Study on the Dynamic Process of the Attachment of a Single Bubble to Rough Surfaces with Different Hydrophobicity

Abstract

A stable attachment between bubbles and solid particles is essential for flotation. Therefore, it is particularly necessary to study the dynamic process that occur in the attachment of bubbles to a solid surface. In this paper, Teflon and plexiglass plates were used as hydrophobic and hydrophilic solid surfaces, respectively, and solid surfaces with roughness of 0.018 μm to 5.33 μm were prepared by polishing with sandpaper. The influence of roughness on the dynamic process in bubble attachment to solid surfaces with different hydrophobicity was studied via a high-speed camera (750 frames per second). It was found that roughness played a positive role in the attachment to the hydrophobic Teflon surface while a negative role in the attachment to the hydrophilic plexiglass surface in terms of the bubble’s attachment to the solid plates. For a smooth Teflon surface, the formation of three–phase contact (i.e., the drainage of wetting film) took up to 95 ms, whereas for a very rough Teflon surface it took only 5 milliseconds. On the contrary, the high roughness prevented the bubble from attaching to the hydrophilic plexiglass surface. It was concluded that the increased roughness of Teflon plates was conductive to air entrapment in surface irregularities, inducing the rapid rupture of the wetting film on a very rough Teflon surface, while the increased roughness of the plexiglass plates resulted in “water pockets” in surface grooves, making the wetting film on the plexiglass surface stable.

1. Introduction

Flotation technology is one of the most effective methods for separating fine coal/mineral particles, and is based on the differences between the surface properties of the target minerals and gangue minerals [1,2,3]. The theory of the wetting contact angle developed by Thomas Young is an important criterion in flotation, which describes the dependence of the contact angle on specific surface energies [4]. According to Young’s equation, materials with a contact angle greater than 90° are considered to be hydrophobic, while materials with a contact angle less than 90° are considered to be hydrophilic. Bubbles are formed by the dispersion of air in a pulp with a strong hydrophobicity, and mainly play a role in carrying hydrophobic mineral particles to the foam layer during flotation. Essentially, flotation is the process of a solid–air interface displacing a solid–liquid interface to achieve a selective attachment between hydrophobic mineral particles and the bubble. Or rather, three–phase contact (TPC) forms as the wetting liquid film separating the bubble and target mineral particle ruptures, forming a stable bubble–particle aggregate [4,5,6]. Obviously, it is particularly necessary to focus on the process of bubbles’ attachment (TPC formation and expansion) to the solid surface.

High-speed cameras are an effective means of studying and analyzing the attachment process between particles/solid substrates and bubbles. One scenario is to use a high–speed camera to study the interaction of dropping particles with a stationary bubble. Numerous details of the sliding behavior of a particle onto a bubble can be monitored with fast video recordings. An intervening or wetting film forms as particles approach bubbles after collision, and the attachment of the particle to the bubble mainly depends on the stability of the wetting film. The attachment event between the particle and the bubble is mainly divided into three sub–processes (see Figure 1 [5,7]): (1) the thinning of the wetting film, (2) the wetting film rupture and TPC formation, and (3) the expansion of TPC. The study herein involves an important concept, the time taken for attachment to be induced in flotation, which is the total time required for the attachment of a particle to an air bubble [4,5,8]. A visualization setup called the CSIRO Milli–Timer apparatus was proposed by Verrelli et al. [9], which permitted the direct observation of the interaction and attachment between particles and a single bubble at the microscale. It was found that a number of particles jumped towards the bubble after a period of sliding and were subsequently attached to the bubble, i.e., achieving the particle–bubble attachment. Based on the jump–in events, the induction time required for the glass sphere to adhere to the bubble was determined to be from 6 ms to 70 ms. Moreover, Verrelli et al. [10] found that the regularity of particles had a significant impact on the particle–bubble attachment, and the induction periods of angular particles were an order of magnitude lower than those of spheres. A similar conclusion was obtained by Vaziri Hassas et al. [11], who indicated that the sharp edges of particles can stimulate the rupture of the wetting film during the very first moments of attachment, and the roughness can strengthen the existing bubble–particle interaction.

