Physicochemical properties, bioactivity and oxidative stability of Riceberry germ antioxidants fortified soybean oil complex coacervate

Abstract

This study examined the effect of pH on the complex coacervation of soy protein isolate and gum arabic for encapsulating soybean oil fortified with antioxidants from Riceberry germ. The optimal pH of 3.5 resulted in high encapsulation yield of 90.64 % and encapsulation efficiency of 79.71 %. Antioxidant stability and release profiles showed that microcapsules effectively preserved tocopherol, γ-oryzanol, and antioxidant property during simulated gastrointestinal digestion. Microcapsules also exhibited enhanced pH and thermal stability compared to unencapsulated oil. Shelf-life tests demonstrated that microcapsules surpassed unencapsulated oil in retaining chemical properties and antioxidant property, with pH 4.0 formulation exhibiting the lowest peroxide value at the end of a 45-day storage period at room temperature. These findings demonstrate that SPI–GA complex coacervation not only significantly enhances the stability and functionality of Riceberry germ antioxidants, but also provides a practical strategy for developing stable, nutritious, and versatile soybean oil products for food applications.

Highlights

• •SPI-GA complex coacervation achieved high oil encapsulation (90.64 % yield).
• •Microcapsules showed controlled release of antioxidants during digestion.
• •Encapsulation boosted stability (pH, thermal, shelf-life) compared to free oil.
• •Microcapsule formulated at pH 4.0 effectively protected oil during 45-day storage.
• •Customized microcapsules offer diverse uses for food applications.

1. Introduction

Soybean oil is widely used across the food, cosmetic, and pharmaceutical industries, owing to its rich content of essential fatty acids, especially polyunsaturated fatty acids (PUFAs). However, its high PUFA content makes it prone to oxidative instability, leading to off-flavours and harmful compounds when exposed to heat, light, and oxygen (Rakariyatham et al., 2024). To address this, natural lipophilic antioxidants, such as vitamin E and γ-oryzanol, can significantly improve oxidative stability of oil (Wangsuntornpakdee et al., 2023). Given the increasing consumer demand for natural and sustainable solutions, Riceberry germ, a valuable by-product of rice processing, stands out as a rich and readily available source of these potent lipophilic antioxidants.

Protecting these bioactive compounds during processing and storage remains a critical challenge. Moreover, the inherent lipophilicity of oils, including soybean oil, limits their direct incorporation into water-based or low-fat food systems, which are increasingly popular among health-conscious consumers. Encapsulation, particularly through complex coacervation, offers an effective solution to both these challenges. This method utilizes electrostatic interactions between polymers to form microcapsules that not only significantly enhance oxidative stability and control antioxidant release but also improve oil dispersibility in water-based environments, expanding their applicability in health-oriented food products (Chen et al., 2025).

In our previous study (Wangsuntornpakdee et al., 2023), optimal conditions for enriching soybean oil with lipophilic antioxidants from three Thai rice germ varieties, including Riceberry, were successfully identified. While high levels of vitamin E (tocopherol), γ-oryzanol, and antioxidant properties were achieved from ultrasonic-assisted extraction of Riceberry germ, a significant limitation emerged. A low encapsulation yield (less than 50 %) was observed when attempting to encapsulate the enriched soybean oil using soy protein isolate (SPI) and inulin as wall materials. The low yield restricts practical use and antioxidant protection, posing a key barrier to developing stable, nutrient-rich oils for food applications.

To overcome the limitations observed in our previous study, the current research employed SPI and gum Arabic (GA) as carrier agents for encapsulating the enriched soybean oil. SPI, known for its water solubility and emulsifying properties, forms stable emulsions and offers high nutritional value, making it an excellent encapsulation matrix (Chen et al., 2022). GA is a highly water-soluble polysaccharide and an effective emulsifier. When combined with SPI, it forms a strong coacervate shell around oil droplets, significantly enhancing encapsulation efficiency and microcapsule stability (Tavares & Noreña, 2020). This synergistic interaction has been demonstrated in other systems. Complex coacervation of chia seed oil using SPI and GA has been described by Bordón et al. (2021). They reported that microencapsulation process showed high encapsulation efficiency over 80 % and could preserve the quality of chia seed oil, as reflected by enhanced oxidative stability index. Furthermore, Zhang, Wang, et al. (2024) investigated the interaction between SPI and GA in coacervates and found that GA chains create a network surrounding SPI molecules, resulting in coacervates that have a spherical shape and viscoelastic properties.

Beyond optimizing encapsulation, this study innovatively extends the characterization of the encapsulated oil to practical application scenarios, particularly focusing on its suitability for various food formulations. The in vitro gastrointestinal digestion of the encapsulated oil was critically investigated to assess the precise release and bioavailability of antioxidants. This is essential for understanding their health benefits post-consumption and is often an overlooked aspect in encapsulation studies. Furthermore, this research uniquely examined the stability and dispersibility of the encapsulated soybean oil under various pH conditions, simulating diverse food and beverage environments from acidic to slightly alkaline, which is crucial for its integration into aqueous and low-fat food matrices. Its thermal stability at 180 °C was also rigorously assessed, directly mimicking common frying and baking temperatures. These comprehensive investigations provide unprecedented insights into the robustness and versatility of this novel encapsulated product for diverse real-world food systems, addressing current gaps in the literature.

Considering this rationale, it was hypothesized that complex coacervation using SPI and GA would achieve higher encapsulation efficiency and improved stability of Riceberry germ–enriched soybean oil compared with the previously tested SPI–inulin system, and that the microcapsules would retain antioxidant activity and dispersibility under conditions food-relevant conditions. This approach represents a novel application of SPI–GA coacervation with Riceberry germ antioxidants, which has not been previously explored for enhancing antioxidant stability in soybean oil and addressing the challenge of preserving lipophilic antioxidants under varying pH and thermal conditions.

The aims of this study were to (1) optimize pH conditions for maximum coacervation efficiency and stability using SPI and GA, (2) evaluate the effectiveness of these microcapsules in preserving lipophilic antioxidants during simulated gastrointestinal digestion, and (3) investigate the stability of the encapsulated oil under various pH and thermal treatments. The overall purpose of this research is to contribute significantly to the development of more stable, nutritious, and versatile soybean oil products, leveraging the protective benefits of natural Riceberry germ antioxidants through an optimized complex coacervation system.

