Multiscale structural modulation of plant proteins via ultrasound and electro-physical processing: from extraction to functional food matrices

1 Introduction

1.1 The rise of plant proteins in sustainable food systems

Growing concerns about environmental sustainability, population growth, and long-term food security have intensified the search for alternative protein sources to meet future dietary needs. Conventional livestock production plays an important role in global nutrition, yet it is also associated with substantial environmental impacts, including greenhouse gas emissions, high land requirements, and intensive water use. These pressures have encouraged researchers and industry stakeholders to explore protein sources that can provide nutritional value while reducing environmental burden. Plant proteins have emerged as one of the most promising candidates in this transition. Compared with animal-based proteins, plant-derived proteins typically require fewer natural resources during production and contribute less to greenhouse gas emissions. Life-cycle analyses of food production systems consistently indicate that plant-based protein sources generally have a smaller environmental footprint than animal agriculture (Poore and Nemecek, 2018). As a result, plant proteins are increasingly incorporated into strategies to develop more sustainable and resilient food systems.

A wide variety of crops serve as potential sources of plant protein, including soybeans, peas, lentils, chickpeas, faba beans, and other legumes. These protein sources are already widely used in the production of plant-based beverages, meat analogues, protein supplements, and functional food ingredients. Their availability, nutritional value, and adaptability to different agricultural systems make them attractive components of emerging plant-based diets. The rising demand for plant-based foods has also stimulated rapid innovation in the food industry. Advances in processing technologies and ingredient design have enabled the creation of products that more closely resemble the texture and sensory characteristics of conventional animal-derived foods. Nevertheless, significant challenges remain when incorporating plant proteins into complex food matrices. Their structural properties often differ considerably from those of animal proteins, which can influence their behaviour during processing and their performance in finished products.

1.2 Structural challenges of plant proteins

Although plant proteins offer important environmental and nutritional advantages, their broader use in food systems is frequently constrained by intrinsic structural properties. Many plant proteins exhibit compact globular conformations stabilized by strong intermolecular interactions, including hydrogen bonding, hydrophobic interactions, and disulfide linkages. These structural characteristics provide stability but can also limit proteins’ ability to interact effectively with surrounding components in food systems. One of the most frequently reported challenges associated with plant proteins is their relatively low solubility under certain conditions. Solubility plays a central role in determining other functional properties, including emulsification, foaming, and gelation. When proteins are poorly soluble, their dispersibility in aqueous systems is reduced, potentially leading to aggregation, sedimentation, or uneven distribution in food formulations (Lam et al., 2016).

Protein aggregation represents another important factor influencing plant protein functionality. During extraction, drying, or processing, plant proteins may form large aggregates through intermolecular interactions. These aggregates can restrict the accessibility of functional groups required for interfacial adsorption and network formation. As a result, plant proteins may exhibit weaker emulsifying or foaming performance than some animal proteins. The presence of anti-nutritional compounds also contributes to the complexity of plant protein utilization. Substances such as trypsin inhibitors, lectins, and phytates may interfere with protein digestion and reduce the bioavailability of certain nutrients (Gilani et al., 2012). In addition, interactions between proteins and naturally occurring phenolic compounds may influence flavor, color, and overall sensory perception of plant-based foods.

These limitations highlight the need for processing approaches that can modify plant protein structures in a controlled manner. By altering molecular conformation, aggregation behavior, and interfacial properties, it becomes possible to improve the functional performance of plant proteins in food systems and expand their range of applications.

1.3 Non-thermal and electro-thermal processing as enabling technologies

In recent years, non-thermal processing technologies have attracted considerable attention as potential tools for improving the functional performance of plant proteins (Rout et al., 2025; Akhi et al., 2025; Yücetepe et al., 2026). Conventional processing methods often rely on thermal treatment or chemical modification to alter protein structures. While these approaches can be effective, they may also lead to undesirable changes such as nutrient loss, excessive protein denaturation, or the formation of off-flavors (Hülsebusch et al., 2025; Karabulut et al., 2024; Yücetepe et al., 2026). In contrast, non-thermal technologies introduce physical energy into food systems, enabling structural modifications without exposing the entire system to high temperatures. It is important to distinguish, within this broad category, technologies that are strictly non-thermal in mechanism, ultrasound, pulsed electric fields (PEF), and cold plasma, which act primarily through mechanical, electrical, or reactive-species pathways without deliberately generating bulk heat, from ohmic heating, which is an electro-thermal process: it uses electrical energy specifically to generate internal Joule heating in a controlled and uniform manner. Ohmic heating is therefore discussed in this review alongside the non-thermal technologies because both categories minimise the severe, non-uniform thermal damage associated with conventional heating, but the two should not be conflated. Several non-thermal processing techniques have been explored for their ability to modify plant protein structures (Justino et al., 2024; Ampofo and Ngadi, 2022; Karabulut et al., 2024). Among these, ultrasound processing, PEF, cold plasma treatment, and ohmic heating are among the most widely studied. Each of these technologies operates through distinct physical mechanisms, but they share the common objective of altering protein conformation and interactions in a controlled manner (Younis et al., 2025).

Ultrasound processing is based on acoustic cavitation, in which microscopic bubbles form and collapse in a liquid medium. The collapse of these bubbles generates localized shear forces and turbulence that can disrupt protein aggregates and promote partial unfolding of protein structures. These structural changes may increase the accessibility of functional groups and improve properties such as solubility and emulsifying capacity (Jambrak et al., 2014). Next, studies involving pea and soy proteins showed that ultrasound treatment reduced particle size and improved dispersion in aqueous systems (Li et al., 2020).

Pulsed electric field (PEF) processing represents another important non-thermal technology used in food processing. In this technique, short bursts of high-voltage electrical pulses are applied to biological materials, creating temporary pores in cellular membranes via electroporation. This effect can facilitate the release of intracellular proteins and other biomolecules while also influencing protein structure at the molecular level. Electric fields may induce molecular polarization and alter intermolecular interactions, which can lead to changes in protein conformation and functional behavior (Zhang et al., 2021).

Cold plasma technology has also been explored as a method for modifying protein surfaces. Plasma contains reactive oxygen and nitrogen species that can interact with exposed amino acid residues. These interactions may introduce new functional groups or alter surface charge distribution, which can influence protein hydrophobicity and interfacial behavior (Misra et al., 2019). Such modifications are particularly relevant in applications in which proteins act as emulsifiers or foaming agents.

Ohmic heating is fundamentally an electro-thermal, rather than a non-thermal, technology: it involves passing an electrical current through the food material, resulting in rapid and uniform heating due to electrical resistance (Joule heating). Because heat is generated internally rather than transferred from external surfaces, temperature gradients are minimized, and protein denaturation can be more carefully controlled (Sakr and Liu, 2014). This approach may support the formation of more uniform protein networks during gelation processes.

Together, these technologies provide new opportunities to tailor plant protein structures for improved functionality in food systems. By influencing molecular interactions, aggregation behaviour, and interfacial properties, non-thermal processing methods can help overcome some of the limitations traditionally associated with plant proteins. As research in this area continues to expand, these technologies are expected to play an increasingly important role in the development of plant-based food products with improved nutritional and functional performance (Younis et al., 2025).

Recent advances in alternative protein research have led to several reviews on non-thermal physical modifications. For example, Kong et al. (2025) categorized physical treatments for techno-functional improvements, while related structural changes to protein bioavailability. Yücetepe et al. (2026) compared cold plasma, high-pressure, ultraviolet, and ultrasound treatments, emphasizing their effects on structure and digestibility. Despite these macro-level reviews, a knowledge gap remains on how coupled ultrasound and electro-physical processing influence interfacial dynamics and phase behavior. This review offers a multiscale analysis that links extraction efficiency to microstructural design of complex food matrices, focusing on molecular mechanisms to optimize emulsion and gelation networks.

This review aims to provide a comprehensive overview of how non-thermal and electro-thermal processing technologies influence the structural organization of plant proteins across multiple length scales, explicitly linking molecular conformation to micro- and macro-structural aggregation. Particular emphasis is placed on understanding molecular-level modifications, including changes in protein folding, intermolecular interactions, and aggregation behavior, and how these translate into functional and nutritional outcomes. In addition, the review evaluates the impact of these technologies on techno-functional properties relevant to food applications, alongside recent advances in computational modeling and considerations for industrial implementation. By integrating current findings, this work seeks to highlight the potential of non-thermal and electro-thermal processing in enhancing the performance of plant proteins for sustainable food systems.

2 Mechanisms of non-thermal and electro-thermal protein modification

Non-thermal processing technologies have emerged as promising tools for modifying plant proteins in ways that are more controlled than conventional heat- or chemical-based treatments (Akharume et al., 2021). Traditional processing methods often rely on prolonged heating or strong chemical conditions to alter protein functionality. While such approaches can be effective, they may also lead to undesirable consequences, including nutrient degradation, excessive protein denaturation, or the formation of unwanted reaction products such as those generated during Maillard reactions (Carbonaro et al., 2015). In contrast, non-thermal technologies introduce physical energy into the system through mechanisms such as sound waves, electric fields, or reactive plasma species, while ohmic heating achieves comparably controlled outcomes through uniform internal Joule heating rather than surface-driven conduction (Icier and Ilicali, 2005). These approaches allow targeted modifications of protein structure while largely preserving nutritional quality. To appreciate the significance of these technologies, it is important to consider the structural organization of plant proteins. Plant proteins consist of amino acid chains that fold into defined secondary structures such as α-helices and β-sheets, which subsequently assemble into complex three-dimensional conformations (Akharume et al., 2021). These tertiary structures may further associate into larger aggregates or networks. The structural arrangement of proteins strongly influences their behaviour in food systems, including their ability to dissolve in water, stabilize emulsions, form foams, or create gel networks. Consequently, processing methods that can adjust these structural arrangements in a controlled manner provide opportunities to improve protein functionality in specific food applications.

