Effects of chlorogenic acid on the interfacial properties and physical stability of sunflower oil body emulsions: Insights from a reconstructed protein-emulsion model

Abstract

The stability of sunflower oil body emulsions is mainly governed by interfacial phospholipids and proteins, whereas the role of endogenous polyphenols remains unclear. This study aimed to evaluate the effects of chlorogenic acid (CGA) on physical stability. Reconstructed oil body proteins (ROBPs) and emulsions (ROBEs) were used as model systems. CGA decreased β-turn content from 43.06% in 0 mg-CGA-ROBPs to 26.50% in 38 mg-CGA-ROBPs. For ROBPs, as CGA addition increased from 0 to 38 mg, the emulsifying activity index increased from 27.91 to 77.32 m²/g and the emulsion stability index from 11.04 to 57.37 min. At 38 mg CGA, ROBEs reached a zeta potential of −47.13 mV and an interfacial protein concentration of 43.18 mg/m², together with increased viscosity and viscoelasticity and improved physical stability. These findings provide insight into how endogenous CGA affects protein conformation and emulsion stability and may guide the formulation of stable oil body-based foods.

1. Introduction

Oil body emulsions have attracted increasing attention in the food industry owing to their intact structure, excellent biocompatibility, and self-emulsifying ability without external emulsifiers (Hao et al., 2022; Zaaboul et al., 2022). These unique characteristics allow oil body emulsions to serve as natural emulsion systems in the development of food products, such as cheese, cream, and salad dressings, as well as delivery systems for lipophilic bioactive compounds, including curcumin and carotenoids (Shi et al., 2024; Zhang, McClements, et al., 2025). However, extracted oil bodies are sensitive to processing and storage conditions. Changes in pH, temperature, and ionic conditions can affect the surface charge and arrangement of interfacial proteins and may also alter protein–phospholipid interactions, leading to droplet aggregation, coalescence, or even phase separation (Abdullah et al., 2020; Ding et al., 2020; Romero-Guzmán et al., 2020). The physical stability of oil bodies therefore depends largely on the composition and structure of the oil–water interface. Proteins and phospholipids are generally regarded as the main components responsible for maintaining interfacial integrity (Li et al., 2022; Yang, Plankensteiner, et al., 2025). Recent studies further suggest that other endogenous components may also contribute to oil body stability.

The endogenous components associated with natural oil body emulsions vary among oilseed species. For example, in camellia seed and pumpkin seed oil body emulsions, endogenous compounds such as tea saponins and protochlorophyll have been reported to interact with proteins, polysaccharides, and other constituents at the oil-water interface, thereby affecting emulsion stability (Cui et al., 2023; Lu et al., 2023). Similarly, pectic polysaccharides released from rapeseed cell walls during wet milling can be co-extracted with oil bodies and interact with proteins, thereby influencing oil body aggregation and resistance to coalescence (Bleibach Alpiger & Corredig, 2024). These findings highlight the potential contribution of endogenous components to oil body emulsion stability. Sunflower is one of the major oilseed crops worldwide, and its seeds are rich in oil and protein, making them an important source of plant oil bodies (Laaraj et al., 2025; Murphy, 2025). In addition to oil and protein, sunflower seeds contain various phenolic compounds, among which chlorogenic acid (CGA) is one of the major phenolic constituents (Laaraj et al., 2025; Tarahi et al., 2024). Previous studies have shown that the addition of polyphenols, such as tannic acid, quercetin, and chlorogenic acid, can alter the structural and functional properties of proteins, thereby affecting oil body emulsion stability (Farooq et al., 2022; Gao et al., 2024; Liao et al., 2024). In sunflower protein systems, exogenous CGA has also been reported to alter protein structure and functionality (Karefyllakis et al., 2017; Karefyllakis et al., 2018). However, studies based on exogenous polyphenol addition cannot fully reveal the specific contribution of endogenous polyphenols within the multicomponent interface of oil body emulsions.

In recent years, the role of endogenous polyphenols in oil bodies has received increasing attention. Li and Zhou (2025) compared walnut oil body emulsions prepared with and without removal of the pellicle and found that endogenous polyphenols influenced their physical stability and rheological properties. For sunflower oil body emulsions, Plankensteiner et al. (2025) investigated the combined effects of co-extracted phenolic compounds and storage proteins on the oxidative stability of sunflower oil body emulsions. These studies support the potential role of endogenous polyphenols in oil body emulsions. However, because polyphenols, proteins, phospholipids, and other components coexist and interact in natural oil bodies, the contributions of individual endogenous polyphenols to emulsion stability cannot be readily distinguished. Reconstructed oil body emulsions allow the composition and proportions of specific components to be adjusted under relatively controlled conditions, providing a practical approach to investigating their effects on emulsion properties (Yuan et al., 2024). Using reconstructed rapeseed oil bodies, Deleu et al. (2010) demonstrated that phospholipids and proteins jointly contribute to interfacial stabilization. Yang et al. (2025) further reported that phosphatidylcholine (PC) plays an important role in maintaining the stability of reconstructed peanut oil body emulsions. Although reconstructed systems do not fully reproduce the complex interfacial composition of natural oil bodies, they can help distinguish the contributions of specific components to oil body emulsion properties. Accordingly, a reconstructed oil body system provides a useful approach to evaluating the contribution of CGA to the properties and physical stability of sunflower oil body emulsions.

Our previous study showed that sunflower oil body emulsions extracted using the AM-NaHCO₃ method exhibited a distinct green color, suggesting that changes in endogenous polyphenols during extraction may be associated with the properties of oil body emulsions (Guan et al., 2026). In the present study, the contents of CGA and its isomers in oil body proteins obtained using different extraction methods were first compared. Based on its pronounced variation among the extraction methods, CGA was selected for further investigation, and reconstructed oil body proteins (ROBPs) and reconstructed oil body emulsions (ROBEs) were prepared. The effects of CGA on the properties of oil body proteins and the physical stability of the emulsions were then investigated using the reconstructed systems.

2. Materials and methods

2.1. Materials

Sunflower seeds used in this study were provided by Alashankou Jinmu Biotechnology Co., Ltd. (Xinjiang, China). Fluorescein isothiocyanate (FITC), Nile red, and three chlorogenic acid isomers, including chlorogenic acid (CAS 327–97-9), neochlorogenic acid (CAS 906–33-2), and cryptochlorogenic acid (CAS 905–99-7), all with purities above 98%, were purchased from Yuanye Bio-Technology Co., Ltd. (Shanghai, China). Analytical-grade chemicals and reagents were commercially obtained unless otherwise noted.