The other scenario is using fast video recordings to study the attachment of a single rising bubble to stationary solid substrates. Malysa et al. [12] explored the influence of surfactants on the rising velocity and dynamics of bubble collisions with liquid/solid interfaces with a high–speed camera. It indicated that in pure water the bubble colliding with the Teflon surface bounced a few times without attachment and simultaneously its shape pulsated rapidly with a frequency over 1000 Hz. The presence of a frother can significantly reduce the pulsation frequency of the bubble and facilitate the bubble’s attachment to a hydrophobic solid surface. Krasowska et al. [13,14,15] found that, with an increase in the roughness of a Teflon surface from below 1 μm to over 50 μm, the number of bounces was decreased from 5 to 0 for the bubble colliding with the Teflon surface, and the attachment time declined from over 80 ms to 3 ms. It was mainly attributed to the fact that higher surface roughness resulted in larger amounts of air entrapment during the Teflon plates’ immersion in water, and the presence of micro–bubbles over the Teflon surface facilitated the attachment of the colliding bubble. Zawala et al. [16,17,18] further explored the trajectory, shape deformation, added mass variation of bubbles and the energy balance during bubbles’ collision with the solid surfaces using a high–speed camera and numerical simulations. Moreover, Niecikowska et al. [19] studied the effect of the formation of dynamic adsorption layer on the time taken for the bubble to attach to a mica surface in n–dodecyltrimethylammonium bromide and n–hexadecyltrimethylammonium bromide solutions. It was concluded that the successful attachment of the bubble to the mica was attributed to the reversal of the electric charge of the bubble surface caused by the adsorption of a cationic surfactant onto the bubble’s surface, which led to electrostatically attractive interactions between the bubble and the mica, resulting in the wetting film rupturing. Furthermore, it was found that the bubble’s attachment to a quartz surface as well as the flotation of quartz were strongly affected by pH of hexylamine solutions [20].

The properties of the solid surface, such as its hydrophobic/hydrophilic character, roughness, surface electric charge, etc., are of crucial importance to the stability and rupture kinetics of the wetting film, and thereby for the attachment of bubbles, and so are the compositions of the surrounding continuous liquid and gas phase [21,22]. The present study focuses on the effects of the hydrophobicity and roughness of a solid surface on the dynamic process of bubble attachment to solid surface as well as the expansion of the TPC in order to understand the characteristics of flotation attachment process more comprehensively.

2. Experimental Materials and Methods

2.1. Materials

During the attachment of a rising bubble to a solid surface, the contact area between the bubble, whose diameter is within the order of millimeters, and the solid surface is small, and therefore the high uniformity of the surface of the solid is crucial to the reliability of experimental results. As a result, commercial plexiglass plates and polytetrafluoroethylene (Teflon) plates instead of the mineral plates were adopted as solid substrates in the present work. The measurement of the contact angles for the Teflon plates and plexiglass plates was conducted with a contact angle measurement instrument (Shanghai Zhongchen JC2000C1, Shanghai, China). The measurement indicated that the contact angle of the plexiglass plate was about 65.11°, while that of the Teflon plate was about 102.88°, as shown in Figure 2. Note that, in the present study, the plexiglass plate was considered as a hydrophilic surface, while the Teflon plates were used as a hydrophobic surface. To prepare solid plates with different surface roughness, the plexiglass and Teflon plates’ surfaces were fully polished using sandpaper with meshes of 120, 220, 400, 600, 1000, 1500, 2000 and 3000. The polished plexiglass plates were, respectively, recorded as P120, P220, P400, P600, P1000, P1500, P2000 and P3000, whereas the polished Teflon plates were recorded as T120, T220, T400, T600, T1000, T1500, T2000 and T3000, respectively. The surface topographies of the plexiglass and Teflon plates were characterized by an optical microscope and scanning electron microscope, while the surface roughness of the plates was quantitatively determined by a surface roughness measurer (Mitutoyo SJ–210, Mitutoyo Measuring Instruments (Shanghai) Co., Ltd, Shanghai, China). Figure 3 presents the surface topography of the polished Teflon plates, and Figure 4 gives the surface topography of the polished plexiglass plates. It can be intuitively observed that there is a significant difference in the surface roughness between the solid plates polished with different sandpapers. Figure 5 shows the roughness of the Teflon and plexiglass plates after polishing with sandpaper. It indicates that the surfaces of T3000 and P3000 were smooth, with a roughness of 0.102 and 0.089 μm, respectively. In contrast, T120 and P120 had rough surfaces, with a roughness of 5.331 and 2.824 μm, respectively.

