Xiaoxiang
Li
,
Ying
Zhang
,
Weichen
Tang
,
Xin
Chen
and
Fei
Dong
*
Jiangsu University, School of Automotive and Traffic Engineering, 301 Xuefu Road, Zhenjiang, 212013, China. E-mail: lixiaoxiangujs@163.com; jsdxdf@163.com
First published on 10th April 2025
Particle–bubble flows are commonly found in industrial processes such as mineral flotation, catalytic reactors, and fluidized beds. This study reports the behavior of particles colliding with adhered bubbles on surfaces, resulting in their detachment. The effects of particle diameter, particle collision velocity, and bubble height on bubble dynamics were investigated. To connect the three factors together, the Weber number of the particles was fitted with the different heights of bubbles to derive a critical detachment curve for bubbles at varying particle diameters. The results indicate that the detachment height of bubbles is inversely proportional to the collision velocity and diameter of the particles. Additionally, among these three factors, the primary force influencing the bubble detachment time is the capillary force of the particles.
The interactions between particles and bubbles are typically categorized into collision, attachment, and detachment, which can alter the size, rise velocity, and residence time of the bubbles, thereby directly influencing the dynamics of gas–liquid flow.3,4 In order to study the detachment behavior of particles during the collision between bubbles and the air–water interface, Wang et al.5 placed hydrophobic particles at the bottom of a water tank and observed their behavior. Bubbles were generated at the capillary tube outlet and attached to the particles, forming mineralized bubbles that rose to the gas–liquid interface. The results showed that the differences in particle hydrophobicity led to inconsistent bubble velocity decay rates. Hydrophilic particles caused a slower decrease in bubble velocity, while hydrophobic particles led to a rapid decline in bubble speed. Mohan et al.6 studied the dynamics of rising bubbles containing particles under the influence of surfactants and turbulence, focusing on different bubble surface loads. The experimental data were compared with the behavior of particle-laden bubbles rising in quiescent water without surfactants. In quiescent water, the addition of methyl isobutyl carbinol surfactant reduced the average rise velocity of bubbles with a certain surface load, while turbulence consistently led to an increase in bubble rise velocity under all testing conditions. Wang et al.7 studied the effect of particle coverage on the dynamics of rising bubbles. This was done by calculating the buoyancy and drag forces acting on the bubbles, as well as the influence of particles on bubble behavior. The overall velocity of the bubbles is directly related to their aspect ratio and is independent of whether the bubbles are covered by particles. As the shape of the bubbles becomes more spherical, their velocity decreases. Xia et al.8 used a three-phase dynamics model of particle–bubble–liquid, this model includes the collision, attachment, and detachment behavior between particles and bubbles. Numerical results indicate that hydrophilic particles can enhance the rise velocity of bubbles, even exceeding that of bubbles rising in pure liquid. However, the attachment behavior significantly hinders bubble rise but enhances particle transport intensity. Particles driven by a higher bubble frequency exhibit greater lift height and a stronger dispersion effect.
However, the dynamic mechanisms underlying the interaction between particles and bubbles leading to bubble detachment are not yet fully understood, as the forces exerted on the bubbles by particle collisions are constantly changing, making the bubble separation process highly complex. Therefore, appropriate experimental setups and correct force theories are crucial for addressing this issue. Mohammadi et al.9 designed an experimental setup that allows particles to accurately collide with bubbles every time, using a vacuum pump to secure particles and releasing them at appropriate times for bubble–particle collision experiments. However, they only studied the different forces acting on particles of various sizes during their attachment process to the bubbles, and they did not study the detachment behavior of the bubbles. Chen et al.10 studied the influence of a single particle on the dynamics of bubble detachment by altering the properties of the liquid as well as the size and position of the particles. The results showed that when the particle is located in the neck region of the bubble, the detachment process is hindered; however, if the particle is positioned above or below the neck of the bubble, it facilitates the detachment process. Nonetheless, the description of the force conditions during bubble detachment is not very clear. Therefore, we can draw on their research theory from the study of Zhu et al.11 They observed the bubble detachment process at the gas nozzle exit using a jet-type microbubble generator device. They established a force balance model that takes into account the fluctuations in the flow field, analyzing the relationship between the bubble detachment size, growth parameters, and forces. This model considers various forces acting on the bubble and categorizes them into separating forces that promote bubble detachment and adhesive forces that hinder it. Wang et al.12 focused on studying the effect of varying DDA concentrations on the changes in particle contact angle and their influence on the dynamics of bubble neck rupture. By using a high-speed camera to capture the variation in the minimum radius of the bubble neck, they fitted the experimental data with a power law equation that describes the changes in the radius of the bubble neck, thereby determining the surface pressure and surface tension. Ultimately, we can integrate the experimental equipment of the aforementioned scholars and the theory of particle collision with bubbles as the research approach for this paper, which will ultimately allow us to establish our innovative method.
