Food Hydrocolloids 54 (2016) 139e150

Contents lists available at ScienceDirect

Food Hydrocolloids
journal homepage: www.elsevier.com/locate/foodhyd

Nanostructure, morphology and functionality of cassava starch after

pulsed electric fields assisted acetylation
Jing Hong a, Xin-An Zeng a, *, Roman Buckow b, Zhong Han a, **, Man-sheng Wang a
a
College of Light Industry and Food Sciences, South China University of Technology, Guangzhou 510641, China
b
CSIRO, Food and Nutrition Flagship, Werribee 3030, VIC, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Pulsed electric fields (PEF) assisted acetylation of cassava starch with a higher degree of substitution (DS)
Received 3 July 2015 of 0.131 was obtained in the presence of 10% (w/w) acetic anhydride after PEF treatment at 2 kV/cm for
Received in revised form 51.0 ms. Fourier-transformed infrared spectroscopy showed a characteristic peak of the acetyl group in
22 September 2015
modified starch at 1730 cm
1. The crystalline region of acetylated starch was destroyed and rearranged as
Accepted 24 September 2015
Available online 1 October 2015
confirmed by X-ray diffraction. Small-angle X-ray scattering indicated that the fractal structure was
changed from a surface fractal structure to a mass fractal structure with increasing DS probed. Atomic
force microscopy showed that PEF assisted acetylation of starch markedly increased the surface
Keywords:
Acetylation
roughness. The prominent freeze-thaw stability of PEF-assisted, acetylated starch demonstrated its po-
Cassava starch tential and superior application in frozen food products.
Pulsed electric fields © 2015 Elsevier Ltd. All rights reserved.
Fractal
Nanostructure

1. Introduction introduction of acetyl groups, which interrupt the ordered struc-

Native cassava starch has limited industrial applications due to
its hydrophilic nature of hydroxyl groups, which limits its use in
starch-based products. Thus, various physical, chemical, enzymatic
and genetic techniques have been used to modify the starch mol-
ecules and improve its physiochemical properties to satisfy in-
dustry requirements (Kavlani Neelam and Singh, 2012). Chemical
modification of starch is often carried out to improve functional and
physicochemical parameters of starch including solubility, pasting
properties, thickening power, shear stability or digestibility (Miao
et al., 2014). The modification of starch with acetic anhydride at
low degree of substitution (DS) of 0.01e0.2 is commonly prepared
by esterification in aqueous medium in the presence of an alkaline
catalyst (Wang & Wang, 2002). Acetylated (AC) starch at a low DS
offers desirable properties such as low-temperature stability
(reduced syneresis), high-thickening power, low-gelatinization
temperature, clarity of cooked food, good film-forming properties
and reduced tendency to retrogradation (Chi et al., 2008; Lawal,
2004; Zie˛ ba, Szumny, & Kapelko, 2011). This is mainly due to the
* Corresponding author.
** Corresponding author.
E-mail addresses: xazeng@scut.edu.cn (X.-A. Zeng), fezhonghan@scut.edu.cn
(Z. Han).
ture of native starch and reduce the electrostatic interaction of
amylose and amylopectin chains. In addition, AC starch is an
effective emulsifier due to the addition of dual functional hydro-
philic and hydrophobic groups (Bao, Xing, Phillips, & Corke, 2003).
The transformation of AC starch was actually attributed to the
alternated nanostructure, fractal and lamellae revealed by
combining molecular investigations in nanostructure (SAXS) and
microscopic characterization (AFM). The semi-crystalline shells
were characterized by alternating high- and low-density lamellae
with a repeat distance of 9e10 nm (Vermeylen et al., 2006).
Recently, AFM observations showed that starch esters exhibited an
aggregative nature, with aggregates being comprised of small
spherulites, lamellae of a few microns and some other ill-defined
structures (Zabar, Lesmes, Katz, Shimoni, & Bianco-Peled, 2010;
Zhu, Li, Chen, & Li, 2013).
In recent years, the method of preparation AC starch has
changed from simply exposing starch granules to acetic anhydride
at room temperature (Bhosale & Singhal, 2006) to methods
applying micro-wave (Biswas et al., 2008; Jyothi, Rajasekharan,
Moorthy, & Sreekumar, 2005), enzymes (Huang et al., 2010), high
pressure (Shogren, 2003), high temperature (Kim, Sandhu, Lee, Lim,
& Lim, 2010), ball milling pre-treatment (Zhang, Zhao, & Xiong,
2010) as well as chemical synthesis (Aini & Purwiyatno, 2010)
before or during acetylation in order to achieve higher DS in

