ARTICLE pubs.acs.org/Langmuir
Measurement of the Surface Tension of Liquid Marbles Tina Arbatan and Wei Shen* Australian Pulp and Paper Institute, Department of Chemical Engineering, Monash University, Clayton, 3800, Australia ABSTRACT: The capillary rise and Wilhelmy plate methods have been used to study the “surface tension” of water marbles encapsulated with polytetrafluoroethylene (PTFE) powders of 1-, 35-, and 100-μm particle size. With the capillary rise technique, a glass capillary tube was inserted into a water marble to measure the capillary rise of the water. The Laplace pressure exerted by the water marble was directly measured by comparing the heights of the capillary rise from the marble and from a flat water surface in a beaker. An equation based on Marmur’s model was proposed to calculate the water marble surface tension. This method does not require the water contact angle with the supporting solid surface to be considered; it is therefore a simple but efficient method for determining liquid marble surface tension. The Wilhelmy method was used to measure the surface tension of a flat water surface covered by PTFE powder. This method offers a new angle for investigating liquid marble shell properties. A discussion on the nature and the realistic magnitude of liquid marble surface tension is offered.
1. INTRODUCTION Liquid marbles, first reported by Aussillous and Quere a decade ago, are an interesting interfacial phenomenon where liquid droplets are enwrapped by hydrophobic powder particles that adhere to the liquid droplet.1 Ever since their discovery, liquid marbles have attracted much attention toward their underlying science,2 properties,3 and applications.4 Recently, further new applications have been reported on using liquid marbles to build micropumps5 and miniature reactors.4a These reports sparked a renewed interest in the understanding of the interfacial properties of liquid marbles, such as their “surface tension” or “effective surface tension”,2a,6 which can greatly influence phenomena such as the formation of liquid marbles using low surface tension liquids,4a and using liquid marbles as micropumps. The “effective surface tension” of a liquid marble is a term describing the surface tension of the liquid droplet, under the influence of the wrapping powder. A simplistic explanation of the origin of liquid surface tension is that the molecules at the liquidair interface experience unbalanced attractive intermolecular forces resulting in a net pulling force toward the bulk of the liquid. For the shell of a liquid marble, the situation is more complicated, as a significant fraction of the liquid/air interface is covered by powder particles and becomes the liquid/solid interface.3b On the other hand, the powder particles on the liquid surface may significantly change the curvature of the liquid drop surface at the microscopic scale, since the particles are not wettable by the liquid and therefore are “floating” on the liquid surface, making micro indents on the liquid surface. The interactions between hydrophobic particles by capillary force may affect the net balance of forces on the shell of a liquid marble.7 These forces may alter the surface tension of the liquid and cause the marble to have an “effective surface tension” that may or may not be the same as the surface tension of the liquid that forms the marble core.2a r 2011 American Chemical Society
There have been several reports in the literature on the measurement of the “effective surface tension” of liquid marbles.2a,6 Methods employed in those studies include the mathematical curve fitting of the marble shape, the characterization of the response of liquid marble to vibration, and the measurement of liquid marble “puddle height”.6b It is necessary to explore other methods for determining the liquid marble surface tension that do not rely on the use of the imaginary liquid/solid contact line or contact angle. The theory we used in this study is based on our hypothesis that a near-spherical water/powder interface of the marble shell exerts a Laplace pressure and this pressure can be measured. When a capillary tube of an internal radius r is inserted vertically into a flat water surface, a capillary rise (hw) of water can be obtained in the tube: 2γw cos θ Fghw ¼ 0 r
ð1Þ
According to Marmur,8 if this capillary tube is connected to a spherical water drop of radius Re in the way shown in Figure 1a, assuming that Re does not change as water goes from the droplet into the tube, the water capillary rise is higher than that from the flat water surface. This additional capillary rise (Δhw) is caused by the curved surface of the water droplet. Marmur proposed the following equation to describe the capillary rise of the above system: 2γw 2γ cos θ Fgðhw þ Δhw Þ ¼ 0 þ w r Re
ð2Þ
where γw is the liquid surface tension, θ is the liquid/capillary wall contact angle, F and g are the liquid density and the Received: April 21, 2011 Revised: August 31, 2011 Published: September 12, 2011 12923
dx.doi.org/10.1021/la2014682 | Langmuir 2011, 27, 12923–12929
Langmuir
ARTICLE
made using inert hydrophobic powder particles. Discussion on the effects of powder particle size on the marble surface tension is offered.
