Experimental and CFD investigation of the dynamic characteristics of particle clusters under cavitation bubble near a rigid wall Authors: Xiaobo Shen, Wei Han, Rennian Li, Shiqi Yang, Qingduo Meng Published: 2026 ABSTRACT The motion of particles under the influence of cavitation bubbles is a nonlinear process. This study investigates the dynamic behavior of particle clusters in particulate flows subjected to cavitation bubble activity near a rigid wall. With experiments and numerical simulations, the effects of cavitation bubbles of varying sizes on particle clusters located at different distances from the rigid boundary are systematically investigated and analyzed. Results demonstrate that, during the bubble expansion phase, the maximum radius Rmax reached by the bubble determines the peak velocity at which the particle cluster is pushed away from the bubble center, particles with greater mass exhibit weaker dynamic responses to the bubble’s expansion and collapse near the rigid wall. The VOF multiphase flow model, coupled with the DPM particle tracking method, produces simulation results that closely align with the experimental findings. This approach proves valuable for investigating particle motion in experimental setups where the presence of gas clusters may obstruct direct observation of particle behavior. The impact intensity of the particle group on the wall, resulting from the cavitation-induced jet, is primarily governed by the velocity at which the jet, generated by the collapse of the cavitation bubble, strikes the rigid wall. When the characteristic parameter λ < 1, which represents the ratio of the cavitation center position to the bubble radius, and the bubble size is relatively small, particles are readily accelerated to between 1/4 and 3/4 of the jet impact velocity, and the energy transfer and dissipation of cavitation bubbles during their evolution in particle clusters are discussed. This particular condition is the dominant factor responsible for wall damage in such cases. Introduction Cavitation is a prevalent phenomenon in fluid dynamics. In the presence of sediment or particles, the flow evolves into a complex multiphase flow, encompassing gas, solid, and liquid phases. Such phenomena are commonly encountered in hydraulic machinery, separators, oil–gas separation systems, and pipeline networks [1]. In these complex flow conditions, the interaction between cavitation bubbles and particles within the fluid medium results in both cavitation erosion [2], [3] and particle induced erosion on the wall surface [4], [5], [6], [7], [8], [9]. The mechanisms underlying the interaction between bubbles and particles are pivotal to advancing the predictive modeling of wall surface damage in such flow systems. Scholars have explored the interaction between cavitation bubbles and particles at the microscopic level [10], [11], [12]. Gonzalez-Avila [13] and Poulain [14], through their investigations on the interaction between particles of different materials and cavitation bubbles, demonstrated that the motion of particles under the influence of cavitation is significantly influenced by the relative distance between the bubbles and the particles. Teran [15] and Li [16] conducted experimental and numerical simulations to investigate the interaction between cavitation bubbles and particles. Their findings revealed that, at the moment of bubble collapse, the bubbles exert an accelerating effect on the nearby particles. Borkent [17], Wagterveld [18], and Wu [19] examined the acceleration of particles during the expansion and collapse of cavitation bubbles. Their studies demonstrated that particles with smaller volumes and lower masses are more susceptible to acceleration. Shen [20], [21] investigated the interaction between a single cavitation bubble and a single suspended particle. His research demonstrated that, under conditions where the particle and bubble sizes are comparable, the suspended particle is predominantly influenced by the radiation force exerted by the bubble during its evolution, with no jet impact on the particle's surface. Furthermore, Shen conducted experimental studies on the dynamics of particles on a wall under the influence of cavitation bubbles. He introduced a characteristic parameter for classifying the interaction of particles on rigid walls, distinguishing whether the particles are affected by the jet produced by cavitation bubble collapse or the radiation force generated during bubble evolution. This parameter is related to the physical properties of the particles and the ratio of particles to bubble sizes. In addition to the aforementioned studies on the microscopic interactions between cavitation bubbles and particles, methods such as disk wear testing, slurry pot experiments, water tunnel experiments, Venturi tube tests, and numerical simulations have been employed [22], [23], [24], [25], [26], [27]. These approaches, by comparing the mass loss and surface morphology changes of specimens before and after the experiments, define the intensity of cavitation-particle interactions from a macroscopic perspective. Amarendra [28] utilized a slurry erosion testing apparatus to investigate the synergistic effects of particle erosion and cavitation erosion. His findings revealed that the combined effect of both erosions led to a greater material mass loss than particle erosion alone. He posited that the presence of cavitation bubbles could potentially enhance the velocity at which particles impact the wall. Thapa [29] investigated the synergistic erosion effects of particle erosion and cavitation erosion using a rotary test rig equipped with a cavitation inducer. His findings similarly demonstrated that the combined effect of both erosions led to material damage that was substantially greater than the cumulative effect of particle erosion and cavitation erosion individually. In contrast, other studies have reached opposing conclusions. Wang [30] examined the influence of particles on the cavitation process using a rotating disc apparatus with Q235 as the sample material at a rotational speed of 2800 rpm. His findings revealed that the presence of particles altered the cavitation erosion process, with smaller particles (100 nm, 500 nm) exhibiting a suppressive effect on synergistic damage. Chen [31] and Guo [32] found that by adding small diameter particles, the cavitation zone area was significantly reduced, thereby suppressing the cavitation effect. Moreover, the suppression effect diminished markedly as the particle diameter increased. In contrast, the addition of larger diameter particles exacerbated the erosion process. In summary, existing research at the microscopic level predominantly focuses on the interaction between individual particles and cavitation bubbles, which significantly diverge from actual physical processes. At the macroscopic level, studies fail to clearly reveal the mechanisms of particle bubble interactions in multiphase flows. Furthermore, due to the involvement of numerous variables and the complexity of the flow states, no consistent conclusions have been reached [33], [34], [35]. In contrast to previous studies that have primarily focused on single bubble dynamics, cavitation jets near rigid walls, or the response of individual particles, the present study investigates the dynamic behavior of particle clusters subjected to the action of a single cavitation bubble near a rigid wall. Particular emphasis is placed on the evolution of particle velocity, kinetic energy, and wall impact intensity during bubble expansion, collapse, and jet impingement. The novelty of the present study is reflected in three main aspects. First, high speed imaging experiments are integrated with coupled VOF and DPM numerical simulations to characterize the dynamic response of particle clusters induced by a cavitation bubble near a rigid wall, covering the entire process from bubble expansion to repeated collapse. Second, the effects of the dimensionless distance parameter λ, maximum bubble radius Rmax, and particle diameter dp on particle velocity and impact kinetic energy are systematically quantified. Third, the mechanisms responsible for the variation in particle velocity and the formation of wall impact intensity are elucidated in terms of particle inertial response, jet induced acceleration, and energy transfer. These results improve the understanding of momentum and energy transfer between bubble induced flow fields and particle clusters, and provide a basis for further studies on erosion mechanisms associated with the coupled effects of cavitation and particles. In the present study, experimental observations and numerical simulations are conducted to investigate the interaction between a single cavitation bubble and particle clusters in a particle laden fluid medium during the expansion and collapse of a bubble near a rigid wall. The primary discussion centers on the dynamic characteristics of particles in the fluid medium under the influence of cavitation bubble. The objective of this paper is to explore the mechanisms governing the interaction between cavitation bubbles and particles in particle-laden media under cavitation conditions, providing a theoretical basis for predicting flow surface damage under the synergistic effects of cavitation and erosion. The present study focuses on the dynamic response of particle clusters induced by a cavitation bubble near a rigid wall. Section 2 introduces the experimental apparatus, the bubble generation method, the particle parameters, and the experimental procedure. Section 3 establishes the coupled VOF and DPM numerical model and presents the numerical calculations and analysis. Section 4 investigates the velocity characteristics of particle clusters under the action of a near wall cavitation bubble. Section 5 evaluates the impact intensity of particle clusters on the rigid wall and analyzes the relevant influencing factors. Section 6 discusses the transfer and dissipation of bubble energy into the surrounding fluid, the jet, and the particle clusters. Section 7 discusses the limitations of the present work and outlines future research directions. Experimental Experimental system An experimental investigation was conducted to study the interaction between cavitation bubbles and particle clusters near the wall surface in a particle flow. The experimental system primarily consists of the following key equipment: a transparent acrylic water tank with dimensions of 300 mm × 300 mm × 300 mm, a low-voltage spark cavitation bubble generator, a particle feeding device a high-smoothness rigid wall surface, as well as the high-speed imaging system and illumination system. The high-speed imaging system consists of a Phantom v1210 high-speed camera, optionally equipped with a Canon macro lens EF 100 mm f/2.8 L IS USM, to ensure accurate focusing and magnification of the experimental region. The illumination system consists of a 300 W LED light source, with a light diffusion panel mounted on the side of the water tank near the light to achieve uniform illumination of the experimental area. The high-speed camera is connected to a computer, where image acquisition and data recording are controlled using Phantom PCC software. A DC adjustable power supply with an output voltage range of 0–100 V serves as the input power source for the low-voltage spark cavitation bubble generator, the output voltage of the power supply is regulated to charge the internal capacitor of the low-voltage electric spark cavitation bubble generator and maintain it at a predefined level. Cavitation bubbles are subsequently generated through controlled capacitor discharge under various voltage conditions, facilitating the adjustment of the cavitation bubble radius. The particle dispersion system is designed to accommodate particles of varying sizes and shapes. By adjusting the position and tilt of the discharge plate, the system regulates the entry position of the particles into the water. The overall configuration is as follows: the water tank is positioned on an adjustable support stand, while the rigid wall is placed inside the tank using a clamping device. The particle dispersion system is mounted at the upper section of the water tank. A schematic representation of the experimental setup is provided in Fig. 1.Fig. 1Schematic diagram of the experimental system for investigating the interaction characteristics between cavitation bubbles and particle clusters near the wall in a particle flow. Schematic diagram of the experimental system for investigating the interaction characteristics between cavitation bubbles and particle clusters near the wall in a particle flow. As shown in Fig. 1, the bubble generation device employs a capacitor bank composed of parallel connected electrolytic capacitors rated at 100 V, with a total capacitance of 7600 μF. For electrolytic capacitors with an individual capacitance exceeding 1000 μF, accurate voltage rating and capacitance, together with a low leakage current, are essential for circuit reliability and directly affect the repeatability of the bubble parameters generated under identical experimental conditions. Accordingly, high precision electrolytic capacitors were used, and oxygen free