Enhancing low-fat frankfurter texture and sensory properties with almond protein emulsion gels

Abstract

Health concerns related to saturated fats intake have encouraged efforts to reduce or replace animal fat in meat products. Direct substitution of saturated fats with liquid vegetable oils often compromises texture and processing stability. Emulsion gels (EGs) can structure liquid oils into solid-like fats. Due to its emulsifying capacity, almond protein isolate (API) was combined with κ-carrageenan (KC) to reinforce the gel matrix, and the API/KC EGs remain unexplored as pork fat replacers in meat products. EGs (6% w/v API, 20% v/v sunflower oil, 0–6% w/v KC) were prepared and characterized. The 6% KC group (AEG6) showed a dense microstructure, enhanced gel strength, improved thermal stability and oil-binding capacity, and the highest viscoelasticity. Replacing 50% pork fat with AEG6 maintained instrumental texture and sensory scores were comparable to conventional frankfurters. Overall, API/KC EGs effectively replace pork fat in low-fat frankfurters, owing to the emulsifying role of API.

Graphical abstract

Highlights

• •Novel almond protein isolate/κ-carrageenan emulsion gels (AEG) were prepared.
• •AEG6 (6% API/6% KC w/v) showed high viscoelastic and thermal stability.
• •AEG6 provided high gel strength, oil-binding capacity and uniform fat distribution.
• •Replacing 50% pork fat with AEG6 showed good emulsion stability in frankfurters.
• •Frankfurters formulated with AEG6 showed good texture and sensory characteristics.

1. Introduction

Meat products are widely consumed for their nutritional value and appealing sensory qualities (Kim et al., 2022). Animal fat contributes considerably to flavor and texture (Shin et al., 2022). However, its high saturated fat content has been associated with various health concerns, including cardiovascular diseases (Briggs et al., 2017). Consequently, research efforts have focused on replacing animal fats with healthier vegetable oils in meat products (Kim et al., 2022; Shin et al., 2022). Vegetable oils are attractive alternatives owing to their beneficial fatty acid profiles, low saturated fat content, and high unsaturated fat content. The key difference between saturated and unsaturated fats is their melting points. Unlike saturated fats, unsaturated fats are typically liquid at room temperature. The direct substitution of animal fat with vegetable oil often negatively impacts meat product quality, leading to an undesirable texture and reduced oxidative stability (Nacak et al., 2021; Ren et al., 2022). Previous research has extensively examined technological strategies for replacing animal fat with lipids of plant or marine origin in processed meat systems (Jiménez-Colmenero et al., 2015). These approaches involve incorporating liquid or modified oils directly, or introducing them as emulsified, encapsulated, or structured components within the meat matrix. Although such lipids offer improved nutritional profiles, their distinct physicochemical properties can adversely influence the textural integrity and sensory perception of the final product. Consequently, recent studies have focused on developing structured lipid systems that transform liquid oils into solid-like materials with desirable plasticity and balanced fatty acid composition (Co & Marangoni, 2012; Dickinson, 2012).

Vegetable oils are frequently incorporated into structured systems, such as emulsion gels and oleogels, to improve their stability and create well-defined structures (López-Pedrouso et al., 2021). Compared to oleogels, emulsion gels typically have a lower oil content, making them more suitable for the development of low-fat products (Ren et al., 2022). In addition, while oleogels primarily carry lipophilic compounds, emulsion gels were capable of incorporating both hydrophilic and lipophilic substances, enabling wider application in foods (Guo et al., 2023). Emulsion gels are complex systems in which oil droplets are dispersed within a continuous gelling matrix, resulting in a network that combines the properties of both emulsions and gels (Jeong et al., 2023). The formation of emulsion gels typically involves the immobilization of oil droplets within a viscoelastic matrix generated through protein or polysaccharide gelation, which can be induced by thermal, ionic, or pH-dependent mechanisms (Dickinson, 2012). The microstructure and rheological behavior of these systems are strongly influenced by the oil volume fraction, droplet size, and interfacial composition, which collectively determine their solid-like characteristics (Dickinson, 2012; Sun & Gunasekaran, 2009). These gel systems possess three-dimensional network structures and exhibit solid-like textural characteristics (Guo et al., 2023). Their soft solid texture mimics the physical behavior of fat in emulsified meat products (Guo et al., 2023). For example, emulsion gels can effectively replace fat in meat products, leading to improvements in cooking yields, oxidative stability, nutritional profiles, and sensory attributes (Fontes-Candia et al., 2023).

Frankfurters are popular processed meat products worldwide due to their convenience and affordable price. Nevertheless, pork frankfurters contain up to 23% fat and 8.7% saturated fatty acids (SFA), and their saturated fat levels often do not align with dietary recommendations (Fontes-Candia et al., 2023; Franco et al., 2019). Accordingly, many reformulation strategies have focused on lowering total fat or replacing animal fat with alternative lipid sources to improve the nutritional profile of frankfurters. Technologically, frankfurters are representative emulsion-type cooked sausages, in which product texture and sensory quality largely depend on the heat-induced formation of a three-dimensional protein matrix during cooking and cooling (Cao et al., 2021). In emulsified meat products, fat is essential for emulsion formation and contributes to the rheological and structural properties of the batter (Crehan et al., 2000). As a result, product quality is highly sensitive to batter/emulsion stability and phase separation under thermal processing (Shin et al., 2022). Therefore, frankfurters provide a practical model system to evaluate whether emulsion gels can mimic the functional role of animal fat while maintaining processing stability and key quality attributes.

Almonds are a rich source of protein (16%–22%) and contain the most essential amino acids (Zhang, Tian, et al., 2021). Almond protein isolate (API) is widely used in various food products owing to its desirable functional properties, including heat stability and oil-absorption capacity (Devnani et al., 2021). Sze-Tao and Sathe (2000) also reported that API exhibits a higher oil absorption capacity than soy protein isolate (SPI), indicating its strong affinity for lipid molecules and oil retention ability. In particular, API has been reported to exhibit superior emulsifying properties compared to other plant proteins such as SPI (Zhang, Tian, et al., 2021), which is one of the most well-studied and commonly used emulsifiers in food systems. Due to its emulsifying potential, API is considered a promising alternative for emulsion-based applications. In addition, API shows high solubility in neutral to alkaline conditions (pH 7–9; 70–80%), comparable to that of popular legume proteins such as soy, pea, and chickpea isolates (70–90%) (Devnani et al., 2021). Remarkably, under acidic conditions (pH 3–4), API maintains substantially higher solubility (80–90%) than most other plant protein isolates, which typically show limited solubility (∼10–50%). This wide pH applicability highlights the potential of API as a multifunctional ingredient suitable for diverse formulations, ranging from acidic fruit- or yogurt-based products to near-neutral plant-based beverages. The amphoteric nature of a protein allows API to interact with molecules carrying opposite charges, enabling electrostatic interactions with anionic polysaccharides (Zang et al., 2019). Proteins participate in other intermolecular forces, such as hydrogen bonding, to enhance the structural stability and functional properties of emulsion systems (Zhang et al., 2022). However, emulsions stabilized solely by proteins often exhibit poor emulsification, leading to instability and a tendency to separate (Zhang et al., 2022). Combining proteins with anionic polysaccharides could overcome these limitations (Zang et al., 2019; Zhang et al., 2022).

κ-Carrageenan (KC) is a sulfated linear anionic polysaccharide composed of alternating 1,3-β-d-galactopyranose and 1,4-α-d-galactopyranose residues (Li, Li, Wang, et al., 2024). KC is frequently employed as an emulsifier, thickener, and stabilizer in various food applications (Li, Li, Wang, et al., 2024). KC is known for its strong and thermo-reversible gelling ability, and it also interacts with proteins to enhance rheological, structural, and textural properties (Jeong et al., 2023). Due to these functional characteristics, KC has been widely applied to improve the stability of emulsion gels and to encapsulate functional nutrients in food systems.