Most of the scholars in the previous section studied the bubble separation mechanisms in bubble-fluid systems or analyzed the sliding of particles on bubbles. However, the mechanism of particles colliding with bubbles and causing detachment remains unclear, and there is relatively little research on the regularity of particle collisions on bubbles in fluid. Therefore, the innovation of this paper lies in the study of the dynamic phenomena of particle collision and bubble separation, including the conditions required for bubble detachment and the detachment time, as well as the proposal of a critical separation curve for bubbles.
![]() | (1) |
First, a bubble of a certain height is generated using a syringe. Then, particles of a selected diameter are allowed to fall from the top of the tube without any initial velocity. The particles will collide with the bubble along the inner wall of the tube. The entire experiment is recorded by observing the relevant motion of the bubble as it detaches. The images of the particle collision on the bubble are captured using a high-speed camera and stored on a PC (Personal Computer), and then processed and analyzed using image analysis software.
When air is injected from the syringe, a bubble forms that gradually increases in height and width. To facilitate the measurement of the bubble size, the height of the bubble is used as the measurement parameter in this study. Measurement shows that the maximum bubble height before it detaches from the copper surface is 3.8 mm. Therefore, the experiment selects a maximum bubble height of 3.5 mm, with the minimum bubble height being 2.0 mm.
Scheme | u p (m s−1) | D p (mm) | H (mm) |
---|---|---|---|
1 | 0.1 | 1.4 | 3.5 |
2 | 0.2 | 1.4 | 3.5 |
3 | 0.3 | 1.4 | 3.5 |
4 | 0.2 | 1.4 | 2.5 |
5 | 0.2 | 1.4 | 3.0 |
6 | 0.2 | 1.0 | 3.5 |
7 | 0.2 | 1.2 | 3.5 |
![]() | (2) |
The buoyancy expression for particles in deionized water is
![]() | (3) |
According to the relative motion between bubbles and particles, the axial component of the drag prevents the collision process between particles and bubbles. The expression for the drag is14
![]() | (4) |
C D is the drag coefficient of the bubble.14 For rigid spherical particles without wall effects, the drag coefficient is determined using eqn (5):
![]() | (5) |
Pressure is expressed as15
fp = πRp2![]() | (6) |
![]() | (7) |
(fp)y = πRp2![]() ![]() | (8) |
The particle–bubble attachment to the interface forms a three-phase contact line. For the force analysis illustrated in Fig. 3, the capillary force can be expressed as15
fc = 2πRpσ(sin![]() ![]() | (9) |
![]() | (10) |
TPCL is the three-phase contact line, representing the distance at which particles come into contact with bubbles at the interface, measurable using image software. Here, RTPCL is half the length of the three-phase contact line, θ is the contact angle, and Rb is the radius of the bubble.
Regarding the bubble and the base, the solid–liquid surface tension is the horizontal force at the pore edge and naturally does not appear in the vertical capillary force. Thus, the capillary force at the pore edge is the solid–gas surface tension.16
![]() | (11) |
The expression for the buoyancy of the bubble is
![]() | (12) |
To facilitate analysis, this study focuses exclusively on the forces in the vertical direction. The forces under consideration in Fig. 3 are Fc, Fd, Fp, Fg and Fb.