copper conductors were adopted to minimize losses between circuit components. A 1kΩ, 5 W cement resistor was connected in series in the charging circuit as a protective resistor to reduce the charging current of the capacitor bank and thereby prolong the service life of the capacitors. The discharge circuit connected to an electrode made of 0.15 mm diameter oxygen-free copper wire; and the circuit input connected to a DC adjustable power supply (output voltage: 0–100 V, maximum output power: 300 W). The circuit diagram is shown in Fig. 1(b). During the experimental preparation phase, the rigid wall was first secured onto the fixture. The wall material, composed of high-strength aluminum alloy (AL7075), was subjected to precision surface machining to ensure high smoothness, thereby facilitating effective reflection of illumination introduced from beneath the cavitation bubble adjacent to the wall. This optical configuration ensured adequate lighting for the visualization of the cavitation bubble evolution and significantly enhanced the fidelity of detail capture throughout the experimental process. Subsequently, the entire assembly was immersed in the water tank. With the aid of a level gauge, the position of the adjustable stand, the discharge outlet of the particle dispersion system, the tilt angle, and the electrode support frame are adjusted to set the distance between the cavitation bubble center and the surface of the rigid wall to the required value for the experiment. Deionized water with a density of 1000 kg/m3, a viscosity of 1.002 mPa·s, and a surface tension of 0.0728 N/m was then introduced into the water tank. During the experiment, the output voltage of the DC adjustable power supply is adjusted according to the required cavitation bubble diameter to charge the low-voltage spark cavitation bubble generator. Before each experiment, the capacitor was charged by a DC power supply to a preset voltage. After the target voltage was reached, the charging process was terminated, and the stored electrical energy was subsequently released through the electrode pair. The resulting low-voltage spark discharge generated a cavitation bubble at the electrode gap. By controlling the charging voltage, cavitation bubbles with different initial energies and maximum radii could be produced. Subsequently, adjust the focus ring of the high-speed camera lens and the angle of the illumination source to ensure that the electrode center is precisely positioned within the focal plane of the field of view, clearly visible on the computer monitor, and the shooting parameters are configured on the computer. Finally, the particle dispersion system is activated. Once the particles are uniformly distributed between the center of the electrode and the rigid wall, the shooting is initiated on the computer. The switch is then rapidly toggled to the discharge position, generating cavitation bubbles and completing the data acquisition for the experiment. The particles used in the experiment are high-purity green silicon carbide with a density of ρp = 3210 kg/m3. The particle size ranges are 16–26 mesh (0.71 mm to 1.18 mm), 26–40 mesh (0.43 mm to 0.71 mm), and 40–70 mesh (0.21 mm to 0.43 mm). Experimental system validation Before commencing the experiment, the repeatability and validity of the results are ensured by first verifying the repeatability of the experimental system. In the experimental system, the positions and physical properties of the particles, as well as the arrangement of the specimens and their material characteristics, are controllable. Therefore, the primary factor determining the overall reliability of the system is the consistency of the cavitation bubble size produced by the low-voltage spark cavitation bubble generator. To ensure this, the reliability of the cavitation bubble generator is verified in a free domain. The free domain cavitation bubble expansion and collapse experiments are conducted in deionized water with a density of ρ = 1000 kg/m3. The experiments are conducted under six distinct input voltages of the low-voltage spark cavitation bubble generator, with the high-speed camera capturing images at a resolution of 128 × 128 pixels and a frame rate of 240,000 frames per second. The average value and standard deviation of the maximum radius, Rmax, were determined, the average value of the cavitation bubble maximum radius Rmax is 7.84 mm, with the maximum value reaching 0.414 under an input voltage of 65 V, as illustrated in Fig. 2. This demonstrates that, under a fixed input voltage, the cavitation bubble parameters exhibit a high degree of consistency and repeatability. Consequently, the experimental system is deemed reliable for use in subsequent experiments. Although inherent random fluctuations exist in the capacitor discharge process, repeated experiments under a fixed input voltage showed only a small dispersion in the maximum bubble size. This indicates that the bubble generation system is sufficiently repeatable for the subsequent comparative analysis of bubble and particle cluster interactions. It should be noted that, considering the spatial resolution of the high-speed images and the uncertainty associated with discharge repeatability, the dimensionless distance parameter λ in different cases was calculated using the experimentally measured Rmax for each corresponding case. This approach reduces parameter deviations caused by fluctuations in the capacitor discharge process.Fig. 2Verification of the experimental system stability. Verification of the experimental system stability. Conduct of the experiment In the experiment investigating the interaction between cavitation bubbles and particle clusters near the wall in a particle flow, the positional relationship between the particles, bubbles, and the wall is depicted in Fig. 3. The cavitation bubble is generated and evolves within the particle fluid medium.Fig. 3Schematic representation of the experimental process (Where D denotes the vertical distance from the cavitation bubble center to the wall, Rmax is the maximum radius attained during bubble expansion, and vpz represents the vertical velocity of the particles as they fall freely under the influence of gravity.). Schematic representation of the experimental process (Where D denotes the vertical distance from the cavitation bubble center to the wall, Rmax is the maximum radius attained during bubble expansion, and vpz represents the vertical velocity of the particles as they fall freely under the influence of gravity.). For the purpose of facilitating the analysis, a dimensionless parameter, λ, is introduced to represent the relationship between the distance D from the cavitation bubble center to the wall surface and the maximum bubble radius Rmax:(1)λ=D/Rmax By adjusting the output voltage of the DC adjustable power supply, cavitation bubbles of varying sizes are generated at different wall distances D, with particles of different sizes and irregular shapes being dispersed accordingly. The high-speed camera is set to a resolution of 512 × 512 pixels, with a frame rate of 37,050 frames per second, corresponding to a frame interval of 27 μs, to capture the experimental phenomena. Experimental results The experimental results are processed using Phantom PCC 3.6 Software for video processing and ImageJ software for image analysis. Classifying based on the characteristic parameter λ, which represents the distance from the cavitation bubble center to the rigid wall, it can be considered that for λ ≥ 1, when the cavitation bubble center is sufficiently distant from the rigid wall, the bubble maintains a nearly perfect spherical shape when it reaches its maximum radius, Rmax. When λ < 1, where the cavitation bubble center is in closer proximity to the wall, the influence of the rigid wall causes the bubble to deviate from a perfect spherical shape upon reaching its maximum radius, Rmax. To facilitate the analysis of bubble size under different characteristic parameters λ, the equivalent spherical diameter [36] is used as the Rmax for each experimental group. In conjunction with the aforementioned experimental methods, particles are introduced into the fluid domain of dimensions 40 mm × 40 mm × 40 mm on the side opposite the rigid wall for each experimental group. The particle feeding rate is controlled at 1.0 g to 1.5 g per second, with the particles falling freely in water. The particle number represents the number of particles identified within the central 40 mm × 40 mm observation region of the high-speed camera prior to bubble generation. It does not correspond to the total number of particles in the entire experimental system. Specific experimental parameters are provided in Table 1.Table 1Experimental parameters for cavitation bubble and particle cluster interaction near rigid wall.CaseD(mm)Rmax(mm)Particle size(mm)Number of particlesλ15.748.310.21 ∼ 0.433000 ∼ 40000.6928.018.430.71 ∼ 1.182000 ∼ 25000.9539.758.260.71 ∼ 1.182000 ∼ 25001.18411.618.470.43 ∼ 0.711000 ∼ 12001.37512.437.910.43 ∼ 0.711500 ∼ 20001.57613.588.570.71 ∼ 1.181000 ∼ 15001.58715.348.590.21 ∼ 0.433000 ∼ 40001.79 Table 1Experimental parameters for cavitation bubble and particle cluster interaction near rigid wall.CaseD(mm)Rmax(mm)Particle size(mm)Number of particlesλ15.748.310.21 ∼ 0.433000 ∼ 40000.6928.018.430.71 ∼ 1.182000 ∼ 25000.9539.758.260.71 ∼ 1.182000 ∼ 25001.18411.618.470.43 ∼ 0.711000 ∼ 12001.37512.437.910.43 ∼ 0.711500 ∼ 20001.57613.588.570.71 ∼ 1.181000 ∼ 15001.58715.348.590.21 ∼ 0.433000 ∼ 40001.79 Experimental parameters for cavitation bubble and particle cluster interaction near rigid wall. Based on the experimental data presented in Table 1, the motion of the particle cluster during the expansion and collapse of the cavitation bubble near the rigid wall is captured, as depicted in Fig. 4.Fig. 4Process of cavitation bubble and particle cluster interaction near the rigid wall(The vertical dashed line represents the surface of the highly reflective rigid wall, with the left side of the dashed line corresponding to the reflected image of the right side. Rmax denotes the moment when the bubble reaches its maximum expansion size. In the experimental group where the cavitation bubble center is positioned relatively far from the rigid wall, the bubble undergoes multiple collapse events. In each experimental group, the same color of arrowheads indicates the direction of motion of the same particle at different time intervals, with the length of the arrow being independent of the particle's velocity.). Process of cavitation bubble and particle cluster interaction near the rigid wall(The vertical dashed line represents the surface of the highly reflective rigid wall, with the left side of the dashed line corresponding to the reflected image of the right side. Rmax denotes the moment when the bubble reaches its maximum expansion size. In the experimental group where the cavitation bubble center is positioned relatively far from the rigid wall, the bubble undergoes multiple collapse events. In each experimental group, the same color of arrowheads indicates the direction of motion of the same particle at different time intervals, with the length of the arrow being independent of the particle's velocity.). The stability of particle feeding was evaluated based on the temporal variation in the number of particles within the observation region obtained from the high speed image sequences. For a falling height of 10 cm, the particle input amount and the initial particle distribution were maintained within a similar range among different experimental cases. The particle velocity before entering the water was estimated from the free fall relation to be approximately 1.4 m/s, which is considerably lower than the particle velocity generated under the action of the cavitation jet. Therefore, when the maximum bubble radius and particle diameter distribution are comparable, the particle feeding process is not expected to introduce a significant systematic difference in the comparison among different cases. As evidenced in Fig. 4, in the different experimental groups, the challenge of capturing the displacement variations of individual particles at different time instances is notably greater in Case 1 and Case 7, where the particles are finer and the particle density is higher, compared to Case 4 ∼ Case 6, where the particles are coarser and the particle density is lower. When the characteristic parameter λ < 1, only a single expansion and collapse cycle of the cavitation bubble is observed (as seen in Case 1 and Case2). In contrast, when λ > 1, an increase in λ leads to the observation of two to three more intense collision and collapse events. Subsequently, the cavitation bubble disperses into a gas cluster and ceases further expansion and collapse. To illustrate the dynamic behavior of particle clusters under the influence of cavitation bubbles, the motion of particle clusters during the bubble evolution process in a free field was included, as shown in Fig. 5. In contrast to the interaction characteristics between cavitation bubbles and particles near the flat wall as illustrated in Fig. 4, cavitation bubbles in the free field exhibit purely symmetric expansion and collapse about