2. EXPERIMENTAL SECTION
Figure 1. (a) Schematic illustration of the experimental setup for capillary rise measurements. (b) Photograph of a capillary tube inserted into a PTFE water marble.
gravitational constant, and Re and hw are the equilibrium values of the radius of the spherical liquid drop and the height of capillary rise, respectively. In a real situation when the radius of the drop is greater than the capillary length of the liquid, deformation of the liquid drop will occur due to gravity. Equation 2 must be revised to include the deformation: 1 1 2γ cos θ Fgðhw þ Δhw Þ ¼ 0 ð3Þ þ þ w γw r1 r2 r where r1 and r2 are the principal radii of the curvature of the drop surface. A simple experimental approach of using the hydraulic radius, defined by eq 4, can be employed to describe the curvature of the drop surface.9 1 1 1 ¼ þ m r1 r2
ð4Þ
Substituting eq 4 into eq 3 gives eq 5. Equation 5 can then be used to experimentally measure the m value of a droplet, by using a liquid of known surface tension and a capillary of known properties. 1 2γ cos θ Fgðhw þ Δhw Þ ¼ 0 ð5Þ þ w γw m r Marmur’s model (eq 2) can be adapted to measure the effective surface tension of a liquid marble by considering the experimental convenience offered by eq 5. In this study the capillary rise method was used to determine the effective surface tension of polytetrafluoroethylene (PTFE) water marbles. Another method we have employed is the Wilhelmy plate method. Powder particles were spread on the water surface in a beaker until the entire water surface was covered. Such a particle-covered water surface should have the same effective surface tension as the water marble shell. The Wilhelmy plate method as a routine surface tension measurement method offers an easy way to measure the liquid marble effective surface tension, although attention must be paid to some experimental details and result interpretation, especially when dealing with powders of large particle sizes. The results obtained using the two methods also provided new insights to the nature of the water marble effective surface tension
2.1. Materials. PTFE powders of 100-, 35-, and 1-μm nominal particle sizes were obtained from Sigma-Aldrich. Millipore water was used in all the experiments. Capillary tubes of an internal radius of 0.45 mm and a wall thickness of 0.1 mm were used in all the experiments. 2.2. Marble Formation. A PTFE powder bed was prepared in a Petri dish. Droplets of water of the controlled sizes were deposited on the powder bed with a micropipet to form marbles by allowing the droplets to roll. The Petri dish was then placed on a moving stage for capillary rise measurement (Figure 1). The marble sizes used in this study were 30, 50, 100, 200, and 300 μL. Water droplets of the same volumes were also measured for their capillary rises. The water droplets were gently placed on the powder bed without shaking. Since PTFE powder does not spontaneously spread over the water surface,7 around 70% of the water droplet surface was free of powder. Since the area around the maximum latitude circle of the droplet is free of powder, the capillary rise data obtained from the “powder-free” water droplet were used to experimentally calculate the hydraulic radius of the droplet (eqs 4 and 5). 2.3. Capillary Rise Method. Figure 1a shows the schematic experimental setup for the capillary rise measurement. The capillary tube was fixed vertically using Scotch tape on a ruler, which was clamped to a lab stand. A water marble or water droplet carried by the moving stage was used to insert the capillary into the marble or droplet for measurement (Figure 1b). The height of capillary rise was measured by taking a reading on the ruler. Before being mounted, the capillary tubes were immersed in a solution of laboratory detergent (RBS 35) and were sonicated for 15 min. This was followed by three cycles of rinsing with Millipore water and 15 min of sonication in Millipore water. The cleaned capillaries were stored immersed in Millipore water before use. In capillary rise measurements of the water droplets and water marbles, capillary rise from the flat water surface in a beaker (hw, eq 1) was first measured. The same capillary tube was used to measure the additional capillary rise of water droplet (Δhw, eq 5) and then of the water marble (Δhm, eq 7 ). Since a significant length of water still remained in the capillary after it was removed from the beaker, when the same capillary was inserted into the water droplet or the marble, the amount of water driven into the capillary was small. This minimized the excessive loss of water from the droplet and the marble; the amount of water that entered the capillary was considered in the final calculation of the droplet and marble surface tension. Capillary rise data obtained from the above three situations were used for the final calculation of the water marble effective surface tension. 