eqn (6)) to obtain the lamellar order. The correlation function was calculated using a cosine transformation of the scattering intensity, I(q),²⁰ where x represents the distance in real space, and denominator is the scattering invariant. The method used to determine the various structural parameters according to γ₁(x) is shown in the inset of Fig. 4.

3. Results and discussion

3.1 Morphological changes

The morphological changes in canna starch upon thermal and enzymatic degradation are shown in Fig. 1 and 2. In the TDCS samples, no significant changes were found between samples treated at different temperatures, with the exception of some swollen starch granules found in 60-TDCS. Swollen starch granules were found at the critical gelatinization temperature, demonstrating that at 60 °C canna starch had already reached the onset of gelatinization, which was confirmed by the DSC data (shown in Table 2). Therefore, this process could not be considered annealing which happened within the gelatinization temperature.²¹ In the samples of EDCS, a considerable disruption of canna starch granules (cracks) was found. With the increasing temperature, the number of cracks increased; the highest degree of damage was observed with 60-EDCS. Fortunately, most of the starch granules retained their oval shape and Maltese cross characteristics. These results further confirmed that the process of annealing and enzymatic degradation was rather mild.

Fig. 1 Light micrographs (left) and polarized light micrographs (right) of canna starch samples upon enzymatic degradation (A: 50-EDCS; B: 55-EDCS; C: 60-EDCS).

Fig. 2 Light micrographs (left) and polarized light micrographs (right) of canna starch samples upon thermal degradation (A: 50-TDCS; B: 55-TDCS; C: 60-TDCS).

3.2 Physicochemical changes

The physicochemical properties of the TDCS and EDCS samples are presented in Table 1. The highest SP (20.83 g g⁻¹) was observed in the sample of NCS, demonstrating that both processes, thermal and enzymatic degradation, reduced the SP of canna starch. The reduction in SP following thermal and enzymatic degradation has been attributed to internal rearrangement within the starch granules. These processes facilitate further interaction among the starch bonding functional groups, causing it to form more ordered double helical amylopectin side chain clusters. Our results and analysis agree with previous results.^15,21 In addition, we conclude that both degradation processes reduced the hydrophobicity of native starch by leaching amylose, a result that has been reported by others.^22–24 In contrast to the reduced WSI in TDCS, all the EDCS samples exhibit an increased WSI with increasing temperature. The increased WSI upon enzymatic degradation might be attributed to some other factors (e.g. breakdown of amylopectin) overwhelming the leaching of amylose.

Table 1 Physicochemical properties of TDCS and EDCS

Starch	DE	WSI (g/100 g)	SP (g g^−1)	AML (g/100 g)	Soluble amylose (g/100 g)	Insoluble amylose (g/100 g)	Total amylose
NCS	0	17.83 ± 0.00	20.83 ± 0.24	36.98 ± 0.43	10.91 ± 0.23	19.97 ± 0.05	30.88 ± 0.17
50-TDCS	0.08	16.39 ± 0.98	13.96 ± 0.29	26.48 ± 0.52	4.77 ± 0.09	22.61 ± 0.44	27.38 ± 0.35
55-TDCS	0.32	15.48 ± 0.28	9.99 ± 0.10	39.44 ± 0.29	7.10 ± 0.05	24.59 ± 0.07	31.69 ± 0.12
60-TDCS	0.42	12.58 ± 0.63	10.87 ± 0.57	25.80 ± 0.32	4.01 ± 0.22	23.24 ± 0.86	27.25 ± 0.64
50-EDCS	23.38	16.88 ± 1.29	17.51 ± 1.99	1.24 ± 0.26	0.23 ± 0.05	8.09 ± 0.08	8.31 ± 0.13
55-EDCS	23.48	39.78 ± 1.34	18.08 ± 0.52	0	0	14.93 ± 2.70	14.93 ± 2.70
60-EDCS	31.35	40.59 ± 3.40	9.93 ± 0.47	0	0	3.37 ± 0.34	3.37 ± 0.34

