Air-Facilitated Three-Phase Contact Formation at Hydrophobic Solid

Haiman Wang , Jia Liu , Weihua He , Youpeng Qu , Da Li , Qing Jiang , and Yujie Feng ... Shiwei Wang , Jifeng Guo , Longfei Tang , Huan He , Xiuxiang ...
0 downloads 0 Views 756KB Size
Download PDF
Langmuir 2007, 23, 549-557

549

Air-Facilitated Three-Phase Contact Formation at Hydrophobic Solid Surfaces under Dynamic Conditions M. Krasowska,† R. Krastev,‡ M. Rogalski,§ and K. Malysa*,† Institute of Catalysis and Surface Chemistry, Polish Academy of Sciences, 30-239 Cracow, Poland, Max Planck Institute fu¨r Kolloid- und Grenzflachenforschung, D-14424 Golm/Potsdam, Germany, and Lab Thermodynam Milieux Polyphases, UniVersite´ de Metz, Technopoˆ le, Metz 2000, 57078 Metz Cedex 3, France ReceiVed August 7, 2006. In Final Form: September 29, 2006 The paper presents results documenting the mechanism of facilitation of the three-phase contact (TPC) formation due to gas entrapped during immersion of hydrophobic (Teflon) plates into distilled water and n-octanol solutions. Collisions, bouncing, the time scale of the TPC formation, and bubble attachment to Teflon plates of different surface roughness were studied using a high-speed camera. Processes occurring during the microscopic wetting film formation at the Teflon plates were monitored using the microinterferometric method (Scheludko-Exerowa cell). A strong relation between the time necessary to form a stable TPC and the roughness of the Teflon was observed. The higher the Teflon roughness was the shorter the time for the TPC formation. This effect can be attributed to two factors: (i) local differences in the thickness of the thinning intervening liquid layer (quicker attainment of rupture thickness at pillars of rough surface) and/or (ii) the presence of gas at the hydrophobic surface. Experimental findings, that (i) prolongation of the plate immersion time resulted in quicker TPC formation, (ii) white irregular and disappearing spots (air pockets) were recorded during the wetting film formation, and (iii) high n-octanol concentration caused prolongation of the time of the TPC formation, show that attachment (TPC formation) of the colliding bubble to hydrophobic surfaces was facilitated by air entrapped at the Teflon plates (and re-distributed) during their immersion into water phase. Thus, on collision instead of solid/gas wetting liquid film a thin gas/liquid/gas foam film was formed which facilitated the TPC formation.

Introduction Formation of the three-phase contact solid/liquid/gas involves always wetting or de-wetting of a solid surface. De-wetting means rupture of the thin liquid film separating the solid surface from the gas phase. Stability of thin liquid films is determined by the interfacial interaction forces. According to the DerjaguinLandau-Verwey-Overbeek (DLVO) theory1,2 the film stability is due to the balance between the repulsive and attractive components of the disjoining pressure Π.3,4 The disjoining pressure is a sum of the electrostatic double layer repulsive component (ΠEL) having a range from a few up to 100 nm; the van der Waals component (ΠVW) ranging in about 1 nm; and the non-DLVO components (Πnon-DLVO) not accounted in the classical DLVO theory:

Π ) ΠVW + Πel + Πnon-DLVO

(1)

The non-DLVO components of the disjoining pressure5 are considered to be related mainly to structural and steric interactions. Many authors reported the existence of long-range attractive interactions between hydrophobic bodies immersed into solution6-8 * Corresponding author. E-mail: [email protected]. † Polish Academy of Sciences. ‡ Max Planck Institute fu ¨ r Kolloid- und Grenzflachenforschung. § Universite ´ de Metz. (1) Derjaguin, B. Theory of Colloids and Thin Films; Consultants Bureau: New York, 1989. (2) Verwey, E. J. W.; Overbeek, J. T. G. Theory of the Stability of Lyophobic Colloids; Dover Publications: Minneola, NY, 1948. (3) Scheludko, A. AdV. Colloid Interface Sci. 1967, 1, 391. (4) Exerowa, D.; Kruglyakov, P. Foams and Foam Films; Elsevier: Amsterdam, 1998. (5) Israelachvili, J. Intermolecular and Surface Forces; Academic Press: London, 1991. (6) Blake, T. D.; Kitchener, J. A. J. Chem. Soc., Faraday Trans. 1 1972, 68, 1435-1442.