http://dx.doi.org/10.1016/j.foodhyd.2015.09.025
0268-005X/© 2015 Elsevier Ltd. All rights reserved.
140 J. Hong et al. / Food Hydrocolloids 54 (2016) 139e150
starches and reduce the use of regents and processing time
(Matsumoto, Tsubaki, Sakamoto, & Azuma, 2011). Pulsed electric
fields (PEF), a non-thermal, low energy processing technique, has
been used to modify properties of starches from different origins
(Han et al., 2012; Han, Zeng, Zhang, & Yu, 2009; Han, Zeng, Yu,
Zhang, & Chen, 2009). It has been shown that the PEF treatment
can enhance starch esterification under appropriate conditions
(Lin, Zeng, Yu, & Sun, 2012; Zeng, Han, & Zi, 2010). Therefore,
acetylation of starch combined with PEF treatment could be a novel
technique to improve the DS at the same acetic anhydride content.
The objective of the present study was to investigate the effect
of different PEF conditions on acetylation of cassava starch.
Furthermore, the crystalline structure and selected physicochem-
ical properties of obtained AC starches will be investigated by
spectroscopic, microscopic, thermal, and physical techniques and
potential applications in industry discussed.
2. Materials and methods
2.1. Materials
Cassava starch was obtained from Guangzhou Mingyu Food
Development Co., Ltd. (Guangzhou, China). Acetic anhydride and
ethanol (100%) were purchased from Beihua Fine Chemicals Co.,
Ltd. (Beijing, China). All chemicals used in the study were of
analytical grade.
2.2. PEF-assisted acetylation
AC starch was prepared using a method with modification
(Sodhi & Singh, 2005). The cassava starch was dried at 45  C for 24 h
to a moistrure content of approximately 8.0% (w/w). Cassava starch
was dispersed in distilled water at a concentration of 35% (w/w).
The starch slurry was agitated at 30  C for 5 min in a 250 mL three-
neck flask fitted with a mechanical stirrer at 300 rpm (IKA, RW 20
digital, Germany) to activate starch. Subsequently, diverse amounts
of acetic anhydride was added drop-wise within 30 min while the
pH was maintained at 8.0e8.5 using 0.75 M NaOH solution. The
electrical conductivity of reaction mixtures were in 9.4e10.6 mS/
cm according to the different addition of 6e10% acetic anhydride.
The electrical conductivity of the suspensions was uniformly
adjusted to 11 mS/cm by adding 2.0 mol/L sodium chloride solution.
Starch samples were then pumped through the PEF system with a
peristaltic pump (Yz1515x, Longer Precision Pump Co. Ltd., Baod-
ing, China). Native and AC starch samples were treated in a bench-
scale, continuous PEF system as illustrated in Fig. 1 (SCUT PEF Team,
the South China University of Technology, Guangzhou, China). The
PEF system features a co-linear treatment chamber with an inner
diameter of the electrodes and insulator of 3 mm and an electrode
gap of 3 mm. PEF treatment was performed at a flow rate of 60 mL/
min applying bi-polar, rectangular pulses of 40 ms at 2000 Hz and
the delay time between the positive and negative pulses was
460 ms. The input voltage of the system was 450e750 V corre-
sponding to electric field strengths of 1.5e2.5 kV/cm. The samples
were cooled to 20  C before being pumped to the PEF chamber for
next cycle. During the whole reaction, pH was sustained at 8.0e8.5.
After the PEF treatment, samples were washed three times with
distilled water and twice with 100% ethanol. The solid was oven
dried at 45  C for 12 h (LinpinIsotemp 202-00 A, Linpin Scientific
Instrument Co., Shanghai, China). After drying, the samples were
passed through a 120-mesh sieve for subsequent use. According to
the conventional method of Ayucitra (2012), samples of 35% starch
concentration, which were added 6%, 8%, 10% acetic anhydride
content and reacted for 60 min, were defined as control samples.
2.3. Experimental design
To optimize the extraction conditions and verify the synergetic
influence of variables, the BoxeBehnken experimental design
(BBD) of the response surface methodology (RSM) by the Design
Expert Version 8.0.5 (Stat-Ease Inc., Minneapolis, USA) was used
with three variables at three levels, respectively (Table 1a,b). The
independent variables and their optimization ranges, such as
strength of PEF (X1, 1.5e2.5 kV/cm), total cycle times in PEF system
(X2, 27e51 ms), and amount of acetic anhydride (X3, 6e10%) were
selected as experimental factors. It is worth noting that the effec-
tive PEF treatment time were 27.2 ms (16 cycles passed through the
PEF system), 39.1 ms (23 cycles passed through the PEF system) and
51.0 ms (30 cycles passed through the PEF system), respectively as
calculated by Eq. (1-3). The specific PEF energy applied to the fluid
during one cycle were 42 kJ/L (1.5 kV/cm), 72 kJ/L (2 kV/cm), 117 kJ/L
(2.5 kV/cm), respectively. The samples were cooled to 20  C before
being pumped to the PEF chamber for next cycle. Thus, the tem-
perature of starch slurry increased from 20  C to 30  C (1.5 kV/cm),
37  C (2 kV/cm), 48  C (2.5 kV/cm), respectively which can make
use of the synergetic effects of mild heat on the PEF treatment ef-
ficiency (Buckow, Schroeder, Berres, Baumann, & Knoerzer, 2010).
PEF treatment time (t) was calculated with the number of pulses
(Np) received in the chambers (Nc ¼ 1), which is obtained from the
residence time in a chamber (tv) as Eq. (1).
tv ¼ V=F (1-1)
where V is the volume of PEF chamber (m3) and F is the flow rate
(m3/s).
NP ¼ tv  f (1-2)
where f is the pulse frequency (Hz).
t ¼ Np  Nc  t (1-3)
where Nc is the number of treatment chambers and t is the pulse
width (s).
The energy input (Q) was calculated by the following equation
used by Zhao et al. (2008):

. 
Q J m3 ¼ E2 st (1-4)
where E is the electric field strength (V/m), s is the electrical con-
ductivity of the product (S/m) calculated for each process temper-
ature, and t is the PEF treatment time (s).
A second order polynomial equation model was used to fit the
yield with interaction terms as given in Eq. (2). And each experi-
ment was replicated at least three times and the DS (%, Yi) was
taken as the responses of the experiments. Sample 0 was the
control of raw cassava starch which was washed and dried as AC
starches.
X X X
Yi ¼ A0 þAi Xi þAij Xi Xj þAii Xi2 ðisjÞ (2)
where Yi are the predicted response values; A0, Ai, Aij and Aii are the
regression coefficients of the model for the intercept, linear cross
product terms and quadratic, respectively; Xi are the variables.
2.4. DS determination
DS value was determined titrimetrically, following the method
outlined by Singh, Chawla, and Singh (2004). AC starch (1.0 g, dry
basis) was placed in a 250 mL flask and 50 mL of 75% (v/v) ethanol
 J. Hong et al. / Food Hydrocolloids 54 (2016) 139e150 141

Fig. 1. Schematic diagram of the continuous PEF system used in this study.

Table 1a 2.5. Fourier transform infrared spectroscopy (FT-IR)

Values of the independent variables and their codes used in response surface
analysis for optimization of PEF-assisted acetylation of cassava starch.
The FT-IR analysis was performed using a Bruker Vertex 70 FTIR
Independent variable Units Variable Code spectrometer (Rheinstetten, Germany). The native, AC and PEF-

1 0 1 assisted AC starch samples were collected using the KBr pellet
method. FT-IR spectra were recorded at a resolution of 4 cm
1 and
PEF strength kV/cm X1 1.5 2 2.5
PEF treatment time ms X2 27.2 39.1 51.0 with a total of 32 scans, and the wave number ranged between 400
Acetic anhydride content % (w/w) X3 6 8 10 and 4000 cm
1. The native and AC starches were equilibrated at

Table 1b
BoxeBehnken design matrix of process variables and predicted values of DS levels of cassava starch.

Standard order Run order X1 kV/cm X2 (ms) X3 (%,w/w) DS (%) Reaction efficiency (%)

Actual value Predicted value

9 1 1
1 0 0.095 ± 0.0017 0.096 74.83 ± 0.34

1 2
1 0
1 0.088 ± 0.0034 0.090 92.43 ± 2.07
15 3 0 0 0 0.099 ± 0.0188 0.098 77.99 ± 1.42
8 4 0 0 0 0.098 ± 0.0022 0.098 77.20 ± 1.73
17 5 0
1 1 0.110 ± 0.0035 0.110 69.32 ± 2.02
14 6
1 1 0 0.099 ± 0.0138 0.097 77.99 ± 0.88
11 7 0 1 1 0.131 ± 0.0017 0.130 82.55 ± 1.70
13 8 1 0
1 0.085 ± 0.0057 0.083 89.28 ± 0.66
6 9
1
1 0 0.086 ± 0.0006 0.084 67.74 ± 0.47
3 10 0 0 0 0.098 ± 0.0040 0.098 77.20 ± 3.14
2 11 0
1
1 0.089 ± 0.0034 0.089 93.48 ± 2.57
4 12
1 0 1 0.110 ± 0.0041 0.110 69.32 ± 2.55
10 13 0 1
1 0.088 ± 0.0026 0.086 92.43 ± 2.73
16 14 1 0 1 0.130 ± 0.0011 0.130 81.92 ± 0.69
12 15 0 0 0 0.093 ± 0.0030 0.098 73.26 ± 2.36
7 16 1 1 0 0.092 ± 0.0023 0.095 72.47 ± 1.81
5 17 0 0 0 0.095 ± 0.0009 0.098 74.83 ± 0.71

X1, X2 and X3 represent PEF strength, PEF treatment time and acetic anhydride content, respectively. Data are expressed as the mean ± standard deviation (SD).