![]() | ||
Fig. 4 Trajectory snapshots of a particle colliding with a bubble at different time points under velocities of (a) 0.1 m s−1, (b) 0.2 m s−1, and (c) 0.3 m s−1. |
To better analyze quantitatively the detachment characteristics of bubbles, Fig. 5 illustrates the forces acting on the particles in the vertical direction during the bubble compression stage. It can be observed that Fg and Fc are the dominant forces, but they act in opposite directions. Fc increases over time, while Fd decreases. According to eqn (10), Fc is proportional to RTPCL2. During the bubble compression phase, as the particle continually interacts with the bubble, TPCL lengthens until the particle's kinetic energy approaches zero, approximating a critical value. From eqn (4), Fd is proportional to up2, and in Fig. 4, particle velocity decreases over time during the compression phase. Fig. 5 examines the collision forces of different particle velocities on bubble detachment. Observing these graphs reveals that Fc is proportional to up; this relationship arises because greater up results in higher particle kinetic energy and more frequent particle-bubble contacts, thus increasing TPCL. Due to the small and relatively constant values of Fp, it does not appear in the graph.
![]() | ||
Fig. 5 Forces acting on the bubble during particle-bubble collisions at velocities of (a) 0.1 m s−1, (b) 0.2 m s−1, and (c) 0.3 m s−1 in the compression stage. |
Wang et al.12 studied the influence of different DDA (dodecylamine) concentrations and particle contact angles on the variation of bubble necks. In this study, to investigate whether different particle velocities affect the pinch-off time of bubbles, particles with different collision velocities with the bubble were selected, and the results are shown in Fig. 6. By analyzing the images using software, the length of R at each moment was measured. Upon comparing the images, the detachment times of the bubbles were similar, regardless of the selected particle velocities. However, at higher particle velocities, R started to change earlier. As can be seen from Fig. 5, the capillary force Fc increases with the particle velocity. Therefore, the greater the capillary force of the particles, the earlier the bubble begins to be affected and generates the detachment behavior.
![]() | ||
Fig. 7 Snapshots of particle trajectories at different time points during collisions with bubbles of (a) 2.5 mm, (b) 3.0 mm, and (c) 3.5 mm heights. |
In Fig. 8, Fc exhibit a trend of initially increasing and then decreasing. Combining this with Fig. 8, it is observed that the increase in Fc is due to the continuous lengthening of the TPCL as particles interact with the bubble. However, Fc is also influenced by φ; as particles begin sliding along the bubble surface, φ increases. When φ reaches , indicating that particles are approaching the equator of the bubble, Fc becomes zero. Moreover, as the bubble size increases, the entire process of Fc variation takes longer, suggesting that the time needed by particles to move from the top to the equator of the bubble also increases. Meanwhile, Fd remains proportional to up2 overall, showing a trend of initially increasing and then decreasing. Notably, the point at which Fd starts decreasing coincides with the peak of Fc, indicating that particles begin decelerating and reduced contact between particles and the bubble, indirectly verifying the accuracy of Fc.
![]() | ||
Fig. 8 The forces acting on bubbles with heights of (a) 2.5 mm, (b) 3.0 mm, and (c) 3.5 mm during the particle-bubble collision process in the bubble compression phase. |
In Fig. 9, the variations in neck radius of these three types of bubbles are observed. It is noted that for H = 2.5 mm, R exhibits oscillatory fluctuations and returns to its original length at t = 24 ms, corresponding to the final image in Fig. 6(a). This indicates that particles have moved close to the bottom surface, hence the bubble is no longer influenced by particles. For the other two bubble scenarios, although they detach as well, the time required for detachment is inversely proportional to the height of the bubble. This is primarily attributed to the combined effects of the buoyancy force and the bottom capillary force
.
is related to D0, and since capillary force at the bottom does not change, the overall force decreases with larger bubbles. Therefore, larger bubbles detach more quickly due to the reduced combined forces acting on them.