2. Materials and methods

2.1. Material

Riceberry germ was purchased from Nana Rice, Surin province, Thailand. The sample was prepared following the method described by Wangsuntornpakdee et al. (2023). Soybean oil was sourced from a supermarket in Japan (Ajinomoto®, J-Oil Mills Inc., Tokyo, Japan). Its fatty acid profile is detailed in Supplementary Table S1. SPI with a protein content of at least 90 % (Myvegan, UK) and GA from Wako Pure Chemical Industries, Ltd. (Japan) were utilized as carrier agents. All other reagents and chemicals employed in the experiments were of analytical grade.

2.2. Preparation of soybean oil enriched with lipophilic antioxidants (SOR)

SOR was prepared using ultrasonic-assisted extraction (UAE) based on the method described by Wangsuntornpakdee et al. (2023). Briefly, Riceberry germ powder (20.0 g) was combined with 100.0 mL of soybean oil (1:5 ratio) and subjected to ultrasonic treatment in an ultrasonic bath (ASU-10D, AS ONE, Japan) at 40 °C for 40 min. After centrifugation at 5170×g for 15 min (25 °C), the supernatant oil fraction was collected and stored in a brown glass bottle at −20 °C.

2.3. Complex coacervation of SOR

2.3.1. Preparation of wall material solutions

A 2.5 g of SPI was dissolved in 100 mL deionized water, and the pH was adjusted to 8.0 using 1.0 M NaOH (the maximum pH for solubility) and stirred for 30 min at ambient temperature. A GA solution was obtained by adding 2.5 g of GA with 100 mL deionized water and stirring until completely dissolved.

2.3.2. Preparation of microcapsules

Complex coacervation was performed using a modified method based on Nori et al. (2011). A 2.5 g of oil extract was emulsified with 100 mL of SPI solution (2.5 %, w/v) and homogenised at 16,000 rpm for 2 min using a homogeniser (Physcotron, Microtec, Tokyo, Japan). The GA solution (2.5 %, w/v) 100 mL was then gently incorporated into the emulsion, and the pH was adjusted to 3.0, 3.5, and 4.0 using 1.0 M HCl. The mixture (Wi ≈ 7.5 g) was stored at 4 °C for 48 h to allow sedimentation. The resulting coacervates were separated by centrifugation (18,900×g, 10 min, 4 °C) and freeze-dried for 24 h using a FREEZONE 4.5 (LABCONCO, MO, US), with a condenser temperature of −50 °C and a chamber pressure of 0.06 mbar. The dried microcapsules were weight (W) before stored in aluminum/polyethylene (Al/PE) bags at −20 °C until further use. Encapsulation yield (EY) was calculated as follows:

$$\mathrm{EY}\left(\%\right)=\frac{\mathrm{W}}{{\mathrm{W}}_{\mathrm{i}}}\times 100$$ (1)

where W is the weight of the dried microcapsules and W_i is the initial weight of the SPI, GA and SOR.

2.4. Encapsulation efficiency (EE)

EE were examined using the method modified from Wangsuntornpakdee et al. (2023). In brief, 0.1 g of microcapsules was dissolved in 10 mL of ethanol and stirred for 10 min at room temperature (∼25 °C), followed by filtration through Whatman No.1 filter paper. The residue was dried and weighed to calculate the weight of surface oil (W_SO). To determine total oil content, 0.1 g of coacervates was mixed with ethanol and sonicated for 40 min (40 °C, 40 kHz, 500 W). The mixture was centrifuged at 12,000×g for 30 min at 25 °C. The weight of total oil (W_TO) was recorded. EE were calculated using the following equations:

$$\mathrm{EE}\left(\%\right)=\frac{{\mathrm{W}}_{\text{TO}}-{\mathrm{W}}_{\mathrm{SO}}}{{\mathrm{W}}_{\text{TO}}}\times 100$$ (2)

where W_TO is the total oil weight in the microcapsules and W_SO is the weight of the oil at the surface of the microcapsules.

2.5. Determination of the chemical properties

The samples were prepared by dissolving 1.0 g of unencapsulated oil in 4 mL of hexane, while 0.1 g of microcapsules was dissolved in 10 mL of deionized water and filtered through a 0.45 μm filter prior to analysis.

2.5.1. Tocopherol

The tocopherol content was evaluated using the modified method of Yilmaz et al. (2004). A 10 mL of ethanol was added to 0.05 mL of sample. The absorbance was read at 290 nm. The tocopherol content was determined using a calibration curve ranging from 0 to 100 μg/mL of α-tocopherol and expressed as milligrams of α-tocopherol per gram of sample (mg α-tocopherol/g sample).

2.5.2. γ-Oryzanol

The γ-oryzanol was determined as described by Bucci et al. (2003). A 0.05 mL of sample was mixed with 10 mL of isopropanol. The absorbance was measured at 325 nm. The γ-oryzanol concentration was quantified using a calibration curve prepared with standard concentrations of γ-oryzanol (0–200 μg/mL). Results were expressed as milligrams of γ-oryzanol per gram of sample (mg γ-oryzanol/g sample).

2.5.3. 2,2′-Azino-bis (3-ethylbenzthiazoline-6-sulphonic acid) (ABTS) radical scavenging activity

ABTS reagent was prepared using the modified method of Re et al. (1999). The samples (0.3 mL) were mixed with ABTS reagent (3 mL) and measured at 734 nm. A standard calibration curve was generated using Trolox at concentrations ranging from 0 to 250 μM. The ABTS radical scavenging activity of the samples was expressed as micromoles of Trolox equivalents per gram of sample (μmol Trolox/g sample).

2.5.4. Ferric reducing antioxidant power (FRAP)

The samples were evaluated according to the method outlined by Peanparkdee et al. (2017). A standard curve was constructed with Trolox concentrations ranging from 0 to 250 μM. FRAP values were expressed as micromoles of Trolox equivalent per gram of sample (μmol Trolox/g sample).

2.6. Mean particle diameter, polydispersity index (PDI), and zeta-potential (ζ-potential)

The mean particle diameter and PDI of SPI, GA, and microcapsules were measured by dynamic light scattering (DLS) using a Zetasizer Nano ZS (Malvern Instruments, UK) at 25 °C. The samples, including wall materials and microcapsules, were diluted with deionized water at a 1:10 ratio and analysed in triplicate.

