Dual‐frequency ultrasound for ultrasonic‐assisted esterification

Abstract The optimization of wheat starch esterification (acetylation) with a high degree of substitution was performed through response surface methodology (RSM) via various concentrations of reagents (acetic anhydride), pHs, and temperatures under various ultrasonication frequencies (25, 40, and 25 + 40 kHz). According to RSM methodology, optimized samples were selected by achieving high degrees of substitution at various frequencies, temperatures, and pHs. Solubility, swelling, X‐ray, RVA, DSC, freeze–thaw stability, texture, and SEM analysis of the optimized samples were performed at three frequencies. X‐ray pattern exhibited a more significant reduction in the crystallinity percentage of esterified starch at frequency 25 + 40 kHz compared with 25 kHz, 40 kHz, and native starch. According to DSC analysis, To, Tp, Tc, and enthalpy of gelatinization (ΔH gel) were lower in AC at frequency 25 + 40 kHz compared with AC at frequency 25 and 40 kHz and N starches. According to morphology analysis, in acetylated starches at 25 and 40 kHz, the surfaces and small granules underwent more damage, whereas in 25 + 40 kHz, large granules were more affected than small granules. Upon acetylation, freeze–thaw stability and textural properties of the starch significantly increased and decreased, respectively. The peak and final viscosity of acetylated starch increased (25 + 40 kHz ˃ 25 kHz ˃ 40 kHz ˃ N starch).

hardness, adhesiveness, cohesiveness, and translucency of the gels are increased, while the initial gelatinization temperature is reduced (González & Perez, 2002). The replacement of the modifier groups may boost the free movement of the starch chains within the amorphous regions in the granule, owing to the disruptions taking place among the inter-and intramolecular hydrogen bonds. The weakened internal bond structure in the starch granules, due to the derivatized modifier groups, enhances the freeze-thaw stability, reduces the gelatinization temperature, generates high levels of peak viscosity, and leads to starch paste clarity (Han, 2010;Han, Lee, Lim, & Lim, 2005).
Ultrasonic treatments can accelerate chemical reaction and increase yields. Under sonication, the surface area of starch granules increases due to the formation of channels and/or holes on the surface and inside the granules (Huang, Li, & Fu, 2007). Several modifications have been performed upon ultrasonication, including acetylation of dioscorea starch following ultrasound treatment (Zhang, Zuo, Wu, Wang, & Gao, 2012), octenyl succinate starch (Chen, Huang, Fu, & Luo, 2014), carboxymethyl starch (Gao et al., 2011;Shi & Hu, 2013), and octenylsuccinates of carboxymethyl starch (Čížová, Sroková, Sasinková, Malovíková, & Ebringerová, 2008). However, there is no information about the acetylation of starch under various frequencies of ultrasonication. The aim of this study was to optimize the AC wheat starch in various parameters such as acetic anhydride concentration, pH, and temperature under frequencies (25, 40, and 25 + 40 kHz) of bath ultrasound; the optimization was further investigated on the amount of modification (molar substitution) in the modified starch. Based on the minimal reagent consumption, the optimized samples were selected and their physicochemical properties (solubility, water absorption), X-ray diffraction, scanning electron microscopy (SEM), Rapid Visco Analyzer (RVA), differential scanning calorimetry (DSC), and Texture Profile Analyzer (TPA) were investigated at three frequencies (25,40,and 25 + 40 kHz). The freeze and thaw stability of AC wheat starch was further compared with that made with native wheat starch.

| Materials
Native wheat starch was obtained from Fars Glucosin Company, Shiraz, Iran. The moisture, fat, ash, and protein of native (N) and AC wheat starch prepared under ultrasonication were specified along with the standard methods of AACC, 2000. Amylose content (%) was calculated by iodine method (Pourmohammadi, Abedi, Hashemi, & Torri, 2018).