Non-thermal processing technologies act primarily by influencing the intermolecular forces that maintain protein structure. Hydrogen bonds, hydrophobic interactions, and disulfide linkages contribute to the stability of folded proteins and aggregated assemblies (Liang et al., 2020). By modifying these interactions, non-thermal treatments can promote partial unfolding of proteins, increase interactions with water molecules, and disrupt aggregates that limit functional performance. In some cases, previously buried amino acid residues become exposed on the protein surface, which can enhance interactions with other components in food systems (Chandrapala et al., 2012). Several mechanisms contribute to these structural changes. Ultrasound treatment, for example, generates acoustic cavitation in liquid media. During this process, microscopic bubbles form and collapse, producing localized mechanical forces that can disrupt protein aggregates and modify tertiary structures. Pulsed electric field (PEF) processing relies on short bursts of high-voltage electricity that induce electroporation in cellular membranes and influence protein conformation through electrical polarization. Cold plasma treatments introduce reactive oxygen and nitrogen species that interact with amino acid residues on protein surfaces, altering surface charge and hydrophilicity. Ohmic heating, although fundamentally an electro-thermal process, produces rapid and uniform heating through electrical resistance, enabling more controlled structural modifications compared with conventional heating methods (Sakr and Liu, 2014).

These technologies can also be applied sequentially or in combination to achieve synergistic effects (Boukid et al., 2021; Gharibzahedi and Smith, 2021). For instance, ultrasound pretreatment may expose protein surfaces and increase accessibility before additional treatments such as PEF or plasma processing are applied. Such combinations allow structural modifications to occur across multiple levels, from molecular rearrangements to changes in larger protein assemblies. The flexibility of these approaches provides opportunities to tailor plant protein functionality for a wide range of food applications, including meat analogues, dairy alternatives, protein-enriched beverages, and bakery products.

Recent research has demonstrated the effectiveness of non-thermal processing for improving the functionality of several plant protein sources, including soy, pea, rice, and lentil proteins. Ultrasound treatment has been shown to significantly enhance the solubility and emulsifying properties of pea proteins by promoting partial unfolding and exposing hydrophobic regions that facilitate interactions with oil droplets in emulsions (Li et al., 2020). Similarly, PEF processing has been reported to improve protein extraction efficiency and induce structural rearrangements that enhance functional performance while avoiding the extensive degradation often associated with high-temperature treatments (Zhang et al., 2021). These findings highlight the importance of understanding the underlying mechanisms through which non-thermal processing technologies influence protein structure and functionality.

2.1 Ultrasound-assisted protein modification

Among the various non-thermal technologies explored for food applications, ultrasound processing has received considerable attention for its ability to modify plant protein structures effectively. Ultrasound treatments are typically applied at frequencies ranging from approximately 20–100 kHz, where high-intensity sound waves propagate through liquid media and initiate a series of physical events. The most important of these events is acoustic cavitation, which involves the formation, growth, and collapse of microscopic bubbles (Mason et al., 1996; Kentish and Ashokkumar, 2011). The collapse of these bubbles generates localized regions of high temperature and pressure, producing strong shear forces and microturbulence in the surrounding medium. Although these extreme conditions occur only momentarily and within microscopic regions, they can significantly influence the structure of nearby protein molecules. Cavitation-induced forces are capable of disrupting protein aggregates, weakening intermolecular interactions, and promoting partial unfolding of protein structures (Chemat et al., 2017).

The structural modifications induced by ultrasound can improve the functional properties of plant proteins in several ways. One important effect is the increase in protein solubility. By exposing hydrophilic and hydrophobic regions that were previously buried within the protein structure, ultrasound treatment enhances interactions between proteins and water molecules (Tang and Ma, 2009). Improved solubility is particularly valuable for applications such as protein beverages and plant-based milk alternatives, where stable dispersion of proteins is required.

Ultrasound treatment can also enhance the ability of proteins to stabilize emulsions and foams. The exposure of hydrophobic groups increases the affinity of proteins for oil–water interfaces, enabling them to act more effectively as natural emulsifiers. Partial unfolding of proteins may also increase molecular flexibility, allowing proteins to reorganize at interfaces and form more stable interfacial films (Arzeni et al., 2012). These characteristics are especially important in products such as dressings, sauces, bakery foams, and whipped desserts. Another potential benefit of ultrasound processing is the improvement of protein digestibility. Structural rearrangements can expose enzymatic cleavage sites that were previously inaccessible, making proteins more susceptible to enzymatic hydrolysis during digestion. As a result, the bioavailability of amino acids may increase, which contributes to improved nutritional quality (Liang et al., 2020).

Numerous studies support these observations. Research involving soy, pea, and lentil proteins has demonstrated that ultrasound treatment can enhance solubility, emulsifying activity, and foaming stability (Jambrak et al., 2008; Jambrak et al., 2014; O’Sullivan et al., 2016). Importantly, ultrasound processing conditions can be adjusted to achieve specific outcomes. Parameters such as treatment intensity, frequency, duration, and temperature can be optimized depending on the type of protein and the intended application (Chandrapala et al., 2012). Overall, ultrasound-assisted modification represents a versatile strategy for improving plant protein functionality. By carefully controlling processing parameters, it is possible to tailor protein properties for different food formulations while maintaining nutritional quality and supporting the development of minimally processed plant-based products.

2.1.1 Acoustic cavitation as the key driving force

The central mechanism underlying ultrasound processing is acoustic cavitation. When ultrasonic waves propagate through a liquid medium, alternating pressure cycles lead to the formation of microscopic gas bubbles. These bubbles expand during low-pressure cycles and subsequently collapse during high-pressure phases. The collapse of cavitation bubbles produces extremely localized energy release, generating high shear forces, microjets, and shock waves within the surrounding liquid (Kentish and Ashokkumar, 2011). Although these effects occur on a microscopic scale and last only fractions of a second, they can strongly influence the structure of nearby molecules.

In protein systems, cavitation-generated mechanical forces can disrupt weak intermolecular interactions that stabilize protein aggregates. Hydrogen bonds and hydrophobic interactions may be weakened, allowing proteins to undergo partial unfolding without the extensive thermal denaturation associated with conventional heat treatments (Jahan et al., 2024). One important advantage of this mechanism is that structural changes occur primarily through mechanical forces rather than through bulk heating. As a result, the overall temperature of the system remains relatively low, which helps preserve sensitive nutrients and maintain protein quality (Mason et al., 1996).

Experimental studies illustrate the practical implications of acoustic cavitation. Jambrak et al. (2014) observed that ultrasound treatment altered the secondary structure of soy proteins, reducing the proportion of α-helical structures while increasing random coil configurations. These structural changes enhanced protein flexibility and improved solubility and emulsifying capacity. Similarly, Tang and Ma (2009) reported improved functional characteristics of soy protein isolate after high-intensity ultrasound treatment. The increased flexibility allowed proteins to interact more effectively with oil droplets, which is critical for stabilizing emulsions in products such as dressings and plant-based dairy alternatives. Li et al. (2020) further reported that ultrasound treatment reduced the particle size of pea protein aggregates. The resulting smaller particles exhibited improved interfacial activity and produced more stable emulsions over time. Similar findings were reported by Karki et al. (2010), who demonstrated that ultrasound-assisted processing enhanced protein release and dispersion in soy systems. Such improvements highlight the potential of ultrasound to enhance the performance of plant proteins as natural emulsifiers, offering alternatives to synthetic additives in clean-label food formulations.

Overall, acoustic cavitation provides a powerful yet controllable mechanism for modifying plant protein structures. By harnessing the mechanical forces generated during bubble collapse, researchers can influence protein conformation, aggregation behaviour, and functional performance. These capabilities make ultrasound processing an important tool in the development of next-generation plant-based food ingredients designed to meet growing consumer demand for sustainable and nutritionally balanced products (Xiong, 2017). The overall sequence of events is summarised schematically in Figure 1.

2.1.2 Disruption of non-covalent interactions

The mechanical forces generated during acoustic cavitation primarily influence the non-covalent interactions that stabilize protein structures. These interactions include hydrogen bonds, hydrophobic associations, van der Waals forces, and electrostatic attractions between charged amino acid residues. Although these bonds are weaker than covalent linkages, they collectively maintain the secondary and tertiary structures of proteins and therefore play a central role in determining protein functionality (Liang et al., 2020). Among these interactions, hydrogen bonding is particularly important in maintaining secondary structural elements such as α-helices and β-sheets. Cavitation-induced shear forces can weaken or disrupt these bonds, leading to partial unfolding of the protein chain. As the structure loosens, hydrophobic amino acid residues that were previously buried within the protein core may become exposed at the surface. This exposure alters the surface properties of the protein and often results in increased surface hydrophobicity (Tang and Ma, 2009).

Changes in surface hydrophobicity have important consequences for protein functionality in food systems. Proteins with greater surface hydrophobicity generally show improved affinity for oil–water interfaces, which enhances their ability to stabilize emulsions and foams. Such behaviour is particularly relevant in food products such as plant-based beverages, dressings, and dairy alternatives, where stable emulsification is essential. Experimental observations support this relationship. For example, O’Sullivan et al. (2016) reported that increases in surface hydrophobicity of plant proteins following physical processing were associated with improved emulsion stability. Similar observations were made by Arzeni et al. (2012), who noted enhanced interfacial properties following ultrasound-induced structural modifications.

Cavitation can also influence electrostatic interactions within protein systems. Structural rearrangements caused by ultrasound treatment may alter the distribution of charged residues along the protein surface, which in turn affects interactions with surrounding water molecules and neighbouring proteins. These changes can be particularly significant under near-neutral pH conditions, where plant proteins often exhibit a tendency to aggregate because of reduced net surface charge. In such cases, ultrasound-induced structural modifications may either enhance dispersion by reducing aggregation or, if treatment conditions are excessive, promote new interactions that lead to re-aggregation (Chandrapala et al., 2012). Consequently, the balance between protein unfolding and aggregation becomes an important factor in determining the final functional properties of ultrasound-treated proteins.

2.1.3 Modification of protein aggregates

Plant protein isolates frequently contain large aggregates that form during processing steps such as extraction, precipitation, drying, and storage. These aggregates can limit the functional performance of proteins in food systems by reducing solubility, decreasing bioavailability, and restricting the accessibility of reactive functional groups. Large aggregates also interfere with the formation of stable emulsions, foams, and gels, which are critical properties in many plant-based food products.

