Effects of Citric Acid Concentration and Hydrolysis Time on Gembili (Dioscorea esculenta) Starch from Indonesia

Abstract

In this study, we investigated the effects of various citric acid concentrations and hydrolysis times on the physical and chemical characteristics of gembili (Dioscorea esculenta) starch. The gembili starch was hydrolyzed using citric acid at concentrations of 0.1, 0.2, and 0.3 M for 6, 12, 18, and 24 h. The recovery yield was between 90.67% and 96.03%, with the highest amylose and resistant starch, following starch hydrolysis, at a citric acid concentration of 0.1 M and hydrolysis time of 12 h (C1-12). Starch hydrolysis using 0.1 M citric acid increased the oil holding capacity, while the water holding capacity remained the same for all treatments. During hydrolysis at concentrations of 0.2 and 0.3 M, solubility increased, whereas swelling power decreased. The L* value and whiteness index increased at citric acid concentrations of 0.2 and 0.3 M, respectively. The starch exhibited type-A diffraction patterns and polygonal morphology in all treatments. Pasting properties analysis indicated no significant differences with the same citric acid concentrations, except for pasting temperature observations with the value between 77.65°C to 80.95°C. Thermal properties analysis revealed that the onset temperature value was significantly different in the native and citric acid-hydrolyzed starches. Particle size distribution analysis indicated a wider range of particle sizes at increasing concentrations of citric acid. Fourier transform infrared spectroscopy analysis indicated a lack of new peaks after the modifications. The results of this study provide information on the effects of acid hydrolysis on the functional properties and health benefits of gembili starch.

INTRODUCTION

Plant parts, such as tubers, roots, and seeds, contain starch, a source of carbohydrate. Starch is a staple food and an energy source (Wu and Zhou, 2018). Starch granules contain two homopolymers: amylose and amylopectin. Amylose is a linear polymer composed of α-D-glucose units connected by α-D-(1→4) bonds. Amylopectin is a branched polymer that consists of α-D-(1→4) bonds and branching α-D-(1→6) bonds (Bertolini, 2009). Generally, starch consists of 25% to 30% amylose and 70% to 75% amylopectin (Zhao et al., 2018). The amylose and amylopectin content depends on the source of the starch. For example, arrowroot starch contains 24.64% amylose and 73.46% amylopectin (Faridah et al., 2013), whereas Dioscorea alata L. (purple yam) contains 25.77% amylose and 74.23% amylopectin (Oliveira et al., 2021). The structure and function of native starch vary greatly according to the starch source, as do the physicochemical and functional properties (Kim et al., 2020).

Dioscorea, a genus in the Dioscoreaceae family, contains more than 600 species. Dioscorea esculenta (gembili) is commonly found on many islands in Indonesia. Gembili is a nutritional food and a medicinal resource, providing carbohydrates and bioactive substances (Antonio and Buot Jr, 2021; Silalahi, 2022). Gembili can be used as a starch source, with the starch content varying between 80.9% and 82.8%, depending on the processing methods employed (Wanasundera and Ravindran, 1992). The total starch concentrations in gembili flour and starch are 74.66% and 86.3%, respectively (Senanayake et al., 2013).

Many studies have explained the physicochemical characteristics and molecular structure of gembili (Jayakody et al., 2007; Shi et al., 2019). Dioscorea starch has many applications in various processing situations, including as an edible coating material (Alam et al., 2019), as a binder or disintegrant in tablet and granule formulations for oral administration (Riley et al., 2006), and as a component in the production of maltodextrin (Shalihin et al., 2024). Other studies have investigated the modification of gembili starch via carboxymethylation (Nattapulwat et al., 2009), changes in its characteristics following heat moisture treatment and annealing (Ayuningtyas et al., 2021), and the effectiveness of native and carboxymethyl-modified starch as tablet disintegrants (Nattapulwat et al., 2008). However, few studies have described the properties of gembili starch and investigated the impact of different concentrations and times of hydrolysis using citric acid.

Acid hydrolysis and esterification are common chemical methods used in starch modification. Treating the suspension of starch as an organic or inorganic acid solution involves compulsion, filtration, neutralization, washing, and drying. The process typically uses a temperature lower than the gelatinization temperature of the starch (Zhang et al., 2023). Acid hydrolysis cuts amylopectin branches, forms linear chains, and increases the amylose content (Abdorreza et al., 2012). Amylose and amylopectin undergo simultaneous attacks during the initial phase of hydrolysis due to their location on the surface of the starch granules (Chen et al., 2017). During the hydrolysis process, the amorphous area degrades rapidly, and hydrolysis occur gradually in the crystalline area (Zhang et al., 2023). Hydrolysis also affects the structure of the starch. Different structures of starch result in different gelatinization, pasting, and digestive properties (Guo et al., 2023).

A previous study of Tacca leontopetaloides starch, wherein hydrolysis with 0.2 M citric acid was employed, reported significant improvements in the resistant starch and amylose contents after hydrolysis (Nurhayati et al., 2022). However, the effect of citric acid concentration and hydrolysis time on gembili starch has not been reported. This study aimed to determine the influence of citric acid concentration and hydrolysis time on the physicochemical characteristics of gembili starch. The results provide information related to the changes in the structure of gembili starch caused by citric acid hydrolysis to produce starch with desired functions and properties.

MATERIALS AND METHODS

Materials

D. esculenta, known locally as gembili, was harvested between August and October 2022 from a local farm in Kulon Progo, Special Region of Yogyakarta, Indonesia. The chemical reagents used in the analyses were of analytical grade. For citric acid, sodium hydroxide, sulfuric acid, n-hexane, bromcresol green-methyl red, boric acid, hydrochloric acid, acetic acid, ethanol, iodine, maleic acid, calcium chloride dihydrate, ammonium sulphate, and amylose for standart were acquired from Merck (Germany) and Megazyme assay kit K-RAPRS 11/19 for resistant starch analysis from Megazyme (Ireland).