The ζ-potential of SPI, GA, and microcapsules was also determined using the Zetasizer Nano ZS. All measurements were performed at 25 °C in triplicate, with samples diluted in deionized water at a 1:10 ratio.

2.7. Scanning electron microscope (SEM)

The surface morphology of the freeze-dried microcapsules was examined using a field-emission scanning electron microscope (FE-SEM) (Model SU8020, Hitachi, Tokyo, Japan). The samples were coated with platinum using a sputter coater (Q15R EX, Quorum, UK). SEM images were captured at the desired magnification at room temperature, using an acceleration voltage of 5.0 kV.

2.8. Fourier transform infrared (FTIR) analysis

The structure of unencapsulated SOR (UOR), SPI, GA, and microcapsules was evaluated using an FTIR spectrophotometer (Model Nicolet IR200, Thermo Scientific, Waltham, Massachusetts, USA) as described by Guo et al. (2021) with some modifications. The spectra were scanned from 4000 to 400 cm⁻¹ at a resolution of 4 cm⁻¹ in 16 and 32 scans for microcapsules and oil samples, respectively.

2.9. Stability of microcapsules during in vitro gastrointestinal digestion

In vitro digestion was conducted with slight modifications to the method described by Peanparkdee et al. (2021). Oil (0.1 mL) or microcapsules (50 mg) were dispersed in 10 mL simulated saliva fluid (SSF), with the pH adjusted to 6.75 and α-amylase (200 U) added. The mixture was incubated at 37 °C, shaking at 120 rpm for 10 min. Subsequently, 10 mL simulated gastric fluid (SGF) containing pepsin was added, and the pH was lowered to 1.2 using 1 M HCl before incubating for 2 h. In the final intestinal phase, gastric phase samples were combined with 7.5 mL of simulated intestinal fluid (SIF), 1.25 mL of NaCl (120 mM), and 1.25 mL of KCl (5 mM), with the pH adjusted to 7.0 using 1 M NaOH and shaken for 2 h.

After each digestion phase, the samples were collected and used for subsequent analyses, including the quantification of tocopherol, γ-oryzanol, ABTS radical scavenging activity, and FRAP. The antioxidant property measured by ABTS and FRAP assays were expressed as μmol Trolox equivalents per gram of sample (μmol Trolox/g sample). Meanwhile, the concentrations of tocopherol and γ-oryzanol were expressed as percent release, calculated using the following equation:

$$\text{Release}\left(\%\right)=\frac{{\mathrm{D}}_{\mathrm{A}}}{{\mathrm{D}}_{\mathrm{B}}}\times 100$$ (3)

where D_A is the value of tocopherol or γ-oryzanol of oil or microcapsule after simulated digestion; D_B is the value of tocopherol or γ-oryzanol of oil or microcapsule before simulated digestion.

2.10. Stability of microcapsules under various conditions

2.10.1. Chemical stability under various pH

The pH stability of the oils and microcapsules was evaluated using a modified method based on Peanparkdee et al. (2020). Four buffer solutions were employed to examine the impact of pH on microcapsule stability: 0.1 M citrate–phosphate buffers at pH 3, 5, and 7, and a 0.1 M sodium carbonate–sodium bicarbonate buffer at pH 9. A 0.1 g sample was added to 10 mL of each buffer and stirred for 30 min. The mixtures were then centrifuged at 3000×g and 25 °C for 30 min using a Hermle Z 206 A centrifuge (Hermle Labortechnik GmbH, Wehingen, Germany). The supernatants were analysed to assess the chemical stability of the samples, including the levels of tocopherol, γ-oryzanol, and antioxidant property.

2.10.2. Chemical stability under thermal treatment

Thermal stability was evaluated using a modified method based on López de Dicastillo et al. (2019) and Pires et al. (2022). A 0.1 g of unencapsulated oil and microcapsules was placed in an open glass dish and heated in a hot air oven at 180 °C for 0, 10, 20, and 30 min. At each time interval, 0.1 g of the samples was removed, dissolved in 10 mL of deionized water, and stirred. The mixture was then centrifuged at 3000×g and 25 °C for 10 min. The supernatant was collected and analysed for chemical stability, focusing on tocopherol, γ-oryzanol, and antioxidant property, which were measured using the ABTS and FRAP assays.

2.10.3. Shelf-life stability under various temperatures

The microcapsule samples were stored in Al/PE bags and monitored over 45 days at 25 °C. At the end of the study period, samples were withdrawn and analysed for chemical stability, focusing on tocopherol, γ-oryzanol, and antioxidant property using the FRAP and ABTS methods. The shelf-life stability of UOR, and microcapsules was assessed by calculating the percentage recovery of tocopherol, γ-oryzanol, ABTS and FRAP using the following equation:

$$\text{Recovery}\left(\%\right)=\frac{{\mathrm{S}}_{\mathrm{A}}}{{\mathrm{S}}_{\mathrm{B}}}\times 100$$ (4)

where S_A is the value of TPC, FRAP or ABTS of oil or microcapsules at day 0; S_B is the value of TPC, ABTS, or FRAP of oil or microcapsules at day 45.

Moreover, peroxide value (PV) of microcapsules during storage was evaluated at day 0 and 45. PV of samples was analysed according to the modified method of AOAC official method 965.33. Firstly, 0.5 g of sample was added to 30 mL of acetic acid – chloroform mix solution (3:2 v/v). The mixture was added with fresh saturated KI solution (1 mL) and kept in the dark for 1 min. Added DI (30 mL) and stirred for 1 min. After that add 1 % (w/v) of the starch solution and slowly titrate with 0.01 N sodium thiosulfate until the colour disappears. The peroxide values (milliequivalent peroxide/kg oil) were calculated by using the following calculation:

$$\mathrm{PV}=\frac{\mathrm{S}\times \mathrm{N}\times 1000}{\mathrm{W}}$$ (5)

where S is the volume (mL) of Na₂S₂O₃; N is the concentration of Na₂S₂O₃ solution; W is the weight of the oil (g).

2.11. Statistical analysis

All measurements were conducted in triplicate. Experimental data were carried out using analysis of variance (ANOVA), and differences between treatments were assessed with Duncan's new multiple range test at a significance level of p < 0.05. Statistical analysis was conducted using IBM SPSS Statistics Version 22 (ThaiSoftup Co., Ltd., Bangkok, Thailand).