2.4. Wilhelmy Plate Method. A microbalance (DCA 322, Cahn Instrument, USA) was used for the Wilhelmy plate measurements of the surface tension of water and the water covered with a layer of PTFE powder. The glass coverslips used in the experiment were obtained from Cahn Instrument, USA. The width and the thickness of the coverslips were 18 mm and 0.1 mm, respectively. Plasma treatment (20 W/30 s) was applied to the coverslips for surface cleaning to restore the full water wettability. The plate advancing and receding speeds were set at 24 μm/s. The surface tension values for water and the water covered by PTFE powder were calculated following eq 6 at zero plate immersion depth where buoyancy is zero: γw ¼
F cos θ 2ðl þ tÞ
ð6Þ
where γw and θ are the surface tension of water and the contact angle between water and the coverslip, respectively; l and t are the width and 12924
dx.doi.org/10.1021/la2014682 |Langmuir 2011, 27, 12923–12929
Langmuir
ARTICLE
Table 1. Summary of Capillary Rise Data Measured for the Flat Water Surface, Water Droplets, and Water Marbles, As Well As Corresponding Surface Tension Calculations liquid volume (μL) 30 50 100 200 300 Δh flat water surface (mm)
6.5 5
3.5 3
1.8
Δh marble of 1-μm particles (mm)
6.5 5
3.2 2
1.8
Δh marble of 35-μm particles (mm)
6
Δh marble of 100-μm particles (mm)
5.5 3.5 2.9 2
Δh theoretical spherical droplet (mm) hydraulic radius (1/m) calculated using eq 5
16.1 13.6 10.8 8.6 7.5 903 694 486 417 243
surface tension, water (mN/m)
71 71 71 71 71
effective surface tension, 1-μm particles (mN/m)
71 71 70 69 71
4.5 3.4 1.6 1.5 1.5
effective surface tension, 35-μm particles (mN/m) 70 70 70 68 70 effective surface tension, 100-μm particles (mN/m) 69 68 69 69 70
the thickness of the glass coverslip, respectively. When the contact angle is zero (this condition is more likely to be satisfied by the receding contact angle), cos θ = 1. The surface tension of water can be calculated using the force measured and the perimeter (2(l + t)) of the glass coverslip. The error for measured surface tension was estimated using the balance calibration error.
3. RESULTS AND DISCUSSION 3.1. Water Marble Effective Surface Tension by the Capillary Rise Method. Table 1 presents the capillary rise results by
inserting the capillary tube into a flat water surface in a beaker and into a water droplet. Additional capillary rises caused by water droplets of different volumes are observed. The collection of each datum was repeated three times, and average values are presented. The error associated with the capillary rise height reading was assumed to be the same as the liquid meniscus height reading from a buret in acidbase titration, which is 0.2 mm. Realistically, however, a capillary rise height difference of 0.5 mm can be confidently differentiated using the present experimental setup. Since a 0.5 mm capillary rise height difference corresponds to 1 mN/m, all surface tension data obtained by the capillary rise method are presented in two significant figures. 3.1.1. Capillary Rise Data from a Flat Water Surface and from a Marble. If water droplets are assumed to be spherical, a theoretical Δhw can be calculated using eq 2 to be 16.1 mm for the 30 μL water droplet. However, our experimental Δhw value for a 30 μL water droplet was only 6.5 mm, and the Δhw values for larger droplets were less than 6.5 mm (Table 1). The difference between the theoretical and experimental values of Δhw is caused by the droplet deformation under the influence of gravity; it is therefore necessary to use the hydraulic radius (1/m) (eq 5) to take into account the gravity influence experimentally (eqs 3 and eq 4). The additional capillary rises caused by the PTFEwater marbles (Δhm) of different water volumes are also shown in Table 1. For calculating the effective surface tension of water marbles, we assume that water marble and the water droplet of the same volume have the same 1/m value. This assumption is supported by the following observations: First, the additional capillary rises of the water droplets (Δhw) and the water marbles (Δhm) of the same volume are similar. The similar magnitudes of Δhw and Δhm suggest that PTFE particles deposited on the water
Figure 2. Photograph showing the continuity of powder-covered and powder-free surfaces of a water marble.
