Design of honeycomb structure surfaces with controllable oil adhesion underwater

Abstract

In this paper, we fabricate honeycomb-like poly acrylic acid (PAA) surfaces and achieve oil adhesion transitions underwater. Characterization of the adhesion indicates that the porous honeycomb structure PAA films can serve as high oil adhesive surfaces both in acidic and in basic aqueous phases. Besides, the adhesion can be controlled by changing the size of the pores and the solution pH. The honeycomb structure film prepared using a template with a theoretical diameter of 6 μm has the highest adhesion in basic solution, which can snap some oil droplets from the original oil. In contrast, the smooth films and other honeycomb structure films in acidic and basic solutions cannot snap an oil droplet. The switchable oil adhesion is attributed to the change of the triple-phase liquid/liquid/solid contact line (TCL) continuity and the negative pressure induced by the pores. This highly adhesive porous film was used as a “mechanical hand” to transfer micro-droplets successfully underwater. The unique adhesive phenomenon of the honeycomb structure will be useful for manipulating oil droplet behavior and suitable for the application of controlling liquid collection and transportation underwater.

---

Introduction

The adhesive behavior of oil droplets on solid surfaces in an aqueous medium has recently become a new research focus, inspired by the superoleophobic low-adhesion properties of fish scales and clam shells underwater.^1–8 The reported methods for regulating oil adhesion underwater can be classified into two major categories, both of which depend on controlling the solid phase: the modification of stimulus-responsive materials and the construction of hierarchical surface structures. Our group previously fabricated several smart surfaces for which the adhesion state of oil droplets could also be switched under water, using stimulus-responsive materials on solid substrates^9–11 or by controlling the interactions at the liquid–liquid interface.¹² Currently, the work in this field has mainly focused on the regulation of the oil adhesion underwater from high to low, no attention has been paid to regulating oil adhesion from high to even higher on a honeycomb-structured surface. Our group previously reported for the first time a controllable high water adhesion in air based on honeycomb structure materials¹³ and oil surface adhesion regulation underwater by controlling the pH of the solution.¹² Inspired by these two works, we will tune oil adhesion on honeycomb films by controlling the pH of the solution. This should be of great scientific interest because the high oil adhesive behavior of the honeycomb structure is more suitable for controlling liquid collection and transportation underwater.^14,15

The aim of the present research is to fabricate honeycomb-like PAA surfaces and achieve oil adhesion transitions in acidic and basic aqueous phases. Below, we will describe the preparation of the PAA hexagonal honeycomb structure. Characterization of the adhesion indicates that the porous honeycomb structure PAA films can serve as high oil adhesive surfaces both in acidic and in basic aqueous phases. Besides, the adhesion can be controlled by changing the size of the pores and the solution pH. The honeycomb structure film prepared using a template with a theoretical diameter of 6 μm has the highest adhesion in basic solution, which can snap some oil droplets from the original oil. This highly adhesive porous film was used as a “mechanical hand” to transfer micro-droplets successfully underwater. In contrast, the smooth films and other honeycomb structure films in acidic and basic solutions cannot snap an oil droplet. The switchable oil adhesion is attributed to the change of the TCL continuity and the negative pressure induced by the pores. The unique adhesive phenomenon of the honeycomb structure will be useful for manipulating oil droplet behavior and suitable for the application of controlling liquid collection and transportation underwater.

Experimental

Preparation of the honeycomb structure PAA

Hexagonal silicon pillar templates with different sizes were obtained using a lithographic etching process. The theoretical diameters of the silicon pillars in our experiment were 4, 6, 8 and 10 μm, respectively. The distance between two adjacent pillars was 1.5 μm, and the height of the silicon pillars was 5 μm. The PAA hydrogel was prepared using a photo-initiated polymerization process with acrylic acid monomers (AA), 2,2-dimethoxy-2-phenylacetophenone (DMPA) and N,N-methylene diacrylamide (BIS) as a precursor, initiator and cross-linker, respectively. Distilled water was used to dissolve the AA, DMPA and BIS. Then this solution was poured onto the silicon template. Through a photo-initiated in situ radical polymerization (λ = 365 nm, 40 min), a honeycomb structure film of PAA-hydrogel was obtained.