Although API has been studied for its functional properties and KC has been used in various emulsion systems and structured lipid applications, their combination in emulsion gels for fat replacement in meat products remains largely unexplored. Specifically, the application of API in emulsion systems for fat replacement in meat products has not been extensively studied, and the interaction between API and KC within the gel system has not been investigated. Given the specific functional properties of API, we hypothesized that API would interact favorably with KC and form stable structures in the emulsion gel matrix. Furthermore, we speculated that the interaction between API and KC within the gel system would enhance both emulsion stability and gel strength, making this combination a promising candidate for fat replacement in frankfurters. Therefore, we investigated the interaction between API and KC, focusing on how their combination affects the structural and functional properties of the emulsion gels. In addition, we characterized the API/KC emulsion gels and evaluated their textural and sensory properties as fat substitutes in frankfurters.

2. Material and methods

2.1. Materials

Sunflower oil (SO) was purchased from CJ Cheil Jedang (Seoul, Korea). Almond flour was supplied by the GNM Food System (Gyeongsan, Korea). Tetrasodium pyrophosphate and KC were purchased from ESfood Co. (Gyeonggi-do, Korea). Nile Red and Nile Blue were obtained from Sigma-Aldrich (St. Louis, MO, USA). Lean pork meat from the hind legs and pork back fat was purchased from a local market (Seoul, Korea).

2.2. Preparation of API

API was prepared using a modified version of a previously reported method (Zhang, Tian, et al., 2021). Briefly, defatted almond flour, made from 100% almonds (protein content approximately 16–22% according to previous reports), was dispersed in distilled water (DW) at a ratio of 1:15 (g/g) and adjusted to pH 9.0 with 1 N NaOH. The dispersion was stirred for 1 h at room temperature and then centrifuged at 8000×g for 30 min at 4 °C. The pH of the supernatant was adjusted to pH 4.5 with 1 N HCl (1.0 mol/L), and the slurry was stirred for 1 h at room temperature. The precipitate was separated again by centrifugation at 8000×g for 30 min at 4 °C. The precipitate was redispersed in DW, stirred for 5 h, and the pH was adjusted to 7.0 with 1 N NaOH (1.0 mol/L) before being freeze-dried. API samples were stored at −80 °C. According to analysis conducted by the Animal Resources Research Center at Konkuk University (Seoul, Korea), the protein content of the API was 97.26 ± 0.08%.

2.3. Preparation of API emulsions (AEs) and API emulsion gels (AEGs)

API emulsions (AEs) were prepared using the following procedure. Briefly, API (6% w/v) was dispersed in DW and homogenized using a high-speed homogenizer (HG-15 A; DAIHAN Co., Seoul, Korea) at 10,000 rpm for 2 min. Subsequently, KC was added to the API dispersion at the specified concentrations (0%, 2%, 4%, or 6% w/v) and homogenized at 10,000 rpm for 30 s. SO (20% v/v) was gradually added to the mixture and homogenized at 10,000 rpm for 3 min to form a stable emulsion. The levels of SO (20%) and KC (0%–6%) were selected based on previous studies (Chen et al., 2020; Cui et al., 2022; Shi et al., 2020) and preliminary tests to obtain stable emulsions and gels with appropriate textural properties. For making API emulsion gels (AEGs), the prepared emulsion was heated to 85 °C and maintained at this temperature for 30 min to induce gelation. The sample was cooled rapidly in an ice bath for 15 min. Finally, the prepared AEG was stored at 4 °C overnight before further analysis. The specific formulations and preparation processes for the AEGs are presented in Table S1 and Fig. S1.

2.4. Characterization of API and AEs

2.4.1. Surface hydrophobicity of API

Surface hydrophobicity measurements were performed by adding 200 μL of bromophenol blue (BPB, 1 mg/mL) solution to 1 mL of API solutions (1%–7% w/v) prepared in DW. The mixture was incubated for 10 min at room temperature and centrifuged at 8000×g for 15 min. The supernatant was diluted 10-fold and the absorbance was measured at 595 nm using a UV–Vis spectrophotometer (BioTek Instruments, Winooski, VT, USA). All measurements were performed five times with DW used as a blank. The surface hydrophobicity was calculated using the following Eq. (1):

$$\mathit{Bound}\mathit{BPB}\left(\mathit{\mu g}\right):\frac{200\times \left({\mathit{Ab}}_{\mathit{blank}}-{\mathit{Ab}}_{\mathit{sample}}\right)}{{\mathit{Ab}}_{\mathit{blank}}}$$ (1)

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

EAI and ESI were determined based on previous studies with slight modifications (Li, Wang, Zhao, et al., 2024; Zhang, Liu, et al., 2021). The emulsions were prepared as described in Section 2.3. Each 20 μL aliquot of the emulsion was collected and diluted 250 times using a 1% sodium dodecyl sulfate (SDS) solution. The absorbance of the samples was measured at 500 nm using UV–Vis spectrophotometer, with an SDS solution used as the blank control. EAI and ESI were calculated using Eqs. (2), (3), respectively:

$$\mathit{EAI}\left(m^2/g\right)=\frac{2\times 3.303}{C\times \left(1-\varphi \right)\times {10}^4}\times A_{500}\times D$$ (2)

$$\mathit{ESI}\left(\%\right)=A_{10}/A_0\times 100$$ (3)

where A₅₀₀ is the absorbance at 500 nm, C is the protein concentration (g/mL) before emulsification, ϕ is the volume fraction of oil in emulsion (v/v), D is the dilution ratio of the emulsion, and A₀ and A₁₀ are the absorbance values of the emulsions after standing for 0 and 10 min, respectively.

2.5. Characterization of AEGs

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

The bonding transition of AEG was investigated using an FT-IR spectrophotometer (FT/IR-4100; JASCO, Tokyo, Japan). The spectral data ranged from 600 to 4000 cm⁻¹ with a resolution of 1 cm⁻¹. For secondary structure analysis, the band within the 1600–1700 cm⁻¹ region was deconvoluted using PeakFit 4.12 software (SeaSolve Software Inc., Framingham, MA, USA). The secondary structure composition of the samples was quantified based on the Gaussian area under the curve corresponding to the β-sheet (1615–1640 cm⁻¹, 1690–1700 cm⁻¹), α-helix (1650–1665 cm⁻¹), β-turn (1665–1690 cm⁻¹), and random coil (1640–1650 cm⁻¹) as previously described (Han et al., 2024).

2.5.2. X-ray diffraction (XRD)

XRD analysis of AEGs was conducted following a method adapted from a previous study (Keum et al., 2024). Measurements were performed using an X-ray diffractometer (SmartLab, Rigaku, Tokyo, Japan) with the operational parameters set at 40 kV and 30 mA. The diffraction patterns were recorded over a 2θ range from 5° to 50°, employing a step size of 0.04°.

2.5.3. Microstructure of AEGs

The morphology of AEGs was examined using field emission-scanning electron microscopy (FE-SEM) (Hitachi SU8010; Hitachi Ltd., Tokyo, Japan). The AEGs were individually mounted on an FE-SEM support and sputtered with platinum using a Sputter Coater 180 (Cressington Scientific Instruments, Watford, UK). Morphologies were observed using FE-SEM.

Confocal laser scanning microscopy (CLSM) images of AEGs were captured using a model LSM 900 instrument (Carl Zeiss, Oberkochen, Germany) as previously described (Jeong et al., 2023). For CLSM sample preparation, aqueous continuous phase and oil droplets were stained in the emulsion (AE) state using 0.03% (w/v) Nile Red and Nile Blue A, respectively. Each stained AE sample was placed on a slide glass, covered with a cover glass, and subjected to heating at 85 °C for 30 min, followed by cooling in an ice bath for 15 min to induce gelation and obtain the stained AEGs. Finally, the prepared AEGs were stored at 4 °C overnight before further analysis.

2.5.4. Differential scanning calorimetry (DSC)

The thermal characteristics of AEGs were assessed using DSC (400 Furance; PerkinElmer, Waltham, MA, USA) with slight modifications based on a previous study (Li, Wang, Cui, et al., 2024). Each sample (20–25 mg) was sealed in an aluminum pan, with an empty pan serving as the reference. Samples were heated from 25 °C to 80 °C at a rate of 5 °C/min under a nitrogen atmosphere.