surface do not affect the water surface tension significantly. For marbles of the same volume, capillary rise shows relatively small difference from one particle size to another. The calculation below (eq 8) shows that a 1 mm difference in Δhm corresponds to a difference in the marble effective surface tension of around 2 mN/m. This calculation suggests that our assumption is reasonable. The capillary rise method may have an advantage over methods based on mathematical curve fitting of a liquid marble image in reducing measurement error. This is because a photographic image of a liquid marble shell does not represent the actual particle/water interface. Furthermore, a marble image has a rough border and curve fitting may introduce an uncertain level of error.3b Second, the surface of a water droplet that is partially covered by PTFE powder does not show any visually detectable changes in curvature across the border of particlecovered and particle-free surfaces (Figure 2). Such continuity of curvature across the particle-covered and particle-free surfaces suggests that the curvature of the particle-free water surface can predict that of the particle-covered water surface. Third, the addition of a small amount PTFE powder to water, so that it just covers a small fraction of the water surface, does not change the water surface tension; this point was confirmed by the Wilhelmy plate measurement of the water surface tension in the presence or absence of a small amount of PTFE powder. This eliminates the possible scenario that a difference in water surface tension for the droplet and for the marble is a factor affecting their hydraulic radii. The assumption of the marble and droplet having the same 1/m value simplifies the calculation of the effective surface tension of water marbles. Equation 5 needs to be further modified in order to adapt the Marmur model to calculate the effective surface tension of water marbles. When the capillary is inserted into a water marble, the total capillary rise is driven by the water meniscus inside the capillary tube and the marble shell outside the capillary. Adapting these differences into eq 5, eq 7 is obtained. γm 2γ cos θ ¼ Fghw þ FgΔhm þ w r m
ð7Þ
where γm is the effective surface tension of the water marble, γw is the surface tension of water, F is the water density, g is the gravitational force, hw is the capillary rise of water measured from the flat water surface in the beaker, Δhm is the additional capillary rise caused by the water marble, and 1/m is the hydraulic radius of the marble, which is assumed to be the same as that of the water droplet of the same volume. Since Fghw = (2γw cos θ)/r (eq 1), eq 8 can be obtained by a simple rearrangement of eq 7: γm ¼
FgΔhm ¼ FgΔhm m 1=m
ð8Þ
Calculation of the effective surface tension of water marbles of different sizes can be made using eq 8 and the data of 1/m and Δhm of water marbles. The capillary rise values measured from 12925
dx.doi.org/10.1021/la2014682 |Langmuir 2011, 27, 12923–12929
Langmuir water droplets are also presented in Table 1. The hydraulic radii of the water droplets (1/m) were calculated using the water surface tension value determined by the capillary rise method (eq 1) and the additional capillary rise values of the droplets (Δhw) measured against the flat water surface by eq 5. Table 1 shows that the additional capillary rise data of water droplets (Δhw) decrease as the droplet volume increases. This trend is in agreement with the Laplace principle. The additional capillary rise data for the water marbles (Δhm) of different volumes measured against the flat water surface, as well as the water marble effective surface tension values calculated from the Δhm data using eq 8, are listed in Table 1. The Δhm data for water marbles of different volumes show trends similar to those of the water droplets; i.e., the Δhm values decrease as the volume of the marbles increase. This trend is also in agreement with the Laplace principle. Data in Table 1 show an interesting and consistent trend: All additional capillary rise data of water marbles are slightly lower than those of water droplets of the same volume. This suggests that the Laplace pressure exerted by the water marble shell is slightly lower than that exerted by water droplets of the same volume. 3.1.2. Nature of the PTFEWater Marble Effective Surface Tension. The above observations offer a clue for us to further understand the nature of the liquid marble effective surface tension. PTFE particles flocculate on the water surface and on the marble surface; capillary force by the water meniscus between particles is the major driving force responsible for the twodimensional particle flocculation on water. Monteux et al.1013 investigated the interfacial tension and the particle flocculation behavior of a particle-covered decanewater interface. Their findings also shed light on the particle flocculation behavior on a water marble surface. Monteux et al.11 showed that when the particle-covered interface is in its natural state (i.e., no lateral external force applied onto particles on the interface and the water drop has a spherical shape in decane), the interface is in the fluid state. In this state the particle-covered interface can be described by the Laplace equation and the interfacial tension can be determined by measuring the bulk pressure of the droplet. When a lateral external force was applied to the particles at the interface, the particle layer began to buckle and the interfacial tension was no longer predictable by the Laplace equation. Liquid