A review of ultrasonic pasteurization in milk and milk-based beverages using different ultrasound systems and thermal conditions: from mechanistic insights to regulatory challenges Authors: Rina Yu, Sin-Young Park, Prabhathma Yasasvi Rathnayake, So Eun Yeo, Chemin Nam, Hyun Uk Cho, Ki Ho Baek, Hae In Yong Published: 2026 ABSTRACT AbstractMilk and milk-based beverages are highly susceptible to contamination by foodborne pathogens and spoilage bacteria, making effective pasteurization essential for ensuring microbial safety. Ultrasound has emerged as a promising alternative pasteurization technology because acoustic cavitation generates physical and sonochemical effects capable of microbial inactivation. However, the antimicrobial efficacy of ultrasound varies considerably depending on ultrasound system configuration, processing conditions, and thermal contribution during treatment. This review provides a microbial safety-oriented overview of ultrasonic pasteurization in milk and milk-based beverages, with particular emphasis on differences among ultrasound systems and processing modes. Specifically, microbial inactivation is compared among cavitation-driven non-thermal ultrasound, ultrasound with natural temperature rise, and thermally assisted ultrasonic treatments. In addition, this review discusses the mechanisms of microbial inactivation, regulatory considerations for industrial applicability, and technological challenges associated with ultrasonic pasteurization. These findings provide insights into the optimization and industrial implementation of ultrasound-based pasteurization technologies for milk and milk-based beverages.Keywords: Milk, Milk-based beverage, Ultrasound, Pasteurization, Safety Introduction Milk is widely recognized as a nutritionally balanced food and has long been regarded as an essential component of the human diet (Pereira, 2014). Milk provides 18 essential nutrients, including vitamin D, calcium, and potassium, and its consumption has been associated with health maintenance, growth, anticancer effects, anti-inflammatory effects, and antioxidant effects (Zhang et al., 2021). For these reasons, milk is recommended in dietary guidelines worldwide (DGA, 2020–2025; Zhang et al., 2021). Positive consumer perceptions of milk’s health benefits have contributed to the global expansion of the milk-based beverage market (Holkar et al., 2019). In addition to raw milk, various milk-based beverages have recently been introduced and are now widely consumed. The demand for flavored milk beverages containing chocolate, banana, or strawberry has increased with changing consumer preferences for both taste and nutrition (Hossain et al., 2025), while milk beverages enriched with natural ingredients such as cacao and green tea have gained popularity among health-conscious consumers (Bhagat et al., 2019). Moreover, flavored milk beverages have been shown to enhance overall milk consumption among children and adolescents (Patel et al., 2018). Milk and milk-based beverages contribute to human health due to their rich nutritional composition. However, they also provide a favorable environment for microbial growth, thereby posing potential risks of spoilage and subsequent foodborne illness (Dash et al., 2022). Milk transported and stored at 4 °C is susceptible to spoilage caused by the proliferation of psychrotrophic bacteria, which can lead to quality deterioration and the development of off-flavors and off-tastes (Dash et al., 2022; Meer et al., 1991). Moreover, in recent decades, pathogenic bacteria such as Escherichia coli O157:H7, Salmonella Enteritidis, Campylobacter jejuni, and Listeria monocytogenes have been detected in milk and milk-based beverages, resulting in numerous reported outbreaks of foodborne illness (Dalton et al., 1997; Heuvelink et al., 2009; Jenkins et al., 2022; Zeinhom & Abdel-Latef, 2014). Therefore, pasteurization is employed to ensure the microbiological safety of milk and milk-based beverages (Ataallahi et al., 2024). In the dairy industry, conventional heat treatments are classified as low-temperature long-time (LTLT, 63 °C for 30 min), high-temperature short-time (HTST, 72–75 °C for 15 s), and ultra-high temperature (UHT, 135–150 °C for 1–10 s) processes (Dash et al., 2022). However, heat treatment can cause quality deterioration in milk, including nutrient degradation, flavor alteration, and a reduction in bioactive compounds (Lee et al., 2025; Makararpong et al., 2024). Due to these limitations, the need for innovative pasteurization technologies has been emphasized, and ultrasound has recently attracted considerable attention as a promising alternative (Rathnayake et al. 2025a). D’Amico et al. (2006), Mudgil et al. (2022), and Shamila-Syuhada et al. (2016) reported that ultrasound treatment can effectively inactivate pathogenic bacteria in milk. However, the effectiveness of ultrasound for microbial inactivation strongly depends on various processing conditions. These conditions include the type of equipment (e.g., probe, bath, or continuous-flow types), frequency, power, treatment duration, physicochemical properties of the sample, and processing temperature. Therefore, to systematically evaluate the applicability of ultrasound as a novel pasteurization technology for milk and milk-based beverages, it is essential to investigate its microbial inactivation effects under different ultrasound operational conditions. This review aims to examine the pathogenic and spoilage bacteria present in milk and milk-based beverages and to evaluate ultrasonic pasteurization strategies under different ultrasound systems and processing modes. Specifically, this review compares microbial inactivation under cavitation-driven non-thermal ultrasound, ultrasound with natural temperature rise, and thermally assisted ultrasonic treatments. In addition, the mechanisms of microbial inactivation, industrial applicability, scale-up limitations, and technological challenges associated with ultrasonic pasteurization are discussed. Pathogenic and spoilage bacteria in milk