2.5.5. Gel strength of AEGs

To evaluate the gel strength of AEGs, unconfined compression tests were performed using a TA-XT Plus texture analyzer (Stable Micro Systems Ltd., Godalming, UK) equipped with a 40 mm cylinder probe. The AEGs were cut into 2 × 2 × 2 cm³ pieces and placed between parallel stainless-steel plates. The stress response and elastic recovery were recorded under 60% strain at a rate of 1.0 mm/s. Stress/strain curves were generated using Exponent Connect software (Stable Micro Systems Ltd., Surrey, UK).

2.5.6. Oil binding capacity (OBC)

OBC of AEGs was determined using a centrifugation method based on a previous study (Keum et al., 2024). An empty 1.5 mL tube was weighed (a) and 1 g of AEG was added (b). The tube was centrifuged at 10,000×g for 15 min at 25 °C. After centrifugation, the tube was inverted for 15 min to allow excess oil to drain into the filter paper. After the oil was drained, the tube was weighed (c). The OBC was calculated using the following Eq. (4):

$$\mathit{OBC}\left(\%\right)=100-\left[\frac{\left\{\left(b-a\right)-\left(c-a\right)\right\}}{b-a}\times 100\right]$$ (4)

2.5.7. Rheological properties

The rheological properties of AEGs were determined based on a previous study (Kim et al., 2022). Measurements were performed using an MCR 92 rheometer (Anton Paar, Graz, Austria) equipped with a 25 mm diameter parallel plate geometry, maintaining a 1 mm gap. Apparent viscosity measurements were performed at 25 °C, with shear rates varying from 0.1 to 100 s⁻¹. Viscoelastic characteristics were evaluated through frequency sweep tests at 25 °C, with angular frequencies ranging from 0.1 to 100 rad/s. These tests were performed within the linear viscoelastic region (LVR) at a strain of 1% to ensure structural integrity. A temperature sweep test was performed with parameters set at a heating rate of 5 °C/min, temperature range of 25–80 °C, oscillation frequency of 1 Hz, and strain of 0.5%. Throughout the experiments, both the storage modulus (G′) and loss modulus (G″) were measured. All measurements were conducted in triplicate.

2.6. Preparation of frankfurters

As shown in Table S2, five different frankfurter formulations were prepared: one with only pork back fat (PF), and four with AEG replacing 25% (FEG25), 50% (FEG50), 75% (FEG75), and 100% (FEG100) of pork back fat. To prepare frankfurters, lean meat, and PF were minced using a 3-mm plate on a mincer (PM-70; Mainca, Barcelona, Spain). Minced meat, fat, ice, and other ingredients were emulsified using a silent cutter (Cutter C4 VV; SIRMAN, Venezia, Italy). The frankfurter batter was stuffed into 240-mm collagen casings (NIPPI Inc., Tokyo, Japan) using a stuffer (IS-8; SIRMAN, Marsango, Italy). The stuffed frankfurters were cooked in a smokehouse at 80 °C for 30 min until the core temperature reached 72 °C, then cooled with cold water for 10 min. All frankfurters were equilibrated to room temperature prior to experiments.

2.7. Characterization of frankfurters supplemented with AEGs

2.7.1. Color and pH of frankfurters

The surface color of frankfurters was measured using a CR-210 colorimeter (Konica Minolta, Ltd., Osaka, Japan). Color analysis was based on L* (lightness), a* (redness), and b* (yellowness) values. Each sample was measured seven times. The colorimeter was calibrated using a white plate (L* = +97.27, a* = +5.21, and b* = −3.40).

The pH of meat batter was measured using a LAQUA pH meter (Horiba, Kyoto, Japan). To prepare the samples for pH measurements, 5 g of meat batter was homogenized in 20 mL of DW (10,000 rpm, 60 s). Each pH measurement was replicated four times.

2.7.2. Texture profile analysis (TPA) of frankfurters

TPA was performed using a TA-XT Plus texture analyzer (Stable Micro Systems Ltd.) with a 40 mm cylinder probe. Frankfurter samples were cut into pieces 25 mm in diameter and 20 mm in height. TPA was performed using the following program parameters: pre-test speed (2.0 mm/s), test speed (2.0 mm/s), post-test speed (8.0 mm/s), maximum load (2 kg), distance (20 mm), and force (5 g). The evaluated textural parameters were hardness, springiness, cohesiveness, chewiness, and gumminess. Each measurement was repeated six times for each group.

2.7.3. Cooking loss of frankfurters

Cooking loss was measured with minor modifications based on a previous study (Shin et al., 2022). The weight of a raw meat batter sample was weighed (a) and the sample was heated at 80 °C for 30 min. After heating, the sample was allowed to cool to room temperature (25 ± 1 °C) for 2 h, and its weight was measured again (b). Cooking loss was calculated as follows (5):

$$\mathit{Cooking\; Loss}\left(\%\right)=\frac{a-b}{a}\times 100$$ (5)

Each measurement was performed in triplicate.

2.7.4. Emulsion stability of frankfurters

The emulsion stabilities of the frankfurter samples were measured as previously described (Keum et al., 2024). The frankfurter batter (25 g) was placed in a conical tube (a) and centrifuged at 5000×g for 5 min. The samples were then heated in a water bath at 80 °C for 30 min and subsequently cooled to room temperature. Pre-weighed empty dishes (b) were prepared and exudates (c) were collected from these dishes. The exudates were dried in an oven at 105 °C for 24 h. After drying, dishes were cooled to room temperature and weighed again (d). The total fluid, water, and fat exudations of the batters were calculated using Eqs. (6), (7), (8).

$$\mathit{Total\; fluid\; exudation}\left(\%\right)=\frac{c}{a}\times 100$$ (6)

$$\mathit{Water\; exudation}\left(\%\right)=\frac{\left(c-b\right)-\left(d-b\right)}{a}\times 100$$ (7)

$$\mathit{Fat}\mathit{exudation}\left(\%\right)=\frac{d-b}{a}\times 100$$ (8)

2.7.5. Rheological properties of frankfurters

The apparent viscosity, frequency sweep, and temperature sweep of frankfurter batters supplemented with AEGs were performed as described in Section 2.5.7, except for the following modifications. The tests were conducted within the LVR at a strain of 0.5% to ensure structural integrity. Frequency sweep tests were performed at 25 °C, with angular frequencies ranging from 1 to 100 rad/s. Temperature sweep tests were performed with parameters set at a heating rate of 2 °C/min, temperature range of 25–80 °C, and oscillation frequency of 1 Hz.

2.7.6. Sensory evaluation of frankfurters

The sensory evaluation was conducted with 36 non-trained consumer panelists with prior experience in food sensory evaluations (17 males and 19 females, aged 21–33 years). After cooking, the samples were cut into 2 cm-thick pieces and randomly assigned three-digit coded labels. Panelists were instructed to cleanse their palates with water between samples to minimize carry-over effects. The evaluation used a 9-point scale to assess preferences (appearance, taste, flavor, texture, and overall preference). The preference attributes were rated from 1 (extremely dislike) to 9 (extremely like). The sensory evaluation protocol was approved by the Institutional Review Board (IRB) of Konkuk University (approval number: KKUIRB-202411-HR-021). All participants provided informed consent for their data to be used and analyzed at the beginning of the study.

2.8. Statistical analyses

Statistical analyses were performed using SPSS-PASW software (version 22.0; SPSS Inc., Chicago, IL, USA). Each experiment was independently performed at least three times under the same conditions, and the results are presented as mean ± standard deviation. When the assumptions of normality and homogeneity of variance were satisfied, one-way analysis of variance (ANOVA) was used to determine significant differences among samples, and Duncan's multiple range test was used for post-hoc comparisons. When these assumptions were not met, nonparametric tests were applied using the Kruskal–Wallis test, and pairwise comparisons were performed with the Mann–Whitney U test when necessary. Statistical significance was set at P < 0.05.