in the past 2 decades of the 20th century. The origin of these “long-range hydrophobic interactions (forces)” was not clear, but the concept was rather widely spread. However, in recent years it started to be discussed and documented9-13 that the concept of the existence of additional “hydrophobic forces” was not correct. It was shown using the tapping mode atomic force microscopy (TM AFM)10,13 that the measured additional “hydrophobic interactions” were in reality due to a coalescence of sub-microscopic air bubbles (nanobubbles) attached to the hydrophobic surfaces immersed into aqueous solution. A number of papers documenting the existence of nanobubbles at hydrophobic surfaces is increasing rapidly,10-17 but the controversy over the hydrophobic interactions is not solved yet. Recently, it was reported18 that prior to ethanol-water exchange, nanobubbles were not detected by TM AFM on freshly cleaved highly ordered pyrolitic graphite (HOPG) and on octadedecyltrichlorosilane (OTS) silicon in water. However, after water-ethanol-water exchange there were nanobubbles, which remained at hydrophobic surfaces for hours. Physical properties of nanobubbles were reported18 to be comparable to those of macroscopic bubbles, (7) Israelachvili, J. N.; Pashley, R. M. J. Colloid Interface Sci. 1984, 98, 500. (8) Yoon, R. H. Int. J. Miner. Proc. 2000, 58, 129. (9) Parker, J. L.; Claesson, P. M.; Attard, P. J. Phys. Chem. 1994, 98, 8468. (10) Ishida, N.; Inoue, T.; Miyahara, M.; Higashitani, K. Langmuir 2000, 16, 6377. (11) Schulze, H. J.; Sto¨ckelheuber, K. W.; Wenger, A. Colloids Surf., A 2001, 192, 61. (12) Tyrrel, J. W. G.; Attard, P. Langmuir 2002, 18, 160. (13) Attard, P. AdV. Colloid Interface Sci. 2003, 104, 75. (14) Carambassis, A.; Jonker, L. C.; Attard, P.; Rutland, M. W. Phys. ReV. Lett. 1998, 80, 5357. (15) Nguyen, A. V.; Nalaskowski, J.; Miller, J. D.; Butt, H.-J. Int. J. Miner. Proc. 2003, 72, 215. (16) Steitz,R.; Gutberlet, T.; Hauss, T.; Klo¨sgen, B.; Krastev, R.; Schemmel, S.; Simonsen, A. C.; Findenegg, G. H. Langmuir 2003, 19, 2409. (17) Yang, J.; Duan, J.; Fornasiero, D.; Ralston, J. J. Phys. Chem. B 2003, 107, 6139.

10.1021/la062320n CCC: $37.00 © 2007 American Chemical Society Published on Web 11/22/2006

550 Langmuir, Vol. 23, No. 2, 2007

besides the contact angle values, which were “... surprisingly large for nanobubbles...” and their long life was attributed to smaller curvature due to high contact angle values. Nevertheless, it needs to be taken into consideration that gas presence (submicroscopic bubbles) can be of significant importance in the stability of various dispersed systems. Rupture of a wetting film and formation of a three-phase contact (gas/liquid/solid) is a necessary condition for attachment of hydrophobic grains to gas bubbles. Flotation separation of mineral grains takes advantage of their differences in wettability (hydrophilic/hydrophobic properties). Lack of attachment of the hydrophilic and the attachment of hydrophobic grains to bubbles and formation of stable bubble-grain aggregates are the basis of flotation separation. During the bubble collision with hydrophobic solid the following processes have to take place: (i) thinning of the intervening liquid film between the bubble and solid, (ii) rupture of the liquid film (if the film reaches the critical thickness) and formation of a “hole” of the three-phase contact perimeter, and (iii) expansion of the hole resulting in formation of the threephase contact perimeter (area) ensuring stability of the attachment. Thus, stability of the wetting film and kinetics of spreading of the three-phase contact (de-wetting of the solid surface) are of crucial importance for the attachment. It has been rather common understanding that if the solid surface is hydrophobic enough (contact angle θ > 90°), then every collision should result in a bubble attachment. However, it was demonstrated recently19,20 that a colliding bubble could bounce a few times without attachment even in the case of such model hydrophobic surface as Teflon. This paper presents data documenting the importance of the gas presence for the three-phase contact (TPC) formation and attachment of the colliding bubble to the hydrophobic solid surface (Teflon) under dynamic conditions. The mechanism of facilitation of the colliding bubble attachment to the hydrophobic solids due to presence and re-distribution of air entrapped at the Teflon plates immersed into water is explained. Data on the time scale of the TPC formation and attachment of the bubble during its collision with Teflon plates of different roughness in distilled water and n-octanol solutions are presented. Processes occurring before and during formation of single-model microscopic wetting films at the Teflon plates of different roughness are also shown.

Krasowska et al.

Figure 1. Experimental setup for monitoring the bubble collisions.