their centers throughout multiple oscillation cycles, without the formation of jets. As a result, particles located near the center of the cavitation bubble are initially displaced outward due to the perturbation induced by the bubble interface during expansion, and subsequently drawn inward toward the bubble center during the collapse phase. After undergoing multiple cycles of expansion and collapse, the cavitation bubble ceases further evolution. As illustrated in Fig. 5, the marked particle clusters exhibit minimal displacement following the evolution of the cavitation bubble in the free domain. The presence of a rigid wall and the associated jet formation are not prerequisites for particle motion; however, they significantly enhance particle acceleration toward the wall and increase particle–wall impact intensity. Therefore, the following section provides a detailed analysis of the dynamic behavior of particle clusters under such wall confined conditions.Fig. 5Interaction process between cavitation bubble and particles in free domain (The differently shaped and colored boxes indicate the positions of the same particle clusters at different moments.). Interaction process between cavitation bubble and particles in free domain (The differently shaped and colored boxes indicate the positions of the same particle clusters at different moments.). For different characteristic parameters λ, during the expansion phase of the first cavitation bubble expansion and collapse cycle, the particle cluster in the vicinity of the maximum bubble radius Rmax (within the red circle in Case 1 of Fig. 4) is displaced in the direction away from the bubble center, as the bubble wall displaces the surrounding fluid medium. During the collapse, the particle cluster moves towards the bubble center, driven by the fluid medium displaced by the contracting bubble wall. To further elucidate the mechanism governing particle acceleration during jet impact, this study refers to the particle dynamics model established in our previous work on the interaction between a single particle and a single cavitation bubble. According to this model, the motion of particles subjected to a cavitation bubble is not controlled solely by hydrodynamic drag. Rather, individual particles in the cluster are mainly affected by the radiation force associated with bubble expansion and collapse, gravity, buoyancy, drag force, and added mass force [20], as expressed in Eq. (2). This formulation further indicates that the force exerted by the bubble decreases with increasing distance between the particle and the bubble center.(2)F=mpρR2R˙2+RR¨πdp2/L0-1/2ρvp2CDπdp2/mp+2/3πdP3ρWhere, mp is the particle mass; ρ is the fluid medium density; L0 is the distance between the particle center and the bubble center; R is the bubble radius; R˙ is the bubble wall velocity; R¨ is the bubble wall acceleration; CD is the drag coefficient of the fluid; dp is the particle's equivalent radius; vp is the particle's velocity. It should be emphasized that the above relation was derived for the interaction between a single particle and a single cavitation bubble. The present study, however, focuses on the motion of particle clusters subjected to a cavitation bubble. In this case, particle shielding, local concentration variations, and wall confinement can further complicate the force balance on individual particles. Therefore, the relation is not used here for a precise prediction of the force acting on each particle in the cluster. Rather, it serves as a mechanistic basis for explaining that particle acceleration is governed by the combined effects of bubble induced pressure forcing, hydrodynamic drag, and added mass. For cases with the characteristic parameter λ < 1 or λ ≈ 1, namely Case 1, Case 2, and Case 3, at the moment of cavitation collapse, a strong jet directed towards the rigid wall is generated. This jet drives the particles in the high-speed jet region, which then impacts the rigid wall. For the experimental groups with characteristic parameter λ > 1, it is observed that during the second and third cavitation expansion and collapse cycles, the bubble is unable to retain its regular spherical shape. The gas volume and energy contained within the bubble are significantly dissipated, leading to a progressive weakening of the collapse intensity. Consequently, the force exerted by the bubble on the particle cluster also diminishes during the subsequent expansion and collapse phases. In experimental groups with smaller particle sizes (Case 1 and Case7), the relatively small mass of individual particles enables them to overcome inertial resistance under weaker flow field disturbances. In contrast, for the experimental groups with larger particle sizes (Case2, Case 3 and Case 6), the disturbances induced by the subsequent cavitation expansion and collapse are insufficient to overcome the inertia of the particles. In all experimental groups, the particles within the particle clusters move in the direction of the jet impact induced by the cavitation bubble. For cases in which the particle size is between those of the above experimental groups, namely Case 4 and Case 5, the particle motion response under the disturbed flow field induced by bubble evolution also exhibits an intermediate behavior. Upon being accelerated by the jet and subsequently impacting the rigid wall, the particles rebound due to the interaction with the wall surface. According to the findings reported in Reference [21], when the ratio of particle diameter to the maximum radius of the cavitation bubble is less than 0.4, the dynamic behavior of particles near the flat wall, particularly during the stage of jet formation, is predominantly influenced by the impact force exerted by the jet. For small particles, the effective impact area exposed to the jet on the spherical surface is minimal, and the probability of the jet striking the particle perpendicularly at its apex is exceedingly low. When the jet deviates from the vertical axis above the particle, the low-mass particle moves in a random direction under the jet’s influence. The rebound velocity of the particle after collision with the wall is substantially lower than the jet impact velocity and decays rapidly. Therefore, this study primarily focuses on the instances when particles with relatively high kinetic energy impact the wall, providing a theoretical basis for assessing wall damage caused by the coupling effects of cavitation and particles in multiphase flow media. The impact intensity of the particle cluster on the wall, induced by the jet, is primarily associated with the kinetic energy of the particles. This is influenced by factors such as particle velocity, particle mass, and particle morphology. As illustrated by the experimental process in Fig. 4, for finer particles, at the moment they reach the rigid wall, the shielding effect of the gas cluster generated by cavitation bubble dissipation complicates the precise measurement of particle velocity. Therefore, in the following sections, numerical simulation methods will be employed to further investigate the dynamical characteristics of the particle clusters and the flow field variations near the cavitation bubble. Numerical calculation and analysis Calculation conditions The numerical simulations conducted in this study are intended to elucidate the influence of the flow field induced by the evolution of a cavitation bubble near a rigid wall on the motion of particle clusters. The initial conditions were prescribed based on the experimentally observed bubble morphology and characteristic dimensions. The VOF method was adopted to resolve the evolution of the gas liquid interface, while the DPM was employed to track the particle trajectories. It should be noted that the present model does not explicitly incorporate a gas liquid phase change mass transfer model. As a result, the evaporation and condensation processes of the gaseous medium inside the bubble, as well as the associated phase change energy consumption, are not directly resolved. Therefore, the numerical model is primarily used to characterize bubble expansion, bubble collapse, near wall jet formation, and the resulting particle motion driven by the induced flow field. It is not intended to provide a precise prediction of the internal phase change process, the local pressure peak during the final stage of collapse, or the shock wave intensity. According to classical bubble dynamics, the radial motion of a bubble is governed by the pressure difference across the bubble interface, liquid inertia, surface tension, and viscous effects. For a cavitation bubble near a rigid wall, the wall confinement breaks the spherical symmetry of the surrounding flow field, leading to asymmetric collapse and the formation of a high speed jet directed toward the wall. Based on the Keller-Miksis [37] cavitation bubble dynamics equation, the initial parameters for numerical computation are determined. Considering the influence of the rigid wall on the cavitation bubble dynamics, the basic equation needs to be modified. Typically, when calculating the cavitation bubble dynamics equation near the wall, it is treated as a problem of interaction between two cavitation bubbles with identical parameters, with one bubble being a mirror image at an equal distance from the wall [38]. It is assumed that when the two parameters are identical, this represents the cavitation bubble dynamics equation near a rigid wall, and the expression is as follows [39]:(3)1-R˙/CRR¨+3/2R˙21-R˙/3C=1+R˙/CPR-p0/ρ+R/ρCd(PR-p0)/dt-R/2D2R˙2+RR¨Where,(4)PR=p0-pv+2δ/R0R03-bR03/R13-bR03k-2δ/R-4μR˙/R+pv(5)d(PR-p0)/dt=-3kR03-bR03kp0-pv+2δ/R0R2R3-bR03-k-1R˙+2δR˙/R2-4μRR¨-R˙2/R2Where, PR the internal pressures of cavitation bubble; C is the speed of sound in the liquid, which is 1500 m/s; R0 is the initial radius of bubble; p0 is ambient pressure, 101,325 Pa; pv is the saturation vapor pressure inside the cavitation bubble, 3250 Pa; μ is the viscosity of the liquid, 1.3127 mPa∙s; σ is the surface tension coefficient of the liquid, 0.0728 N/m; b is van der Waals coefficient, 0.0016; k is the polytropic exponent, 4/3. Bubble expansion, collapse, and jet formation are strongly dependent on the prescribed initial and boundary conditions. The initial bubble radius and internal pressure determine the initial energy of the bubble during the expansion stage, thereby influencing the maximum bubble radius, collapse time, and collapse intensity. A bubble with a larger initial radius or a higher internal pressure can transfer more energy to the surrounding liquid during expansion, which in turn enhances the local pressure gradient and jet velocity during collapse. As a result, the particle cluster can acquire higher velocity and greater kinetic energy. The initial distance between the bubble and the rigid wall is another critical parameter governing near wall bubble dynamics. When the bubble is located close to the rigid wall, the wall constrains the motion of the liquid on the wall side, leading to an asymmetric pressure and velocity field around the bubble. This asymmetry promotes the formation of a high-speed jet directed toward the wall during collapse, accompanied by the migration of the bubble center toward the wall. Such wall induced asymmetric collapse constitutes the dominant mechanism by which a near wall cavitation bubble affects the motion of particle clusters. The equation (3) is solved using the ode45 method [40], and the relationship curves between the cavitation radius R and internal pressure PR as functions of time are obtained, with the maximum cavitation radius Rmax determined from the experimental measurements. As an example, using the maximum cavitation radius Rmax = 8.31 mm observed in experimental Case 1, the curves representing the relationship between the cavitation radius, internal pressure, and time are illustrated in Fig. 6.Fig. 6Curves of the cavitation radius and internal pressure variation with time near the rigid wall (The dashed line represents the internal pressure PR0 at the initial cavitation bubble radius R0.). Curves of the cavitation radius and internal pressure variation with time near the rigid wall (The dashed line represents the internal pressure PR0 at the initial cavitation bubble radius R0.). As depicted in Fig. 6, the first expansion and collapse process of the cavitation bubble near the wall can be characterized by the moment at which the maximum radius Rmax is reached, serving as a boundary that separates the expansion and collapse phases. It is assumed that these two phases are approximately symmetric around this moment. The radius and corresponding internal pressure at a given point during the collapse phase are taken as the initial conditions for the numerical calculation. The initial conditions for each group, as outlined in Table 1, are provided in Table 2.Table 2Initial conditions for numerical calculations.CaseRmax (mm)D (mm)R0 (mm)PR0 (MPa)18.315.740.513028.438.010.513138.269.750.512848.4711.610.513357.9112.430.512068.5713.580.513578.5915.340.5135 Table 2Initial conditions for numerical calculations.CaseRmax (mm)D (mm)R0 (mm)PR0 (MPa)18.315.740.513028.438.010.513138.269.750.512848.4711.610.513357.9112.430.512068.5713.580.513578.5915.340.5135 