3. Results and discussion

3.1. Optimum pH for complex coacervation

The ζ-potential is an indicator of the surface charge and electrical potential of polymers within colloidal systems (Tavares & Noreña, 2020). Fig. 1 illustrates the ζ-potential values of SPI and GA dispersions across a pH range of 2.0 to 5.0. SPI solutions displayed positive ζ-potential values between pH 2.0 and 4.0, which shifted to negative values as the pH increased beyond 4.5. This behaviour is linked to the pI of SPI, approximately 4.2 (Feng et al., 2024). The amphoteric properties of SPI proteins, which contain both amino (NH₂) and carboxyl (–COOH) groups, account for this pH-dependent behaviour. At pH values below the pI, the amino groups (–NH₃⁺) are protonated, conferring a positive charge, while the carboxyl groups remain in their neutral form. When the pH rises above the pI, deprotonation of these functional groups occurs. The amino groups lose protons, converting to their neutral form (−NH₂), while the carboxyl groups dissociate to form negatively charged carboxylate ions (–COO⁻). This shift leads to an increase in negatively charged groups, resulting in a net negative ζ-potential (Zhao et al., 2024)

Fig. 1.

ζ-potential of SPI and GA as a function of pH.

The GA solution, an anionic polysaccharide with a pKa of around 2.2 (Razavi et al., 2021), exhibited a positive ζ-potential at pH 2.0, which became negative between pH 2.5 and 5.0. This negative ζ-otential is due to the deprotonation of carboxylate groups, which imparts a negative charge to the GA solution (Tavares & Noreña, 2020).

Therefore, pH values of 3.0, 3.5, and 4.0 were selected for the complex coacervation of SOR, using SPI and GA as wall materials. At these pH levels, the complex coacervation process is driven by electrostatic forces between the positively charged SPI and negatively charged GA. This interaction promotes charge neutralization, resulting in the formation of an insoluble complex, which effectively encapsulates the oil. The selection of these specific pH values optimizes the balance between protein and polysaccharide charges, enhancing the encapsulation efficiency and the stability of the produced microcapsules.

3.2. Complex coacervation of SOR

3.2.1. EY and EE

Table 1 presents the EY and EE of microcapsules prepared at different pH conditions: 3.0 (E3.0), 3.5 (E3.5), and 4.0 (E4.0). The encapsulation yield was highest for E3.0 and E3.5, with values of 88.38 % and 90.64 %, respectively, whereas E4.0 exhibited a lower yield of 69.17 %. In terms of encapsulation efficiency, E4.0 achieved the highest value of 81.50 %, which was not significantly different from E3.5, which had an EE of 79.71 %. Conversely, E3.0 had the lowest encapsulation efficiency at 75.49 %.

Table 1.

Physical and chemical properties of microcapsules.

Properties	E 3.0	E 3.5	E 4.0
Physical properties
Encapsulation yield (%)	88.38a ± 3.11	90.64a ± 4.64	69.17b ± 3.56
Encapsulation efficiency (%)	75.49b ± 0.34	79.71a ± 1.92	81.50a ± 0.78
Mean particle diameter (μm)	4.04b ± 0.05	7.46a ± 0.07	1.76c ± 0.08
Polydispersity Index (PDI)	0.52b ± 0.02	0.46c ± 0.00	0.86a ± 0.04
ζ-Potential (mV)	−30.95a ± 2.47	−31.65a ± 0.49	−33.30a ± 0.57
Chemical properties
Chemical compounds
Tocopherol (mg/g microcapsule)	0.52c ± 0.02	0.71b ± 0.07	1.00a ± 0.02
γ-Oryzanol (mg/g microcapsule)	7.90a ± 0.10	7.76a ± 0.21	7.72a ± 0.16
Antioxidant activity
ABTS (μmol Trolox/g microcapsule)	1186.68a ± 0.00	1208.66a ± 38.06	1236.12a ± 69.93
FRAP (μmol Trolox/g microcapsule)	7976.19a ± 109.11	7797.62a ± 203.08	7773.81a ± 161.04

Means in the same row marked with different letters are significantly different (p < 0.05).

The complex coacervation process was strongly influenced by the solution pH, which affects the surface charge of the particles and induces conformational changes in the wall materials (Knoerdel et al., 2021). Based on the EY and EE results, pH had a significant effect on the coacervation between SPI and GA. The highest EY and EE values were observed at pH 3.5, suggesting that this pH level optimizes electrostatic interactions between the polymers. At pH 3.5, the balance of charge densities is most favourable, probably corresponding to the electrical equivalence point (EEP) of the SPI-GA system. This balance enhances charge neutralization, thereby promoting the formation of a stable and efficient coacervate.

3.2.2. Tocopherol, γ-oryzanol contents and antioxidant property

The concentrations of key bioactive compounds in the microcapsules are presented in Table 1. The E4.0 formulation showed the highest tocopherol content at 1.00 mg/g sample, followed by E3.5 (0.71 mg/g sample) and E3.0 (0.52 mg/g sample). However, there was no significant difference in γ-oryzanol content among the microcapsules, with values ranging from 7.72 to 7.90 mg/g sample.

The antioxidant properties of microcapsules are summarized in Table 1. The ABTS values for the E3.0, E3.5, and E4.0 formulations were 1186.68, 1208.66, and 1236.12 μmol Trolox/g sample, respectively, with no statistically significant differences observed between the formulations. Similarly, the FRAP values did not exhibit significant variation, with results of 7976.19, 7797.62, and 7773.81 μmol Trolox/g sample for E3.0, E3.5, and E4.0, respectively.

The contrasting results could be attributed to the presence of various other antioxidants in the SOR. In addition to tocopherol and γ-oryzanol, Riceberry germ also contains other lipophilic antioxidants, including phytosterols and fat-soluble vitamins (Sangpradab et al., 2021). These additional bioactive compounds may influence the overall antioxidant property.

3.2.3. Mean particle diameter, PDI and ζ-potential

Table 1 demonstrates that the particle sizes of the E3.0, E3.5, and E4.0 formulations were 4.04, 7.46, and 1.76 μm, respectively, indicating that the pH adjustment had a significant impact on particle size. The PDI values for E3.0, E3.5, and E4.0 were 0.52, 0.46, and 0.86, respectively. High PDI values (> 0.5) observed in this study suggest a non-uniform particle size distribution, particularly in the E4.0 formulation.

