Classification and Comparative Analysis of Acoustic Agglomeration Systems for Fine Particle Removal

Abstract

This study presents a systematic classification of acoustic agglomeration systems, developed on the basis of an extensive review of experimental and numerical studies, specifically addressing fine particles. The classification framework encompasses wave type, geometric orientation, level of functional integration, chamber composition, and auxiliary enhancement mechanisms. By organizing the diverse configurations into consistent categories, this study enables a comparative analysis of system performance and suitability for practical applications. This review highlights typical design features, operational ranges, and implementation contexts, while identifying key advantages and limitations of each system type. Strengths such as scalability, compatibility with filtration units, and enhancement of particle capture are contrasted with challenges including acoustic intensity requirements, resonance sensitivity, and integration constraints. The proposed classification serves as a practical tool for guiding future design, optimization, and application of acoustic agglomeration technologies in air pollution control.

1. Introduction

Fine particulate matter (PM2.5) poses a serious threat to human health due to its ability to penetrate deep into the alveolar regions of the lungs and to enter the bloodstream. These particles, typically smaller than 2.5 microns in diameter, bypass natural respiratory filtration mechanisms and accumulate in the body over time. Chronic exposure to PM2.5 has been directly linked to ischemic heart disease, strokes, chronic respiratory conditions, and reduced life expectancy [1,2].

In 2021, the World Health Organization (WHO) updated its Global Air Quality Guidelines, reducing the recommended annual mean concentration for PM2.5 to 5 μg/m³. This revision was based on growing evidence that even low levels of exposure are harmful to health [3]. Despite these recommendations, average PM2.5 concentrations across many European cities remain significantly above the guideline, often ranging from 10 to 14 μg/m³. Even in cities with comparatively clean air, such as Stockholm and Helsinki, recorded values slightly exceed the WHO threshold [4]. The persistence of elevated PM2.5 levels indicates that existing emission control measures are insufficient. This issue directly relates to the United Nations Sustainable Development Goal 3—ensuring healthy lives and promoting well-being—which includes targets for reducing mortality linked to environmental pollution [5].

Fine particles with diameters of 2.5 μm or less are primarily formed during combustion of fossil fuels in transportation, energy production, and industrial processing. Additional contributions come from non-exhaust emissions such as brake and tire abrasion, construction activity, and resuspension of road dust. In industrial environments, high-temperature operations and chemical aerosol formation are dominant sources [6,7]. The resulting particle mixtures exhibit complex size distributions and chemical compositions, which complicate their separation from air streams.

The reduction in airborne particulate matter can be approached through emission control or post-release removal. Emission control strategies include fuel substitution, modernization of industrial processes, and implementation of low-emission technologies. While effective in the long term, these measures require significant investment and are slow to implement on a wide scale [8]. Immediate removal of PM2.5 from the air is typically achieved in situ using mechanical filters, electrostatic precipitators, or sorbent materials. High-efficiency particulate air (HEPA) filters are effective for submicron particles but introduce high flow resistance and require frequent replacement [9]. Electrostatic systems are efficient in dry gas streams but perform poorly in humid conditions or for particles with low electrical mobility. Adsorption and catalytic methods are material-specific and unsuitable for general-purpose air purification. These limitations create demand for alternative or auxiliary technologies that can improve particle capture while reducing operational cost and energy consumption.

Acoustic coagulation is a method for increasing the aerodynamic diameter of suspended particles by exposing them to an acoustic field. The process is based on the interaction between particles and pressure gradients generated by high-frequency sound waves. As a result, particles are displaced toward pressure nodes or antinodes, depending on their density and compressibility. Increased particle concentration in these regions leads to collisions and the formation of agglomerates. Enlarged particles are more efficiently removed by standard filtration or inertial separation methods. The process does not require chemical agents, operates in continuous flow, and can be applied to a wide range of gas–particle systems [10,11].

The phenomenon of acoustic agglomeration was first observed by Wood in 1927 during early experiments on the behavior of particles in sound fields [12]. Patterson and Cawood confirmed that particle clustering occurs in standing-wave fields [13], and Andrade investigated the effect of sound waves on particle motion trajectories [14]. Between 1949 and 1964, Denser, Fuks, and Moore analyzed the influence of physical parameters on coagulation in aerosol and fog systems [15,16,17]. In 1965, Mednikov introduced a complete mechanistic theory that explained the agglomeration process as a result of relative particle motion under acoustic excitation [10]. During the 1970s to 1990s, Volk Jr. developed models combining orthokinetic interaction and Brownian motion, establishing a quantitative basis for process prediction [18]. Chou investigated the effect of turbulence on agglomeration efficiency [19]. Later, Tiwary proposed the wake effect model [20], and Song expanded the analysis by including radiation pressure and local density gradients [21]. Further theoretical refinements were made by González [22,23].

In the past three years, there has been a clear resurgence of research focused specifically on acoustic agglomeration of fine airborne particles in dedicated agglomeration chambers, with 14 studies published during this period.

Despite extensive research, acoustic agglomeration has not yet reached widespread industrial deployment. Barriers include limited energy efficiency, difficulties in upscaling, and challenges in integration with existing air treatment systems. Meanwhile, numerous studies have explored various chamber geometries, wave configurations, and particle–flow interactions, often yielding isolated findings without a coherent framework. Critically, no standardized classification exists that organizes experimental setups based on acoustic field characteristics, particle types, or structural configurations.

To address this gap, the present study conducts an analysis and classification of existing acoustic agglomeration systems. The objective is to identify dominant design types, assess their operational advantages and limitations, and define the most promising directions for further development and application in air pollution control—particularly with respect to enhancing the removal efficiency of fine particulate matter.

2. Physical Principles of Acoustic Agglomeration

Acoustic agglomeration is governed by a set of physical mechanisms that determine particle motion and coagulation under the influence of a sound field. The dominant mechanisms include the orthokinetic interactions, the radiation pressure force, the acoustic wake effect, the viscous interparticle force, and the secondary acoustic streaming phenomenon. The prevalence of each mechanism depends on the acoustic parameters (frequency and SPL), particle properties (size and density), and medium characteristics.

Orthokinetic agglomeration arises due to differential motion between particles of varying sizes in an oscillating fluid. Larger particles exhibit greater inertia, leading to relative velocities and increased collision probability. This mechanism is particularly effective in polydisperse aerosols, where inter-size motion becomes significant. The theoretical basis for this mechanism was formulated by Mednikov [10] and later applied in modeling studies by Dong et al. [24] and Shi et al. [25].

Radiation pressure, or the time-averaged force exerted by a sound wave on a particle, results in net displacement of particles toward pressure nodes or antinodes in a standing wave. Danilov and Mironov developed analytical expressions describing this force [26]. Its role was further analyzed in González et al. [22] and Zu et al. [27], where particle trapping in nodes was shown to facilitate growth.