Starch extraction

Starch extraction followed the method of Mutmainah et al. (2021) with some modifications. In brief, gembili was washed, peeled, and cut into small pieces. The pieces were then soaked in 5% NaCl solution for 30 to 45 min, rinsed, and then ground with water (1:4 w/v) to obtain starch slurry. Grinding was performed using a home blender (BL-152 GF/PF-AP, Miyako). After grinding, the starch slurry was filtered and pressed. The pressed products were left to stand for 12 to 16 h, until the supernatant and precipitate had separated. The supernatant was then discarded and replaced with water (1:2 v/v) twice every 8 h. The wet starch was dried in a cabinet dryer at 50°C for 12 to 24 h. The dry starch was mashed using a blender and passed through an 80-mesh sieve. The resulting product was named “native starch” and kept in a Ziploc bag until further use.

Starch hydrolysis

Starch hydrolysis was performed following the methods described by Nurhayati et al. (2022) with some modifications. In brief, starch hydrolysis was performed at a native starch to citric acid ratio of 1:4 (w/v, db). This approach employed different citric acid concentrations (0.1, 0.2, and 0.3 M) and hydrolysis times (6, 12, 18, and 24 h). The native starch and citric acid solution mixture was stirred at room temperature (±25°C). After each hydrolysis step, the reaction was neutralized with 1 M NaOH to adjust the pH to 7 and then stored at 4°C for 24 h. The hydrolyzed starch was dried at 50°C for 24 h in a cabinet dryer. The dry starch was then mashed using a blender and passed through an 80-mesh sieve. The recovery yield was calculated using the following equation:

$$\text{Recovery yield (\%)=}\frac{\text{Starch weight after hydrolysis}}{\text{Starch weight before hydrolysis}}\times 100$$

Proximate analysis

Proximate analysis was performed according to the AOAC International (2000) method to determine the moisture, crude protein, crude lipid, and crude ash contents. Carbohydrate content was determined by subtracting the total percentages of moisture, protein, fat, and ash.

Amylose analysis

The amylose content was analyzed according to the method described by Sonia et al. (2019). In brief, a 100 mg sample was dissolved in 1 mL of 95% ethanol and 9 mL of 1 N NaOH and then heated for 10 min until a gel formed. The compound was then transferred to a measuring flask and adjusted to 100 mL. Then, a 5 mL starch solution was added with 1 mL of 1 N acetic acid and 2 mL of iodine reagent, and the solution was adjusted to 100 mL with distilled water. The solution was incubated for 20 min and then measured at a wavelength of 625 nm.

Resistant starch analysis

Resistant starch was measured using the Megazyme assay kit (K-RAPRS 11/19) based on the AOAC Method 2002.02 and AACC Method 32-40 (McCleary et al., 2020).

Water and oil holding capacities

The measurement of the water holding capacity (WHC) and oil holding capacity (OHC) followed the method of Raza et al. (2021). In brief, 10 mL of water or sunflower oil was added to 2 g of the sample, then the sample mixture was homogenized in a vortex for 5 min, followed by centrifugation at 3,000 g for 10 min at room temperature (±25°C). After centrifugation, the supernatant was removed, and the wet weight was calculated. The equations used to calculate WHC and OHC were as follows:

$$\text{WHC or OHC (g/g)=}\frac{\text{Wet weight-dry weight}}{\text{Dry weight}}$$

Swelling power and solubility

The swelling power and solubility were determined as described by da Rosa Zavareze et al. (2010) with slight modifications. In brief, 0.2 g of sample and 10 mL of distilled water were added to a centrifuge tube. Samples were initially homogenized and heated at different temperatures (60°C, 70°C, and 80°C) for 30 min. The samples were manually homogenized every 5 min, and then the heated samples were cooled at room temperature and centrifuged (DLAB-DMO636, DLAB Scientific Instrument LLC) at 1,000 g for 20 min. The supernatants from the sample were transferred to a weighing bottle and dried at 105°C to 110°C until they reached a constant weight. The swelling power and solubility were calculated using the following equations:

$$\begin{array}{l}\text{Swelling power (g/g)=}\frac{\text{Weight of pellet}}{\text{Weight of the initial sample}}\\ \text{Solubility (\%)=}\frac{\text{Weight of dried supernant}}{\text{Weight of the initial sample}}\times 100\end{array}$$

Color

The color of the starch was determined using a chromameter (CR-400, Konica Minolta). The analysis was conducted by observing five areas. The color values were using the CIE-LAB system, where L* represents the lightness of the sample, a* represents the redness to greenness of the sample, and b* represents the yellowness to blueness. The whiteness index was calculated in accordance with Ratnaningsih et al. (2020) using the following equation:

$$\text{Whiteness index=100-}\sqrt{{\left(\text{100-L*}\right)}^{\text{2}}{\text{+a*}}^{\text{2}}{\text{+b*}}^{\text{2}}}$$

X-ray diffraction

Starch analysis via X-ray diffraction (XRD) (D8 Advance, Bruker AXS) was performed using the method described by Zhang et al. (2021). The sample was scanned with a diffraction angle of 2θ. XRD was then operated at 40 kV and 30 mA with a scanning speed of 4°/min.

Pasting properties

Starch pasting properties were determined using a Rapid Visco Analyzer (RVA 4500, Perten Instruments). The sample was dissolved in distilled water, and the moisture content of the starch was corrected to 14%. The stirring speeds were 960 and 160 rpm for 20 and 50 s, respectively. The starch solution was then heated from 50°C to 95°C at a heating rate of 5.2°C/min. The starch was held at 95°C for 5 min, cooled to 50°C at a rate of 5.2°C/min, and then held at 50°C for 2 min. The pasting temperature (PT), peak viscosity (PV), breakdown viscosity (BV), final viscosity (FV), and setback viscosity (SV) were then measured.