2.2. Extraction of sunflower oil bodies and oil body proteins

Sunflower oil body emulsions and oil body proteins were prepared according to our previously reported method (Guan et al., 2026). Briefly, dehulled sunflower seeds were soaked in deionized water at 4 °C for 12 h and extracted using water (AM), 1.5% Viscozyme L (AEM), 0.1 M NaHCO₃ (AM-NaHCO₃), or 0.1 M NaCl (AM-NaCl). The mixtures were then ground, incubated at 50 °C for 1 h, and centrifuged to obtain the oil body emulsions. The collected emulsions were freeze-dried and defatted to obtain oil body proteins.

2.3. Identification of polyphenols from oil body proteins

(1) Extraction of total polyphenols: Polyphenols were extracted from oil body proteins using 50% methanol at a solid-to-solvent ratio of 1:50 (w/v) under continuous stirring at 25 °C for 4 h, followed by centrifugation (5000 rpm, 10 min) to recover the supernatant (Yan & Zhou, 2021). The residue was repeatedly extracted under the same conditions until no detectable phenolic content was observed in the supernatant, as determined by the Folin-Ciocalteu method according to Singleton and Rossi (1965). All supernatants were combined to obtain the total polyphenol extract.

(2) HPLC conditions: Polyphenols were separated on an Agilent Zorbax SB-C18 column (4.6 × 250 mm, 5 μm, Agilent Technologies, Santa Clara, CA, USA). The mobile phase consisted of acetonitrile (A) and 0.1% formic acid in water (B). Gradient separation was carried out according to the elution program specified in GB/T 43733–2024. The flow rate was 1.0 mL/min, the column temperature was maintained at 30 °C, the detection wavelength was 327 nm, and the injection volume was 10 μL (Wang, Cao, et al., 2024).

(3) Preparation of standard curves: The three chlorogenic acid standards were dissolved in 50% methanol to prepare solutions at concentrations ranging from 5 to 60 μg/mL. After filtration through a 0.22 μm membrane filter, each standard solution was analyzed under the HPLC conditions described above. Standard curves were constructed by plotting peak area (y) against concentration (x). The regression equations were as follows: Chlorogenic acid: y = 25,756.725× – 13,339.375 (R² = 0.9995); neochlorogenic acid: y = 25,032.9974× – 10,278.41 (R² = 0.9998); cryptochlorogenic acid: y = 26,222.5517× – 12,554.86 (R² = 0.9997).

(4) Polyphenol quantification: After membrane filtration (0.22 μm), the total polyphenol extract was subjected to chromatographic analysis under the previously described conditions. Individual polyphenol contents were quantified using the corresponding standard calibration curves.

2.4. Preparation of Dephenolized oil body proteins

The AM-OBP sample used for ROBE reconstruction was subjected to repeated polyphenol extraction as described in Section 2.3. The resulting solid residue was collected and freeze-dried to obtain dephenolized oil body proteins, which were subsequently used to prepare the reconstructed oil body emulsions.

2.5. Preparation of reconstructed oil body proteins (ROBPs) and reconstructed oil body emulsions (ROBEs)

Based on the compositional characteristics of sunflower oil bodies (moisture 51.97 ± 0.14%, protein 3.98 ± 0.05%, fat 41.79 ± 0.49%) and the highest CGA content detected in oil body proteins obtained using different extraction methods (approximately 12 mg/g protein, Table 1), reconstructed oil body emulsions (ROBEs) with different CGA addition levels were prepared. For each 80-mL ROBE formulation, 3.2 g of dephenolized oil body protein was used. Therefore, the maximum CGA addition was calculated as 12 mg/g protein × 3.2 g protein = 38.4 mg and was rounded to 38 mg. Accordingly, the CGA addition levels were set at 0, 8, 18, 28, and 38 mg.

Table 1.

Polyphenol content (mg/g) in sunflower oil body proteins extracted by different methods.

Extraction method	Neochlorogenic acid	Chlorogenic acid	Cryptochlorogenic acid
AM-OBP	2.79 ± 0.03ab	11.35 ± 0.12c	1.22 ± 0.01b
AEM-OBP	2.88 ± 0.04a	11.63 ± 0.14b	1.20 ± 0.02b
AM-NaHCO3-OBP	2.73 ± 0.04b	3.44 ± 0.06d	2.84 ± 0.02a
AM-NaCl-OBP	2.71 ± 0.08b	11.89 ± 0.07a	1.21 ± 0.02b

Note: Values are expressed as mean ± SD (n = 3). Different letters within the same column indicate significant differences among extraction methods for each polyphenol (p < 0.05). Here, mg/g refers to mg of polyphenol (e.g., chlorogenic acid) per g of oil body protein. The markedly lower CGA content in AM-NaHCO₃-OBP compared with other extraction methods may be associated with the transformation of CGA under alkaline extraction conditions and possible oxidation reactions during extraction, which could influence its content.

Abbreviations: AM-OBP, aqueous method oil body protein; AEM-OBP, aqueous enzymatic method oil body protein; AM-NaCl-OBP, aqueous method with NaCl; AM-NaHCO₃-OBP, aqueous method with NaHCO₃.

Briefly, 3.2 g of dephenolized oil body protein was weighed into a 100 mL beaker, after which CGA was added. Subsequently, 41.6 g of NaHCO₃ solution (pH 8.0) was added. The same concentration and amount of NaHCO₃ solution were used in all treatments to maintain consistent alkaline conditions across the samples. The mixtures were stirred at room temperature for 16 h under alkaline conditions to allow interaction between CGA and the oil body proteins. The final pH values remained consistent across all treatment groups. The mixed solutions were freeze-dried to obtain reconstructed oil body proteins (0, 8, 18, 28, and 38 mg-CGA-ROBPs). Separately, mixed solutions prepared under identical conditions were blended with 33.6 g of sunflower oil, sheared at 10,000 rpm for 2 min, and homogenized for three cycles at 40 MPa to obtain reconstructed oil body emulsions (0, 8, 18, 28, and 38 mg-CGA-ROBEs) (Fig. 1) (Yang, Li, et al., 2025).

Fig. 1.

Schematic illustration of the preparation process for reconstructed oil body emulsions.

2.6. Characterization of ROBEs

2.6.1. Color

The color of the ROBEs was assessed using a colorimeter (CR-400, Konica Minolta Holdings Co., Ltd., Japan), which was first calibrated against a standard white plate. Sample aliquots of 20 mL were placed in the measuring cell under low-light conditions, and the L*, a*, and b* values were recorded (Wang et al., 2021).

