Preparation of starch-vitamin E complexes using electron beam irradiation pretreatment in synergy with ultra-high pressure: enhancing physicochemical and functional properties

Abstract

The low water solubility and poor thermal stability of Vitamin E (VE) limit its application in food processing. This study investigated the effects of the synergistic combination of electron beam irradiation (EBI) at doses of 0, 5, and 10 KGy and ultra-high pressure (UHP) at 100, 300, and 500 MPa on the preparation of potato starch-VE complexes. EBI pretreatment significantly reduced the degree of branching and enhanced complexation, antioxidant activity, and resistant starch (RS) content in a dose-dependent manner, whereas increasing UHP pressure progressively diminished these attributes. The combination of 10 KGy EBI with 100 MPa UHP yielded the highest RS content (67.24%) and DPPH scavenging activity (85.02%). XRD analysis indicated no V-type crystalline formation, suggesting weak but sustained complexation. The parameter-dependent effects of EBI and UHP offers a promising approach to produce starch-VE complexes with improved antioxidant capacity and digestion resistance, benefiting functional food development for diabetics.

Highlights

• •Starch-vitamin E complexes formed by two combined non-thermal techniques.
• •Complexation degree decreased as pressure increased from 100 to 500 MPa.
• •10 KGy irradiation plus 100 MPa treatment achieved optimal effect.
• •Non-inclusion complexes with no V-type crystalline structure detected.

1. Introduction

Starch, a major dietary carbohydrate, undergoes digestion, absorption and metabolism processes that are closely linked to chronic diseases such as diabetes (Lei et al., 2023). Potatoes, as a staple food, contain relatively high levels of resistant starch (RS), making them more suitable for diabetic diets (Wang et al., 2020). However, the RS content in potato starch remains insufficient to fully meet the dietary requirements of diabetic individuals (Rivas et al., 2023), thus highlighting the need for modification strategies to enhance RS levels. Research has shown that starch can interact with small molecules, such as fatty acids and polyphenols, through distinct mechanisms: (1) forming V-type crystals with amylose and (2) binding via non-covalent interactions, which can inhibit enzymatic activity, increase RS content and reduce digestibility (Wu et al., 2024). Therefore, complexation with small molecules presents a promising approach to further increase the RS content of starch.

Vitamin E (VE) is an indispensable fat-soluble vitamin that supports vital physiological functions in humans (Ribeiro, Estevinho, & Rocha, 2021; Wijekoon, Mahmood, Ariffin, Nafchi, & Zulkurnain, 2023). It has also been shown to alleviate diabetes-related complications and improve metabolic health (Dass, Narayana, & Venkatarathnamma, 2018). However, VE suffers from poor stability and is highly susceptible to oxidation, light and heat, with these limitations not only hindering its application in the food industry but also reducing its effectiveness in biological systems (Wijekoon et al., 2023). The development of food delivery systems for encapsulating VE enhances its solubility and stability, while the preparation of starch-VE complexes achieves this by not only increasing RS content but also effectively dispersing VE within the starch. However, heat treatment has been shown to reduce the VE content and antioxidant activity (Yu et al., 2023). Therefore, non-thermal processing methods are crucial for the preparation of starch-VE complexes to preserve antioxidant activity and increase RS content.

UHP treatment represents a promising non-thermal food processing technology that has gained attention in starch modification research. Indeed, compared with thermal treatments, ultrasound and microwave methods, UHP offers several advantages, including lower energy consumption, shorter processing time, more uniform treatment and better preservation of heat-sensitive compounds (Li & Zhu, 2018; Wang et al., 2025). Nonetheless, UHP alone has certain limitations, such as a narrow application scope and suboptimal effectiveness for certain starch modifications (Wang et al., 2025). Consequently, combining UHP with other modification techniques to achieve synergistic effects has become a research focus. Notably, electron beam irradiation (EBI) is an emerging technology that offers multiple benefits such as rapid and efficient processing, high flexibility and safety, environmental friendliness, simple operation as well as low cost (Lei et al., 2023). Moreover, EBI treatment can break glycosidic bonds, degrade amylopectin and enhance solubility and amylose content (Zhao et al., 2023; Zhao et al., 2023), which are beneficial for the preparation of starch-VE complexes. Compared with traditional radioisotope-based irradiation, EBI is electrically driven, requires no radioactive source, generates no radioactive waste, and its radiation ceases immediately upon power-off; its ultra-high dose rate significantly shortens processing time and reduces energy consumption. In contrast, conventional irradiation relies on expensive isotope sources that require periodic replacement, accompanied by heavy shielding and long-term waste management, resulting in high operational and maintenance costs. EBI offers clear advantages in environmental friendliness, safety, energy efficiency, and cost, making it more suitable for green starch modification.

Research on the preparation of starch-VE complexes using non-thermal treatments is limited, particularly regarding the use of EBI as a starch pretreatment followed by UHP processing to form the complexes. Therefore, this study proposes a dual non-thermal treatment approach combining EBI and UHP for the preparation of starch-VE complexes, aiming to systematically evaluate the synergistic effects of these two technologies on the structural, rheological, antioxidant, and digestive properties of the complexes. The overarching goal is to develop a dual-functional starch-based ingredient that simultaneously enhances VE stability and increases RS content, thereby offering new insights into functional food development for diabetic populations.

2. Materials and methods

2.1. Materials

Native potato starch (biochemical grade) was purchased from Yuanye Biotechnology Co., Ltd. (Shanghai, China), while VE (purity ≥95.0%) and 2,2-diphenyl-1-picrylhydrazyl (DPPH, purity ≥98.5%) were provided by Maclin Biochemical Technology Co., Ltd. (Shanghai, China). In addition, α-amylase (≥ 2000 U·g⁻¹), pepsin (porcine gastric mucosa, USP grade, 1:3000), trypsin (biochemical reagent, 1:4000) and amyloglucosidase (biochemical reagent, 100,000 U·mL⁻¹) were obtained from Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). All other chemicals used in this study were of analytical grade.