Figure 2. Schematic of the cell for wetting film measurements.

Experimental Section Bubble Collisions. A high-speed camera (Weinberger, SpeedCam 512+) of 1182 frames/s was used to monitor velocity variations, bouncing, and time of attachment of the bubble colliding with the Teflon plates positioned horizontally inside a glass column filled with water. The experimental setup is presented in Figure 1, and its main elements were as follows: (i) a square glass column (50 × 50 mm), (ii) a capillary (0.075 mm) connected with a syringe pump for the bubble release in a controlled manner, (iii) the SpeedCam camera, and (iv) a PC with frame grabber and digital image analysis software. The Teflon plates were positioned horizontally ca. 300 mm from the capillary orifice. The movies recorded were transformed into BMP pictures and analyzed frame by frame using the SigmaScanPro Image Analysis Software. The equivalent diameter of the bubble detaching from the capillary was 1.48 ( 0.03 mm in distilled water.21 Further details of the experimental procedure and image analysis of the recordings are described elsewhere.19,20 Single Microscopic Model Wetting Films. Microscopic singlemodel wetting films were studied using a modified version of the (18) Zhang H. X.; Maeda N.; Craig V. S. J. Langmuir 2006, 22, 5025. (19) Krasowska, M.; Malysa, K. Physicochem. Probl. Miner. Process. 2005, 39, 21. (20) Krasowska, M.; Malysa, K. Int. J. Miner. Process., in press (online June 12, 2006).

thin film pressure balance (TFPB) method for thin liquid film studies.4,23,24 The method is based on the original porous plate experimental cell proposed by Scheludko-Exerowa. A porous glass plate (normal pore size 0.9-1.4 µm, Robu Glass Filter-Gerate GmbH, Berlin, Germany) fused with a glass capillary was kept inside a closed vessel exposing the end of the capillary to the atmosphere. There was a hole of 0.8 mm diameter in the porous glass plate. The studied solution was sucked through the porous glass plate prior to the experiment to remove traces of air in the micropores and to wet them homogeneously. The solution height was maintained in the capillary to provide sufficient bulk solution in contact with the hole where the film was formed. The studied Teflon plate was fixed onto the porous plate (see Figure 2), thus, covering the hole. In such configuration a wetting liquid film separating the Teflon plate (on the top) from the liquid/gas interface (on the bottom) could be formed. The closed airtight vessel was thermostated, and the desired constant gas pressure was adjusted. The gas pressure in the vessel was gradually, slowly, and smoothly increased in a controlled way. The pressure in the vessel forced the liquid from the wetting layer to thin until its critical rupture thickness was obtained. The film drainage (21) Krzan, M.; Malysa, K. Colloids Surf., A 2002, 2007, 279. (22) Malysa, K.; Krasowska, M.; Krzan, M. AdV. Colloid Interface Sci. 2005, 114/115, 205.

Air-Facilitated Three-Phase Contact Formation

Langmuir, Vol. 23, No. 2, 2007 551

Table 1. Characteristics of the Teflon Surfaces

process was monitored, and the images were recorded by an inverted video enhanced microscopy system. The setup is equipped with an optical system which allows the intensity reflected from the film light to be measured. Thus, using the microinterferometric method of Scheludko,3 the film thickness could be obtained. The porous plate was thoroughly washed by extensive sucking of Milli-Q water through the cell followed by drying and heating to 500 °C for at least 5 hsthus, burning out any organic contaminations. Materials. Three Teflon plates of different roughness were used in the experiments. The plates were prepared from the same piece of the commercial Teflon, and their surface roughness was modified using the abrasive papers of different grid numbers. The plate with the most smooth surface (called “Teflon I”) was obtained by polishing with abrasive paper No. 2400 and next the diamond grinding DPpaste of 1/4 µm. The surface of the plate called “Teflon III” was, as received from the mechanical shop. The third plate called “Teflon V” had the surface roughened using abrasive paper No. 100. Optical microscope photographs were taken for characterization of surface roughness of the Teflon plates. For the smoothest Teflon surface, the atomic force microscopy (AFM) pictures (Nanoscope III, tapping mode) were obtained, as well. Microscopic photographs of the Teflon surfaces together with information about cavity sizes, as estimated from the photograph image analysis, are presented in Table 1. The AFM picture of the smoothest Teflon surface (Teflon I) is presented in Figure 3. Indeed cavities of the size of at least 200 nm are observed there, and this is in reasonable agreement with data obtained from the optical microscope photoanalysis (Table 1). All Teflon plates were cleaned using a diluted chromic mixture and next carefully washed-out with 4-fold distilled water. Then, the plates were further cleaned by boiling in 4-fold distilled water for 1 h. The Teflon plates were sonicated for 15 min in 3% Mucasol solution and rinsed with Milli-Q water in the case of single-model film experiments. The plates were sonicated again in acetone, rinsed with Mili-Q water, and stored in Mili-Q water in a closed vessel. Immediately before the measurement the Teflon plates were dried in nitrogen atmosphere to avoid even microsize dust grains which

could act as a “defects” inside the wetting film, causing its instability and rupture. High-purity distilled water (4-fold distilled or Milli-Q water) was used in the experiment. n-Octanol was a commercial reagent of high purity. The experiments were carried out at room temperature (22 ( 2°).