and milk-based beverages Bacteria present in milk and milk-based beverages can be broadly classified into two groups: (i) pathogenic bacteria and (ii) spoilage bacteria (Fig. 1). Pathogenic bacteria can cause foodborne illnesses that may result in symptoms such as vomiting, diarrhea, and fever, and in severe cases, may lead to death, thereby posing a significant threat to public health (Heredia and Garcia, 2018). In contrast, spoilage bacteria do not directly cause illness in consumers but are responsible for milk deterioration, leading to undesirable changes in appearance and sensory qualities (Saha et al., 2024). Pathogenic and spoilage bacteria in milk and milk-based beverages Pathogenic bacteria may be introduced into milk and milk-based beverages through direct contact with contamination sources in the farm environment or through excretion from the udders of infected animals (Oliver et al., 2005). When raw milk contaminated with pathogenic bacteria enters a processing facility, the bacteria can persist within biofilms, posing a considerable risk of post-pasteurization contamination and consequently exposing consumers to potential infection (Oliver et al., 2005). Some pathogenic bacteria, including Campylobacter spp., enterohemorrhagic E. coli (EHEC), and Salmonella spp., inhabit the intestinal tracts of ruminants (Heredia and Garcia, 2018). Consumption of unpasteurized or raw milk contaminated with C. jejuni has been associated with outbreaks of campylobacteriosis (Christidis et al., 2016). For example, in 2005, an outbreak involving 22 confirmed cases of C. jejuni infection occurred in the Netherlands following the consumption of raw milk produced on a farm (Heuvelink et al., 2009). In addition, in 2006, an outbreak comprising 1,644 cases of campylobacteriosis was reported in California, United States, linked to the consumption of inadequately pasteurized milk supplied within a correctional facility (Jay-Russell et al., 2013). Outbreaks associated with EHEC infections have also been reported in unpasteurized milk and other dairy products (Heredia and Garcia, 2018). Although most E. coli strains are normal inhabitants of the intestinal tracts of animals, certain serotypes can cause foodborne illness and gastrointestinal disorders. Among these, E. coli O157:H7 produces Shiga toxins that can cause severe diseases such as hemorrhagic colitis and hemolytic uremic syndrome (HUS) and is therefore considered a major concern for both the food industry and public health (Zeinhom & Abdel-Latef, 2014). In 2019, an outbreak occurred in South Yorkshire, United Kingdom, after pasteurized milk became contaminated with unpasteurized milk, resulting in 21 confirmed cases of E. coli O157:H7 infection, three of which progressed to HUS (Jenkins et al., 2022). Salmonella spp. are well-known causes of foodborne illness, typically resulting in diarrhea, abdominal cramps, vomiting, and fever. Outbreaks of Salmonella spp. infections have been reported in association with unpasteurized milk and dairy products (Zeinhom & Abdel-Latef, 2014). According to the California Department of Public Health, at least eight cases of salmonellosis were identified within one month in 2023 among individuals who had consumed a commercially distributed raw milk brand (Weinstein, 2025). Listeria spp. is also one of the major foodborne pathogens detected in milk, although it is not commonly found in the intestinal tracts of ruminants. Listeria spp. is widely distributed in nature and is frequently present in food and soil environments (Heredia and Garcia, 2018). Contamination of dairy products with L. monocytogenes can cause miscarriage or stillbirth in pregnant women and may also lead to septicemia. In 2007, five cases of listeriosis were linked to L. monocytogenes contamination in pasteurized and coffee-flavored milk, resulting in three deaths (Cumming et al. 2009). Additionally, a foodborne illness outbreak occurred at a picnic in Illinois in 1994 due to chocolate milk contaminated with L. monocytogenes, resulting in diarrhea and fever among attendees (Dalton et al., 1997). The spoilage of milk and milk-based beverages is primarily caused by spoilage bacteria such as Pseudomonas spp., lactic acid bacteria, and Serratia spp. Because milk and milk-based beverages are distributed and stored at low temperatures, psychrotrophic bacteria have been a major concern in the dairy industry for several decades (Meer et al., 1991). Among the non–spore-forming psychrotrophic bacteria, Pseudomonas spp. is the most frequently detected in contaminated raw milk (Salgado et al., 2020), with Pseudomonas fluorescens and Pseudomonas putida being the most representative species. These bacteria produce heat-resistant enzymes such as proteases and lipases, which degrade proteins and fats, leading to various quality defects (Tamime, 2009). Lactic acid bacteria metabolize lactose in milk and milk-based beverages, producing lactic acid that lowers pH and promotes curd formation. They also produce acetic and propionic acids that contribute to sour off-flavors (Xu et al., 2020). Spoilage of milk and milk-based beverages by lactic acid bacteria typically occurs under improper storage conditions, particularly when milk is stored at elevated temperatures or when the product composition inhibits the growth of gram-negative aerobic bacteria. This situation is especially common when milk is stored on farms or transported without refrigeration (Tamime, 2009). Serratia spp. can also be present in raw milk stored at 4 °C, contributing to spoilage. Serratia liquefaciens is known to produce heat-resistant proteases that remain active even after UHT treatment, potentially causing spoilage (Salgado et al., 2020). Contamination of dairy cows with Serratia marcescens can result in mastitis and contaminated milk that may appear pink (Rachon et al., 2017). Moreover, S. liquefaciens and S. marcescens are recognized as opportunistic pathogens that can cause infections when the host’s immune system is compromised or when environmental conditions are unfavorable (Givskov et al., 1997; Rachon et al., 2017). Conventional pasteurization methods, such as LTLT and HTST treatments, are commonly used to inactivate microorganisms in milk and milk-based beverages. However, as mentioned above, foodborne outbreaks linked to pathogenic bacteria have still been reported in commercially pasteurized milk and milk-based beverages. Furthermore, conventional thermal treatments may be insufficient to inactivate heat-resistant bacterial spores and enzymes present in milk and milk-based beverages (Khan et al., 2025; Xu et al., 2006). Residual bacterial survival and heat-induced damage to the product can result in quality deterioration during storage. Consequently, there is a growing demand for innovative pasteurization technologies that can effectively inactivate bacteria in milk and milk-based beverages while preserving product quality. Novel ultrasound processing One of the main reasons ultrasound technologies have attracted attention for pasteurizing milk and milk-based beverages is that sound waves are considered a safe, non-toxic, and environmentally friendly technology (Fan et al., 2021). Compared with other modern processing technologies, ultrasound has lower construction costs, consumes less power, and is well-suited for liquid foods (Rathnayake et al. 2025a). Several studies have investigated various types of ultrasounds under different processing conditions and reported that they can be effectively applied to the pasteurization of milk and milk-based beverages, showing potential to replace conventional heat treatments such as LTLT and HTST pasteurization (Abrahamsen & Narvhus, 2022; Chen et al., 2024). Principle of ultrasound generation Ultrasound is a form of energy generated by a transducer that converts electrical energy into acoustic energy (Nam et al., 2026; Ortega-Rivas, 2012). Figure 2 illustrates the principle and effects of cavitation generated during ultrasonic treatment of liquid media. Ultrasonic waves are represented as sinusoidal waves consisting of alternating high-pressure (compression) and low-pressure (rarefaction) regions (Leighton 2007; Rathnayake et al. 2025b). The amplitude refers to the maximum displacement from the baseline, whereas the wavelength represents the distance between two points in the same phase. Frequency denotes the number of wave cycles repeated per unit time. When ultrasound is applied to a liquid, longitudinal waves propagate through the medium, creating periodic regions of compression and rarefaction among the particles (Piyasena et al., 2003). These pressure fluctuations induce acoustic cavitation, leading to the formation of microbubbles within the liquid. In food applications, particular attention is given to inertial (or transient) cavitation, which occurs at high acoustic amplitudes (Kentish & Ashokkumar, 2010). In this process, bubbles undergo rapid expansion followed by violent collapse, generating extreme localized conditions that result in both physical and sonochemical effects (Qin et al., 2024). Mechanisms of microbial inactivation by ultrasound treatment Physically, the asymmetric collapse of bubbles produces intense compression and generates shockwaves, microjets, and microstreaming. Localized hotspots can reach temperatures of up to 5,500 °C and pressures of up to 50,000 kPa (Lauteri et al., 2023). Microjets released during bubble collapse break down larger particles into smaller ones, while interfacial turbulence enhances droplet formation and improves emulsification (Kentish & Ashokkumar, 2010). Sonochemical effects arise from the extreme temperature and pressure conditions generated during bubble collapse, which lead to the formation of reactive radical species (Qin et al., 2024). Primary radicals are produced as a direct result of the high temperatures inside collapsing bubbles. When water vapor is present, hydrogen (H •) and hydroxyl (•OH) radicals are formed according to Eq. (1), and these radicals may subsequently recombine to form various molecular products as shown in Eqs. (2), (3), and (4) (Arefi-Oskoui et al., 2019). (1) H2O Ultrasonic irradiation → H• + •OH. (2) •OH + H• → H2O (3) •OH +• OH → H2O2 (4) H2O2 + H •→ H2O + •OH For a single bubble, a higher internal temperature results in the formation of a greater number of radicals. The internal bubble temperature can be increased by raising the ultrasonic intensity or ambient pressure, or by lowering the bulk liquid temperature (Kentish & Ashokkumar, 2010). At lower frequencies (e.g., 20 kHz), bubble size increases, resulting in greater heat generation (Kentish & Ashokkumar, 2010; Merouani et al., 2014). Consequently, the number of primary radicals produced per bubble increases at lower frequencies. However, in a multi-bubble (multi-cavitation) environment, the total radical yield depends not only on the temperature within each bubble but also on the number of bubbles formed (Kentish & Ashokkumar, 2010). Therefore, for the same liquid volume and ultrasonic power, higher frequencies may produce fewer radicals per bubble; however, because of the greater number of cavitation bubbles, the overall radical yield can be higher. As a result, sonochemical effects are generally most pronounced at intermediate frequencies, typically in the range of 200–500 kHz (Kentish & Ashokkumar, 2010). Regardless of the mechanism by which acoustic energy dissipates, the energy ultimately converts into heat (Kentish & Ashokkumar, 2010). Therefore, the application of ultrasound inevitably causes an increase in temperature within the system. If cooling is not applied simultaneously, this temperature rise can significantly affect food materials and must be carefully considered as a variable during processing. The interaction between ultrasound and heat influences both microbial inactivation and product quality; thus, precise control of processing conditions is essential when applying ultrasound to milk and milk-based beverages. Different types of ultrasounds Ultrasound types can be broadly categorized into probe type, bath type, and continuous flow type systems