2.2. Experiment on Collision and Attachment of Air Bubbles to Solid Surfaces

The experimental device used for observing the collision with and attachment of a bubble to the solid surfaces is shown in Figure 6. It mainly consists of a square column, a high–speed camera (i–SPEED 3, OLYMPUS, Tokyo, Japan) with a monitor, a non–stroboscopic light source, a microsyringe (GS1200, Gilmont, Gilmer, TX, USA) and a flat–headed needle with an inner diameter of 0.91 mm. The microsyringe has an accuracy of 0.002 mL, and it was used to generate an air bubble with a diameter of about 3.5 ± 0.1 mm using a flat–headed needle in the aqueous phase. In a typical experiment, the plexiglass or Teflon plates were horizontally fixed and immersed in deionized water, as shown in Figure 6. The high–speed camera was used to monitor the collision and attachment of bubbles to the plexiglass/Teflon plates, and the frame rate of fast video recordings was fixed at 750 fps. Additionally, it should be pointed out that the lower surface of plexiglass or Teflon plate is about 200 mm away from the point of the bubble formation (i.e., the needle). The software Image Pro Plus 6.0 was adopted for the data processing. Specifically, the video obtained by the high–speed camera was first converted into sequential images with the software i–SPEED Suite. The automatic tracking function in Image Pro Plus 6.0 was used to capture and record the trajectory of the air bubble, and therefore, the instantaneous velocity of bubble was attained. The moment when TPC is formed can be easily and clearly determined in a slow playback of the video using the software Image Pro Plus, and therefore the frame corresponding to this moment is determined as the one when the TPC occurs. Moreover, the measurement function in Image Pro Plus 6.0 was adopted to analyze the TPC line length of the bubble attachment to the solid surface.

3. Results and Discussion

3.1. Effect of Roughness on Bubble Attachment to Hydrophobic Teflon Plates

The collision and attachment of rising bubbles to Teflon plates with a roughness from 0.102 μm to 5.331 μm are shown in Figure 7, Figure 8, Figure 9 and Figure 10, and the instantaneous velocities of bubbles during their interaction with the Teflon plates are shown in Figure 11. It should be noted that the moment of the first collision of the bubble with the solid plate is recorded as t = 0 ms in the data processing. It can be seen in Figure 7, Figure 8, Figure 9 and Figure 10 that the deformation of bubbles was almost the same during their approach towards the Teflon surfaces prior to the first collision with Teflon plates, and the terminal velocity of bubbles during their approaching to Teflon surfaces was about 30 ± 1.5 cm/s. The deformation of a bubble rising in water is mainly caused by the drag forces exerted by the continuous fluid (continuous water phase) on the bubble. The drag forces encountered by the bubble are affected by the bubble’s radius and rising velocity, and the kinematic viscosity and density of water, etc. In the present study, the above parameters were kept as consistent as possible, and therefore a similar deformation of bubbles was observed during their approach towards the Teflon surfaces. However, a significant difference in the bubble motion and attachment behavior after its collision with the hydrophobic Teflon surface with different roughness was observed, as shown in Figure 7, Figure 8, Figure 9 and Figure 10.

As shown in Figure 7, the “collision–bounce” cycles of the bubble, the amplitude of bounce and the degree of deformation of the bubble can be easily and clearly observed through the video obtained by the high–speed camera recording the dynamic process of the attachment between the bubble and the solid plates. Specifically, the moments when the bubble collides with and bounces from the solid surface can be clearly determined by a slow playback of the video using the software Image Pro Plus, and therefore the frames corresponding to these moments can be determined. For the smooth T3000 surface with a roughness of 0.102 μm, the bubble bounced two times before TPC formation. During “collision–bounce” cycles, the velocity and the “amplitude” of the bounce and the degree of deformation in the bubble gradually decreased due to the dissipation of kinetic energy associated with the bubble’s motion. During the third collision with the T3000 surface, permanent contact of the bubble with T3000 surface was attained. Then, the bubble experienced a process of “oscillation in vertical direction” lasting about 21.33 ms at the T3000 surface (see Figure 7 and Figure 11), and during this event, most of the residual kinetic energy of the bubble was consumed. At approximately the 95th ms, the wetting film between the bubble and T3000 was ruptured and TPC nucleus or hole was formed, followed by the expansion of the TPC which lasted for about 31 ms. Finally, the bubble adhered to the T3000 surface, which was reflected by a sudden variation in the instantaneous velocity of the bubble between 95 and 125 ms in Figure 11. In contrast, for the T1500 surface with a roughness of 0.316 μm, the bubble bounced one time before TPC formation (see Figure 8). The bubble was captured by the T1500 surface in the case of the bubble with large residual kinetic energy (with a maximum oscillation velocity exceeding 10 cm/s) after the second collision, as shown in Figure 8 and Figure 11. It was found that the rupture of the wetting film (i.e., TPC formation) occurred at an earlier time (the 53rd ms), which was 42 ms ahead of that for the T3000 surface. At the T600 and T220 surfaces with a surface roughness of 0.729 μm and 5.331 μm, respectively, permanent contact with the bubble was immediately obtained during the first collision (no bubble bounce), as shown in Figure 9 and Figure 10. For the T600 surface, TPC was formed in the 15th ms after a process of bubble oscillation which lasted for about 10 ms. At the T220 surface, TPC was formed in the 5th ms and TPC spreading was completed at the 25th ms.