Results and Discussion Bubble Collisions and Attachment. Table 2 presents three sequences of photographs showing phenomena occurring during bubble collisions with various Teflon plates in distilled water. Clear influence of the Teflon roughness on the collision process can be seen there. In the case of the highest surface roughness (Teflon V), the bubble was attached during the first collision, while in the case of the smoothest one (Teflon I), the bubble bounced four times prior to the attachment. The time needed for the bubble attachment varied by over an order of magnitude from ca. 2.5 to over 80 ms, as can be noted from the time values marked in the photograph sequences. Note, how rapid are phenomena occurring during the collisions and how quick variations of the bubble shape are after the collision (see photographs for Teflon I). It is rather commonly expected that if the solid surface is hydrophobic, the first collision should result in the bubble attachment. However, as seen in Table 2, the bubble colliding with the Teflon I and Teflon III surface was not attached during the first collision but bounced backward. Since all these experiments were carried out under identical conditions (experimental procedure, distilled water of identical quality, identical method of the Teflon plates cleaning, identical size and rising velocity of the bubbles, the same setup), therefore, the results presented in Table 2 indicate the crucial importance of the surface roughness to the time scale of the bubble attachment. In the case of the Teflon I surface a few “approach-bounce” cycles were observed prior to the three-phase contact (gas/liquid/

552 Langmuir, Vol. 23, No. 2, 2007

Figure 3. AFM image (3-D) of the “Teflon I”.

solid) formation and the bubble attachment. The velocity of each subsequent bubble approach to the plate was lower as a result of the energy dissipation.22,25,26 It is worthy to underline here that similar bouncing (oscillation) phenomena were reported by Nguyen et al.27 during their experiments with solid particles falling onto the plane liquid/air interface. Quantitative results showing variations of the bubble local velocity, i.e., the bubble velocity at a given point, during collisions with the Teflon I, Teflon III, and Teflon V plates are presented in Figure 4. The bubble approach velocity was constant (bubble terminal velocity) prior to the first collision and identical in all experiments (ca. 35 cm/s). On collision the bubble was stopped and bounced backward within 2-3 ms. Velocity variations at first collision were identical for the three Teflon surfaces, but then significant differences occurred. The values of the time of the TPC formation (tTPC), i.e., the time periods from a moment of the first collision to the moment of the TPC formation and the bubble attachment, are marked in Figure 4, and they were 2.5, 39.7, and 83.6 ms for Teflon V, Teflon III, and Teflon I plates, respectively. These tTPC values are the averages of 20-40 repeated experiments, for reducing the data scatter observed in the case of Teflon I and Teflon III surfaces. The most probable reasons responsible for scatter of the attachment time measurements are the following: (i) reproducibility of measurements under dynamic conditions is always lower than under static or quasi-static conditions, (ii) homogeneity of most of solid surfaces is rather low, (iii) “spots” (local areas) of the solid surface “hit” by the colliding bubble are always random so their properties can differ significantly, and (iv) there were cavities and asperities (pillars) of different sizes at the plates surface. It needs to be added here that the tTPC formation values, marked in Figure 4, indicate the time of formation of a stable enough TPC to prevent the bubble detachment. Bouncing of the bubble from interface is a result of two competing processes:25,26 (i) the thinning of the liquid layer between the bubble and solid surface and (ii) the increase of the free energy of the system resulting from the increase of the bubble surface area. The free energy of the system increases, at the expense of the kinetic energy of the rising bubble, and the bubble bounces backward if the thinning liquid layer did not reach a critical rupture thickness. Careful analysis of photographs presented in Table 2 shows that indeed for Teflon III there was recorded a situation where the wetting film was ruptured but a perimeter of the three-phase contact was too small to prevent (23) Exerowa, D.; Kolarov, T.; Khristov, K. Colloids Surf. 1987, 22, 171. (24) Bergeron, V.; Radke, C. J. Langmuir 1992, 8, 3020. (25) Chesters, A. K.; Hofman, G. Appl. Sci. Res. 1982, 38, 353.