3. Results and discussion

3.1. Surface hydrophobicity of API and emulsifying properties of AEs

The surface hydrophobicity of a protein is an index that measures the hydrophobic groups on the protein surface that interact with a polar aqueous environment. Surface hydrophobicity is closely related to functional properties, such as emulsification, surface tension, and foaming (Loyeau et al., 2021). In this study, the surface hydrophobicity, ESI, and EAI of API were determined using UV–Vis spectrophotometry (Fig. 1). A significant increase in API surface hydrophobicity was observed with increasing API concentration (Fig. 1A; P < 0.05), with comparable values observed at 6% and 7% (P > 0.05). Similarly, the EAI increased significantly with API concentration (Fig. 1C; P < 0.05), showing the highest values at 6% and 7%, which did not differ significantly. In contrast, the ESI showed no significant differences among the groups (Fig. 1B; P > 0.05). The increase in surface hydrophobicity with increasing API concentration can be explained by greater exposure of the hydrophobic residues on the surface of API. In addition, surface hydrophobicity is closely related to EAI, which represents the ability of protein–lipid and protein–protein interactions to form a stable emulsion (Zhang, Liu, et al., 2021). Accordingly, higher surface hydrophobicity is generally regarded as beneficial for emulsifying functionality, because greater exposure of hydrophobic groups can promote protein adsorption at the oil–water interface, lower interfacial tension, and facilitate the formation of a protective interfacial film around droplets, thereby supporting efficient emulsification (Gao et al., 2024; Zhang et al., 2022). By contrast to EAI, ESI maintains dispersion between the dispersed and continuous phases without coalescence, flocculation, or creaming (Zhang, Liu, et al., 2021). The ESI values of all samples were not significantly different and were maintained at a high level, close to 100%. The findings indicate the sufficient emulsion stability of API, even at low concentrations. These findings are probably attributable to the intrinsic characteristics of the API that contribute to emulsion stability, in addition to surface hydrophobicity, such as molecular interactions within emulsion systems. Considering both surface hydrophobicity and EAI results, 6% and 7% API showed the highest values without significant differences (P > 0.05). Therefore, 6% API was selected as the optimal concentration for preparing API-based emulsion gels for subsequent experiments. Taken together, the presence of hydrophobic groups in the API suggests its potential as an emulsifier. Future investigations should expand the concentration range and examine the effects of processing conditions (e.g., pH and pressure) on API functionality.

Fig. 1.

Surface hydrophobicity and emulsifying properties of almond protein isolate (API)-based emulsions (AEs) at different concentrations (1–7%, w/v). (A) Surface hydrophobicity of API measured using bromophenol blue (BPB) binding. (B) Emulsion stability index (ESI) and (C) Emulsion activity index (EAI) of AEs measured using a UV–Vis spectrophotometer at 500 nm. The error bars represent the standard deviation (n = 5 for A; n = 4 for B and C). Different letters above bars indicate significant differences among treatments (P < 0.05); bars sharing at least one letter are not significantly different (P > 0.05). N.S. indicates no significant difference (P > 0.05). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

3.2. Structural properties of AEGs

The structural properties of AEGs were analyzed using FT-IR spectroscopy, protein secondary structure analysis, and XRD (Fig. 2). The FT-IR spectra included peaks at 3374, 2927, 2854, 1744 cm⁻¹, corresponding to characteristic functional groups (Fig. 2A). The broad peak observed in the 3000–3600 cm⁻¹ range was associated with both free and bound O—H and N—H functional groups, suggesting the presence of hydrogen bonds between API and KC (Taaca et al., 2023). The gradual decrease in peak intensity with increasing KC concentration suggests that the hydroxyl groups of API are increasingly involved in ionic or hydrogen bonding with KC, leading to reduced free O—H groups (Huang et al., 2021). The peaks at 2927 and 2854 cm⁻¹, corresponding to the asymmetric and symmetric stretching vibrations of C—H, respectively, indicate the presence of aliphatic chains commonly found in lipids, fats, and other organic compounds (Zhang, Chen, et al., 2021). Peaks at 2927 and 2854 cm⁻¹ were observed, particularly in AEG6, suggesting that KC may promote intermolecular interactions within the emulsion gel network. These interactions potentially enhance the structural matrix and stabilize the dispersed oil droplets through the integration of aliphatic components. The peak at 1744 cm⁻¹ was attributed to C O stretching of the methyl ester groups. The increased intensity of this peak, particularly in AEG6, supports the formation of a stable emulsion gel network, as ester bond formation is indicative of interactions between API, KC, and oil droplets in the matrix. The peaks in the 1600–1700 cm⁻¹ region (amide I region) represent the C O stretching vibrations of peptide linkages in the backbone of the protein (Zhang, Chen, et al., 2021) and were further analyzed to investigate changes in secondary structure.

Fig. 2.

Structural properties of almond protein isolate/κ-carrageenan emulsion gels (AEGs). (A) Fourier-transform infrared (FT-IR) spectra of AEGs recorded within the range of 600–4000 cm⁻¹, (B) Secondary structure profiles obtained from deconvoluted amide I bands (1600–1700 cm⁻¹), (C) X-ray diffraction (XRD) patterns collected over a 2θ range of 5°–50°. Each analysis was performed in triplicate (n = 3). AEG0: AEG without κ-carrageenan (KC), AEG2: AEG supplemented with 2% KC, AEG4: AEG supplemented with 4% KC, and AEG6: AEG supplemented with 6% KC.

The secondary structure of AEGs was quantified using deconvoluted FT-IR spectra in the amide I region (Fig. 2B). AEG0 had an average of 28.26% β-sheet, 18.24% α-helix, 23.03% β-turn, and 30.47% random coil. As the content of KC increased, only β-sheet content increased, whereas the content of α-helix, β-turn, and random coil decreased. The gelation process induces a structural transition in proteins from α-helical to β-sheet structures, facilitating the formation of an organized network (Kobayashi et al., 2017). The observed increase in β-sheet content suggests enhanced oil–protein and protein–protein molecular interactions that facilitate a more ordered protein arrangement, contributing to improved structural stability of AEGs (Li, Wang, Cui, et al., 2024). Moreover, the increase in β-sheet content might be related to the reduction of β-turns, as the weak hydrogen bonds transform into the strong bond characteristic of β-sheets (Aziznia et al., 2024). Previous studies have also demonstrated that anionic polysaccharides such as κ-carrageenan promote the exposure of hydrophobic groups and the formation of β-sheet–rich domains, thereby strengthening intermolecular hydrogen bonding and facilitating the development of stable gel networks (Chen et al., 2025). In addition, a higher proportion of β-sheet structures has been associated with increased protein aggregation and the formation of homogeneous gel matrices, which contribute to improvements in gel strength, water-holding capacity, and overall structural stability (Chen et al., 2025). Additionally, the reduced content of random coil with increasing KC content suggests a transition to a more ordered emulsion gel structure (Han et al., 2024). Overall, these structural rearrangements indicate that API–KC interactions contribute to the formation of a cohesive and thermally stable network, which may enhance the overall organization and physicochemical stability of AEGs. Although these results suggest that API–KC interactions contribute to network stabilization, the binding modes and their relative contributions remain unresolved. Future work should clarify these mechanisms using molecular docking, supported by complementary experimental approaches, to strengthen the structure–function interpretation of AEGs.

The crystalline properties of AEGs were evaluated using XRD (Fig. 2C). A strong peak was observed at approximately 20° for AEG0, whereas the peak intensity decreased and became broader in AEG2–AEG6. This attenuation and broadening of diffraction peaks indicate a reduction in crystallinity and a transition toward a more amorphous structure with increasing KC content. According to previous findings, the addition of KC disrupted the crystalline region of the walnut protein-KC emulsion gel and was reported to potentially increase the number of binding sites at the oil–water interface (Li, Wang, Lv, et al., 2024). In the present study, this decrease in crystallinity can be explained by the strong binding of API molecules to the chains of KC, forming amorphous complexes through intermolecular interactions such as electrostatic attraction and hydrogen bonding (Dong et al., 2023). Consequently, such an increase in interfacial binding sites, together with electrostatic attraction and hydrogen bonding between API and KC, may have contributed to a more stable emulsion gel structure, as visually demonstrated by KC-containing AEGs (Fig. S2).