in distilled water was added. The stoppered flask was heated to
50  C for 30 min under agitation, cooled to room temperature
(25  C) and then 40 mL of 0.5 M NaOH was added. The excess alkali
was titrated to pH 7.0 with 0.5 M HCl using phenolphthalein as
indicator. The solution was left to equilibrate for 2 h, and then any
additional alkali, which may have leached from the sample, was
titrated. The original unmodified starch was used as blank sample.
Titration volumes of blank and treated samples were recorded in
mL. The calculation of DS was performed as the research of Singh
et al. (2004) reaction efficiency (RE) was calculated as the report
of Miao et al. (2014).
45  C for 24 h prior to FT-IR analysis.
2.6. X-ray diffraction (XRD)
X-ray diffraction was measured using a D8 ADVANCE X-ray
diffractometer (Siemens, Karlsuhe, Germany) with a target voltage
of 40 kV, current of 40 mA. Samples were tightly packed in the
background less sample holder and scanned with Cu-Ka radiation
(l ¼ 0.15418 nm) in the angular range of 4e50 with step intervals
of 0.04 and a scanning rate of 0.2 s on each step. Crystallinity (%)
was defined as the percentage ratio of diffraction peak area to total
142 J. Hong et al. / Food Hydrocolloids 54 (2016) 139e150

Table 1c
Regression coefficient estimate and significance test for the quadratic polynomial model.

Source Sum of squares df Mean square F-value Probably(P) > F

Model 2.6  10
3 9 2.9  10
4 43.875 <0.0001 significant

X1 4.8  10
5 1 4.8  10
5 7.413 0.0297 *
X2 7.5  10
5 1 7.5  10
5 11.443 0.0117 **
X3 1.9  10
3 1 1.9  10
3 292.954 <0.0001 **
X1  X2 5.8  10
5 1 5.8  10
5 8.868 0.0206 *
X1  X3 1.4  10
4 1 1.4  10
4 22.101 0.0022 **
X2  X3 6.9  10
5 1 6.9  10
5 10.648 0.0138 *
X12 3.0  10
5 1 3.0  10
5 4.651 0.0679
X22 1.6  10
5 1 1.6  10
5 2.525 0.1561
X32 2.4  10
4 1 2.4  10
4 36.151 0.0005 **
Residual 4.6  10
5 7 0.7  10
5
Lack of Fit 3.5  10
5 3 1.2  10
5 4.140 0.1018 not significant
Pure Error 1.1  10
5 4 0.3  10
5
Corrected Total 2.6  10
3 16 43.875 <0.0001
R2 ¼ 0.9826

*P＜0.05.
**P＜0.01.

diffraction area.
2.7. Small angel X-ray scattering (SAXS)
The measurements of small angel X-ray scattering were per-
formed with proper modification using a SAXS camera (Anton-Paar,
Graz, Austria) (Zhu et al., 2013). The system was operated at 50 mA
and 40 kV, using Cu-Ka radiation with a wavelength of
l ¼ 0.1542 nm as X-ray source. Samples were configured to 60% (w/
v) slurry then placed into a capillary of 1 mm diameter and 0.01 mm
wall thickness after being equilibrated at 20  C for 24 h. In addition,
samples were exposed at the incident X-ray monochromatic beam
for 10 min. The sample-to-detector distance was 261.2 mm, and the
temperature was kept at 26.0  C. The average repeat distance of the
amorphous and crystalline lamellae of each sample was calculated
by Eq. (3). The relationship between q and q can be calculated by Eq.
(4).
2p
d¼
q rate of 10  C/min. A sealed empty pan was used as a reference. The
4p sin q
q¼
l were determined. Two parallel replications were measured for each
where d (nm) is the lamellar repeat distance and q (1/nm) is the
scattering vector. l (nm) is the wavelength of the X-ray source.
All I(q) data were normalized so as to have the uniform primary
intensity at q ¼ 0 for transmission calibration. The background
scattering contributions from capillary and solvent were corrected.
Desmearing is necessary because of the line collimation.
2.8. Atomic force microscopy (AFM)
The starch morphology was observed using an AFM (Vecco
Metrology Group, Digital Instruments, USA) (Deng et al., 2014).
Aliquots of sample suspensions (0.05%, w/v) were deposited onto
smooth mica plates and adhered to the stainless plate by double-
sided tape after being dried at room temperature. Imaging was
taken in tapping mode with a cantilever resonant frequency of
around 330 kHz and scan rate of 2.441 Hz. All samples were
recorded at ambient conditions (25  C and 50% relative humidity)
and all AFM images contained 512  512 data points. The raw data
images were processed using the Nanoscope 5.31r1 software
(Digital Instruments, America) to obtain the quantitative and
qualitative information of AC starch aggregate morphology. Surface
roughness was measured using Ra, the arithmetic average of ab-
solute values, which is defined as Eq. (5).
1X n
Ra ¼ jy j (5)
n i¼1 i
where yi is the height of the surface at ith location, n is the number
of data points in the image.
2.9. Differential scanning calorimetry (DSC)
Thermal properties were determined using a DSC-7 analyzer
(Perkin Elmer, Connecticut, USA) as proposed by Han et al. (2012)
with minor modification. Starch samples (2.5 mg, dry basis) were
weighted directly into an aluminum pan, and distilled water
(7.5 mL) was added to obtain a starch-water ratio of 1:3 (w/w). The
pan was hermetically sealed and allowed to equilibrate for 2 h
before analysis. The pans were then heated from 20 to 100  C at the
(3)
onset temperature (To), the endothermic peak temperature (Tp), the
final temperature (Tc) and the enthalpy of gelatinization (DH, J/g)
(4)
experiment.
2.10. Thermogravimetric analysis (TGA)
Thermogravimetric analysis of native and substituted AC starch
was performed using a Perkin Elmer Pyris 1 Thermogravimetric
system (Perkin Elmer Inc, USA). Samples were subjected to a
heating rate of 10 C/min in the range of 30e600  C with Al2O3 as
reference material. Nitrogen was used as the purge gas at a flow
rate of 20 mL/min.
2.11. Viscosity behavior
Rheological properties of starches were measured by a Bra-
bender Viscograph-V (Brabender GmbH & Co. KG, Duisburg, Ger-
many). A starch suspension (6%, w/w) was prepared by weighing
starch (6 g, dry basis) into the Brabender canister with distilled
water. The starch suspension was heated at a rate of 7.5  C/min from
30  C to 95  C, maintained at 95  C for 5 min, cooled to 50  C at a
rate of 7.5  C/min, and then maintained at 50  C for 5min.
 J. Hong et al. / Food Hydrocolloids 54 (2016) 139e150
2.12. Solubility and swelling power
The solubility (S) and swelling power (SP) was determined using
the method reported by Collado and Corke (1997) with minor
amendments. Approximately 0.5 g (dry basis) of each sample was
mixed with distilled water to obtain a 1% (w/w) starch slurry in a
centrifugal tube. The suspensions were heated at 95  C for 30 min.
The cooked samples were then cooled to room temperature at
centrifugal force of 146 g for 20 min. The supernatants were placed
in glass Petri dishes and dried at 105  C for 12 h to constant weight.
The calculation of S and SP was performed as reported by Collado
and Corke (1997). Three replications were performed for each
experiment.
2.13. Freeze-thaw stability
6.0 g (dry basis) of starch was weighted to prepare a 6% (w/w)
suspension with distilled water in a conical flask with stopper. AC
starch suspensions were heated in boiling water for 30 min. The
suspensions were then transferred to cylindrical plastic tubes after
being cooled to room temperature. Finally, the cylindrical plastic
tubes (50 mL, Anke Equipment Ltd., Shanghai, China) were frozen
in a chest freezer (Sanyo refrigerator, model SF-C1497) at
20  C for
24 h and then thawed at room temperature for 6 h. If no water was
separated out after thawing, the samples were freeze-thawed a
second time. In this study, all the samples were subjected to up to 5
freeze-thaw cycles. Syneresis was measured using a modified
method of Arunyanart and Charoenrein (2008). The cylindrical
plastic tube was placed into centrifuge tubes. The tube was then
centrifuged at 146 g centrifugal force (TDL-5-A, Anke Equipment
Ltd., Shanghai, China) at 25  C for 20 min. The amount of liquid
separated from the gel was measured in a burette with filter paper
(Whatman No. 41). The percentage of syneresis (Y (%)) was calcu-
lated as the ratio of the amount of separated liquid (g) to the total
weight (g) of the gel before centrifugation and multiplied 100 by
Eq. (6). The data was reported as averages of three measurements.
m1
m2
Yð%Þ ¼  100 (6)
m1
where m1, m2 refers to the quantity of starch paste after thawing
and sediment after centrifugation, respectively.
2.14. Statistical analysis
All treatments and measurements were replicated two or three
times. The RSM design was analyzed using the statistical package
Design Expert 8.0 (Stat-Ease, Inc. Minneapolis, USA). Analysis of
variance (ANOVA) was performed to compare treatment means and
differences were considered as significant at P＜0.05.
3. Results and discussion
3.1. Combined effects of processing parameters on DS
From the experimental conditions and corresponding response
values shown in Table 1b, the models for the DS ranged from 0.085
to 0.131 in the coded and actual form were obtained in the
quadratic prediction function. The DS of control samples by con-
ventional method were 0.066, 0.085 and 0.093 with the acetic
anhydride content of 6%, 8%, 10%, respectively. There is an obvious
and prominent improvement in DS after synergy of PEF compared
to the above data and other researches using conventional method
(Ayucitra, 2012; Bhosale & Singhal, 2006; Miao et al., 2014). The
fitted quadratic model for DS is given in Eq. (7).
143
DS ¼ 0:064
0:038  X1
0:012  X2 þ 0:039  X3