2.2. Methods

2.2.1. EBI treatment

Native potato starch in dry powder form was uniformly spread in polyethylene bags (thickness approximately 2–3 mm) and subjected to EBI at doses of 0, 5, and 10 KGy using a 10 MeV/20 kW high-energy electron linear accelerator (Yangling Hesheng Radiation Technology Co., Ltd., Shaanxi, China). The irradiation was performed at an accelerating voltage of 21.3 MeV, beam power of 19 kW, and pulse repetition frequency of 452 Hz. The absorbed dose was measured using potassium dichromate dosimeters calibrated against the National Institute of Metrology, China (Certificate No. DLjs2025–01404, expanded uncertainty 4.0%, k = 2), with dosimeters placed on both sides of the bags to ensure uniformity. Samples (average weight 200 g) were sealed in HDPE bags and positioned within the homogeneous irradiation zone at a distance of 52–54 cm from the scanning window under ambient conditions. The resulting samples were designated as NS (0 KGy), 5 KGy, and 10 KGy, respectively.

2.2.2. Degree of branching (DB)

The degree of branching of the NS, 5 KGy and 10 KGy was determined by ¹H nuclear magnetic resonance (NMR), following the method described by Chen et al. (2024). Firstly, 5 mg of starch was dissolved in DMSO‑d₆, and then sonicated at 85 °C for 60 s to completely dissolve it before placing it in a NMR tube for measurement. The NMR spectra of the samples were obtained using a Bruker spectrometer (ADVANCE, Bruker, Germany) at 600 MHz. The DB of the samples was calculated according to the following equation:

$$\mathit{DB}\left(\%\right)=\frac{I_{\alpha -1,6}}{I_{\alpha -1,4}+I_{\alpha -1,6}}\times 100\%$$ (1)

where ${\mathrm{I}}_{\alpha -1,4}$and ${\mathrm{I}}_{\alpha -1,6}$ denote the peak areas of α-1,4 and α-1,6 linkages, respectively.

2.2.3. Preparation of starch-VE complexes

The complexes were prepared based on the method of Zhao et al. (2019), with slight modifications. Each starch sample (10 g; NS, 5, 10 KGy) was first suspended in distilled water to obtain 10% (w/w) starch suspensions. Separately, 0.5 g of VE was dissolved in 50 mL of anhydrous ethanol before being thoroughly mixed with the starch suspensions. The resulting mixtures were then treated with UHP at 100, 300 and 500 MPa for 10 min at 25 °C prior to a 10-min centrifugation at 10,619g. The precipitates were subsequently washed twice with 50% ethanol to remove unbound VE, dried at 40 °C, ground into powder and passed through a 100-mesh sieve before being stored under dry conditions. The obtained samples were eventually labeled as follows: Control-100 MPa (native starch without EBI pretreatment, complexed with VE under 100 MPa UHP), Control-300 MPa, and Control-500 MPa, 5 KGy-100 MPa, 5 KGy-300 MPa, 5 KGy-500 MPa, 10 KGy-100 MPa, 10 KGy-300 MPa and 10 KGy-500 MPa. The same UHP treatment and drying procedure was carried out for the starch suspension without VE, with the resulting samples labeled as NS-100 MPa, NS-300 MPa and NS-500 MPa.

2.2.4. Scanning electron microscopy (SEM)

A small amount of each dried sample was fixed onto an aluminum stub using conductive adhesive and coated with a thin layer of gold. The microstructure was then observed using a JCM-7000 scanning electron microscope (SEM) at an accelerating voltage of 5 kV and a magnification of 1000×.

2.2.5. Confocal laser scanning microscopy (CLSM)

CLSM analysis was conducted according to the method of Liu et al. (2024), with slight modifications. Each sample (2 mg) was fully dispersed in a Nile Red-methanol solution (1 g·L⁻¹) and, following a 12-h incubation at 4 °C, it was centrifuged at 1873g for 30 min. Excess dye was then removed by washing with methanol, after which the stained samples were air-dried at room temperature and visualized using CLSM (Carl Zeiss AG, LSCM 800, Germany) with an excitation wavelength of 633 nm.

2.2.6. Degree of complexation

The relative amount of complexed VE was determined using the Folin-Ciocalteu method, with slight modifications based on Wang et al. (2024). A 0.2 g portion of each prepared sample was dispersed in 5 mL of distilled water and heated in a water bath at 90 °C for 10 min. After cooling to room temperature, the samples were centrifuged at 4000 g for 10 min, after which 2 mL of the supernatant was mixed with 2 mL of 10-fold diluted Folin-Ciocalteu reagent and allowed to react for 5 min. Subsequently, 2 mL of 6% (w/v) sodium carbonate solution was added, and the mixture was incubated in a 65 °C water bath for 10 min before being cooled to room temperature. The absorbance values of the samples were eventually measured at 725 nm using a UV–Vis spectrophotometer (Varian Inc., Palo Alto, CA).

2.2.7. Antioxidant properties

The antioxidant capacity of the samples was assessed via DPPH radical scavenging activity using a commercial DPPH assay kit (Suzhou Grace Biotechnology Co., Ltd., China).

2.2.8. Differential scanning calorimetry (DSC)

The thermal properties of the samples were determined using DSC (Q20; TA Instruments, New Castle, DE) according to the method of Wang et al. (2024). The instrument was calibrated using indium (purity >99.99%, melting point 156.6 °C), and baseline drift (< ±0.05 mW) was corrected by blank baseline subtraction. Briefly, 2.5 mg of each sample was mixed with 7.5 μL of distilled water in a sealed aluminum crucible. After equilibration at room temperature for 12 h, the samples were scanned from 30 to 210 °C at a rate of 10 °C·min⁻¹ and under a nitrogen flow rate of 2 mL·min⁻¹. The gelatinization onset temperature (To), peak temperature (Tp), conclusion temperature (Tc) and enthalpy change (ΔH) were then recorded.