Ultrasound processing offers an effective strategy for modifying these aggregates. The intense shear forces generated during cavitation can mechanically disrupt large protein clusters and break them into smaller particles (Karki et al., 2010). This reduction in particle size increases the overall surface area available for interaction with water and other components in the food matrix. As a result, proteins tend to disperse more uniformly, which improves hydration behaviour and enhances functional properties. Several studies have reported these effects in different plant protein systems. Hu et al. (2013) observed that ultrasound treatment significantly reduced the particle size of soy protein aggregates, which was associated with improved water-holding capacity and enhanced emulsifying performance. Similarly, Zhao et al. (2018) reported increased solubility and improved foaming properties in rice proteins following ultrasound treatment, suggesting that aggregate disruption contributed to improved interfacial behaviour. Additional evidence provided by Liang et al. (2020) demonstrated that ultrasound-assisted treatments can effectively improve the dispersibility and stability of plant protein systems across multiple botanical sources.

Despite these benefits, ultrasound treatment must be carefully controlled. Excessive processing intensity or prolonged exposure can lead to unintended re-aggregation of proteins. This phenomenon often occurs because newly exposed hydrophobic regions on the protein surface interact with each other, forming new aggregates. Therefore, parameters such as ultrasound intensity, treatment time, and frequency must be optimized to achieve a balance between aggregate disruption and the prevention of re-aggregation (Chandrapala et al., 2012). Careful control of these conditions allows the structural properties of plant proteins to be tailored to the functional requirements of specific food applications.

2.2 Electro-physical processing technologies

Electro-physical processing technologies represent another important category of non-thermal methods used to modify plant protein structures. Unlike ultrasound processing, which primarily introduces mechanical energy into the system, electro-physical approaches involve the application of electrical energy in different forms. These technologies may operate through electric fields, reactive chemical species, or localized electrical heating, each producing distinct effects on protein structure and functionality (Sakr and Liu, 2014; Misra et al., 2019). Among the most widely investigated electro-physical techniques are pulsed electric fields (PEF), cold plasma treatment, and ohmic heating. Each of these methods offers specific advantages for modifying protein conformation, improving extraction efficiency, and enhancing functional properties of plant proteins (Barba et al., 2015; Zhang et al., 2021). Consequently, electro-physical processing technologies provide valuable tools for food scientists seeking to optimize the performance of plant-based protein ingredients while maintaining nutritional quality and reducing processing intensity.

2.2.1 Pulsed electric fields: electroporation and molecular polarization

PEF technology involves the application of short bursts of high-voltage electrical pulses to biological materials. Typical electric field strengths range from approximately 5–50 kV/cm, with pulse durations generally in the microsecond range. One of the primary effects of these pulses is electroporation, a phenomenon in which strong electric fields temporarily disrupt cell membranes and create nanoscale pores (Toepfl et al., 2007). Electroporation increases the permeability of plant cell structures and facilitates the release of intracellular compounds, including proteins, peptides, and other bioactive substances. This mechanism has important implications for protein extraction processes because it enables greater recovery of proteins without the need for harsh solvents or severe thermal treatments. Barba et al. (2015) reported that PEF pretreatment improved protein extraction yields from plant materials by approximately 30%. Similarly, Zhang et al. (2021) demonstrated that PEF treatment enhanced both the extraction efficiency and functional properties of pea proteins. Additional studies by Puértolas and de Marañón (2015) showed that PEF-assisted extraction improved mass transfer and reduced processing time while preserving nutritional quality.

Beyond its effects on cellular structures, PEF can directly influence protein molecules. Electric fields interact with molecular dipoles and charged amino acid residues, causing polarization and reorientation within the protein matrix (Jaeger et al., 2012). These interactions may disrupt hydrogen bond networks and modify the hydration layer surrounding proteins, leading to partial unfolding of protein structures. Such conformational changes can improve protein functionality by exposing hydrophilic and reactive functional groups that enhance solubility, emulsifying activity, and gelation behaviour (Wouters et al., 2001). Importantly, these modifications occur without the extensive thermal degradation associated with conventional heating methods. Consequently, PEF processing is particularly attractive for the development of clean-label food systems and minimally processed protein ingredients. By adjusting electric field strength, pulse duration, treatment frequency, and total energy input, it is possible to tailor structural modifications according to the requirements of specific food applications (Raso and Heinz, 2006). The ability of PEF technology to combine extraction enhancement with functional modification makes it a highly promising strategy in sustainable protein processing. These mechanisms are summarised schematically in Figure 2.

2.2.2 Cold plasma: surface redox chemistry

Cold plasma processing has emerged as another promising non-thermal approach for modifying the surface properties of food proteins. Plasma is generally described as a partially ionized gas containing energetic particles such as electrons, ions, ultraviolet photons, and reactive oxygen and nitrogen species (Misra et al., 2019). When generated under atmospheric or low-pressure conditions at relatively low temperatures, cold plasma can interact with biological materials without producing significant thermal damage. The interaction between cold plasma and proteins occurs primarily at the surface level. Reactive oxygen and nitrogen species generated within the plasma environment readily interact with exposed amino acid residues, initiating oxidation–reduction reactions that modify protein side chains (Pankaj et al., 2018). Sulphur-containing amino acids such as cysteine and methionine are especially susceptible to oxidative modifications. These reactions may lead to the formation of sulfoxides, disulfide cross-links, or carbonyl derivatives, all of which influence protein conformation and intermolecular interactions. In addition to these modifications, plasma exposure can introduce new oxygen-containing functional groups onto protein surfaces. Functional groups such as hydroxyl and carbonyl moieties may increase surface polarity, alter hydrophobicity, and modify electrostatic interactions between proteins and surrounding molecules (Segat et al., 2016). Such changes strongly influence interfacial behaviour and therefore affect the ability of proteins to stabilize emulsions and foams.

Experimental studies demonstrate the practical significance of these effects. Misra et al. (2019) observed that cold plasma treatment increased the emulsifying activity of soy protein isolates because the introduction of oxygen-containing groups enhanced protein surface polarity and interfacial adsorption. Similarly, Segat et al. (2016) reported improved foaming capacity in plasma-treated wheat proteins, which was associated with modifications in flexibility and surface hydrophobicity. Comparable findings were reported by Ji et al. (2018), who demonstrated enhanced solubility and dispersibility of plasma-treated peanut proteins. Because cold plasma primarily modifies protein surfaces rather than inducing extensive unfolding throughout the entire molecular structure, it allows fine control over functional properties while minimizing structural degradation (Bahrami and Khodaei, 2020). This selective mode of action makes cold plasma particularly useful for tailoring protein interfacial behaviour in plant-based beverages, whipped systems, and dairy alternatives.

2.2.3 Ohmic heating: controlled electro-thermal effects

Ohmic heating represents another electro-physical processing approach capable of modifying plant protein structure through controlled electrical energy transfer. In this technique, alternating electric current passes directly through the food material, and electrical resistance generates heat internally within the sample (Sakr and Liu, 2014). Unlike conventional heating, where thermal energy transfers gradually from external surfaces inward, ohmic heating produces rapid and uniform volumetric heating throughout the entire food matrix. The uniform heating profile generated during ohmic processing minimizes temperature gradients and prevents localized overheating, thereby allowing greater control over protein denaturation and aggregation behaviour (Icier and Ilicali, 2005). Moderate unfolding during ohmic heating may expose reactive groups that participate in intermolecular interactions, facilitating the formation of protein networks important for gelation and texture development.

Research demonstrates that ohmic heating can improve the gelation characteristics of plant proteins. Sakr and Liu (2014) observed that ohmic heating promoted more homogeneous protein unfolding and network formation compared with conventional thermal methods. The resulting protein gels exhibited improved water-holding capacity, elasticity, and structural stability. Similar observations were reported by Pereira et al. (2007), who found that ohmic treatment enhanced protein functionality while reducing thermal damage to sensitive nutrients. The ability to control heating rates and structural transitions is particularly valuable for the development of plant-based meat analogues and high-moisture protein systems (Kyriakopoulou et al., 2019). Furthermore, ohmic heating can be integrated with other non-thermal processing technologies, including ultrasound and PEF, to achieve complementary structural modifications and enhanced functional performance.

2.3 Synergistic and hybrid processing strategies

Recent research increasingly emphasizes the advantages of combining multiple non-thermal processing technologies to achieve synergistic effects in plant protein modification. Individual technologies alter protein structures through distinct mechanisms; therefore, their integration enables targeted structural transformations across multiple scales, from cellular disruption to molecular rearrangement (Chemat et al., 2017).

For example, PEF processing may first induce electroporation and increase membrane permeability, thereby improving protein release from plant tissues. Subsequent ultrasound treatment can further disrupt protein aggregates and promote partial unfolding through cavitation-induced shear forces. Cold plasma treatment may then modify surface chemistry by introducing reactive functional groups that improve interfacial behaviour and protein reactivity (Wang et al., 2022). A growing body of literature supports the effectiveness of such hybrid approaches. Wang et al. (2022) demonstrated that combined PEF and ultrasound processing significantly improved protein extraction yield and emulsifying activity in soybean systems. Similarly, hybrid ultrasound–plasma treatments enhanced foaming stability and interfacial functionality by simultaneously modifying protein conformation and surface chemistry. Further studies by Boukid et al. (2021) showed that integrating ultrasound with enzymatic or pH-shifting treatments further enhanced protein solubility and flexibility.

Recent reviews also indicate that combining green processing technologies can maximize protein yield and functionality while minimizing energy consumption and processing severity (Gharibzahedi and Smith, 2021). Hybrid approaches involving PEF, ultrasound, enzymatic hydrolysis, and plasma treatment have demonstrated improvements in gelation behaviour, water-holding capacity, and structural organization of plant protein systems (Liang et al., 2020). Emerging computational and machine-learning approaches further support the optimization of hybrid processing conditions. Wang et al. (2023) highlighted that machine-learning-assisted optimization can predict the effects of combined treatments on protein functionality, thereby improving process efficiency and product consistency.

Overall, synergistic non-thermal processing strategies provide a highly versatile platform for tailoring plant protein functionality. By integrating complementary mechanisms such as electroporation, cavitation, redox modification, and controlled electro-thermal effects, hybrid technologies can achieve greater improvements in solubility, interfacial activity, gelation, and texture than individual treatments alone. Consequently, integrated electro-physical processing represents a promising direction for the development of next-generation plant protein ingredients for applications in meat analogues, dairy alternatives, and functional food systems. Table 1 summarises and compares the mechanisms, structural effects, and industrial potential of the four technologies discussed above.