Aqueous phase pH control

The basic aqueous phase with a pH of 11.12 was prepared by dissolving NaOH in deionized water at a predetermined concentration. The acidic aqueous phase with a pH of 1.92 was prepared by dissolving H₂SO₄ at a predetermined concentration. A pH meter was used to measure the pH of the aqueous phases. Then, the smooth films and the honeycomb structure films were immersed in water with the different pH values to study their oil droplet contact angle (CA) and adhesion.

Characterization

The PAA hydrogels were placed in acid and alkaline solution for 20 min respectively. Then the PAA hydrogels were transferred to liquid nitrogen. After being completely frozen, we put the PAA hydrogel into a freeze drier for 24 hours to remove the water from the hydrogel for scanning electron microscopy (SEM) investigation. SEM images were obtained using a field-emission scanning electron microscope (JSM-6700F, Japan). Oil contact angles were measured on an OCA20 machine (Data Physics, Germany) at ambient temperature. The oil droplets (1,2-dichloroethane (DCE), about 2 μL) were dropped carefully onto the materials, which were immersed in water. The average value of five measurements performed at different positions on the same sample was adopted as the contact angle. The force required to take the oil droplet away from the substrate was measured using a high-sensitivity microelectromechanical balance system (Data Physics DCAT 11, Germany) in a water environment. An oil droplet (about 5 μL) was first suspended with a metal ring, and then the substrate was placed on the balance table. The substrate was moved upward at a constant speed of 0.005 mm s⁻¹ until the substrate came into contact with the oil droplet. Then the substrate was moved down. The force increased, and the shape of the oil droplet changed from spherical to elliptical. When the oil droplet was about to leave the substrate, the contact force sharply decreased and the shape of the droplet returned to spherical.

Results and discussion

Morphology of the honeycomb structure films both in acidic and in basic solution

Fig. S1† is the SEM images of the used hexagonal columnar silicon templates with different theoretical diameters and theoretical spaces. The theoretical diameters chosen in this experiment are 4 μm, 6 μm, 8 μm and 10 μm, respectively. The spaces between two posts are the same, ca. 1.5 μm (Table 1). Fig. 1 shows SEM images of the honeycomb structure PAA surface prepared using the theoretical diameter of 6 μm template both in acidic and in basic aqueous phases. Fig. 1a is the top-view SEM image of the honeycomb structure PAA film in the acidic aqueous phase. A film with a uniform hexagonal porous structure can be found over large areas. The pore diameter is 5.3 μm, while the average wall thickness between two pores is approximately 1.7 μm. Fig. 1b is the top-view SEM image of the honeycomb structure PAA film in the basic aqueous phase. The pore diameter is 4.7 μm, while the average wall thickness between two pores is approximately 2.0 μm. The results show that the actual pore diameters are smaller than that of the theoretical pore diameters and the actual wall thicknesses are larger than that of the theoretical walls because of the water swelling of the PAA gels. Especially in basic solution, this swelling is more significant. A similar phenomenon can be found in the other samples (Table 1). The pore diameters are 3.4 μm and 3.2 μm in the acidic and basic aqueous phases for the theoretical diameter of 4 μm template, respectively. The walls are approximately 1.8 μm and 1.9 μm, respectively. The pore diameters are 7.3 μm and 6.9 μm in the acidic and basic aqueous phases for the theoretical diameter of 8 μm template, respectively. The walls are approximately 1.8 μm and 1.9 μm, respectively. The pore diameters are 8.9 μm and 8.4 μm in the acidic and basic aqueous phases for the theoretical diameter of 10 μm template, respectively. The walls are approximately 1.7 μm and 1.9 μm, respectively.

Table 1 Average pore size and wall thickness between two pores of different honeycomb films in acid and basic solution

Sample no.	Theoretical diameter of the pore (μm)	Wall thickness (μm) in pH 1.92 water	Wall thickness (μm) in pH 11.12 water	Pore diameter (μm) in pH 1.92 water	Pore diameter (μm) in pH 11.12 water
1	4	1.8	1.9	3.4	3.2
2	6	1.7	2.0	5.3	4.7
3	8	1.8	1.9	7.3	6.9
4	10	1.7	1.9	8.9	8.4

---

Fig. 1 SEM images of the honeycomb films prepared using the template with a 6 μm theoretical pore size. (a) The actual diameter is 5.3 μm and the wall thickness is 1.7 μm in pH 1.92 solution and (b) the actual diameter is 4.7 μm and the wall thickness is 2.0 μm in pH 11.12 solution.