0:00761  X1 X2 þ 0:012  X1 X3 þ 0:00417  X2 X3

0:011  X21
0:00198  X22 þ 0:00748  X23
(7)
where X1, X2 and X3 is the coded value of the test variables, namely,
the PEF strength, PEF treatment time and acetic anhydride content,
respectively.
The results of the response surface model fitting and ANOVA are
given in Table 1c. The ANOVA of the quadratic regression model
indicates that the model is significant, as evident from the Fisher's
F-test value of 43.87 and a very low probability value P-value (＜
0.0001) for the overall model and individual model parameters
(Table 1c). The goodness of the fit of the model is good as shown by
a determination coefficient (R2) value of 0.960. The 3D response
and contour plot shown in Fig. 2 are graphical representations of
the regression equation from which DS values can be predicted
under different reaction conditions. Each contour curve represents
an infinite number of combinations of two different variables with
the third variable at zero level (PEF strength of 2 kV/cm, acetic
anhydride content of 8%) and the maximum DS value was confined
at the peak ellipse in the contour diagrams.
In summary, the maximum DS of 0.131 was obtained at a PEF
strength (X1) of 2 kV/cm, PEF total treatment time of 51 ms (X2) and
acetic anhydride content (X3) of 10%. Experimental DS values were
in agreement with the predicted DS values shown in Table 1b.
3.2. FT-IR analysis
The FT-IR spectra of native starch, AC starches with DS 0.085, DS
0. 092, DS 0.099 and DS 0.131 are shown in Fig. 3. For native potato
starch spectrum, the peak around 3000e3600 cm
1 was attributed
to the complex stretching vibration associated with free inter- and
intra-molecular bound hydroxyl groups. The observed band at
2929 cm
1 was attributed to CeH bond stretching. The band that
appeared at 1650 cm
1 was ascribed to H2O bending vibration
(Hong, Chen, Zeng, & Han, 2016). In comparison with native starch
(Fig. 3a), FT-IR spectra of AC starches with different DS (Fig. 3bee)
exhibited a new band at 1733 cm
1 (C]O stretching of acetyl
group) and 1560 cm
1 (carboxyl group stretching of acetyl group),
which originates from carboxyl group stretching vibration
providing evidence of acetylation (Halal et al., 2015). In this work,
there was only a slight intensity increase of the ester characteristic
peak at 1733 cm
1 with increasing DS which is due to the relatively
low DS value of the esters (0.086e0.131) (Lu, Luo, Fu, & Xiao, 2013;
Shogren, 2003).
3.3. XRD analysis
XRD spectra of native and AC starch with various DS are shown
in Fig. 4. Native cassava starch exhibited sharp diffraction peaks at
2q ¼ 15.1, 17.3 , 18.0 and 23.0 and small peaks at 11.25 , 19 and
30 which was similar to Klein, Vanier, Moomand, Pinto, Colussi, da
Rosa Zavareze et al. (2014). In contrast to the native starch, the
relative crystallinity of PEF-treated AC starches is reduced from
33.3% (native) to 27.6% (Fig. 4b). The crystallinity of starch is based
on the formation of double helices in the amylopectin chains (Diop,
Li, Xie, & Shi, 2011). As shown in Figs. 3 and 4, some hydroxyl
groups in AC starch were replaced by acetyl groups, which reduced
the formation of inter- and intra-molecular hydrogen bonds and
resulted in a partial destruction of the ordered crystal structure of
starch (Halal et al., 2015).
In this study, the relative crystallinity of AC samples with a DS of
144 J. Hong et al. / Food Hydrocolloids 54 (2016) 139e150

Fig. 2. The effects of PEF strength (X1), PEF treatment time (X2) and acetic anhydride (X3) on DS of three-dimensional (3D) response surface curves (a, c, e) and contours (b, d, f).