2.2.9. X-ray diffraction (XRD) analysis

The crystalline structure of the samples was analyzed using an X-ray diffractometer (Shimadzu, XRD-6000, Japan), operated at 40 kV and 40 mA. In this case, diffraction patterns were recorded over a 2θ range of 5° to 40° at a scanning speed of 4°·min⁻¹ and a step size of 0.02°. The relative crystallinity (RC) was subsequently calculated according to the following equation (Lei et al., 2023):

$$\mathit{RC}\left(\%\right)=\frac{\mathit{Area\; of\; crystalline\; peaks}}{\mathit{Area\; of}\mathit{all}\mathit{peaks}}\times 100\%$$ (2)

2.2.10. Fourier transform-infrared (FT-IR) spectroscopy

FT-IR analysis was performed according to the method of Zhao, Liang, et al. (2023) and Zhao, Wang, et al. (2023). Briefly, 1 mg of each dried sample was thoroughly mixed with 100 mg of potassium bromide, ground and eventually pressed into a thin disc. Spectra were then acquired using a Nicolet 170-SX FT-IR spectrometer (Thermo Fisher Scientific, Waltham, MA) over the range of 4000–500 cm⁻¹.

2.2.11. Rheological properties

The rheological properties of the samples were determined using a DHR-1 rheometer (TA Instruments, USA) as previously described by Wang et al. (2024). Briefly, a 0.3 g sample was mixed with 3.6 mL of distilled water and heated in a 95 °C water bath for 30 min. After cooling to room temperature, the resulting paste was placed on the rheometer plate, and a thin layer of silicone oil was applied to the surface to prevent moisture loss. Rheological measurements were then conducted, with the parameters set to a frequency range of 0.1–10 Hz and a strain of 2%.

2.2.12. In vitro digestibility

The in vitro digestibility of the starch samples was determined based on the method of Chen et al. (2024), with slight modifications. Briefly, 100 mg of each dried sample was dispersed in 5.0 mL of simulated saliva solution and stirred continuously at 60 rpm for 5 min in a constant-temperature water bath (37 °C). Subsequently, 5.0 mL of simulated gastric solution (containing 3.5 mg·mL⁻¹ pepsin in 1 mol·L⁻¹ HCl, pH = 1.8 ± 0.1) was added, with the mixture incubated for 2 h at 37 °C in a constant-temperature water bath. Following the gastric digestion, the pH was adjusted to 7.0 ± 0.1 before adding 5.0 mL of an enzyme mixture (containing 200 U·mL⁻¹ α-amylase and 160 U·mL⁻¹ glucoamylase). The samples were then incubated at 37 °C for 120 min with agitation at 120 rpm. At specific time points (0, 10, 20, 30, 60, 90 and 120 min), corresponding to the end of oral digestion, end of gastric digestion and during intestinal digestion, 1 mL of sample was collected for analysis. Specifically, the enzymes were inactivated by adding 4 mL of anhydrous ethanol, and following a 10-min centrifugation at 2654g, the amount of reducing sugar in the hydrolysate was determined using the 3,5-dinitrosalicylic acid method. Each sample was analyzed in triplicate.

The hydrolysis index (HI) as well as rapidly digestible starch (RDS), slowly digestible starch (SDS) and resistant starch (RS) content were eventually calculated as follows:

$$\mathrm{HI}/\%=\frac{\mathrm{Rt}}{W}\times 100\%$$ (3)

$$\mathit{RDS}/\%=\frac{\left(R20-\mathit{FG}\right)\times 0.9}{W}\times 100\%$$ (4)

$$\mathit{SDS}/\%=\frac{\left(R120-R20\right)\times 0)9}{W}\times 100\%$$ (5)

$$\mathit{RS}/\%=100-\mathit{RDS}-\mathit{SDS}$$ (6)

where Rt represents the amount of reducing sugar at time t, R20 and R120 represent the amounts of glucose released at 20 and 120 min respectively, FG is the free reducing glucose in the starch, W is the total starch content and 0.9 is the conversion factor from glucose to starch equivalent.

2.2.13. Determination of VE retention rate

VE content was determined by UV spectrophotometry with slight modifications (Yılmaz, Öztürk, & Kadıoğlu, 2004). VE standard (50.0 mg) was dissolved in anhydrous ethanol and made up to 50 mL to obtain a 1.0 mg/mL stock solution. Aliquots of the stock solution (0, 0.25, 0.50, 1.00, 1.50, 2.00, 2.50 mL) were transferred to 10 mL volumetric flasks and diluted to volume with anhydrous ethanol to obtain standard solutions at concentrations of 0, 25, 50, 100, 150, 200, and 250 μg/mL. Absorbance was measured at 284 nm with anhydrous ethanol as the blank, and the calibration curve was constructed.

Approximately 0.1 g of VE–starch complex was accurately weighed into a centrifuge tube, mixed with 5.0 mL of anhydrous ethanol, vortexed for 1 min, and subjected to ultrasonic-assisted extraction for 15 min (200 W, 25 °C). The extract was centrifuged at 10000 rpm for 10 min, and the supernatant was filtered through a 0.45 μm organic membrane before absorbance measurement at 284 nm. All samples were analyzed in triplicate. Blank starch samples (without VE) were processed in parallel as background controls.

The VE content was calculated from the standard curve, and the VE retention rate was calculated as:

$$\mathit{Retention\; rate}\left(\%\right)=\left(\frac{m}{m_0}\right)\times 100\%$$ (7)

where m is the mass of VE in the complex, and m₀ is the initial amount of VE added (0.5 g).

2.2.14. Statistical analysis

All experiments were performed in triplicate, and results are expressed as mean ± standard deviation. Statistical comparisons were performed using two-way ANOVA followed by Duncan's multiple range test (p < 0.05) in SPSS 23.0, and figures were generated using Origin 2021.

3. Results and discussion

3.1. Degree of branching (DB)

The ¹H NMR spectra of NS, 5 KGy, 10 KGy are exhibited in Fig. 1, where the degree of branching of the samples can be quantified by calculating the proportion of α-1,6 glycosidic bonds in the hydrogen spectra (Zhang et al., 2018). NS, 5 KGy, 10 KGy both showed distinct peak signals around 5.1 ppm and 4.9 ppm, indicating the hydrogen atoms on the α-1,4 glycosidic bond and α-1,6 glycosidic bond-connected glucose units, respectively. The DB values of the samples are also shown in Fig. 1, with the leftmost number in Fig. 1. It can be clearly seen that after EBI irradiation, the DB value of starch decreased significantly with increasing irradiation dose (10 KGy < 5 KGy < NS). This may be due to the fact that the high-energy rays produced by irradiation will cause the glycosidic bonds in the molecular chains of starch to be attacked and broken, in which the α-1,6 glycosidic bonds may be broken more easily. In addition, EBI treatment degraded both amylopectin and amylose, disrupting the starch structure, this is consistent with previous studies, that is, both Lei et al. (2023) and Zhao, Liang, et al. (2023) and Zhao, Wang, et al. (2023) found that irradiation treatment would induce degradation of amylopectin more easily, resulting in an increase in the content of amylose.