Thermal properties

A differential scanning calorimeter (DSC) (DSC-60Plus, Shimadzu) and TA-60WS collection monitor software were used to determine the thermal properties of the starch. First, 2 mg of the starch sample was placed in a pan. Distilled water (10 µL) was then added, and the sample was heated from 30°C up to 100°C at a rate of 10°C/min. The onset temperature (To), peak temperature (Tp), final temperature (Tc), and temperature range (Tc－ To) were noted.

Morphological properties

Scanning electron microscopy (SEM) (JEOL JSM-6510LA, Jeol Ltd.) and a JEOL JEC-3000FC autocoater were used to investigate the morphological properties of the starch. Starch samples were placed in a stub specimen, covered with carbon tape, and coated with a layer of gold. The gold coating technique involved placing a sample in the autocoater at 3.2 Pa for 120 s. The morphological characteristics were observed at 10 kV using potential acceleration.

Particle size distribution

The particle size distribution of the starch was measured using a particle size analyzer (PSA) (Mastersizer 3000, Malvern). The data from the PSA were reported at D10, D50, and D90.

Fourier transform infrared

The sample spectra were read using fourier transform infrared spectroscopy (FTIR) (Nicolet iS10 FTIR Spectrophotometer, Thermo Scientific), and the detector type was deuterated triglycine sulfate. Sample preparation involved the mixture of 1 mg of the sample and 100 mg of potassium bromide (KBr). The mixture was formed into pellets and then measured at wavelength of 4000－400 cm⁻¹ at resolutions of 8 cm⁻¹.

Experimental design and statistical analysis

The starch samples were evaluated with a 3×4 factorial experimental design to examine the effects of citric acid concentration and hydrolysis time. The first factor consisted of three different concentrations of citric acid (0.1, 0.2, and 0.3 M); the second factor consisted of four different hydrolysis times (6, 12, 18, and 24 h). Data were expressed as mean±standard deviation. Data were subjected to analysis of variance and Duncan’s multiple range test to determine the significance of the differences. A significance level of P<0.05 was used. Statistical analyses were conducted using SPSS software (ver.25, IBM Corp.).

RESULTS AND DISCUSSION

Recovery yield and chemical composition

The recovery yield of the starch hydrolysis using citric acid varied between 90.67% and 96.03% (Fig. 1A). The results indicated no significant differences among the hydrolysis treatments; however, a significant difference was observed compared with the native starch. A previous study showed that sweet potato starch displayed a recovery yield range of 85.05% to 92.93% (Babu et al., 2015a) and sago starch displayed a range of 94.3% to 99.2% (Abdorreza et al., 2012). Table 1 presents the results of the proximate analysis of the native and hydrolyzed starches. The low levels of lipid, protein, and ash indicate a high level of purity in the native and hydrolyzed starches. Starch hydrolysis with 0.1 M citric acid exhibited significant differences from the native starch in terms of lipid, protein, ash, and carbohydrate contents. The moisture contents of the native and citric acid-hydrolyzed starch ranged from 4.67% to 9.97%, demonstrating a reduction with the increase in citric acid at concentrations of 0.2 and 0.3 M, Babu et al. (2015a) revealing that the moisture content of the native and hydrolyzed starches ranged from 6.62% to 14.11%. The addition of citric acid to starch decreased the moisture content. Moreover, the interaction between the hydroxyl groups of glucose in starch and the hydroxyl group in citric acid contributes to the reduction of the moisture content.

Fig. 1.

(A) Recovery yield, (B) oil holding capacity (OHC) and water holding capacity (WHC), (C) solubility, and (D) swelling power of native and hydrolyzed starch samples. Values are presented as mean±SD. Nat, native starch; C1, citric acid hydrolysis concentration of 0.1 M; C2, citric acid hydrolysis concentration of 0.2 M; C3, citric acid hydrolysis concentration of 0.3 M. 6, hydrolysis time of 6 h; 12, hydrolysis time of 12 h; 18, hydrolysis time of 18 h; 24, hydrolysis time of 24 h. Lowercase letters indicate significant differences between citric acid concentrations (P<0.05). Uppercase letters show significant differences between hydrolysis times (P<0.05). Native starch was used as a control for citric acid concentrations and hydrolysis times, respectively.  

Table 1.