marble shell is not fully covered by consisting particles; only one layer is in contact with the liquid core and this particle layer affects the effective surface tension of the marble. A freshly formed PTFEwater marble is nearly spherical; if gravity is neglected, the particlewater interface is not under any external lateral force. Therefore, PTFE particles on the water marble are likely to be in a state similar to the fluid state described by Monteux et al.10,11 In this state PTFE particles are held by capillary forces; however, capillary forces between particles are not uniform across the marble surface due to the random and nonuniform distribution of particles on the marble surface. Such nonuniform particle distributions on liquid marble surfaces have been repeatedly reported.2c,3a,4h,4i Monteux et al.11 observed that, on a particle-covered decanewater interface, particles formed a closely packed slab when a lateral compression force was applied at the interface. However, when the interface was decompressed, particles formed patches that moved away from one another as the interface expanded. Although capillary force tightly holds particles to form many integrated multiparticle patches, each patch does not respond to the increase of interface area by increasing the interparticle distance; instead, the
ARTICLE
distances between patches become larger. This behavior also suggests that the particle-free interface dominantly responds to the increase in interface area. Monteux et al.11 also showed that interfacial tension values of interfaces covered by particle patches show a very small differences from those of interfaces covered by well-dispersed particles. Their results support our observation that water surfaces free of PTFE particles possess the property of surface tension, whereas surfaces covered by particles (or integrated multiparticle patches) do not have the property of surface tension, as they do not exhibit the “elastic” behavior that the water surface does. This analysis therefore suggests that the effective surface tension of PTFEwater marble is basically the surface tension of water under the influence of the inert PTFE particles deposited on the water surface. 3.1.3. Realistic Magnitude of PTFEWater Marble Effective Surface Tension. The magnitude of the effective surface tension for marbles formed with 1-μm PTFE particles is slightly higher than those formed with large particles. Although the magnitude of this trend is only around 12 mN/m, it corresponds to 0.5 1 mm difference in the capillary rise that is measurable using the capillary rise method. A similar trend was also reported by Monteux et al. for the particle-covered decanewater interface.11 A calculation using data from the Laplace pressure vs the radius of the particle-covered water droplet in decane by Monteux et al.11 showed that the interfacial tension of the particle-covered decanewater interface was lower than that of the same interface but free of particles. In the PTFEwater marble study, the marble surface tension slightly lower than the water surface tension may be related to the structure of the marble shells formed with particles of different sizes. We observed that marbles coated with 1-μm PTFE powder were less stable than those made of other two powders, suggesting that the 1-μm PTFE water marble surface may have a larger fraction of area not covered by PTFE powder. Another observation is that the 1-μm PTFE particle coating layer on water is more compressible than that of the 100-μm PTFE particle layer. When two equal-volume marbles, one coated with 1-μm and the other with 100-μm PTFE particles, merge to form a bigger marble, 100-μm particles push forward as the marbles merge and occupy ∼75% of the surface of the newly formed marble, whereas the 1-μm particles only occupy ∼25% of the new marble surface. The greater compressibility of the 1-μm PTFE particles also suggests that the 1-μm PTFEwater marble could have a greater fraction of powder-free area. A marble surface that has a larger fraction of powder-free water surface is expected to have an effective surface tension closer to water surface tension. 3.2. Effective Surface Tension of Water Covered by PTFE Particles Measured by the Wilhelmy Method. 3.2.1. Observations Specifically Related to the Wilhelmy Method. The effective surface tension of a flat water surface covered by a layer of inert hydrophobic powder should be the same as that of a water marble covered by the same powder. Any method that can measure the surface tension of a powder-covered flat water surface therefore has the potential to provide insights into the influence of inert powder particles on the water surface. In this work we used the Wilhelmy method. A simple Wilhelmy cycle containing an advancing phase and a receding phase was used to measure the water surface tension and the effective water surface tension when a thin layer of powder (visually similar to that on the marble surface) was spread onto the water surface. The force measured by the Wilhelmy method shows a sharp increase as the glass coverslip 12926
dx.doi.org/10.1021/la2014682 |Langmuir 2011, 27, 12923–12929
Langmuir
ARTICLE
Figure 3. Wilhelmy results of advancingreceding cycles of (a) water, (b) water covered by 1-μm PTFE particles, (c) water covered by 35-μm PTFE particles, and (d) water covered by 100-μm PTFE particles. The force values shown by red arrows are presented in Table 2. These values are used to calculate the effective surface tension of water and PTFEwater marbles, which are also presented in Table 2.