Initial conditions for numerical calculations. Calculation methods and boundary conditions The numerical simulation of the interaction between the near-wall cavitation bubble and the particle cluster involves capturing the evolution of the cavitation bubble and the particle motion trajectories. The VOF multiphase flow model is chosen to capture the boundary changes of the cavitation bubble [41], and the laminar and unsteady methods are used to calculate the flow field. Except for the left side, which is the wall surface, the other surfaces of the fluid computational domain are set as pressure outlets, with the pressure value equal to the ambient pressure. The velocity coupling uses the Coupled method, pressure is computed using PRESTO!, while density, momentum, and energy are calculated using the Second Order Upwind scheme. The volume fraction is calculated using the Compressive method. The internal medium of the cavity is water vapor, with the density type set as ideal-gas. The selection of the time step was guided by the Courant number, which represents the ratio between the fluid displacement within a single time step and the local grid size. For the explicit multiphase flow formulation adopted in this study, the volume fraction at the current time step is updated based on the solution obtained at the previous time step. Accordingly, the time step must be sufficiently small to ensure stable and accurate interface capturing. Considering numerical accuracy, convergence stability, and computational cost, the Courant number was controlled to be less than 0.5. Under this constraint, the time step used in the simulations was in the range of 1e-8 s to 1e-7 [39], [42]. For the interaction between the fluid medium and the particle clusters, the Discrete Phase Model (DPM), utilizing the Euler-Lagrange approach, is employed for the computation. In this approach, the fluid phase is treated as a continuous medium and solved using the Navier-Stokes equations, while the differential equations governing particle motion are solved in the Lagrangian framework. The particle trajectory is determined by the interaction between the discrete phase and the fluid phase. The interactions between particles are neglected, and the force balance equation for the particles is expressed as follows [43]:(6)mpdvp/dt=FD+FP+FVM+FBWhere,(7)FD=1/8CDρπdp2v-vpv-vp(8)FP=1/4πdp3ΔP(9)FVM=-1/12πdp3ρPdvp/dt(10)FB=1/6πdp3(ρP-ρ)gWhere, FD is the drag force acting on the particle; FP is the force due to the pressure gradient; FVM is the virtual mass force acting on the particle; FB is the buoyancy force acting on the particle; v is the velocity of the fluid medium; ΔP is the pressure difference on the surface of the particle. The heat exchange between the solid and liquid phases is disregarded, with only momentum and turbulent kinetic energy exchange considered. Two-phase coupling is realized by alternately solving the governing equations for the continuous liquid phase and the discrete particle phase. The momentum exchange occurs at each control volume as the particles move along their trajectory through the fluid phase. The expression for the momentum exchange term, SM, is given by [44]:(11)SM=∑FD+FP+FVM+FBMp△tWhere, Mp is the mass flow rate of the particles; Δt is the time step. The momentum source term in Eq. (11) represents the reaction force exerted by the particles on the continuous phase and is obtained from the volume-averaged particle–fluid interaction force within the computational cell. Mesh independent verification For the simulation of the interaction between the near-wall cavitation bubble and the particle clusters, three-dimensional calculations were conducted using Fluent software. The computational domain was defined as a cubic region with dimensions of 40 mm × 40 mm × 40 mm, while grid refinement was applied specifically to the region of 30 mm × 30 mm × 30 mm near the rigid wall. A grid independence study was conducted using an initial bubble radius of R0 = 0.5 mm and an initial internal pressure of PR0 = 130 MPa. Six different grid configurations were compared, with the results summarized in Table 3. When the number of radial grid cells exceeds 50, the variation in the predicted maximum bubble radius becomes significantly smaller, and further mesh refinement has a limited effect on the numerical results. Considering both computational accuracy and computational cost, the mesh with 50 radial grid cells was selected for the subsequent simulations. Consequently, Table 3, which consists of a total of 9.74 million grid cells, was selected for the final grid partitioning. To further illustrate the mesh generation strategy, the overall computational mesh and the locally refined regions are presented in Fig. 7. The computational domain includes the sand layer surface and outlet boundary, while local mesh refinement is applied in the bubble and near-wall regions to improve the resolution of the bubble evolution and jet formation.Table 3Mesh independent analysis.CaseCells in radial direction inside bubbleRmax(mm)1308.622408.473458.384508.295558.356608.27Fig. 7Computational domain mesh generation. Table 3Mesh independent analysis.CaseCells in radial direction inside bubbleRmax(mm)1308.622408.473458.384508.295558.356608.27 Mesh independent analysis. Computational domain mesh generation. Particle arrangement of the calculation area In accordance with the actual experimental conditions and the particle distribution, particles are placed within the computational domain. In accordance with the experimental setup, the initial velocity of the particles during free fall in the fluid medium, as well as the effect of gravity on the particles, are considered. Particles are uniformly distributed within the 40 mm × 40 mm × 40 mm computational domain, according to the particle sizes listed in Table 1. For each calculation, the particles are categorized into three groups based on their sizes within the experimental particle size range, with each group accounting for 1/3 of the total number of particles. Based on the position and dimensions of the computational domain, a program was developed using Python software to implement the aforementioned arrangement rules, generating the.inj particle injection files. This process completed the initial placement of the particle group for each calculation case. The initial particle arrangements corresponding to the experimental groups in the numerical simulations are shown in Table 4.Table 4Numerical simulation of the initial arrangement of particle clusters in each group.CaseParticle size distribution(mm)Number of particles10.2/0.3/0.4300020.8/1.0/1.2200030.8/1.0/1.2200040.5/0.6/0.7100050.5/0.6/0.7150060.8/1.0/1.2100070.2/0.3/0.43500 Table 4Numerical simulation of the initial arrangement of particle clusters in each group.CaseParticle size distribution(mm)Number of particles10.2/0.3/0.4300020.8/1.0/1.2200030.8/1.0/1.2200040.5/0.6/0.7100050.5/0.6/0.7150060.8/1.0/1.2100070.2/0.3/0.43500 Numerical simulation of the initial arrangement of particle clusters in each group. For the tracking of the particle phase, interactions with the continuous phase are implemented. The DPM (Discrete Phase Model) iteration interval is set to 10, and the maximum number of tracking steps is limited to 500. The physical model employed for tracking the discrete phase incorporates the Saffman lift force, virtual mass force, and pressure gradient force. The drag force criterion for particle momentum exchange is defined as nonspherical, with a shape factor of 0.3. Numerical calculation results Based on the conditions outlined above and in conjunction with the experimental process images, numerical simulations are performed to investigate the interaction between the near-wall cavitation bubbles and the particle clusters for Case1 to Case7. A normalized time parameter T* was introduced, with total duration from bubble inception to the impact of the jet on the rigid wall used as the characteristic time scale. This treatment allows the numerical simulations and experimental observations to be compared at corresponding dimensionless time instants, thereby enabling a clearer comparison of the key morphological features during bubble evolution. This normalization enables a coherent comparison between experimental observations and numerical simulations and facilitates the systematic analysis of particle kinematic behavior, cavitation bubble evolution, jet morphology, and the dynamic characteristics of the jet within a unified temporal framework. To facilitate comparison, the time interval from cavitation bubble generation to jet impact on the rigid wall is normalized for both the experimental measurements and numerical simulation results. The changes in the geometric morphology of the cavitation bubbles, the velocity variations of the particle clusters, and the fluid velocity distribution near the cavitation bubbles center during the expansion and collapse process for each experimental group are presented in Fig. 8, presenting a two-dimensional projection of the three-dimensional simulation results.Fig. 8Numerical simulation results of the interaction between the cavitation bubble and the particle clusters near the rigid wall (In the figure, panel A illustrates the volume fraction of water vapor within the cavitation bubble region; panel B depicts the velocity distribution of the particle clusters influenced by the cavitation bubble; and panel C shows the velocity distribution of the fluid medium in the cavitation bubble region. The corresponding numerical scale for each simulation result is provided at the bottom.). Numerical simulation results of the interaction between the cavitation bubble and the particle clusters near the rigid wall (In the figure, panel A illustrates the volume fraction of water vapor within the cavitation bubble region; panel B depicts the velocity distribution of the particle clusters influenced by the cavitation bubble; and panel C shows the velocity distribution of the fluid medium in the cavitation bubble region. The corresponding numerical scale for each simulation result is provided at the bottom.). As shown in Fig. 8 the numerical simulation results of the cavitation bubble morphology (A) at the same normalized time T* demonstrates a good agreement with the experimental results for each group. However, following the initial collapse, a discrepancy is observed between the morphology of the gas cluster formed in the simulation and the experimental results. This discrepancy can be attributed to the omission of cavitation bubble wall evaporation and condensation effects during the oscillation process, which causes deviations in the subsequent expansion and collapse phases when compared to experimental observations [45]. The experimentally and numerically observed bubble center migration, formation of a wall directed jet, and acceleration of the particle cluster toward the wall agree with the established theory of near wall bubble dynamics. As shown in the particle velocity distribution results (B) of each experimental group in Fig. 8, it can be observed that, during the process of cavitation bubble formation and expansion to the maximum radius Rmax, the particles are displaced away from the center of the cavitation bubble. Particle velocity is primarily influenced by the expansion of the cavitation bubble wall, which propels the surrounding fluid medium. As the cavitation bubble radius increases, the velocity of the particles progressively diminishes. The characteristics of particle motion are also closely related to particle mass. In contrast to the motion of the particles, the evolution of the cavitation bubble wall occurs with greater intensity and speed. Additionally, after the first expansion and collapse cycle, the impact of the subsequent cavitation bubble evolution on the surrounding flow field substantially weakens, leading to a significant reduction in its influence on the dynamic state of the particle clusters. Thus, for particles with larger mass, their greater inertia results in a weaker response to the expansion and collapse processes of the cavitation bubble. This is clearly evident in the velocity distribution maps of the particles across the different groups. At any given moment, the simulation results for smaller particles consistently show higher velocities compared to their larger counterparts. As depicted in the particle velocity distribution maps surrounding the cavitation bubble in Fig. 8, the moment of maximum particle velocity coincides with the initial rapid expansion phase of the cavitation bubble. During this stage, the maximum velocity of the particle clusters is primarily determined by their proximity to the cavitation bubble center. Particles located closer to the cavitation bubble center are subjected to a stronger push, resulting in a higher velocity away from the cavitation bubble. For Case 1 to Case 7, the maximum particle velocities across different particle sizes and cavitation bubble dimensions range from 15 m/s to 40 m/s. In terms of the particle velocity upon impact with the rigid wall under the influence of the cavitation bubble, as illustrated by the flow field velocity distribution map (C), at the moment of cavitation