At pH 3.0 and 3.5, the microcapsules exhibited large particle size due to the formation of larger SPI–GA aggregates. This occurred because the surface charge was reduced, leading to decreased electrostatic repulsion and allowing the complexes to form larger particles (Li et al., 2018). In contrast, at a pH of 4.0, the particle size distribution of the SPI–GA complexes reached a minimum. This is because the electrostatic repulsion between SPI and GA increased at these pH levels, preventing excessive aggregation and resulting in the formation of smaller microcapsules (Cao et al., 2023).

Despite the variability in particle size, the ζ-potential values for all samples ranged from −30.95 to −33.30 mV, indicating moderate electrostatic stability of the microcapsules (Fraj et al., 2019). These ζ-potential values are within the range typically associated with sufficient repulsion between particles to prevent aggregation, contributing to the overall stability of the microcapsules in suspension (Aslan Türker & Işçimen, 2025). Thus, while the particle size distribution varied, the formulations maintained a reasonable level of stability, which is crucial for their intended application.

3.2.4. Surface morphology

Fig. 2 displays the SEM images highlighting the surface morphologies of the E3.0, E3.5, and E4.0 microcapsules at magnifications of 400×, 4000×, and 10,000×. The microcapsules formed at pH 3.0 and 3.5 exhibited irregular agglomerates with larger particle sizes. This could be attributed to the weaker electrostatic interactions at these pH values (Lv et al., 2013), leading to incomplete or unstable coacervation between SPI and GA. The surface texture of these microcapsules was rough and wrinkled, suggesting non-uniform particle formation that could affect the stability and release characteristics of the microcapsules.

Fig. 2.

Scanning electron microscope (SEM) images of microcapsules prepared at pH 3.0 (E3.0), pH 3.5 (E3.5) and pH 4.0 (E4.0) at magnification 400×, 4000×, and 10,000×.

In contrast, the E4.0 microcapsules displayed smaller particle sizes with more consistent, nearly spherical shapes and smoother surface textures. This improvement can be explained by the more favourable charge density balance at pH 4.0, where the electrostatic interactions between SPI and GA are most effective, resulting in better complexation. The smoother and more uniform morphology at pH 4.0 indicates more efficient encapsulation, which could enhance both the stability and controlled release of the encapsulated bioactive compounds.

Overall, the microcapsules exhibited aggregated morphologies, which is consistent with previous study. Cao et al. (2023) investigated the effect of pH on the interaction between soy whey protein (SWP) and GA at the oil–water interface and reported that the pH of the system had a significant impact on the morphology and stability of the SWP–GA emulsion. At low pH values (2.0–4.0), the droplets gradually enlarged and partially aggregated, reflecting a less organized network structure. As the pH increased, the droplets became smaller and more uniformly distributed, with a dense network forming between them, which enhanced emulsion stability and effectively resisted aggregation.

3.2.5. FTIR spectra

FTIR was employed to analyse the chemical interactions between the core material and the carrier agents within the particles. Fig. 3 presents the FTIR spectra of UOR, SPI, GA, E3.0, E3.5, and E4.0. The UOR spectrum exhibited peaks at 2922.92 and 2855.44 cm⁻¹, corresponding to the asymmetric stretching of C—H(—CH₂). Additionally, absorption bands were observed around 1743.58, 1456.15, and 1157.80 cm⁻¹, which can be attributed to C—H(—CH₂) symmetric stretching, C O stretching, C—H(CH₂) bending, and C—O stretching, respectively (Wangsuntornpakdee et al., 2023).

Fig. 3.

FTIR spectra of UOR, SPI, GA and microcapsules prepared at different pH (E3.0, E3.5, and E4.0).

The SPI spectrum revealed several characteristic peaks associated with protein structures. The band at 3467.43 cm⁻¹ was identified as the N—H stretch within the amino group of primary amines. Moreover, peaks recorded at 1647.41 and 1541.72 cm⁻¹ were linked to the C O stretch in the amide I band and the N—H stretch in amide II, respectively (Milošević et al., 2020; Wangsuntornpakdee et al., 2023).

For GA, the spectrum demonstrated absorbance around 3434.96 cm⁻¹, indicating the O—H stretching within the carboxyl group. The peak at 2929.86 cm⁻¹ was associated with free carboxyl groups, which carry a negative charge (Sharifi et al., 2021). Additionally, bands at 1637.26 and 1420.04 cm⁻¹ corresponded to the asymmetric and symmetric stretching of carboxylate salts (—−COO⁻), while the band at 1068 cm⁻¹ was attributed to C—O—C stretching vibrations (de Gulão et al., 2016).

In the case of microcapsules, a shift in the peaks corresponding to the amino and carboxylic groups was observed at around 3400 cm⁻¹ and 2900 cm⁻¹, respectively. This shift may be attributed to the electrostatic interactions between the amino groups of SPI molecules and the carboxyl groups in GA. Additionally, the intensity of the amide II band in the microcapsules, which appears around 1500 cm⁻¹, was notably reduced compared to the SPI spectrum. On the other hand, the amide III band, ranging from 300 to 1200 cm⁻¹ (associated with C—N stretching and N—H deformation), exhibited a stronger signal. This suggests that complexation with GA not only affects the electrostatic interactions but also has a significant impact on the protein conformation, potentially altering its secondary structure (Feng et al., 2023).

3.3. Bioactive compounds and antioxidant property during in vitro gastrointestinal digestion

Fig. 4 illustrates the percentage release of bioactive compounds from UOR and microcapsules during simulated digestion. The results indicate that tocopherol and γ-oryzanol in UOR exhibited reduced stability under simulated digestion conditions. After simulated intestinal digestion, the release of tocopherol was minimal at only 1.36 %, while γ-oryzanol was completely undetectable.

Fig. 4.

Percent release of tocopherol (a), γ-oryzanol (b), ABTS (c) and FRAP (d) assays of UOR, SPI, GA and microcapsules prepared at different pH (E3.0, E3.5, and E4.0) during in vitro gastrointestinal digestion.