based on their application method and structural design (Fig. 3). Each type differs in energy delivery, processing uniformity, and suitability for industrial applications (Ortega-Rivas, 2012). Understanding the characteristics of these various ultrasound systems provides a useful basis for evaluating their potential to replace conventional thermal pasteurization processes in milk and milk-based beverages. Types of ultrasound devices: (a) probe type, (b) bath type, and (c) continuous flow type A probe-type ultrasound system consists of a generator, transducer, booster horn, and sonotrode (Soltani Firouz et al., 2023). The generator receives electrical input and converts it into high-frequency electrical energy. The transducer then converts this electrical energy into mechanical vibrations (Ortega-Rivas, 2012). The booster horn amplifies the vibration amplitude, and the sonotrode—attached to the end of the booster—directly contacts the sample to transmit ultrasound energy (Soltani Firouz et al., 2023). An ultrasonic bath employs sandwich-type transducers mounted on the outer walls or bottom of the tank to indirectly irradiate the sample, with water typically serving as the coupling medium. A limitation of the ultrasonic bath is that the water within the bath attenuates ultrasound intensity, resulting in lower power delivery to the target material compared with a probe system (Chemat et al., 2017). Because ultrasound energy is dispersed over a wider area, the power intensity of an ultrasonic bath is approximately 100 times lower than that of a probe system (Soltani Firouz et al., 2023). Both probe type and bath type ultrasound devices deliver energy in a single direction, resulting in non-uniform energy distribution within the system (Kim et al., 2024). Furthermore, if the heat generated during operation is not effectively controlled, it can adversely affect product quality. Recently, continuous flow type ultrasound systems have been developed. Some devices combine the probe type with a continuous flow type system, while others are designed with a cylindrical structure capable of generating ultrasonic energy in a 360° pattern under a continuous flow type system (Chemat et al., 2017; Kim et al., 2024). Compared with conventional ultrasound systems, these devices enable the circulating sample to receive stronger and more uniform energy transmission (Kim et al., 2024). Continuous flow type ultrasound devices allow adjustment not only of fundamental parameters such as power, amplitude, and frequency but also of flow-related variables, including flow rate and number of cycles, thereby influencing the overall ultrasonic effect (Crudo et al., 2014; Van Hekken et al., 2019; Villamiel and de Jon, 2000). Effect of ultrasound on the microbiological safety of milk and milk-based beverages Microbial inactivation mechanism of ultrasound Ultrasound has been identified as a promising pasteurization technology for milk and milk-based beverages. The microbial inactivation effect of ultrasound is attributed to the combined action of physical effects—such as shockwaves, microjets, and shear forces—and sonochemical effects, including the generation of reactive oxygen species (ROS; Fig. 2). Structural disruption of bacterial cell membranes is considered a major antimicrobial mechanism of ultrasound because it facilitates the penetration of antimicrobial agents into the cell (Lin et al., 2019). The physical effects generated by ultrasound—such as shockwaves, microjets, and shear forces produced during cavitation bubble oscillation and collapse, as well as microstreaming around the bubbles—can cause structural damage to bacterial cell walls and membranes (Dai et al., 2020). This physical damage is further enhanced when ultrasound is combined with mild heat, which weakens the protective function of the cell envelope and increases cavitation efficiency (Li et al., 2017). ROS generated through sonochemical effects can penetrate bacterial cells via membrane pores and react with intracellular biomolecules such as proteins, lipids, polysaccharides, and nucleic acids, ultimately leading to cell death (Dai et al., 2020). •OH can cleave the DNA double helix and induce base modifications. ROS promote oxidative degradation of cellular components, resulting in disruption of the bacterial cell wall. For instance, •OH reacts with polyunsaturated fatty acids in the phospholipid bilayer of the membrane, triggering lipid peroxidation (Jana et al., 1990). Additionally, ROS can also be produced through sonoluminescence—a phenomenon in which light is emitted across a broad spectrum during acoustic cavitation in liquids (Harris et al., 2014). However, this antimicrobial effect occurs specifically in the presence of a sonosensitizer with photocatalytic activity, as the light emitted by sonoluminescence excites the sonosensitizer, thereby inducing ROS generation and promoting microbial inactivation (Dai et al., 2020). This oxidative process by ROS compromises membrane integrity and function. Furthermore, it leads to the formation of secondary reactive products that exert additional damage on both the cell wall and membrane. Application of ultrasound for pasteurization of milk and milk-based beverages Microbial inactivation in milk and milk-based beverages is an important aspect for both consumers and the dairy industry. Consequently, the application of ultrasound as a pasteurization technology has attracted considerable attention from researchers. The efficiency of microbial inactivation by ultrasound is generally affected by several factors, including processing parameters (e.g., power, frequency, and treatment time), equipment type, and temperature. Furthermore, the microbial inactivation effect of ultrasound increases with the intensity of processing parameters. Numerous studies have applied ultrasound to milk and milk-based beverages for microbial inactivation under different processing modes, which can be broadly classified into (i) cavitation-driven ultrasound treatment under controlled temperature conditions, (ii) ultrasound