TABLE 1. Comparative analysis of non-thermal processing technologies for plant protein modification.

2.4 Comparative responses of different plant protein sources to the same processing technology

The preceding subsections describe ultrasound, PEF, cold plasma, and ohmic heating largely in generic terms, yet a given technology can produce markedly different structural outcomes depending on the botanical source of the protein being treated. Direct comparisons under matched processing conditions illustrate this clearly. Under ultrasound treatment, soy and pea globulins (7S vicilin- and 11S legumin-type proteins) generally show a pronounced, dose-dependent unfolding response, with the more loosely associated 7S/vicilin fraction unfolding and solubilising more readily than the disulfide-rich 11S/legumin fraction (Sha and Xiong, 2022). Oat protein, which is dominated by a compact, highly disulfide-stabilised 12S globulin with unusually high thermal and structural stability, requires substantially higher ultrasound intensities to achieve comparable particle-size reduction and solubility gains, and only moderate intensities yield a favourable balance between unfolding and re-aggregation, whereas higher intensities increasingly fragment and re-aggregate the protein rather than continuing to improve functionality (Li and Xiong, 2021; Rafique et al., 2024). Wheat gluten proteins, being prolamins rather than globulins, respond to ultrasound primarily through changes in disulfide crosslink density and aggregate/network structure rather than through classical α-helix-to-β-sheet transitions, since their native structure already contains little ordered secondary structure to begin with (Akharume et al., 2021).

A similar source dependence is observed for PEF and cold plasma. PEF-induced electroporation and dipole-driven unfolding act most efficiently on proteins with a substantial net helical dipole moment, which favours pea and soy globulins over prolamins, whose limited helical content offers fewer dipole-coupled sites for the electric field to act upon (Zhang et al., 2021; Sha and Xiong, 2022). Cold plasma treatment, which acts through surface-localised reactive oxygen and nitrogen species, tends to show the most pronounced surface-hydrophobicity and foaming improvements in proteins with a high proportion of surface-exposed aromatic and sulfur-containing residues; oat protein, for example, shows measurable gains in foaming capacity under atmospheric cold plasma despite its high thermal stability, because the treatment acts on surface chemistry rather than requiring bulk unfolding (Eazhumalai et al., 2023). Faba bean globulins occupy an intermediate position between soy/pea and oat, sharing the 7S/11S architecture of legumes but with a subunit composition that produces a somewhat lower unfolding threshold and a correspondingly higher risk of re-aggregation if ultrasound or PEF treatment is not carefully controlled (Alavi et al., 2021). These comparisons indicate that no single set of processing parameters is optimal across all plant protein sources. Globular, disulfide-poor fractions such as 7S vicilin respond efficiently to moderate ultrasound or PEF treatment, disulfide-rich globulins such as 11S legumin and oat 12S globulin require higher energy inputs or longer treatment times to achieve comparable unfolding, and prolamins such as wheat gluten proteins respond through fundamentally different structural pathways that are not well captured by α-helix/β-sheet metrics alone. Tables 1, 2 summarise representative processing conditions and outcomes across these protein sources.

TABLE 2. Structural modifications and techno-functional improvements of plant proteins induced by non-thermal processing technologies.

3 Multiscale structural transformation of plant proteins

A major advantage of modern non-thermal processing technologies is their ability to induce structural modifications in plant proteins across multiple length scales. Protein functionality in food systems is determined not only by the molecular arrangement of amino acids but also by how these molecules assemble into larger structures that ultimately influence texture, stability, and rheological behaviour. Therefore, understanding the effects of processing requires examining structural transformations at the molecular, supramolecular, and macroscopic levels simultaneously. Plant proteins possess complex hierarchical structures. At the most basic level, the amino acid sequence determines how the polypeptide chain folds into secondary structural elements such as α-helices and β-sheets. These elements then assemble into tertiary structures that define the three-dimensional conformation of the protein.

In many cases, individual protein molecules further associate to form larger aggregates or networks that determine the physical characteristics of food systems. Changes occurring at any of these structural levels can influence functional properties such as solubility, emulsification, foaming, and gelation. Non-thermal processing technologies, including ultrasound, pulsed electric fields (PEF), cold plasma, and ohmic heating, provide opportunities to modify these structural levels in a controlled manner. Unlike conventional thermal treatments, which often cause extensive denaturation and aggregation, non-thermal methods allow targeted structural adjustments while preserving nutritional quality. Several studies have demonstrated that these treatments can improve protein solubility, interfacial activity, gelation behaviour, and digestibility (Jambrak et al., 2014; O’Sullivan et al., 2016). Understanding how such structural transformations occur is essential for designing plant protein ingredients with improved functional performance.

3.1 Molecular-scale transformations

3.1.1 Rearrangement of secondary and tertiary structures

At the molecular level, non-thermal processing primarily affects the secondary and tertiary structures of proteins. Structural modifications are commonly investigated using spectroscopic techniques such as Fourier transform infrared spectroscopy (FTIR), circular dichroism (CD), fluorescence spectroscopy, Raman spectroscopy, and differential scanning calorimetry (DSC). These analytical approaches enable researchers to observe changes in protein folding patterns, conformational stability, and intermolecular interactions following processing treatments (Tang and Ma, 2009; Liang et al., 2020). Numerous studies have reported that non-thermal treatments, particularly ultrasound and PEF, can significantly alter the distribution of secondary structural elements within plant proteins. In many cases, α-helical structures decrease while β-sheet and random coil structures increase, indicating partial unfolding and increased molecular flexibility (Jambrak et al., 2013; Arzeni et al., 2012). Such conformational rearrangements expose previously buried functional regions and improve molecular mobility.

Jambrak et al. (2013) reported that ultrasound treatment of soy protein isolates reduced α-helix content while increasing β-sheet structures. This partial unfolding exposed hydrophobic regions that enhanced the ability of proteins to stabilize emulsions and foams. Increased flexibility also enabled proteins to reorganize more effectively at oil–water interfaces, thereby improving interfacial stability. Similar observations were reported for PEF processing. Zhang et al. (2021) demonstrated that PEF treatment induced partial unfolding of pea proteins by influencing electrostatic interactions and molecular polarization. These structural modifications increased protein solubility and enhanced emulsifying capacity.

Additional studies support these findings. Wang et al. (2018) demonstrated that ultrasound-induced conformational changes enhanced the thermal stability and flexibility of soy protein isolates. Likewise, Chen et al. (2019) observed that non-thermal treatments modified hydrogen bonding and tertiary structure organization, contributing to improved gelation behaviour in plant protein systems. Moderate increases in β-sheet structures may promote the formation of stronger protein networks during gelation, which is particularly important for plant-based meat and dairy alternatives where structural integrity and texture are essential (Kyriakopoulou et al., 2019).

However, excessive β-sheet formation can lead to irreversible aggregation and reduced solubility. Consequently, careful optimization of processing parameters such as ultrasound intensity, electric field strength, and treatment duration is necessary to balance beneficial unfolding with the prevention of undesirable aggregation (Hu et al., 2013). To better understand these transformations, researchers increasingly employ computational approaches such as molecular dynamics simulations and predictive modelling. Such approaches provide insights into how energy inputs from non-thermal processing influence protein folding pathways and molecular interactions (Wang et al., 2023). These tools may ultimately guide the design of optimized processing strategies for plant protein ingredients.

The preferential loss of α-helical structure relative to β-sheet structure under ultrasound and PEF treatment can be rationalised in terms of the hydrogen-bonding topology and dipole characteristics. α-Helices are stabilised by a regular, intramolecular pattern of backbone hydrogen bonds linking each carbonyl oxygen to the amide hydrogen of the residue four positions further along the chain. Because each hydrogen bond acts independently of its neighbours, the network is mechanically ‘low-cooperativity’: disrupting a small number of bonds is sufficient to destabilise an entire helical turn. Acoustic cavitation generates highly localised shock waves and shear gradients at the collapse of cavitation bubbles; these forces act over nanometre-to-micrometre length scales that are well matched to unravelling this loosely coupled hydrogen-bond network. β-Sheets, in contrast, are stabilised by a more extensively cross-linked, cooperative network of inter-strand hydrogen bonds combined with tight packing of hydrophobic side chains between adjacent strands; disrupting this requires simultaneous failure of many bonds across the sheet, making it comparatively more resistant to brief, localised mechanical perturbation (Liang et al., 2020; Jambrak et al., 2013). A parallel argument applies to PEF processing: the α-helix carries a substantial net dipole moment along its axis because the peptide-bond dipoles are aligned in the same direction, whereas the anti-parallel or parallel hydrogen-bonding pattern of a β-sheet leaves a comparatively small net dipole. α-Helical segments are therefore more strongly coupled to, and more readily reoriented or distorted by, an externally applied electric field than β-sheet segments, which offers a physical explanation for the electric-field-induced reduction in helical content reported for PEF-treated pea and soy proteins (Zhang et al., 2021; Sha and Xiong, 2022).

It is important to note that these transitions are not simply a direct α-helix-to-β-sheet conversion. Circular dichroism and FTIR data across multiple studies indicate that β-turn content typically remains relatively stable or decreases only modestly, while random coil content consistently increases alongside the reduction in α-helix content (Jambrak et al., 2013). This pattern suggests that the dominant process is a loss of ordered secondary structure overall, with polypeptide segments released from disrupted α-helices distributing between newly formed, loosely organised β-sheet-like aggregates and disordered random-coil conformations, depending on local hydrophobicity, ionic strength, and the residual energy input. Moderate processing intensities tend to favour the random-coil pathway, which increases molecular flexibility and solvent accessibility without extensive re-aggregation, whereas higher-intensity or prolonged treatment favours the formation of new, often intermolecular, β-sheet structures that can drive irreversible aggregation (Hu et al., 2013; Chen et al., 2022).