| Acetylation of wheat starch
We added 100 g of starch to 224 ml of distilled water and stirred for 1 hr at 24°C. The pH was set to 6.5, 8, and 9.5 with NaOH solution (1 M), and acetic anhydride (4, 6 and 8%) was slowly added to the suspension at various pHs. Whole acetylation reaction was performed at 10 min. The suspensions were sonicated at an amplitude of 24% (with an input power of up to 400 W and working frequencies of 25, 40, and 25 + 40 kHz) at 45, 60, and 75°C. The time duration of sonication in each frequency was 5 min (25 kHz: 5 min, 40 kHz: 5 min, and 25 + 40 kHz: 5 min). pH was then maintained at 4.5 with HCl solution (1 M) for 5 min. All the experiments were performed in bakers placed at known positions in ultrasonic bath (Pacisa SA) equipped with a temperature controlling system. The ultrasonic bath with internal dimensions of 300 × 150 × 150 mm 3 consisted of a rectangular tank containing four transducers at the bottom. The suspensions were completely washed, lyophilized, and milled using a laboratory mill (AlexanderWerk, Model WEL82), and manually sieved to obtain particle sizes < 40 μm (Wojeicchowski, Siqueira, Lacerda, Schnitzler, & Demiate, 2018).

| Acetyl percentage
The titrimetric method of Mbougueng, Tenin, Scher, and Tchiégang (2012) was employed in order to specify the acetyl group percentage (AC%) and (DS). AC (5.0 g), 50 ml of distilled water, and 24 ml of 0.45 M NaOH were added in a 240-mL flask and mixed for 30 min at room temperature. The surplus alkali was back-titrated using 0.2 M HCl and phenolphthalein as an indicator. The reaction mixture was put to stand for 2 hr, where the alkali was removed from the titrated sample. The native starch was also used as a blank sample. Initially, the acetyl (%) and DS were calculated according to Equations (1)  W is the weight of the sample.

| X-ray diffraction
To obtain a relative humidity of 75%, all starch samples were primarily placed, for 5 days, in a relative humidity box containing supersaturated NaCl at room temperature. X-ray diffraction pattern of the samples was determined by use of an X-ray diffractometer (Model D8 Advance, Germany) (Li et al., 2014). Utilizing the instrument software (EVA, Version 9.0), the degree of starch crystallinity was specified through dividing the area under the peaks by the total curve area.

| Solubility and water absorption
The method of Li et al. (2014) was used to determine water swelling and solubility of the samples with a slight modification. Native and modified wheat starches (1 g) and distilled water (30 ml) were added into a centrifuge tube, and, using a vortex mixer set at high speed, they were shaken at room temperature for 5 min. Each tube was incubated in water bath for 30 min at 95°C followed by centrifugation for 15 min at 3000 g. The weights of the pellet and dried supernatants (at 120°C for about 2 hr) were further obtained. The water solubility and swelling values of the native and modified samples were calculated using Equations (3) and (4), respectively.

| Pasting properties
Using the method described by Pourmohammadi et al. (2018), the pasting properties of the native and modified wheat starches were determined through RVA (Newport Scientific Pty. Ltd) interfaced with a personal computer.

| Thermal properties
The gelatinization properties of the native and modified wheat starches were analyzed using a DSC instrument (OIT-5000; Sanaf) equipped with STAR software making using of  method. The device was calibrated with indium and mercury, and an empty pan was used as a reference. Native and modified wheat starches (10 g) were weighed into the standard aluminum pan. Distilled water (30 ml) was added, and the pans were sealed and equilibrated for 24 hr at room temperature. After that, the starch slurry was gelatinized in the DSC instrument, by heating at 6°C/ min, from 10 to 160°C. Following heat treatment, the samples were cooled to 25°C and then removed from the DSC. The changes in enthalpy (ΔH in J/g of dry starch), onset temperature (T o ), peak temperature (T p ), and conclusion temperature (T c ) for gelatinization were obtained from the exotherm DSC curves.

| Microstructure determination
The microstructure of the modified starch was specified using SEM (Model Leica Cambridge) by use of the method proposed by Pourmohammadi et al. (2018).