with natural temperature rise, and (iii) thermally assisted ultrasonic treatments (Table 1). Table 1Comparison of microbial inactivation under different ultrasound systems and thermal conditions in milk and milk-based beveragesUltrasound system(Thermal condition)*Milk typeTreatment conditionBacterial reduction effectReferenceFixed settingsParametersProbe-type ultrasoundRaw milk- Frequency: 20 kHz- Temperature: 22 ~ 26 °C- Time: 10 minPower: 105 W- Staphylococcus spp.: 0.45 Log CFU/mL- Total aerobic bacteria: 0.68 Log CFU/mL- Lactic acid bacteria: 0.66 Log CFU/mLMudgil et al. (2022)Power: 130 W- Staphylococcus spp.: 0.72 Log CFU/mL- Total aerobic bacteria: 0.91 Log CFU/mL- Lactic acid bacteria: 0.91 Log CFU/mLPower: 140 W- Staphylococcus spp.: 1.39 Log CFU/mL- Total aerobic bacteria: 1.33 Log CFU/mL- Lactic acid bacteria: 1.61 Log CFU/mLPower: 160 W- Staphylococcus spp.: 3.4 Log CFU/mL- Total aerobic bacteria: 4.16 Log CFU/mL- Lactic acid bacteria: 1.50 Log CFU/mLPower: 170 W- Staphylococcus spp.: 3.4 Log CFU/.mL- Total aerobic bacteria: 4.16 Log CFU/mL- Lactic acid bacteria: 1.58 Log CFU/mLPower: 210 W- Staphylococcus spp.: 3.4 Log CFU/mL- Total aerobic bacteria: 4.16 Log CFU/mL- Lactic acid bacteria: 1.49 Log CFU/mLProbe-type ultrasoundRaw milk- Frequency: 24 kHz- Power: 200 W- Temperature: 15 ~ 25 °CTime: 8 min- Total viable counts: 1.22 Log CFU/mL- Psychrotrophic bacteria counts: 1.38 Log CFU/mLChouliara et al. (2010)Time: 16 min- Total viable counts: 2.18 Log CFU/mL- Psychrotrophic bacteria counts: 2.29 Log CFU/mLProbe-type ultrasoundChocolate milk- Power: 400 W- Frequency: 19 kHz- Final temperature: 42 °C (3.0 kJ/cm3)Energy density: 0.3 kJ/cm3- Total aerobic bacteria: approximately 0.8 Log CFU/mLMonteiro et al. (2018)Energy density: 0.9 kJ/cm3- Total aerobic bacteria: approximately 1.2 Log CFU/mLEnergy density: 1.8 kJ/cm3- Total aerobic bacteria: approximately 1.8 Log CFU/mLEnergy density: 2.4 kJ/cm3- Total aerobic bacteria: approximately 2.2 Log CFU/mLEnergy density: 3.0 kJ/cm3- Total aerobic bacteria: 3.56 Log CFU/mLProbe-type ultrasoundWhey beverage- Frequency: 19 kHz- Time: 3 min- Final temperature: 53 °CPower: 200 W- Total aerobic mesophilic bacteria: 0.03 Log CFU/mL- Molds and yeasts:0.2 Log CFU/mLGuimarães et al. (2018)Power: 400 W- Total aerobic mesophilic bacteria: 1 Log CFU/mL- Molds and yeasts:0.4 Log CFU/mLPower: 600 W- Total aerobic mesophilic bacteria: 2 Log CFU/mL- Molds and yeasts:0.4 Log CFU/mLProbe-type ultrasoundKinnow- whey beverage- Power: 400 W (amplitude 70%)- Frequency: 20 kHz- Final temperature: 23.5 °CTime: 5 min- Total aerobic bacteria: 0.6 Log CFU/mL- Molds and yeasts:1.41 Log CFU/mLSiddique et al. (2023)Time: 10 min- Total aerobic bacteria: 0.7 Log CFU/mL- Molds and yeasts:1.46 Log CFU/mLTime: 15 min- Total aerobic bacteria: 0.99 Log CFU/mL- Molds and yeasts:1.61 Log CFU/mLContinuous flow type ultrasoundRaw milk- Frequency: 35 kHz- Temperature: 30–37 °C- Time: 10 minPower: 320 W- Total aerobic bacteria: 1 Log CFU/mLCrudo et al. (2014)Power: 370 W- Total aerobic bacteria: 1.3 Log CFU/mLProbe-type ultrasound(Natural temperature rise during treatment)Raw milk- Frequency: 19 kHz- Power: 475 W- Time: 158 s- Final temperature: 55 °C- Total aerobic mesophilic bacteria: 2.6 Log CFU/mL- Lactic acid bacteria: 2.1 Log CFU/mLScudino et al. (2022)Probe-type ultrasound(Natural temperature rise during treatment)Raw milk- Frequency: 19 kHz- Power: 475 WEnergy density: 1 kJ/mL- Time: 53 s- Final temperature: 34 °C- Total aerobic mesophilic bacteria: 0 Log CFU/ mLScudino et al. (2020)Energy density: 3 kJ/mL- Time: 2 m 38 s- Final temperature: 55 °C- Total aerobic mesophilic bacteria: 1.8 Log CFU/ mLEnergy density: 5 kJ/mL- Time: 4 m 23 s- Final temperature: 76 °C- Total aerobic mesophilic bacteria: 3.92 Log CFU/ mLEnergy density: 7 kJ/mL- Time: 6 m 9 s- Final temperature: 85 °C- Total aerobic mesophilic bacteria: 3.92 Log CFU/ mLProbe-type ultrasound(Natural temperature rise during treatment)Raw milk- Frequency: 24 kHz- Power: 200 W- Time: 15 minAmplitude: 125 μm- Final temperature: 60 °C- Staphylococcus aureus, Listeria monocytogenes, Salmonella Typhimurium, Escherichia coli, Pseudomonas fluorescens, Lactobacillus plantarum, and Lactobacillus pentosus: 8.1 Log CFU/mLShamila-Syuhada et al. (2016)Amplitude 62.5 μm- Final temperature: 50 °C- Staphylococcus aureus: 3.8 Log CFU/mL- Listeria monocytogenes: 3.6 Log CFU/mL- Lactobacillus plantarum: 4.5 Log CFU/mL- Lactobacillus pentosus: 4.6 Log CFU/mL- Salmonella Typhimurium, Escherichia coli, Pseudomonas fluorescens: 5 Log CFU/mLProbe-type ultrasound(Natural temperature rise during treatment)Human milk- Frequency: 20 kHz- Temperature: 40 °C- Final temperature: 59 °C (60 W, 6 min)Energy density: 1,080 kJ/L- Power: 40 W- Time: 9 min- Enterobacter cloacae: 6.31 Log CFU/mL- Staphylococcus epidermidis: 6.07 Log CFU/mLKontopodi et al. (2022)Energy density: 1,080 kJ/L- Power: 60 W- Time: 6 min- Enterobacter cloacae: 6.63 Log CFU/mL- Staphylococcus epidermidis: 6.50 Log CFU/mLEnergy density: 1,628 kJ/L- Power: 60 W- Time: 9 min- Enterobacter cloacae: 6.52 Log CFU/mL- Staphylococcus epidermidis: 6.21 Log CFU/mLContinuous flow type ultrasound(Natural temperature rise during treatment)Raw milk- Frequency: 20 kHz- Power: 150 W- Inlet temperature: 24 °C- Flow rate: 11 mL/min- Time: 102.3 sOutput intensity: Level 5- Final temperature: 51 °C- Pseudomonas fluorescens: 0 Log CFU/mL- Streptococcus thermophilus: 0.5 Log CFU/mLVillamiel and de Jon (2000)Output intensity: Level 10- Final temperature: 76.1 °C- Pseudomonas fluorescens: 4.2 Log CFU/mL- Streptococcus thermophilus: 2.7 Log CFU/mLContinuous flow type ultrasound(Thermally assisted ultrasonic treatment)Raw milk- Dual frequency: 16 & 20 kHz- Power: 1,200 W- Flow rate: 0.45 L/min- Time: 14 minInlet temperature: 42 °C- Total mesophilic bacteria: 0 Log CFU/mL- Total psychrophilic bacteria: 0.5 Log CFU/mLVan Hekken et al. (2019)Inlet temperature: 54 °C- Total mesophilic bacteria: 3.36 Log CFU/mL- Total psychrophilic bacteria: 4.88 Log CFU/mLProbe-type ultrasound(Thermally assisted