Fig. 1.

¹H NMR spectra of NS (a), 5 KGy (b) and 10 KGy (c). The rightmost number represents the degree of branching (DB) for each sample. The intense signals in the δ 3.3–3.6 ppm region (H2–H5 of glucose ring protons) have been vertically truncated to enhance visualization of the diagnostically relevant anomeric proton signals (α-1,4 at ∼δ 5.3 ppm and α-1,6 at ∼δ 4.7–5.1 ppm); this truncation does not affect the integration ratios used for DB calculation.

3.2. Morphological structure analysis

The SEM images of the samples are shown in Fig. 2. Overall, the native potato starch granules exhibited oval, round or spherical shapes with smooth surfaces (Wani et al., 2016), and irradiation alone did not significantly alter their surface morphology (Zhao, Liang, et al., 2023; Zhao, Wang, et al., 2023). Similarly, after UHP treatment, starch morphology remained mostly intact under low-pressure conditions (Fig. 2a1). However, at 300 MPa, small surface depressions appeared (Fig. 2a2), with visible cracks further observed at 500 MPa (Fig. 2a3). Additionally, the granule surfaces exhibited mild swelling after UHP treatment, likely due to the infiltration of water molecules into the starch granules during the pressurization process. These results were consistent with observations reported for pressure-treated lotus starch (Guo et al., 2015) and fava bean starch (Zhao, Liang, et al., 2023; Zhao, Wang, et al., 2023). Following the addition of VE, the starch granules largely retained their original morphology. Nonetheless, in the 500 MPa group, although the initial cracks disappeared, some surface depressions and pores were still evident. This could be attributed to the fat-soluble nature of VE, which facilitates hydrogen bonding interactions between VE and starch, resulting in a stronger hydrogen-bonding network that protects the starch granules from pressure-induced damage (Zhang, Yang, Li, & Gao, 2011). A similar structural enhancement was reported for rice and maize starch complexes with apple polyphenols following UHP (Chou et al., 2020), where the addition of the polyphenols led to more compact and stable microstructures.

Fig. 2.

Scanning electron micrographs of the starch-VE complexes (×1000), with the red arrows indicating deformation, pores and cracks on the starch granule surfaces. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

3.3. CLSM analysis

CLSM images of the complexes are presented in Fig. 3, providing insight into the distribution and relative amount of VE within the samples. Nile red, a lipophilic fluorescent dye, stains fat-soluble components (such as VE) red, and as shown in Fig. 3 (Groups b, c, and d), samples containing VE displayed clear red fluorescence, confirming the formation of amylose-VE complexes (Jia, Sun, Chen, Zheng, & Guo, 2018). In addition, the intensity of red fluorescence increased after electron beam irradiation (EBI), which may be attributed to the cleavage of α-1,4 and α-1,6 glycosidic bonds in starch molecules during EBI, resulting in a significant increase in amylose content and an increase in binding sites for VE within the starch matrix, thereby promoting the formation of complexes (Lei et al., 2023). However, with increasing UHP pressure, the red fluorescence decreased significantly, suggesting that the UHP disrupted the internal starch structure, weakening its capacity to form complexes with VE (Guo et al., 2015).

Fig. 3.

Confocal laser scanning microscopy images of the starch-VE complexes, with the white arrows indicating VE.

3.4. Degree of complexation in starch-VE complexes

The complexing degree of potato starch with VE was quantified using the Folin-Ciocalteu method, with the results presented in Fig. 4. This method relies on the reduction of molybdate and tungstate in the reagent by phenolic hydroxyl groups, resulting in the formation of a blue complex (Zhao et al., 2019). In this case, the absorbance intensity is proportional to the degree of complexation.

Fig. 4.

Complexing degree of the different starch-VE complexes.

Overall, the irradiated samples showed significantly higher absorbance values compared with the non-irradiated ones, with this effect being dose-dependent, thus indicating that the EBI treatment significantly enhanced the complexing degree. This could potentially be attributed to EBI-induced disruption of the amylose structure, which promoted amylopectin degradation into shorter chains, thereby facilitating their interaction with VE. These observations were consistent with those reported by Liu et al. (2024). In contrast, UHP treatment caused a significant reduction in absorbance with increasing pressure. For example, at 100 MPa, the absorbance values were 0.562 (Control), 0.597 (5 KGy) and 0.816 (10 KGy), while at 500 MPa, they decreased to 0.396, 0.418 and 0.437, respectively. These results suggested that UHP was unfavorable for complex formation, possibly due to the disruption of crystalline regions and swelling of the starch granules, which may be broken hydrogen bonds and weakened starch-VE interactions (Li & Zhu, 2018). It should be noted that the moisture content of the samples before and after UHP treatment was not quantitatively determined in this study; therefore, the specific contribution of water infiltration to the starch–VE interactions could not be precisely quantified. Future studies should incorporate moisture content determination to further elucidate the independent and interactive effects of water infiltration and pressure during UHP treatment on the formation of starch–small molecule complexes. A similar trend was observed by Guo, Zhao, Chen, Chen, and Zheng (2019) in starch-tea polyphenol complexes following UHP treatment. Interestingly, even under UHP conditions, irradiated samples (e.g., 10 KGy-100 MPa, 0.816) exhibited higher absorbance than their non-irradiated counterparts (Control-100 MPa, 0.562), indicating that the EBI pretreatment partially counteracted the negative effects of UHP on the complexes and reflecting a synergistic effect. These results were consistent with those of CLSM (Fig. 3). Two-way analysis of variance (ANOVA) (Table 3) demonstrated that both EBI dose (F (2,24) =1142.69, p < 0.001) and UHP pressure (F (2,24) =1379.58, p < 0.001) exerted significant main effects on the degree of complexation. Moreover, a significant interaction between the two factors was observed (F (4,24) =222.23, p < 0.001), suggesting that the promoting effect of EBI pretreatment on complex formation was regulated by the subsequent UHP treatment level. The significant interaction between EBI and UHP (F (4,24) =222.23, p < 0.001) statistically validates a true multiplicative synergistic effect of the two factors on complex formation, rather than a mere additive combination of their individual effects.