touches the water surface (Figure 3). Continued advancing shows an irregular pattern of force variation due to the physical or chemical nonhomogeneities on the coverslip.11 The receding force curve is smooth, as a thin film of water completely covers the inhomogeneous surface regions over the coverslip surface during receding. It is reasonable to assume that the receding contact angle for water with the hydrophilic surface glass surface is zero or at least very close to zero. Since the first advancing force reading equals the receding force reading at the same position (see arrow), we decided to use the first advancing force reading to calculate the water surface tension. The so-calculated water surface tension using eq 6 was 72.3 ( 0.3 mN/m, which agrees reasonably well with the literature value (72.14 mN/m) for 23 °C.12 The wetting force data of the particle-covered water surface were taken as the zero plate immersion depth in advancing. The wetting force data show three differences from that of the powder-free water surface. First, as the coverslip advances toward the PTFE-particle-covered water surface, the coverslip first touches particles on the surface, generating a negative force reading (Figure 3). The coverslip then pierces through the particle layer, making contact with water underneath the particle layer. Second, the magnitude of the negative force readings increases as the particle size increases. Third, the first advancing force reading
as the coverslip pierces through the particle layer decreases as the size of the particles increases. Interestingly, the Wilhelmy measurements of the effective surface tension of PTFE-particlecovered water surface shows a consistent trend with the effective surface tension values measured from water marbles using the capillary rise method, although the values of the measurements by the two methods are significantly different for the larger particles. Figure 3 shows that it requires a smaller force to pierce through the layer of 1-μm particles but a larger force for larger particles. This result is in qualitative agreement with our observation that water marbles made of 1-μm powder particles were the most unstable and those made of 100-μm particles were the most stable. The only surprising difference shown by the Wilhelmy method is that the effective surface tension for the 100-μm PTFE particles was significantly lower than those for the 1- and 35-μm particles (Table 2). A close examination reveals that the PTFE powder particles on the water meniscus of the glass coverslip may affect the height of the water meniscus (Figure 4). One-micrometer particles formed a layer over water with a greater fraction of the powder-free surface than 100-μm particles; the 1-μm particle layer over the water surface might have been pushed away from the wall of the coverslip by the curved meniscus, making the particle layer have less influence on the meniscus. While the 12927
dx.doi.org/10.1021/la2014682 |Langmuir 2011, 27, 12923–12929
Langmuir
ARTICLE
Table 2. Calculated Values of Wetting Forces and Surface Tension Results from the Data Extracted from the Graphs Obtained from Wilhelmy Plate Method sample
wetting force (mg)
wetting force (mN)
surface tension (mN/m)
water water covered by 1-μm PTFE particles water covered by 35-μm PTFE particles
269 ( 1
2.63 ( 0.01
72.3 ( 0.3
259 ( 1 258 ( 1
2.54 ( 0.01 2.53 ( 0.01
70.2 ( 0.3 69.9 ( 0.3
water covered by 100-μm PTFE particles
213 ( 1
2.08 ( 0.01
57.6 ( 0.3
Figure 4. Schematic showing the possible influence of PTFE particles of different sizes on the meniscus height of water (H) on the wall of the glass coverslip. Thinner and more compressible 1-μm PTFE particle layer has less influence on the meniscus height, as it can be pushed away by the rising meniscus during the advancing of the coverslip. On the other hand, 100-μm PTFE particles affect the meniscus height more significantly.