Transmittance
3000-3600, -OH
4000 3500
Fig. 3. FT-IR spectra of native (a) and PEF assisted AC cassava starches with DS of 0.085 (b), 0.092 (c), 0.099 (d) and 0.131 (e).
Diffraction Intensity
17.3 18.0
15.1 23.0
10 20 30
Two Theta (deg)
Fig. 4. XRD spectra at wavelength l ¼ 0.154 nm for native (a) and PEF assisted AC
cassava starches with DS of 0.085 (b), 0.092 (c), 0.099 (d) and 0.131 (e).
0.085, 0.092, 0.099, and 0.131 were 27.6%, 28.5%, 31.1% and 29.7%,
respectively. Fig. 4 shows that AC starch had a lower crystallinity
than native starch. This has also been proved by others (Chi et al.,
2008; Diop et al., 2011). This is possibly the result of the energy
input of the PEF treatment which possibly changed the crystalline
region of starch granules assisting acetyl groups to replace hydroxyl
groups. As DS decreased, the crystallinity of AC starch was
decreased (Fig. 4). This effect was more pronounced in starches
treated at higher PEF strength (i.e., 2.5 kV/cm) resulting in
increased damage in native starch granule and enhanced acetyla-
tion (Han, Zeng, Yu et al., 2009, 2012).
3.4. SAXS
As the internal structure of starch is proposed to be fractal, a
power-law decay of the scattering pattern can be observed in Fig. 5,
showing the relationship I ~ qa at scattering vector q on a double-
J. Hong et al. / Food Hydrocolloids 54 (2016) 139e150 145
d 1733, C=O
c
2933, C-H
b 1650, H 2O
1733 1560, O=C-O
1560 a
1650
1650, H 2O
3000 2500 2000 1500 1000 500 1800 1700 1600 1500
-1 -1
Wave numbers(cm ) Wave numbers(cm )
logarithmic scale. The exponent a is calculated from the linear
relationship between I and q on the double-logarithmic scale
(Vermeylen et al., 2006). The average repeat distance d of the semi-
crystalline lamellae was predicted by the peak value of q at round
e 0.5e1 nm
1 and the fractal dimension was calculated as follows:
29.7%
Ds ¼ 6 þ a (
4＜a＜
3) and Dm ¼
a (
3＜a＜
1), which rep-
resents the surface fractal structure and the mass fractal structure,
d respectively (Zhang, Li, Liu, Xie, & Chen, 2013).
31.1%
AC starch with a DS of 0.131 had a slope a ¼
4.83 and a range
of
5 <a <
4 which is in fair agreement with a ¼
4 expected for
c
28.5% the Porod region of structures with sharp phase boundaries and
deviations from transition regions (Vermeylen et al., 2006). It
b
27.6% shows that more hydroxyl groups are substituted by acetyl groups
with the increasing DS resulting in more hydrophobicity of the
starch granules and reduced swelling in water. For native starch, Ds
was 2.51 with the semi-crystalline lamellae d of 9.66 nm which is
a
33.3% reported frequently (Zhang et al., 2013). As DS increased from 0.085
to 0.131, the fractal structure dimension was increased from 1.74 to
40 50 2.69. In Fig. 5b, the slope of native starch was higher than AC starch
with low DS of 0.085 and 0.092 but lower than AC starch with low
DS of 0.099 and 0.131 which was identified with the structure di-
mensions. The slope and fractal dimension are directly related to
the compactness of the scattering particles, where densely packed
structures give rise to a scattering intensity (Schmitt et al., 2010;
Zhu et al., 2013). Therefore, the compact structure corresponds to
a larger fractal dimension as well as a steeper slope. On the basis of
the fractal geometry, the gradually imposed real line slope a
demonstrated that starch micelles with higher DS swelled less in
water and became compacter during the sample preparation pro-
cess (24 h swelling in excess water).
AC starch with higher DS had a larger slope as well as more
compacter micelles. However, as shown in Fig. 5b shows that the
slope a was 1.74 for AC starch with DS of 0.085. This is smaller than
the slope of 2.41 for octenyl succinic anhydride (OSA) modified
starch micelle with a DS of 0.02 as reported by other researchers
(Zhu et al., 2013). Another phenomenon observed was that PEF
treatment resulted in grooves and cavities on the starch granule
surface, which can provide effective channels for water diffusion
(Han et al., 2012). Therefore, PEF treated starch micelles swell more
in water resulting in a lower slope a and less compacter micelle
detected by SAXS (Han et al., 2012; Zhu, Li, Chen, & Li, 2012).
146 J. Hong et al. / Food Hydrocolloids 54 (2016) 139e150
Fig. 5. The SAXS patterns (logelog) of native and PEF-assisted AC starches. The
intensities are shifted along the ordinate in order to avoid overlapping of the symbols.
(a) SAXS patterns in the range of 0.06<q<5. The real lines show the relationship I~qa.
(b) The comparation of SAXS patterns of native (a) and PEF assisted AC cassava starch
with DS of 0.085 (b), 0.092 (c), 0.099 (d) and 0.131 (e).
3.5. Starch micelle aggregates morphology observed by AFM
AFM was applied as a novel tool to help elucidate the micro and
nano-scale attributed to acetyl groups. Several studies have
explored the nanostructure of native starch (Juszczak, Fortuna, &
Krok, 2003, Juszczak, Fortuna, Krok et al, 2003), phsically mod-
ificated starch (Deng et al., 2014) and enzymatic starch hydrolysis
(Sujka & Jamroz, 2009). However, only a few studies have used AFM
to observe chemical modification of starch, especially acetylation.
For example, OSA starch was only examined by the cross section of
AFM and the analysis to study the structual roughness of nano-
structure was limited (Zhu et al., 2013).
Therefore, in this study, the topographic images of native starch
and AC starch with a DS of 0.131 were studied and the structural
changes are presented in Fig. 6. In Fig. 6a and e, blockets are clearly
visible on the surface of the native starch, and these blockets are
clustered or fused together to form nodules. The blocklets can be
attributed to the alternative amorphous and semi-crystalline
growth rings of starch granules and the diameter of the blocklets
varied from 20 to 500 nm depending on starch origins and the
processing methods (Deng et al., 2014).
In contrast to native starch, AC starch granules had more and
larger nodules on the surface. Gaps can also be observed on the
surface of native and AC granules. In the phase images of Fig. 6ceg,
it was found that the mean surface roughness parameters of mean
roughness (Ra) was increased from 47 nm (native starch) to 58 nm
(AC starch). This shows that the compatibility between acetyl
groups and starch granules was good, and the acetyl groups were
unevenly dispersed in the starch matrix. The cross section analysis
of native and AC starch granules is shown in Fig. 6d, h. After being
treated by acetylation, aggregates of spheroids can be seen in
Fig. 6h. This is similar to other AFM work on the complexation of
starch granule (Zabar et al., 2010; Zhu et al., 2012). Arrows 1 and 2
marked in the images represent a peak and valley can be seen in the
line of Fig. 6d, h. Furthermore, the surface roughness Ra as well as
the root mean square of Z data (RMS) can be used to characterize
the roughness of the cross section. The starch matrix in Fig. 6h was
substituted by hydrophobic groups, resulting in a higher roughness
(Ra ¼ 1.893 nm, RMS ¼ 4.564 nm) than that of the native starch
(Ra ¼ 1.148 nm, RMS ¼ 1.733 nm).
Researches have pointed out that water molecules play an
important role in macromolecules (Rumpold, 2005). As PEF was
applied in acetylation, efficient energy was supplied to accelerating
the breakage of bonds and accelerated the replacement of acetyl
groups which resulted in more hydrophobic groups as well as
rougher surface of AC starch granule nanostructure. Therefore, as
DS increased, more acetyl groups are linked to starch chains
resulting in additional space between the semicrystalline and
amorphous lamellae of the starch molecule which can be occupied
by water molecules (Deng et al., 2014). In addition, based on the
Van der Waals force, in aqueous solution the hydrophobic part
forms the core of the micelle and hydrophilic part is oriented to-
wards the water phase. As a consequence, PEF treatment of cassava
starch allowed the introduction of hydrophobic groups and helped
to form clusters in granules which increased the diameter of the
blockets as well as the surface roughness.
3.6. Thermal and thermogravimetry analysis
The gelatinization temperature of the native and PEF-assisted
AC starch determined by DSC is shown in Table 2. The onset tem-
perature (To), peak temperature (Tp), endset temperature (Te) and
enthalpy (DH) of AC starches were lower than that of native starch
which in agreement with other studies (Adebowale, Afolabi, & Olu-
Owolabi, 2006; Lawal, 2004). As for AC starches, the gelatinization
temperature and enthalpy were decreased with increasing DS. AC
starch with a DS of 0.131 exhibited the lowest thermal stability
(indicated by Te) and with a DH of 10.1 J/g compared to other AC
starches tested in this study. This can be explained as a weakness of
internal hydrogen bonding due to the reduction of hydroxyl groups
substituted by hydrophobic acetyl groups. Beyond that, based on
our previous work (Han et al., 2012), it was reported that the
enthalpy of cassava starch decreased after being treated by PEF.
Single acetylation can also decreased its thermal stability
(Adebowale et al., 2006). Therefore, the decrease in the enthalpy of
PEF-assisted AC starch might be the effect of both modifications.
The results of the thermogravimetry (TG) analysis are shown in
Fig. 7. In Fig. 7a, a slight mass loss at 40e130  C is seen which due to
the volatilization of physically retained water in the starch sample
(Danilovas, Rutkaite, & Zemaitaitis, 2014). At 250e390  C, a sig-
nificant mass reduction from 86% to 20% occurs due to the degra-
dation of starch (Zhu, Li, Huang, Chen, & Li, 2014). The temperature
at 50% weight loss was 360  C for the native and 340  C for AC
 J. Hong et al. / Food Hydrocolloids 54 (2016) 139e150 147