Table 3.

Two-way ANOVA results for the effects of EBI dose, UHP pressure, and their interaction on the physicochemical and functional properties of starch–VE complexes.

Response variable	EBI dose	UHP pressure	EBI*UHP
Complexing degree	F (2,24) =1142.69 p < 0.001	F (2,24) =1379.58 p < 0.001	F (4,24) =222.23 p < 0.001
DPPH scavenging activity (%)	F (2,24) =15.605 p < 0.001	F (2,24) =0.579 p = 0.567	F (4,24) =0.107 p = 0.979
ΔH (J·g−1)	F (2,24) =36.569 p < 0.001	F (2,24) =57.325 p < 0.001	F (4,24) =0.776 p = 0.550
VE retention (%)	F (2,18) =6.564 p = 0.005	F (2,18) =7.116 p = 0.003	F (4,18) =0.613 p = 0.657

Significant effects (p < 0.05) are indicated in bold. df = (2, 24) for all variables except VE retention (df = (2, 18)), reflecting the exclusion of native starch (NS) samples from the VE retention analysis.

3.5. Assessment of antioxidant capacity

DPPH is a stable free radical that appears dark purple in color, and in the presence of antioxidants, hydrogen atoms from these compounds bind to the unpaired electron of DPPH, reducing it to a stable non-radical form which is detected as a color change from dark purple to light yellow (Wang, Zhao, Li, Wang, & Shen, 2015). In this study, the free radical scavenging activity of the different complexes was calculated based on absorbance measurements, as illustrated in Fig. 5. Overall, the DPPH scavenging activity was significantly enhanced following EBI treatment. For instance, the scavenging rate rose from 64.92% in Control-100 MPa to 85.02% in 10 KGy-100 MPa, indicating enhanced antioxidant capacity due to irradiation. This improvement could be attributed to the cleavage of glycosidic bonds and the degradation of starch molecular chains, which reduce spatial hindrance and facilitate the incorporation of VE between starch molecules (Lei et al., 2023), thus promoting better interaction with free radicals and enhancing antioxidant activity. Additionally, previous studies have shown that irradiation could promote the decomposition of polyphenolic compounds (Adamo et al., 2004; Hwang et al., 2021). In contrast to EBI, increasing UHP pressure reduced the antioxidant activity of the complexes in the following order: 100 MPa > 300 MPa > 500 MPa. This reduction could be due to the disruption of hydrogen bonds within starch as well as between starch and VE at higher pressures (Li & Zhu, 2018), leading to weakened molecular interactions and reduced antioxidant effects. This is consistent with previous findings where polyphenol content was positively and significantly correlated with DPPH scavenging activity (Chou et al., 2020; Yu et al., 2023), thus aligning with the complexing degree results shown in Fig. 4. Two-way ANOVA (Table 3) revealed that EBI dose had a significant main effect on DPPH scavenging activity (F (2,24) =15.61, p < 0.001). In contrast, neither UHP pressure (F (2,24) =0.58, p = 0.567) nor the EBI × UHP interaction (F (4,24) =0.11, p = 0.979) reached statistical significance. These results suggest that the antioxidant properties of the starch–VE complexes are primarily governed by EBI pretreatment, with no detectable contribution from UHP treatment or synergistic effects between the two factors. The non-significant interaction (F (4,24) =0.11, p = 0.979) confirms that the antioxidant activity is exclusively governed by EBI, with no detectable contribution from UHP either additively or synergistically, representing an EBI-dominant effect.

Fig. 5.

DPPH radical scavenging activity of the different starch-VE complexes.

3.6. XRD analysis

The XRD spectra of NS and starch-VE complexes are shown in Fig. 6. Overall, NS exhibited characteristic B-type crystalline diffraction peaks at 17.2°, 22.2° and 24°. It has been documented that small molecules such as polyphenols and fatty acids can interact with starch through two distinct modes: forming V-type inclusion complexes, or forming non-inclusion complexes through hydrogen bonding and hydrophobic interactions in the amorphous regions (Wu et al., 2024; Yu et al., 2022). Classical V-type complexes typically exhibit characteristic diffraction peaks at approximately 7.5°, 12.8°, and 19.8° (2θ) in XRD patterns (Wang et al., 2025). In the present study, none of these characteristic peaks were observed in any of the samples, indicating that VE did not form classical V-type inclusion complexes with amylose, which is likely attributable to the relatively large molecular volume of VE (∼473 Da) that may impose steric hindrance (Jiang et al., 2019). Non-inclusion-type starch–small molecule complexes formed primarily through hydrogen bonding and hydrophobic interactions in amorphous regions have also been reported in other systems (Guo et al., 2019), providing a literature basis for the mechanistic interpretation herein. The addition of VE did not introduce new diffraction peaks, suggesting that the complexes were formed via non-covalent interactions rather than through V-type inclusion crystals, which is consistent with earlier reports (Guo et al., 2019).

Fig. 6.

X-ray diffraction patterns of native starch (NS) and starch-VE complexes: 100 MPa group (A), 300 MPa group (B) and 500 MPa group. (C). RC: relative crystallinity.