2.6.2. Particle size

Particle size distributions were determined by laser diffraction (BT-9300H, Dandong Baxter, China) after 1 g of each ROBE was dispersed in 10 mL of deionized water. All measurements were conducted at 25 °C, using a sample refractive index of 1.46 and a dispersant (water) refractive index of 1.33, with obscuration maintained between 8 and 10% (Wang et al., 2020).

2.6.3. Zeta potential

Prior to zeta potential measurement, emulsion samples were diluted 100-fold with deionized water and thoroughly vortexed. Analysis was carried out at 25 °C using a Zetasizer Nano ZSP (Malvern Instruments, UK), with refractive index values of 1.47 for the dispersed phase and 1.33 for the continuous phase. An equilibration period of 60 s was allowed before each measurement (Lv et al., 2020).

2.6.4. Electrical conductivity

The electrical conductivity of ROBEs (0, 8, 18, 28, and 38 mg-CGA-ROBEs) was measured directly without dilution using a DDS-11 A conductivity meter (INESA Scientific Instrument Co., Ltd., Shanghai, China).

2.6.5. Multiple light scattering

For stability analysis, 20 mL of each ROBE was placed in a cylindrical glass tube without visible air bubbles. Measurements were conducted using an MS20 Stability Analyzer (Beijing Dataphys Instruments Co., Ltd., China). Dynamic monitoring was conducted at 25 °C for 120 min, with scans along the tube height every 100 s. Transmittance (0°) and backscattering (135°) signals were recorded simultaneously (Wang et al., 2020).

2.6.6. Laser confocal scanning microscopy (CLSM)

Prior to microscopic observation, ROBEs were diluted 100-fold with deionized water. Subsequently, 1 mL of the diluted suspension was incubated with a fluorescent dye mixture consisting of FITC (10 μL, 1 mg/mL in acetone) and Nile Red (10 μL, 0.1 mg/mL in ethanol). After incubation in the dark for 15 min, a 10 μL aliquot was mounted on a glass slide for imaging. CLSM (FV3000, Olympus, Tokyo, Japan) was used for observation. FITC and Nile Red were excited at 488 nm and 545 nm, respectively, with emission collected at 510–530 nm (green) and 570–620 nm (red) (Lv et al., 2020).

2.6.7. Rheological properties

Rheological characterization of the ROBEs was performed using a rheometer (Discovery HR-10, TA Instruments, USA) equipped with a parallel-plate geometry (40 mm diameter, 1 mm gap). Each sample was loaded onto the lower plate, and viscosity was measured over a shear rate range of 0.1–100 s⁻¹ at 25 °C. A strain sweep test was first conducted to determine the linear viscoelastic region (LVR). A strain amplitude of 0.5% was selected for subsequent frequency sweep tests conducted over a frequency range of 0.1–100 Hz (Feng et al., 2023).

2.6.8. Interfacial protein concentration and percentage of adsorbed proteins

Fresh ROBEs were centrifuged in 10 mL tubes at 5000 rpm for 30 min to achieve phase separation. The aqueous fraction was withdrawn using a syringe and filtered through a 0.45 μm membrane. Protein contents in the emulsion and the aqueous phases were quantified using the Kjeldahl and biuret methods, respectively (Guo et al., 2021). Interfacial protein concentration (Γ) and the percentage of adsorbed proteins (AP) were calculated using Eqs. (1), (2):

$$\mathrm{\Gamma }\left(\frac{\mathit{mg}}{m^2}\right)=\frac{\left({\mathrm{C}}_{\mathrm{T}}-{\mathrm{C}}_{\mathrm{A}}\right)\times {\mathrm{d}}_{3,2}}{6\times \mathrm{\phi }}$$ (1)

$$\mathit{AP}\left(\%\right)=\frac{\left(C_T-C_A\right)\times 100}{C_T}$$ (2)

where C_T and C_A represent the total protein concentration and the protein concentration in the aqueous phase after centrifugation, respectively (mg/mL). d_3,2 denotes the surface-area-weighted mean diameter, while ϕ refers to the oil volume fraction.

2.6.9. Contact angle

Contact angle measurements were performed on ROBEs using a contact angle analyzer (OCA20, DataPhysics Instruments GmbH, Germany). Diluted emulsions (1%, w/v) were coated onto glass slides and air-dried for 12–16 h at room temperature. The resulting films were subjected to sessile drop testing at 25 °C. A 7.5 μL water droplet was placed on each film, and its profile was imaged. The contact angle was then determined by fitting each profile to the Young-Laplace equation (Hao et al., 2024).

2.6.10. Interfacial tension

Interfacial tension of the ROBEs was determined with a contact angle analyzer (OCA20, DataPhysics Instruments GmbH, Germany). Emulsions were diluted to 1% (w/v) in deionized water and loaded into a syringe fitted with a stainless-steel needle (inner diameter 1.823 mm). The needle was placed in a quartz cuvette containing sunflower oil, and a 30 μL drop was formed at the tip by an autosampler. Interfacial tension was then recorded by the pendant drop technique at 10 s intervals for 120 min, yielding 720 measurements (Hao et al., 2024).

2.6.11. Environmental stability of ROBEs

The environmental stability of ROBEs containing different amounts of CGA was evaluated under salt, pH, thermal, and freeze–thaw stresses (Geng et al., 2024). For salt treatment, NaCl was added to the samples to obtain final concentrations of 0, 100, and 300 mmol/L. For pH treatment, the samples were adjusted to pH 3.0 or 7.0 using 0.01 mol/L HCl or NaOH. For thermal treatment, the samples were heated at 60 or 90 °C for 30 min and then cooled to room temperature. For freeze–thaw treatment, the samples were frozen at −20 °C for 22 h and subsequently thawed at 37 °C for 2 h, constituting one freeze–thaw cycle. After each treatment, the particle size of the ROBEs was measured as described in Section 2.6.2.

2.7. Characterization of ROBPs

2.7.1. Zeta potential

The zeta potential of ROBPs was determined according to Lv et al. (2020). Protein samples were prepared as a 10 mg/mL solution in deionized water. Zeta potential was determined using a Zetasizer (Nano ZSP, Malvern Instruments Ltd., Malvern, UK), following the procedure described in Section 2.6.3.