---

All of the above changes in physicochemical properties can be explained in terms of changes in amylose. Amylose can be classified into soluble and insoluble amylose, and these can be separated using centrifugation. Soluble amylose can be quantified using the amylose in the supernatant, while insoluble amylose can be quantified from the amylose retained in the sediment. In the samples of TDCS, soluble amylose decreased and insoluble amylose increased, which gave further support to our previous hypothesis that soluble amylose leaches from the starch and part of the leachate becomes insoluble during thermal degradation. Moreover, heat preferentially degrades soluble amylose over insoluble amylose. This finally leads to a reduction in WSI, SP and AML. However, in EDCS, high WSI coupled to negligible AML and soluble amylose was observed. This demonstration demonstrates that amylopectin breakdown overwhelms amylose leaching; it is the amylopectin that is responsible for the high WSI rather than amylose.²⁵ In addition, amylopectin breakdown yields short chains less than 10 to 20 glucose units or branched materials that cannot bind iodine, as it requires amylose chains of 10 to 20 glucose units to achieve a good blue color.²⁶

3.3 Thermal property changes

Besides changes in physicochemical properties, thermal and enzymatic degradation also significantly altered the thermal properties of canna starch. The gelatinization transition temperatures and enthalpies of TDCS and EDCS are shown in Table 2. Higher temperatures narrow the gelatinization range and increase gelatinization temperatures during both thermal and enzymatic degradation. This indicates that there was a strengthening of the bonds requiring higher temperatures for the gelatinization of the granules. However, compared to NCS, thermal degradation decreased the gelatinization temperature. This was contradictory to previous reports²⁷ that annealing (a treatment process similar to our thermal degradation) increased the peak gelatinization temperature of starches. This difference was likely due to the different treatment times used: more than 24 hours of annealing versus 3 hours of thermal degradation in our system. Increases in stability and pasting temperatures were observed when the annealing time was increased,^7,28 and the annealing time had to be more than 6 h to induce this change in the gelatinization properties.²⁹ Thus, the different treatments produced differences in the gelatinization properties of TDCS and annealing starches was observed. In the samples of EDCS, crystallites of canna starch are strengthened (reflected in higher T_o, T_p, and T_c values) and the crystalline homogeneity increased (reflected in lower T_c − T_o values) compared to that of NCS. The results are consistent with our proposed mechanism of soluble-insoluble amylose transformation.²⁵ Insoluble amylose is more stable than soluble amylose, and hence it helped perfect the starch granules.

Table 2 The DSC characteristics of TDCS and EDCS

Starch	TO^a (°C)	TP^a (°C)	TC^a (°C)	TC − TO (°C)	ΔH^b (J g^−1)
a TO, TP, TC: onset, peak and completed temperatures.b Enthalpy of the endothermic peak.
NCS	59.3	63.4	69.4	10.1	10.04
50-TDCS	54.3	60.6	67.8	13.5	9.05
55-TDCS	57.8	65.3	70.2	12.4	5.54
60-TDCS	69.6	73.5	79.1	9.5	2.71
50-EDCS	60.4	66.7	71.3	10.9	7.25
55-EDCS	61.9	65.7	71.9	10.0	10.72
60-EDCS	63.2	67.5	72.5	9.3	10.94

---

Additionally, thermal degradation also significantly decreased ΔH in DSC profiles, a result that is consistent with previous reports.²¹ This suggests that starches treated at higher temperatures are more fragile than those treated with lower temperatures. However, when the temperature was increased for the enzymatic degradation, an increase in ΔH was observed. The enthalpy change during starch gelatinization is related to the melting of crystalline zones. With more crystalline samples, a larger amount of energy is required to melt those crystals. Therefore, enzymatic degradation helped the perfection of the starch granules, and this phenomenon tends to increase with increasing temperature with temperatures between 50–60 °C.

3.4 Internal structural analysis via SAXS pattern

The internal structural changes upon thermal and enzymatic degradation were examined by SAXS. The scattering data for both TDCS and EDCS were plotted as I (intensity) vs. q (scattering vector) in Fig. 3. The well-resolved main scattering peak around the scattering vector of 0.6 nm⁻¹ is thought to arise from the periodic arrangement of alternating crystalline and amorphous lamellae of amylopectin. This corresponds to the lamellar repeat distance or Bragg spacing³⁰ and was used to quantify the degree of lamellar order in the starches.

Fig. 3 Small-angle X-ray scattering (SAXS) spectra of canna starch samples conducted by thermal and enzymatic degradation.