bubble collapse when a jet is generated, particles move toward the rigid wall in response to the jet directed towards it. The velocity of these particles is governed by the characteristic parameter λ and the specific properties of the particles. The detailed characteristics of this process are discussed in subsequent sections. Particle clusters velocity characteristics Based on the experimental and numerical simulation results, the relationship between the particle clusters’ velocity and the normalized time under the influence of the cavitation bubble near the rigid wall is illustrated in Fig. 9. The experimental values correspond to the average velocity of at least five particles within the closest proximity to the cavitation bubble, where their velocities can be clearly captured at each specific moment. In the experiments, particularly for the smaller particle sizes in Case 1 and Case 7, as well as the experimental groups with a large number of particles, the velocity data for particles in the central region of the jet cannot be analyzed at the moment of cavitation bubble collapse into a gas cluster. This is due to the shielding effect of the gas cluster, which obstructs clear observation and measurement. At the moment of bubble formation, the interference from the high-intensity light produced by the electrode discharge also prevents the capture of the particle group velocity at the cavitation bubble's center. Therefore, there is a lack of experimental data on the particle group velocity during this specific time point. In contrast, the numerical simulation effectively circumvents this limitation. By directly extracting the velocity component from the particle tracking results at each time point, the maximum velocity of the particle cluster can be accurately obtained. This is one of the key reasons why the numerical simulation method is employed in this study as a complementary tool to experimental measurements.Fig. 9Curve between the particle clusters’ velocity and normalized time under the influence of a rigid wall cavitation bubble (The figure depicts key moments of significant changes in the particle group velocity, including the inception of cavitation bubble, the first collapse, and the second collapse. The red arrows denote the instant when the cavitation jet reaches the rigid wall, representing the corresponding particle group velocity at that specific time.). Curve between the particle clusters’ velocity and normalized time under the influence of a rigid wall cavitation bubble (The figure depicts key moments of significant changes in the particle group velocity, including the inception of cavitation bubble, the first collapse, and the second collapse. The red arrows denote the instant when the cavitation jet reaches the rigid wall, representing the corresponding particle group velocity at that specific time.). As observed in Fig. 9, the particle velocity shown in Fig. 9 represents the average velocity of the particles located within the bubble-affected region. The reported values were obtained by statistical averaging over all selected particles rather than by tracking a single particle. Therefore, the velocity curves characterize the collective dynamic response of the particle cluster to bubble evolution. However, due to the experimental values being the average velocity of multiple particles within the focal plane, while the numerical simulation captures the maximum velocity of the entire computational domain, a notable discrepancy in the values between the two is inevitable. Despite this, the overall velocity trends align closely, and the velocities at corresponding time points are within the same order of magnitude. Therefore, it can be concluded that the numerical method employed in this study effectively simulates the interaction process between the near a rigid wall cavitation bubble and the particle clusters. The discrepancy cannot be attributed solely to the difference between experimentally averaged quantities and numerically obtained peak values. It is also related to experimental measurement uncertainties, simplifications inherent in the numerical model, and the intrinsic uncertainty associated with the highly transient evolution of the cavitation bubble. Experimentally, particle velocities were determined from particle displacements extracted from high-speed image sequences. Therefore, the measured velocities are affected by the image resolution, frame rate, exposure time, particle identification threshold, and uncertainty in particle centroid localization. During bubble collapse and jet impact, the peak particle velocity may occur over an extremely short time interval. As a result, the finite temporal resolution of the high-speed images may smooth or underestimate the instantaneous peak velocity. In addition, bubble shielding, particle overlapping, and motion blur may introduce further uncertainty in particle trajectory extraction. From a numerical perspective, the present model does not explicitly incorporate gas liquid phase change mass transfer. This simplification may affect the prediction of the pressure peak and shock wave intensity during the final stage of bubble collapse. Nevertheless, the experimental and numerical results show good consistency in terms of the overall particle motion direction, velocity evolution trend, and order of magnitude. This indicates that the coupled VOF and DPM model is capable of capturing the dominant dynamic characteristics of particle clusters subjected to a cavitation bubble, although limitations remain in the accurate prediction of local instantaneous peak velocities and individual particle trajectories. Shen [39] studied the impact load characteristics of jet formation induced by the collapse of cavitation bubbles in the vicinity of flat rigid walls, a geometric model was developed, indicating that the jet assumes a conical configuration during the collapse process. It was found that the jet impact velocity reaches its maximum at the apex of the conical jet head. Vortical structures form around the jet center; however, the velocity of the fluid within these vortices is significantly lower than the velocity of the jet head upon reaching the rigid wall, approximately 1/10 of the jet core velocity. Therefore, the acceleration effect of the vortical region on the particles is limited, and the most significant impact on the wall is still caused by the acceleration of particles driven by the main body of the jet generated during cavitation bubble collapse. Relative to particle collisions, the localized high-speed flow and pressure gradients generated by the cavitation collapse jet within a short time scale exert a more dominant influence on particle acceleration. Consequently, neglecting particle collisions is expected to have only a limited effect on the overall particle motion direction, the velocity evolution trend, and the identification of the primary impact cases addressed in this study. Nevertheless, interparticle collisions may still influence local particle dynamics in cases with relatively high local particle concentrations, such as Case 1 and Case 7. The omission of particle collisions may introduce several uncertainties. First, the lateral dispersion of the particle cluster may be underestimated. Second, the trajectories of individual particles in locally concentrated regions may deviate from the predicted paths. Third, the variability of particle impact velocity and impact kinetic energy near the wall may be underestimated. Therefore, the present numerical results are primarily intended to reveal the overall motion characteristics of particle clusters driven by the bubble induced flow field. Limitations remain in resolving local particle accumulation, collision induced trajectory deflection, and local energy exchange in high concentration particle regions. Despite these simplifications, the experimental observations and numerical predictions show good agreement with respect to the overall motion characteristics of the particle cluster. Specifically, both the experiments and simulations demonstrate that the particle cluster is displaced away from the bubble center during the bubble expansion stage, whereas during bubble collapse, the near wall jet accelerates the particles toward the rigid wall. Smaller particles exhibit a stronger response to the transient flow field than larger particles. Moreover, a smaller dimensionless distance parameter leads to a higher jet impact velocity on the wall and a more pronounced tendency for particle wall impact. The numerical results reproduce, with reasonable accuracy, the principal particle motion direction, the velocity evolution trend, and the representative wall impact cases observed in the experiments. This agreement indicates that, under the present experimental and numerical conditions, the bubble induced flow field remains the dominant mechanism governing the overall dynamic response of the particle cluster. Accordingly, although the neglect of particle particle collisions may introduce uncertainties in local particle dispersion, individual particle trajectories, and velocity fluctuations in regions of high particle concentration, its influence on the principal conclusions of this study is expected to be limited. These conclusions mainly concern the controlling effects of the maximum bubble radius, the dimensionless distance parameter, the jet velocity, and the particle size on the overall motion behavior of particle clusters. From the simulated particle clusters velocities during the high-speed expansion phase at the early stage of cavitation bubble formation in Fig. 9, and by comparing with the cavitation bubble parameters in Table 1, it can be observed that, due to the similar Rmax sizes of Case1, Case2, Case3, Case4, Case6 and Case7, which are significantly larger than that of Case5, the maximum particle velocity in Case5 is noticeably smaller than that in the larger cavitation bubbles during this phase. This suggests that, during the first and most intense expansion and collapse process of the cavitation bubble, although larger particle sizes result in weaker acceleration of the particles, the dominant factor remains the maximum size (Rmax) to which the cavitation bubble can expand. This directly determines the energy carried by the cavity, which in turn drives the surrounding fluid medium and particle clusters. In comparison with the experimental groups containing smaller particles (Case1, Case7), the response of the larger particle groups (Case2, Case3, Case4, Case5, Case6) to the rapid expansion and contraction of the cavitation bubble wall, which induces a sudden variation in the surrounding flow field, is more delayed. This is evident in the velocity curves, where the curve for smaller particles exhibits a sharper peak at the maximum velocity, while for the larger particles, which must overcome greater inertial forces, the velocity curve shows a smoother variation at the corresponding time. From Fig. 9, it can also be observed that for the small-sized particles corresponding to Case1 and Case7, when the characteristic parameter λ < 1, the maximum impact velocity of the particles on the wall is greater than that in the case where λ > 1. Specifically, in Case1 (λ < 1), the maximum velocity of the particle group impacting the wall occurs during the first collapse of the cavitation bubble near the rigid wall, with a velocity of 43.5 m/s. At this moment, the velocity of the jet generated by the cavitation bubble collapse reaching the rigid wall (as shown in the velocity field contour map (C) for Case1) decreases from a maximum of 89 m/s to 70 m/s. This indicates that the moment of maximum particle impact velocity on the rigid wall lags behind the moment when the jet impacts the rigid wall. In Case 7, where λ > 1, the jet generated by the cavitation bubble reaches the rigid wall with a significant reduction in velocity, with the maximum speed decreasing to 15 m/s. This results in a considerable reduction in the fluid kinetic energy driving the particles toward the rigid wall, causing the maximum particle impact velocity to be approximately 3 m/s. As the particle size increases, in the Case 2 to Case 7 groups where the λ values are similar, the maximum velocity of the particle group reaching the rigid wall is lower than in cases with smaller particle sizes. Impact strength of particle clusters on rigid wall surface The impact of the particle clusters on the rigid wall under the influence of the near rigid wall cavitation bubble is a significant factor contributing to wall damage. In conjunction with numerical simulations, particle trajectories at the moment when the velocity of near wall particle clusters reaches its maximum under the impact of cavitation bubble jet, as shown in Fig. 10.Fig. 10Maximum velocity of the particle cluster upon impacting the rigid wall under the influence of the near a rigid wall cavitation bubble (The direction of the arrows represents the motion direction of individual particles.). Maximum velocity of the particle cluster upon impacting the rigid wall under the influence of the near a rigid wall cavitation bubble (The direction of the arrows represents the motion direction of individual particles.). As shown in Fig. 