![]() | ||
Fig. 10 Trajectory snapshots at different time points for particles with diameters of (a) 1.0 mm, (b) 1.2 mm, and (c) 1.4 mm during collision with a bubble. |
In Fig. 11(b) and (c), Fc is significantly greater than in (a), yet the difference between Fc values in the former two cases is not substantial. This is because larger particles, despite having greater kinetic energy upon collision and thereby increasing contact with the bubble to some extent, experience increased upward forces Fd and Fb as well as downward force Fg as Dp increases. Consequently, the variation in Fc may not necessarily correlate linearly with Dp. The trend in Fd still shows an initial increase followed by a decrease, with the key distinction being that initial Fd is directly proportional to Dp.
![]() | ||
Fig. 11 Forces acting on the particle during the collision of particles with diameters of (a) 1.0 mm, (b) 1.2 mm, and (c) 1.4 mm in the bubble compression phase. |
Fig. 12 illustrates the time taken for bubbles to detach after collision with particles of different diameters, corresponding to the instantaneous detachment moments of bubbles in Fig. 9. This demonstrates that the detachment time of bubbles is inversely proportional to Dp. This is because larger particles generate a greater capillary forc Fc when colliding with bubbles, as shown in Fig. 11, which results in a shorter bubble detachment time.
![]() | (13) |
Fig. 13 represents different motion state diagrams of particles colliding with bubbles. The horizontal axis denotes different bubble sizes, while the vertical axis indicates the particle Weber number at 0 ms. The three graphs in Fig. 13 depict various states of particle–bubble interaction for different Dp values. Experimental data indicate substantial agreement between the boundaries of the two distinct collision behaviors and scaling laws. Our scaling laws suggest that, under identical bubble conditions at the same height, smaller diameter particles exhibit higher inertia forces compared to larger diameter particles, resulting in equivalent behavioral effects. Therefore, extensive experiments were conducted to record the critical We numbers for particle collision leading to bubble detachment or non-detachment, followed by curve fitting of these critical detachment Weber numbers for each bubble at different heights. Smaller particles require a larger We (up to a maximum of 15) to detach bubbles in deionized water, whereas larger particles can detach bubbles with a smaller We (as low as 0.2) when colliding with the same bubbles. The data results indicate that particles colliding with bubbles at higher H require a larger We for detachment, whereas the opposite is true for collisions with bubbles at smaller H. Furthermore, larger Dp particles require a smaller We for bubble detachment, while smaller Dp particles require a larger We for detachment.
![]() | ||
Fig. 13 The critical detachment diagram of the bubble illustrates the relationship between the Weber number and bubble height for particles with diameters of (a) 1.0 mm, (b) 1.2 mm, and (c) 1.4 mm. |
(1) Bubble detachment requires the colliding particles to reach a certain kinetic energy, and the conditions for detachment of bubbles of different sizes are provided in the Weber number diagram. The detachment process can be divided into the stages of bubble compression and rising based on the motion and force conditions of particles and bubbles. During the compression stage, the kinetic energy of the particles is transferred to the bubble. In the rising stage, the time taken for R to change from 1000 μm to 0 μm determines the time required for bubble detachment.
(2) The experimental results indicate that the time for the bubbles to detach is inversely proportional to Dp and H. However, an interesting observation is that under the conditions of this study (up = 0.1, 0.2, and 0.3 m s−1), the detachment time is not directly related to up. When particles collide with larger bubbles at higher up, although the generated Fc has increased but this only leads to a rapid change in the bubble neck during the initial rising period (18–25 ms). Finally, the detachment process begins to lag in the subsequent time. The maximum value of Fc is also directly proportional to Dp, as larger Dp has greater kinetic energy, resulting in a larger RTPCL during collision. In the H factor, the duration of Fc is proportional to H, because more time is required for particles to slide over larger bubbles.
(3) The critical curve relating H and We in bubble detachment is presented in this paper. In the three different Dp factors, the critical detachment curves of the bubbles are all similar inverse proportional curves. Although the We number required for particles with different Dp values to detach from bubbles at H = 2 mm is relatively high (ranging from 8 to 14), the variation in We needed for detachment from bubbles at H = 3.5 mm is minimal (ranging from 0.5 to 1).
Footnote |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5cp00938c |
This journal is © the Owner Societies 2025 |