Oil wettability of the smooth films and honeycomb structure films underwater

First, the oil 1,2-dichloroethane was used to study the wettability transition in the aqueous phase upon a pH change. Fig. 2a and b show the oil contact angles on the smooth PAA surfaces in the aqueous phases. These surfaces showed oleophobicity with a contact angle of 115.4 ± 2.2° in acidic solution (pH 1.92) and a contact angle of 116.8 ± 1.4° in basic solution (pH 11.12) for the same surface. Similarly, the porous surfaces showed oleophobicity with a contact angle of 145.3 ± 3.4° in acidic solution (pH 1.92) and a contact angle of 139.2 ± 2.1° in basic solution (pH 11.12) for the same surface. Here the porous film was prepared using the template with a 6 μm theoretical pore size. The contact angles of the oil droplets on these smooth PAA surfaces and honeycomb structure surfaces were larger than 90° for both acidic and basic solutions. This result confirmed that these surfaces exhibit oleophobic properties under water. As compared to the smooth PAA films, the increased CAs of the porous films indicate that the hydrophobicity of the porous film is increased as the pores are added. Yet the porous films in different pH solutions have shown similar CA data, which indicates these porous films have the same roughness. The reason for the increase of the hydrophobicity is mainly ascribed to water trapped in the pores, which can prevent the intrusion of oil into the pores, resulting in the large contact angle.

Fig. 2 Oil CA photos for the smooth films and honeycomb films underwater. (a) Smooth film with a CA of about 115.4 ± 2.2° in pH 1.92 solution, (b) smooth film with a CA of about 116.8 ± 1.4° in pH 11.12 solution, (c) porous film with a CA of about 145.3 ± 3.4° in pH 1.92 solution, (d) porous film with a CA of about 139.2 ± 2.1° in pH 11.12 solution. Here the porous film was prepared using the template with a 6 μm theoretical pore size.

For a rough surface between a solid and air, Cassie et al. proposed a model describing the contact angle in a water/air/solid system. In an oil/water/solid system, on the other hand, where the rough surface is between a solid and water, the Cassie model is expressed as follows:^1,16

cos θ′ = f cos θ + f − 1

where f is the area fraction of the solid, θ is the contact angle of an oil droplet on a smooth surface in water, and θ′ is the contact angle of an oil droplet on a rough surface in water. In our study, taking 1,2-dichloroethane as an example, the contact angles θ and θ′ were measured to be 145.3 ± 3.4°and 115.4 ± 2.2° when the oil droplet was in contact with the honeycomb structure PAA surface and smooth PAA surface in acidic solution, and 139.2 ± 2.1° and 116.8 ± 1.4°in basic solution. We could calculate f according to the equation. The solid area fraction f is ca. 0.32 in acidic solution and 0.39 in basic solution. The theoretical calculations fit with our SEM images (Fig. 1) very well.

Oil adhesive behavior of the honeycomb structure PAA surface at different pHs

As reported above, the oil droplets on these surfaces are ball-like in shape, resulting from oleophobicity. However, they cannot roll off when the surface is tilted vertically, even if the surface is turned upside down. The shapes of the oil droplets on the surfaces are shown in Fig. 3. The volume of the oil droplet on the surface is as high as 5 μL. From these photos, it can be seen that these surfaces have a high adhesion for oil droplets underwater both in acidic and in basic solution.

Fig. 3 The oil droplet on the oleophobic surface remained attached even when the surface was tilted 180° underwater. (a) Smooth film in pH 1.92 solution, (b) smooth film in pH 11.12 solution, (c) porous film in pH 1.92 solution, (d) porous film in pH 11.12 solution. It is shown that these surfaces have high adhesion for oil underwater. The oil used is 1,2-dichloroethane with a specific gravity of about 1.26, which is higher than that of water. Here the porous film was prepared using the template with a 6 μm theoretical pore size.