10, the direction of the impact velocity of the particle cluster near the rigid wall under the influence of the cavitation bubble aligns with the direction of the jet generated by the cavitation bubble collapse. Only the particle distribution is displayed for clarity, whereas the gas–liquid interface is not shown. The particle field is presented as a two-dimensional projection of the three-dimensional simulation results; therefore, apparent overlap between particles and the bubble region may occur due to projection effects. For the experimental group with the characteristic parameter λ < 1, the velocity of the jet generated by cavitation bubble collapse reaching the wall is significantly higher than that of the experimental group with λ > 1, which is consistent with the conclusion in [46]. According to the findings in [21], near a rigid wall particles experience significant acceleration due to the jet generated by cavitation collapse only when the ratio of particle size to cavitation bubble size is less than 0.4. Additionally, the smaller the particle size, the closer the maximum velocity achieved by the particles under the jet's influence is to the velocity of the jet impacting the wall [47]. By comparing Fig. 4, Fig. 5, Fig. 6, it can be observed that at the moment when the cavitation bubble collapses and the resulting jet reach the wall, the motion directions of particles near the vapor phase structure of the bubble are consistent between the two. This consistency confirms the reliability of the coupled VOF and DPM method in capturing the particle motion characteristics induced by near wall cavitation bubble dynamics. The impact intensity of the particle cluster on the rigid wall is directly related to the kinetic energy of the particles when they reach the surface near the rigid wall. According to the kinetic energy theorem, the kinetic energy carried by the particles under the action of cavitation bubble is calculated as follows:(12)Ep=1/2Vpρpvp2Where, Vp is the volume of a single particle. From equation (12), it can be observed that for particles made of the same material, their kinetic energy when impacting the rigid wall under the influence of the cavitation bubble is a direct proportional function of particle volume and particle velocity. Moreover, the impact of particle velocity on kinetic energy is significantly greater than the effect of particle volume. The particle velocity, under the condition of characteristic parameter λ < 1, can reach approximately 1/4 to 3/4 of the maximum jet velocity vjet prior to the jet front impacting the opposite bubble wall, depending on the size of the particles. Specifically, the calculation of the jet impact velocity generated by the collapse of the cavitation bubble near the rigid wall can be referred to by the following equation [48]:(13)vjet=87.94γ0.51+1/3η-κ1.2-2(14)η=H/Rmax(15)γ=d/Rmax(16)κ=Ry,m/RmaxWhere, η is the ratio of the distance between the two walls H to the maximum bubble radius Rmax; γ is the ratio of the initial distance d from the bubble center to the nearby wall to Rmax; κ is the ratio of the distance Ry,m from the top of the bubble to the nearby wall at the moment of maximum expansion to Rmax. Under the boundary condition where one of the walls is considered to be infinitely distant from the cavitation bubble (i.e., H→∞, leaving only the bottom wall), the above equation reduces to the following form:(17)vjet=87.94γ0.5 Therefore, when evaluating the impact load on the fluid medium wall containing particle cluster under the influence of the cavitation bubble, in addition to accounting for the jet impact load, the range of kinetic energy carried by particles of varying sizes can be rapidly and efficiently determined based on the cavitation bubble's position, characteristic parameter λ, and particle properties. This approach offers a theoretical foundation for calculating the wall impact load and predicting the extent of damage to the wall due to the interaction between the cavitation and the particles. To further elucidate the relationship between particle velocity and the velocity of the collapse induced jet, the particle acceleration process is analyzed from the perspective of inertial response. The near wall jet generated during bubble collapse is characterized by a highly transient nature, pronounced velocity gradients, and strong spatial localization. Consequently, particles subjected to this jet are unable to instantaneously follow the surrounding fluid motion. Instead, their acceleration is governed by the combined effects of hydrodynamic drag, pressure gradient force, and added mass force. As a result, the peak particle velocity remains lower than the jet impact velocity. The range in which the particle velocity reaches approximately 1/4 to 3/4 of the jet impact velocity is not assumed as an empirical constant. Rather, it is obtained from a statistical comparison between the peak particle velocity and the corresponding jet impact velocity in different cases. Specifically, for each case, the peak velocity of representative particles, vpmax, and the jet impact velocity at the corresponding stage, vjet, were extracted, and their ratio was evaluated as follows:(18)β=vpmax/vjetWhere, β denotes the ratio of the peak particle velocity to the jet impact velocity; vpmax is the peak particle velocity; vjet is the jet impact velocity. The statistical results indicate that, within the ranges of particle size and bubble jet intensity considered in this study, β is mainly distributed between 1/4 and 3/4. The occurrence of this velocity ratio range is closely related to the finite inertial response of the particles. The particle response to variations in the surrounding fluid velocity can be quantified using the particle relaxation time:(19)τp=ρpdp2/18μWhere, τp is the particle relaxation time, s. According to Eq. (19), for a given particle density and liquid viscosity, the particle relaxation time scales with the square of the particle diameter. Therefore, increasing the particle size substantially enhances particle inertia, leading to a delayed response to the collapse induced jet. To further characterize this inertial response, the particle Stokes number is defined as follows:(20)St=τp/tjetWhere, tjet is the jet action time, s. In the present study, the jet action time is defined as the interval between the initiation of collapse induced jet formation, during which the jet begins to produce a noticeable acceleration of the particles, and the stage at which the jet impacts the rigid wall and subsequently attenuates. This parameter characterizes the effective duration of momentum transfer from the high-speed jet to the particles. It should be emphasized that the particle relaxation time expressed in Eq. (20) is derived under the assumption of Stokes drag at low particle Reynolds numbers. In the present study, the jet generated by bubble collapse is characterized by high velocity, strong transience, and pronounced spatial localization. Under such conditions, the actual particle force is influenced not only by drag but also by nonlinear drag effects, pressure gradient force, added mass force, and other unsteady flow effects. Therefore, the Stokes number calculated from Eq. (20) is used here primarily to quantify the relative inertial response among particles of different diameters, rather than to represent the absolute particle response time under high-speed jet conditions. To further describe the inertial response intensity associated with the observed particle velocity ratio, a first order inertial response model is introduced. The particle velocity response can be approximated as follows:(21)dvp/dt=vjet-vp/τp(22)vp=vjet1-exp-tjet/τp Therefore, the ratio of the particle velocity to the jet velocity can be expressed as:(23)vp/vjet=1-exp-tjet/τp To characterize the effective inertial response of particles under the transient jet action, an effective Stokes number is introduced:(24)Steff=τp/tjet(25)β=1-exp-1/Steff Accordingly, the effective Stokes number, Steff, is given by:(26)Steff=-1/ln1-β For β is 1/4 to 3/4, the corresponding Steff ranges from 3.48 to 0.72. This indicates that, under the present operating conditions, the particle response time is of the same order of magnitude as the jet action time. During the jet action period, particles are able to acquire only a fraction of the jet momentum, but cannot completely follow the jet motion. As a result, the particle velocity remains lower than the jet velocity, giving rise to the observed velocity ratio range of approximately 1/4 to 3/4. From a mechanistic perspective, the lower bound of this velocity ratio range is primarily associated with larger particles or particles located away from the jet core region. Owing to their greater inertia and longer relaxation times, these particles cannot be sufficiently accelerated within the short duration of jet action, resulting in a relatively low velocity ratio. By contrast, the upper bound is mainly associated with smaller particles or particles located near the jet core region. These particles possess lower mass and exhibit a faster inertial response, enabling them to acquire more momentum within the limited jet action duration and thus attain a higher velocity ratio. Therefore, the observed range in which the particle velocity reaches approximately 1/4 to 3/4 of the jet impact velocity results from the combined effects of particle size, particle density, jet action duration, jet velocity, and the initial particle position. Based on experimental observations and numerical simulations, the inertial response characteristics of the particles were obtained and are summarized in Table 5.Table 5Particle inertial response parameters and velocity ratios.CaseParticle size distribution(mm)SteffResponse characteristics10.2/0.3/0.40.72 ∼ 1.44Rapid response; higher jet momentum acquisition.20.8/1.0/1.21.44 ∼ 3.48Strong inertia; delayed response.30.8/1.0/1.21.44 ∼ 3.48Strong inertia; delayed response.40.5/0.6/0.7≈1.44Moderate inertia; partial jet momentum acquisition.50.5/0.6/0.7≈1.44Moderate inertia; partial jet momentum acquisition.60.8/1.0/1.21.44 ∼ 3.48Strong inertia; delayed response.70.2/0.3/0.40.72 ∼ 1.44Rapid response; higher jet momentum acquisition. Table 5Particle inertial response parameters and velocity ratios.CaseParticle size distribution(mm)SteffResponse characteristics10.2/0.3/0.40.72 ∼ 1.44Rapid response; higher jet momentum acquisition.20.8/1.0/1.21.44 ∼ 3.48Strong inertia; delayed response.30.8/1.0/1.21.44 ∼ 3.48Strong inertia; delayed response.40.5/0.6/0.7≈1.44Moderate inertia; partial jet momentum acquisition.50.5/0.6/0.7≈1.44Moderate inertia; partial jet momentum acquisition.60.8/1.0/1.21.44 ∼ 3.48Strong inertia; delayed response.70.2/0.3/0.40.72 ∼ 1.44Rapid response; higher jet momentum acquisition. Particle inertial response parameters and velocity ratios. Transfer and dissipation of cavitation bubble energy A detailed analysis is further conducted on the transfer and dissipation of the energy carried by cavitation bubbles during their collapse, in which the energy is transmitted to the surrounding fluid medium and particle clusters. The energy carried by a cavitation bubble with a known maximum radius can be calculated using the following expression [49]:(27)Eb=4/3πp∞-p0Rmax3 Neglecting the energy consumed by the phase transition of the gaseous medium inside the cavitation bubble, the energy carried by the bubble is primarily transformed and dissipated in the forms of shock waves, radiation forces, jet kinetic energy, and particle kinetic energy. From an energy perspective, Wang [33] captured the variations in energy within the bubble medium, at the interface, and in the surrounding fluid region, and found that the energy in the region remained nearly conserved during most of the cavitation bubble oscillation cycle. Therefore, shock waves, jet kinetic energy, particle kinetic energy, and the energy transmitted to the surrounding fluid medium constitute the major portion of the energy carried by cavitation bubbles. Regarding the energy of shock waves generated during cavitation bubble collapse, at least two shock waves are typically released during bubble inception and collapse. Capturing the energy released by these shock waves remains challenging and is usually estimated by subtracting other major forms of energy transfer and dissipation from the total energy carried by the bubble. The kinetic energy carried by an individual particle can be calculated according to Eq. (12), whereas the total kinetic energy of a particle cluster requires summation over multiple particles. The energy carried by the cavitation bubble jet can be estimated based on the bubble size and the corresponding jet dimensions, and is given by the following expression [42]:(28)Ejet=1/6πρvjet2tanα/22εRmax3(29)ε=h/RmaxWhere, h is the jet length; α is the angle at the jet head. The dissipation of energy is primarily due to the rapid change in flow state following the jet impact on the wall, which generates vortices and thereby dissipates a portion of the energy carried by the cavitation bubble. Taking Case 6 and Case 7 as representative examples, the Q criterion was used to extract the vortex core regions and thereby identify the near