3.3. Microstructure of AEGs

The microstructures of AEGs were visualized by FE-SEM and CLSM (Fig. 3). As shown in Fig. 3A, AEG0 exhibited a relatively homogeneous and fine-textured surface, whereas AEG2 showed a more heterogeneous and discontinuous matrix. As the KC concentration increased, the matrix of the AEGs exhibited a more compact and robust structure, characterized by a more continuous network and thicker matrix regions between pores in AEG4–AEG6 (magnified insets in Fig. 3A). These structural changes likely resulted from increased hydrogen bonding and electrostatic interactions between API and KC with increasing KC concentration (Dong et al., 2023). These findings are consistent with the results of FT-IR, protein secondary structure analysis, and XRD, which also demonstrated enhanced structural organization with increasing KC concentrations. The collective findings indicate that KC may interact with API to form a stable, interconnected network, thereby enhancing the structural integrity of AEGs.

Fig. 3.

Microstructure of almond protein isolate/κ-carrageenan emulsion gels (AEGs). (A) Representative field-emission scanning electron microscopy (FE-SEM) images showing the surface morphology of AEGs (scale bar = 1.00 mm). To show representative microstructural details, circled regions are magnified in insets for each KC concentration tested. (B) Representative confocal laser scanning microscopy (CLSM) images of AEGs. The continuous aqueous-phase gel matrix formed by API and KC was stained with Nile Blue A (shown in green) and the oil droplets with Nile Red (shown in red); merged images are shown in the bottom row. Yellow circles indicate representative regions of interest, which are shown as magnified insets. CLSM images were captured at 20× magnification (scale bar = 40 μm). These images highlight differences in oil droplet distribution and matrix continuity among AEGs as a function of κ-carrageenan (KC) concentration. AEG0: AEG without KC, AEG2: AEG supplemented with 2% KC, AEG4: AEG supplemented with 4% KC, and AEG6: AEG supplemented with 6% KC. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Representative CLSM images of the emulsion gels with various KC concentrations are shown in Fig. 3B. All AEG samples were stained using the same staining procedure and dye concentrations prior to CLSM imaging. Fluorescence images obtained for each AEG included the continuous aqueous-phase gel matrix formed by API and KC stained with Nile Blue A (green channel) and oil droplets stained with Nile Red (red channel), allowing visualization of the spatial distribution of the gel matrix and oil droplets, and merged images. Consistent with the FE-SEM data, the CLSM images demonstrated that as the KC concentration increased, a denser matrix structure formed. Specifically, the green fluorescence became more continuous and spatially interconnected, indicating enhanced matrix continuity at higher KC concentrations. As dense gel networks formed, the oil droplets decreased in size and became more uniformly distributed throughout the matrix. These images indicate KC-dependent changes in matrix continuity/compactness and oil droplet dispersion, providing visual evidence that increasing KC promotes a more integrated API–KC network consistent with enhanced emulsion gel stability. These observations suggest that the interaction between API and KC contributes to the stabilization of the oil droplets (Shi et al., 2025). Our results align with those of previous studies, demonstrating that smaller oil droplets improve emulsion gel stability (Shi et al., 2025). Taken together, our data suggest that the API–KC interaction promotes the formation of a denser gel network and has the potential to improve the emulsion stability of AEGs by reducing the size of oil droplets in the matrix.

3.4. Thermal property, gel strength, and OBC of AEGs

The thermal properties of AEGs were analyzed by DSC (Fig. 4A). The lack of peaks between 25 °C and 80 °C indicates the thermal stability of AEGs within this temperature range. These results can be attributed to the amorphous structures of AEG2, AEG4, and AEG6, as demonstrated by the XRD data (Fig. 2C). Although the XRD results suggest some crystallinity in AEG0, the absence of corresponding thermal transitions in the DSC analysis indicates that these crystalline regions are insufficient to affect the thermal stability within the studied temperature range. Amorphous materials typically exhibit featureless DSC curves lacking sharp peaks owing to their disordered molecular structures and absence of distinct phase transitions. This observation is consistent with previous research, where the absence of thermal transition peaks in the DSC analysis was interpreted as evidence of thermal stability (Keum et al., 2024). The absence of thermal transitions from 25 to 80 °C indicates that AEGs maintain structural integrity at these temperatures, making them suitable for various meat product applications. Such thermal stability further indicates that the emulsion gel would remain its structural integrity under the thermal conditions of frankfurter processing, thereby allowing it to interact with the frankfurter matrix as a single phase rather than as separated components.

Fig. 4.

Functional properties of almond protein isolate/κ-carrageenan emulsion gels (AEGs). (A) Differential scanning calorimetry (DSC) thermograms of AEGs recorded from 25 °C to 80 °C at a heating rate of 5 °C/min under a nitrogen atmosphere (n = 3), (B) (B) Gel strength of AEGs determined by unconfined compression using a 40 mm cylindrical probe at a strain rate of 1.0 mm/s and 60% deformation (n = 4), and (C) Oil binding capacity (OBC) of AEGs measured after centrifugation at 10,000×g for 15 min at 25 °C (n = 4). The error bars represent the standard deviation. The different letters indicate significant differences (P < 0.05). AEG0: AEG without κ-carrageenan (KC), AEG2: AEG supplemented with 2% KC, AEG4: AEG supplemented with 4% KC, and AEG6: AEG supplemented with 6% KC.

The gel strengths of AEGs are shown in Fig. 4B. AEG0 exhibited no measurable gel strength, likely due to insufficient structural integrity resulting from the absence of KC. The gel strength of AEG6 was significantly higher than that of the other AEG groups (P < 0.05). This increase in gel strength can be attributed to the enhanced interactions between API and KC. During emulsion gel formation, protein and polysaccharide unfolding facilitates strong hydrophobic interactions and hydrogen bonding (Li, Feng, Dai, et al., 2024). Increasing the KC concentration led to more extensive protein–polysaccharide interactions, forming an interconnected and mechanically robust gel network. Similar observations have been reported in SPI–KC systems, where electrostatic attraction between the positively charged regions of the protein and the negatively charged sulfate groups of KC, together with hydrophobic and hydrogen-bonding interactions, contributed to the reinforcement of the gel matrix (Ortiz et al., 2004; de Souza Paglarini et al., 2019). In these systems, the addition of KC regulated protein-protein interactions, where the negatively charged KC binds to the positively charged protein residues, strengthening electrostatic interactions. As the concentration of KC increases, these interactions become more pronounced, stabilizing the particle interactions and contributing to the formation of a more stable and cohesive structure (Lu et al., 2023). Meanwhile, these increasing electrostatic interactions may reduce the interparticle distance fluctuations, which could further improve the excluded volume interactions, and then the strain-hardening effects, ultimately leading to the formation of a more robust gel network (Lu et al., 2023). Given that API has a globular conformation and a comparable isoelectric point (pI ≈ 4.5) to SPI, it likely exhibits similar surface charge characteristics, suggesting that analogous intermolecular associations occur between API and KC, contributing to the formation of a similar gel matrix structure. FE-SEM observations of AEG0 indicate a compact arrangement that may result from network shrinkage and compaction during dehydration, producing a structure with limited connectivity between protein strands. In contrast, AEG6 shows a more open microstructure in FE-SEM, while CLSM reveals an extensively interconnected protein–polysaccharide network (Fig. 3A and B), which corresponds with its higher gel strength. These complementary electrostatic and hydrophobic interactions are considered responsible for the reinforced gel network and the superior gel strength observed in AEG6. The 6% KC emulsion gel lacking API (KEG) also exhibited a gel strength similar to that of AEG6 (Fig. S3A), suggesting that KC acts as a gelling agent and contributes to gel formation and structural integrity. Collectively, our data demonstrate that stronger API–KC interactions, evidenced by the high gel strength of AEG6, result in a more stable and robust emulsion gel.

The OBC of the AEGs were evaluated to determine their ability to retain the oil phase within their structures (Fig. 4C). OBC indicates the tightness of the internal network structure and its ability to retain oil droplets. Consistent with the gel strength trend (Fig. 4B), the OBC values increased from 63% in AEG0 to 96% in AEG6; however, as the data did not fully satisfy the assumption of normality, nonparametric tests were applied. The Kruskal–Wallis test revealed significant differences among the four samples (P < 0.05), and pairwise comparisons using the Mann–Whitney U test confirmed significant differences between all groups (P < 0.05). This trend is consistent with the enhanced API–KC interactions described in the analysis of gel strength, where electrostatic and hydrophobic associations strengthen the three-dimensional gel network, reducing the mobility of oil droplets within the matrix. Such enhanced API–KC interactions likely led to the formation of a dense and cohesive microstructure capable of entrapping small oil droplets, as evidenced by FE-SEM and CLSM observations (Fig. 3A and B) (Fontes-Candia et al., 2022). In comparison, KEG exhibited a significantly lower OBC than that exhibited by AEG6 (Fig. S3B), indicating that API plays a crucial role in enhancing oil retention. As shown in Table S1, KEG and AEG6 contained the same KC concentration (6%), but the absence of API in KEG limited the formation of the dense microstructure necessary for effective oil entrapment. Collectively, these results highlight the cooperative role of API and KC in forming a cohesive and stable gel matrix capable of immobilizing the oil phase, suggesting the potential of AEGs as effective fat substitutes in meat products due to their thermal stability, robust structure, and oil retention capability.