3.1.2 Exposure of functional moieties

Partial unfolding of plant proteins during non-thermal processing can expose chemical groups that are normally buried within the interior of the protein structure. These groups include hydrophobic amino acid residues, sulfhydryl (–SH) groups, aromatic rings, and charged side chains. The exposure of these functional moieties strongly influences the functional behaviour of proteins in food systems (Li et al., 2020). Sulfhydryl groups are particularly important because they can form disulfide bonds during subsequent processing or storage. The formation of new disulfide linkages stabilizes protein networks and enhances gel strength. Hu et al. (2013) observed that ultrasound treatment increased the availability of free sulfhydryl groups in soy protein isolates, resulting in improved gel formation and increased water-holding capacity. Similar findings were reported by Xiong (2017), who highlighted the importance of sulfhydryl-mediated crosslinking in plant protein structuring for meat analogues.

Changes in surface hydrophobicity also influence protein functionality. Li et al. (2020) reported that ultrasound-treated pea proteins exhibited increased surface hydrophobicity because of the exposure of previously buried hydrophobic residues. These exposed regions enhanced the ability of proteins to interact with lipid phases and improved their capacity to stabilize oil–water emulsions. O’Sullivan et al. (2016) similarly observed improved interfacial activity and stronger viscoelastic films in ultrasound-treated pea protein systems. Cold plasma treatment introduces additional chemical modifications through the generation of reactive oxygen species that interact with protein surfaces. These oxidative reactions may introduce carbonyl or hydroxyl groups, alter surface charge distribution, and increase protein polarity (Misra et al., 2019). Ji et al. (2018) reported that cold plasma treatment increased the dispersibility and solubility of peanut proteins through oxidative modification of surface residues. Such changes strongly affect hydration behaviour, protein–protein interactions, and protein–lipid interactions in complex food matrices. Together, these molecular-level modifications determine how proteins interact within food systems and ultimately influence the stability, texture, and functionality of plant-based formulations. These molecular-level changes are summarised in Figure 3.

Taken together, the molecular-level events described above form a linked mechanistic cascade rather than a set of isolated observations. Cavitational shear or electric-field-induced polarisation first disrupts the intramolecular hydrogen bonds stabilising α-helices, which drives partial unfolding of the polypeptide chain; this unfolding exposes previously buried hydrophobic residues and free sulfhydryl groups to the aqueous environment; the newly exposed hydrophobic patches then drive intermolecular association, either as controlled, functionally beneficial aggregation or, if processing intensity is excessive, as extensive hydrophobic and disulfide-mediated aggregation that reduces solubility (Hu et al., 2013; Ampofo and Ngadi, 2022). This cascade—hydrogen-bond disruption → unfolding → hydrophobic/thiol exposure → controlled or excessive aggregation—is the mechanistic thread that connects the molecular-scale changes.

3.1.3 Influence of native protein conformation and botanical source

The discussion above treats ‘plant protein’ as a broadly uniform substrate, yet proteins from different botanical sources possess markedly different native secondary and tertiary architectures that dictate their structural response and resistance to non-thermal and electro-thermal processing. Soy and pea proteins are dominated by the globulin fractions 7S and 11S, which are compact, largely β-sheet-rich, disulfide-stabilised oligomeric proteins assembled from multiple non-covalently and covalently linked subunits (Akharume et al., 2021). Because the 11S fraction contains a substantially higher density of intramolecular disulfide bonds than the 7S fraction, it is generally more resistant to cavitation- and field-induced unfolding, while the more loosely associated 7S fraction unfolds more readily and shows greater increases in surface hydrophobicity and solubility following identical ultrasound treatment (Sha and Xiong, 2022). This means that the overall structural response reported for a ‘pea protein isolate’ or ‘soy protein isolate’ is a composite of two structurally distinct globulins that respond differently to the same processing conditions.

Cereal storage proteins such as wheat gluten proteins (gliadins and glutenins) belong to the structurally distinct prolamin class, which is characterised by a high proportion of glutamine and proline residues, limited α-helical or β-sheet content in the native state, and stabilisation dominated by extensive inter-chain disulfide crosslinking and hydrogen bonding within repetitive, proline-rich domains rather than by the compact globular folds seen in legume globulins (Akharume et al., 2021). As prolamins are only sparsely organised into classical secondary structural elements to begin with, ultrasound- or PEF-induced disruption manifests less as a loss of α-helix or β-sheet content and more as changes in disulfide crosslink density, aggregate size, and viscoelastic network structure, as has been reported for cold-plasma-treated wheat protein (Segat et al., 2016). Legume proteins from other sources, such as faba bean globulins, occupy an intermediate position; they share the 7S/11S globulin architecture of soy and pea but differ in subunit composition and disulfide distribution, which alters their unfolding threshold and the resulting balance between improved solubility and susceptibility to re-aggregation under ultrasound-assisted alkaline treatment (Alavi et al., 2021).

These differences have direct practical implications for process design. Processing parameters (ultrasound intensity and duration, PEF field strength and pulse number, plasma exposure time) that are optimal for unfolding a disulfide-poor 7S globulin may be insufficient to meaningfully modify a disulfide-rich 11S globulin or a prolamin network, while conditions aggressive enough to unfold the latter may over-process and re-aggregate the former (Kong et al., 2025). A critical, source-aware interpretation of structural outcomes is therefore necessary before generalising processing recommendations across the plant protein category as a whole, and future studies should report globulin subunit composition or prolamin subclass alongside processing conditions to enable meaningful cross-study comparison.

3.2 Supramolecular and micro-scale assembly

3.2.1 Aggregation and self-assembly

At the supramolecular level, non-thermal processing technologies influence how protein molecules associate to form larger structural assemblies. Plant protein isolates often contain large aggregates generated during extraction, precipitation, and drying processes. These aggregates can reduce solubility and limit functional performance in food systems (Karki et al., 2010).

Ultrasound treatment is particularly effective in modifying such aggregates. Cavitation-induced shear forces disrupt large protein clusters and reduce them to smaller particles with greater surface area. Zhao et al. (2018) demonstrated that ultrasound treatment reduced the size of rice protein aggregates and improved both solubility and foaming capacity. Similar observations were reported by Jambrak et al. (2008), who found that ultrasound treatment improved protein dispersion and hydration behaviour by reducing aggregate size.

PEF processing influences protein assembly through a different mechanism. By inducing partial unfolding and altering electrostatic interactions, PEF treatments can promote controlled reassembly of proteins into more ordered structures. In some systems, these processes encourage the formation of fibrous protein networks suitable for plant-based meat analogues. Wang et al. (2022) demonstrated that PEF pretreatment improved the formation of structured soy protein networks, resulting in enhanced gel strength and elasticity. Similar findings were reported by Toepfl et al. (2007), who showed the capacity of PEF processing to modify biopolymer organization via electroporation-induced structural changes.

Self-assembly behaviour is also influenced by environmental factors such as pH, ionic strength, and protein concentration. Recent studies by Gharibzahedi and Smith (2021) indicate that combining non-thermal treatments with controlled environmental conditions can further improve protein organization and functionality. These supramolecular assemblies ultimately determine the texture, stability, and mouthfeel of protein-based foods. Representative changes in aggregation and microstructural organisation are illustrated in Figure 4.

3.2.2 Interfacial behaviour

The ability of proteins to stabilize emulsions and foams depends on their behaviour at oil–water and air–water interfaces. Non-thermal processing technologies boost this behaviour by collective protein flexibility and modifying surface chemistry (Arzeni et al., 2012).

Following ultrasound or PEF treatments, proteins often display improved mobility and increased adsorption at interfaces. Once adsorbed, proteins reorganize to form viscoelastic interfacial films that stabilize droplets and air bubbles. O’Sullivan et al. (2016) observed that ultrasound-treated pea proteins formed stronger interfacial films than untreated proteins, leading to improved emulsion stability and reduced droplet coalescence.

Cold plasma treatments can further modify interfacial behaviour through oxidative surface reactions that increase polarity and protein reactivity. Misra et al. (2019) demonstrated that plasma-treated proteins exhibited enhanced emulsifying activity because oxidative modifications improved interfacial adsorption and film formation. Similar observations were reported by Bahrami and Khodaei (2020), who noted that plasma-induced surface oxidation improved foam stability and emulsion formation in plant proteins.

Such improvements are particularly important for products such as plant-based beverages, dairy alternatives, whipped toppings, and aerated desserts, where stable emulsions and foams are essential for product quality and shelf stability.

3.3 Macro-scale structure–property relationships

Structural changes occurring at molecular and supramolecular levels ultimately influence the macroscopic properties of protein-based food systems. These properties include viscosity, yield stress, gel strength, elasticity, water-holding capacity, and rheological behaviour. Rheological measurements such as storage modulus (G′), loss modulus (G″), and viscosity profiles are commonly used to evaluate these characteristics (Sakr and Liu, 2014).

Non-thermal processing can significantly modify these macroscopic properties by influencing protein network formation and intermolecular interactions. Partial unfolding of proteins promotes intermolecular associations that strengthen gel networks and improve elasticity. Sakr and Liu (2014) demonstrated that ohmic heating produced more uniform protein gels compared with conventional thermal treatments. Ultrasound-treated soy and pea proteins have similarly been shown to form stronger and more elastic gels because of improved hydration and controlled aggregation behaviour (Hu et al., 2013; Li et al., 2020).

Recent investigations also indicate that hybrid processing technologies can further improve macroscopic functionality. Wang et al. (2022) reported that combined ultrasound and PEF processing enhanced gel strength and water retention in soy protein systems. Boukid et al. (2021) additionally highlighted the potential of integrating non-thermal technologies to tailor rheological properties for specific food applications.

These macroscopic properties are critical for consumer acceptance because they determine texture, mouthfeel, and sensory quality. Improved protein networks also contribute to better processing stability, reduced phase separation, enhanced water retention, and extended shelf life. Through careful application of non-thermal processing technologies, plant protein ingredients can therefore be engineered with tailored structural and functional properties suitable for next-generation sustainable food systems. These macro-scale outcomes are summarised in Figure 5.

In this whole section, it is thus observed how a single processing event initiates changes that propagate across scales: molecular-level hydrogen-bond disruption and partial unfolding expose hydrophobic and reactive groups that drive supramolecular aggregation and altered interfacial behaviour, which in turn determine the macroscopic solubility, emulsification, foaming, gelation and functional properties. Figure 6 summarises this multiscale cascade in a single conceptual diagram, explicitly linking molecular conformational change to aggregate-level structural change and, ultimately, to functional outcome, and highlighting processing intensity as the key variable that determines whether the cascade terminates in improved functionality or in the overprocessing-related deterioration discussed in the next section.