Fig. 6. AFM topography images (500 nm  500 nm1 mm  1 mm) of native starch (a, b, c, d) and PEF-assisted AC starches with DS of 0.131 (e, f, g, h). Figures (a), (b), (e), and (f) are
topography images, whereas figures (c) and (g) are phase images in surface plot. Figures (d) and (h) represent the cross section analysis.

starch with a DS of 0.131. Correspondingly, the decomposition stability of the AC starches was lower than that of the native starch,
temperature corresponding to the maximum rate of weight loss which can be explained by the instability of carboxylic groups and
(Tmax) of the native and AC starch (DS of 0.131) were 350  C and breakage of hydrogen bonds during acetylation (Singh & Nath,
323  C, respectively. These results indicated that the thermal 2012). Since the main decomposition mechanism of starch is the
148 J. Hong et al. / Food Hydrocolloids 54 (2016) 139e150

Fig. 6. (continued).

dehydration reaction between starch hydroxyls, AC starch with
lower DS, which means less unstable hydroxyl groups, has better
thermal stability among AC starches with different DS. These re-
sults are in agreement with earlier studies about starch esters
(Peng, Ren, Zhong, & Sun, 2011; Rudnik, Matuschek, Milanov, &
Kettrup, 2005). Derivative thermogravimetry (DTG) curves in
Fig. 7b show that the maximum rate of weight loss rate of AC starch
(DS ¼ 0.131) at about 330  C was much higher than other starches
except for native starch. This indicated that AC starch with higher
DS would loss more weight at the same temperature and exhibited
worse thermal stability.
tested samples. All AC starches had lower DSC peak temperatures
(Tp) than native starch (Bao et al., 2003). However, peak viscosity
(Pv), breakdown (BD) and setback value (SB) of AC starches were
lower than those of native starch, which is different to previous
findings (Bao et al., 2003). This difference is likely the result of the
substantially greater substitution of AC starch with acetic anhy-
dride residue compared to other studies. Furthermore, the reaction
mechanism for relatively low Pv, SD and SB of PEF-assisted AC
starch needs further investigation of the rheological properties of
resultant pastes and gels.

3.8. Solubility and swelling power

3.7. Viscosity behavior
The pasting properties of native and AC cassava starches are
shown in Table 2. The peak viscosity obtained from native starch
was 238 Brabner viscosity (BU) at 71.9  C. The peak viscosity of AC
starch with DS of 0.086 was 296 BU which was higher than other
Table 2 shows the solubility and swelling power (SP) of native
cassava starch and AC starch with various DS. The solubility and SP
of native starch is 10.5% and 24.2 g/g, respectively. After PEF-
assisted acetylation increased, its the solubility and SP increased
of cassava starch up to 17.0% and 33.9 g/g, respectively. And the

Table 2
Characteristics of native and AC cassava starches.

Sample Solubility (%) Swelling power (g/g) Thermal stability Brabender viscosity
  
To ( C) Tp ( C) Te ( C) DH (J/g) Pv (BU) BD (BU) SB (BU)

Native 10.5 ± 0.75c 24.2 ± 0.67c 65.8 ± 0.4a 71.9 ± 0.1a 82.5 ± 0.3a 18.1 ± 2.5a 238 ± 8.20c 142 ± 4.28b 89 ± 2.17b
DS of 0.085 13.7 ± 0.19b 26.2 ± 0.20b 62.4 ± 0.1b 68.3 ± 0.4b 77.9 ± 0.8b 14.7 ± 1.3b 296 ± 4.86a 171 ± 2.13a 102 ± 0.12a
DS of 0.092 10.5 ± 0.59c 21.7 ± 1.14d 62.6 ± 0.3b 67.3 ± 1.1b 77.1 ± 0.1b 12.1 ± 2.2bc 295 ± 5.73a 169 ± 7.10a 102 ± 3.73a
DS of 0.099 17.0 ± 0.86a 33.4 ± 0.77a 61.5 ± 0.6c 67.1 ± 0.2b 75.2 ± 1.2c 11.9 ± 1.0bc 264 ± 4.10b 146 ± 6.08b 86 ± 1.90b
DS of 0.131 15.9 ± 0.15a 33.9 ± 1.17a 61.5 ± 0.3c 67.2 ± 0.7b 73.9 ± 0.8c 10.1 ± 0.6c 286 ± 7.00a 165 ± 0.40a 87 ± 3.33b