All microcapsule formulations exhibited significantly higher release rates for tocopherol and γ-oryzanol compared to UOR, particularly during the simulated small intestinal digestion phase. During the simulated oral digestion phase, the release of tocopherol ranged from approximately 3 % to 6 %, and γ-oryzanol from 13 % to 14 %. The brief exposure (10 min) to the oral environment was insufficient to disintegrate the microcapsule matrix. Thus, only surface oil was released during this phase, indicating that the α-amylase presented in SSF was unable to break down the coacervate matrix formed by the SPI-GA complex (Timilsena et al., 2017). In the simulated gastric digestion phase, the release of tocopherol and γ-oryzanol from microcapsules increased to 18 %–35 % and 15 %–17 %, respectively. The increased release of oil in the gastric environment can be attributed to the enzymatic breakdown of the protein shell surrounding the oil (Timilsena et al., 2017). The presence of pepsin in the gastric digestion conditions promotes the hydrolysis of some proteins present in the SPI, leading to a more porous microcapsule structure and enhanced release of the encapsulated antioxidants (Pham et al., 2021).

The majority of bioactive compounds were released during the simulated intestinal digestion phase, attributed to the reduction in the protein molecular weight, which contributes to the breakdown of the microcapsule shell (Pham et al., 2021). Among the formulations, E3.0 achieved the highest release of tocopherol (Fig. 4a), with a release rate of 81.86 %, followed by E3.5 at 52.97 % and E4.0 at 43.19 %. Conversely, E3.5 exhibited the highest release of γ-oryzanol (Fig. 4b) at 29.79 %, with E4.0 at 27.85 % and E3.0 at 26.58 %.

The ABTS assay results (Fig. 4c) show that the intestinal fraction of UOR retained only 1.01 μmol Trolox/g sample of its antioxidant property, indicating a minimal retention of activity compared to the microcapsules. In contrast, the microcapsules demonstrated improved protection of lipophilic antioxidants, with antioxidant property ranging from 8.53 to 8.98 μmol Trolox/g sample. Similarly, the release values obtained through the FRAP assay (Fig. 4d) showed that antioxidant property was undetectable in UOR during simulated intestinal digestion, highlighting its instability under these conditions. However, the microcapsules exhibited release values between 8.16 and 10.15 μmol Trolox/g sample, reflecting their enhanced ability to retain antioxidant property throughout the digestion process. During the simulated intestinal digestion at pH 7.0, SPI and GA become negatively charged. This alteration in charge weakens the electrostatic interactions between the two components, leading to the breakdown of the microcapsule shell. As a result, this structural degradation enables the gradual release of antioxidant compounds, maintaining their bioactivity (Li, Wang, & Mei, 2021).

The results demonstrate that the complex coacervation technique significantly enhances the stability of UOR during simulated gastrointestinal digestion. However, the optimal microcapsule formulation cannot be fully determined based on these findings alone. Further evaluation is necessary, considering both shelf-life stability and resilience during processing. These additional factors are essential to ensure that the microcapsules retain their structural integrity and functional properties throughout storage and various processing conditions.

3.4. pH stability

The pH of food matrices varies widely, from acidic to slightly alkaline, and the pH sensitivity of microcapsules is critical in determining their quality and stability in food and beverage systems. To address this, the stability of UOR and microcapsules was assessed under different pH conditions that mimic common food and beverage environments. In this study, the values of chemical properties were evaluated across a pH range from acidic (pH 3.0–4.0, typical of fruit juices and carbonated beverages) to neutral (pH 7.0, representative of tea, coffee, and milk), and slightly alkaline (pH 8.0–9.0, as found in alkaline water).

From the results in Fig. 5, UOR displayed considerable variability in its tocopherol and γ-oryzanol content across different pH conditions. Tocopherol levels in UOR were lowest at pH 3.0 (0.44 mg/g sample) and pH 5.0 (0.45 mg/g sample), significantly increasing at pH 7.0 (1.81 mg/g sample), and then decreasing at pH 9.0 (0.64 mg/g sample). For γ-oryzanol, UOR showed a low initial value at pH 3.0 (1.90 mg/g sample), followed by a substantial increase at pH 5.0 (4.15 mg/g sample) and a peak at pH 7.0 (22.38 mg/g sample), before decreasing at pH 9.0 (14.23 mg/g sample). Regarding antioxidant activity, UOR maintained relatively low FRAP values across all pH levels, ranging from 0.59 to 0.75 mg/g sample. However, ABTS radical scavenging activity for UOR was not detectable at pH 5.0 and 7.0, despite a highly elevated value observed at pH 9.0 (408.45 mM Trolox/g sample). This highlights the complex nature of UOR's antioxidant response under specific pH conditions, or potential limitations of the assay when applied directly to unencapsulated oil.

Fig. 5.

Release of tocopherol (a), γ-oryzanol (b), antioxidant property measured using ABTS (c) and FRAP (d) assays of UOR, SPI, GA and microcapsules prepared at different pH (E3.0, E3.5, and E4.0) under different pH environments.

All encapsulated formulations (E3.0, E3.5, and E4.0) demonstrated an enhanced ability to preserve antioxidant properties across the tested pH conditions compared to UOR. Generally, tocopherol and γ-oryzanol concentrations within the microcapsules increased as the pH shifted from acidic (pH 3.0) to neutral (pH 7.0), followed by a slight decrease at pH 9.0. At pH 3.0, tocopherol contents ranged from 0.82 mg/g sample in E3.0 to 0.91 mg/g sample in E4.0, while γ-oryzanol ranged from 0.77 mg/g sample in E3.0 to 1.45 mg/g in E4.0. The highest concentrations for both compounds were observed at pH 9.0, where tocopherol reached 3.75 mg/g sample in E3.0 and γ-oryzanol reached 4.14 mg/g sample in E3.5. Specifically, formulation E3.0 generally showed the highest tocopherol retention at pH 9.0 (3.75 mg/g sample), while E3.5 exhibited the highest γ-oryzanol content at the same pH (4.14 mg/g sample). FRAP values for encapsulated samples showed varied responses to pH. E4.0 consistently maintained higher FRAP values across acidic to neutral pH (1.12 mg/g sample at pH 3.0, 1.13 mg/g sample at pH 5.0, 1.05 mg/g sample at pH 7.0), indicating robust reducing power under these conditions. All encapsulated samples generally showed a decrease in FRAP activity at pH 9.0. Similarly, ABTS radical scavenging activity generally increased from acidic to neutral pH for encapsulated samples, with the highest values observed at pH 7.0. E4.0 demonstrated the highest ABTS activity at pH 7.0 (139.63 mM Trolox/g sample), signifying its superior radical scavenging capability under neutral conditions. Similar to FRAP, ABTS activity tended to decrease at pH 9.0 for most formulations, though E4.0 still showed strong activity (121.14 mM Trolox/g sample).