2.7.2. Scanning Electron microscopy (SEM)

A small amount of ROBPs (1–2 mg) was mounted onto the sample stage using tweezers and conductive double-sided tape. The samples were then coated with a thin layer of gold via ion sputtering (10 mA, 30 s) to improve conductivity. After positioning to obtain an optimal field of view, microstructural morphology was observed using a scanning electron microscope (Quanta 250 FEG, FEI Corporation, USA) at an accelerating voltage of 10 kV and a magnification of 1000× (Yan & Zhou, 2021).

2.7.3. SDS-page

ROBPs were prepared as 5 mg/mL solutions in a reducing SDS buffer (0.1 M Tris-HCl at pH 6.8, 2% SDS, 0.025% bromophenol blue, 20% glycerol, and 10% β-mercaptoethanol). Samples were vortexed in 1.5 mL tubes until fully dispersed, heated at 100 °C for 5 min, and then cooled to room temperature before loading. For each lane, 10 μL of sample or marker was applied. The stacking gel (5%) was run at 60 V, and the separating gel (12%) at 100 V. The run continued until the dye front had moved to within 0.5–1 cm of the gel bottom. After electrophoresis, the gel was stained for 2 h with Coomassie Brilliant Blue R-250 (2.5 mg/mL in 35% ethanol and 7% acetic acid), and then destained with a dye-free ethanol/acetic acid/water mixture until the background became clear. The gel was then imaged (Guo et al., 2021).

2.7.4. Fourier transform infrared (FTIR) spectroscopy

For FTIR analysis, 2 mg of ROBPs was thoroughly ground with 200 mg of anhydrous KBr in an agate mortar under an infrared lamp, and the mixture was pressed into transparent pellets at a pressure of 60 MPa. Spectra were recorded using an FTIR spectrometer (Shanghai Scientific Instrument Co., Ltd., China) over the range of 400–4000 cm⁻¹ at a resolution of 4 cm⁻¹ with 32 scans per sample. A KBr pellet without sample was used as the background. The amide I region (1600–1700 cm⁻¹) was analyzed using OMNIC and PeakFit software for curve fitting and deconvolution, and secondary structure contents were subsequently determined (Guo et al., 2021).

2.7.5. Intrinsic fluorescence spectroscopy

ROBPs were suspended in 10 mM phosphate-buffered saline (PBS, pH 7.0) at 0.2 mg/mL, vortexed thoroughly, and transferred into a quartz cuvette. Fluorescence spectra were recorded on a spectrofluorometer (F-7000, Hitachi, Japan) at an excitation wavelength of 280 nm, and emission was monitored over the range of 290 to 500 nm with slit widths set to 5 nm (Karefyllakis et al., 2018).

2.7.6. Surface hydrophobicity

Protein samples were diluted with 0.1 M phosphate buffer (pH 7.0) to reach final concentrations of 0, 0.05, 0.10, 0.15, 0.20, and 0.25 mg/mL. A 5 mL aliquot of each dilution was mixed with 50 μL of 8 mM ANS, followed by gentle vortexing for 30 s. The mixtures were incubated in the dark for 15 min before fluorescence measurements. The experimental protocol followed that described by Guan et al. (2026).

2.7.7. Emulsifying activity index (EAI) and emulsion stability index (ESI)

ROBPs were suspended in 0.1 M phosphate buffer (pH 7.0) at 5 mg/mL and stirred at 500 rpm for 2 h at 25 °C. The resulting protein dispersion was mixed with sunflower oil at a 3:1 (v/v) ratio and homogenized at 10,000 rpm for 2 min to prepare an emulsion. The freshly prepared emulsion was diluted 100-fold with 0.1% SDS either immediately or after standing for 10 min. Absorbance was measured at 500 nm, and the EAI and ESI were calculated using the equations described by Lu et al. (2023).

$$\mathrm{EAI}\left(\raisebox{1ex}{${\mathrm{m}}^2$}\!\left/ \!\raisebox{-1ex}{$\mathrm{g}$}\right.\right)=\frac{2\times 2.303\times A_0\times \mathit{DF}\times 10⁻⁶}{C\times 0.25\times L}$$ (3)

$$\mathrm{ESI}\left(\min \right)=\frac{A_0\times 10}{A_0-A_{10}}$$ (4)

where A₀ and A₁₀ are the absorbance values recorded at 0 and 10 min, respectively, while DF indicates the dilution factor. C corresponds to the protein concentration (g/mL), and L represents the optical path length (0.01 m).

2.8. Statistical analysis

All experiments were conducted using three independently prepared samples for each treatment (n = 3). Data are expressed as mean ± standard deviation (SD). SPSS 26.0 was used for statistical analysis. One-way analysis of variance (ANOVA) followed by Duncan's multiple range test was used to evaluate significant differences among means (p < 0.05). Figures were plotted with Origin 2021.

3. Results and discussion

3.1. Identification of polyphenols in sunflower oil body proteins

As shown in Table 1, chlorogenic acid (CGA), neochlorogenic acid, and cryptochlorogenic acid were detected in oil body proteins (OBPs) obtained using different extraction methods. Among the detected phenolic compounds, CGA was the predominant component in OBPs. The extractable CGA content in AM-NaHCO₃-OBP was 3.44 ± 0.06 mg/g, which was significantly lower than that in AM-OBP, AEM-OBP, and AM-NaCl-OBP, indicating that extraction conditions markedly affected the phenolic composition of OBPs. The highest CGA content detected in the OBP samples was approximately 12 mg/g (Table 1). Accordingly, 38 mg was selected as the maximum CGA addition level for the reconstructed formulations. The lower CGA content in AM-NaHCO₃-OBP may be associated with altered CGA extractability under alkaline conditions and possible interactions between CGA and OBPs, which could affect CGA distribution during extraction (Guan et al., 2026). Neochlorogenic acid and cryptochlorogenic acid showed relatively minor variations across extraction methods. Among these detected phenolic compounds, CGA exhibited the highest abundance and the most pronounced changes under different extraction conditions. Therefore, although CGA does not fully represent the diversity of endogenous phenolic compounds naturally present in sunflower oil bodies, it was selected as the target compound for subsequent reconstruction studies.