The data (mean ± standard deviation) had three replicates as described in materials and methods. To, onset temperature; Tp, pasting temperature; Te, endset temperature; Pv,
peak viscosity; BD, breakdown; SB, setback; BU, Brabender unit. Means in the same column followed by different letters (aef) are significantly different (P＜0.05).
 J. Hong et al. / Food Hydrocolloids 54 (2016) 139e150
Fig. 7. TGA curves of native (a) and PEF assisted AC cassava starches with DS of 0.085
(b), 0.092 (c), 0.099 (d) and 0.131 (e).
higher DS is, the higher SP exhibits. For instance, the AC starch with
DS of 0.086 showed lower solubility (13.86%) and lower SP (26.2 g/
g) compared to solubility of 15.9% and SP of 33.9 g/g which origi-
nated from AC starch with the highest DS of 0.131.
Similar results have been reported earlier in other studies (Liu,
Ramsden, & Corke, 1997; Singh et al., 2004). The changes
observed in solubility and SP by acetylation can be attributed to the
introduction of hydrophobic substituting groups. The introduction
of acetyl groups in starch can electrstatically bind water molecules
and facilitate access of water to amorphous areas (Sodhi & Singh,
2005).
3.9. Freeze-thaw stability
The freeze-thaw stability of cassava starches with different DS is
displayed in Fig. 8. Native starch had higher syneresis and lower
freeze-thaw stability than AC starches. Syneresis in freeze-thawed
starch gels is due to the re-crystallization of amylose molecules in
starch granules at reduced temperature resulting in a loss of water
from the gel structure (Arunyanart & Charoenrein, 2008). AC
starches with PEF treatment showed good freeze-thaw stability for
up to 5 cycles (Fig. 8). The reduction tendency in AC starches is due
149
a
b
c
40 d
e
Syneresis/(%)
20
1 2 3 4 5
Cycle times
Fig. 8. Freeze-thaw stability of native (a) and PEF assisted AC cassava starches with DS
of 0.085 (b), 0.092 (c), 0.099 (d) and 0.131 (e).
to the prevention of alignment and association between macro-
molecules by the acetyl groups (Gonzalez & Perez, 2002).
Ayucitra (2012) indicated that AC starch with DS of 0.080
exhibited syneresis from 21% (cycle 1) to 45% (cycle 5). This is
significantly larger than the 0.04% (cycle 1) to 9.8% (cycle 5) found
for PEF-assisted AC starch with a DS of 0.085 in this study (Fig. 8b).
Furthermore, the syneresis values of PEF-assisted AC starches were
no more than 10% after 5 cycles which revealed its extremely good
stability in freeze-thaw ability. Hence, it can be speculated that
freeze-thaw stability of PEF-assisted AC starch is better than that of
AC starch prepared by conventional method. Therefore, PEF-
assisted acetylation has the potential to enable stabilization of
starch gels in frozen foods.
4. Conclusions
In the present study, the effects of PEF treatment on nano-
structure, morphology and functionality of acetylated starch
were investigated. The granular and crystalline structures were
destroyed and rearranged and a mass fractal structure was formed
with the increasing DS. These results showed that PEF-assisted,
acetylated starch had better potential to stabilize starch gels at
low temperature condition in frozen food industry. In addition, as
the hydrophobic side chains are added to the originally merely
hydrophilic starch molecule by acetyl groups and starch may
adsorb to the interface of water and oil thus becoming a stabilized
emulsion. Therefore, the information discovered in the current
paper will be useful for promoting the use of the AC starch in food
industry and the stabilization of emulsions, digestibility as well as
its rheology of PEF-assisted AC starch need to be further studied.
Acknowledgments
This research was supported by Chinese National “863 Project”
(2011AA100801), Chinese National Science Fund (21376094,
31301559), S&T project of Guangdong province (2012A020200002,
2013B020203001, 2013B020311008), Open Project Program of
State Key Laboratory of Food Science and Technology, Nanchang
University (No. SKLF-KF-201407), Fundamental Research Funds for
the Central Universities, China (2015). The authors are grateful to
150 J. Hong et al. / Food Hydrocolloids 54 (2016) 139e150
the Guangdong Province Government (China) for support through
the program of “Leading Talent of Guangdong Province (Da-Wen
Sun)”.
References
Adebowale, K. O., Afolabi, T. A., & Olu-Owolabi, B. I. (2006). Functional, physico-
chemical and retrogradation properties of sword bean (Canavalia gladiata)
acetylated and oxidized starches. Carbohydate Polymers, 65, 93e101.
Aini, N., & Purwiyatno, H. (2010). Gelatinization properties of white maize starch
from three varieties of corn subject to oxidized and acetylated-oxidized
modification. International Food Research Journal, 17, 961e968.
Arunyanart, T., & Charoenrein, S. (2008). Effect of sucrose on the freeze-thaw sta-
bility of rice starch gels: correlation with microstructure and freezable water.
Carbohydrate Polymer, 74, 514e518.
Ayucitra, A. (2012). Preparation and characterization of acetylated corn starches.
International Journal of Chemical Engineering and Applications, 3, 156e159.
Bao, J. S., Xing, J., Phillips, D. L., & Corke, H. (2003). Physical properties of octenyl
succinic anhydride modified rice, wheat, and potato starches. Jounal of Agri-
cultural and Food Chemistry, 51, 2283e2287.
Bhosale, R., & Singhal, R. (2006). Process optimization for the synthesis of octenyl
succinyl derivative of waxy corn and amaranth starches. Carbohydrate Polymer,
66, 521e527.
Biswas, A., Shogren, R. L., Selling, G., Salch, J., Willett, J. L., & Buchanan, C. M. (2008).
Rapid and environmentally friendly preparation of starch esters. Carbohydrate
Polymers, 74, 137e141.
Buckow, R., Schroeder, S., Berres, P., Baumann, P., & Knoerzer, K. (2010). Simulation
and evaluation of pilot-scale pulsed electric field (PEF) processing. Jounal of
Food Engineering, 101, 67e77.
Chi, H., Xu, K., Wu, X., Chen, Q., Xue, D., Song, C., et al. (2008). Effect of acetylation on
the properties of corn starch. Food Chemistry, 106, 923e928.
Collado, L. S., & Corke, H. (1997). Properties of starch noodles as affected by sweet
potato genotype. Cereal Chemistry, 74, 182e187.
Danilovas, P. P., Rutkaite, R., & Zemaitaitis, A. (2014). Thermal degradation and
stability of cationic starches and their complexes with iodine. Carbohydrate
Polymers, 112, 721e728.
Deng, Y., Jin, Y., Luo, Y., Zhong, Y., Yue, J., Song, X., et al. (2014). Impact of continuous
or cycle high hydrostatic pressure on the ultrastructure and digestibility of rice
starch granules. Journal of Cereal Science, 60, 302e310.
Diop, C. I. K., Li, H. L., Xie, B. J., & Shi, J. (2011). Effects of acetic acid/acetic anhydride
ratios on the properties of corn starch acetates. Food Chemistry, 126, 1662e1669.
Gonzalez, Z., & Perez, E. (2002). Effect of acetylation on some properties of rice
starch. Starch-Sta €rke, 54, 148e154.
Halal, S. L. M. E., Colussi, R., Pinto, V. Z., Bartz, J., Radunz, M., Carreno, N. L. V., et al.
(2015). Structure, morphology and functionality of acetylated and oxidised
barley starches. Food Chemistry, 168, 247e256.