Figure 12 illustrates the characteristics of the bubbles’ attachment to the Teflon plates. As shown in Figure 12, the timescales of bubble attachment to Teflon plates strongly depended on the surface roughness, and the time point of TPC formation at the Teflon surfaces was gradually advanced with the increased surface roughness. Additionally, it was found that it took approximately 31 ms to finish TPC expansion at the smooth T3000 surface. With the increased roughness, the time required for TPC expansion on the T1500, T600 and T220 surfaces was decreased to 29, 25 and 19 ms, respectively. On the contrary, the final lengths of the TPC formed on the Teflon surfaces was increased from 3.96 mm to 5.85 mm with the increase in roughness (see Figure 12), improving the strength of the attachment between the bubbles and the Teflon surface. It should be noted that, in this study, the length of the TPC was determined as the horizontal projection of the TPC (i.e., TPC diameter) formed by the bubble adhering to the solid surface. In summary, for the hydrophobic Teflon plates, the increase in roughness greatly accelerated the thinning and rupture of the wetting film formed on the Teflon plate, and thereby promoted the formation and expansion of TPC, enhancing the attachment of the bubbles to the Teflon surface.

3.2. Effect of Roughness on Bubble Attachment to Hydrophilic Plexiglass Plates

Figure 13, Figure 14, Figure 15 and Figure 16 show the collision and attachment of rising bubbles to plexiglass plates with different roughness, and Figure 17 presents instantaneous velocities of bubbles during their interaction with plexiglass plates. Compared with Teflon plates with highly hydrophobic surfaces, roughness showed quite contrary influence on the attachment of bubbles to the plexiglass plates with hydrophilic surfaces. It can be seen in Figure 13 that at the smooth P3000 surface with a roughness of 0.089 μm, the bubble bounced two times during its attachment to the plexiglass plate. After the bubble’s third collision with the P3000 surface, it experienced a process of “oscillation” lasting about 230 ms at the P3000 surface (see Figure 13 and Figure 17). At the 353rd ms, the wetting film ruptured and TPC was formed, followed by the expansion of the TPC lasting about 50 ms (see Figure 17). Compared with the instantaneous velocities of the bubble during its interaction with the T3000 surface, it can be concluded that the variation in the amplitude of the instantaneous velocities of the bubble during TPC formation and expansion on the P3000 surface was much smaller. It is suggested that the net attraction between the bubble and the T3000 surface was much greater than that between the bubble and the P3000 surface. In addition, it became increasingly difficult for bubbles to attach themselves to plexiglass plates as the surface roughness increased. For the P1000 surface with a roughness of 0.405 μm, the bubble bounced three times before TPC formation. A process of “oscillation” lasting longer (up to 428 ms) for the bubble was observed after the bubble’s fourth collision with the P1000 surface, and TPC was formed at about the 531st ms, much later than that at the P3000 surface. Moreover, no obvious variation in the instantaneous velocity of bubble was observed during TPC formation and expansion, i.e., the dynamic process of TPC spreading of the bubble on the P1000 surface was almost imperceptible (see Figure 14 and Figure 17). As illustrated in Figure 18, which gives the characteristics of the attachment of bubbles to the P3000 and P1000 surfaces, the final TPC line of the bubble adhering to P3000 surface was slightly longer than that of the bubble adhering to P1000 surface. The time required for the TPC to spread across at P1000 surface was about 90 ms, much greater than that for the P3000 surface. In contrast, for the rough P400 and P220 surfaces, no formation of TPC was observed after the permanent contact of the bubble with the plexiglass plates. In other words, the intervening film between the bubble and P400/P220 surface did not rupture, and no attachment of the bubble to the P400/P220 surface occurred. In summary, increased roughness enhanced the stability of the wetting film on the hydrophilic plexiglass surface, which is not conducive to the wetting film’s drainage and thereby the bubble’s attachment. Obviously, roughness played a positive role in the attachment of bubbles to the hydrophobic Teflon surface while a negative role in the attachment of bubbles to the hydrophilic plexiglass surface.