The acoustic wake effect is associated with disturbances in the fluid medium trailing a particle oscillating in a sound field. These wakes induce attractive forces on nearby particles, aligning them along the wave path. This hydrodynamic interaction was modeled by Tiwary and Reethof [20] and expanded by Dong et al. [24] to confirm its role for coarse-mode particles.

Viscous interparticle forces arise when sound-induced velocity gradients produce shear flows around suspended particles. González-Gómez et al. [22] offered a detailed theoretical analysis of these viscous forces, with numerical reproduction in works such as Shi et al. [25] and Liu et al. [28].

Secondary effects such as acoustic streaming also play a role, particularly in enhancing local turbulence and increasing collision probability. These effects are non-negligible at higher SPLs and can supplement primary mechanisms under specific flow conditions.

2.1. Frequency Dependence

The efficiency of acoustic agglomeration strongly depends on the match between particle size and acoustic frequency. As established in the numerical analysis by Liu et al. [28], particles smaller than 1 μm exhibit more efficient agglomeration at higher frequencies, typically in the ultrasonic range (20–100 kHz), due to enhanced orthokinetic and viscous interactions. Conversely, larger particles (e.g., >2 μm) respond better to lower frequencies (below 5–10 kHz), where inertial and wake-related effects dominate. These conclusions align with findings by Caperan et al. [29] and Shi et al. [25]. Most experimental setups reviewed operate below 20 kHz, matching conventional loudspeaker capability.

2.2. Sound Pressure Level

Sound pressure level (SPL) significantly influences agglomeration strength. According to Dong et al. [24] and Zhou et al. [30], meaningful effects begin above 120–130 dB, with optimal ranges near 140–150 dB. Zu et al. [27] confirmed that growth rates saturate above this range.

2.3. Particle Characteristics and Concentration

The properties of particles strongly influence agglomeration efficiency. Particle size distribution is one of the most critical parameters. Polydisperse systems benefit from orthokinetic and wake-induced collisions, as larger and smaller particles interact dynamically [10,25,28]. For monodisperse aerosols, the efficiency is lower and more sensitive to frequency tuning. Particle shape and phase (solid or liquid) also affect the outcome. Most experimental studies consider spherical, solid particles such as fly ash [27,30,31] or aerosols of liquids like glycol [29]. Particle concentration has a direct effect on collision probability. Increased number density enhances agglomeration rates due to more frequent encounters between particles. Caperan et al. [29] showed a non-linear dependence between concentration and coagulation rate, confirming the importance of optimizing inlet loading. Similarly, Shi et al. [25] included concentration as a parameter in their turbulence-enhanced agglomeration model.

2.4. Influence of Ambient Conditions

Humidity and temperature significantly affect acoustic agglomeration efficiency. Higher humidity enhances interparticle adhesion through capillary forces, leading to more stable agglomerates, while low humidity may hinder bonding. Temperature influences air viscosity and sound propagation, affecting both particle motion and acoustic forces. These effects were noted in experimental studies [31] and confirmed in simulations [28]. Therefore, ambient conditions must be considered in practical applications and system design.

3. Classification of Acoustic Agglomeration Systems

This section introduces a classification of acoustic agglomeration systems. The classification categories were developed based on the analysis of studies involving various types of acoustic chambers, coagulation mechanisms, and system configurations. The division into categories such as functional integration, wave type, or chamber geometry emerged naturally during the structuring of the research, as these parameters directly influence performance and applicability.

Classification of the acoustic agglomeration system is shown in Figure 1.

Many agglomeration systems fall into multiple categories simultaneously. For this reason, the classification was designed as a cross-sectional framework rather than a rigid grouping. This approach captures the full range of investigated designs and provides flexibility for comparative analysis.

3.1. Classification by Functional Integration

3.1.1. Standalone Acoustic Agglomeration

Standalone acoustic agglomeration is applied as an independent particle treatment method in systems where acoustic forces alone are used to enlarge particle size and promote gravitational or inertial separation (Figure 2). Typical setups include cylindrical or rectangular resonant chambers operating in the audible frequency range (1–6 kHz) with SPLs up to 160 dB [29,32,33,34,35,36,37,38,39]. Coagulation efficiency typically ranges from 60% to 85% for particles in the 0.3–5 μm range, depending on the SPL, frequency, and residence time. For example, Sadighzadeh et al. reported a reduction of over 60% in sulfuric acid mist (0.3–2 μm) at 155 dB, while Noorpoor et al. achieved a reduction of up to 80% for fly ash particles between 0.3 and 5 μm [33,34].

3.1.2. Acoustic Pre-Conditioning

Acoustic pre-conditioning uses the acoustic field to increase particle size before they reach the main purification unit, such as a filter, electrostatic precipitator, or scrubber (Figure 1). This approach enhances capture efficiency, especially for fine and ultrafine particles, by forming larger agglomerates that are easier to remove in the subsequent stage [30,40,41,42,43]. Efficiency gains of 20–40% have been reported for particle sizes ranging from 0.1 to 10 μm, particularly in the submicron range. For instance, Gallego-Juárez et al. achieved significant PM₁₀ reduction when coupling a standing-wave chamber with an electrostatic precipitator, while Liu et al. observed a 25–35% improvement in filter efficiency for 0.3–2 μm particles at 130 dB SPL.

3.2. Classification by Acoustic Wave Type

3.2.1. Standing Wave

Standing-wave resonant chambers generate a stationary acoustic field between an emitter and a reflective surface, forming regions of alternating pressure nodes and antinodes. Particles suspended in the medium experience a set of forces that drive motion and agglomeration, with the dominant mechanism being the primary acoustic radiation force [44,45,46].

This force acts on particles with nonzero acoustic contrast and causes net displacement toward either pressure nodes or antinodes, depending on the material properties. For spherical particles much smaller than the wavelength, the axial component of the radiation force is [45]

$$F_x^{rad}=V_{0}k{E}_{ac}\mathrm{\Phi }\sin (2kx),$$

(1)

where V₀ is the particle volume, k the wave number, E_ac the acoustic energy density, and Φ the acoustic contrast factor.

In polydisperse systems, orthokinetic interactions also contribute, as particles of different sizes respond differently to the oscillating fluid field [47]. Larger particles exhibit higher inertia and lag behind, creating relative velocities and increasing collision frequency. This mechanism is particularly effective when small and large particles coexist.

Additional contributions arise from the acoustic wake effect and viscous interparticle forces [35,48]. The former stems from fluid disturbances trailing oscillating particles, inducing alignment and attraction along the wave direction. The latter originates from velocity gradients in the surrounding medium and becomes relevant at high sound pressure levels.

The typical configuration of a standing-wave resonant chamber consists of a rectangular or cylindrical enclosure with one or more high-power transducers mounted on rigid walls to generate stationary pressure waves (Figure 3a). A reflector or closed end creates acoustic resonance by supporting constructive interference of the incident and reflected waves. Particle-laden gas flows horizontally or vertically through the resonant zone, where the standing wave forms discrete pressure nodes and antinodes.