Krasowska et al.

detachment of the bouncing bubble. In Table 2, the photograph at t ) 4.2 ms for Teflon III shows that there was a “necking” formed between the Teflon III surface and the detaching bubble. Then, the necking was ruptured (photograph at t ) 5.1 ms), and a satellite-bubble attached to the Teflon III surface was left. During the second approach the bouncing bubble collided with the satellite bubble left and was attached to the Teflon III surface. It can be also observed in Table 2 that the diameter of the threephase contact formed was enlarging rapidly. Thus, the rupture of the separating liquid layer (wetting film) is a necessary, but not sufficient, condition to prevent the bubble bouncing from the liquid/solid interface,22 i.e., for the bubble attachment to hydrophobic solid surface. Leja pointed out28 in his discussion of formation of the stable bubble-grain aggregates that the following substeps need to happen: (i) syneresis and rupture of the liquid film, (ii) formation of a hole of the three-phase contact, and (iii) expansion of the hole and formation of the perimeter of the three-phase contact, ensuring that detachment will not occur. We discuss possible reasons of such significant differences in time of the TPC formation and the bubble attachment to Teflon surfaces of different roughness. As seen in Figure 4 a course of the bubble local velocity variations during its first collision was identical. Differentiation started to be noted only on further stages, where either the bubble was attached or bounced backward. This is an indication that during the first collision the bubble velocity variations were determined only by the bubble shape variations and hydrodynamics of thinning of the intervening liquid layer (here distilled water in all cases). Thus, it shows that properties of the solid surface (roughness and nonhomogeneities) affected the bubble velocity variations only when the rupture of the intervening thin liquid film occurred; i.e., there were formed local areas (“holes”) of the three-phase contact. It needs to be underlined that in the case of Teflon V the attachment occurred always during the first collision, and the reproducibility of measurements was excellent. With decreasing roughness of the Teflon surface the time needed for the bubble attachment was significantly prolonged (from ca. 2.5 ms to 40 and 84 ms) and reproducibility of the measurements was significantly lowered. Thus, in our opinion there are two main factors responsible for such significant difference in the time of the TPC formation and bubble attachment to Teflon surfaces of different roughness: (i) local differences in thickness of the thinning intervening liquid layer and (ii) the presence of gas at the hydrophobic surface, because with increasing surface roughness more air could be entrapped there. When the solid surface is not smooth, there are crevices (pillars) of different sizes (height and width) in different local areas (see Table 1 and Figure 3). Thus, the thinning liquid layer can have locally (on pillars) much smaller thickness than an average one, and the probability of reaching there a critical thickness of rupture is much higher. So, at such areas the points (holes) of the TPC can be formed during the collision time, i.e., during the time of 2-3 ms. It seems reasonable to assume that with increasing roughness the size (height and width) of such asperities is increasing (see Table 1) and so the higher is the probability of formation of a long enough TPC perimeter to arrest the bouncing of the bubble. However, with increasing surface roughness, there can be also more air adhered to hydrophobic Teflon surface. It is worthwhile to recall here that the now rather commonly accepted idea explaining so-called long-range “hydrophobic forces” describes the effect as coalescence of nanobubbles attached to the hydrophobic surfaces immersed into aqueous solution. Further experimental evidence (26) Krzan, M.; Lunkenheimer, K.; Malysa, K. Langmuir 2003, 19, 6586.

Air-Facilitated Three-Phase Contact Formation

Langmuir, Vol. 23, No. 2, 2007 553

Table 2. Sequences of Photographs Showing Bubble Shape Pulsation and Bouncing during Subsequent Collisions with “Teflon I”, “Teflon III”, and “Teflon V” in Distilled Watera

a

Bubble equivalent diameter was 1.48 mm.

Figure 4. Variations of the bubble local velocity during collision with “Teflon I” (black circles), “Teflon III” (white squares), and “Teflon V” (gray diamonds) surfaces in distilled water.

documenting importance of the air presence for kinetics of the bubble attachment is presented below. Figure 5 presents the collision numbers during which the attachment to the Teflon III surface occurred as a function of the immersion time. The immersion time is defined as the time interval from a moment of the plate immersion into water until the moment of the first collision of the rising bubble with the solid surface. The attachment of the bubble occurred during the third or fourth collision only, at immersion times shorter than 10 s (see Figure 5). Prolongation of the plate immersion time, until 20 s and longer, resulted in a quicker attachment. Some colliding bubbles happened to be attached even during the first collisions. Prolongation of the immersion time means that there is a longer time available for re-distribution of the air entrapped during the Teflon plate immersion into water. Moreover, it also means a