ultrasonic treatment)Raw milk- Power: 150 W- Frequency: 20 kHz- Time: 18 minTemperature: 20 °C- Total aerobic bacteria: 2.6 Log CFU/mL- Listeria monocytogenes: 1.93 Log CFU/mL- Escherichia coli O157:H7: 4.63 Log CFU/mLD’amico et al. (2006)Temperature: 57 °C- Total aerobic bacteria: 2.96 Log CFU/mL- Listeria monocytogenes: 2.32 Log CFU/mL- Escherichia coli O157:H7: 5.91 Log CFU/mLBath type ultrasound(Thermally assisted ultrasonic treatment)Huma milk- Power: 110 W- Frequency: 40 kHz- Time: 10 minTemperature: 20 °C °- Staphylococcus aureus: 0 Log CFU/mLGomes et al. (2021)Temperature: 30 °C- Staphylococcus aureus: 0.07 Log CFU/mLTemperature: 40 °C- Staphylococcus aureus: 0.13 Log CFU/mLTemperature: 50 °C- Staphylococcus aureus: 0.74 Log CFU/mLTemperature: 60 °C- Staphylococcus aureus: 4.58 Log CFU/mL* Thermal conditions were categorized based on information explicitly reported in the original studies. Some studies did not clearly distinguish between non-thermal and thermally assisted ultrasonic effects Comparison of microbial inactivation under different ultrasound systems and thermal conditions in milk and milk-based beverages - Frequency: 20 kHz - Temperature: 22 ~ 26 °C - Time: 10 min - Staphylococcus spp.: 0.45 Log CFU/mL - Total aerobic bacteria: 0.68 Log CFU/mL - Lactic acid bacteria: 0.66 Log CFU/mL - Staphylococcus spp.: 0.72 Log CFU/mL - Total aerobic bacteria: 0.91 Log CFU/mL - Lactic acid bacteria: 0.91 Log CFU/mL - Staphylococcus spp.: 1.39 Log CFU/mL - Total aerobic bacteria: 1.33 Log CFU/mL - Lactic acid bacteria: 1.61 Log CFU/mL - Staphylococcus spp.: 3.4 Log CFU/mL - Total aerobic bacteria: 4.16 Log CFU/mL - Lactic acid bacteria: 1.50 Log CFU/mL - Staphylococcus spp.: 3.4 Log CFU/.mL - Total aerobic bacteria: 4.16 Log CFU/mL - Lactic acid bacteria: 1.58 Log CFU/mL - Staphylococcus spp.: 3.4 Log CFU/mL - Total aerobic bacteria: 4.16 Log CFU/mL - Lactic acid bacteria: 1.49 Log CFU/mL - Frequency: 24 kHz - Power: 200 W - Temperature: 15 ~ 25 °C - Total viable counts: 1.22 Log CFU/mL - Psychrotrophic bacteria counts: 1.38 Log CFU/mL - Total viable counts: 2.18 Log CFU/mL - Psychrotrophic bacteria counts: 2.29 Log CFU/mL - Power: 400 W - Frequency: 19 kHz - Final temperature: 42 °C (3.0 kJ/cm3) - Frequency: 19 kHz - Time: 3 min - Final temperature: 53 °C - Total aerobic mesophilic bacteria: 0.03 Log CFU/mL - Molds and yeasts:0.2 Log CFU/mL - Total aerobic mesophilic bacteria: 1 Log CFU/mL - Molds and yeasts:0.4 Log CFU/mL - Total aerobic mesophilic bacteria: 2 Log CFU/mL - Molds and yeasts:0.4 Log CFU/mL - Power: 400 W (amplitude 70%) - Frequency: 20 kHz - Final temperature: 23.5 °C - Total aerobic bacteria: 0.6 Log CFU/mL - Molds and yeasts:1.41 Log CFU/mL - Total aerobic bacteria: 0.7 Log CFU/mL - Molds and yeasts:1.46 Log CFU/mL - Total aerobic bacteria: 0.99 Log CFU/mL - Molds and yeasts:1.61 Log CFU/mL - Frequency: 35 kHz - Temperature: 30–37 °C - Time: 10 min Probe-type ultrasound (Natural temperature rise during treatment) - Frequency: 19 kHz - Power: 475 W - Time: 158 s - Final temperature: 55 °C - Total aerobic mesophilic bacteria: 2.6 Log CFU/mL - Lactic acid bacteria: 2.1 Log CFU/mL Probe-type ultrasound (Natural temperature rise during treatment) - Frequency: 19 kHz - Power: 475 W Energy density: 1 kJ/mL - Time: 53 s - Final temperature: 34 °C Energy density: 3 kJ/mL - Time: 2 m 38 s - Final temperature: 55 °C Energy density: 5 kJ/mL - Time: 4 m 23 s - Final temperature: 76 °C Energy density: 7 kJ/mL - Time: 6 m 9 s - Final temperature: 85 °C Probe-type ultrasound (Natural temperature rise during treatment) - Frequency: 24 kHz - Power: 200 W - Time: 15 min Amplitude: 125 μm - Final temperature: 60 °C Amplitude 62.5 μm - Final temperature: 50 °C - Staphylococcus aureus: 3.8 Log CFU/mL - Listeria monocytogenes: 3.6 Log CFU/mL - Lactobacillus plantarum: 4.5 Log CFU/mL - Lactobacillus pentosus: 4.6 Log CFU/mL - Salmonella Typhimurium, Escherichia coli, Pseudomonas fluorescens: 5 Log CFU/mL Probe-type ultrasound (Natural temperature rise during treatment) - Frequency: 20 kHz - Temperature: 40 °C - Final temperature: 59 °C (60 W, 6 min) Energy density: 1,080 kJ/L - Power: 40 W - Time: 9 min - Enterobacter cloacae: 6.31 Log CFU/mL - Staphylococcus epidermidis: 6.07 Log CFU/mL Energy density: 1,080 kJ/L - Power: 60 W - Time: 6 min - Enterobacter cloacae: 6.63 Log CFU/mL - Staphylococcus epidermidis: 6.50 Log CFU/mL Energy density: 1,628 kJ/L - Power: 60 W - Time: 9 min - Enterobacter cloacae: 6.52 Log CFU/mL - Staphylococcus epidermidis: 6.21 Log CFU/mL Continuous flow type ultrasound (Natural temperature rise during treatment) - Frequency: 20 kHz - Power: 150 W - Inlet temperature: 24 °C - Flow rate: 11 mL/min - Time: 102.3 s Output intensity: Level 5 - Final temperature: 51 °C - Pseudomonas fluorescens: 0 Log CFU/mL - Streptococcus thermophilus: 0.5 Log CFU/mL Output intensity: Level 10 - Final temperature: 76.1 °C - Pseudomonas fluorescens: 4.2 Log CFU/mL - Streptococcus thermophilus: 2.7 Log CFU/mL Continuous flow type ultrasound (Thermally assisted ultrasonic treatment) - Dual frequency: 16 & 20 kHz - Power: 1,200 W - Flow rate: 0.45 L/min - Time: 14 min - Total mesophilic bacteria: 0 Log CFU/mL - Total psychrophilic bacteria: 0.5 Log CFU/mL - Total mesophilic bacteria: 3.36 Log CFU/mL - Total psychrophilic bacteria: 4.88 Log CFU/mL Probe-type ultrasound (Thermally assisted ultrasonic treatment) - Power: 150 W - Frequency: 20 kHz - Time: 18 min - Total aerobic bacteria: 2.6 Log CFU/mL - Listeria monocytogenes: 1.93 Log CFU/mL - Escherichia coli O157:H7: 4.63 Log CFU/mL - Total aerobic bacteria: 2.96 Log CFU/mL - Listeria monocytogenes: 2.32 Log CFU/mL - Escherichia coli O157:H7: 5.91 Log CFU/mL Bath type ultrasound (Thermally assisted ultrasonic treatment) - Power: 110 W - Frequency: 40 kHz - Time: 10 min * Thermal conditions were categorized based on information explicitly reported in the original studies. Some studies did not clearly distinguish between non-thermal and thermally assisted ultrasonic effects Several studies have reported microbial inactivation