The standing-wave principle is the fundamental basis for most acoustic agglomeration chamber designs and has been adopted by the majority of researchers in experimental and numerical studies [28,29,30,32,33,34,35,36,37,38,41,42,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63].

In standing-wave acoustic agglomeration studies, tested aerosols included sulfuric acid mist, submicron particulate matter, DOP (dioctyl phthalate) droplets, glycol fog, coal fly ash, silica, Arizona dust, and smoke from cable fires. Particle size ranges varied from 0.023 µm to 30 µm. The most common target ranges were 0.3–10 µm [56], 0.4–3 µm [32], and below 2.5 µm (PM2.5) in several tests. Frequencies used ranged from 343 Hz to 1.103 MHz. Reported efficiency varied from 20% to 97.5% depending on the frequency, SPL, and particle type: for example, 97.5% removal of dust at 123 dB [33], up to 83% removal of DOP aerosol at 21 kHz and 140 dB [32], 40–70% improvement in filtration for Arizona dust [43], and up to 85% agglomeration of glycol fog [29].

3.2.2. Traveling Wave

Unlike standing-wave systems, traveling-wave agglomeration chambers utilize a unidirectional propagating acoustic field without fixed nodes or antinodes (Figure 3b). This configuration avoids resonance and instead generates a continuous momentum transfer along the wave vector [46].

The primary acoustic radiation force acting on a small spherical particle in a traveling-wave field can be expressed as [46]

$$F^{rad}=\pi a^{3}k{f}_2^i\left(\tilde{\rho },\delta \right)E_{ac}{e}_z,$$

(2)

where a is the particle radius, k is the wave number, f ⁱ₂ is the imaginary part of the dipole scattering coefficient (dependent on the density ratio and viscous effects), E_ac is the acoustic energy density, and e_z is the direction of wave propagation.

The primary acoustic radiation force in traveling waves is significantly weaker for small particles, especially when their diameter is much smaller than the wavelength [64]. As a result, agglomeration is dominated by secondary mechanisms such as interparticle wake interactions, shear-induced collisions, and turbulent entrainment. These effects become particularly effective at high sound pressure levels and in polydisperse systems where differential response to oscillatory flow enhances relative motion [42]. A typical traveling-wave setup consists of a cylindrical or rectangular flow duct with a horn-type or piston-type transducer at the inlet and an acoustic absorber at the outlet to suppress reflections. Because of the absence of spatial confinement, traveling-wave systems are better suited for continuous high-flow applications and environments where resonance is difficult to maintain. The axial component of the radiation force can still act on particles, but its contribution is limited unless enhanced by coupling with drag or jet effects.

Traveling-wave acoustic agglomeration has been applied to aerosols with particle sizes ranging from 0.03 μm to 10 μm. Zhou et al. reported up to 70% removal efficiency for fly ash particles within this range using frequencies between 1000 and 2400 Hz and an SPL up to 150 dB [65]. Volk and Moroz demonstrated a reduction in the submicron fraction of carbon black aerosols from 96% to 40% at 3 kHz and 120 dB [18]. Liu et al. confirmed coagulation of charged particles in a traveling-wave field at 130 dB, with performance influenced by the initial particle concentration and electrostatic properties [66]. These results highlight that although traveling-wave systems exert weaker acoustic radiation forces on small particles, effective agglomeration can still be achieved through secondary mechanisms under sufficiently high acoustic intensities.

3.3. Classification by Geometric Orientation

3.3.1. Horizontal Chamber

Horizontal chambers represent the most common configuration for acoustic agglomeration systems, especially in industrial and laboratory setups [60,67,68]. Their geometry facilitates integration into ducted gas flows and supports either standing or traveling-wave fields depending on boundary conditions and transducer placement (Figure 4a). The acoustic wave can propagate along the axis of the chamber or perpendicular to the flow, depending on the orientation of the emitter and reflector surfaces [60,69,70,71]. In most cases, rectangular or cylindrical ducts are equipped with side-mounted or bottom-mounted transducers and reflective terminations to establish resonance. Moreover, multiple transducers can be installed in various configurations, including cross-mode operation, where waves intersect at different angles. Such setups enable the formation of complex pressure distributions, potentially enhancing particle convergence and improving agglomeration efficiency beyond that of a single-wave system.

3.3.2. Vertical Chamber

Vertical acoustic agglomeration chambers are less common but offer advantages in gravitational alignment and particle residence time (Figure 4b). In this geometry, the acoustic field typically acts along or across the vertical axis, often facilitating natural sedimentation or enabling longer exposure durations. Notable implementations include Zhou et al. [30], who used a 1.5 m tall cylindrical chamber to agglomerate fly ash (0.03–10 μm) with up to 70% efficiency at 150 dB and 1400 Hz, and Zhang et al. [31], who reported 85–92% efficiency using liquid binders in a vertical setup under moderate power input (2.5–7.5 W). Compared to horizontal systems, vertical chambers often achieve similar or higher efficiencies for fine particles due to improved control over particle trajectories and better phase separation under gravity. However, they are less suited for integration into existing ductwork and typically require vertical space. Horizontal chambers, while more versatile in industrial integration, tend to exhibit slightly lower standalone efficiency but higher flow throughput.

3.3.3. U-Shaped Chamber

U-shaped acoustic chambers are compact systems that fold the flow path to increase residence time while enabling constructive interference of acoustic waves (Figure 4c). These configurations are especially suited for low-flow environments or when integration with compact filter systems is desired. Zhang et al. demonstrated a U-shaped system with side-mounted speakers and steel reflectors, operating in the 836–4183 Hz range at SPLs up to 116 dB, effectively increasing filtration efficiency of submicron NaCl particles (0.25–1.0 μm) from 61% (pre-conditioning only) to 89% when combined with a MERV-6 filter [71]. A more recent study by Zhang et al. applied the same U-shaped principle with embedded TENG structures and achieved 97.5% capture of PM2.5 at 123 dB, using fundamental and harmonic frequencies between 795 and 2385 Hz [70]. Compared to standard horizontal or vertical chambers, U-shaped systems offer enhanced efficiency under lower SPLs and compact design, though flow rates remain limited.

3.4. Classification by Acoustic Chamber Composition

3.4.1. Hollow Resonant Chamber

Hollow resonant chambers refer to setups with no internal porous or filter elements, relying solely on the standing-wave field within an unobstructed volume (Figure 5a). These systems provide a clean acoustic environment and allow detailed study of fundamental agglomeration mechanisms such as radiation forces, wake interactions, and orthokinetic effects. Compared to porous systems, hollow chambers offer better control over wave distribution and reduced attenuation but lack secondary capture mechanisms, making them more suitable for fundamental research or as pretreatment stages.