The adhesive force was defined as the force required to lift the oil droplet off the substrate and it can be assessed using a highly sensitive microelectromechanical balance system. An optical microscope lens and a charge-coupled-device (CCD) camera system were used to record the images during the experiment. Whole curves of the force versus the distance between the solid surface and the oil droplet are plotted in Fig. 4. The distance is the film movement length from a certain base (zero) position to the oil droplet. Fig. 4a and b are force–distance curves recorded before and after the oil droplet contacted with the smooth surface underwater in acidic and basic solution, respectively. At first, the smooth film was placed on the plate of the balance system underwater, and a 5 μL oil droplet was suspended on a metal ring. Then, the film was brought into contact with the oil droplet (process 1). The film was moved at a rate of 0.01 mm s⁻¹. As the film was leaving the oil droplet after contact, the balance force increased gradually and reached a maximum of 35.2 ± 2.3 μN in acidic solution and 24.5 ± 1.4 μN in basic solution at the end of process 2 (Table 2). Finally, the balance force decreased immediately when the film broke away from the oil droplet in process 3 to finish one cycle of the force measurement. Before the substrate was about to leave the oil droplet, the shape of the oil droplet first changed from spherical to elliptical and then the shape changed back to spherical after the oil droplet was displaced from the surface. These results show that a smooth PAA surface shows high adhesion both in acidic solution and in basic solution. But the adhesion is larger in acidic solution than in basic solution. This adhesion difference is caused by different hydrogen bonding interactions. At low pH, intramolecular hydrogen bonding in PAA occurs, and results in high oil adhesion. As for high pH, the oil droplet adhesion decreases due to the intermolecular hydrogen bonding between PAA and the surrounding water.¹⁰

Fig. 4 Force–distance curves recorded before and after the oil droplet is in contact with the surface underwater. The distance of the x axis is the film movement length from a certain base (zero) position to the oil droplet. Process 1: the surface approaches the oil droplet. Process 2: the surface moves downward after coming into contact with the oil droplet. Process 3: the oil droplet moves away from the surface. Insets: photographs of the oil droplet shapes underwater taken at the corresponding stages during the measurement process. (a) Smooth film in pH 1.92 solution, (b) smooth film in pH 11.12 solution, (c) porous film in pH 1.92 solution, (d) porous film in pH 11.12 solution. Here the porous film was prepared using the template with a 6 μm theoretical pore size.

Table 2 Average adhesive forces of the smooth films and the porous films with different pore diameters in acid and basic solution

Sample no.	Theoretical diameter of the pore (μm)	Adhesion force (μN) in pH 1.92 water	Adhesion force (μN) in pH 11.12 water
1	0	35.2 ± 2.3	24.5 ± 1.4
2	4	33.4 ± 2.8	29.1 ± 3.2
3	6	30.4 ± 1.6	38.1 ± 2.5
4	8	27.6 ± 2.1	34.2 ± 3.5
5	10	24.3 ± 1.3	34.8 ± 1.9

---

The same experiment was conducted to study the adhesion behavior of the porous PAA films with different pore sizes. The porous film with 4 μm theoretical diameter pores has an adhesion of 33.4 ± 2.8 μN in acidic solution and 29.1 ± 3.2 μN in basic solution (Table 2). It has a similar behavior compared to the smooth film in the different pH value solutions. Fig. 4c and d are force–distance curves of the porous film with theoretical diameter pores of 6 μm in pH 1.92 and in pH 11.12 solution, respectively. The porous film with 6 μm theoretical diameter pores has an adhesion of 30.4 ± 1.6 μN in acidic solution and 38.1 ± 2.5 μN in basic solution (Table 2). This surface displayed a large adhesive behavior in basic solution compared to that in acidic solution, which makes the oil droplet able to snap on the surface (Fig. 4d insets). The porous film with 8 μm theoretical diameter pores has an adhesion of 27.6 ± 2.1 μN in acidic solution and 34.2 ± 3.5 μN in basic solution (Table 2). The porous film with 10 μm theoretical diameter pores has an adhesion of 24.3 ± 1.3 μN in acidic solution and 34.8 ± 1.9 μN in basic solution (Table 2). The above data showed that the adhesions of the smooth film and the porous film with 4 μm theoretical diameter pores in acidic solution are larger than that in basic solution, the adhesions of the porous films with 6 μm, 8 μm and 10 μm theoretical diameter pores in acidic solution are smaller than that in basic solution (Fig. 5a). In other words, the adhesion decreases with the increase of the pore diameter in acidic solution, and increases with the increase of the pore diameter in basic solution (Fig. 5b).

Fig. 5 Adhesive force variations of the smooth films and the porous films with different pore diameters in acid solution and basic solution. (a) Same sample comparisons in different solutions and (b) same pH solution comparisons of different samples.