3.2. Color of the ROBEs

Color in emulsions arises from the combined effects of light absorption by chromophores and light scattering from dispersed droplets. As shown in Table 2, the L*, a*, and b* values changed from 63.12 ± 0.19, 0.66 ± 0.01, and 6.89 ± 0.01 to 58.36 ± 0.09, −3.91 ± 0.04, and 3.79 ± 0.03, respectively, as the amount of added CGA increased. The shift in a* from a positive to a negative value indicates that the ROBEs became greener. Under alkaline conditions, CGA is susceptible to oxidation and may generate oxidation-related products, which could alter light absorption in the visible region. Possible interactions between these oxidation products and oil body proteins may also influence the color development of the ROBEs (Akbulut et al., 2012; Xu et al., 2023). The decrease in L* indicates reduced emulsion lightness, which may reflect changes in light absorption and scattering associated with CGA addition and the accompanying alterations in droplet size and interfacial organization. The decrease in b* value may be associated with interactions between the aromatic structures of CGA and endogenous pigments or protein chromophores, which may influence the optical properties of ROBEs (Wildermuth et al., 2016; Xu et al., 2023).

Table 2.

Effect of chlorogenic acid (CGA) addition on the color of reconstructed oil body emulsions.

CGA addition (mg)	L*	a*	b*
0	63.12 ± 0.19a	0.66 ± 0.01a	6.89 ± 0.01a
8	62.26 ± 0.29b	−1.07 ± 0.01b	6.02 ± 0.01b
18	60.22 ± 0.09c	−2.70 ± 0.01c	4.98 ± 0.01c
28	59.55 ± 0.18d	−3.52 ± 0.09d	4.21 ± 0.01d
38	58.36 ± 0.09e	−3.91 ± 0.04e	3.79 ± 0.03e

Note: Values are expressed as mean ± SD (n = 3). Different letters within the same column indicate significant differences (p < 0.05). L*, a*, and b* represent the CIE color parameters of the reconstructed oil body emulsions.

3.3. Physicochemical properties of ROBEs

As shown in Fig. 2A, the ROBEs exhibited a gradual transition from milky white to green with increasing CGA addition (0, 8, 18, 28, and 38 mg), consistent with the shift in a* from a positive to a negative value (Table 2). The particle size distribution of the ROBEs exhibited a bimodal pattern (Fig. 2B), with the first peak at 1–10 μm and the second peak at 10–100 μm. With increasing CGA addition, the second peak shifted toward smaller particle sizes. Consistently, as shown in Fig. 2C, the highest CGA addition (38 mg) significantly reduced the particle size to 18.09 μm and the particle size distribution width (Span) to 2.15 (p < 0.05). The reduction in particle size may be related to changes in the interfacial behavior of OBPs after CGA addition. These changes could improve the coverage of newly formed oil-water interfaces and limit droplet recoalescence. Moreover, the zeta potential of the ROBEs became more negative with increasing CGA addition, reaching −47.13 mV (Fig. 2D, p < 0.05). Under alkaline conditions, CGA may undergo both deprotonation and oxidation. Deprotonation of CGA may produce negatively charged carboxylate and phenolate groups, whereas quinone derivatives formed through CGA oxidation may interact with OBPs. These processes could alter the surface charge characteristics of the ROBEs, contributing to their more negative zeta potential. The improved physical stability of ROBEs may be related to the decreased particle size and stronger electrostatic repulsion between droplets (Li et al., 2020; Yan, Xie, et al., 2021). Tian et al. (2021) also demonstrated that tea polyphenols at low levels (0.01%) can improve emulsion physical stability. As shown in Fig. 2E, the electrical conductivity of the ROBEs significantly decreased with increasing CGA addition (p < 0.05). This decrease may reflect reduced mobility of charged components, possibly associated with CGA-protein interactions (Guo & Xiong, 2021).

Fig. 2.

Effects of CGA addition on the physicochemical properties of reconstructed oil body emulsions. (A) appearance; (B) particle size distribution; (C) particle size (μm) and Span (particle size distribution width); (D) zeta potential (mV); (E) electrical conductivity (mS/cm).

3.4. Rheological properties of ROBEs

All ROBEs exhibited shear-thinning behavior, with viscosity progressively decreasing as the shear rate increased. The initial viscosity of the 0 mg-CGA-ROBEs was lower than that of ROBEs with CGA additions (Fig. 3A). The higher viscosity of the ROBEs with CGA addition indicates greater resistance to droplet movement, which could reduce droplet collisions and recoalescence, consistent with the progressive decrease in particle size with increasing CGA addition observed in Fig. 2 (Xie et al., 2023; Zhang, Li, et al., 2025). As seen in Fig. 3B, G′ and G″ both increased with frequency (0.1–100 Hz), and G′ remained above G″ over the whole range, indicating a weak-gel-like behavior dominated by elastic response. The absence of crossover between G′ and G″ suggests that elastic-dominant behavior was maintained throughout the tested frequency range. The smaller difference between G′ and G″ in the 0 mg-CGA-ROBEs suggests weaker elastic dominance, which may reflect less pronounced structural reinforcement in the absence of CGA (Wang, Liu, et al., 2024; Xie et al., 2023). Overall, the higher viscosity and stronger elastic response of the ROBEs with CGA addition indicate greater resistance to flow and deformation, thereby supporting their improved physical stability.

Fig. 3.

Effects of chlorogenic acid (CGA) addition on the rheological properties of reconstructed oil body emulsions. (A) Viscosity (Pa·s); (B) Frequency sweep showing storage modulus (G', MPa) and loss modulus (G", MPa).

3.5. Interfacial properties of ROBEs

As illustrated in Fig. 4A and B, CGA addition led to gradual increases in both the interfacial protein concentration and the percentage of adsorbed protein in ROBEs (p < 0.05). The increased interfacial protein concentration may be associated with changes in OBP conformation and organization at the oil-water interface following CGA addition. The conformational changes of OBPs may modify the exposure of hydrophilic and hydrophobic regions on the protein surface, which could alter the amphiphilic balance of OBPs and favor adsorption and retention at the interface (Chang et al., 2026; Liao et al., 2024). However, the contribution of other factors, such as protein redistribution or aggregation, cannot be completely excluded.

Fig. 4.

Effects of CGA addition on the interfacial properties of reconstructed oil body emulsions. (A) Interfacial Protein Concentration (mg/m²); (B) Percentage of Adsorbed Proteins (%); (C) Contact Angle (°); (D) Interfacial Tension (mN/m).