3.4.2. With Internal Porous Medium

In systems with porous or fibrous structures directly exposed to the acoustic field, the wave enhances particle deposition by concentrating them into specific regions of the medium or improving their interception efficiency (Figure 5b). Gupta and Feke (1997, 1998) demonstrated that polymer foam and packed beds exposed to standing-wave fields at 1.103 MHz significantly improved particle capture, with retention even for particles 100 times smaller than the pore size [72,73]. Barrio-Zhang et al. numerically showed that standing-wave enhancement in porous filter configurations increased particle concentration near fibers and improved collection efficiency up to 1.5× compared to non-acoustic operation, especially under staggered fiber alignment [74]. These systems benefit from internal gradients and structure–wave coupling but may suffer from saturation effects or reduced airflow. They are effective for submicron particles, with a typical SPL of 140–180 dB and frequencies from tens of kHz up to 1.1 MHz.

3.4.3. With Mechanical Resonators

Acoustic agglomeration systems with mechanical resonators incorporate vibrating structural elements (Figure 5c)—such as rods, plates, or membranes—into the acoustic zone to amplify or modulate the sound field [69,75]. These components are tuned to resonate at specific frequencies, enhancing local pressure gradients and particle interaction zones. Unlike hollow chambers, mechanical resonators generate secondary oscillatory effects that promote particle clustering through localized turbulence, wave scattering, and intensified velocity gradients. This configuration allows better control over the spatial distribution of acoustic forces and can improve agglomeration efficiency, especially for fine and polydisperse aerosols. Such systems are well-suited for compact devices or applications requiring passive amplification without increasing power input.

3.5. Classification by Additional Coupled Mechanisms (in the Acoustic Zone)

3.5.1. With Hydrodynamic Enhancement

Acoustic agglomeration systems with hydrodynamic enhancement combine sound fields with additional flow-induced forces to intensify particle collisions. This approach typically employs a central gas jet or controlled turbulence to generate velocity gradients and shear layers within the acoustic zone. These flow structures increase relative motion between particles, boosting orthokinetic interactions and promoting agglomeration beyond what the acoustic field alone can achieve. In the study by Sun et al., a horizontal cylindrical chamber was equipped with a coaxial gas jet interacting with a standing-wave field [76]. The system demonstrated effective growth of inhalable particles (0.43–10 μm), with the mass median diameter increasing from 2.47 to 6.48 μm at an optimal frequency of 1416 Hz and an SPL of 128 dB. This confirms that coupling acoustic and hydrodynamic fields offers a promising strategy for enhancing fine particle treatment in high-flow conditions.

3.5.2. With Injected Binding Agents

Acoustic agglomeration systems with injected binding agents enhance particle coagulation by introducing droplets of water, vapor, or chemical solutions into the acoustic field [77,78,79]. These agents promote interparticle adhesion via mechanisms such as capillary bridging, liquid-film formation, or surface tension effects [80,81,82]. The acoustic field increases collision frequency, while the presence of droplets facilitates the formation of stable agglomerates [83]. Importantly, the injection of binding agents also increases the overall particle concentration in the active zone, which further amplifies the probability of collisions and accelerates the agglomeration process.

Zhang et al. used XTG, KC, and PFS additives in a vertical standing-wave system and achieved 85–92% agglomeration efficiency for 0.15–9.5 μm particles at 1400 Hz and 143 dB [77]. In a numerical study, Zhao et al. showed that increasing the concentration of droplets from 1.4 × 10⁴ to 2.4 × 10⁴ cm⁻³ raised efficiency from 51.6% to 85.6% [78]. Yan et al. (2016) showed that injecting seed droplets of a wetting agent into the acoustic field increased the removal efficiency of coal-fired PM2.5 particles by 18–25%, especially at higher SPLs up to 157 dB [83].

Depending on the applied method of enhancement, including the use of binding agents, liquid droplets, or modified acoustic fields, agglomerates may form with diverse morphologies and structural properties. The final shape and size of the agglomerates are influenced by multiple factors: the initial particle characteristics (shape, surface roughness, and hygroscopicity), acoustic parameters such as sound pressure level and frequency, the presence of humidity or condensable vapors, and the overall particle concentration.

Typical features include elongated chains, porous clusters, or compact spherical bodies. The use of liquid binders or wetting agents often promotes the formation of more stable and cohesive structures by enhancing surface adhesion (Figure 6).

3.5.3. With Electromagnetic or Electrostatic Assistance

Acoustic agglomeration systems with electromagnetic or electrostatic assistance rely on the combined action of acoustic radiation forces and electric fields to enhance particle clustering and removal. The acoustic field induces oscillatory motion and drives particles toward pressure nodes or antinodes, while the electric field introduces additional forces—such as electrophoretic drift or charge polarization—that increase collision probability and support the transport of agglomerates to collection electrodes. This coupling is particularly useful for submicron particles that exhibit low inertial response to acoustic forces alone. In the study by He et al., a vertical rectangular chamber was equipped with both a standing-wave acoustic field (800–2400 Hz, up to 143 dB) and a pulsed corona discharge (35–55 kV, 100–300 Hz) [84]. The combined effect significantly increased the size of coal-derived particles (0.15–200 μm) and improved their removal by an integrated electrostatic precipitator, demonstrating the potential of hybrid fields for efficient fine particle control.

3.5.4. With Turbulent Flow Enhancement

This category includes systems where acoustic agglomeration is enhanced by introducing turbulent flow structures, such as gas jets or mechanically induced vortices. Turbulence modifies the local particle distribution, creating inhomogeneities and increasing collision probability [25,85,86]. In such environments, the combination of inertial clustering and acoustic radiation forces leads to significantly improved agglomeration rates, particularly for submicron particles. The synergistic effect is most pronounced when the turbulence intensity is sufficient to induce relative motion between particles without fully disrupting acoustic ordering.

Wang demonstrated through laboratory-scale PIV analysis that turbulent microstructures increase local particle concentration and improve acoustic interaction outcomes [85]. Malherbe et al. also confirmed that even acoustically induced turbulence can enhance particle collisions and cluster formation [42].

3.5.5. With Swirl Flow

This group includes systems where vortex motion is intentionally introduced into the acoustic field to enhance particle interactions. Unlike random turbulence, swirl flow structures create organized rotational movement of gas-dispersed particles, increasing their residence time and local concentration within specific zones of the standing-wave field [87,88]. The use of swirl flows supports particle clustering not only in nodal planes but also across low-pressure regions, promoting additional interaction pathways between particles.

The key mechanism is the superposition of vortex motion and acoustic streaming, which leads to enhanced agglomeration through three effects: increased mutual particle motion, prolonged interaction time in confined vortex zones, and forced particle transport across nodal areas. These flows can be initiated by geometrical configuration (e.g., tangential inlets) or by designing ultrasonic transducers that produce inhomogeneous fields with alternating oscillation phases.