100-μm particle layer has a smaller fraction of powder-free water surface; it may have a stronger influence on the meniscus as it could not be pushed as far away as the 1-μm powder layer by the meniscus. Since the surface tension measurement using the Wilhelmy method is sensitively related to the water meniscus height on the wall of the glass coverslip, particles of different sizes may affect the measurement of the effective surface tension of particle-covered water surface using the Wilhelmy method. Our interpretation is supported by the fact that water covered by powders of different sizes show very similar receding wetting forces; this is because the water meniscus on the wall of the coverslip pushes the powders further away during receding than during advancing. When particles are pushed away from the coverslip wall, their influence on the meniscus height on the coverslip wall will be smaller, and this may be the reason why the receding force data are similar for particles of different sizes. 3.2.2. Comments on the Wilhelmy Data of the Effective Surface Tension. It is likely that the use of the coverslip in the Wilhelmy method has disturbed the powder/water interface; consequently the effective surface tension measured this way might not fully reflect the condition of the PTFEwater interface in an undisturbed state. Despite this, the Wilhelmy method still presents qualitative confirmation of the results obtained using the capillary rise method. The Wilhelmy method also shows that the effective surface tension of a water marble covered by PTFE particles is in fact the water surface tension, but influenced by the presence of the particles on the water surface. The Wilhelmy results also suggest that the water surface covered by the inert PTFE powder has a lower surface tension than water.
4. CONCLUSION This work investigated two experimental methods for measuring the effective surface tension of liquid marbles. We chose PTFEwater marble as the system of our investigation. The capillary rise method captures the Laplace pressure exerted by the liquid marble shell; this method requires the measurements
of only the capillary rise differences between the water marble and the flat water surface. The Wilhelmy method offers a new way of studying the surface properties of liquid marble; we proposed that the effective surface tension of a flat particlecovered liquid surface is the same as that of the liquid marble, provided that the particles are inert to the liquid. This proposal will hopefully provide a fresh angle for studying the liquid marble surface properties. The result analysis shows that the capillary rise method gives a more realistic effective surface tension of water marble covered by inert PTFE powder particles. It is likely that data obtained with the Wilhelmy method may not fully reflect the effective surface tension of PTFE-covered water surface, since the Wilhelmy plate disturbs the particlewater interface. Despite this, the Wilhelmy method may still be a simple method that can provide new understanding of the particle-covered liquid surface. This work also shows that the effective surface tension of PTFEwater marble is in fact the water surface tension, but it is influenced by the presence of PTFE particles. From this point of view the effective surface tension of PTFEwater marble is lower than the surface tension of water; our results are in agreement with this analysis.
’ AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected].