Han, Z., Zeng, X. A., Fu, N., Yu, S. J., Chen, X. D., & Kennedy, J. F. (2012). Effects of
pulsed electric field treatments on some properties of tapioca starch. Carbo-
hydrate Polymers, 89, 1012e1017.
Han, Z., Zeng, X. A., Yu, S. J., Zhang, B. S., & Chen, X. D. (2009a). Effects of pulsed
electric fields (PEF) treatment on physicochemical properties of potato starch.
Innovative Food Science and Emerging Technologies, 10, 481e485.
Han, Z., Zeng, X. A., Zhang, B. S., & Yu, S. J. (2009b). Effects of pulsed electric fields
(PEF) treatment on the properties of corn starch. Journal of Food Engineering, 93,
318e323.
Hong, J., Chen, R. J., Zeng, X.-A., & Han, Z. (2016). Effect of pulsed electric fields
assisted acetylation on morphological, structural and functional characteristics
of potato starch. Food Chemistry, 192, 15e24.
Huang, Q., Fu, X., He, X. W., Luo, F. X., Yu, S. J., & Li, L. (2010). The effect of enzymatic
pretreatments on subsequent octenyl succinic anhydride modifications of
cornstarch. Food Hydrocolloids, 24, 60e65.
Juszczak, L., Fortuna, T., & Krok, F. (2003a). Non-contact atomic force microscopy of
starch granules surface. Part I. potato and tapioca starches. Starch-Sta €rke., 55,
1e7.
Juszczak, L., Fortuna, T., & Krok, F. (2003b). Non-contact atomic force microscopy of
starch granules surface. Part II. selected cereal starches. Starch-Sta €rke, 55, 8e16.
Jyothi, A. N., Rajasekharan, K. N., Moorthy, S. N., & Sreekumar, J. (2005). Microwave-
assisted synthesis and characterization of succinate derivatives of cassava
(Manihot esculenta Crantz) starch. Starch-Sta €rke, 57, 556e563.
Kavlani Neelam, S. V., & Singh, L. (2012). Various techniques for the modification of
starch and the applications of its derivatives. International Research Journal of
Pharmacy, 3, 25e31.
Kim, H. N., Sandhu, K. S., Lee, J. H., Lim, H. S., & Lim, S. T. (2010). Characterisation of
2-octen-1-ylsuccinylated waxy rice amylodextrins prepared by dry-heating.
Food Chemistry, 119, 1189e1194.
Klein, B., Vanier, N. L., Moomand, K., Pinto, V. Z., Colussi, R., da Rosa Zavareze, E.,
et al. (2014). Ozone oxidation of cassava starch in aqueous solution at different
pH. Food Chemistry, 155, 167e173.
Lawal, O. S. (2004). Succinyl and acetyl starch derivatives of a hybrid maize:
physicochemical characteristics and retrogradation properties monitored by
differential scanning calorimetry. Carbohydrate Research, 339, 2673e2682.
Lin, Z. R., Zeng, X. A., Yu, S. J., & Sun, D. W. (2012). Enhancement of ethanol-acetic
acid esterification under room temperature and non-catalytic condition via
pulsed electric field application. Food and Bioprocess Technology, 5, 2637e2645.
Liu, H., Ramsden, L., & Corke, H. (1997). Physical properties and enzymatic di-
gestibility of acetylated ae, wx, and normal maize starch. Carbohydrate Poly-
mers, 34, 283e289.
Lu, X., Luo, Z., Fu, X., & Xiao, Z. (2013). Two-step method of enzymatic synthesis of
starch laurate in ionic liquids. Journal of Agricultural and Food Chemistry, 61(41),
9882e9891.
Matsumoto, A., Tsubaki, S., Sakamoto, M., & Azuma, J. (2011). A novel saccharifi-
cation method of starch using microwave irradiation with addition of activated
carbon. Bioresource Technology, 102, 3985e3988.
Miao, M., Li, R., Jiang, B., Cui, S. W., Zhang, T., & Jin, Z. (2014). Structure and phys-
icochemical properties of octenyl succinic esters of sugary maize soluble starch
and waxy maize starch. Food Chemistry, 151, 154e160.
Peng, X., Ren, J., Zhong, L., & Sun, R. (2011). Homogeneous synthesis of hemi-
cellulosic succinates with high degree of substitution in ionic liquid. Carbohy-
drate Polymers, 86, 1768e1774.
Rudnik, E., Matuschek, G., Milanov, N., & Kettrup, A. (2005). Thermal properties of
starch succinates. Thermochimica Acta, 427, 163e166.
Rumpold, B. A. (2005). Impact of high hydrostatic pressure on wheat, tapioca, and
potato starches (p. 120). Berlin: Berlin Technical University. PhD thesis.
Schmitt, C., Moitzi, C., Bovay, C., Rouvet, M., Bovetto, L., Donato, L., et al. (2010).
Internal structure and colloidal behaviour of covalent whey protein microgels
obtained by heat treatment. Soft Matter, 6, 4876e4884.
Shogren, R. L. (2003). Rapid preparation of starch esters by high temperature/
pressure reaction. Carbohydrate Polymers, 52, 319e326.
Singh, N., Chawla, D., & Singh, J. (2004). Influence of acetic anhydride on physico-
chemical, morphological and thermal properties of corn and potato starch. Food
Chemistry, 86, 601e608.
Singh, A. V., & Nath, L. K. (2012). Synthesis, characterization, and compatibility
study of acetylated starch with lamivudine. Journal of Thermal Analysis and
Calorimetry, 108, 307e313.
Sodhi, N. S., & Singh, N. (2005). Characteristics of acetylated starches prepared using
starches separated from different rice cultivars. Journal of Food Engineering, 70,
117e127.
Sujka, M., & Jamroz, J. (2009). a-Amylolysis of native potato and corn starches-SEM,
AFM, nitrogen and iodine sorption investigations. LWT-Food Science and Tech-
nology, 42, 1219e1224.
Vermeylen, R., Derycke, V., Delcour, J. A., Goderis, B., Reynaers, H., & Koch, M. H.
(2006). Gelatinization of starch in excess water: beyond the melting of lamellar
crystallites. A combined wide-and small-angle X-ray scattering study. Bio-
macromolecules, 7, 2624e2630.
Wang, Y. J., & Wang, L. F. (2002). Characterization of acetylated waxy maize starches
prepared under catalysis by different alkali and alkaline-earth hydroxides.
Starch-Sta €rke, 54, 25e30.
Zabar, S., Lesmes, U., Katz, I., Shimoni, E., & Bianco-Peled, H. (2010). Structural
characterization of amylose-long chain fatty acid complexes produced via the
acidification method. Food Hydrocolloids, 24, 347e357.
Zeng, X. A., Han, Z., & Zi, Z. H. (2010). Effects of pulsed electric field treatments on
quality of peanut oil. Food Control, 21, 611e614.
Zhang, B. J., Li, X. X., Liu, J., Xie, F. W., & Chen, L. (2013). Supramolecular structure of
A- and B-type granules of wheat starch. Food Hydrocolloids, 31, 68e73.
Zhang, Z., Zhao, S., & Xiong, S. (2010). Synthesis of octenyl succinic derivative of
mechanically activated indica rice starch. Starch-Sta €rke, 2010(62), 78e85.
Zhao, W., Yang, R., Lu, R., Wang, M., Qian, P., & Yang, W. (2008). Effect of PEF on
microbial inactivation and physical-chemical properties of green tea extracts.
LWT-Food Science and Technology, 41, 425e431.
Zhu, J., Li, L., Chen, L., & Li, X. X. (2012). Study on supramolecular structural changes
of ultrasonic treated potato starch granules. Food Hydrocolloids, 29, 116e122.
Zhu, J., Li, L., Chen, L., & Li, X. X. (2013). Nano-structure of octenyl succinic anhydride
modified starch micelle. Food Hydrocolloids, 32, 1e8.
Zhu, J., Li, X., Huang, C., Chen, L., & Li, L. (2014). Structural changes and triacetin
migration of starch acetate film contacting with distilled water as food simu-
lant. Carbohydrate Polymers, 104, 1e7.
Zie˛ ba, T., Szumny, A., & Kapelko, M. (2011). Properties of retrograded and acetylated
starch preparations: Part 1. Structure, susceptibility to amylase, and pasting
characteristics. LWT-Food Science and Technology, 44, 1314e1320.