A notable implementation is described by Khmelev et al., who developed a resonance-type agglomeration chamber with a bending-mode disk radiator generating inhomogeneous ultrasonic fields. At a 165 dB sound pressure and a gas flow rate of 6.2 m³/h, the system achieved particle collection efficiencies of 95% for PM2.5, 92% for PM1.5, and 85% for PM0.5 [87]. The study demonstrated that vortex formation due to acoustic gradients significantly outperformed homogeneous field conditions. The chamber design also allows for scaling via multiple disk radiators installed in parallel.

3.6. Classification by Acoustic Frequency Range

3.6.1. Audible Range

Agglomeration systems operating in the audible range, typically 500 Hz to 20 kHz, are most effective for coarse particles, typically above 2–3 μm (Figure 7). At these frequencies, larger particles retain high entrainment and respond strongly to the acoustic field, enabling efficient radiation force action and orthokinetic collisions. Most large-scale or pilot systems use this range due to its deeper penetration and lower energy absorption [29,30]. Audible-range excitation is also better suited for integration in industrial ductwork and high-flow applications, where strong particle inertia aids in size growth and separation.

3.6.2. Ultrasound

In contrast, ultrasound-based systems, typically 21–1000 kHz, are suited for compact designs and precise manipulation of submicron particles. They generate stronger gradients and allow for localized pressure control but suffer from higher energy absorption [72,74]. The choice of frequency depends on particle size, system geometry, and target application: audible waves favor bulk-flow agglomeration, while ultrasound enables integration in confined or high-precision environments.

4. Discussion

Among all categories, systems employing acoustic pre-conditioning with porous media demonstrate the highest potential. These systems not only enhance particle concentration upstream of filtration elements but also reduce energy consumption and extend filter lifespan.

Standalone agglomeration chambers remain effective in laboratory-scale demonstrations, particularly for rapid removal or visualization studies. However, their integration into real-world flows remains limited due to space, noise, and flow continuity constraints.

Regarding chamber composition, hollow resonant chambers are the most versatile and scalable, allowing fine control over wave modes and SPLs. Meanwhile, systems that couple acoustics with injected binding agents or vapor condensation exhibit enhanced performance for submicron particles, though they often require additional maintenance or chemical handling.

A notable distinction exists between audible and ultrasonic frequency ranges. Audible-range systems (≤20 kHz) are easier to implement and control, especially in large chambers, but often require higher SPLs to achieve strong agglomeration. Ultrasonic systems (>20 kHz), by contrast, are more effective for targeting submicron particles and allow more compact designs, though they demand precise tuning and often involve higher equipment costs.

It should be noted that a wide variety of classification combinations exists across the reviewed studies, and these do not always overlap consistently for a given particle size range. This indicates that multiple structural and operational configurations can achieve similar agglomeration efficiency, depending on the dominant mechanisms involved. The most effective combinations of acoustic system configurations for different particle size ranges are summarized in

Table 1. Effective acoustic system configurations by particle size range.

Particle Size Range	Most Effective Construction
≤0.5 µm	Vertical chamber + standing wave + porous medium Vertical chamber + standing wave + injected droplets Horizontal chamber + ultrasound + hollow resonator Vertical chamber + ultrasound + long exposure
0.5–1 µm	Horizontal chamber + standing wave + hollow chamber Vertical chamber + standing wave + aerosol with binders Vertical chamber + porous medium + acoustic pre-conditioning Vertical chamber + ultrasound + seed particles
1–3 µm	Horizontal chamber + traveling wave + no filter Vertical chamber + standing wave + orthokinetic mechanism Horizontal chamber + standing wave + turbulence
3–10 µm	Horizontal chamber + traveling wave + dry flow Vertical chamber + standing wave + large binder particles Horizontal chamber + low-frequency acoustic field

.

The main advantages, limitations, and typical application areas of each system type are summarized in

Table 2. Classification-based evaluation of acoustic agglomeration systems.

Subcategory	Advantages	Limitations	Typical Use Cases
Classification by functional integration
Standalone acoustic agglomeration	Operates independently without filtration; flexible for lab and pilot-scale systems.	Requires high acoustic intensity; limited by agglomerate removal efficiency.	Aerosol coagulation studies; dry gas purification without filter elements.
Acoustic pre-conditioning	Enhances performance of downstream filters; reduces particle load and extends filter life.	Cannot operate as standalone system; dependent on integration with filters.	HVAC systems; ESP pretreatment; multistage filtration lines.
Classification by acoustic wave type
Standing wave	Creates stable trapping zones for particles; high control of spatial distribution.	Requires precise geometry and alignment; sensitive to flow disturbances.	Laboratory setups; stationary ducts; resonance-based agglomerators.
Traveling wave	Supports continuous displacement and agglomeration in flow direction.	Lower trapping force compared to standing waves; needs careful damping.	Moving gas streams; inline reactors; flow-through coagulation chambers.
Classification by geometric orientation
Horizontal chamber	Easier integration into ducts and industrial lines; supports wide flow regimes.	May require flow straighteners or diffusers for wave stability.	Industrial exhaust; ESP inlets; dry gas cleaning.
Vertical chamber	Supports gravitational sedimentation; compact footprint.	Flow rate must be controlled to avoid particle re-entrainment.	Pilot-scale agglomerators; stack-mounted treatment units.
U-shaped chamber	Increases residence time via recirculation; fits compact geometries.	More complex wave tuning and interaction at bends.	Portable purifiers; integrated pretreatment modules.
Classification by acoustic chamber composition
Hollow resonant chamber	Simplifies construction and modeling; maintains acoustic quality.	Limited field modulation options; no interaction with physical media.	Base setups for acoustic chambers.
With internal porous medium	Supports capture and deposition after agglomeration; enhances flow resistance control.	Risk of clogging; acoustic field distortion near obstacles.	Hybrid acoustic–filter units; airflow filtration elements.
With mechanical resonators	Improves field uniformity and resonance targeting; enhances trapping precision.	Requires custom tuning; structurally complex.	Compact and high-precision agglomeration systems.
Classification by additional coupled mechanisms (in the acoustic zone)
With hydrodynamic enhancement	Jet mixing increases collisions and agglomeration efficiency.	Sensitive to jet velocity and orientation.	Dust control in ducts with active flow injection.
With injected binding agents	Liquid bridges improve capture rate for fine particles.	Requires droplet control; risk of residue or coalescence.	Fire smoke, vapor co-removal; hybrid agglomeration systems.
With electromagnetic or electrostatic assistance	Synergistic interaction enhances agglomeration; supports lower SPLs.	Integration complexity; field interference risks.	Electroacoustic collectors; fine dust pretreatment.
With turbulent flow enhancement	Enhances particle collision via inertial clustering; effective for submicron particles in dynamic flows.	Less control over particle paths; may disrupt standing-wave structure	Exhaust treatment with acoustic nozzles; aerosol agglomeration in gas jets
With swirl flow	Promotes particle accumulation in vortex zones; enables efficient agglomeration at low concentrations.	Complex chamber geometry; swirl may entrain particles too strongly	Compact and high-precision agglomeration systems; fine aerosol agglomeration.
Classification by acoustic frequency range
Audible range (≤20 kHz)	Lower equipment cost; deeper penetration in dense aerosols.	Less effective on submicron particles; audible noise generation.	Coarse and fine aerosol agglomeration; low-frequency acoustic research.
Ultrasound (>20 kHz)	Effective for ultrafine particles; compact transducer designs.	Higher energy demand; attenuation in large volumes.	Fine aerosol agglomeration.