As shown in Fig. 3, the scattering peaks from TDCS and EDCS become less pronounced with increased temperature. This suggests that both thermal and enzymatic degradation increase the stacking disorder with respect to lamellar order, a result consistent with previous studies.^12,31,32 This result was in contrast to the strengthened gelatinization temperature that was observed with increasing temperature. The structural differences observed in our DSC and SAXS analyzes were probably due to the differences in the short-(DSC) and long-range (SAXS) structure. The more intense scattering in NCS, can be explained by the more pronounced electron density difference between amorphous and crystalline lamellae. Therefore, we can rank the Δρ value (density difference between crystalline and amorphous lamellae) for the samples analyzed as follows: NCS > 50-EDCS > 50-TDCS > 55-EDCS > 60-EDCS > 55-TDCS > 60-TDCS. A number of other studies³³ have also reported an increased Δρ (gained stability) in starches subjected to annealing. However, the decrease in Δρ upon thermal degradation that observed in our system was the first measured, demonstrating that the differences between thermal degradation and annealing. We can attribute this to the formation of a more homogeneous structure within a shorter treating time compared to annealing which performed longer procedures.

3.5 Internal structural analysis via one dimensional correlation functions

The resulting correlation functions of NCS, TDCS and EDCS based on eqn (6) are presented in Fig. 4. The γ₁(x) curves from all starch treatments are rather smooth. Theoretically, the linear correlation within lamellae would not exist when the ordered arrangement is disrupted, therefore a discussion of γ₁(x) is not an applicable method to analyze samples of 60-TDCS. Fig. 4 (inset) shows how the long period (d), crystal thickness (d_c) and amorphous thickness (d_a) are determined. Because the crystallinity of the native starches analyzed was less than 50%,¹² the smaller thickness value was assigned to the crystalline region (d_c). Therefore, amorphous thickness (d_a) can be obtained by subtracting the crystalline thickness from the long period (d_a = d − d_c). As Table 3 indicates, the crystalline lamellae thickness decreases while that of the amorphous lamellae increases in the TDCS and EDCS. The increase in d appears to be due to an increase in d_a, as the increased thickness is greater than its decreased counterpart. Increased thickness means increased lamellar disorder, which is consistent with the less pronounced scattering peaks shown in Fig. 3. Bragg spacing is readily obtained from the SAXS curves via the Bragg equation, with q being the scattering vector at the scattering peak. It is found that the repeat distance calculated from the Bragg equation (D) is always larger than the overall periodicity (d), which has been reported previously.^15,34,35 This is because the overall periodicity (d) is consistent with the Bragg spacing (D) only when the lamellar thickness distribution is narrow. When the lamellar thickness distribution is wide, the Bragg spacing overestimates the overall periodicity. Upon thermal and enzymatic degradation, ΔL (the difference between d and D) of canna starch decreased, possibly due to the homogenization of starch structures at nanometer level.

Fig. 4 Correlation function calculated from SAXS profiles in Fig. 3. The inset demonstrates how the values were determined.

Table 3 The feature dimensions of the lamellar structures of starch samples conducted by enzymatic degradation

Starch	φ^a	d^b (nm)	dc^c (nm)	da^d (nm)	Dbragg^e (nm)	ΔL^f (nm)	Crystallinity (%)
a Inner crystallinity in the semi-crystalline stacks of starch.b Long period, i.e. the lamellar repeat distance.c Crystal thickness.d Amorphous thickness.e Repeat distance calculated from Bragg equation.f The difference between Dbragg and d.
NCS	0.19	9.16	1.75	7.45	9.66	0.50	22.18
50-EDCS	0.18	9.21	1.71	7.50	9.67	0.47	27.68
55-EDCS	0.17	9.37	1.62	7.75	9.82	0.45	35.06
60-EDCS	0.16	9.43	1.51	7.92	9.81	0.38	34.65
50-TDCS	0.18	9.25	1.67	7.58	9.67	0.42	24.10
55-TDCS	0.14	9.94	1.41	8.52	9.82	0.08	42.26
60-TDCS	—	—	—	—	—	—	29.40