The RC of the complexes increased under the addition of VE followed by UHP treatment, thus reflecting enhanced structural stability. This could have been due to VE molecules entering the amorphous regions of starch, where they induced the rearrangement of starch chains and facilitated the formation of a more ordered structure and denser crystalline regions (Zuo, Zhu, Jiang, Sui, & Kong, 2024). Similarly, under identical pressure conditions, RC increased with higher irradiation doses in the following order: 10 KGy > 5 KGy > NS. For instance, in the 100 MPa group, RC increased from 22.26% (Control) to 25.05% (10 KGy), likely due to EBI-induced cleavage of glycosidic bonds and degradation of starch (Zheng et al., 2023), which resulted in shorter and more linear chains that aligned and binded more effectively with VE, thus increasing crystallinity. However, as UHP pressure increased, RC declined significantly from 25.05% for 10 KGy-100 MPa to 23.36% for 10 KGy-500 MPa. These results suggested that high-pressure treatment likely caused structural damage to the starch granules, disrupting the double-helix and crystalline regions and impairing molecular interactions. It has also been hypothesized that the infiltration of water molecules during UHP treatment could open the starch double helices within crystalline regions, rendering them more susceptible to disruption (Colussi et al., 2018).

3.7. FT-IR spectroscopy

The FT-IR spectra of the starch-VE complexes are presented in Fig. 7. Specifically, the broad absorption bands in the 3000–3700 cm⁻¹ region corresponded to O—H stretching vibrations, while the peak near 2940 cm⁻¹ could be attributed to C—H stretching in CH₂ groups (Wang et al., 2024). Furthermore, the absorption peak at 1647 cm⁻¹ indicated water retention in the non-crystalline regions of the starch granules, and the band near 1000 cm⁻¹ corresponded to C—O—H bending, with both being related to hydrogen bonding (Li et al., 2023). Importantly, the addition of VE did not result in new absorption peaks, thereby confirming that no covalent bonds were formed. However, the 1647 cm⁻¹ band exhibited a blue-shift, with the O—H stretching band (3000–3700 cm⁻¹) also showing increased broadening, thus further supporting the involvement of hydrogen bonding in the complexation of VE with starch (Zhang et al., 2024).

Fig. 7.

FT-IR spectra of the starch-VE complexes: 100 MPa group (A), 300 MPa group (B) and 500 MPa group (C).

Short-range molecular ordering of the complexes was also evaluated based on the absorbance ratio R_1047/1022, with higher values indicating increased structural order (Wang et al., 2024). As shown in Table 1, both VE addition and EBI treatment increased the R_1047/1022 ratio in a dose-dependent manner, while UHP treatment led to a decrease. These results suggested that VE promoted a more stable and ordered complex structure, while EBI enhanced crystallinity by increasing amylose content. In contrast, high-pressure conditions likely disrupted the crystalline regions, particularly at higher pressures, as noted in the XRD findings (Fig. 6).

Table 2.

Initial hydrolysis rate and final hydrolysis degree of starch–VE complexes at 240 min.

Samples	Initial rate (%·min−1)	Final hydrolysis degree at 240 min (%)
NS-100 MPa	3.30 ± 0.14c	81.95 ± 0.87bcd
Control-100 MPa	3.04 ± 0.08d	76.22 ± 0.87f
5 KGy-100 MPa	2.54 ± 0.10g	72.81 ± 1.34g
10 KGy-100 MPa	2.61 ± 0.14fg	75.65 ± 1.03f
NS-300 MPa	3.84 ± 0.14b	83.79 ± 1.02b
Control-300 MPa	3.34 ± 0.08c	81.27 ± 0.87cd
5 KGy-300 MPa	2.74 ± 0.08ef	76.01 ± 0.95f
10 KGy-300 MPa	2.49 ± 0.16g	78.8 ± 0.71e
NS-500 MPa	4.25 ± 0.08a	87.1 ± 0.95a
Control-500 MPa	3.66 ± 0.11b	83.79 ± 1.31b
5 KGy-500 MPa	2.88 ± 0.11de	80.22 ± 0.79de
10 KGy-500 MPa	3.31 ± 0.08c	82.53 ± 0.71bc

All values are expressed as mean ± standard deviation. Different superscript letters in the same column indicate that the data are significantly different (p < 0.05).

Table 1.

Molecular ordering and thermal properties of the starch-VE complexes.

Samples	R1047/1022	To (°C)	Tp (°C)	Tc (°C)	ΔH (J·g−1)
NS-100 MPa	0.76 ± 0.015fg	61.34 ± 0.36b	64.78 ± 0.06de	70.46 ± 1.77ab	2.92 ± 0.57de
Control-100 MPa	0.79 ± 0.015cde	62.59 ± 0.10a	64.85 ± 0.07de	68.71 ± 0.06e	3.66 ± 0.17bc
5 KGy-100 MPa	0.82 ± 0.015b	60.47 ± 1.28cd	65.08 ± 0.64cd	69.17 ± 1.83bcd	4.24 ± 0.23b
10 KGy-100 MPa	0.85 ± 0.012a	61.44 ± 0.28b	65.34 ± 0.18bc	68.03 ± 0.47de	5.01 ± 0.14a
NS-300 MPa	0.76 ± 0.006g	61.36 ± 0.22b	64.46 ± 0.04ef	68.17 ± 0.32cd	2.38 ± 0.46e
Control-300 MPa	0.78 ± 0.010def	61.10 ± 0.23bc	64.76 ± 0.05de	71.32 ± 0.27a	3.27 ± 0.14cd
5 KGy-300 MPa	0.81 ± 0.015bc	61.00 ± 0.34bc	65.00 ± 0.06cd	69.64 ± 0.61bc	3.76 ± 0.31bc
10 KGy-300 MPa	0.84 ± 0.012a	61.52 ± 0.29b	66.51 ± 0.32a	71.32 ± 0.26a	3.92 ± 0.27b
NS-500 MPa	0.73 ± 0.006h	59.79 ± 0.42de	64.25 ± 0.45f	71.17 ± 0.14a	1.42 ± 0.24f
Control-500 MPa	0.76 ± 0.010fg	59.60 ± 0.21e	64.17 ± 0.05f	69.61 ± 0.35bc	1.71 ± 0.18f
5 KGy-500 MPa	0.77 ± 0.010efg	62.71 ± 0.18a	64.99 ± 0.01cd	71.68 ± 0.09a	2.58 ± 0.22e
10 KGy-500 MPa	0.79 ± 0.012cd	61.29 ± 0.09b	65.55 ± 0.18b	71.89 ± 0.57a	2.75 ± 0.60de

All values are expressed as mean ± standard deviation. Different superscript letters in the same column indicate that the data are significantly different (p < 0.05).