. A complete list of system parameters and the classification of reviewed studies is provided in

Table A1. Acoustic agglomeration system parameters and classification.

Reference	Particle Type	Particle Size and Distribution	Frequency (kHz)	Intensity (dB)	Results	Classification
Lai et al. [57]	–	Polydisperse, 10 µm	6.858, 10.287, 20.574	70, 160	Effective agglomeration of PM10 at ~10 kHz, 160 dB, and flow 0.1–0.3 m/s; potential to reduce energy use and pressure drop in ventilation systems	Acoustic pre-conditioning; standing wave; horizontal chamber; hollow resonant chamber
Zhang et al. [71]	NaCl, balsa wood dust (preliminary)	Polydisperse, 0.25–10 µm	0.836, 1.637, 2.510, 3.346, 4.183	108–116	Prefiltration efficiency for 1.0 µm particles reached 61% without filter and 89% with MERV-6; system significantly improves capture of submicron particles	Acoustic pre-conditioning; standing wave; U-shaped chamber; hollow resonant chamber
Caperan et al. [29]	Glycol aerosol	Polydisperse, 0.5–1.5 µm, lognormal, geometric mean 0.8 µm	10, 21	100–200 W (electric power)	Initial agglomeration rate increases proportionally to square of acoustic velocity amplitude; highest efficiency achieved with standing-wave field	Standalone acoustic agglomeration; traveling wave; horizontal chamber; hollow resonant chamber
Caperan et al. [49]	Fly ash	Polydisperse, dae ≈ 1.6 µm	21	200 W (electric power)	Agglomeration rate K0 = 2.5 × 10−2 s−1, 3.3 times higher than Brownian agglomeration rate; fine fraction efficiently removed by acoustic field	Standalone acoustic agglomeration; standing wave; hollow resonant chamber
Zhang et al. [70]	Ambient aerosol	Polydisperse, 0.3–3.0 µm	0.795, 1.590, 2.385	104–123	Filtration efficiency for PM2.5 increased with SPL, reaching up to 97.5% at 123 dB; highest efficiency observed under low airflow conditions	Acoustic pre-conditioning; standing wave; U-shaped chamber; with electromagnetic or electrostatic assistance
Gupta et al. [72]	Polystyrene particles	Polydisperse, 4–30 µm, spherical	1103	20, 30 W (electric power)	Particle retention up to 60% during early phase, 35% at saturation, and 70–80% in aluminum mesh under optimal conditions	Standalone acoustic agglomeration; standing wave; horizontal chamber; with internal porous medium
Barrio-Zhang et al. [74]	Polystyrene	Monodisperse, 500 nm	343	180	Capture efficiency improved by up to 123 times in staggered configuration at low flow rate under 343 kHz and 180 dB	Acoustic pre-conditioning; standing wave; horizontal chamber; with internal porous medium
Kilikevičienė et al. [61]	Silica test dust (ISO 12103-1 A1)	Polydisperse, 0.3–10 µm	0.5–3.0	129–135	Reduction in particles <1 µm: 11.4–79.0%; increase in >2 µm particles: 14.3–331.7% depending on frequency	Acoustic pre-conditioning; horizontal chamber; standing wave
Zhang et al. [79]	Arizona dust, water droplets	Polydisperse, 0.006–10 µm (dust); D10 = 1.4 µm and D90 = 5.4 µm (droplets)	21	133.4, 135.7, 136.9	PM concentration reduced to 30% of original value with AA+DA; filtration efficiency increased by up to 57% compared to baseline	Acoustic pre-conditioning; standing wave; horizontal chamber; hollow resonant chamber; with injected binding agents
Zhao et al. [78]	Solid particles and liquid droplets	Not specified	1–26	140, 148	Agglomeration efficiency increased from 51.6% to 77.1% (140→148 dB) at 1.4 × 104 drops/cm3; from 68.2% to 85.6% at 2.4 × 104 drops/cm3	Acoustic pre-conditioning; standing wave; horizontal chamber; hollow resonant chamber; with injected binding agents
Sadighzadeh et al. [33]	Sulfuric acid mist	Polydisperse, 0.4–20 µm	0.132, 0.245, 0.458, 0.852, 1.410, 3.530, 7.150	115, 135, 155, 165	Removal efficiency reached 86% at 852 Hz and 165 dB; increased efficiency with higher inlet concentration and lower airflow	Standalone acoustic agglomeration; standing wave; horizontal chamber; hollow resonant chamber
Garbarienė et al. [58]	Diesel exhaust aerosol	Polydisperse, 10–470 nm	21.3	144.1	Particle concentration (10–70 nm) reduced by 21.7%, 100–180 nm by 8.2%; overall 11.0% reduction for NExBTL100 without EGR	Acoustic pre-conditioning; standing wave; horizontal chamber; hollow resonant chamber
Ng et al. [42]	Arizona dust	Polydisperse, 0.25–32 µm	6.4	140	Filtration efficiency of MERV 11 increased from 73% to 83%, and MERV 13 from 88% to 92%; PM0.4–0.5 concentration reduced by 16%	Acoustic pre-conditioning; standing wave; horizontal chamber; with internal porous medium
Gallego Juárez et al. [50]	Coal combustion aerosol	Polydisperse, 0.005–30 µm	10, 20	145–165	Mass removal efficiency up to 37%; submicron number reduction up to 40%; better results at 20 kHz	Acoustic pre-conditioning; standing wave; horizontal chamber; hollow resonant chamber; with electrostatic assistance
Sun et al. [76]	Fly ash	Polydisperse, 0.43–10 µm	0.28–3.5	100–130	Maximum agglomeration efficiency of 36.9% achieved at 1416 Hz, 128 dB, and 22.5 m/s jet velocity; mass median diameter increased from 2.47 µm to 6.48 µm	Standalone acoustic agglomeration; standing wave; horizontal chamber; hollow resonant chamber; with hydrodynamic enhancement
Liu et al. [63]	Organic waste combustion PM	Polydisperse, 0.009–2.5 µm	21	150	Filtration efficiency increased up to 22.4% at 30 °C; higher temperatures reduced efficiency; biomass > MSW > polypropylene	Acoustic pre-conditioning; standing wave; vertical chamber; hollow resonant chamber; with injected binding agents
Moldavsky et al. [52]	Arizona fine dust	Polydisperse, 0.19–>5 µm	0.05, 0.1, 0.3, 0.5, 1.0	90–130	Filter lifespan doubled at 300 Hz and 125 dB; extended 5–10× at 50–100 Hz and 125–130 dB; no reduction in filtration efficiency, reduced pressure drop	Acoustic pre-conditioning; standing wave; horizontal chamber; hollow resonant chamber
Amiri et al. [32]	Dioctyl phthalate (D.O.P.) aerosol	Polydisperse, 0.26–2 µm	0.204, 0.550, 0.650, 0.749	140, 150, 155, 162	Coagulation efficiency reached 83% at 749 Hz and 162 dB; effectiveness threshold observed at ≥155 dB	Acoustic pre-conditioning; standing wave; vertical chamber; hollow resonant chamber
Zhou et al. [30]	Fly ash	Polydisperse, 0.3–10 µm	1.2, 1.4, 2.0, 2.2, 2.4	130–148	Filtration efficiency of bag filter increased from 91.29% to 99.19%; ESP efficiency increased from 89.05% to 99.28% at 1400 Hz, 148 dB	Acoustic pre-conditioning; traveling wave; vertical chamber; hollow resonant chamber; with mechanical filters
Gupta & Feke [73]	Polystyrene	Polydisperse, 2–30 µm	1103	50 W (electric power)	Filtration efficiency reached 80–90% after 1–2 min; optimal power 20–40 W; too high power (50 W) reduced efficiency due to acoustic streaming	Acoustic pre-conditioning; standing wave; rectangular chamber; with internal porous medium
Liu et al. [53]	Fly ash	Polydisperse, 0.03–10 µm	0.2, 1.0, 1.4, 1.6, 1.8, 20	150	Efficiency of 75.3% achieved at 1400 Hz, 148 dB, 5 s; high-frequency mode (20 kHz) reduced 0.2–0.25 µm particles by 10.38%	Standalone acoustic agglomeration; traveling wave; vertical chamber; hollow resonant chamber