Mechanism affecting the adhesive force

In general, the adhesion of a surface can be governed by both the chemical composition and geometrical microstructures.¹⁷ From a horizontal comparison of Table 2, it can be seen that the adhesion of the smooth film and the honeycomb structure film with the theoretical diameter pores of 4 μm is larger in acidic solution than in basic solution. At low pH, intramolecular hydrogen bonding in PAA is formed, and results in high oil adhesion. As for high pH, intermolecular hydrogen bonding between PAA and the surrounding water is formed, which decreases the oil adhesion. The adhesion of the honeycomb structure films with the theoretical diameter pores of 6 μm, 8 μm and 10 μm is smaller in acid solution than in basic solution. But from a vertical comparison, we can see that the adhesion decreases with the increase of the pore size in acid solution and increases with the increase of the pore size in basic solution. This unusual adhesion phenomenon is caused by the special pore structure (Fig. 6). As for the acid solution, the adhesion decrease with the increase of the pore size is mainly caused by a decreasing continuity of the three-phase contact line (TCL)^18,19 (Fig. 6a). The increase of the adhesion on the honeycomb film in basic solution is mainly caused by the decrease of the pore size (Fig. 6b). The only possible explanation is: when an oil droplet is placed on the honeycomb surface, there are sealed pockets of water trapped in the pores (Fig. 6c). Because capillary water exists in the pore, the surface of the oil/water is a meniscus (Fig. 6c). To some extent, the oil droplet is an elastic body. When we applied an external force to it, a deformation occurred along the direction of the external force. We also observed droplet deformation along the direction of the force during the adhesion measurement process. Thus, when the droplet is gradually retracted from the sample surface, the meniscus on each pore is changed from concave to convex. This can result in an increased volume of water sealed in each pore by the liquid/liquid interface. A pore with a smaller diameter has a higher expansion ratio, so the negative pressure would be rather large. That is, the force produced by this is large. In this case, the volume of water sealed in a pore varies with its diameter. Larger pores would be expected to have lower expansion ratios, and thus a lower negative pressure. From the above analysis, we can see that a single honeycomb structural pore with a small diameter would require a large droplet pulling-off force, and the density is higher when the diameter is smaller. Therefore, the total surface adhesive force would be larger for the honeycomb structure film with a smaller pore diameter in basic solution.

Fig. 6 Effect of the surface structure change on the oil adhesive behavior of the honeycomb film in different pH solutions. (a) The gecko state between the Wenzel and the Cassie states. The oil droplet partially wets the rough features that continue to trap water on the surface. The TCL is discontinuous in pH 1.92 solution. (b) TCL becomes more continuous in the pH 11.12 solution than that of the pH 1.92 solution. (c) Schematic illustrations of the interfaces between the oil and a single honeycomb structural pore, and the volume change of the sealed water in one PAA pore upon the action of an external force. Capillary adhesion arises when a oil droplet sitting on the pore is gradually drawn upward because the convex oil/water interface produces an inward pressure ΔP. W represents a water droplet.

Potential applications for liquid transfer underwater

In our work, the porous film with 6 μm theoretical diameter pores in basic solution shows the highest adhesive force for oil droplets. Thus this film can be used as a “mechanical hand” to transfer a certain amount of oil droplets from a superoleophobic surface to a oleophilic surface (Fig. 7a). At first an oil droplet of 5 μL was placed on the superoleophobic square-pillar-structured silicon substrate in basic solution (Fig. S2a†) (step 1), then the porous film with 6 μm theoretical diameter pores was moved to come into contact with the oil droplet (step 2). The whole oil droplet was completely adhered to the porous film because of its high adhesive force for the oil in basic solution (step 3). At last, the oil droplet was transferred to the oleophilic square-pillar-structured silicon substrate (Fig. S2b†) by touching (steps 3 and 4). Similarly, oil droplet transfer from an oleophobic smooth PAA surface to an oleophilic surface (Fig. 7b) can been achieved using the porous film with 6 μm theoretical diameter pores in basic solution as a “mechanical hand”. Therefore, oil transfer underwater from the superoleophobic or oleophobic surface to the oleophilic surface was successfully achieved via the porous PAA film.

Fig. 7 Transfer of (a) oil droplet from the low adhesive superoleophobic square-pillar-structured silicon substrate to the high adhesive honeycomb structure PAA surface in pH 11.12 solution, and then to another square-pillar-structured silicon substrate with oleophilic properties; (b) oil droplet from the high adhesive oleophobic smooth PAA surface to the more highly adhesive honeycomb structure PAA surface in pH 11.12 solution, and then to another square-pillar-structured silicon substrate with oleophilic properties. 5 μL oil droplets were used in this process.