in milk and milk-based beverages using ultrasound treatment alone under room-temperature conditions, where microbial inactivation primarily resulted from cavitation-driven physical and sonochemical effects under controlled non-thermal conditions. Typically, the temperature of milk beverages increases gradually during ultrasonic processing; however, this natural temperature rise can be controlled by employing a cooling system or an ice bath during treatment. In a study applying probe-type ultrasound at room temperature to raw milk, Mudgil et al. (2022) reported that increasing power levels reduced the counts of total aerobic bacteria, Staphylococcus spp., and lactic acid bacteria, with reductions of approximately 4.16, 3.40, and 1.50 Log CFU/mL, respectively, after treatment at 160 W and 20 kHz for 10 min. Chouliara et al. (2010) also found that extending the treatment time enhanced microbial inactivation, resulting in reductions of approximately 2 Log CFU/mL in total viable and psychrotrophic counts after ultrasound treatment (200 W, 24 kHz, without heat) for 16 min. In addition, Crudo et al. (2014), using a continuous flow type ultrasound system, reported that under conditions of 35 kHz for 10 min at room temperature, increasing power further reduced the total aerobic bacteria in raw milk. Moreover, in a study applying 400 W, 20 kHz ultrasound to Kinnow whey beverage, the counts of total aerobic bacteria and molds/yeasts decreased with increasing treatment time, showing reductions of approximately 1.0 and 1.61 Log CFU/mL, respectively, after 15 min of treatment (Siddique et al., 2023). These findings indicate that microbial inactivation under ultrasound processing cannot be interpreted solely as a non-thermal cavitation effect, because thermal contribution during treatment substantially influences antimicrobial efficacy. During ultrasound treatment, the natural temperature rise can be directly utilized to achieve microbial inactivation in milk and milk-based beverages. In such cases, a cooling system is not used in the ultrasound setup. The natural increase in temperature can act synergistically with acoustic cavitation effects, thereby reducing the treatment time required to reach a specific level of microbial reduction or enabling a greater degree of microbial inactivation compared with ultrasound treatment alone, as described earlier. When probe-type ultrasound (19 kHz, 475 W) was applied to raw milk for 158 s, the final temperature reached 54 °C, resulting in reductions of approximately 2.6 Log CFU/mL in total aerobic mesophiles and 2.1 Log CFU/mL in lactic acid bacteria (Scudino et al., 2022). Shamila-Syuhada et al. (2016) reported that when raw milk was treated with ultrasound at 200 W and 24 kHz for 15 min with amplitudes of 62.5 μm and 125 μm, the final temperatures reached 50 °C and 60 °C, respectively. Consequently, microbial reductions of approximately 8 Log CFU/mL were observed at 125 μm for Staphylococcus aureus, L. monocytogenes, Salmonella Typhimurium, E. coli, P. fluorescens, Lactobacillus plantarum, and Lactobacillus pentosus, whereas reductions of about 3.8–5 Log CFU/mL were reported at 62.5 μm. The microbial inactivation effect associated with ultrasound-induced heating has also been demonstrated in whey-based beverages. Guimarães et al. (2018) observed that applying ultrasound to whey beverages at 600 W and 19 kHz for 3 min increased the final temperature to 53 °C, achieving an approximate 2 Log CFU/mL reduction. Scudino et al. (2020) reported that when raw milk was treated with ultrasound at 475 W and 19 kHz, an energy density of 1 kJ/mL resulted in a final temperature of 34 °C with no detectable reduction in total aerobic mesophilic bacteria, whereas 3 kJ/mL increased the final temperature to 55 °C and achieved approximately a 1.8 Log CFU/mL reduction. In contrast, treatment at 5 kJ/mL elevated the final temperature to 76 °C, producing about a 4 Log CFU/mL reduction, and the corresponding processing time was only 4 min 23 s. Monteiro et al. (2018) reported that treatment of chocolate milk at 400 W and 19 kHz with an energy density of 3 kJ/cm³ resulted in an approximately 3.56 Log CFU/mL reduction in total aerobic bacteria, with a final temperature of 42 °C and a processing time of 750 s. However, an excessive temperature rise during ultrasound processing may adversely affect the physical and nutritional properties of the product. Consequently, several studies have focused on integrating ultrasound with mild heat under controlled temperature conditions. In general, the simultaneous application of mild heat and ultrasound has been shown to achieve greater microbial inactivation than ultrasound treatment alone at room temperature, owing to the synergistic interaction between thermal and cavitation effects. Van Hekken et al. (2019) reported that applying continuous flow type ultrasound (1,200 W, dual frequencies of 16 and 20 kHz) to raw milk at 54 °C for 14 min resulted in reductions of 3.36 Log CFU/mL in total mesophilic bacteria and 4.88 Log CFU/mL in total psychrotrophic bacteria, whereas no bactericidal effect was observed when the same ultrasound treatment was conducted at 42 °C. In human milk, ultrasound treatment at 20 °C did not reduce S. aureus; however, the application of bath type ultrasound at 60 °C resulted in an approximately 4.58 Log CFU/mL reduction (Gomes et al., 2021). In summary, microbial inactivation efficiency during ultrasonic pasteurization of milk and milk-based beverages is strongly influenced by ultrasound system configuration, processing conditions, and thermal contribution during treatment. Non-thermal ultrasound, ultrasound with natural temperature rise, and thermally assisted ultrasonic treatments demonstrated distinct microbial inactivation behaviors as well as different practical limitations regarding quality preservation and process efficiency. Therefore, optimization of processing parameters is essential to maintain a balance between microbial inactivation efficiency and quality preservation. Regulations and challenges of