4 Impact on functional and nutritional attributes

Non-thermal processing technologies have significantly broadened the application potential of plant proteins in modern food systems. In addition to inducing structural transformations at molecular and supramolecular levels, these technologies influence how proteins interact with water, lipids, carbohydrates, and other food constituents. Such interactions determine essential techno-functional properties including solubility, emulsification, foaming, gelation, viscosity, and water-holding capacity (Day, 2013; Lam et al., 2016). Consequently, technologies such as ultrasound, pulsed electric fields (PEF), cold plasma, high-pressure processing, and ohmic heating are increasingly employed to improve both the functional and nutritional quality of plant-derived proteins.

These developments have contributed substantially to the rapid expansion of plant-based meat analogues, dairy alternatives, protein-fortified beverages, and functional foods designed to mimic the texture and sensory properties of animal-derived products while offering sustainable and nutritionally balanced alternatives (Poore and Nemecek, 2018; Willett et al., 2019). Moreover, hybrid processing strategies integrating multiple non-thermal technologies have demonstrated synergistic effects capable of further enhancing protein functionality and stability (Boukid et al., 2021).

4.1 Techno-functional properties

4.1.1 Solubility enhancement

Protein solubility is one of the most critical functional attributes because it directly affects emulsification, foaming, digestibility, and gelation behaviour. Plant proteins frequently exhibit poor solubility because protein aggregation occurs during extraction, precipitation, and drying operations. Aggregation is primarily driven by hydrophobic interactions and intermolecular bonding, which produce compact structures that disperse poorly in aqueous environments (Carbonaro et al., 2015).

Non-thermal processing technologies offer effective approaches for improving protein solubility by inducing partial unfolding and reducing aggregate size. Ultrasound processing, in particular, enhances solubility through cavitation-induced shear forces that disrupt protein aggregates and expose hydrophilic amino acid residues (Jambrak et al., 2008). Li et al. (2020) demonstrated that ultrasound treatment significantly increased the solubility of pea protein isolates through particle-size reduction and structural rearrangement. Similar improvements were observed in rice protein systems, where Zhao et al. (2018) reported increased hydration and dispersibility after ultrasound treatment.

Additional studies have shown that PEF processing can improve protein dispersion by modifying electrostatic interactions and surface charge distribution (Jaeger et al., 2012). Likewise, high-pressure processing has been reported to improve the hydration characteristics of legume proteins by disrupting non-covalent interactions that stabilize aggregates (Campus, 2010). Cold plasma treatments may further improve solubility by introducing oxygen-containing functional groups onto protein surfaces, thereby increasing polarity and water affinity (Bahrami and Khodaei, 2020).

Improved solubility is particularly important in protein beverages and dairy alternatives because poor dispersion often causes sedimentation and gritty textures. Enhanced dispersibility contributes to smoother mouthfeel, improved product consistency, and greater storage stability (Gharibzahedi and Smith, 2021).

4.1.2 Emulsification and foaming

Proteins possess amphiphilic characteristics because they contain both hydrophilic and hydrophobic regions. This dual nature enables proteins to adsorb at oil–water and air–water interfaces, allowing them to function as natural emulsifiers and foaming agents. The efficiency of plant proteins in these applications depends strongly on molecular flexibility and interfacial adsorption behaviour (Arzeni et al., 2012).

Non-thermal processing technologies improve emulsification and foaming by inducing partial unfolding and exposing hydrophobic residues that interact more effectively with lipid phases. Tang and Ma (2009) observed that ultrasound-treated soy proteins displayed improved interfacial activity and enhanced emulsion stability because of increased molecular flexibility. Similarly, O’Sullivan et al. (2016) demonstrated that ultrasound-treated pea proteins formed stronger viscoelastic films around air bubbles, resulting in superior foam stability.

PEF processing also enhances interfacial behaviour. Zhang et al. (2021) reported that PEF-treated pea proteins produced emulsions with smaller droplet sizes and greater kinetic stability than untreated proteins. Wang et al. (2022) further demonstrated that combined ultrasound and PEF treatments generated synergistic improvements in emulsification performance and foam stability.

Cold plasma processing modifies interfacial properties through oxidative reactions that alter surface polarity and molecular charge distribution (Misra et al., 2019). Ji et al. (2018) reported that plasma-treated peanut proteins exhibited improved foaming capacity and emulsion stability because oxidative surface modifications enhanced interfacial adsorption. Similar findings were described by Pankaj et al. (2018), who noted that controlled plasma treatment improved protein surface reactivity without causing extensive structural degradation.

These functional improvements are especially valuable in products such as vegan mayonnaise, whipped toppings, dairy alternatives, aerated desserts, and salad dressings. Enhanced protein functionality also reduces dependence on synthetic emulsifiers and stabilizers, thereby supporting clean-label product development.

4.1.3 Gelation behaviour

Gelation is a fundamental functional property that determines the texture and structural integrity of many protein-rich foods, including meat analogues, yogurt substitutes, and protein gels. Protein gelation involves partial unfolding followed by intermolecular interactions such as hydrophobic association, hydrogen bonding, electrostatic attraction, and disulfide cross-linking to form a three-dimensional network capable of entrapping water (Xiong, 2017).

Non-thermal technologies strongly influence gelation behaviour because they expose reactive groups while minimizing thermal degradation. Hu et al. (2013) demonstrated that ultrasound treatment increased the availability of sulfhydryl groups in soy protein isolates, promoting stronger gels with improved water-holding capacity. Similar observations were reported by Li et al. (2020), who found that ultrasound-treated pea proteins formed more elastic gel structures because of enhanced intermolecular interactions.

PEF treatments can also promote controlled protein unfolding and network formation. Wang et al. (2022) observed that PEF improved gel cohesiveness and elasticity in soy protein systems. Ohmic heating has additionally shown promising effects on gel structure because rapid and uniform internal heating minimizes localized overheating and promotes homogeneous protein network formation (Sakr and Liu, 2014).

Recent studies indicate that hybrid processing approaches combining ultrasound with enzymatic treatments or pH-shifting strategies can further improve gelation behaviour and structural uniformity (Boukid et al., 2021). Such improvements are highly relevant for plant-based meat analogues, where protein networks must replicate the fibrous texture and chewiness of animal muscle tissue (Kyriakopoulou, Dekkers, and van der Goot, 2019). Table 2 consolidates the processing conditions, structural modifications, and functional improvements reported for the protein sources discussed above.

Recent studies further emphasize the growing importance of emerging non-thermal technologies in improving the functional and nutritional quality of plant proteins. Novel processing approaches such as ultrasound, PEF, and hybrid electro-physical treatments have demonstrated considerable potential for enhancing protein dispersion, stability, and structural flexibility while minimizing thermal degradation (Aguiló-Aguayo et al., 2011). These technologies are increasingly viewed as sustainable alternatives to conventional thermal processing because they preserve heat-sensitive nutrients and reduce energy consumption.

High-intensity ultrasound treatments have been shown to improve the functional behaviour of soy proteins by altering molecular conformation and reducing aggregate size. Deng et al. (2018) reported that ultrasound processing enhanced water-holding capacity, emulsifying stability, and gelation behaviour in soy protein systems through controlled structural unfolding. Similar observations were discussed by Knorr et al. (2004), who highlighted the broad applicability of ultrasonics in food processing for improving hydration behaviour, extraction efficiency, and interfacial functionality of food biopolymers.

PEF processing has also attracted increasing attention for its ability to influence protein conformation and intermolecular interactions. Qin et al. (2016) demonstrated that PEF treatment modified the structural organization of food proteins, leading to improved solubility and enhanced emulsifying properties. These effects were associated with molecular polarization and partial unfolding induced by electric field exposure.

More recently, comprehensive reviews have emphasized the role of integrated non-thermal technologies in tailoring plant protein functionality for next-generation food applications. Sharma et al. (2021) concluded that combining ultrasound, PEF, cold plasma, and other innovative processing strategies can significantly improve techno-functional properties such as gelation, foaming, digestibility, and interfacial stability. Such advances are expected to support the development of high-quality plant-based foods with improved sensory and nutritional attributes.

4.2 Nutritional bioavailability

4.2.1 Reduction of anti-nutritional factors

Plant proteins often contain anti-nutritional compounds such as phytates, tannins, lectins, and trypsin inhibitors that interfere with digestion and reduce mineral bioavailability (Gilani et al., 2012). Traditional heat treatments may reduce these compounds but frequently cause nutrient degradation and undesirable flavour changes.

Non-thermal technologies provide alternative strategies for reducing anti-nutritional factors while preserving nutritional quality. Ultrasound processing generates cavitation forces capable of disrupting protein–inhibitor complexes and weakening interactions that stabilize anti-nutritional compounds (Karki et al., 2010). Gilani et al. (2012) reported that ultrasound treatment reduced trypsin inhibitor activity in legume proteins and significantly improved digestibility.

PEF processing may also reduce anti-nutritional interactions by permeabilizing cellular structures and increasing enzyme accessibility (Barba et al., 2015). Toepfl et al. (2007) further demonstrated that PEF processing preserved heat-sensitive nutrients while improving extraction efficiency. High-pressure processing has likewise been shown to reduce lectin activity and improve mineral bioavailability in legumes (Campus, 2010).

The reduction of anti-nutritional compounds is particularly important for populations heavily dependent on plant-based diets because enhanced nutrient availability improves dietary protein quality and supports long-term nutritional adequacy.

4.2.2 Improved digestibility

Protein digestibility is closely related to the accessibility of peptide bonds to digestive enzymes such as pepsin and trypsin. In tightly folded or highly aggregated proteins, many peptide bonds remain inaccessible, reducing hydrolysis efficiency during digestion (Carbonaro et al., 2015).

Non-thermal processing technologies improve digestibility by modifying protein conformation and exposing previously hidden enzymatic cleavage sites. Ultrasound and PEF treatments disrupt tertiary and quaternary structures, thereby increasing susceptibility to enzymatic hydrolysis (Barba et al., 2015).