| Textural analysis
Textural properties of starch gels were specified using a Texture Profile Analyzer (TA Plus; Stable Micro System). To prepare the paste of gelatinized starch, 10 g of starch was added to 100 ml of distilled water, where the slurry remained in water bath (95°C for 30 min). The hot paste was collected and transferred into a cylindrical plastic container of 10 mm diameter and 10 mm height, and kept at 4°C for 24 hr. The gels were then removed from the container and tested for their textural properties. Utilizing a cylindrical aluminum plunger with a diameter of 10 mm, the gels were packed at a test speed of 2 mm/s, pretest and post-test speed of 5 mm/s, time interval of 10 s, and strain deformation of 24%.
Data were analyzed three times, and the results were averaged.

| Freeze-thaw stability
With slight modifications in the methods described by Lawal (2010), we determined the syneresis of the starch gels belonging to the native and modified starches during cold and frozen storage. A starch suspension (5% dry basis, w/w) was heated at 95°C under constant agitation for 1 hr. 20 g of noodle strands belonging to N and AC starch was centrifuged at 2000 g for 10 min to remove free water. The free water (supernatant) was decanted, and the tubes containing starch paste were exposed to freeze and thaw cycles, followed by centrifugation at 3,500 g for 30 min. The freezing process was done at −18°C for 24 hr, and the melting process was carried out at 30°C for 4 hr. The two processes were repeated for 8 cycles, in each of which the isolated water was determined. The weight of water was taken, and the amount of syneresis was calculated as the percentage of water separated as Equation (5):

| Statistical analysis
So as to specify the regression coefficients, the Design-Expert (version 6.0.5) methodology (Abedi, Sahari, Barzegar, & Azizi, 2015) and statistical software packages SAS 9.1 (SAS Institute) were made. A Box-Behnken design from the response surface was used to study the effect of three different factors on the acetylation degrees of wheat starch. Factors associated with acetylation were acetic anhydride (4, 6, and 8%), pH (6.5, 8, and 9.5), and temperature (45, 60, and 75°C). The experimental data were fitted in accordance with Equation (6) as a second-order polynomial equation, including the linear and interaction effects of each factor: where Y is the predicted response, X i and X j are independent factors, b0 is the offset term, bi is the ith linear coefficient, bii is the ith (3) Water solubility (%) = weight of dissolved solids in supernatant/weight of dry sample solids in the original sample × 100 (4) Water swelling (%) = weight of sediment × 100 ∕ weight of dry sample solids × 100 − solubility quadratic coefficient, and bij is the ijth interaction coefficient. All analyses were obtained in triplicates and reported as mean values.

| Molar substitution of AC wheat starch
Concerning the AC wheat starch at sonication frequencies of 25, 40, and 25 + 40 kHz, optimization was done in order to investigate the effect of acetic anhydride content, pH, temperature, and ultrasonic frequency on the amount of modification or molar substitution of the modified starch ( Figure 1).
As a result of Equations (7-9), the effect of A (acetic anhydride content) > B (pH) > C (reaction temperature) is on route to the maximum acetylation. The negative effect of temperature at dual frequency (25 + 40 kHz) was more than 25 and 40 kHz; however, the effect of temperature at 25 kHz frequency was the same as 40 kHz. The addition-elimination is the main mechanism for acetylation reaction, in which hydroxyl groups are converted to acetyl. The hydroxyl groups presented different levels of reactivity. The most reactive group was the hydroxyl of Carbon 6, followed by those of Carbons 2 and 3 (Wojeicchowski et al., 2018).
The C = O bond of the acetyl group was substituted in all amorphous areas and at the outer lamellae of crystalline sections due to the poor ability of acetic anhydride to penetrate granule structures (Singh et al., 2007).