higher probability of air nucleation (adsorption) at the hydrophobic Teflon surface. Figure 6 presents additional data documenting that there indeed were sub-microscopic air bubbles adhered to the hydrophobic Teflon surface. There are presented the bubble local velocity variations during collisions with the Teflon V surface in distilled water and in two n-octanol (OctOH) solutions of different concentrations. The bubble approach velocity was identical (ca. 15 cm/s) for both n-octanol concentrations, and it was significantly lower than in distilled water. This lowering in the bubble velocity in n-octanol solutions is due to the presence of surfactant which immobilizes the gas/liquid interface.29-31 It was shown recently21 that there exists minimum adsorption coverage needed for full immobilization of the bubble interface and that above this coverage the bubble terminal velocity starts to be practically independent of the surfactant concentration. The bubble was always attached to the Teflon V surface during the first collision in distilled water, indicating that the wetting film was unstable and ruptured very quickly. Despite much lower velocity and kinetic energy of the bubble motion a similar instability was observed for the intervening liquid film in 3 × 10-5 M OctOH solution. This proves that the origin of the film instability cannot be found in the different kinetic energy of the approaching bubble. The attachment did not happen during the first collision in 6 × 10-4 M OctOH solution, despite that the bubble approach velocity was identical to that of the low concentrated n-octanol solutions. Thus, at high n-octanol concentration (6 × 10-4 M) the time for the TPC formation was prolonged from 2.5 ms (in distilled water and 3 × 10-5 M OctOH solution) to ca. 18 ms. This shift in the (27) Nguyen, A. V.; Schulze, H. J.; Stechemesser, H.; Zobel, G. Int. J. Miner. Process. 1997, 150, 113. (28) Leja, J. Surface Chemistry of Froth Flotation; Plenum Press: New York and London, 1982. (29) Frumkin, A. N.; Levich, V. G. Zh. Phys. Chim. 1947, 21, 1183.

554 Langmuir, Vol. 23, No. 2, 2007

Krasowska et al. Table 3. Sequences of Photographs Illustrating Phenomena Occurring during Formation of the Wetting Films at the “Teflon I” and “Teflon V” Surfaces

Figure 5. Influence of the immersion time of the “Teflon III” plate into distilled water on the collision number during which the bubble was attached to the plate surface.

Figure 6. Variations of the bubble local velocity with time during collision with the “Teflon V” plate in distilled water (gray diamonds), 0.000 03 M n-octanol (black triangles), and 0.0006 M n-octanol (white circles) solutions.

time for the TPC formation is indeed meaningful because in the case of Teflon V the attachment in distilled water occurred always during the first collision and the reproducibility of measurements was excellent. In the discussion above of the possible reasons of facilitation of the TPC formation by increasing surface roughness of the hydrophobic surface the hypothesis was put forward that it might also be a result of an increased amount of air (sub microscopic bubbles) adhering to a more rough Teflon surface. If the hypothesis is correct, that would mean that in reality the thin intervening film was not the wetting (nonsymmetric) film between the solid surface and liquid/air interface of the colliding air bubble but a symmetric foam film between bubbles. It is well-known that the stability of the symmetric foam film increases with surfactant concentration.4 Generally the prolongation of the time of the TPC formation at high n-octanol concentration could be discussed as a reasult of n-octanol adsorption on both liquid/gas and liquid/ solid interfaces. In flotation aliphatic alcohols are used as frothers, which are added for lowering the bubble velocity and ensuring formation of a froth layer, due to their preferential adsorption at the liquid/gas interface. Nevertheless, n-octanol adsorption at the Teflon surface cannot be disregarded completely. To check it, we measured the apparent contact angle values at Teflon for

different n-octanol concentrations. It was found that there was practically no influence of increasing OctOH concentration on the apparent contact angle values. There was noted a very weak trend of a decrease in apparent contact angle values, but it was within experimental scatter and most probably was due to lowering of the solution surface tension. Therefore, we think that OctOH adsorption at the liquid/solid interface was of no or only minor importance for the TPC formation. Thus, when the bubble collided with the Teflon V surface, the intervening liquid layer was a foam film formed between the approaching bubble and air adhered to the Teflon V surface. This intervening liquid film (foam film, we believe) started to drain until its critical thickness of rupture was reached. At high n-octanol concentration the foam film formed was more stable,4,32 and therefore the bubble bounced and attachment occurred only during the second collision. In distilled water and in 3 × 10-5 M n-octanol solution the time (30) Levich, V. G. Physicochemical Hydrodymanics; Prentice-Hall: Englewood Clifts, NJ, 1962. (31) Dukhin, S. S.; Kretzschmar, G.; Miller, R. Dynamics of Adsorption at Liquid Interfaces. Theory, Experiments, Application; Elsevier: Amsterdam, 1995. (32) Jachimska, B.; Warszynski, P.; Malysa, K. Colloids Surf., A 2001, 192, 177.