---

The inner crystallinity of the lamellae (φ), defined by the ratio of the crystal thickness to the overall periodicity, is presented in Table 3. The inner crystallinity was obtained by SAXS and represents the crystalline volume fraction within the lamellae stacks.³⁶ This method can be used when the morphology is paracrystalline or highly disordered.³⁷ We observed that NCS has the highest φ value. Starch subjected to thermal and enzymatic degradation displayed a φ value that decreased sharply with increasing treatment temperature. Also listed in Table 3 is the crystallinity, which was obtained from the XRD pattern. From these data we found that the starch crystalline region from EDCS is perfected with increasing temperature, while that of TDCS has a more complicated situation. Samples of 55-TDCS have the highest crystallinity (42.26%). The increase in crystallinity from 50 to 55 °C can be attributed to the initial hydrolysis of the amorphous region, which resulted in a relatively high degree of crystallinity or reordering of the chain segments. This was driven by the cleavage of amorphous starch chains, in which starch molecules being degraded into small molecular fractions but persisting in a radial arrangement.³⁸ In contrast, the decrease in crystallinity that from temperature of 55 to 60 °C observed was probably due to the initial corruption of starch crystallites.

The relative crystallinity from XRD was calculated using the ratio of the areas of crystalline and amorphous contributions to the diffraction. The ratios of the scattering intensities due to the crystalline and amorphous regions are nearly equal to the ratio of the masses of the two regions. This is because the average electron densities of substances are proportional to the densities of the substances.³⁹ Thus, XRD analysis yields the crystalline weight fraction. The crystalline volume fraction within the lamellae, determined by γ₁(x), did not align with the larger crystalline weight fraction. It is surprising that the highest crystalline volume fraction and lowest crystalline weight fraction were both observed in NCS, demonstrating a lower sample density. It appears that the two degradation procedures helped compact the starch granules resulting in higher densities. The discrepancy between the two different crystalline forms may be an effective way to explore the preferential degradation of the amorphous region, as this region has lower initial density profile.

4 Conclusions

Thermal degradation is similar to enzymatic degradation, and can be considered a control for this process, it is easy to conclude that enzymatic degradation helped the native starch maintain its values (density difference between crystalline and amorphous lamellae). Upon thermal and enzymatic degradation, it is possible to obtain starch with enhanced physicochemical properties and controlled structures. Elevated temperatures decrease the amount of soluble amylose, narrow the gelatinization range, and reduce the crystalline volume fraction and lamellar disordering of canna starch.

Conflicts of interests

The authors declare no competing financial interest and agreed to submit the manuscript.

Abbreviations

(EDCS)	Enzymatic degraded canna starch
(TDCS)	Thermal degraded canna starch
(SP)	Swelling power
(NCS)	Native canna starch
(SAXS)	Small angle X-ray scattering
(WSI)	Water solubility index
(AML)	Amylose leaching
(DSC)	Differential scanning calorimetry
(XRD)	X-ray diffraction

Acknowledgements

The authors are grateful to the National Natural Science Foundation of China (Grant No: 21276154, 31171642, 31571763) for financial support and fellow scientists at the lab for assistance with the manuscript. The SAXS experiments were performed at Synchrotron Radiation Facility, Shanghai, China (BL16B1), and the authors wish to thank the technical staff for their help.