The observed trends in tocopherol and γ-oryzanol release, and antioxidant activity, can be attributed to the pH-dependent properties of the SPI-GA complex coacervate wall. The specific pH used during the coacervation process would have influenced the final structure and integrity of the microcapsule wall, thereby affecting its permeability and release characteristics in subsequent pH environments. At higher pH levels (pH 7.0 and 9.0), the increased solubility of both SPI and GA is known to induce structural changes in the proteins, such as conformational shifts and reduced steric hindrance (Heydari et al., 2021). This can lead to a more efficient release of the core materials as the wall matrix dissolves or becomes more permeable, resulting in higher measured antioxidant concentrations. However, the decrease in ABTS and FRAP antioxidant activity at pH 9.0, despite higher measured tocopherol and γ-oryzanol concentrations, suggests a more complex interplay. At elevated pH, the ionization and electron transfer properties of SPI can be altered (Mirzaei et al., 2020), potentially disrupting electrostatic interactions and hydrogen bonds within the wall structure. This disruption might lead to protein denaturation and a destabilization of the SPI-GA matrix, which could reduce the overall encapsulation efficiency or allow for increased oxidation of the released oil after its release from the microcapsule. While more antioxidants are released, their immediate activity might be compromised due to the altered environment or subsequent degradation, especially under prolonged exposure at higher pH.

These varied release profiles are essential for the functional application of the microcapsules. Formulations E3.0 and E3.5, which showed a tendency for faster release of bioactive compounds, as indicated by higher measured antioxidant content at higher pH where the wall structure is more permeable. This characteristic makes them well-suited for applications such as acidic beverages or heat-processed foods. In these systems, a rapid release of antioxidants into the food matrix is desired to provide immediate protective effects or antioxidant activity upon consumption.

Conversely, formulation E4.0 exhibited superior overall antioxidant retention and activity across a wider pH range, especially under neutral conditions, suggesting a more controlled or slower release profile. This characteristic makes E4.0 particularly suitable for applications such as functional foods and nutritional supplements. For such products, the primary goal is often to protect the bioactive compounds within the microcapsule throughout the food matrix and during storage, ensuring their delivery and bioavailability only after consumption and digestion, rather than releasing them prematurely within the food product itself. Premature release within the food matrix could expose the antioxidants to degradation, thus compromising their intended health benefits and the overall efficacy of the fortified product. This tailored release capability, driven by the pH-dependent properties of the SPI-GA encapsulation system, allows for versatile applications based on specific product and delivery requirements.

3.5. Thermal stability

UOR and microcapsule samples were exposed to convective heating at 180 °C for 0, 10, 20, and 30 min to simulate frying conditions. The thermal stability of samples was assessed by analysing key chemical properties.

Fig. 6 illustrates that during thermal treatment, the concentrations of tocopherol (Fig. 6a), γ-oryzanol (Fig. 6b), and antioxidant property, as measured by ABTS (Fig. 6c) and FRAP (Fig. 6d), were significantly lower and showed a general decline in UOR compared to the microcapsules across all heating times. For instance, tocopherol content in UOR started at 0.15 mg/g sample and ended at 0.20 mg/g sample, with γ-oryzanol at 0.10 mg/g sample reducing to 0.33 mg/g sample after 30 min. The FRAP values remained low (0.14 μmol Trolox/g sample), and ABTS activity, while initially high, decreased over time (from 103.24 to 99.56 μmol Trolox/g sample) suggesting degradation. This indicates a high susceptibility of unencapsulated oil to degradation when directly exposed to high temperatures and oxidative reactions.

Fig. 6.

Release of tocopherol (a), γ-oryzanol (b), antioxidant property measured using ABTS (c) and FRAP (d) assays of UOR, SPI, GA and microcapsules prepared at different pH: E3.0, E3.5, and E4.0 during thermal treatment at different times.

In contrast, the microcapsules generally exhibited higher levels of these compounds and antioxidant activities throughout the heating process. For the encapsulated samples, tocopherol and γ-oryzanol levels tended to increase with extended heating time, suggesting a potential protective effect or a temperature-induced release mechanism. After 30 min of heating, formulation E3.5 showed the highest tocopherol content at 2.11 mg/g sample, followed by E4.0 at 1.81 mg/g sample and E3.0 at 1.34 mg/g sample. E4.0 exhibited the highest values for γ-oryzanol (3.08 mg/g sample), ABTS (72.85 μmol Trolox/g sample), and FRAP (0.71 μmol Trolox/g sample) after 30 min of thermal treatment.

The superior retention of antioxidants in microcapsules compared to UOR during thermal treatment can be attributed to the protective barrier formed by the SPI-GA wall materials. This matrix shields the encapsulated tocopherol and γ-oryzanol from direct exposure to high temperatures and external oxygen, thereby mitigating thermal degradation. This protective effect is consistent with findings by Šaponjac et al. (2016). They reported enhanced thermal stability of SPI-based microcapsules due to high protein content of SPI forming a strong barrier. Additionally, GA has been shown to enhance the thermal resistance of microcapsules, highlighting its role as a valuable wall material for high-temperature applications (Ferreira et al., 2016).

The observed increase in antioxidant activities (tocopherol, γ-oryzanol, ABTS, and FRAP) within the microcapsule samples after heating at 180 °C can be attributed to multiple factors. Heat treatment may induce partial thermal degradation or increased permeability of the microcapsule wall, facilitating the release of previously encapsulated antioxidants into the analytical medium, which leads to higher measured values. In addition, the wall materials of the microcapsules, comprising SPI and GA, contain amino groups and reducing sugars, which can undergo Maillard reactions under high-temperature conditions. The resulting Maillard reaction products (MRPs) are known to exhibit antioxidant properties (Li, Li, et al., 2021; Kan et al., 2024), contributing additional radical-scavenging activity beyond the originally encapsulated compounds.