’ ACKNOWLEDGMENT Funding received from the Australian Research Council through Grants ARC LP0989823 and ARC LP0990526 is acknowledged. ’ REFERENCES (1) Aussillous, P.; Quere, D. Liquid marbles. Nature 2001, 411 (6840), 924–927. (2) (a) Aussillous, P.; Quere, D. Properties of liquid marbles. Proc. R. Soc. A: Math., Phys. Eng. Sci. 2006, 462 (2067), 973–999. (b) Mahadevan, L. Non-stick water. Nature 2001, 411 (6840), 895–896. (c) McEleney, P.; Walker, G. M.; Larmour, I. A.; Bell, S. E. J. Liquid marble formation using hydrophobic powders. Chem. Eng. J. 2009, 147 (23), 373–382. (d) Bormashenko, E.; Pogreb, R.; Bormashenko, Y.; Musin, A.; Stein, T. New investigations on ferrofluidics: ferrofluidic marbles and magneticfield-driven drops on superhydrophobic surfaces. Langmuir 2008, 24 (21), 12119–12122. (3) (a) Eshtiaghi, N.; Liu, J. S.; Shen, W.; Hapgood, K. P. Liquid marble formation: spreading coefficients or kinetic energy? Powder Technol. 2009, 196 (2), 126–132. (b) Nguyen, T. H.; Hapgood, K.; Shen, W. Observation of the liquid marble morphology using confocal microscopy. Chem. Eng. J. 2010, 162 (1), 396–405. (c) Newton, M. I.; et al. Electrowetting of liquid marbles. J.Phys. D: Appl. Phys. 2007, 40 (1), 20. (d) McHale, G.; Elliott, S. J.; Newton, M. I.; Herbertson, D. L.; Esmer, K. Levitation-free vibrated droplets: resonant oscillations of 12928
dx.doi.org/10.1021/la2014682 |Langmuir 2011, 27, 12923–12929
Langmuir
ARTICLE
liquid marbles. Langmuir 2008, 25 (1), 529–533. (e) McHale, G.; Herbertson, D. L.; Elliott, S. J.; Shirtcliffe, N. J.; Newton, M. I. Electrowetting of nonwetting liquids and liquid marbles. Langmuir 2006, 23 (2), 918–924. (4) (a) Xue, Y.; Wang, H.; Zhao, Y.; Dai, L.; Feng, L.; Wang, X.; Lin, T. Magnetic liquid marbles: a “precise” miniature reactor. Adv. Mater. 2010, 22 (43), 4814–4818. (b) Tian, J.; Arbatan, T.; Li, X.; Shen, W. Liquid marble for gas sensing. Chem. Commun. 2010, 46 (26), 4734– 4736. (c) Bormashenko, E.; Bormashenko, Y.; Pogreb, R.; Gendelman, O. Janus droplets: liquid marbles coated with dielectric/semiconductor particles. Langmuir 2010, 27 (1), 7–10. (d) Gao, L.; McCarthy, T. J. Langmuir 2007, 23, 10445. (e) Bhosale, P. S.; Panchagnula, M. V.; Stretz, H. A. Mechanically robust nanoparticle stabilized transparent liquid marbles. Appl. Phys. Lett. 2008, 93 (3), No. 034109. (f) Zhao, Y.; Fang, J.; Wang, H.; Wang, X.; Lin, T. Magnetic liquid marbles: manipulation of liquid droplets using highly hydrophobic Fe3O4 nanoparticles. Adv. Mater. 2010, 22 (6), 707–710. (g) Bormashenko, E.; Bormashenko, Y. Non-stick droplet surgery with a superhydrophobic scalpel. Langmuir 2011, 27 (7), 3266–3270. (h) MaHale, G.; Newton, M. I. Liquid marbles: principles and applications. Soft Matter 2011, 7 (12), 5473– 5481, DOI: . DOI: 10.1039/C1SM05066D. (i) Bormashenko, E. Liquid marbles: properties and applications. Curr. Opin. Colloid Interface Sci. 2011, 16 (4), 266–271, DOI: . DOI: 10.1016/j.cocis.2010.12.002. (5) Bormashenko, E.; Balter, R.; Aurbach, D. Micropump based on liquid marbles. Appl. Phys. Lett. 2010, 97 (9), 091908. (6) (a) Bormashenko, E.; Pogreb, R.; Whyman, G.; Musin, A. Surface tension of liquid marbles. Colloids Surf., A 2009, 351 (13), 78–82. (b) Bormashenko, E.; Pogreb, R.; Whyman, G.; Musin, A.; Bormashenko, Y.; Barkay, Z. Shape, vibrations, and effective surface tension of water marbles. Langmuir 2009, 25 (4), 1893–1896. (7) Nguyen, T. H.; Eshtiaghi, N.; Hapgood, K. P.; Shen, W. An analysis of the thermodynamic conditions for solid powder particles spreading over liquid surface. Powder Technol. 2010, 201 (3), 306–310. (8) Marmur, A. Penetration of a small drop into a capillary. J. Colloid Interface Sci. 1988, 122 (1), 209–219. (9) Scheidegger, A. E. The Physics of Flow through Porous Media, 3rd ed.; University of Toronto Press: Toronto, 1974; p 58. (10) Monteux, C.; Jung, E.; Fuller, G. G. Mechanical properties and structure of particle coated interfaces: influence of particles size and bidisperse 2D suspensions. Langmuir 2007, 23, 3975–3980. (11) Monteux, C.; Kirkwood, J.; Xu, H.; Jung, E.; Fuller, G. G. Determining the mechanical response of particle-laden fluid interfaces using surface pressure isotherms and bulk pressure measurements of droplets. Phys. Chem. Chem. Phys. 2007, 9, 6344–6350. (12) CRC Handbook of Chemistry and Physics, 67th ed.; Weast, R. C., Astle, M. J., Beyer, W. H., Eds.; CRC Press: Boca Raton, FL, USA, 1987; p F-33. (13) Kralchevsky, P. A.; Nagayama, K. Capillary interactions between particles bound to interfaces, liquid films and biomembranes. Adv. Colloid Interface Sci. 2000, 85, 145–192.
12929
dx.doi.org/10.1021/la2014682 |Langmuir 2011, 27, 12923–12929