The contrasting role of roughness in the attachment to the hydrophobic Teflon surface and hydrophilic plexiglass surface is interesting and worth discussing. It is well known that hydrophobic surfaces have a natural affinity for air. When the dry hydrophobic Teflon plates are immersed in aqueous solutions, the increased roughness of Teflon surface, which have more surface irregularities (grooves, scratches and gaps), improves the probability that some microscopic air bubbles will be entrapped at the Teflon surface [15,23], as seen in Figure 3 and Figure 10. Specifically, the higher surface roughness on the Teflon plates provides more and bigger cavities, grooves and gaps, making rough Teflon surface to entrap more gas when immersed into aqueous solutions. The cavitation effect caused by this was believed to be the origin of long–range (up to hundreds of nanometers) hydrophobic forces between air bubbles and Teflon plates immersed in aqueous solutions [24,25,26]. These long–range hydrophobic forces are rather important for the stability and drainage kinetics of the wetting film [27]. It is precisely the hydrophobic interaction that facilitates the thinning process of the wetting film during the attachment of rising bubbles to Teflon plates, so TPC is formed (i.e., the wetting film ruptures) at an earlier time point in time for the rough T1500/T600/T220 surfaces compared with the smooth T3000 surface. Also, there is a greater final TPC length for the T1500/T600/T220 surfaces, meaning a stronger attachment of bubbles to rough Teflon plates. Additionally, the hydrophobic attraction can also be attributed to the bridging of interfacial submicron (nano–) bubbles which were easily formed at rough hydrophobic solid surface [15,28,29,30,31]. Despite there still being a debate about whether nanobubbles can be thermodynamically stable, their existance at hydrophobic surfaces immersed into aqueous solutions has been well established by experiments by many researchers [30,31,32,33]. Meanwhile, both nanobubbles and trapped air could significantly increase the boundary slip near the rough hydrophobic solid surfaces immersed in aqueous solutions, which affects the drainage kinetics of wetting film [15,16,28,29,34,35,36]. On the contrary, an increased roughness makes hydrophilic plexiglass surfaces more hydrophilic. The hydrophilic plexiglass surfaces have a natural affinity for water, and therefore the plexiglass surfaces are easily covered by water molecules when immersed in aqueous solutions. Moreover, for rough plexiglass surfaces immersed into aqueous solutions, “water pockets” rather than “gas pockets” are formed in the surface irregularities, which makes the covered wetting film more thermodynamically stable. Consequently, the increased roughness was unconducive to the drainage of the wetting film at the hydrophilic plexiglass surface, and thereby no rupture of the wetting film and no attachment of bubbles were observed on the rough P400 and P220 surfaces, as shown in Figure 15 and Figure 16. The inspiration for this study is the fact that, during flotation, the hydrophobicity of highly hydrophobic minerals can be further improved by increasing their surface roughness. On the contrary, for hydrophilic low–rank coal, efforts could be made to obtain a smooth surface in order to restrain its wetting behavior in flotation pulp. However, controlling for the surface roughness of coal particles in flotation in practice is unrealistic. From a practical perspective, the wetting of a coal surface with water molecules could be restrained by shortening the prewetting and conditioning times for coal particles during low–rank coal flotation.

4. Conclusions

This study focused on the effect of roughness on the dynamic process in a bubble’s attachment to hydrophobic Teflon plates and hydrophilic plexiglass plates. It clearly indicated that the role played by surface roughness in a bubble’s attachment to hydrophobic Teflon and hydrophilic plexiglass plates is almost opposite. At the smooth T3000 surface, the drainage of wetting film took up to 95 ms to form TPC. Moreover, the TPC expansion took about 31 ms to attain a final TPC line of 3.96 mm in length. In comparison, for the very rough T220 surface, the wetting film rupture only took 5 ms and a rather larger TPC line (5.85 mm) was obtained. This suggests the bubble’s attachment was strongly promoted by the increasing roughness of the hydrophobic Teflon surface. However, the opposite was true for the bubble’s attachment to the plexiglass plates as the surface roughness increased. The high roughness of the P400 and P220 surfaces prevented the bubble from adhering itself to the plexiglass surfaces. The dramatic difference in the bubble’s attachment behavior at Teflon and plexiglass surfaces resulted from roughness could be attributed to the air entrapment generated due to the hydrophobic Teflon surface’s irregularities, which was believed to be the dominant factor inducing the wetting film’s drainage and rupture. For the rough plexiglass surfaces immersed in aqueous solutions, “water pockets” rather than “gas pockets” formed in the grooves of the hydrophilic surface, which made the wetting film more thermodynamically stable.