As reported by Zhang, Zhang, et al. (2024), performing protein–polysaccharide coacervation at temperatures that favor Maillard reactions can enhance the physicochemical and functional properties of microcapsules. This supports our observation that heating at 180 °C not only promotes thermal degradation of wall materials but also facilitates the formation of MRPs, together resulting in the measured increase in antioxidant activities. These findings highlight that both physical release of encapsulated antioxidants and generation of novel antioxidant compounds via Maillard reactions are responsible for the enhanced activity observed in the heated microcapsules.

From an application perspective, this heat-induced release profile can be advantageous for certain food products. For instance, in fried or baked goods, the goal is often to deliver antioxidants that become available during or immediately after the cooking process. The release of active compounds under thermal conditions would provide protection to the food matrix or enhance the nutritional profile of the cooked product upon consumption. However, if the primary objective is to maintain the integrity of the microcapsule within the food matrix throughout processing and storage, then significant release during heating could compromise the long-term protective efficacy of the encapsulation. Therefore, understanding this nuanced heat-induced release behaviour is crucial for selecting the appropriate microcapsule formulation based on the specific processing conditions and desired end-product function.

3.6. Shelf-life stability

The results shown in Fig. 7a indicated that complex coacervation significantly improved the shelf-life stability of soybean oil over a 45-day storage period at room temperature, which otherwise exhibited low release rates of tocopherol, γ-oryzanol, ABTS, and FRAP activity at 1.77 %, 27.84 %, 29.76 %, and 48.40 %, respectively. In contrast, microcapsules formulated with E3.0 and E3.5 demonstrated high release rates, with tocopherol at 85.36 % and 76.51 %, respectively, and γ-oryzanol at 79.37 % and 73.38 %, respectively. Additionally, all microcapsules showed elevated antioxidant property, with ABTS values ranging from 91.34 % to 95.38 % and FRAP values ranging from 82.21 % to 85.27 %. These findings underscore the efficacy of the complex coacervation technique in enhancing the stability and retention of bioactive compounds during storage.

Fig. 7.

Percent release of tocopherol, γ-oryzanol and antioxidant property measured using ABTS and FRAP assays (a) and peroxide value (b) of UOR, SPI, GA and microcapsules prepared at different pH: E3.0, E3.5, and E4.0 after 45-day storage at 25 °C.

To assess oxidative stability under ambient conditions, UOR and microcapsules were stored at 25 °C for 45 days. The PV of UOR and E3.0, E3.5, and E4.0 formulations were measured on day 0 and day 45 (Fig. 7b).

At the beginning of the storage period, the PV of UOR was 5.70 meq PV/kg oil. This initial value was lower than that of E3.0 (7.97 meq PV/kg oil) and E3.5 (9.99 meq PV/kg oil), but slightly higher than E4.0 (5.99 meq PV/kg oil). However, after 45 days, UOR showed a dramatic increase in PV, reaching 41.43 meq PV/kg oil. This clearly indicates severe lipid oxidation in the absence of encapsulation. In contrast, encapsulated formulations exhibited significantly lower PVs: 14.99 meq PV/kg oil for E3.0, 13.98 meq PV/kg oil for E3.5, and 7.99 meq PV/kg oil for E4.0.

Among the encapsulated samples, E4.0 demonstrated the greatest oxidative stability. It maintained the lowest PV on day 45, with its value remaining within the acceptable limit established by the Codex Alimentarius (10 meq PV/kg oil; CXS 210–1999). These results clearly demonstrate the protective effect of encapsulation, particularly for formulation E4.0, in significantly reducing oxidative degradation during storage at room temperature. The sharp increase in PV observed in UOR further emphasizes the high susceptibility of unencapsulated oil to oxidation when directly exposed to atmospheric oxygen.

The lower PV observed in microcapsules indicates that the SPI–GA matrix acts as an effective oxygen barrier, reducing diffusion and protecting bioactive compounds from degradation during storage. This result is consistent with previous researches showing that microencapsulation enhances oil stability. Ferreira et al. (2016) showed that crude palm oil microencapsulated with cassava starch and GA maintained good stability during 5 weeks of storage at 45 °C, with PV increasing only slightly from 11.16 to 12.54 meq/kg oil. Similarly, Wang et al. (2025) reported that tuna oil microcapsules prepared using whey protein isolate and GA exhibited lower PV (4.93 meq/kg oil) compared to free oil (3.59 meq/kg oil) after 16 days of storage at 55 °C. These findings confirm that encapsulation provides an effective barrier against oxygen diffusion and contributes both physical protection and antioxidant effects, thereby enhancing oil stability and extending shelf life.

Compared to these studies, our findings further demonstrate that formulation parameters, such as pH during coacervation, can fine-tune the protective effect of microcapsules, with E4.0 showing superior performance in maintaining antioxidant activity and oxidative stability.

4. Conclusion

The highest encapsulation yield (90.64 %) and efficiency (79.71 %) were achieved at pH 3.5, while the smallest and most uniform microcapsules formed at pH 4.0. The microcapsules exhibited stable ζ-potential values, with FTIR confirming key electrostatic interactions between SPI and GA. They effectively preserved antioxidants and displayed controlled, digestion-dependent release, with minimal release in oral and gastric phases and predominant release in the intestinal phase, enhancing bioavailability. Compared with unencapsulated oil, the microcapsules showed improved oxidative stability and higher retention of bioactive compounds. Formulations E3.0 and E3.5 favored rapid antioxidant availability, whereas E4.0 provided more sustained protection, suitable for functional foods and supplements. Shelf-life evaluation further confirmed the enhanced stability of encapsulated antioxidants. Depending on the preparation pH, formulations can be tailored for specific applications, such as rapid antioxidant availability in acidic or heat-processed foods, or prolonged protection and targeted release in functional foods and nutritional supplements.

CRediT authorship contribution statement

Palita Wangsuntornpakdee: Writing – original draft, Investigation, Formal analysis, Conceptualization. Tanwarat Aksornsri: Writing – original draft, Investigation, Formal analysis. Utai Klinkesorn: Writing – review & editing, Supervision. Satoshi Iwamoto: Writing – review & editing, Supervision. Methavee Peanparkdee: Writing – review & editing, Supervision, Methodology, 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.

Appendix A. Supplementary data

Supplementary Table S1

Data availability

Data will be made available on request.