3.8. DSC analysis

The thermal properties of the samples, including To, Tp, Tc and ΔH, are summarized in Table 1. The values for To, Tp and Tc correspond to the melting temperatures of the least stable microcrystals, the majority of starch crystals and the most stable microcrystals, respectively (Karunaratne & Zhu, 2016), and among these, Tp serves as an indicator of thermal stability, with higher values suggesting increased structural stability of the complex (Chou et al., 2020). After the addition of VE, the To and Tc values remained relatively unchanged, while that of Tp increased significantly, indicating that VE interacted with the starch within this temperature range to form more thermally stable complexes.

ΔH represents the energy absorbed during the transition of starch granules from a crystalline to a gelatinized state (Wang et al., 2021), with higher values indicating a more ordered crystal structure (Kang et al., 2023). Overall, the ΔH of irradiated samples increased in a dose-dependent manner compared with NS, possibly due to EBI-induced degradation of amylopectin into shorter amylose chains, thus promoting rearrangement and the formation of a more stable crystal structure (Zheng et al., 2023). Conversely, the ΔH values decreased significantly with increasing UHP pressure, reaching a minimum of 1.71 J·g⁻¹ in the Control-500 MPa sample. In this case, the UHP likely disrupted the double helix and crystalline regions of starch (Zhao, Liang, et al., 2023; Zhao, Wang, et al., 2023), thereby reducing the energy required for gelatinization. Furthermore, a previous study noted that partial gelatinization of starch may also occur at 500 MPa (Oh, Pinder, Hemar, Anema, & Wong, 2008), which could further weaken the hydrogen bonds between starch and VE molecules within the gelatinization temperature range. However, this effect is confined to the low-temperature gelatinization process and does not involve the high-temperature (>100 °C) interactions discussed in some DSC studies. Two-way ANOVA (Table 3) revealed that both EBI dose (F (2,24) =36.57, p < 0.001) and UHP pressure (F (2,24) =57.33, p < 0.001) exerted significant main effects on the enthalpy change (ΔH), while their interaction did not reach statistical significance (F (4,24) =0.78, p = 0.550). This suggests that EBI and UHP acted independently in modulating the thermal stability of the starch–VE complexes, with no synergistic or antagonistic interplay between the two factors. The non-significant interaction (F (4,24) =0.78, p = 0.550) indicates that EBI and UHP independently modulate the thermal stability of the complexes, representing independent additive effects rather than synergy.

3.9. Rheological analysis

The storage modulus (G'), loss modulus (G") and loss tangent (tan δ) of the samples, which reflect gel structure and viscoelastic properties, are shown in Fig. 8. G' indicates the elastic component as well as the energy stored and recovered in the sample, while G" represents the viscous component and energy lost during deformation under external forces (Liu et al., 2023; Wu et al., 2024). Overall, for all samples, G' was greater than G" across the frequency range, indicating that elasticity dominated over viscosity. Furthermore, the tan δ, defined as the ratio of G" to G', was less than 1 in all cases, suggesting a weak gel-like behavior and a tendency toward solid-like characteristics (Liu et al., 2024; Zuo et al., 2024). Following the addition of VE, both G' and G" decreased relative to native starch, possibly due to hydrogen bonding between VE and starch, which limited water penetration into the granules, thus inhibiting elastic gel formation (Wu, Niu, & Xu, 2020). In addition, EBI treatment enhanced the viscoelasticity of the samples, with both G' and G" following the order: 10 KGy > 5 KGy > NS. These results suggested that EBI-induced structural modifications likely favored the formation of a stable gel network, thereby improving the rheological performance of the complexes, in agreement with previous reports (Zheng et al., 2023). However, as UHP pressure increased, G' and G" both decreased in the following order: 100 MPa > 300 MPa > 500 MPa. These changes reflected a reduction in viscoelasticity under UHP, likely due to UHP-induced weakening of hydrogen bonding and disruption of the crystalline structure, which resulted in a shift from an ordered to a disordered state.

Fig. 8.

The storage modulus (G'), loss modulus (G") and loss factor (tan δ) of the starch-VE complexes.

3.10. In vitro digestibility

The in vitro digestibility of the starch-VE complexes is illustrated in Fig. 9. As digestion time increased, all samples were gradually hydrolyzed by amylase and generally reached a plateau after 180 min. Furthermore, both the hydrolysis rate (Fig. 9A) and digestibility of the complexes (Fig. 9B) decreased with the addition of VE under different treatments, while the RS content increased. For example, the RS content of NS-100 MPa was 63.12%, and it increased to 65.70% upon VE addition (Control-100 MPa). This could be attributed to VE's interaction with starch, primarily through hydrogen bonding, which potentially inhibited amylase activity and enhanced starch resistance to enzymatic digestion (Zheng et al., 2021). These results were consistent with the XRD findings where no V-type crystalline structures were observed in the complexes. Furthermore, previous studies have suggested that increasing the VE content in the complex could result in reduced digestibility (Cai, Zhang, Zhang, Shan, & Cheng, 2013), thus aligning with the observed complexing degree (Fig. 4).

Fig. 9.

The hydrolysis rate (A) and digestibility characteristics (RDS, SDS, RS) (B) of starch-VE complexes.

On the other hand, with increasing UHP pressure, the hydrolysis rate increased, while the RS content decreased significantly. Notably, the RS content followed the order: 100 MPa > 300 MPa > 500 MPa, which corroborated previous findings (Chou et al., 2020). The digestibility of starch is influenced by its crystalline structure, with denser and more ordered crystalline regions exhibiting greater resistance to enzymatic attack compared with amorphous ones (Guo et al., 2019). Therefore, it is likely that the higher pressure caused more severe disruption to the crystalline zones, leading to reduced RS content. In contrast, EBI treatment increased the RS content with a dose-dependent effect. For instance, the RS content of Control-300 MPa, 5KGy-300 MPa and 10 KGy-300 MPa were 63.43%, 64.54% and 65.80%, respectively. This was likely due to EBI-induced depolymerization of amylopectin into shorter amylose chains, which then rearranged to form more resistant structures, thereby enhancing the starch's resistance to enzymatic digestion. Notably, the highest RS content (67.24%) was achieved when the starch-VE complex was prepared using 10 KGy EBI pretreatment in combination with 100 MPa UHP treatment.