References

1. E. G. White, US Pat., 4,179,527, Washington, DC, 1979.
2. R. F. Tester and S. J. Debon, Int. J. Biol. Macromol., 2000, 27, 1–12.
3. S.-H. Yoo and J.-l. Jane, Carbohydr. Polym., 2002, 49, 307–314.
4. X. Liu, L. Yu, H. Liu, L. Chen and L. Li, Polym. Degrad. Stab., 2008, 93, 260–262.
5. D. Sievert, Z. Czuchajowska and Y. Pomeranz, Cereal Chem., 1991, 68, 86–91.
6. S. H. Imam, S. H. Gordon, A. Mohamed, R. Harry-O'kuru, B. S. Chiou, G. M. Glenn and W. J. Orts, Polym. Degrad. Stab., 2006, 91, 2894–2900.
7. A. M. M. Gomes, C. E. M. da Silva, N. M. P. S. Ricardo, J. M. Sasaki and R. Germani, Starch–Stärke, 2004, 56, 419–423.
8. Y. Shariffa, A. Karim, A. Fazilah and I. Zaidul, Food Hydrocolloids, 2009, 23, 434–440.
9. Y. Ma, C. Cai, J. Wang and D.-W. Sun, J. Food Eng., 2006, 73, 297–303.
10. R. Waduge, R. Hoover, T. Vasanthan, J. Gao and J. Li, Food Res. Int., 2006, 39, 59–77.
11. R. Hoover and Y. Zhou, Carbohydr. Polym., 2003, 54, 401–417.
12. R. F. Tester, X. Qi and J. Karkalas, Anim. Feed Sci. Technol., 2006, 130, 39–54.
13. K. Kasemwong, K. Piyachomkwan, R. Wansuksri and K. Sriroth, Starch–Stärke, 2008, 60, 624–633.
14. F. H. Cisneros, R. Zevillanos and L. Cisneros-Zevallos, J. Agric. Food Chem., 2009, 57, 7363–7368.
15. X. Lan, Y. Li, S. Xie and Z. Wang, Food Chem., 2015, 188, 632–640.
16. T. Schoch, Methods Carbohydr. Chem., 1964, 4, 106–108.
17. Y. Watcharatewinkul, C. Puttanlek, V. Rungsardthong and D. Uttapap, Carbohydr. Polym., 2009, 75, 505–511.
18. Y. Jin, K. Zhang and J. Du, J. Inst. Brew., 2008, 114, 289–293.
19. S. Nara and T. Komiya, Starch–Stärke, 1983, 35, 407–410.
20. B. Goderis, H. Reynaers, R. Scherrenberg, V. B. F. Mathot and M. H. J. Koch, Macromolecules, 2001, 34, 1779–1787.
21. L. Jayakody and R. Hoover, Carbohydr. Polym., 2008, 74, 691–703.
22. K. Adebowale, T. Henle, U. Schwarzenbolz and T. Doert, Food Hydrocolloids, 2009, 23, 1947–1957.
23. R. Cui and C. Oates, Food Chem., 1999, 65, 417–425.
24. O. O. Olayinka, K. O. Adebowale and B. I. Olu-Owolabi, Food Hydrocolloids, 2008, 22, 225–230.
25. X. Lan, S. Xie, F. Xie, X. Liu, J. Wu and Z. Wang, Starch–Stärke, 2015, 67, 1–10.
26. J. A. Putseys, L. J. Derde, L. Lamberts, E. Östman, I. M. BJOrck and J. A. Delcour, J. Agric. Food Chem., 2009, 58, 1939–1945.
27. H.-J. Chung, Q. Liu and R. Hoover, Carbohydr. Polym., 2009, 75, 436–447.
28. A. M. M. Gomes, C. E. Mendes da Silva and N. M. P. S. Ricardo, Carbohydr. Polym., 2005, 60, 1–6.
29. I. Larsson and A. C. Eliasson, Starch–Stärke, 1991, 43, 227–231.
30. T. A. Waigh, A. M. Donald, F. Heidelbach, C. Riekel and M. J. Gidley, Biopolymers, 1999, 49, 91–105.
31. M. L. Foresti, M. del Pilar Williams, R. Martínez-García and A. Vázquez, Carbohydr. Polym., 2014, 102, 80–87.
32. L. Xia, G. Wenyuan, J. Qianqian, W. Yanli, G. Xinhua and H. Luqi, Starch–Stärke, 2011, 63, 616–624.
33. P. J. Jenkins and A. M. Donald, Carbohydr. Res., 1998, 308, 133–147.
34. M. B. Cardoso and H. Westfahl, Carbohydr. Polym., 2010, 81, 21–28.
35. B. Crist, J. Macromol. Sci., Part B: Phys., 2000, 39, 493–518.
36. Y. Tang, Z. Jiang, Y. Men, L. An, H.-F. Enderle, D. Lilge, S. V. Roth, R. Gehrke and J. Rieger, Polymer, 2007, 48, 5125–5132.
37. S. V. Gomand, L. Lamberts, C. J. Gommes, R. G. Visser, J. A. Delcour and B. Goderis, Biomacromolecules, 2012, 13, 1361–1370.
38. Y. Bai, L. Cai, J. Doutch, E. P. Gilbert and Y.-C. Shi, J. Agric. Food Chem., 2014, 62, 4186–4194.
39. S. Manohara, S. Hanagodimath and L. Gerward, Phys. Med. Biol., 2008, 53, N377–N386.

---