| Effect of ultrasonication
Under sonification, DS was increased since the solubility and swelling of granules were improved as a result of sufficient contact with reagents, which enhanced the homogeneity of the reactants. Collapse of cavitation bubbles upon ultrasonic treatment is capable of mechanically damaging and rupturing the starch granules and breaking the chains of polymers due to high-pressure gradients, high local velocities of liquid layers, shearing force, and microstreaming. According to Amini, Razavi, and Mortazavi (2015) and Zhu (2015), ultrasound creating fracture, pore, and crack on glanular surfaces may enhance the surface area of starch, facilitate the penetration of esterifying agents (acetyl) into the granular structure, and accelerate the chemical reactions. The frequency of ultrasound is a key factor for enhancing cavitational bubble due to certain reasons (Abedi, Sahari, & Hashemi, 2017): (a) Cavitation yield decreases with the increase in frequency, which is attributed to the scattering, attenuation, and shortening of the acoustic cycle at high frequencies (Abedi et al., 2017). (b) Instable cavitation at low frequencies causes the bubble to collapse very quickly and violently, resulting in more rapid agitation and mass transfer (Abedi et al., 2017).

| Effect of temperature
Temperature had a progressive effect on the DS of the modified starch, which is attributed to the swelling of starch granules, diffusion of the esterifying agents, mobility of reactant molecules for reaction with amorphous regions, destruction of the crystalline regions of the starch granules, and conversion to amorphous regions (Han et al., 2012). This phenomenon can simplify the acceptability of granules in acetic anhydride and propylene glycol percolation.
Temperature is an important factor, yet a very high temperature is not conducive to the absorption of the esterifying agent (acetyl) by the starch. This is because of the exothermic esterification reaction of starch (Han et al., 2012;Singh et al., 2007)

| Effect of pH
DS increased following the addition of NaOH to the reaction medium, when the pH was set to 6.5, 8, and 9.5. Alkaline pH was able to enhance the DS owing to the disruption in the hydrogen bonds between molecules in amorphous and crystalline regions, and the facilitation of the percolation and penetration of the esterifying groups into the starch granular structure Han et al., 2012). Although pH plays a major factor in the DS improvement of acetylation, very high pHs can diminish DS due to the hydrolyzation of acetic anhydride and the inhibition effect of gelatinized layer on the efficient contact between starch and acetic anhydride Han et al., 2012). Various botanical starches have been acetylated with different degrees of acetylation, such as 0.104%-0.184% for maize starch , 0.081 for waxy maize starch (Wang & Wang, 2002), 0.087%-0.118% (Sodhi & Singh, 2005)  The high molar substitution acetylation in this study was interpreted by reaction medium under ultrasonic process, which affects the amylose regions. The extent of modification by ultrasonication in the crystallinity of starch granules depends on experimental conditions and the type of starch (Zhu, 2015), and requires the introduction of acetyl group throughout the amylose or amylopectin sections. According to Singh et al. (2004), with low amylose content (low amylose/amylopectin ratio) a very high DS of acetylation can be observed. Surface of granule is another major factor affecting DS. Furthermore, González and Perez (2002) interpreted that the low degree of substitution in rice starch might be due to the lack of enough large inner channels or granular surface pores which can facilitate the access of acetic anhydride to the interior of the granule.

| The crystalline structure
The X-ray diffraction pattern of native and modified acetylated wheat starches is presented in Figure 2. X-ray diffraction analysis estimates the crystalline structure and the amount of crystal existing in the starch granules. The structure is related to amylopectin double helix. The results showed that the X-ray pattern of wheat starch was the same as other cereals (A pattern). The native starch had sharp diffraction peaks at 15°, 17°, 18°, 23° (2θ), indicating the typical A pattern of cereal starch (Zobel, Young, & Rocca, 1988     These findings are in accordance with the results of Li et al. (2014) where the starch with higher amylose content reduced the gelatinization temperatures due to less crystalline and more amorphous regions (Li et al., 2014). These results are in accordance with certain other research (González & Perez, 2002;Van Hung & Morita, 2005a, b;Liu et al., 1999;Salcedo Mendoza et al., 2016). In these studies, the modification of starch by acetylation increased the peak viscosity due to (a) the decrease in the strength of associative intermolecular forces in the amorphous regions of the granules, (b) lower degrees of crystallinity in AC compared with N wheat starches, (c) increase in solubility and swelling power (Figures 3 and 4), (d) the solubilizing amylose released outside the swollen starch granule (Han, 2010;Salcedo Mendoza et al., 2016), and (e) improvement in the hydrophilic capacity of the reorganized structures (Han, 2010;Salcedo Mendoza et al., 2016). On the other hand, compared with the native starches, the AC wheat starches presented better stability during heating. Reassociation of amylose molecules occur when leached out from swollen starch granule, thus providing a higher resistance to thermal or mechanical forces .   Singh, and Singh (2004).