wall vortex structures generated after jet impact on the wall. The Q criterion is defined as follows:(30)Q=1/2Ω2-S2Where, Ω and S represent the antisymmetric and symmetric parts of the velocity gradient tensor, respectively, s−1. In the present study, a threshold of Q is 4.0 × 106s−2was used to identify vortex structures, and the near wall vortex core regions satisfying this criterion were extracted. It should be noted that the color scale in Fig. 10 denotes the velocity magnitude within the extracted vortex core regions, rather than the value of Q itself. This visualization enables simultaneous representation of the spatial location of the vortex structures and the flow intensity within them. The vorticity distribution in the near wall flow field at the moment of particle impact, induced by the bubble generated jet, is presented in Fig. 11.Fig. 11Vorticity distribution in the vicinity of the wall. Vorticity distribution in the vicinity of the wall. As shown in Fig. 11, when the cavitation bubble collapses and generates a jet that reaches the rigid wall, a large number of vortices are formed in the fluid medium within the bubble’s gaseous region. Compared with the particle impact velocities on the wall after being accelerated by the jet, as shown in Fig. 10, it is evident that the velocity of the fluid in the vortex regions is significantly higher than that of the particles. Consequently, the jet impact alters the flow state near the wall, dissipating a substantial portion of the energy carried by the cavitation bubble, which gradually dissipates with the subsequent fluid motion. After bubble collapse, the energy associated with shock waves and vortex related dissipation is indirectly estimated as the residual energy obtained by subtracting the jet kinetic energy and particle kinetic energy from the initial bubble energy, as expressed by:(31)Eres=Eb-Ejet+EpWhere, Eb is the energy carried by the bubble, J; Ejet is the jet kinetic energy, J. It should be emphasized that this residual term comprises not only the energy associated with shock waves, but also vortex related dissipation, viscous dissipation, and other unresolved energy conversion contributions. Therefore, it should not be interpreted as a direct measure of vortex dissipation alone. To provide a more direct characterization of the development of vortex structures in the flow field after jet impact on the wall, the vorticity volume integral was introduced in the revised manuscript as a post processing indicator of vortex intensity. Vorticity is defined as the curl of the velocity field:(32)ω=ωx2+ωy2+ωz2 To quantify the overall intensity of the vortex structures within the computational domain after jet impact on the wall, the vorticity volume integral is defined as follows:(33)Iω=∫ΩVωdVWhere, ΩV is the characteristic region near the wall where jet impact and vortex structure development occur, m3. It should be noted that the vorticity volume integral is not directly equivalent to the vortex dissipation energy. A rigorous evaluation of viscous dissipation energy would require further integration of the velocity gradient tensor and the viscous dissipation rate. Based on the numerical simulation results, the vorticity volume integrals over the selected near wall vortex region at the bubble collapse moment for Case 6 and Case 7 are listed in Table 6. To ensure that the vorticity integral results obtained at different stages are comparable, three representative instants were selected for analysis. The instant t1 denotes the stage before jet formation, at which the jet velocity has not yet increased appreciably and satisfies vjet < 0.1 vjetmax. The instant t2 is defined as the moment when the jet front first impinges on the rigid wall. The instant t3 represents the post impact stage, during which the near wall vortex structures are fully developed, and are taken as the time when the vorticity integral reaches, or is close to, its peak value. Comparison of the vorticity integrals at t1, t2 and t3 enables the evolution of the flow rotational intensity during jet formation, wall impact, and subsequent vortex development to be quantified.Table 6Variation of vortex intensity in the jet region.CaseStage of bubble evolutionIω (m3/s)6t18.11 × 10−6t22.31 × 10−5t34.99 × 10−67t17.28 × 10−6t22.30 × 10−5t31.70 × 10−6 Table 6Variation of vortex intensity in the jet region.CaseStage of bubble evolutionIω (m3/s)6t18.11 × 10−6t22.31 × 10−5t34.99 × 10−67t17.28 × 10−6t22.30 × 10−5t31.70 × 10−6 Variation of vortex intensity in the jet region. In this study, the vorticity integral is introduced to provide a direct quantitative indicator of the flow field structure, thereby verifying the enhancement of vortex structures and the associated energy dissipation trend after jet impact on the wall. Eq. (31), in contrast, is retained as an indirect estimate of the residual energy of the bubble. Discussion and outlook In the present study, experimental observations and coupled VOF and DPM simulations were performed to investigate the motion characteristics of particle clusters subjected to the action of a cavitation bubble near a rigid wall. Emphasis was placed on the evolution of particle cluster velocity, kinetic energy, and wall impact intensity during bubble expansion, collapse, and jet impingement. The results demonstrate that the high-speed jet induced by near wall bubble collapse constitutes a primary driving mechanism for the motion of particle clusters toward the wall. The dynamic response of the particle cluster is governed by the combined effects of particle size, the dimensionless distance parameter λ, the maximum bubble radius Rmax, and the jet impact velocity. Although the experimental observations and numerical predictions show good agreement in terms of particle motion direction, velocity evolution trend, and order of magnitude, the bubble particle cluster wall interaction involves strong transient behavior, multiscale flow structures, and multiphase coupling. Consequently, the present study still has several limitations that should be addressed in future work. Future investigations may focus on the following aspects:(1)Incorporation of gas liquid phase change mass transfer. During the evolution of spark induced cavitation bubbles, the gaseous medium inside the bubble, vapor pressure, non-condensable gas, and interfacial evaporation and condensation processes may influence the collapse time, pressure peak, and shock wave intensity. Future work should incorporate a gas liquid phase change mass transfer model, together with the evolution of internal bubble pressure and shock wave propagation, to quantitatively evaluate the influence of phase change on bubble collapse intensity and the corresponding dynamic response of particle clusters. It should be noted that the present VOF model does not explicitly account for gas–liquid phase transition. Therefore, while the model can reasonably reproduce bubble evolution and jet formation validated by experiments, its ability to predict collapse-induced pressure peaks and shock-wave intensity during the final collapse stage remains limited.(2)Incorporation of interparticle collisions. Under conditions of relatively high local particle concentration, interparticle collisions may influence the lateral dispersion of particle clusters, local accumulation, individual particle trajectories, and the dispersion of particle velocities near the wall. For the local dynamics of high concentration particle clusters, future work should introduce particle collision models or CFD DEM coupled approaches to more accurately describe interparticle momentum exchange, collision induced energy dissipation, and the evolution of local particle concentration.(3)Measurement and incorporation of particle wall restitution coefficients. Future studies may extract the normal and tangential instantaneous particle velocities before and after wall impact from high-speed image sequences to determine the particle wall restitution coefficients. These coefficients can then be introduced into DPM wall boundary conditions or CFD DEM coupled models, thereby enabling a more detailed investigation of the coupling among cavitation jet impact, particle rebound, secondary impact, and cumulative wall damage.(4)The conclusions obtained in this study are mainly applicable to the ranges of particle size, bubble scale, and bubble–rigid wall distance investigated in the present experiments and numerical simulations. However, when the particle size, bubble radius, or wall distance extends beyond the ranges investigated in this study, the coupling between the particles and the transient flow field, as well as the bubble-collapse and jet characteristics, may change. Therefore, the quantitative results obtained herein should not be directly extrapolated to all operating conditions. Future studies will extend the ranges of particle size, bubble scale, and wall distance and establish more systematic parameter relationships based on key dimensionless parameters, such as the Stokes number and dimensionless wall distance, to further validate and extend the applicability of the present conclusions. (1)Incorporation of gas liquid phase change mass transfer. During the evolution of spark induced cavitation bubbles, the gaseous medium inside the bubble, vapor pressure, non-condensable gas, and interfacial evaporation and condensation processes may influence the collapse time, pressure peak, and shock wave intensity. Future work should incorporate a gas liquid phase change mass transfer model, together with the evolution of internal bubble pressure and shock wave propagation, to quantitatively evaluate the influence of phase change on bubble collapse intensity and the corresponding dynamic response of particle clusters. It should be noted that the present VOF model does not explicitly account for gas–liquid phase transition. Therefore, while the model can reasonably reproduce bubble evolution and jet formation validated by experiments, its ability to predict collapse-induced pressure peaks and shock-wave intensity during the final collapse stage remains limited. Incorporation of gas liquid phase change mass transfer. During the evolution of spark induced cavitation bubbles, the gaseous medium inside the bubble, vapor pressure, non-condensable gas, and interfacial evaporation and condensation processes may influence the collapse time, pressure peak, and shock wave intensity. Future work should incorporate a gas liquid phase change mass transfer model, together with the evolution of internal bubble pressure and shock wave propagation, to quantitatively evaluate the influence of phase change on bubble collapse intensity and the corresponding dynamic response of particle clusters. It should be noted that the present VOF model does not explicitly account for gas–liquid phase transition. Therefore, while the model can reasonably reproduce bubble evolution and jet formation validated by experiments, its ability to predict collapse-induced pressure peaks and shock-wave intensity during the final collapse stage remains limited. (2)Incorporation of interparticle collisions. Under conditions of relatively high local particle concentration, interparticle collisions may influence the lateral dispersion of particle clusters, local accumulation, individual particle trajectories, and the dispersion of particle velocities near the wall. For the local dynamics of high concentration particle clusters, future work should introduce particle collision models or CFD DEM coupled approaches to more accurately describe interparticle momentum exchange, collision induced energy dissipation, and the evolution of local particle concentration. Incorporation of interparticle collisions. Under conditions of relatively high local particle concentration, interparticle collisions may influence the lateral dispersion of particle clusters, local accumulation, individual particle trajectories, and the dispersion of particle velocities near the wall. For the local dynamics of high concentration particle clusters, future work should introduce particle collision models or CFD DEM coupled approaches to more accurately describe interparticle momentum exchange, collision induced energy dissipation, and the evolution of local particle concentration. (3)Measurement and incorporation of particle wall restitution coefficients. Future studies may extract the normal and tangential instantaneous particle velocities before and after wall impact from high-speed image sequences to determine the particle wall restitution coefficients. These coefficients can then be introduced into DPM wall boundary conditions or CFD DEM coupled models, thereby enabling a more detailed investigation of the coupling among cavitation jet impact, particle rebound, secondary impact, and cumulative wall damage. Measurement and incorporation of particle wall restitution coefficients. Future studies may extract the normal and tangential instantaneous particle velocities before and after wall impact from high-speed image sequences to determine the particle wall restitution coefficients. These coefficients can then be introduced into DPM wall boundary conditions or CFD DEM coupled models, thereby enabling a more detailed investigation