Chemical composition of native and hydrolyzed starches

Sample	Moisture	Crude lipid (%db)	Crude protein (%db)	Crude ash (%db)	Carbohydrate	Amylose (%db)	Resistant starch (%db)
Nat	9.97±2.01bB	0.51±0.04aA	0.98±0.05bA	0.35±0.17aA	88.21±1.84cC	18.83±0.04aA	39.46±13.53aA
C1-6	9.04±2.18bA	1.17±0.40bAB	1.23±0.06cA	2.33±1.59bB	86.24±0.13aC	19.73±0.41bA	48.04±6.36bA
C1-12	9.44±1.88bA	1.16±0.25bAB	1.11±0.45cA	2.59±1.95bB	85.70±0.76aC	22.16±0.67bA	52.88±0.21bA
C1-18	9.12±1.42bA	0.96±0.66bAB	1.93±0.89cA	2.28±1.76bC	85.70±0.10aA	22.08±0.15bA	50.13±0.75bA
C1-24	9.05±2.35bA	1.39±0.16bB	2.02±0.32cA	2.28±1.65bB	85.26±0.22aB	20.31±0.50bA	49.04±4.72bA
C2-6	4.75±0.29aA	0.41±0.03aAB	0.36±0.00aA	5.22±0.05cB	89.26±0.27cC	20.10±3.78abA	41.01±2.03aA
C2-12	4.67±0.35aA	0.48±0.01aAB	0.35±0.01aA	4.41±0.04cB	90.10±0.30cC	20.15±0.16abA	35.59±3.83aA
C2-18	5.23±0.36aA	0.57±0.20aAB	0.52±0.23aA	7.64±0.01cC	86.04±0.08cA	20.39±0.64abA	32.80±2.81aA
C2-24	5.37±0.35aA	0.96±0.05aB	0.31±0.07aA	5.20±0.00cB	88.17±0.32cB	20.14±2.54abA	34.10±0.94aA
C3-6	5.09±0.43aA	0.59±0.01aAB	0.17±0.01aA	6.17±0.01cB	87.98±0.43bC	20.93±0.31abA	37.28±1.89aA
C3-12	5.55±0.42aA	0.18±0.08aAB	0.18±0.00aA	5.46±0.03cB	88.63±0.48bC	20.41±0.40abA	38.67±0.17aA
C3-18	5.84±0.16aA	0.35±0.00aAB	0.18±0.00aA	8.21±0.01cC	85.43±0.17bA	20.62±1.38abA	39.06±1.20aA
C3-24	6.62±0.20aA	0.33±0.04aB	0.18±0.00aA	5.80±0.02cB	87.07±0.18bB	20.79±0.34abA	35.76±0.56aA

Values are presented as mean±SD.

Nat, native starch; C1, citric acid hydrolysis concentration of 0.1 M; C2, citric acid hydrolysis concentration of 0.2 M; C3, citric acid hydrolysis concentration of 0.3 M; 6, hydrolysis time of 6 h; 12, hydrolysis time of 12 h; 18, hydrolysis time of 18 h; 24, hydrolysis time of 24 h.

Lowercase letters in the same column indicate significant differences between citric acid concentrations (P<0.05). Uppercase letters in the same column show significant differences between hydrolysis times (P<0.05). Native starch was used as a control for citric acid concentrations and hydrolysis times, respectively.

The application of citric acid in the hydrolysis treatment affected the synthesis of amylose and resistant starch. A significant difference was observed in the amylose and resistant starch levels between the hydrolyzed starch treated with 0.1 M citric acid and the native starch. However, an elevated citric acid concentration during the hydrolysis process did not result in any significant differences in the amylose and resistant starch values compared with the native starch. Acid hydrolysis breaks amylopectin branches and produces linear chains, thereby increasing the amylose levels (Abdorreza et al., 2012). The acid hydrolysis process reduces amylose levels by breaking down the amorphous components and glycosidic linkages (Akin-Ajani et al., 2014). The use of high concentrations of citric acid for hydrolysis reduced amylose levels. A correlation was noted between the levels of amylose and resistant starch. Increased amylose levels were indicated by increased resistant starch levels. Resistant starch exists in the crystalline area and during acid hydrolysis (Nagahata et al., 2013).

Functional properties

The functional characteristics of gembili starch were significantly altered by citric acid hydrolysis. We evaluated the following functional characteristics in this study: OHC, WHC (Fig. 1B), solubility, and swelling power (Fig. 1C and 1D). The OHC values for the native and hydrolyzed starch ranged from 1.35 to 1.97 g/g, and there was a significant difference between the starch hydrolyzed with 0.1 M citric acid and the native starch. The native starch of D. esculenta has an OHC value of 1.86 mL/g (Chiranthika et al., 2022). The OHC is associated with additional chemical components such as fibers, which increase the hydrophobic characteristics of a material (Rodríguez-Ambriz et al., 2008). The WHC values for the native and hydrolyzed starches ranged from 0.9 to 1.34 g/g and no significant differences in WHC were observed between the native and hydrolyzed starches. The WHC of starch is affected by several factors, including the granule shape, amylose content, component source, and treatment methods (Alqah et al., 2022).

The solubility and swelling power of all the samples increased in line with the temperature. When heating above the gelatinization temperature, hydrogen bonds are broken, allowing the entrance of water into the granules and the hydration of free hydroxyl groups (da Rosa Zavareze et al., 2010). The broken hydrogen bonds allow for greater interaction between the starch chains in the amorphous and crystalline regions. This results in the absorption of more water, and the starch granules become swollen, which leads to the solubilization and leaching out of amylose, the final result when solubility is enhanced (Kumoro et al., 2019). At concentrations of 0.2 and 0.3 M citric acid, the samples showed higher solubility than native starch and starch hydrolyzed at a concentration of 0.1 M citric acid but had low swelling power values. Acid treatment can damage the microcrystalline structure of the starch, resulting in changes in the intermolecular structure, amylose leaching, and degradation of amylopectin, which increase the solubility. Amylose leaching after acid treatment causes starch damage and decreased swelling power values (Li et al., 2020). Similar results were shown in studies of starches from banana, lotus stem, sweet potato, and wheat, which all exhibited increased solubility and decreased swelling power after acid modification (Kaur et al., 2011). Citric acid modification to several types of tubers, including white yam, water yam, yellow yam, and bitter yam, when compared to their native starch, showed similar results (Falade and Ayetigbo, 2017).

Color analysis

Color observation is an important parameter related to quality and consumer acceptance. The results show that the L*, a*, and b* values and the whiteness index are important parameters (Table 2). The L* value ranged from 91.77 to 98.58, and there was no significant difference in the L* value between the starch hydrolyzed with 0.1 M citric acid and native starch. A previous study reported that the L* value of the native starch from Dioscorea sp. is 98.1 and that of D. alata is 98.12 (Hornung et al., 2018; Oliveira et al., 2021). No significant differences in the value of a* were observed between the native and hydrolyzed starches. The b* values of the native starch were different than those of the starch hydrolyzed with 0.1 M citric acid. The modification procedure induced changes in the values of L*, a*, and b* of the samples, which were related to the purification and separation of starch components, including proteins, fiber, sugar, salt, and other compounds (Falade and Ayetigbo, 2015). The whiteness index exhibited a significant difference between the native starch and the starches hydrolyzed at citric acid concentrations of 0.2 and 0.3 M.