| The effects of modification on starch morphology
The morphological patterns of N and modified starches are shown in Figure 5 where     (2003), Babic et al. (2007), and .

| Freeze-thaw stability
The purpose of the present research was to stabilize the starch gels from syneresis during freezing and thawing. The tendency to retrogradation of gels prepared from N and AC wheat starches was calculated by determining the syneresis (percentage of water loss) during storage at 4°C ( Figure 6). During the freezing cycles, the crystal size was enlarged with the reduction in their number. In the thawing stage, the ice crystals were converted to water masses easily removed from the polymer  (Smith & Schwartzberg, 1985). Furthermore, water leakage led to the formation of sponge texture in starch. N starch gels started to retrograde following 24 hr, indicated by the increase in the percentage of water from 42.64% to 63.47% during storage (8 cycles). Acetylated starch gels prevented the separation of water from starch noodle even after five freeze-thaw treatments, showing a better freeze-thaw stability due to high DS. The percentage of water separated from AC starch gels at frequencies 25 kHz, 40 kHz, and 25 + 40 kHz reached 1.10%, 3.36, and 0.39%, respectively, indicating the lower increment in syneresis compared with the N wheat starch gel, as illustrated in Figure 6. The highest resistance to freeze-thaw cycle(s) was observed in AC wheat starch at frequency 25 + 40 kHz (lowest % separated water), while the lowest belonged to N wheat starch (highest % separated water). Syneresis in freeze-thawed gels is due to the reassociation of amylose molecules together and with amylopectin chain in starch granules at reduced temperatures which accelerates the exclusion of water from the gel structure. Amylose association takes place at the initial step of storage, while amylopectin association occurs in the following stages. Retrogradation may occur when a gelatinized starch is cooled. The increase in the amount of separated water indicates the lack of freeze-thaw stability. After acetylation, water molecules were absorbed by functional groups and were not able to freeze, thereby reducing syneresis (Han et al., 2012). On the other hand, the introduction of large groups of acetyl into the starch chains reduced the inner and outer bonds. Moreover, the acetyl group is smaller in size, which is effective on water absorption and stability against water exclusion.
By inserting the acetyl group, the hydrophilic nature of these reagents resulted in the electrostatic repulsion between the chains, preventing the amylose-amylose and amylose-amylopectin interactions (Kaur et al., 2004;Lawal, 2010). When the amylose molecules carrying acetyl group are mingled with amylopectin, the acetyl group sterically prevents the aggregation of amylopectin, entailing lower retrogradation and reduction in the percentage of water separated during freezethaw cycles (Kaur et al., 2004;Lawal, 2010).

| CON CLUS IONS
In order to increase the DS and consumption of minimum reagents, a reaction mixture was carried out under ultrasonic processes. The molar substitution optimizing AC wheat starch was 1.22%, 1.37%, and 1.02% for 25 kHz, 25 + 40 kHz, and 40 kHz, consuming 5% of acetic anhydride. High DS is due to the ultrasonic conditions at dual frequency (25 + 40 kHz) which form fractures, pores, and cracks on the surface of granules, facilitating the penetration of reagents into granule starch. Acetylation ameliorates the solubility, water absorption, resistance to freeze and thaw cycle, and viscosity peak, and reduces the gelatinization temperature.

ACK N OWLED G M ENTS
The authors would like to thank Dr. Mohsen Ajdari, University of Fasa, for his technical assistance.

CO N FLI C T O F I NTE R E S T
The authors declare that they do not have any conflict of interest.

E TH I C S S TATEM ENT
Human and animal testing is unnecessary in this study.