As shown in Fig. 4C, the contact angles of all ROBEs were below 90° and decreased with increasing CGA addition. This decrease may be associated with CGA-induced changes in the conformation and surface properties of OBPs. Accordingly, the lower contact angles indicate increased surface wettability of reconstructed oil bodies (Liao et al., 2024; Wang et al., 2022). Furthermore, the interfacial tension of ROBEs decreased significantly with increasing CGA addition (Fig. 4D). A lower interfacial tension indicates that less energy is required to form the oil-water interface. Together with the increase in interfacial protein concentration, this may improve the coverage of newly formed droplets during homogenization and reduce recoalescence (Işçimen, 2025; Wang et al., 2022). However, the distribution of CGA between the aqueous phase and the oil-water interface was not directly determined; therefore, these interfacial changes do not provide direct evidence of CGA localization or enrichment at the interface.

Beyond CGA, other food-derived ingredients, including arabinoxylans, mangiferin, and carminic acid, could also be investigated in reconstructed oil body systems (Lan et al., 2026; Yatheshappa et al., 2025). Arabinoxylans, for example, possess thickening, gelling, and emulsion-stabilizing properties (Hussain & Simsek, 2026). Future studies could evaluate the effects of other functional ingredients on interfacial assembly and emulsion stability.

3.6. Multiple light scattering stability analysis of ROBEs

The physical stability of ROBEs at different CGA addition levels was assessed using multiple light scattering, with changes in backscattering intensity (ΔBS) as the analytical parameter. The ΔBS profile of the 0 mg-CGA-ROBEs fluctuated sharply along the sample height (Fig. 5), suggesting greater physical instability associated with droplet migration and possible aggregation or phase separation. In contrast, increasing CGA addition resulted in smaller and more uniform variations in ΔBS, consistent with less pronounced droplet migration and greater physical stability of the ROBEs during the measurement period (Chen et al., 2023). Three factors may contribute to this improvement. First, CGA addition increased the negative surface charge of ROBEs. The 38 mg-CGA-ROBEs exhibited the most negative zeta potential (−47.13 mV) (Fig. 2D), which may enhance electrostatic repulsion between droplets and reduce the likelihood of aggregation after droplet collisions (Zhang et al., 2025). In addition, the viscosity of the 38 mg-CGA-ROBEs increased to 420.717 Pa·s (Fig. 3A). According to Stokes' law, the increased viscosity of ROBEs may reduce droplet movement and retard gravitational separation, thereby decreasing the tendency toward phase separation during the measurement period (Wei & Gao, 2016). Moreover, the increased interfacial protein concentration and adsorbed protein percentage (Fig. 4A and B) may be associated with changes in protein organization following CGA addition, which could contribute to the development of a more organized interfacial layer. Such an interfacial layer may provide additional protection around the oil droplets, thereby improving resistance to droplet aggregation and phase separation. These results suggest that CGA addition improved the physical stability of ROBEs through the combined effects of stronger electrostatic repulsion, increased viscosity, and altered interfacial organization.

Fig. 5.

Effects of CGA addition on the backscattering intensity (ΔBS) profile of reconstructed oil body emulsions.

3.7. Microstructure of ROBEs

Confocal laser scanning microscopy (CLSM) images showed that proteins (green fluorescence) were mainly distributed at the oil-water interface surrounding the oil droplets, whereas the oil phase (red fluorescence) was dispersed throughout the system (Fig. 6). These observations indicated that the ROBEs formed typical oil-in-water (O/W) emulsions (Yang, Li, et al., 2025). The CLSM images further showed that CGA addition altered droplet size and spatial distribution. In the 0 mg-CGA-ROBEs, the oil droplets were relatively large and exhibited apparent aggregation and a heterogeneous distribution. With increasing CGA addition, the droplets appeared smaller and more uniformly distributed. CGA addition may alter OBP conformation and the distribution of surface groups, favoring more uniform protein coverage at the oil-water interface. Such coverage could reduce interfacial heterogeneity and limit direct contact between adjacent droplets, thereby decreasing their tendency to aggregate. The observed microstructural trend was consistent with the progressive decrease in particle size shown in Fig. 2C. Overall, the improved interfacial properties and physical stability of ROBEs suggest their potential as ingredients in oil body-based food formulations, although their performance in actual food systems remains to be evaluated.

Fig. 6.

Effects of CGA addition on the microstructure of reconstructed oil body emulsions. (A) 0 mg CGA; (B) 8 mg CGA; (C) 18 mg CGA; (D) 28 mg CGA; (E) 38 mg CGA. Note: CLSM images showing (1) protein (green), (2) triacylglycerol core (red), and (3) merged image; scale bar: 20 μm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

3.8. Environmental stability of ROBEs

The effects of environmental conditions on the particle size of ROBEs are shown in Fig. 7. At different NaCl concentrations, particle size generally decreased with increasing CGA addition (Fig. 7A), indicating that CGA partially inhibited salt-induced droplet aggregation. ROBEs generally exhibited larger particle sizes at 300 mmol/L NaCl, possibly because ionic screening weakened the electrostatic repulsion between droplets (Geng et al., 2024). The pH also affected the particle size of ROBEs (Fig. 7B). All samples exhibited smaller particle sizes at pH 7.0 than at pH 3.0, with particle size decreasing as CGA addition increased. The larger particle size at pH 3.0 may be associated with increased protein protonation and reduced electrostatic repulsion (Geng et al., 2024; Wang et al., 2025). These results suggest that ROBEs are more stable under neutral conditions. Heating at 90 °C resulted in larger particle sizes than heating at 60 °C (Fig. 7C), possibly due to protein conformational changes and droplet aggregation (Wang et al., 2025). Nevertheless, particle size decreased with increasing CGA addition under both conditions, indicating that CGA alleviated heat-induced aggregation and improved thermal tolerance. Freeze–thaw treatment substantially increased the particle size of all ROBEs (Fig. 7D), indicating disruption of the emulsion structure by ice crystal formation. The 38 mg CGA group exhibited a significantly smaller particle size than the other groups, although the overall freeze–thaw stability remained poor (Chen et al., 2025). This destabilization may provide a basis for exploring freeze–thaw treatment as a potential approach for demulsification and oil recovery. Overall, these findings indicate the potential application of CGA-containing ROBEs in food processing, although their processing and storage stability should be further validated in actual food systems.

Fig. 7.

Effects of CGA addition on the particle size of reconstructed oil body emulsions under different environmental conditions. (A) NaCl concentration; (B) pH; (C) heating temperature; and (D) freeze–thaw treatment. Different letters of the same type indicate significant differences among CGA addition levels under the same treatment condition (p < 0.05).

3.9. Microstructure and protein profiles of ROBPs

As shown in Fig. 8A, the 0 mg-CGA-ROBPs exhibited large aggregates with relatively smooth surfaces. As CGA addition increased, these aggregates gradually fragmented and developed more irregular, loosely organized morphologies. These morphological changes may reflect alterations in protein conformation and molecular organization following CGA addition (Guo et al., 2026; Li et al., 2025).