Carbonaro et al. (2015) reported that ultrasound-induced structural changes enhanced the digestibility of plant proteins by increasing enzyme accessibility. Liang et al. (2020) similarly observed that non-thermal processing improved amino acid bioavailability because partial unfolding increased structural flexibility and reduced aggregate formation.

Recent investigations by Piccolomini et al. (2022) verified that combined ultrasound and enzymatic treatments improved in vitro digestibility and peptide release in legume proteins. Likewise, high-pressure processing has been shown to enhance proteolytic susceptibility in soy and pea proteins by disrupting intermolecular interactions (Panozzo et al., 2014).

Improved digestibility has major nutritional implications for infants, elderly populations, athletes, and individuals recovering from illness because increased amino acid availability supports tissue repair, metabolic regulation, and muscle synthesis.

4.3 Sensory quality and stability

Beyond influencing structural and nutritional characteristics, non-thermal processing technologies can significantly affect sensory quality and product stability. Plant proteins are often associated with undesirable flavour characteristics such as bitterness, grassy notes, and beany aromas arising from volatile compounds and protein–phenolic interactions (Day, 2013).

Non-thermal treatments can alter protein conformation and surface chemistry in ways that influence flavour binding and volatile release. Structural modifications may reduce interactions between proteins and flavour-active molecules, thereby improving the overall sensory profile of plant-based foods.

Cold plasma processing has attracted considerable interest because of its influence on oxidative stability and antioxidant activity. Misra et al. (2019) reported that cold plasma treatment increased antioxidant activity and delayed lipid oxidation in protein-containing systems. Reduced oxidation helps preserve flavour quality and extend shelf life. Similar findings were reported by Bahrami and Khodaei (2020), who observed improved oxidative stability in plasma-treated proteins.

Improved interfacial properties also pay to enhanced physical stability in emulsified systems. Proteins with stronger interfacial films reduce creaming, phase separation, and foam collapse during storage (O’Sullivan et al., 2016). Such advances are essential for maintaining the texture, consistency, and visual appearance of food products throughout shelf life.

However, excessive treatment intensity may induce undesirable oxidation and off-flavour development. Therefore, optimization of processing conditions remains essential to balance improvements in functionality and stability with preservation of nutritional and sensory quality (Pankaj et al., 2018). Table 3 provides a chronological summary of representative studies discussed throughout this review.

TABLE 3. Representative studies on non-thermal processing of plant proteins and their structural and functional outcomes.

4.4 Risks of overprocessing: Balancing structural modification and functional deterioration

The earlier section has emphasised the beneficial structural modifications achievable through non-thermal and electro-thermal processing, but these benefits are not unlimited, and processing intensity and duration must be kept within a defined favourable range. Beyond this range, the same mechanisms that improve functionality at moderate intensity begin to cause functional deterioration. For ultrasound treatment, water sonolysis at high acoustic intensities generates hydroxyl and other free radicals that oxidise newly exposed free sulfhydryl groups back into disulfide bonds, promoting irreversible protein re-aggregation rather than continued unfolding; this oxidative re-aggregation pathway has been documented for soy, black bean, and other legume proteins subjected to prolonged or high-power ultrasound, and typically manifests as a decline in solubility and emulsifying capacity after an initial improvement (Hu et al., 2013; Ampofo and Ngadi, 2022). A closely related effect occurs in cold plasma processing, where excessive exposure to reactive oxygen and nitrogen species can over-oxidise surface amino acid residues, producing carbonylation and cross-linking reactions that stiffen the protein surface and reduce, rather than improve, interfacial activity (Segat et al., 2016).

Overprocessing also has a direct effect on digestibility and sensory quality. Once excessive aggregation has occurred, the resulting dense, disulfide- and hydrophobically-linked aggregates can become less accessible to digestive proteases than either the native protein or the moderately processed, partially unfolded form, producing a characteristic inverted U-shaped relationship between processing intensity and digestibility, with an intermediate optimum rather than a monotonic improvement (Carbonaro et al., 2015). Oxidative modification of aromatic and sulfur-containing amino acids under high-intensity ultrasound or extended cold plasma exposure can also generate off-flavour compounds and altered colour, limiting consumer acceptance even when technical functional metrics such as solubility remain favourable (Xiong and Guo, 2020). These findings indicate that the point at which structural modification becomes detrimental rather than beneficial is technology- and protein-source-specific, and is generally reached well before complete protein unfolding or maximal cavitation intensity.

From a practical standpoint, these overprocessing risks define the operating window within which non-thermal and electro-thermal technologies should be applied. For ultrasound, this typically corresponds to moderate acoustic power and treatment duration sufficient to promote unfolding and hydrophobic exposure without triggering extensive sonolytic oxidation (Rafique et al., 2024). Analogous optimal-intensity windows have been reported for PEF field strength and pulse number and for cold plasma exposure time. Establishing and reporting these windows explicitly, rather than reporting only the single most favourable condition tested, would substantially improve the practical value of future structural studies for process design.

5 Modelling and predictive optimization of non-thermal protein processing

Non-thermal protein processing involves complex structural transformations that occur across multiple scales, from molecular interactions to macroscopic functional properties. Optimizing these processes is challenging because many processing variables interact simultaneously, often producing non-linear and difficult-to-predict outcomes. Traditional experimental approaches based on trial-and-error can therefore be time-consuming and resource-intensive. To address these challenges, researchers increasingly employ computational modelling and data-driven approaches to better understand and optimize non-thermal protein processing. These tools help identify relationships between processing parameters and protein structural changes, allowing more efficient design of processing conditions that achieve desired functional outcomes.

5.1 Molecular dynamics and computational modelling

Molecular dynamics simulations have become an important tool for investigating protein behaviour under non-thermal processing conditions. These simulations allow researchers to observe atomic-level structural changes in proteins when subjected to external forces such as mechanical stress, electric fields, or pressure variations. By tracking molecular movements over time, molecular dynamics models provide insights into how protein folding patterns respond to different processing environments. Studies using molecular dynamics simulations have shown that ultrasound treatment can disrupt hydrogen bonds that stabilize α-helices and β-sheets, resulting in partial unfolding of protein structures. Such structural modifications often increase the exposure of hydrophobic or reactive groups, which correlates with improvements in solubility and interfacial activity observed experimentally.

Similarly, simulations of proteins exposed to PEF suggest that strong electric fields can influence molecular dipole orientation and modify hydration shells surrounding proteins. These effects may lead to changes in electrostatic interactions and promote structural changes. Although molecular dynamics simulations provide valuable insights, they can be computationally demanding, particularly when modelling large proteins or long-time scales. To overcome these limits, researchers often use coarse-grained modelling approaches that simplify the depiction of molecular systems while preserving key structural features. Multiscale modelling techniques that combine detailed atomistic models with basic depictions allow the study of larger protein systems and more complex processing scenarios.

5.2 Machine learning in process optimization

Machine learning approaches are increasingly applied to optimize non-thermal protein processing by analysing complex experimental datasets. Algorithms such as artificial neural networks, random forest models, and support vector machines are capable of identifying relationships between processing parameters and functional outcomes that may not be evident through conventional statistical analysis. For example, machine learning models can predict how variations in ultrasound intensity, treatment time, or PEF strength influence functional properties such as solubility, emulsifying capacity, and gel strength. By analysing large datasets, these models can identify optimal processing conditions more rapidly than traditional experimental screening methods. Recent studies have demonstrated that integrating machine learning into experimental workflows can significantly accelerate process optimization and reduce the number of required experiments (Wang et al., 2023). Furthermore, the integration of real-time sensors with machine learning algorithms enables adaptive process control. In such systems, processing parameters can be adjusted dynamically based on incoming data, ensuring consistent product quality despite variations in raw materials or operating conditions.

5.3 Predictive structure–property relationships

An important objective in non-thermal protein research is to establish reliable relationships between structural changes at the molecular level and functional properties at the macroscopic level. Structural features such as β-sheet content, surface hydrophobicity, sulfhydryl availability, and degree of aggregation are closely associated with functional attributes, including gel strength, emulsification capacity, and viscosity. For instance, moderate increases in β-sheet content can strengthen protein networks and improve gel firmness, which is beneficial for products such as plant-based meat analogues. However, excessive β-sheet formation may lead to aggregation and reduced solubility. Understanding these relationships allows researchers to identify processing conditions that achieve optimal structural modifications. To develop predictive models, researchers combine experimental techniques such as spectroscopy, particle size analysis, and rheological measurements with computational modelling. This integrated approach enables the development of frameworks that link structural signatures to functional outcomes. These models support the rational design of plant protein ingredients tailored to specific applications, including dairy alternatives, structured meat analogues, and protein-enriched beverages.

6 Industrial feasibility and sustainability considerations

The successful implementation of non-thermal protein processing technologies requires careful evaluation of their feasibility at an industrial scale. Although many studies demonstrate promising results under laboratory conditions, scaling these processes for commercial food production introduces additional technical, economic, and regulatory challenges. Industrial food production systems must operate efficiently at high throughput while maintaining consistent product quality. Therefore, processing technologies must be compatible with existing manufacturing infrastructure and capable of delivering uniform treatment across large volumes of material.

6.1 Scalability of non-thermal processing technologies

Scaling non-thermal processing technologies from laboratory experiments to industrial applications requires significant engineering considerations. Laboratory systems often operate under highly controlled conditions and process relatively small sample volumes, whereas industrial food production requires continuous processing of large quantities of raw materials. Ultrasound processing illustrates these challenges. Industrial-scale ultrasound systems must be designed to deliver acoustic energy uniformly across large volumes of liquid or semi-solid materials. Flow-through reactor designs equipped with multiple transducers are commonly used to achieve more consistent cavitation. However, maintaining uniform energy distribution remains challenging, and uneven cavitation may lead to inconsistent protein modification. Research has shown that large-scale ultrasound systems can improve extraction efficiency and reduce processing time. For example, Chemat et al. (2017) demonstrated that industrial ultrasound reactors increased extraction yields of plant proteins while reducing solvent consumption and processing duration. PEF technology offers advantages for industrial scalability because it can be integrated into continuous-flow processing systems. Industrial PEF systems consist of treatment chambers equipped with electrodes that deliver high-voltage pulses to flowing materials. These systems are already used in food applications such as juice pasteurization, suggesting strong potential for protein processing applications. However, high equipment costs and electrode wear remain important economic considerations. Cold plasma processing presents additional challenges because the treatment requires controlled gas compositions and specialized reactor designs. Ensuring uniform treatment of powders or dense protein materials can be difficult at large scales. Nevertheless, pilot-scale plasma reactors have been developed, and studies such as those by Misra et al. (2019) demonstrate the feasibility of plasma treatment for protein powders. Advances in reactor design, automation, and process monitoring are gradually improving the scalability of non-thermal technologies, bringing them closer to widespread industrial implementation.