Air-Facilitated Three-Phase Contact Formation

Langmuir, Vol. 23, No. 2, 2007 555

Figure 7. Schematic illustration of air entrapment and re-distribution during the Teflon plate immersion into water.

needed for the drainage of the film, its rupture, and TPC formation was 2.5 ms, while in 6 × 10-4 M n-octanol solution it was 18 ms. High n-octanol adsorption coverage at both interfaces of the foam film formed prevented the film rupture. In our opinion the results presented above supply convincing evidence that there were indeed sub microscopic air bubbles attached to hydrophobic solid surfaces immersed into the water phase and they were a factor of crucial importance for the TPC formation under dynamic conditions and attachment of the colliding bubble. Investigations on formation and rupture of the wetting film at Teflon surfaces supplied additional data confirming the correctness of this hypothesis. Thin Liquid Films at Teflon Surfaces. The wetting films at Teflon plates were formed by “sucking off” slowly an excess of solution from the porous plate by applying external pressure. First, thick films were formed, they drain, and at the end of this process the thin wetting films were formed. Photographs taken during the drainage of the films at the Teflon I and Teflon V surfaces are presented in Table 3. The moments of the film lifetime at which the photographs were shot are also given. The moment of the film rupture was chosen as a zero point of the time scale because this is a well-defined reproducible point. Thus, the preceding photographs show pictures of the films at identical time intervals prior to the moments of their rupture. The striking difference between the pictures of the wetting films on the Teflon I and Teflon V surfaces is seen for long times prior to the films’ rupture (see photographs for times t ) 66, 68, 79, and 88 s). At the rough Teflon V surface there are clearly seen white irregular areas (spots). As the film drains, these white irregular spots become smaller and smaller and finally they disappear. In the optical geometry of the experiment, these spots could be thinner parts of the film created as a result of the formation of nonhomogeneous parts of the film or to be air cavities entrapped between the film and the support. We assume that these spots were formed from air entrapped in the large cavities of the rough Teflon V surface. During the film thinning because of the liquid flow out of the film, the air was removed and the white irregular spots disappeared. The fact that such white spots were not recorded at the smooth Teflon I surface does not mean that there was not air entrapped. The roughness of the Teflon I surface is below 1

µm. It is probable that there could be air entrapped inside these cavities also, but the dimension of such spots is below the resolution of the optical microscope and therefore they were not recorded. Mechanism of the Bubble Attachment due to Air Entrapped in Cavities of Hydrophobic Surfaces. Four sets of experimental findings show the important role of the gas presence at the hydrophobic surface: (i) prolongation the plate immersion time, over ca. 20 s, resulted in a quicker attachment of the colliding bubble (during first or second instead of fourth or fifth collision); (ii) the time of attachment was shortened with the increasing roughness of the Teflon surface; (iii) white irregular spots (air pockets) were recorded during wetting films formation at the Teflon surface; (iv) the attachment occurred always during the first collision in the case of the very rough Teflon V surface, but in high concentration of n-octanol there was bouncing and attachment occurring only during the second collision. This was recorded to take place despite the fact that the bubble approach velocity was thereby over 50% lower. Moreover, when a satellite bubble was left at the Teflon surface during the first collision, then during the second one the attachment occurred if the bubble “hit” the surface spot containing this satellite bubble. On the basis of this experimental evidence the following mechanism of facilitation by air entrapped the three-phase contact formation and the bubble attachment to the hydrophobic surface under dynamic conditions is postulated. When the Teflon plate is being immersed into water, then air is entrapped inside its surface cavities (pores, holes, and scratches). The higher the surface roughness was the more air was entrapped inside the cavities. Figure 7 presents schematically the processes occurring during and after immersion of the Teflon plate into water. It seems reasonable to assume that at the moment of the plate immersion and horizontal positioning there was approximately a planar interface between water and air entrapped inside the cavities. However, this is definitely a nonequilibrium position of the interface. The equilibrium state of the interface between water and the pockets of air entrapped inside Teflon cavities is determined by the Young-Laplace equation:

∆P ) 2(σ/r) cos θ

(2)