Fig. 8.

Effects of CGA addition on the microstructure and molecular composition of reconstructed oil body proteins. (A) Scanning Electron Microscopy (SEM) images: (a) 0 mg; (b) 8 mg; (c) 18 mg; (d) 28 mg; (e) 38 mg. All SEM images were captured at 1000× magnification; scale bar = 100 μm. (B) SDS-PAGE analysis: Lane 1, molecular weight marker; Lane 2, 0 mg; Lane 3, 8 mg; Lane 4, 18 mg; Lane 5, 28 mg; Lane 6, 38 mg.

To further investigate the effects of CGA on the molecular weight distribution of oil body proteins, SDS-PAGE was used to analyze changes in the protein profiles of ROBPs (Fig. 8B). Compared with those of the 0 mg-CGA-ROBPs, the intensities of protein bands in the low-molecular-weight region (0–17 kDa) gradually decreased as CGA addition increased. Meanwhile, band intensities increased in the high-molecular-weight region, and protein signals accumulated near the loading wells.

These changes may reflect alterations in the aggregation state and electrophoretic behavior of OBPs, potentially involving protein aggregation, structural rearrangements, or oxidation-related changes under alkaline conditions (Guo et al., 2021; Guo et al., 2026). He et al. (2023) reported the formation of new high-molecular-weight aggregates (>130 kDa) in peanut protein samples treated with CGA, accompanied by the disappearance of protein bands at approximately 33 and 95 kDa. In contrast, although protein signals accumulated near the loading wells, no distinct bands above 130 kDa were observed in the present sunflower OBP samples treated with CGA. This difference suggests that changes in SDS-PAGE profiles following CGA addition may depend on the protein source, structural characteristics, and treatment conditions.

The pronounced changes in the 0–17 kDa region may involve oleosins, which are major structural proteins of oil body membranes and are generally found in the low-molecular-weight region (Plankensteiner et al., 2025). However, SDS-PAGE reflects band distribution and electrophoretic behavior but cannot identify the specific protein components or determine the nature of their interactions with CGA. Further studies are therefore needed to identify the affected proteins and clarify their involvement in the structural changes associated with CGA addition.

3.10. FT-IR of ROBPs

FT-IR spectroscopy was used to examine changes in the molecular environment of OBP functional groups following CGA addition. Compared with 0 mg-CGA-ROBPs, samples with CGA addition exhibited notable changes in the amide A band (3200–3600 cm⁻¹), which includes contributions from N—H and O—H stretching vibrations and is sensitive to the hydrogen-bonding environment (Fig. 9A). These spectral changes may reflect alterations in the overall hydrogen-bonding environment arising from both CGA hydroxyl groups and OBP functional groups, but do not by themselves demonstrate direct hydrogen bonding between CGA and OBPs (Chang et al., 2026). The amide I region (1600–1700 cm⁻¹), mainly associated with C O stretching vibrations of peptide bonds and closely related to protein secondary structure, and the amide II region (1500–1600 cm⁻¹), originating from coupled C—N stretching and N—H bending vibrations, both showed noticeable changes following CGA addition (Guo et al., 2021). These spectral changes suggest alterations in the local molecular environment of the peptide backbone, which could affect the vibrational behavior of amide bonds (Yan, Gao, et al., 2021). Such molecular rearrangements could alter the exposure and local environment of protein functional groups, potentially affecting protein organization at the oil-water interface. However, FT-IR alone cannot identify the specific interactions responsible for these changes or distinguish between covalent and non-covalent CGA–protein interactions.

Fig. 9.

Effects of CGA addition on the structural properties of oil body protein. (A) Fourier-transform infrared (FT-IR) spectra; (B) Secondary Structure Composition; (C) Intrinsic Fluorescence Spectra (a.u); (D) Surface Hydrophobicity(H₀).

3.11. Secondary structure of ROBPs

Fig. 9B shows that CGA addition altered the secondary-structure composition of ROBPs. FT-IR fitting analysis of the amide I band revealed that the β-turn content decreased from 43.06% in the 0 mg-CGA-ROBPs to 26.50% in the 38 mg-CGA-ROBPs, whereas the α-helix, β-sheet, and random coil contents increased to 21.23%, 27.20%, and 23.30%, respectively. The decrease in β-turn content accompanied by increases in α-helix, β-sheet, and random coil contents indicates a redistribution toward both ordered and flexible conformations. Because these secondary structures are stabilized by distinct backbone hydrogen-bonding arrangements, their redistribution may reflect localized reorganization of the polypeptide chains following CGA addition (Wang et al., 2023). Yan, Gao, et al. (2021) reported that polyphenol binding to camphor seed protein decreased the random coil content while increasing α-helix, β-sheet, and β-turn contents, indicating a more ordered structure. By contrast, Zhang et al. (2024) found that EGCG binding to wheat germ protein reduced β-sheet content while increasing the α-helix and random coil contents, suggesting a more disordered conformation. Overall, CGA addition was associated with a redistribution of the secondary structures of OBPs.

3.12. Intrinsic protein fluorescence of ROBPs

As CGA addition increased from 0 to 38 mg, the intrinsic fluorescence intensity of ROBPs progressively decreased from 13,711 to 6528 a.u., indicating pronounced fluorescence quenching (Fig. 9C). This fluorescence quenching may be associated with changes in the microenvironment of intrinsic fluorophores following CGA-protein interactions. CGA-protein interactions may contribute to conformational changes in ROBPs, thereby altering the local environment and polarity surrounding aromatic amino acid residues, particularly tryptophan and tyrosine. Changes in the microenvironment surrounding these fluorophores could modify their fluorescence behavior and contribute to the observed decrease in intrinsic fluorescence (He et al., 2023; Lv et al., 2025). The progressive decrease in fluorescence intensity suggests that increasing CGA addition was associated with greater changes in the microenvironment of intrinsic fluorophores, which may reflect conformational alterations of oil body proteins (Wang et al., 2023).