Zhang et al. [77]	Fly ash	Polydisperse, 0.15–9.5 µm	1.4	135, 141, 143	Filtration efficiency increased from 63% (baseline) to 91% with XTG binder at 7.5 W	Acoustic pre-conditioning; traveling wave; vertical chamber; hollow resonant chamber; with injected binding agents
Noorpoor et al. [34]	D.O.P. droplets	Monodisperse and polydisperse, 260–3000 nm	0.2, 0.65, 0.83	120–145	Removal efficiency of 260 nm particles reached 93.35% at 830 Hz, 145 dB, and 10 L/min; dropped to 43.2% at 50 L/min	Standalone acoustic agglomeration; standing-wave field; vertical chamber; hollow resonant chamber
González et al. [35]	Glass spheres	Monodisperse, 8.0 ± 0.9 µm	0.2–5	–	Collisions observed only for qp < 0.5; 100% interaction for qp = 0.05–0.15	Standalone acoustic agglomeration; standing wave; horizontal chamber; hollow resonant chamber
Lu et al. [36]	Fly ash	Polydisperse, 10–100 μm	2.0	158	Observed both agglomeration and breakup of particles; stable agglomerates for size ratio a = 0.73–1.37 at 10 μm	Standalone acoustic agglomeration; standing-wave field; horizontal chamber; hollow resonant chamber; without additional mechanisms
Gallego-Juárez et al. [41]	Fly ash	Polydisperse, <0.005–30 μm	10, 20	165	Mass concentration reduced up to 37%; number of particles reduced up to 40%, best at 20 kHz and SPL > 160 dB	Acoustic pre-conditioning; standing-wave field; horizontal chamber; hollow resonant chamber; with electrostatic assistance
Zhou et al. [65]	Fly ash	Polydisperse, 0.03–10 µm	1.0, 1.2, 1.4, 1.6, 1.8, 2.0, 2.4	133–144	At 1400 Hz, 142 dB: particle concentration reduced to 35% of initial; higher SPL and concentration increase efficiency	Acoustic pre-conditioning; vertical chamber; traveling-wave field; hollow resonant chamber
Yan et al. [80]	Fly ash	Polydisperse, 0.023–9.314 µm	2	120–157	Particle concentration reduced by 56.7% at 150 dB with SDS droplets; 73.3% efficiency at droplet conc. of 2 × 104 cm−3	Acoustic pre-conditioning; standing-wave field; vertical chamber; with injected binding agents
Volk & Moroz [18]	Carbon black	Polydisperse, 0.01–1.0 µm	1.3, 2, 3, 4, 5, 6	100, 110, 120	Fraction of particles < 1 µm reduced from 96% to 40% at 3 kHz, 120 dB	Standalone acoustic agglomeration; traveling-wave field; horizontal chamber; hollow resonant chamber
Liu et al. [43]	Arizona dust	Polydisperse, 0.1–10 µm	21	133, 136	Filtration pressure drop reduced by 67.3%; filtration efficiency increased; filter lifetime extended by more than 50%	Acoustic pre-conditioning; standing-wave field; vertical chamber; hollow resonant chamber; with injected binding agents
Yan et al. [83]	Carbonaceous dust and silico-aluminate minerals	Polydisperse; 0.023–9.314 µm	1.8	120, 130, 140, 150, 153, 158	Efficiency increased from 10–23% (acoustic only) to 53–80% with vapor condensation at SPL = 150 dB, S = 1.2	Acoustic pre-conditioning; standing-wave field; vertical chamber; hollow resonant chamber; with injected binding agents
Liu et al. [66]	Smoke particles	Polydisperse, 0.2–0.5 µm (exp); 0.01–1 µm (sim)	0.8, 1.0, 1.2, 1.4, 1.6	131, 133, 135 (exp); 160 (sim)	Coagulation efficiency increased from 28.6% (acoustic only) and 50.6% (electric only) to 62% under combined field	Standalone acoustic agglomeration; with electromagnetic or electrostatic assistance; standing-wave field; traveling-wave field; horizontal chamber; hollow resonant chamber
He et al. [84]	Fly ash	Polydisperse; 0.15–200 μm	0.8, 1.2, 1.6, 2.0, 2.4	135, 139, 143	Removal efficiency increased from 80.7% (ESP only) to 98.3% (combined field); D50 increased from 23.39 μm to 73.28 μm	Acoustic pre-conditioning; standing-wave field; vertical chamber; hollow resonant chamber; with electromagnetic or electrostatic assistance
Yuan et al. [56]	Cable fire smoke	Polydisperse, 0.3–10 µm	1.5, 2, 4, 6	141–145	Transmittance increased from 10% to 60% in 11 s at 1.5 kHz, 12 W; full smoke removal in ~180 s	Standalone acoustic agglomeration; standing-wave field; horizontal chamber; hollow resonant chamber
Kilikevičienė et al. [60]	Diesel exhaust particles	Polydisperse, 0.3–10 µm	21.4, 33.8	139–141	For D100: 44.3% reduction at 0.3 µm, 92.45% at 10 µm; for ROMEP: 6.0% reduction at 0.3 µm, 87.4% at 2 µm	Acoustic pre-conditioning; standing-wave field; horizontal chamber; hollow resonant chamber; with injected binding agents
Hoffmann et al. [91]	Fly ash; limestone; flue gas particles	Polydisperse; submicron to tens of microns	0.044, 0.64–1.07, 1, 10, 19.6, 20, 21	140–165	Particle mass shifted to >11 µm (23%) at 44 Hz; up to 70% reduction in micron particles at 10 kHz, 140–165 dB	Standalone acoustic agglomeration; standing-wave field; traveling-wave field; horizontal chamber; hollow resonant chamber
Mao et al. [82]	Cable fire smoke; water droplets	Polydisperse, 0.3–1 µm (smoke); 5–20 µm (droplets)	1–6	137, 151	Visibility threshold reached 30 s faster with droplets; same efficiency at 137 dB with droplets as 151 dB without	Standalone acoustic agglomeration; standing wave; vertical chamber; hollow resonant chamber; with injected binding agents
Komarov et al. [51]	Zn condensate	Polydisperse; 0.1–80 µm	0.098, 0.210, 0.359, 0.459, 0.645, 0.991	142–160	Particle size increased by 50%; number concentration decreased by 60%; mass concentration decreased by 40% at 160 dB	Standalone acoustic agglomeration; standing-wave field; vertical chamber; hollow resonant chamber
Guo et al. [81]	Water droplets, glycerin mist, cable fire smoke	Polydisperse; 0.3–25 µm	15	143–150	Transparency increased to 60% in 4–8 s depending on aerosol type	Standalone acoustic agglomeration; standing-wave field; vertical chamber; hollow resonant chamber
Sadighzadeh et al. [54]	Sulfuric acid mist	Polydisperse; 0.4–20 µm	0.852	115, 135, 155, 165	Removal efficiency increased from 22.75% to 78.69% with SPL increase	Acoustic pre-conditioning; standing-wave field; horizontal chamber; hollow resonant chamber
Knoop et al. [48]	Glass spheres	Polydisperse, 58–242 µm (primary), 1–3 mm (agglomerates)	20	—	Agglomerates fragmented under high-intensity ultrasound; fragmentation depends on primary size, humidity, and hydrophobicity	Standalone acoustic agglomeration; standing-wave field; vertical chamber; hollow resonant chamber
Li et al. [68]	Coal-fired ash	Polydisperse, 0.03–9.97 µm	1.0–3.0	111–141	Removal efficiency reached 70% at 1500 Hz, 141 dB, and 560 L/h cooling water flow	Acoustic pre-conditioning; standing-wave field; vertical chamber; with internal porous medium; with injected binding agents
Zhao et al. [67]	Coal dust	Polydisperse, 0.6–110 µm	0.1–2.2	80–120	Total dust concentration reduced from 18.25 to 9.5 mg/m3; sedimentation time reduced by 33.64%; respirable dust removal efficiency increased by 21.93% at 1300 Hz and 120 dB	Acoustic pre-conditioning; standing-wave field; horizontal chamber; hollow resonant chamber; with injected binding agents
Yang et al. [59]	Oil droplets	Polydisperse, 20–60% size increase depending on frequency	1.3	not specified	Drop diameter increased by ≈60% at 1300 Hz; aggregation enhanced at standing-wave antinodes	Acoustic pre-conditioning; standing-wave field; vertical chamber; hollow resonant chamber; with injected binding agents