Table 2.

Color of native and hydrolyzed starches

Sample	L*	a*	b*	Whiteness index (%)
Nat	91.77±1.02aA	—5.21±2.24aA	9.08±1.04bA	86.56±0.95aA
C1-6	91.98±1.05aB	—3.78±2.64aA	8.39±0.09aA	87.63±0.06aB
C1-12	92.04±0.74aB	—3.95±2.60aA	8.24±0.20aA	87.73±0.22aB
C1-18	92.21±0.81aB	—4.13±2.55aA	8.39±0.12aA	87.69±0.26aB
C1-24	92.34±0.70aB	—4.34±2.37aA	8.52±0.05aA	87.63±0.43aB
C2-6	98.25±0.66bB	—6.41±1.23aA	8.73±0.29abA	89.00±0.84bB
C2-12	97.90±1.08bB	—6.59±1.32aA	8.56±0.36abA	88.94±0.86bB
C2-18	98.40±0.52bB	—6.41±1.76aA	8.90±0.16abA	88.86±1.07bB
C2-24	98.54±0.36bB	—6.44±1.85aA	8.87±0.14abA	88.89±1.14bB
C3-6	98.57±0.41bB	—6.62±1.47aA	8.85±0.08abA	88.82±0.88bB
C3-12	98.41±0.35bB	—6.42±1.51aA	8.90±0.04abA	88.87±0.85bB
C3-18	98.29±0.10bB	—6.31±1.47aA	8.93±0.08abA	88.90±0.88bB
C3-24	98.58±0.20bB	—6.08±1.76aA	8.73±0.11abA	89.22±1.05bB

Values are presented as mean±SD.

Nat, native starch; C1, citric acid hydrolysis concentration of 0.1 M; C2, citric acid hydrolysis concentration of 0.2 M; C3, citric acid hydrolysis concentration of 0.3 M; 6, hydrolysis time of 6 h; 12, hydrolysis time of 12 h; 18, hydrolysis time of 18 h; 24, hydrolysis time of 24 h; L*, lightness; a*, redness to greenness; b*, yellowness to blueness.

Lowercase letters in the same column indicate significant differences between citric acid concentrations (P<0.05). Uppercase letters in the same column show significant differences between hydrolysis times (P<0.05). Native starch was used as a control for citric acid concentrations and hydrolysis times, respectively.

X-ray diffraction

The diffraction and relative crystallinity of the starch samples were measured using XRD (Raza et al., 2021). In addition to a double peak at approximately 17° and 18° 2θ, the type-A starch crystals showed a strong diffraction peak between 15° and 23° 2θ. The type-B starch crystals also demonstrated a strong diffraction peak at approximately 17° 2θ, and several small peaks at approximately 15°, 20°, 22°, and 24° 2θ, with a characteristic peak at 5.6° 2θ. The type-C crystals are a combination of type-A and type-B crystalline structures (Shi et al., 2019). The XRD patterns of the native and hydrolyzed starches showed no significant differences (Fig. 2). Similar results were obtained in sweet potato starch modified with acid at several different concentrations, but the peak intensities and 2θ values showed clear variations between the native and acid-modified starch. The addition of citric acid can increase the 2θ value in sweet potato starch (Babu et al., 2015b). The particles in the native and hydrolyzed starches exhibited type-A diffraction patterns. In addition to the small diffraction peaks at 17.9°, these patterns are characterized by strong peaks at 2θ values of approximately 15°, 17°, and 23°. A slightly sharp peak was observed in the diffraction peaks at 17° after acid hydrolysis of gembili starch. Acid-modified sorghum starch retains the type-A diffraction pattern and sharp peaks appear at 2θ=17.5° and 23.5° (Olayinka et al., 2013). Type-A pathways have strong peak fractions at approximately 15° and 23° along with incomplete peaks at approximately 17° and 18° (Zhai et al., 2017). Different results were obtained from yam (D. alata L.) starch with a type-B diffraction pattern, revealing the presence of five peaks at 2θ (5.63°, 15.5°, 17.10°, 22.29°, and 24.03°) (Oliveira et al., 2021) as well as starch from D. opposita Thunb., which exhibited a typical type-C diffraction pattern and showed diffraction peaks at 15°, 17°, and 22° of 2θ (Qian et al., 2019). D. opposita Thunb. flour and starch have type-A XRD patterns, with strong peaks at 2θ=15.3°, 17.1°, 18.2°, and 23.5° (Yu et al., 2021). Chinese yam starch exhibit strong peaks at 15°, 17°, and 23° (2θ). Variations in the yams can influence the structure of starch crystallization (Zou et al., 2022).

Fig. 2.

The type-A X-ray diffraction patterns of native and hydrolyzed starches. Nat, native starch; C1, citric acid hydrolysis concentration of 0.1 M; C2, citric acid hydrolysis concentration of 0.2 M; C3, citric acid hydrolysis concentration of 0.3 M; 6, hydrolysis time of 6 h; 12, hydrolysis time of 12 h; 18, hydrolysis time of 18 h; 24, hydrolysis time of 24 h.