Conclusions

A honeycomb-like PAA surface with high oil adhesion underwater was achieved. Characterization of the adhesion indicated that the porous honeycomb structure PAA films can serve as high adhesive surfaces both in acidic and in basic aqueous phases. Besides, the adhesion can be controlled by changing the size of the pore and the solution pH. The switchable oil adhesion is attributed to regulation of the TCL continuity and the negative pressure induced by the pores. The unique adhesive phenomenon of the honeycomb structure will be useful for manipulating oil droplet behavior and suitable for the application of controlling liquid collection and transportation underwater. This highly adhesive porous film was used as a “mechanical hand” to transfer micro-droplets successfully underwater. These findings will help us to design novel highly adhesive materials in multiphase systems, which will be used for the construction of the future generation of microdevices and will have potential applications in liquid transportation, biochemical separation, in situ detection, microfluid systems, and other areas. This study also gives us a more profound understanding of the wettability of solid surfaces underwater, which has a bright future in underwater bionic applications.

Acknowledgements

This work was supported by the National Research Fund for Fundamental Key Projects (2014CB931802 and 2013CB834705).

Notes and references

1. M. J. Liu, S. T. Wang, Z. X. Wei, Y. L. Song and L. Jiang, Adv. Mater., 2009, 21, 665–669.
2. M. J. Liu, Y. M. Zheng, J. Zhai and L. Jiang, Acc. Chem. Res., 2010, 43, 368–377.
3. M. Nosonovsky and B. Bhushan, Philos. Trans. R. Soc., A, 2009, 367, 1511–1539.
4. Y. C. Jung and B. Bhushan, Langmuir, 2009, 25, 14165–14173.
5. L. Chen, M. J. Liu, H. Bai, P. P. Chen, F. Xia, D. Han and L. Jiang, J. Am. Chem. Soc., 2009, 131, 10467–10472.
6. Y. Huang, M. J. Liu, J. X. Wang, J. M. Zhou, L. B. Wang, Y. L. Song and L. Jiang, Adv. Funct. Mater., 2011, 21, 4436–4441.
7. L. P. Xu, J. Zhao, B. Su, X. L. Liu, J. T. Peng, Y. B. Liu, H. L. Liu, G. Yang, L. Jiang, Y. Q. Wen, X. J. Zhang and S. T. Wang, Adv. Mater., 2013, 25, 606–611.
8. X. L. Liu, J. Zhou, Z. X. Xue, J. Gao, J. X. Meng, S. T. Wang and L. Jiang, Adv. Mater., 2012, 24, 3401–3405.
9. L. Chen, M. J. Liu, L. Lin, T. Zhang, J. Ma, Y. L. Song and L. Jiang, Soft Matter, 2010, 6, 2708–2712.
10. Q. F. Cheng, M. Z. Li, F. Yang, M. J. Liu, L. Li, S. T. Wang and L. Jiang, Soft Matter, 2012, 8, 6740–6743.
11. C. M. Ding, Y. Zhu, M. J. Liu, L. Feng, M. X. Wan and L. Jiang, Soft Matter, 2012, 8, 9064–9068.
12. M. C. Li, B. Wang, L. P. Heng and L. Jiang, Adv. Mater. Interfaces, 2014, 1400298.
13. L. P. Heng, X. F. Meng, B. Wang and L. Jiang, Langmuir, 2013, 29, 9491–9498.
14. M. H. Jin, X. J. Feng, L. Feng, T. L. Sun, J. Zhai, T. J. Li and L. Jiang, Adv. Mater., 2005, 17, 1977–1981.
15. X. Yao, J. Gao, Y. L. Song and L. Jiang, Adv. Funct. Mater., 2011, 21, 4270–4276.
16. A. B. D. Cassie and S. Baxter, Trans. Faraday Soc., 1944, 40, 546.
17. W. Lee, B. G. Park, D. H. Kim, D. J. Ahn, Y. Park, S. H. Lee and K. B. Lee, Langmuir, 2010, 26, 8233–8238.
18. L. C. Gao and T. J. McCarthy, Langmuir, 2007, 23, 3762–3765.
19. W. Chen, A. Y. Fadeev, M. C. Hsieh, D. Öner, J. Youngblood and T. J. McCarthy, Langmuir, 1999, 15, 3395–3399.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra10514e

---