milk pasteurization technology Ultrasound has emerged as a promising alternative to conventional pasteurization for milk and milk-based beverages; however, several challenges remain before its full industrial implementation can be achieved. According to the international standard Codex Alimentarius for validating milk pasteurization technologies, milk pasteurization must achieve at least a 5 Log CFU/mL reduction in non-spore-forming pathogenic bacteria. Furthermore, any alternative technology must demonstrate microbial inactivation equivalent to or greater than that achieved by conventional heat pasteurization to be recognized as an acceptable process (Čapla et al., 2022). However, only a few studies have reported reductions exceeding 5 Log CFU/mL using ultrasound alone. Excessive ultrasound intensity required to meet pasteurization standards can negatively affect the sensory and nutritional properties of milk products. Therefore, establishing optimal processing parameters that ensure both microbial safety and product quality is essential to satisfy consumers’ growing demand for fresh flavor and natural quality (Prosekov, 2025). Ultrasound-induced radicals can cause lipid oxidation and generate volatile compounds that result in off-flavors (Guimarães et al., 2019; Riener et al., 2009). Off-flavors such as metallic, burnt, and rubbery notes have been reported in ultrasound-treated milk products (Bui et al., 2020; Juliano et al., 2014). Hence, it is necessary to optimize processing conditions to simultaneously achieve microbial inactivation while minimizing the sensory and nutritional degradation that may occur during ultrasound treatment. Consumer safety concerns, such as the generation and corrosion of metallic materials during ultrasound treatment, must also be addressed. In probe type systems, erosion of the sonotrode surface may produce metallic particles that can react with radicals to form additional chemical compounds (Mawson et al., 2014). Furthermore, stainless steel, which is commonly used in food processing equipment, is susceptible to electrochemical corrosion in the presence of chlorides, and the sonochemical and physical effects of ultrasound can exacerbate this phenomenon (Mawson et al., 2014). In food manufacturing and processing, product contamination with metallic particles or the occurrence of equipment corrosion is strictly prohibited, making the implementation of technical measures to prevent these issues essential. The development of large-scale ultrasound equipment for industrial applications is therefore necessary. Such systems should be capable of achieving microbial inactivation while addressing safety and quality concerns in milk and milk-based beverages. Although ultrasound can be applied relatively uniformly under controlled laboratory conditions, technical limitations arise during scale-up to industrial operations. In particular, probe and bath type systems often deliver insufficient energy intensity or lack uniformity (Kim et al., 2024), making it difficult to ensure consistent product quality and safety. Hence, future research should focus on developing continuous flow systems suitable for large-scale processing and on optimizing ultrasound parameters for milk and milk-based beverage processing. In addition to equipment development, integration of ultrasound with other hurdle technologies may represent an important future direction for industrial implementation. The combination of ultrasound with mild heat, pulsed electric fields, ultraviolet treatment, or natural antimicrobials may enhance microbial inactivation while minimizing adverse effects on product quality (Cho et al. 2026; Rathnayake et al. 2025b). Furthermore, advances in real-time monitoring systems and digital process control technologies could facilitate the development of intelligent ultrasonic pasteurization platforms that maintain both microbial safety and product quality under industrial processing conditions. Future research should also focus on establishing standardized validation protocols, evaluating long-term economic feasibility, and generating regulatory data required for commercial approval in different countries. Conclusion Ultrasound has attracted increasing attention as a promising alternative processing technology to conventional pasteurization in milk and milk-based beverages. Previous studies have demonstrated that microbial inactivation by ultrasound is strongly influenced by ultrasound system configuration, processing conditions, and thermal contribution during treatment. In particular, non-thermal cavitation-driven ultrasound, ultrasound with natural temperature rise, and thermally assisted ultrasonic treatments showed distinct microbial inactivation behaviors and practical limitations. In addition, challenges related to consumer safety, quality preservation, regulatory validation, and large-scale continuous processing systems must be addressed for successful commercialization. Future advances in ultrasound system design and process control may support the development of ultrasound-based pasteurization technologies as next-generation microbial safety strategies for milk and milk-based beverages. Author contributions Yu RN and Park SY wrote the original draft of the manuscript. Rathnayake PY, Yeo SE, Nam C, Cho HU, and Baek KH performed the investigation and prepared all figure and tables. Yong HI wrote, reviewed, and edited the manuscript. All authors reviewed and approved the final version of the manuscript. Funding This review was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (Project No. RS-2025-24523920). Data availability No datasets were generated or analysed during the current study. Declarations Competing interests The authors declare no competing interests. Ethics approval No ethics approval was required for this study. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Rina Yu and Sin-Young Park contributed equally to this work. References Associated Data Data Availability Statement No datasets were generated or analysed during the current study.