556 Langmuir, Vol. 23, No. 2, 2007

where θ is the contact angle, σ is the surface tension, r is the radius of the cavity (pore), and ∆P is the pressure difference between the gas (in the “pocket“) and water phases. Equilibrium contact angles (advancing) at the Teflon plates studied were within 100-128°, so cos θ < 0 and ∆P < 0. It shows that there should be a convex meniscus at equilibrium and the excessive amounts of air entrapped inside the cavities had to be transferred elsewhere. There were two possible “directions”: either dissolution in water or spreading over the Teflon surface. Dissolution in water was rather unlikely because water was saturated with air. Thus, spreading over the Teflon surface seems to be the most probable. Since the cavities were overfilled, either air films or sub-microsopic bubbles were formed at the prolonged immersion time. When the colliding bubble approached such Teflon surface having air films and/or sub-microsopic bubbles at its pillars (asperities), i.e., at the outmost elements of the surface, then a rupture of the intervening liquid film and a coalescence of large colliding bubble with a number of sub-microsopic bubbles (already being in contact with the Teflon surface) or de-wetted areas (air films at Teflon asperities) occurred. In such a way a long enough perimeter (a large area) of the three-phase contact was obtained immediately during the collision. The bubble backward motion was arrested, and the bubble stayed attached to the Teflon surface. Thus, there was no need for rapid dewetting of the Teflon surface during the short time of ca. 2-3 ms of the bubble-solid collision, which is a necessary condition for formation of a stable TPC. Microscope photographs (metalographic microscope Nikon EPIPHOT 200) presented in Figure 8 provide additional evidence supporting the mechanism proposed. White, irregular shapes (air pockets) can be easily spotted on images of two areas (A and B) of the Teflon V surface immersed into water. It can be also noted on photographs B1 and B2, showing the same area of the surface at different times after the plate immersion into water, that shapes of these white spots were evolving with time. In our opinion these images document that indeed air can be entrapped during the Teflon plates’ immersion and the air entrapped can be re-distributed over the hydrophobic surface.

Concluding Remarks The formation of three-phase contact (TPC) solid/liquid/gas is a step which occurs in flotation processes prior to formation the bubble-grains aggregates. The stability of the thin liquid film formed between the solid phase and the approaching gas bubble governs the outcome and velocity of the process. It is a common understanding that if the solid surface is hydrophobic enough (unstable wetting films), then the collision should result in the bubble attachment. It was observed, however, that the bubble could bounce even a few times from a smooth hydrophobic (Teflon) surface prior to the TPC formation in distilled water and n-octanol solutions. As a result of the bouncing, the time needed for the TPC formation was 83.6 ms in the case of a smooth Teflon surface (roughness < 1 µm), while only ca. 2.5 ms for a much rougher surface (roughness g 50 µm). In our opinion the increasing roughness of the hydrophobic surface can influence the kinetics of the TPC formation in a 2-fold manner: (i) surface asperities can facilitate rupture of the wetting film due to faster attainment locally (pillars, asperities) of the rupture thickness, and/or (ii) surface defects (cavities, gaps, etc.) can act as sites where the gas can be trapped at the hydrophobic solid surface and the bubble attachment is facilitated by sub-microscopic air bubbles present there. The higher the

Krasowska et al.

Figure 8. Microscope photographs of the “Teflon V” surface immersed into water. Photographs A and B show different area of the surface. Photographs B1 and B2 show the same area but at different times from the plate immersion into water.

surface roughness was the more air was entrapped, and therefore the time of the bubble attachment was shortened. The following experimental findings indicate that facilitation of the TPC formation and the bubble attachment was due to the presence and re-distribution of air at the Teflon surfaces, namely: (i) prolongation the plate immersion time, over ca. 20 s, resulted in quicker attachment of the colliding bubble (during first or second instead of fourth or fifth collision); (ii) white irregular and disappearing spots (air pockets) were recorded during the wetting film formation; (iii) a satellite bubble left at the Teflon surface during the first collision facilitated the attachment during the second one; (iv) attachment occurred always during the first collision in the case of the very rough Teflon V surface, but in a high concentration of n-octanol there was bouncing and attachment that occurred only during the second collision. Thus, on collision instead of a solid/gas wetting liquid film there was a thin gas/liquid/gas foam film(s), between a number of submicrosopic bubbles (already being in contact with the Teflon surface) or de-wetted areas (air films at Teflon asperities), and the three-phase contact was obtained immediately after the foam film rupture. Thus, there was no need for rapid de-wetting during

Air-Facilitated Three-Phase Contact Formation

Langmuir, Vol. 23, No. 2, 2007 557

the short time of ca. 2-3 ms of the bubble-solid collision, and the bubble was attached immediately.

Potsdam, Germany, for financial support and hospitality to perform the experiments with microscopic wetting films.

Acknowledgment. M.K. acknowledges with gratitude the Max Planck Institute for Colloid and Surface Chemistry, Golm/

LA062320N