3.5. Rheological properties of AEGs

The apparent viscosity and viscoelasticity of the AEGs were analyzed to evaluate their rheological properties (Fig. 5). The apparent viscosity gradually decreased with increasing shear rate, indicating that all AEGs exhibited shear-thinning behavior (Ren et al., 2024). Increased KC concentrations led to an increased apparent viscosity across all shear rates, suggesting the formation of a more complex molecular network (Fig. 5A). These observations are consistent with those of previous studies that reported similar trends with increasing polysaccharide concentrations in emulsion gels (Shi et al., 2025). These studies suggest that KC has a thickening effect and that a higher KC content promotes droplet aggregation, leading to a denser network structure within the emulsion gel. Additionally, the abundance of hydroxyl groups in KC may contribute to the increased viscosity of the emulsion gels through hydrogen bonding with water molecules (Abu Bakar et al., 2020; Li et al., 2023).

Fig. 5.

Rheological properties of almond protein isolate/κ-carrageenan emulsion gels (AEGs). (A) Apparent viscosity curves of AEGs measured at 25 °C over a shear-rate range of 0.1–100 s⁻¹, (B) Frequency-sweep test for viscoelasticity performed at 25 °C over an angular-frequency range of 0.1–100 rad/s to determine the storage (G′) and loss (G″) moduli, and (C) Temperature-sweep test for viscoelasticity conducted from 25 °C to 80 °C to evaluate temperature-dependent variations in the storage (G′) and loss (G″) moduli. The black line indicates temperature-dependent changes in rheological properties. Each analysis was performed in triplicate (n = 3). AEG0: AEG without κ-carrageenan (KC), AEG2: AEG supplemented with 2% KC, AEG4: AEG supplemented with 4% KC, and AEG6: AEG supplemented with 6% KC. Each colored line represents a different sample.

In the viscoelasticity tests, the storage modulus (G′) exceeded the loss modulus (G″) indicating a solid-like state for the material. As shown in Fig. 5B, all AEGs, except AEG0 that lacked KC, exhibited higher G′ than G″, indicating the formation of three-dimensional networks during gelation due to API–KC interactions. With increasing KC concentration, both G′ and G″ increased, suggesting stronger API–KC interactions and the formation of denser and more interconnected gel networks. The increasing G′ value indicates that KC enhances the mechanical strength of the AEG networks. Previous studies support the observation that protein–polysaccharide interactions enhance gel rigidity and elasticity (Shi et al., 2025). Furthermore, the increased excluded volume effects due to the higher concentration of KC contributed to the formation of stronger connection regions within the 3D networks, which leads to the formation of a more robust gel structure, as indicated by the increase in G′ (Lu et al., 2023). Collectively, our results suggest that KC enhances the elasticity and viscosity of AEGs, which enhances the solid-like properties and promotes formation of stable gel structures.

A temperature sweep test was performed to investigate the viscoelastic changes in AEGs caused by temperature variations (Fig. 5C). The variation in G′ value of the AEGs was measured over a temperature range of 25 to 80 °C. The G′ value for all AEGs decreased as the temperature reached 80 °C, but increased significantly during cooling from 80 to 25 °C. The increased temperature caused the breakdown of the double-helical aggregates of KC, leading to their decomposition into individual KC molecules. This process results in the dissolution of the network structure, forming a randomly coiled polymer solution and reducing the viscoelasticity of the emulsion gels (Li, Li, Wang, et al., 2024). Additionally, the decrease in G′ value during the heating phase might be associated with the disruption of hydrogen bonds and the dissociation of protein aggregates (Xu et al., 2023). During the cooling phase, the G′ value of AEGs increased markedly as the temperature decreased from 80 to 25 °C. In particular, AEG6 exhibited higher G′ values than that of other groups, indicating stronger reformation of the gel network. The increase in G′ values observed during the cooling phase suggests that KC rapidly reforms its helical structure. The regeneration of hydrogen bonds and repolymerization of KC molecular chains in the cooling phase further supports the formation of a structured gel (Li, Li, Guo, et al., 2024). Additionally, the thermal gelation of API involves partial unfolding of their compact molecular strands at high temperatures. Upon cooling, these unfolded proteins aggregate and crosslink, further enhancing the gel network structure. Overall, rheological analysis demonstrated that the AEGs exhibited enhanced viscosity, viscoelastic properties, and thermo-reversible gelation behavior, with AEG6 showing the highest values. These rheological properties suggest that AEGs are versatile and stable gel systems that are suitable for food applications requiring structural integrity and elasticity.

3.6. Color, pH, and texture profile of frankfurters supplemented with AEG

Following the evaluation of AEG characteristics, including thermal stability, gel strength, OBC, and rheological properties, AEG6 was selected as a fat substitute for frankfurter supplementation due to its superior physicochemical properties. The resulting color, pH, and texture profiles are shown in Table 1. Color is a critical sensory attribute that drives consumer preference and purchase intent. Therefore, ensuring color stability is vital for overall product acceptability. Frankfurters were prepared with PF partially or fully replaced by AEGs (0%, 25%, 50%, 75%, and 100%), and their color was evaluated by measuring L*, a*, and b*. According to the Kruskal–Wallis test, no significant difference was found in lightness (P > 0.05), indicating that the assumption of normality was not met for this parameter. This result was further supported by the Mann–Whitney U test, which confirmed no significant differences between the groups for lightness (P > 0.05). Similarly, redness (a*) and yellowness (b*) values measured on the cut cross-sections showed no significant differences among the groups (P > 0.05). These results are visually supported by Fig. 6, which shows a generally uniform cross-sectional appearance across treatments. These data suggest that replacing pork fat with AEGs can maintain the color profile of meat products, potentially minimizing concerns regarding visual aesthetic changes. The consistent pH values (P > 0.05) across all frankfurter samples indicate that AEG can successfully be used as a substitute for pork fat, while preserving key quality attributes. Collectively, these data suggest that AEGs could effectively replace pork fat in frankfurters without compromising quality attributes, including color and pH.

Table 1.

Color, pH, texture profile analysis (TPA), cooking loss and emulsion stability of frankfurters with almond protein isolate/κ-carrageenan emulsion gels (AEGs) substituted at proportions ranging from 0% to 100%.

Parameters		Frankfurters
		PF	FEG25	FEG50	FEG75	FEG100
L*		80.30 ± 0.95a	80.18 ± 0.56a	80.58 ± 0.61a	80.95 ± 0.69a	80.10 ± 0.56a
a*		2.35 ± 0.17a	2.34 ± 0.05a	2.32 ± 0.19a	2.37 ± 0.11a	2.36 ± 0.11a
b*		10.18 ± 0.18a	10.05 ± 0.19a	9.84 ± 0.12a	9.86 ± 0.09a	9.90 ± 0.30a
pH		6.64 ± 0.00a	6.64 ± 0.02a	6.64 ± 0.00a	6.64 ± 0.01a	6.64 ± 0.00a
Hardness (N)		68.06 ± 3.89a	63.58 ± 3.02ab	62.95 ± 6.49ab	61.04 ± 4.31bc	53.50 ± 3.41c
Springiness		0.58 ± 0.02a	0.60 ± 0.03a	0.57 ± 0.02ab	0.51 ± 0.03b	0.50 ± 0.04c
Cohesiveness		0.14 ± 0.02a	0.16 ± 0.01a	0.13 ± 0.01b	0.11 ± 0.02bc	0.12 ± 0.01c
Chewiness (N)		5.17 ± 1.21a	4.85 ± 0.63ab	4.63 ± 0.94ab	3.95 ± 0.80bc	3.35 ± 0.63c
Gumminess (N)		8.94 ± 1.81a	8.75 ± 0.86a	8.13 ± 1.45ab	6.79 ± 1.56bc	6.47 ± 0.36c
Cooking loss (%)		7.22 ± 0.38a	7.15 ± 0.32a	7.22 ± 0.26a	7.23 ± 0.54a	7.56 ± 0.78a
Exudation (%)	Total	1.01 ± 0.26a	1.18 ± 0.11a	1.01 ± 0.07a	1.10 ± 0.04a	0.99 ± 0.07a
	Oil	0.31 ± 0.17ab	0.39 ± 0.07a	0.32 ± 0.11ab	0.15 ± 0.11bc	0.06 ± 0.03c
	Water	0.72 ± 0.06b	0.68 ± 0.02b	0.69 ± 0.05b	0.86 ± 0.10a	0.93 ± 0.09a