in Appendix A.

A substantial number of CFD simulation studies were reviewed, and their findings support the logic of the proposed classification framework [51,57,71,74,89]. These studies typically focus not on entire systems, but on specific physical phenomena or isolated components within the setup, due to the high computational demands of full-scale chamber modeling. Nevertheless, the simulation results consistently align with the classification, particularly in terms of wave behavior, particle dynamics, and the influence of structural features.

Scalability and Integration Issues

Several acoustic agglomeration configurations demonstrate strong potential for upscaling and integration into HVAC systems or industrial air treatment units. Horizontal resonant chambers with standing-wave fields have already been tested at pilot scale with high flow rates and are compatible with duct geometries. U-shaped channels, as proposed in recent designs, offer compactness and minimal flow disruption, making them suitable for retrofitting. Systems based on acoustic pre-conditioning—where the sound field enhances particle capture before filtration—are especially promising, as they can be added to existing filters without modifying the core system design.

Integration of acoustic agglomeration systems into real-world environments faces several challenges related to scaling and practical implementation. Maintaining stable resonance conditions under variable airflow rates is technically demanding, particularly in dynamic ventilation scenarios. Additionally, high-intensity acoustic fields (SPL ≥ 140–165 dB), essential for effective agglomeration, pose significant challenges regarding energy consumption, noise management, and sound isolation requirements.

The presence of additional acoustic elements—such as emitters or internal resonators—may increase pressure drop or interact adversely with certain filter materials sensitive to vibration or temperature variations. Careful optimization of chamber geometry, localized acoustic fields, narrowband excitation, and integration with passive noise control methods (absorptive linings and Helmholtz resonators) can help mitigate these drawbacks. For submicron particles, employing high-frequency ultrasound offers the advantage of strong attenuation near the treatment area, effectively limiting noise propagation.

Another critical barrier to widespread adoption is the absence of a standardized evaluation methodology that balances particle removal efficiency with energy-related metrics. Current indicators, such as the quality factor (QF) [90], used for filtration systems, do not fully reflect acoustic system specifics, where pressure drop might remain minimal while significant acoustic power is required. Developing a universally accepted performance metric, considering both filtration effectiveness and overall energy input, remains essential for fair benchmarking and informed decisions on practical implementation.

5. Conclusions

This study presents a comprehensive review and systematic classification of acoustic agglomeration systems, consolidating current research into a unified analytical framework. The proposed categorization—based on wave type, geometric orientation, functional integration, chamber composition, and auxiliary mechanisms—facilitates consistent comparison across experimental and modeling studies.

The analysis reveals that certain configurations, such as standing-wave horizontal chambers and U-shaped channels, offer high potential for integration into industrial air treatment and HVAC systems due to their compatibility with flow geometry and scalability. Acoustic pre-conditioning units, positioned upstream of filtration media, demonstrate particular promise in enhancing particle capture efficiency without requiring structural modifications to the filter itself.

However, the practical deployment of such systems is constrained by several technical factors, including sensitivity to airflow variability, high-intensity acoustic field requirements, and potential resonance instability. Limitations related to pressure drop, spatial constraints, and filter material response must also be addressed.

This work offers a structured and comparative reference for the development, selection, and optimization of acoustic agglomeration systems across various application contexts. Special emphasis is placed on improving the removal efficiency of fine particulate matter in air pollution control.