3.11. Determination of VE retention rate analysis

The VE retention rate in VE-starch composite was determined by UV–vis spectrophotometry, and the results are shown in Fig. 10. Both EBI pretreatment and UHP treatment exerted significant effects on VE retention (p < 0.05, Table 3), while their interactive effect did not reach statistical significance. Under the same UHP pressure, VE retention increased significantly with increasing EBI dose. At 100 MPa, the 10 KGy EBI-pretreated group exhibited the highest VE retention of 88.13 ± 1.40%; at 300 MPa and 500 MPa, the VE retention of the 10 KGy groups reached 81.50 ± 1.64% and 62.23 ± 1.50%, respectively, which were significantly higher than those of the corresponding control groups under the same pressures (61.60 ± 1.15% and 43.77 ± 1.19%). The promoting effect of EBI on VE retention could be attributed to its degradative action on starch molecules: electron beam irradiation cleaves glycosidic bonds, depolymerizes amylopectin, increases the proportion of amylose and improves the homogeneity of chain-length distribution, thereby providing more binding sites and hydrophobic cavities for VE, enhancing non-covalent interactions (e.g., hydrophobic interactions and hydrogen bonding) between starch and VE, and facilitating the stable entrapment of VE. It has been reported that EBI treatment can degrade starch molecular chains and expose more active sites, thus strengthening the binding capacity toward small-molecule ligands (Chung & Liu, 2010). Under the same EBI dose, VE retention decreased significantly with increasing UHP pressure. In the 10 KGy pretreatment group, VE retention at 100, 300 and 500 MPa were 88.13%, 81.50% and 62.23%, respectively, showing a clear pressure-dependent decline. This trend may be associated with excessive gelatinization and structural collapse of starch induced by high pressure, leading to dissociation of the pre-formed starch-VE complexes, leakage of VE and subsequent loss during washing; in addition, the high-pressure environment may also accelerate oxidative degradation of VE (Barba, Esteve, & Frigola, 2012; Oey, Van der Plancken, Van Loey, & Hendrickx, 2008). Notably, EBI pretreatment alleviated the negative effects of high pressure to some extent: the VE retention of the 500 MPa-10 KGy group (62.23%) was even higher than that of the 300 MPa control group (61.60%), indicating that EBI-induced structural modification of starch could enhance the pressure stability of the complexes, which is consistent with previous reports that EBI pretreatment can improve the subsequent modification efficiency of starch (Shen et al., 2025; Zhao, Liang, et al., 2023; Zhao, Wang, et al., 2023). Duncan's multiple range test showed that the VE retention of the nine treatment groups could be divided into seven significantly different subsets (a–g), among which the 10 KGy-100 MPa combination exhibited significantly higher VE retention than all other treatment groups (p < 0.05), while the control-500 MPa group showed the lowest value. These results indicate that the synergistic effect of EBI pretreatment and low-pressure UHP is most favorable for VE retention, and may provide a basis for process optimization in the preparation of starch-VE composites with high retention using non-thermal processing technologies. Two-way ANOVA (Table 3) revealed that both EBI dose (F (2,18) =6.56, p = 0.005) and UHP pressure (F (2,18) =7.12, p = 0.003) exerted significant main effects on VE retention, whereas their interaction did not reach statistical significance (F (4,18) =0.61, p = 0.657). This suggests that EBI and UHP independently regulated VE retention without synergistic or antagonistic interplay between the two factors. This suggests that EBI and UHP independently regulate VE retention with additive rather than synergistic effects, as the non-significant interaction (p = 0.657) indicates no multiplicative enhancement beyond the sum of individual contributions.

Fig. 10.

VE retention rate in starch-VE composites subjected to different EBI doses (0, 5, 10 KGy) and UHP levels (100, 300, 500 MPa). Different letters indicate significant differences (p < 0.05).

4. Conclusion

This study explored the preparation of potato starch-VE complexes via a synergistic approach involving EBI pretreatment and UHP processing. XRD analysis revealed the absence of V-type crystalline structures between VE and starch, suggesting a sustained but weak interaction mechanism. EBI pretreatment significantly increased both the complexing degree and RS content in a dose-dependent manner (10 KGy > 5 KGy), and also improved the short-range ordered structure, thermal stability, and viscoelastic properties of starch. In contrast, increasing UHP pressure caused considerable disruption to the crystalline regions of the starch-VE complexes, thus leading to decreases in the complexing degree, relative crystallinity, antioxidant capacity and RS content. The combination of high-dose EBI (10 KGy) with low-pressure UHP (100 MPa) yielded the highest antioxidant activity (85.02%) and RS content (67.24%). Notably, two-way ANOVA revealed that the EBI–UHP combination was not universally synergistic but rather parameter-dependent: true synergy was observed for complexing degree, while independent additive effects were found for ΔH and VE retention, and an EBI-dominant effect was observed for DPPH scavenging activity. This dual non-thermal strategy shows promise for enhancing starch–VE complex functionality, improving VE solubility and stability, and developing antioxidant/anti-digestive functional foods. However, high costs, process complexity, and safety requirements hinder industrial application. Future research should systematically evaluate storage stability under various conditions. These findings provide new insights into dietary interventions for diabetes management. However, further in vivo studies are required to evaluate the bioavailability and safety of these modified complexes in humans. This approach provides a non-thermal alternative to chemical or enzymatic modifications for producing diabetic-friendly starch-based ingredients, combining nutritional benefits with consumer-friendly clean-label processing.

CRediT authorship contribution statement

Jianyu Yuan: Writing – review & editing, Supervision, Resources, Methodology, Data curation. Miaomiao Tian: Writing – original draft, Validation, Formal analysis, Data curation. Wenting Chen: Validation, Methodology. Macao Zhuo: Validation. Guang Li: Supervision. Yue Wang: Resources, Project administration, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

Data will be made available on request.