The AEG used in all formulations was AEG6 (almond protein isolate 6% + κ-carrageenan 6%). PF: pork back fat 100%, FEG25: pork back fat 75% + AEG 25%, FEG50: pork back 50% + AEG 50%, FEG75: pork back fat 25% + AEG 75%, and FEG100: pork back 0% + AEG 100%. ^a–c Mean values in the same row are significantly different (P < 0.05). Data are presented as mean ± standard deviation. Sample sizes: color (n = 7), pH (n = 4), TPA (n = 6), cooking loss (n = 6), and emulsion stability (n = 3).

Fig. 6.

Visual appearance of frankfurters formulated with varying proportions of pork back fat and almond protein/κ-carrageenan-based emulsion gels (AEGs). The AEG used in all frankfurters was AEG6 (almond protein isolate 6% + κ-carrageenan 6%). FEG: frankfurter with emulsion gel–based fat replacement, PF: pork back fat 100%, FEG25: pork back fat 75% + AEG 25%, FEG50: pork back 50% + AEG 50%, FEG75: pork back fat 25% + AEG 75%, FEG100: AEG 100%.

Texture is a key quality parameter for meat product consumers and is often significantly affected by the replacement of pork fat with vegetable oils. Therefore, structured vegetable oils, including emulsion gels and oleogels, have been employed as fat substitutes (López-Pedrouso et al., 2021). Our TPA data showed that while PF had the highest hardness, the hardness of FEG25 and FEG50 was not significantly different (Table 1). This indicates that substitution up to 50% can maintain the characteristic firmness expected for frankfurters. By contrast, FEG75 and FEG100 had significantly lower hardness values compared to PF (P < 0.05). The reduced hardness at higher AEG substitution levels (above 50%) is likely due to the resulting increase in water content, which leads to a softer texture (Zhao et al., 2023). The decreased hardness and springiness at higher substitution levels may also be due to the differences in the physicochemical properties of pork fat and AEG. Other textural parameters (cohesiveness, chewiness, and gumminess) followed trends similar to those observed for hardness. The decrease in cohesiveness, chewiness, and gumminess with increasing degrees of substitution can be attributed to the formation of a weaker structural network, likely due to increased moisture content (Nacak et al., 2021). While chewiness and gumminess were statistically similar to those of PF up to 50% substitution (FEG25 and FEG50), higher substitution levels showed differences. Overall, these trends indicate that fat substitution (up to 50%) can maintain texture attributes relevant to sensory quality, whereas higher substitution levels tend to yield a softer texture and reduced structural integrity. Accordingly, replacing up to half of the pork fat with AEGs appears to be a practical strategy for preserving the desired textural characteristics of the frankfurters.

3.7. Cooking loss and emulsion stability of frankfurters supplemented with AEG

Data of cooking loss and emulsion stability of frankfurters are presented in Table 1. Cooking loss represents a reduction in moisture and fat content during heating and is a crucial quality parameter for meat products. In our study, no significant differences in cooking loss values were evident among frankfurters, ranging from 7.15% to 7.56% (P > 0.05). The low cooking loss observed in all frankfurters may be attributed to the gel-forming properties of the API and KC, which form a structural matrix that supports the overall integrity of the product during thermal processing. These data indicate that the substitution of pork fat with AEG can maintain the stability of frankfurters without significant variations in cooking loss.

Emulsion stability, as measured by the total amount of liquid, fat, and water released, is a crucial factor that influences the quality of frankfurters, along with cooking loss. Most importantly, total exudation analysis revealed no significant differences among the frankfurters, regardless of the AEG substitution proportion, indicating that the overall emulsion stability was maintained (P > 0.05). However, at substitution proportions higher than FEG50, oil exudation was significantly reduced compared to PF (P < 0.05), whereas water exudation was significantly increased (P < 0.05). As the proportion of pork fat-substituted with AEG increased, the actual fat content in the frankfurters decreased accordingly, with the trend becoming particularly pronounced at substitution proportions higher than FEG50 (P < 0.05). Additionally, the previously observed high OBC values of AEG may have contributed to stabilizing oil droplets within the matrix and reducing oil exudation (Fig. 4B). The significant increase in water exudation in FEG75 and FEG100 may be attributed to the increased water content in the frankfurters as the proportion of pork fat replaced by AEG increased (P < 0.05). Although the amounts of oil and water exudation changed with increasing AEG substitution, the overall emulsion system remained stable across substitution levels. This finding suggests that AEG maintained structural stability within the frankfurter matrix and likely acted as a single phase during processing rather than existing as separated components. In summary, these emulsion stability data indicate that AEG exhibits cooking and emulsion stability comparable to that of pork fat, suggesting its potential as a substitute for frankfurters.

3.8. Rheological properties of frankfurter batters formulated with AEG

The rheological properties of frankfurter batters were analyzed by measuring their apparent viscosity and viscoelasticity (Fig. 7A and B). As shown in Fig. 7A, the viscosity of all frankfurter batters decreased as the shear rate increased from 0.1 to 100 s⁻¹, exhibiting pseudoplastic behavior. This is because a higher shear rate disrupts the hydrogen bonding between protein molecules, which weakens the gel networks and reduces the resistance to flow (Wang et al., 2023). Increasing the proportion of pork fat replaced with AEG resulted in a decrease in viscosity at the same shear rate. The lower apparent viscosities observed in FEG75 and FEG100 likely resulted from their higher water content and altered structural networks, leading to a softer and less cohesive gel matrix. Notably, FEG25 and FEG50 showed similar apparent viscosities to PF, suggesting that replacing up to 50% pork fat with AEG maintained comparable rheological behavior. These results indicate that AEG can substitute up to 50% of pork fat without significantly affecting the apparent viscosity of frankfurter batters, demonstrating its potential as a fat substitute.

Fig. 7.

Rheological properties of frankfurter batters with almond protein isolate/κ-carrageenan emulsion gels (AEGs) substituted at proportions ranging from 0% to 100%. (A) Apparent viscosity of frankfurters measured at 25 °C over a shear-rate range of 0.1–100 s⁻¹, (B) Frequency-sweep test for viscoelasticity performed at 25 °C over an angular-frequency range of 1–100 rad/s to determine the storage (G′) and loss (G″) moduli, and (C) Temperature sweep test for viscoelasticity of frankfurters conducted from 25 °C to 80 °C to evaluate temperature-dependent variations in the storage (G′) and loss (G″) moduli. The black line indicates temperature-dependent changes in rheological properties. Each analysis was performed in triplicate (n = 3). The AEG used in all formulations was AEG6 (almond protein isolate 6% + κ-carrageenan 6%). The black line indicates temperature-dependent changes in rheological properties. PF: pork back fat 100%, FEG25: pork back fat 75% + AEG 25%, FEG50: pork back + AEG 50%, FEG75: pork back fat 75% + AEG 25%, and FEG100: AEG 100%. Each colored line represents a different sample.