Pasting properties

Table 3 provides a summary of the pasting properties of the native and hydrolyzed starches. The observed results were the PT, PV, BV, FV, and SV. The pasting properties of starch can be influenced by the source of the plant, as well as its purity, granule size, shape, amylose, proteins, lipids, and fiber contents, distribution of the chain length of amylopectin, and the interactions between the existing components in the starch (Oliveira et al., 2021). The PT, PV, BV, FV, and SV did not show significant differences when the citric acid concentration remained the same and the hydrolysis duration varied. However, the PT of the native starch and the starch hydrolyzed at citric acid concentrations of 0.2 and 0.3 M differed significantly. The PT values improved after the implementation of the acid hydrolysis process. Hydrolysis facilitates the reorientation of molecules or particles, resulting in the production of starch with remarkable organization. The strengthening of the intragranular bonded forces increases the heat requirement of starch before structural disintegration and paste formation (Sun et al., 2022).

Table 3.

Paste viscosities of native and hydrolyzed starches

Sample	PT (°C)	PV (cP)	BV (cP)	FV (cP)	SV (cP)
Nat	77.65±0.00aA	6,389.00±56.57cB	3,155.50±79.90cB	4,400.00±26.87bB	1,166.50±3.54aA
C1-6	78.68±0.04aBC	4,803.00±108.89bA	1,925.50±286.38bA	4,342.00±43.48bA	1,464.50±133.64bA
C1-12	78.98±0.25aBC	5,049.50±185.97bA	1,733.00±309.71bA	4,881.50±399.52bA	1,565.00±96.17bA
C1-18	78.35±0.57aC	4,877.00±653.37bA	2,084.50±127.99bA	4,402.50±726.20bA	1,610.00±200.82bA
C1-24	78.03±0.46aB	5,050.00±919.24bA	2,452.00±65.05bA	4,390.50±963.79bA	1,792.50±109.60bA
C2-6	79.65±0.78bBC	3,744.00±575.58aA	1,844.00±441.23aA	3,251.50±867.62aA	1,351.50±149.20aA
C2-12	78.93±0.25bBC	3,602.00±230.52aA	2,007.50±143.54aA	2,972.00±241.83aA	1,377.50±132.23aA
C2-18	80.75±1.27bC	2,682.00±644.88aA	1,234.50±816.71aA	2,423.00±214.96aA	975.50±386.79aA
C2-24	78.53±0.32bB	3,841.00±135.76aA	807.00±8.49aA	3,805.00±91.92aA	771.00±35.36aA
C3-6	80.03±0.25bBC	3,703.00±325.27aA	1,244.50±490.02aA	3,265.50±307.59aA	807.00±472.35aA
C3-12	79.85±0.00bBC	3,334.00±77.78aA	790.50±30.41aA	3,223.50±10.61aA	680.00±97.58aA
C3-18	80.95±0.49bC	3,140.00±609.53aA	1,340.00±89.10aA	2,980.50±201.53aA	1,180.50±497.10aA
C3-24	79.98±1.31bB	2,809.00±483.66aA	1,301.50±132.23aA	2,482.50±171.83aA	975.00±179.61aA

Values are presented as mean±SD.

Nat, native starch; C1, citric acid hydrolysis concentration of 0.1 M; C2, citric acid hydrolysis concentration of 0.2 M; C3, citric acid hydrolysis concentration of 0.3 M; 6, hydrolysis time of 6 h; 12, hydrolysis time of 12 h; 18, hydrolysis time of 18 h; 24, hydrolysis time of 24 h; PT, pasting temperature; PV, peak viscosity; BV, breakdown viscosity; FV, final viscosity; SV, setback viscosity.

Lowercase letters in the same column indicate significant differences between citric acid concentrations (P<0.05). Uppercase letters in the same column show significant differences between hydrolysis times (P<0.05). Native starch was used as a control for citric acid concentrations and hydrolysis times, respectively.

The values of PV, BV, and FV decreased after hydrolysis with citric acid. The results indicated no significant differences in the hydrolysis results at the concentrations of 0.2 and 0.3 M. Similarly, the duration of the hydrolysis process did not exhibit significant variations. Acid concentrations and hydrolysis times affect the reduction in PV, BV, and FV values, as shown in the study conducted by Babu et al. (2015a). The decrease in the PV values is associated with the hydrolysis in the non-crystalline regions and the formation of low-molecular-weight dextrin (Singh et al., 2009), which also reduces the swelling power (Mehboob et al., 2015). The reduction in the BV values was attributed to the reorientation of starch during the hydrolysis process and the changes in the amorphous and crystalline regions after acid modification. The decrease in the FV values correlates with the loss of amylose during the pasting process (Falade and Ayetigbo, 2017). Meanwhile, the SV values showed a significant difference when the citric acid concentration reached 0.1 M, but no significant changes were observed at concentrations of 0.2 and 0.3 M when compared with the native starch. No significant differences were conferred by the duration of hydrolysis. SV occurs during the cooling process to facilitate the reassociation of starch molecules, especially amylose. Low SV levels correlate with decreased retrogradation and syneresis (Ragaee and Abdel-Aal, 2006).