Beyond reactor design, the economics of scale-up remain a major barrier to industrial adoption. Ultrasound systems for continuous protein processing require custom-engineered, multi-transducer flow cells, and the absence of standardised, off-the-shelf industrial equipment keeps capital costs high and process optimisation largely empirical, since equipment geometry strongly affects the achievable cavitation intensity and its uniformity across the treated volume (Ampofo and Ngadi, 2022). PEF faces a related challenge: no fully standardised continuous PEF processing configuration yet exists for protein-rich, particulate-containing food matrices, and treatment chamber design must account for electrode fouling, gas bubble formation, and non-uniform field distribution, all of which can compromise processing consistency at scale. High initial capital investment for PEF generators and treatment chambers is consistently identified as the single largest impediment to industrial adoption, even though running costs are comparatively low once a system is installed (Alkanan et al., 2024). Ohmic heating systems face their own limits, including electrode corrosion and fouling during extended operation and the need to tailor electrode geometry and voltage to the electrical conductivity of each specific protein formulation, which limits the direct transferability of a given industrial configuration between different protein sources and product formats (Pereira et al., 2007).

6.2 Energy efficiency and life cycle sustainability

Non-thermal processing technologies are often considered more sustainable alternatives to conventional thermal or chemical processing methods. These technologies can reduce energy consumption, minimize solvent use, and decrease the generation of waste streams during protein processing. Life cycle assessment studies indicate that non-thermal processing can significantly reduce the environmental footprint of plant protein production. Ultrasound-assisted extraction, for example, has been shown to reduce processing time by 50%–70% compared with traditional extraction techniques (Chemat et al., 2017). Reduced processing time translates into lower energy consumption and improved operational efficiency. PEF processing can also enhance sustainability by increasing extraction efficiency while reducing reliance on chemical solvents. Barba et al. (2015) reported that PEF-assisted extraction required less water and produced less wastewater than conventional alkaline extraction processes. Although certain non-thermal technologies require high instantaneous power inputs, the overall energy consumption per unit of processed material is often lower because of improved efficiency and shorter processing times. Additionally, reduced use of chemicals and lower wastewater generation contribute to improved environmental performance. Future improvements in reactor design and integration with renewable energy sources may further enhance the sustainability of non-thermal processing technologies.

Quantitative energy comparisons remain limited but are beginning to emerge. PEF systems are generally reported to be more energy-efficient than thermal pasteurisation for liquid and semi-liquid matrices because energy is delivered directly and briefly to the product rather than to the surrounding processing environment, but the specific energy requirement (kJ per kg of product) rises sharply with the field strength and pulse number needed to achieve substantial protein structural modification, which is typically higher than the field strength required for microbial inactivation alone (Alkanan et al., 2024). Ultrasound processing shows a comparable pattern: energy efficiency is highest at moderate acoustic intensities that favour controlled unfolding, while the disproportionately higher energy demand of high-intensity, extended-duration treatment yields diminishing structural and functional returns once the overprocessing regime is reached (Ampofo and Ngadi, 2022). These considerations mean that the most energy-efficient operating point for a given technology is generally the same moderate-intensity window that avoids overprocessing, so that energy efficiency and functional quality are not competing objectives but are, in practice, optimised together.

6.3 Regulatory and safety considerations

Before new food processing technologies can be widely adopted, they must meet regulatory requirements designed to ensure consumer safety and product quality. Regulatory frameworks vary across regions, but they generally require a comprehensive evaluation of potential health risks, nutritional changes, and chemical modifications resulting from processing. Ultrasound and PEF technologies are typically classified as physical processing methods because they modify food structures without introducing chemical additives. As a result, these technologies may face fewer regulatory barriers compared with chemical modification techniques. Cold plasma processing presents additional regulatory considerations because reactive species generated during plasma treatment can induce chemical modifications in proteins and other food components. Regulatory agencies must therefore evaluate the potential formation of novel compounds and assess their safety. Organizations such as the European Food Safety Authority (EFSA) and the United States Food and Drug Administration (FDA) require extensive safety assessments for new food processing technologies. These evaluations may include toxicological studies, allergenicity assessments, nutritional analyses, and shelf-life stability tests. Transparent reporting of safety data and adherence to regulatory standards will be essential for gaining both regulatory approval and consumer acceptance of foods produced using non-thermal processing technologies.

7 Future perspectives and conclusion

Non-thermal processing technologies are increasingly recognized as chief tools for advancing plant protein utilization in modern food systems. As global demand for sustainable protein sources continues to grow, these technologies provide new opportunities to enhance the functionality, nutritional value, and sensory quality of plant-derived proteins. Techniques such as ultrasound, PEF, cold plasma, and ohmic heating allow controlled modification of protein structures without the extensive thermal damage often associated with conventional processing methods. While significant progress has been made in understanding the mechanisms and benefits of these technologies, several scientific and technological challenges remain. Addressing these challenges will be essential for fully realizing the potential of non-thermal processing in the development of next-generation plant-based foods.

Recent advances in food processing emphasize the integration of sustainable and intelligent non-thermal technologies for plant protein functionality. Emerging tactics focus not only on improving conventional techno-functional properties such as solubility, gelation, and emulsification, but also on developing highly controlled structural modifications that support clean-label formulations and resource-efficient manufacturing systems. Researchers have highlighted that next-generation strategies combine physical, electrochemical, and data-driven approaches to optimize protein behaviour while minimizing nutrient degradation and environmental impact (Alexiou et al., 2022). Green processing technologies are receiving particular attention because of their ability to improve protein functionality while reducing water, solvent, and energy consumption. Gao et al. (2022) reported that innovative non-thermal methods can precisely modify intermolecular interactions and hydration properties in plant proteins, thereby improving dispersion stability and rheological performance in complex food systems. Such improvements are especially relevant for the production of plant-based beverages, dairy alternatives, and high-moisture meat analogues.

Recent developments also indicate that hybrid non-thermal systems integrating ultrasound, PEF, enzymatic modification, and computational optimization can generate synergistic effects that exceed the performance of individual treatments. Kumar et al. (2022) observed that combining multiple emerging technologies enables more efficient control over protein unfolding, aggregation, and network formation, resulting in improved textural and sensory properties of plant-based foods. In addition, growing interest in digital food engineering and predictive modelling has accelerated the application of machine learning and process optimization tools in protein processing research. Zhang et al. (2022) emphasized that hybrid intelligent processing systems may allow researchers to predict structure–function relationships more accurately and tailor protein ingredients for specific industrial applications. These advances suggest that future non-thermal processing strategies will increasingly rely on integrated and adaptive technologies capable of producing highly functional, nutritionally enhanced, and sustainable plant protein ingredients.

7.1 Standardization of multiscale characterization

One of the major challenges in current research on plant protein modification is the lack of standardized methods for characterizing structural changes across different scales. Studies investigating non-thermal processing effects often employ different analytical techniques, experimental conditions, and reporting formats. As a result, direct comparison between studies can be difficult, which limits the ability to develop unified conclusions regarding processing–structure–function relationships. Establishing standardized characterization protocols would greatly improve reproducibility and comparability of research findings. Techniques such as spectroscopy, microscopy, particle size analysis, rheology, and interfacial measurements are commonly used to evaluate structural changes in proteins, but consistent application of these tools across studies would enable more reliable data interpretation. Standardized methodologies would also facilitate the creation of shared datasets that could support the development of computational models and machine learning approaches for predicting protein behaviour under different processing conditions. Improved integration of experimental and computational data will be particularly valuable for understanding how structural changes at the molecular level translate into functional properties in complex food systems. Such efforts would help accelerate the development of predictive frameworks that guide the design of plant protein ingredients with targeted functional characteristics.

7.2 Tailor-made proteins and digital food manufacturing

Advances in protein engineering and processing technologies are creating new opportunities to design plant proteins with specific structural and functional properties. By combining non-thermal processing techniques with molecular-level insights into protein behaviour, it becomes possible to tailor plant protein ingredients for particular food applications. This approach is especially relevant for emerging food manufacturing technologies such as digital food fabrication and three-dimensional food printing. These systems require ingredients with carefully controlled rheological properties, including appropriate flow behaviour, rapid structural setting, and the ability to maintain defined shapes after processing. Non-thermal processing methods can modify protein structures in ways that enable these characteristics, allowing proteins to function as effective structural components in digitally manufactured foods. The integration of protein engineering, advanced processing technologies, and digital food manufacturing also opens new possibilities for personalized nutrition. By tailoring protein structures and functionalities, food products can be designed to meet specific nutritional requirements, sensory preferences, or health objectives. Such developments could play an important role in future food systems where customization, sustainability, and nutritional optimization are increasingly prioritized.

7.3 Conclusion

Non-thermal processing technologies represent an important advancement in the field of plant protein research and food engineering. Techniques such as ultrasound, PEF, cold plasma, and ohmic heating enable controlled modification of protein structures across multiple scales, from molecular conformations to macroscopic food structures. These modifications can enhance functional performance, improve digestibility, and contribute to better sensory characteristics in plant-based foods. The integration of computational modelling, machine learning, and advanced analytical methods is further strengthening our ability to understand and control these processes. By linking structural transformations with functional outcomes, researchers and food technologists can design plant protein ingredients that meet the requirements of diverse food applications. As research and technological development continue, non-thermal processing is expected to play an increasingly important role in the production of sustainable protein ingredients. By combining innovations in molecular science, process engineering, and digital manufacturing technologies, the food industry will be better equipped to develop plant-based products that are nutritionally balanced, environmentally sustainable, and appealing to consumers.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Daniela Lamas, Institute of Marine and Coastal Research (IIMyC), Argentina

Reviewed by: Alkatuzzakia Akhi, Bangladesh Agricultural University, Bangladesh

Ge Xu, Nanjing Normal University, China