The G′ and G″ values were measured during frequency sweeps to evaluate the viscoelastic stability of the frankfurter batters under deformation (Fig. 7B). The meat batters exhibited higher G′ values than G″ values across all frequency ranges (1–100 rad/s), indicating that elastic behavior dominated. Notably, the G′ values of FEG25 and FEG50 were comparable to those of PF, suggesting that up to 50% replacement of pork fat with AEG does not significantly compromise the elastic properties of the meat batters. The ability of FEG25 and FEG50 to maintain elastic properties comparable to those of PF may be because of the presence of KC in the AEG formulation (Shin et al., 2022). KC, which improves elasticity in emulsified meat products (Shin et al., 2022), likely contributed to the comparable viscoelastic properties of FEG25 and FEG50 relative to PF, despite their reduced fat content. This suggests that replacing up to 50% pork fat with AEG in conjunction with KC maintains sufficient stability against deformation, further supporting its potential as a fat substitute for frankfurter batters.

The G′ and G″ values across the temperature variations for the frankfurter batters are shown in Fig. 7C. The changes in G′ values are closely associated with the protein denaturation process during heating. In all samples, a slight increase in G′ was observed in the initial heating stage (25–53 °C; 0–14 min), likely due to the unfolding of myosin molecules and the formation of nascent gel networks. This observation aligns with the findings reported by (Wu et al., 2009), who reported a gradual increase in elasticity during the initial heating phase owing to the sequential unfolding of myosin. Subsequently, as the temperature increased from 53 to 62 °C (14–18 min), a slight decline in G′ values was observed, indicating the disruption of weak gel structures formed earlier. These results are likely due to the denaturation of myosin, which leads to the unfolding of protein tails and subsequent breakdown of gel networks (Zhao et al., 2023). Notably, the FEG groups showed no significant deviation from the PF in this temperature range, suggesting that the thermal stability of the gel network was not significantly affected by the proportion of AEG substitution. In the final stage (62–80 °C; 14–28 min), the G′ values sharply increased, indicating the formation of a highly elastic and dense three-dimensional gel network (Alvarez et al., 2012). FEG100 exhibited the lowest G′ values with slow increases during this phase, indicating that complete substitution of pork fat with AEG weakened the formation of a dense three-dimensional gel network. By contrast, other FEG groups showed G′ values similar to PF, maintaining the viscoelastic properties. The differing G′ values observed during cooling (80–25 °C; 28–44 min) between PF and the FEG groups are likely due to the distinct properties of pork fat and AEG. Pork fat, through the insertion of meat protein chains between its acyl chains, enhances lipid–protein interactions (Herrero et al., 2012), resulting in a strong gel network and high G′ values. Although AEG can also interact with meat proteins and contribute to the structure of the FEG groups, the resulting network may differ because of the inherent differences between AEG and pork fat. The protein–AEG interactions in FEG25 and FEG50 appeared to mimic the lipid–protein interactions in pork fat, resulting in similar matrix structures and contributing to the maintenance of elasticity and structural stability. However, the lower G′ values observed in FEG75 and FEG100 likely reflect the limited ability of AEG to fully replicate these lipid–protein interactions at higher substitution levels. These results suggest that AEG can replace up to 50% pork fat without compromising the viscoelastic properties or structural integrity of frankfurter batters, demonstrating its potential as a partial fat replacement.

3.9. Sensory evaluation of frankfurters supplemented with AEG

Reducing or replacing fat in meat products presents a challenge because of its potential impact on sensory quality. Therefore, it is crucial to evaluate the sensory attributes of fat-substituted products (Nacak et al., 2021). We assessed the sensory properties of the PF and FEG groups based on taste, flavor, appearance, texture, and overall acceptability (Table 2). Previous studies using high-pressure-modified quinoa protein emulsions demonstrated that up to 50% fat substitution could maintain or even enhance sensory qualities (Zhao et al., 2023). Similarly, in the present study, sensory attributes were maintained after partial or total replacement of pork back fat with AEG. According to the Kruskal–Wallis test, no significant differences (P > 0.05) were observed among the five groups for all sensory parameters, and pairwise comparisons using the Mann–Whitney U test also showed no significant variations in appearance, flavor, taste, or overall preference. Regarding appearance, the sensory evaluation was consistent with the instrumental color measurements (Table 1) and the comparable cross-sectional appearance shown in Fig. 6. However, texture was the only attribute showing a significant difference (P < 0.05), where the FEG100 group received a lower score compared with the PF group. These results indicate that partial replacement of pork fat with AEG did not adversely affect the sensory perception of frankfurters. The reduction in texture perception was particularly evident in the FEG100 group compared with the PF, which may be associated with the increased water content and the formation of a softer gel network in the emulsion matrix, leading to a less firm mouthfeel. Nevertheless, these changes were not substantial enough to affect the overall sensory acceptability. Similarly, other quality parameters, including color, pH, cooking loss, and emulsion stability, showed comparable results across all substitution levels (Table 1), indicating that AEG replacement was sufficient to maintain similar physicochemical properties. Overall, these results indicate that AEG can serve as an effective pork fat replacer in frankfurters, maintaining comparable sensory quality and consumer acceptability, even at high substitution levels.

Table 2.

Sensory evaluation of frankfurters with almond protein isolate/κ-carrageenan emulsion gels (AEGs) substituted at proportions ranging from 0% to 100%.

Parameters	PF	FEG25	FEG50	FEG75	FEG100
Appearance	6.61 ± 1.40a	6.31 ± 1.53a	6.42 ± 1.59a	6.44 ± 1.48a	6.42 ± 1.66a
Flavor	5.72 ± 1.75a	5.94 ± 1.84a	5.78 ± 1.71a	5.69 ± 2.03a	5.50 ± 1.73a
Taste	6.19 ± 1.92a	6.53 ± 1.52a	6.53 ± 1.54a	6.11 ± 1.65a	5.97 ± 1.68a
Texture	5.72 ± 1.75a	5.94 ± 1.84a	5.78 ± 1.71a	5.69 ± 2.03a	5.50 ± 1.73a⁎
Overall preference	6.33 ± 1.51a	6.33 ± 1.64a	6.58 ± 1.63a	6.08 ± 1.66a	5.86 ± 1.78a

The AEG used in all formulations was AEG6 (almond protein isolate 6% + κ-carrageenan 6%). PF: pork back fat 100%, FEG25: pork back fat 75% + AEG 25%, FEG50: pork back 50% + AEG 50%, FEG75: pork back fat 25% + AEG 75%, and FEG100: pork back 0% + AEG 100%. ^a–c Mean values in the same row are significantly different (P < 0.05). An asterisk (*) indicates a significant difference between PF and FEG groups (P < 0.05, Mann–Whitney U test). Data are presented as mean ± standard deviation (n = 36; 17 males and 19 females, aged 21–33 years).

4. Conclusions

The emulsifying properties of API enabled the formation of stable emulsion gel structures through surface hydrophobicity and emulsification activity. The physicochemical, thermal, rheological, and structural properties of AEGs were also evaluated. Among the tested formulations, AEG6 exhibited superior structural stability and a dense microstructure, enhancing oil entrapment efficiency. The thermal stability, gel strength, and OBC of AEG6 were enhanced, along with improved rheological properties. The application of AEG6 as a pork back fat substitute in frankfurters demonstrated significant effects on rheological and textural properties at replacement proportions of up to 50%. Notably, sensory evaluation revealed that FEG75 and FEG100 maintained comparable taste, flavor, and overall preference to those of pork back fat, although FEG100 exhibited a softer texture. These findings indicate that while textural and rheological properties were maintained up to 50% replacement, sensory acceptability remained comparable even at higher substitution levels, suggesting that AEG6 can serve as an effective fat replacer in frankfurters without compromising the key attributes.

CRediT authorship contribution statement

Seong Joon Hong: Writing – review & editing, Writing – original draft, Methodology, Investigation, Conceptualization. Jong Hyeon Han: Writing – review & editing, Investigation. Dong Hyun Keum: Writing – review & editing, Investigation. Sung Gu Han: Writing – review & editing, Writing – original draft, Supervision, Conceptualization.

Institutional review board statement

The Institutional Review Board approved the sensory evaluation procedure (KKUIRB-202411-HR-021).

Funding sources

This research was supported by a National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (RS-2022-NR069435) and the Bio&Medical Technology Development Program of the National Research Foundation (NRF) funded by the Korean government (MSIT) (RS-2022-NR067491).

Declaration of competing interest

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

Data availability

Data will be made available on request.