Thermal properties

DSC is an important technique for characterizing the gelatinization of starch due to its accuracy, high reproducibility, and simple operation (Shi et al., 2019). DSC data are presented with the transition onset and endset temperatures: the paste viscosity starts to increase at onset, endset is Tc of the viscosity increment, To is the temperature at maximum viscosity, and (Tc－To) is the transition temperature range (Ahmad et al., 2020). Table 4 summarizes the DSC parameters from the native and citric acid-hydrolyzed starches. A significant difference was observed in the To values between the native and citric acid-hydrolyzed starches, with an increase in the value following hydrolysis. There was a significant difference in the Tp value between the native and hydrolyzed starches at concentration of 0.2, and 0.3 M citric acid. The Tc value showed significant differences between the native starch and starch hydrolyzed at concentrations of 0.2 and 0.3 M. Different results were noted for sweet potato starch modified with hydrochloric acid and citric acid at concentrations of 1% and 5%, which demonstrated a decrease in the To value. In contrast, the Tp and Tc values showed the same pattern, increasing in value compared to native starch (Babu et al., 2015b). Sorghum starch modified with acid showed an increase in Tc value at all hydrolysis times, whereas the To and Tp values increased at hydrolysis times of 8 and 16 h (Singh et al., 2009). To, Tc, and Tp values are influenced by various factors associated with changes in the structure of starch granules, such as the interaction of amylose–amylose and amylose-lipid. The presence of these interactions suppresses the mobility of starch chains in amorphous lamellae. Starch that forms crystals and new bonds requires a higher temperature for this process to occur (Barretti et al., 2022). Increased Tc values after the acid hydrolysis process resulted in the disappearance of the amorphous area of the starch granules. This phenomenon is related to the melting of the crystalline area due to high temperatures but does not change the amylopectin content or double helix (Utrilla-Coello et al., 2014). Sweet potato starches that were hydrolyzed with citric acid at high concentrations and for longer periods demonstrated increased Tp and Tc values due to the high rate of amorphous area hydrolysis, which increases the relative crystallinity and gelatinization temperature (Babu et al., 2015a). The temperature range (Tc－To) was significantly different in the native starch compared with the starch hydrolyzed at a citric acid concentration of 0.3 M. The temperature range indicates heterogeneity in the recrystallized amylopectin (Xing et al., 2017).

Table 4.

Thermal properties of native and hydrolyzed starches (unit: °C)

Sample	To	Tp	Tc	Tc－To
Nat	69.64±0.04aA	73.02±0.08aA	75.83±0.14aA	6.20±0.18aA
C1-6	70.20±0.18bB	74.24±0.12aB	77.85±0.34abAB	7.66±0.16abA
C1-12	71.14±0.17bB	75.01±0.49aB	77.99±0.37abB	6.85±0.21abA
C1-18	70.77±0.99bB	74.93±1.14aB	78.06±1.68abB	7.29±0.69abA
C1-24	71.32±0.54bB	75.19±0.18aB	78.26±0.40abB	6.94±0.13abA
C2-6	71.63±0.34cB	78.34±1.20bB	81.48±0.33bcAB	9.85±0.66abA
C2-12	72.09±0.23cB	77.75±0.84bB	80.12±0.11bcB	8.03±0.11abA
C2-18	72.87±0.23cB	79.08±1.97bB	79.73±1.33bcB	6.87±1.10abA
C2-24	71.99±0.89cB	79.59±1.30bB	81.50±2.50bcB	9.51±1.61abA
C3-6	72.05±0.93cB	79.76±3.58bB	79.88±3.73bcAB	7.83±2.80bA
C3-12	71.91±0.57cB	78.42±0.52bB	82.63±4.53cB	10.72±5.10bA
C3-18	72.62±0.41cB	82.80±6.05bB	87.00±6.67cB	14.38±6.26bA
C3-24	72.48±1.51cB	79.38±0.11bB	80.73±1.68cB	8.25±0.16bA

Values are presented as mean±SD.

Nat, native starch; C1, citric acid hydrolysis concentration of 0.1 M; C2, citric acid hydrolysis concentration of 0.2 M; C3, citric acid hydrolysis concentration of 0.3 M; 6, hydrolysis time of 6 h; 12, hydrolysis time of 12 h; 18, hydrolysis time of 18 h; 24, hydrolysis time of 24 h; To, onset temperature; Tp, peak temperature, Tc, final temperature; Tc-To, temperature range.

Lowercase letters in the same column indicate significant differences between citric acid concentrations (P<0.05). Uppercase letters in the same column show significant differences between hydrolysis times (P<0.05). Native starch was used as a control for citric acid concentrations and hydrolysis times, respectively.

Morphological properties and particle size distribution

The morphological characteristics of the starch samples were examined under SEM. The results are presented in Fig. 3. The granule size was analyzed using a PSA, and the results are shown in Table 5. Starches isolated from different botanical sources exhibit specific characteristic granule morphology. The native and hydrolyzed starches of D. esculenta showed polygonal morphologies for all treatments. Similar results were noted by Chiranthika et al. (2022) and Jayakody et al. (2007). The native starch of D. esculenta has a polygonal morphological form with sharp edges. Starch that was hydrolyzed with 0.1 M citric acid showed no significant morphological changes compared with the native starch. However, the surfaces of the starch samples hydrolyzed with 0.2 and 0.3 M citric acid were rougher and the intergranules were more stacked compared to the native starch. On the first day, hydrolysis typically produces only slight surface exo-erosion, which increases after the third day; however, the granules can still be preserved (Utrilla-Coello et al., 2014). Acid hydrolysis processes yield rough surfaces, and amorphous starch degrades in the early stages of hydrolysis (Wang et al., 2017). A rough surface indicates the degradation of starch granules, which affects the functional properties, such as increased solubility. The increased solubility is due to the increase in the amount of amylose out of the granules (Javadian et al., 2021). Table 5 shows the determination of the particle size distribution, with values of D10, D50, and D90, indicating that 10%, 50%, and 90% of the diameters of the particle granules were smaller than these values, respectively. The values of D10, D50, and D90 in the native starch and hydrolyzed starch were 4.91－7.74, 11.06－88.05, and 166.65－ 245.56 µm, respectively. Significant differences were observed in the particle size distribution values between the treatments; a high citric acid concentration induced a large distribution of particle sizes. Quinoa starch hydrolyzed with citric acid at concentrations of 0%, 10%, and 40% and further modified by extrusion showed opposite results, namely a decrease in the particle size distribution value with an increasing citric acid concentration (Rueda et al., 2022). The variety and the growth environment both have impacts on the average particle size of yam starch (Zou et al., 2022). The size distribution of starch can affect the processing of food products, such as the preparation of dough. The particle size distribution of starch affects the properties and microstructure of the dough (Hu et al., 2025).