3.13. Surface hydrophobicity of ROBPs

ANS is a hydrophobic fluorescent probe that binds to exposed hydrophobic patches on proteins, generating a measurable fluorescent signal. With increasing CGA addition, the surface hydrophobicity (H₀) of ROBPs decreased progressively (p < 0.05), indicating diminished availability of surface hydrophobic sites for ANS binding. This decrease may be associated with CGA-protein interactions and the accompanying conformational changes in ROBPs, which could reduce the exposure of hydrophobic residues and decrease the number of accessible ANS binding sites (Guo et al., 2026). Additionally, the hydrophilic phenolic hydroxyl groups of CGA may contribute to increased surface polarity, which could further reduce the apparent surface hydrophobicity of ROBPs (Zhang et al., 2024). Reduced surface hydrophobicity may decrease excessive hydrophobic association among proteins, thereby limiting aggregation and favoring the formation of a more uniform interfacial layer.

3.14. Physicochemical and functional properties of ROBPs

The absolute zeta potential values of ROBPs with CGA addition were significantly higher than those of the 0 mg-CGA-ROBPs (p < 0.05) (Fig. 10A). The increased absolute zeta potential may be associated with the deprotonation of CGA under weakly alkaline conditions, which could contribute to a more negative surface charge of ROBPs. Moreover, CGA-associated changes in protein surface properties may contribute to alterations in the surface charge characteristics of ROBPs. As shown in Fig. 10B and C, the emulsifying properties of ROBPs improved progressively with increasing CGA addition (0–38 mg). The emulsifying activity index (EAI) increased from 27.91 to 77.32 m²/g, while the emulsion stability index (ESI) increased from 11.04 to 57.37 min (p < 0.05). These results indicate that CGA addition enhanced the emulsifying activity and emulsion stability of ROBPs. Several factors may contribute to this improvement. Owing to its aromatic moiety and multiple hydrophilic groups, CGA may influence the interfacial behavior of ROBPs by modifying their surface properties and molecular organization (Wang, Li, et al., 2024). Furthermore, the changes in secondary structure distribution, including increased α-helix and β-sheet contents accompanied by a moderate increase in random coil content, suggest that CGA addition was associated with structural rearrangement of ROBPs. The moderate increase in random coil content may reflect greater conformational flexibility of ROBPs (Tang et al., 2024). Meanwhile, the decreased surface hydrophobicity of ROBPs after CGA addition suggests reduced exposure or accessibility of surface hydrophobic regions, which may limit excessive hydrophobic association among proteins while still allowing their adsorption at the oil-water interface. These changes in surface properties may contribute to the observed increases in EAI and ESI, indicating that ROBPs are more effective in forming and stabilizing emulsions. In addition, the increased absolute zeta potential suggests stronger electrostatic repulsion associated with ROBPs, which may further contribute to interfacial stabilization and reduce droplet aggregation (Yang, Li, et al., 2026).

Fig. 10.

Effects of CGA addition on the physicochemical and functional properties of reconstructed oil body proteins. (A) Zeta potential (mV); (B) Emulsifying Activity Index (EAI, m²/g); (C) Emulsion Stability Index (ESI, min).

3.15. Mechanistic investigation of Chlorogenic acid on the stability of ROBEs

The enhanced physical stability of ROBEs after CGA addition may be associated with the modulation of protein conformation and interfacial properties (Fig.11). At the protein level, CGA-protein interactions may contribute to conformational rearrangement of OBPs, as suggested by the redistribution of secondary structures, including increased α-helix and β-sheet contents together with a moderate increase in random coil content relative to the 0 mg-CGA-ROBPs. The moderate increase in random coil content suggests a modest increase in the conformational flexibility of OBPs after CGA addition. CGA addition also increased EAI and ESI, indicating enhanced emulsifying activity and emulsion stability of ROBPs. The conformational and surface-property changes of OBPs may have contributed to these improvements. Consistent with these functional changes, the interfacial protein concentration and percentage of adsorbed protein also increased in ROBEs, indicating greater adsorption and retention of OBPs at the oil–water interface. In addition, CGA addition increased the viscosity and absolute zeta potential of ROBEs, which may enhance resistance to droplet movement and electrostatic repulsion, respectively, thereby contributing to improved emulsion stability. These changes may collectively reduce droplet mobility, collisions, and coalescence, thereby limiting phase separation and contributing to the improved physical stability of ROBEs. These findings provide insight into how CGA regulates the physical stability of oil body emulsions in a controlled reconstructed system. However, in natural oil body emulsions, the effects of CGA may be further modulated by its interactions with other endogenous components.

Fig. 11.

Proposed mechanism by which CGA enhances the stability of reconstructed oil body emulsions.

Previous studies have shown that other functional components, such as polyphenols and polysaccharides, can also influence the physical stability of oil body emulsions (Supplementary Table 1). It should be noted that the effects of these components depend on the oil body source, the characteristics and concentration of functional components, and the preparation procedures. The introduction of other functional ingredients, such as polysaccharides or polyphenols, may further modify interactions among the components present in the system, thereby influencing interfacial structure and emulsion stability. Therefore, the stability of multicomponent oil body emulsions is likely to depend on interactions among the components, including possible interactions between CGA carboxyl groups and protein residues. Further studies are needed to clarify how these component interactions affect interfacial organization under more complex food conditions.

4. Conclusion

This study investigated the role of endogenous chlorogenic acid (CGA) in the physical stability of sunflower oil body emulsions using reconstructed oil body proteins and emulsions as model systems. CGA addition altered the conformational characteristics of oil body proteins and improved their emulsifying properties, with EAI and ESI increasing from 27.91 to 77.32 m²/g and from 11.04 to 57.37 min, respectively. At the emulsion level, CGA addition increased the zeta potential (−47.13 mV), interfacial protein concentration (43.18 mg/m²), viscosity, and viscoelasticity of ROBEs, which may contribute to their improved physical stability. The environmental stability tests further showed that CGA addition improved the tolerance of ROBEs to salt, pH variation, and thermal treatment, although their freeze-thaw stability remained limited. Although the reconstructed oil body emulsion model allowed the role of CGA in emulsion stabilization to be evaluated, the interfacial organization and component interactions in this simplified system may differ from those in natural oil body emulsions. Future studies should investigate the contributions of other endogenous sunflower polyphenols and possible CGA oxidation products, as well as their interactions with phospholipids and other endogenous components. Their long-term physical and oxidative stability, formulation performance, and sensory properties should also be evaluated in actual sunflower oil body-based foods.

CRediT authorship contribution statement

Mengzhen Guan: Writing – original draft, Visualization, Validation, Software, Resources, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Fusheng Chen: Supervision, Project administration, Funding acquisition, Conceptualization. Dingyang Lv: Writing – review & editing, Supervision, Formal analysis. Yimiao Xia: Supervision, Formal analysis.

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

Research data are not shared.