of the coupling among cavitation jet impact, particle rebound, secondary impact, and cumulative wall damage. (4)The conclusions obtained in this study are mainly applicable to the ranges of particle size, bubble scale, and bubble–rigid wall distance investigated in the present experiments and numerical simulations. However, when the particle size, bubble radius, or wall distance extends beyond the ranges investigated in this study, the coupling between the particles and the transient flow field, as well as the bubble-collapse and jet characteristics, may change. Therefore, the quantitative results obtained herein should not be directly extrapolated to all operating conditions. Future studies will extend the ranges of particle size, bubble scale, and wall distance and establish more systematic parameter relationships based on key dimensionless parameters, such as the Stokes number and dimensionless wall distance, to further validate and extend the applicability of the present conclusions. The conclusions obtained in this study are mainly applicable to the ranges of particle size, bubble scale, and bubble–rigid wall distance investigated in the present experiments and numerical simulations. However, when the particle size, bubble radius, or wall distance extends beyond the ranges investigated in this study, the coupling between the particles and the transient flow field, as well as the bubble-collapse and jet characteristics, may change. Therefore, the quantitative results obtained herein should not be directly extrapolated to all operating conditions. Future studies will extend the ranges of particle size, bubble scale, and wall distance and establish more systematic parameter relationships based on key dimensionless parameters, such as the Stokes number and dimensionless wall distance, to further validate and extend the applicability of the present conclusions. Conclusions In the present study, experimental observations and coupled VOF and DPM numerical simulations were performed to investigate the motion characteristics of particle clusters induced by a cavitation bubble near a rigid wall. The evolution of particle velocity, displacement, and wall impact kinetic energy during bubble expansion, collapse, and jet impingement was analyzed. The main conclusions are summarized as follows:(1)During the bubble expansion stage, the particle cluster is driven away from the bubble center by the outward motion of the expanding bubble wall. The maximum bubble radius Rmax is identified as the dominant parameter controlling particle acceleration during this stage. An increase in Rmax enhances the momentum transferred from the bubble to the surrounding liquid, thereby resulting in a higher peak velocity of the particle cluster. With increasing particle mass, inertial effects become more significant, and the particle response to the transient flow field induced by bubble expansion and collapse becomes weaker. This behavior is characterized by a smaller velocity increment and a more pronounced delay in particle motion.(2)The experimental observations and numerical predictions exhibit good consistency in terms of particle motion direction, velocity evolution, and velocity magnitude. The coupled VOF and DPM model can effectively reproduce the principal motion characteristics of particle clusters during bubble evolution. Moreover, it provides additional information on particle velocities, trajectories, and the flow field distribution around the bubble center in regions that are difficult to resolve experimentally because of bubble shielding.(3)During the bubble collapse stage, the rigid wall induces asymmetric bubble collapse and promotes the formation of a high-speed jet directed toward the wall. The wall impact kinetic energy of the particles is mainly governed by the jet impact velocity, particle size, and initial particle position. When the characteristic parameter λ < 1, the jet impact velocity is relatively high, and small particles are more readily accelerated to approximately 1/4 to 3/4 of the jet impact velocity, corresponding to a higher risk of wall impact. The energy carried by the bubble is mainly transferred through shock waves, jet kinetic energy, fluid motion, and particle kinetic energy. The near wall vortex structures formed after jet impact on the wall represent an important manifestation of bubble energy dissipation. (1)During the bubble expansion stage, the particle cluster is driven away from the bubble center by the outward motion of the expanding bubble wall. The maximum bubble radius Rmax is identified as the dominant parameter controlling particle acceleration during this stage. An increase in Rmax enhances the momentum transferred from the bubble to the surrounding liquid, thereby resulting in a higher peak velocity of the particle cluster. With increasing particle mass, inertial effects become more significant, and the particle response to the transient flow field induced by bubble expansion and collapse becomes weaker. This behavior is characterized by a smaller velocity increment and a more pronounced delay in particle motion. During the bubble expansion stage, the particle cluster is driven away from the bubble center by the outward motion of the expanding bubble wall. The maximum bubble radius Rmax is identified as the dominant parameter controlling particle acceleration during this stage. An increase in Rmax enhances the momentum transferred from the bubble to the surrounding liquid, thereby resulting in a higher peak velocity of the particle cluster. With increasing particle mass, inertial effects become more significant, and the particle response to the transient flow field induced by bubble expansion and collapse becomes weaker. This behavior is characterized by a smaller velocity increment and a more pronounced delay in particle motion. (2)The experimental observations and numerical predictions exhibit good consistency in terms of particle motion direction, velocity evolution, and velocity magnitude. The coupled VOF and DPM model can effectively reproduce the principal motion characteristics of particle clusters during bubble evolution. Moreover, it provides additional information on particle velocities, trajectories, and the flow field distribution around the bubble center in regions that are difficult to resolve experimentally because of bubble shielding. The experimental observations and numerical predictions exhibit good consistency in terms of particle motion direction, velocity evolution, and velocity magnitude. The coupled VOF and DPM model can effectively reproduce the principal motion characteristics of particle clusters during bubble evolution. Moreover, it provides additional information on particle velocities, trajectories, and the flow field distribution around the bubble center in regions that are difficult to resolve experimentally because of bubble shielding. (3)During the bubble collapse stage, the rigid wall induces asymmetric bubble collapse and promotes the formation of a high-speed jet directed toward the wall. The wall impact kinetic energy of the particles is mainly governed by the jet impact velocity, particle size, and initial particle position. When the characteristic parameter λ < 1, the jet impact velocity is relatively high, and small particles are more readily accelerated to approximately 1/4 to 3/4 of the jet impact velocity, corresponding to a higher risk of wall impact. The energy carried by the bubble is mainly transferred through shock waves, jet kinetic energy, fluid motion, and particle kinetic energy. The near wall vortex structures formed after jet impact on the wall represent an important manifestation of bubble energy dissipation. During the bubble collapse stage, the rigid wall induces asymmetric bubble collapse and promotes the formation of a high-speed jet directed toward the wall. The wall impact kinetic energy of the particles is mainly governed by the jet impact velocity, particle size, and initial particle position. When the characteristic parameter λ < 1, the jet impact velocity is relatively high, and small particles are more readily accelerated to approximately 1/4 to 3/4 of the jet impact velocity, corresponding to a higher risk of wall impact. The energy carried by the bubble is mainly transferred through shock waves, jet kinetic energy, fluid motion, and particle kinetic energy. The near wall vortex structures formed after jet impact on the wall represent an important manifestation of bubble energy dissipation. GlossaryNotationsbDer Waals coefficient (−)CSpeed of sound in the fluid medium (m/s)dpRadius of particles (mm)CDDrag coefficient (−)DDistance from the cavitation bubble center to the wall surface (mm)EbEnergy carried by the bubble(J)EjetEnergy carried by the jet(J)EpKinetic energy carried by the particles (J)EresResidual energy(J)FBBuoyant force on the buoyancy ball (N)FDDrag force on the buoyancy ball (N)FpForce due to the pressure gradient (N)FVMVirtual mass force on the buoyancy ball (N)hJet length(mm)HDistance between the two walls(mm)kPolytropic exponent (−)IωVorticity volume integral(m3/s)L0Distance between the particle center and the bubble center (mm)mpMass of the particle (kg)MpMass flow rate of the particles (kg/s)PRInternal pressure of the cavitation bubble near the wall (Pa)PR0Initial pressure inside the cavitation bubble (MPa)p0Ambient pressure (Pa)pvVapor pressure (Pa)Qvortex structures(s−2)RRadius of the bubble (mm)RmaxMaximum bubble radius (mm)Ry,mDistance between top of the bubble and wall(mm)R˙Wall velocity of cavitation bubble (m/s)R¨Wall acceleration of cavitation bubble (m/s2)SSymmetric parts of the velocity gradient tensor(s−1)SMMomentum exchange term (Kg·m/s)SteffEffective Stokes number(−)t1Stage before jet formation(s)t2The jet front first impinges on the rigid wall(s)t3Post impact stage(s)tjetJet action time(s)T*Normalization time (−)vVelocity of the fluid medium (m/s)vpVelocity of the particle (m/s)vpmaxThe peak particle velocity(m/s)vjetVelocity of the micro-jet (m/s)VpVolume of the particle (m3)ΔPPressure difference on the surface of the particle (Pa)ΔtTime step (μs)Greek lettersαAngle at the jet head (°)σSurface tension coefficient (N/m)βRatio of the peak particle velocity to the jet impact velocityεCharacteristic parameter of jet length(−)λDimensionless parameter between the bubble center and the wall (−)γd to the Rmax(−)ηH to the Rmax (−)κRy,m to the Rmax(−)τpParticle relaxation time(s)μViscosity of the liquid (mPa∙s)ωVorticity of the velocity field(s−1)ΩAntisymmetric parts of the velocity gradient tensor(s−1)ρDensity of the fluid medium(kg/m3)ρpDensity of the particle (kg/m3)AcronymsPRESTO!PREssure STaggering OptionVOFVolume of fluidDPMDiscrete phase model NotationsbDer Waals coefficient (−)CSpeed of sound in the fluid medium (m/s)dpRadius of particles (mm)CDDrag coefficient (−)DDistance from the cavitation bubble center to the wall surface (mm)EbEnergy carried by the bubble(J)EjetEnergy carried by the jet(J)EpKinetic energy carried by the particles (J)EresResidual energy(J)FBBuoyant force on the buoyancy ball (N)FDDrag force on the buoyancy ball (N)FpForce due to the pressure gradient (N)FVMVirtual mass force on the buoyancy ball (N)hJet length(mm)HDistance between the two walls(mm)kPolytropic exponent (−)IωVorticity volume integral(m3/s)L0Distance between the particle center and the bubble center (mm)mpMass of the particle (kg)MpMass flow rate of the particles (kg/s)PRInternal pressure of the cavitation bubble near the wall (Pa)PR0Initial pressure inside the cavitation bubble (MPa)p0Ambient pressure (Pa)pvVapor pressure (Pa)Qvortex structures(s−2)RRadius of the bubble (mm)RmaxMaximum bubble radius (mm)Ry,mDistance between top of the bubble and wall(mm)R˙Wall velocity of cavitation bubble (m/s)R¨Wall acceleration of cavitation bubble (m/s2)SSymmetric parts of the velocity gradient tensor(s−1)SMMomentum exchange term (Kg·m/s)SteffEffective Stokes number(−)t1Stage before jet formation(s)t2The jet front first impinges on the rigid wall(s)t3Post impact stage(s)tjetJet action time(s)T*Normalization time (−)vVelocity of the fluid medium (m/s)vpVelocity of the particle (m/s)vpmaxThe peak particle velocity(m/s)vjetVelocity of the micro-jet (m/s)VpVolume of the particle (m3)ΔPPressure difference on the surface of the particle (Pa)ΔtTime step (μs)Greek lettersαAngle at the jet head (°)σSurface tension coefficient (N/m)βRatio of the peak particle velocity to the jet impact velocityεCharacteristic parameter of jet length(−)λDimensionless parameter between the bubble center and the wall (−)γd to the Rmax(−)ηH to the Rmax (−)κRy,m to the Rmax(−)τpParticle relaxation time(s)μViscosity of the liquid (mPa∙s)ωVorticity of the velocity field(s−1)ΩAntisymmetric parts of the velocity gradient tensor(s−1)ρDensity of the fluid medium(kg/m3)ρpDensity of the particle (kg/m3)AcronymsPRESTO!PREssure STaggering OptionVOFVolume of fluidDPMDiscrete phase model CRediT authorship contribution statement Xiaobo Shen: Writing – original draft, Software, Methodology, Investigation, Data curation. Wei Han: Writing – review & editing, Supervision. Rennian Li: Validation, Resources. Shiqi Yang: Data curation. Qingduo Meng: Data curation. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.