Fig. 3.

Scanning electron microscopy images of native and hydrolyzed starches. Nat, native starch; C1, citric acid hydrolysis concentration of 0.1 M; C2, citric acid hydrolysis concentration of 0.2 M; C3, citric acid hydrolysis concentration of 0.3 M; 6, hydrolysis time of 6 h; 12, hydrolysis time of 12 h; 18, hydrolysis time of 18 h; 24, hydrolysis time of 24 h.

Table 5.

Particle size distribution of native and hydrolyzed starches (unit: µm)

Sample	D10	D50	D90
Nat	4.91±0.02aA	11.06±0.10aA	166.65±3.17aA
C1-6	6.10±0.02bB	55.38±0.79bC	240.56±1.39bC
C1-12	5.89±0.03bC	36.88±1.24bB	219.57±2.16bB
C1-18	5.77±0.03bB	43.79±1.30bC	217.62±0.24bB
C1-24	5.81±0.01bD	39.18±0.48bD	225.18±1.67bC
C2-6	5.98±0.01cB	53.22±0.87cC	229.91±0.41cC
C2-12	6.25±0.02cC	54.28±0.93cB	225.71±0.73cB
C2-18	6.12±0.02cB	56.81±0.82cC	229.64±1.16cB
C2-24	6.47±0.01cD	61.02±0.11cD	228.98±0.60cC
C3-6	5.95±0.01dB	49.87±0.75dC	227.98±2.08dC
C3-12	6.41±0.04dC	63.12±0.40dB	235.05±1.07dB
C3-18	6.09±0.02dB	58.77±0.62dC	237.72±1.87dB
C3-24	7.74±0.01dD	88.05±0.65dD	245.56±1.06dC

Values are presented as mean±SD.

Nat, native starch; C1, citric acid hydrolysis concentration of 0.1 M; C2, citric acid hydrolysis concentration of 0.2 M; C3, citric acid hydrolysis concentration of 0.3 M; 6, hydrolysis time of 6 h; 12, hydrolysis time of 12 h; 18, hydrolysis time of 18 h; 24, hydrolysis time of 24 h.

Lowercase letters in the same column indicate significant differences between citric acid concentrations (P<0.05). Uppercase letters in the same column show significant differences between hydrolysis times (P<0.05). Native starch was used as a control for citric acid concentrations and hydrolysis times, respectively. D10, D50, and D90 indicate the specific particle sizes in which 10%, 50%, and 90% of the sample particles are smaller, respectively.

FTIR spectroscopy

The FTIR spectroscopy observations (Fig. 4) revealed structural changes at the molecular level (e.g., changes in starch chain conformation, crystallinity, and retrogradation) (Miao et al., 2011). Based on Ayo-Omogie et al. (2022), tell about the wavenumbers less than 1500 cm⁻¹ represented the fingerprint region while wavenumbers from 1500 cm⁻¹ to 4500 cm⁻¹ identify the functional groups. The FTIR results from the native and citric acid-hydrolyzed starches were demonstrated in the 4000－400 cm⁻¹ area, showing several functional groups with bands and peaks. The bands at 3398, 2929, 1639, and 1200－ 800 cm⁻¹ appeared for all treatments. The 3398 cm⁻¹ band indicated hydroxyl groups (Garg and Jana, 2011). The 2930 cm⁻¹ band was assigned to the C-CH₂-C asymmetric stretching vibration (Aalim and Luo, 2021). The vibration of the hydroxyl group that appeared at a wavelength below 3800 cm⁻¹ was attributed to the amylose spectrum of starch. The peak at 2927.10 cm⁻¹ was attributed to the hydrocarbon framework structure of amylose and amylopectin. Cyrtosperma merkusii starch modified with acid and continued NaOH neutralization showed starch decomposition, where the final result was the formation of amylose, sodium citrate, and sodium citrate dihydrate (Dompo et al., 2024). The characteristics that appear on the 1639 cm⁻¹ band relate to the water bound in the starch (Zhai et al., 2017). The 1200－800 cm⁻¹ region is related to C-O, C-C, and C-H stretching vibrations, as well as C-O-H bending modes (Borah et al., 2017). The peaks at 1155 and 1078 cm⁻¹ were attributed to the vibrations of C-O, C-C stretching, and C-O-H bending (Ogunmolasuyi et al., 2016). The peak at 1020 cm⁻¹ was attributed to the stretching vibration of C-O in the C-O-C groups (Ren et al., 2017). The starch hydrolyzed with citric acid at concentrations of 0.2 and 0.3 M for all hydrolysis times exhibited a new band at 1591 cm⁻¹. This band was related to the C=O asymmetric stretching vibration absorption peaks that appeared at 1591, 1585, 1583, and 1587 cm⁻¹ (Sun et al., 2022). In this study, no new peaks appeared after citric acid hydrolysis. Several previous studies have shown that starch hydrolyzed with 10% to 50% citric acid create the emergence of a new peak in the 1720 cm⁻¹ region, which confirms the presence of the carbonyl ester in the crosslink reaction product. Starch samples modified using higher concentrations of citric acid showed greater peak intensities (Utomo et al., 2020).

Fig. 4.

Fourier transform infrared spectra of native and hydrolyzed starches. Nat, native starch; C1, citric acid hydrolysis concentration of 0.1 M; C2, citric acid hydrolysis concentration of 0.2 M; C3, citric acid hydrolysis concentration of 0.3 M; 6, hydrolysis time of 6 h; 12, hydrolysis time of 12 h; 18, hydrolysis time of 18 h; 24, hydrolysis time of 24 h.