Rapid deposition of superhydrophilic stalagmite-like protrusions for underwater selective superwettability

Abstract

Selective superwettability has attracted much attention for various applications, especially for fabricating membranes for oil-water separation. However, most approaches were conducted using complicated chemical reactions or processes. In the present study, we developed a facile one-step method to coat hierarchically stalagmite-like protrusions on stainless steel mesh which was repellent to not only oil, but various organic solvents underwater. The modified stainless steel mesh, prepared using an atmospheric pressure plasma technique, showed superhydrophilicity without any further surface elaboration. Due to its strong affinity toward water molecules, it enables the stable entrapment of a water layer on the surfaces of mesh wires against oils. High underwater static contact angles and extremely low roll-off angles towards various oil were exhibited. Notably, the selective superwettability was well preserved even under harsh environments, such as saline water and strong acid solution. Satisfactory separation efficiency for oil-water mixtures was demonstrated in this study. Furthermore, the effect of surface roughness in terms of intrusion pressure was investigated. The effect of surface roughness on intrusion pressure presented here can be applied for various coating materials used in oil-water separation.

---

Introduction

Separation of oil from oily wastewater has recently caught much attention^1,2 because oil pollution has frequently happened, such as from domestic sewage, industrial wastewater, and offshore oil spills. From the 1907 Thomas W. Lawson oil spill to the latest 2016 Shell Gulf of Mexico oil spill,³ these catastrophic accidents have never ended. The toxic chemicals and the oil-contaminated water cause serious problems to wildlife and result in potential risk to human health as well. Consequently, it is highly desirable to develop an advanced methodology for effectively separating oil and water.

For conventional oily wastewater treatments, the method of gravity separation followed by skimming⁴ is regarded as a cost-effective way to eliminate the free oil. However, the commercial membranes hardly remove smaller oil droplets dispersing in water. Nowadays, the advanced materials utilized for oil-water separation can be classified into three types of surfaces, superhydrophobic/superoleophilic surfaces,^5–9 superhydrophilic/superoleophobic surfaces^10–12 and underwater superoleophobic surfaces.^13–18 Through these rational designs of surface texture and chemical composition, satisfactory efficiency of oil-water separation can be achieved. However, the superhydrophobic/superoleophilic surfaces are commonly contaminated by oil or oil-based contaminants. Additionally, because of the higher density of water, the settlement of water will form a barrier layer against oil permeation which is not suitable for gravity separation. On the other hand, superhydrophilic/superoleophobic surfaces could be the ideal candidate for separation of water from oil¹⁹ whereas the fabrication processes are usually required complicated chemical reactions. As for the underwater superoleophobic surfaces, several natural surfaces, such as shark skin²⁰ and the backside surface of lotus leaf,²¹ display more resistant to oily contamination.¹⁰

Recently, the work done by Waghmare et al.²² indicated that the underwater superoleophobicity of fish scale is not solely contributed by the hierarchically structured surface of scales. The synergistic effect of geometrical asperities and the mucus layer provides excellent oil-repellent property of fish scales. Without the fish's mucus layer, the performance has been evidenced to be obviously compromised. Inspired by the mechanism of fish scales, we report a rapid and facile method to fabricate a highly efficient and robust coating for gravity-driven oil-water separation by atmospheric pressure plasma technique.

In our previous work,²³ we proposed a fabrication method that can synthesize a hierarchically stalagmite-like tungsten oxide protrusions with good uniformity within a few seconds. Here we utilized this technique and demonstrated superior and long-term stable superhydrophilicity on coated stainless steel mesh. No further chemical decoration was conducted. Excellent underwater superoleophobicity toward various organic solvents was observed after pre-wetted the surface by water. The synthesized surfaces exhibited high efficiency for extracting water from the oil-water mixture. Due to strong surface-coating attachment of the texture, the coating preserved the oil-repellent property under harsh environments. The role of nanostructures on the surface is studied in this research. Moreover, how nanostructures affect the oil-water separation process, which is scarcely discussed in related researches, was also investigated.

Experimental

Materials

Stainless steel meshes (SS316 grade, 26 μm aperture size with 0.025 mm wire diameter, 45 μm aperture size with 0.04 mm wire diameter, and 154 μm aperture size with 0.1 mm wire diameter) and soybean oil were purchased from local vendors. Hexane (>99%, Sigma Aldrich, U.S.A), n-dodecane (95%, Tedia, U.S.A), sulfuric acid (95–98%, Sharlau, Spain), sodium chloride (99.5%, Showa, Japan) were used as received without further purification. For the clear observation during gravity-driven separation process, water was stained by methylene blue trihydrate (95%, Acros Organics, USA) and the organic solvent droplets were stained by Oil Red O (Acros Organics, USA).

Methods

Fabrication of underwater superoleophobic surfaces. The stainless steel meshes (2.5 × 2.5 cm²) were cleaned by acetone, ethanol and deionized water and dried by the air gun, prior to plasma treatment. The flat silicon substrates (1.5 × 1.5 cm²) were cleaned by diluted hydrofluoric acid (2%) and deionized water. For entry the chamber, the cleaned substrates were fixed on the stage by polyimide film silicone tapes (3M™ Polyimide Film Tape 5413). A bluish glow discharge appeared from the apex of nozzle when injecting the working gas with pure argon (Ar) or 0.2% of O₂/Ar and applying power of 50 W. The plasma jet, approximately 2 mm in diameter and 1 cm in length, is generated in the nozzle and blown out by the gas flow. The plasma jet scanned through the whole area of meshes under identical scanning rate (0.8 s cm⁻²). The distance between the jet and the substrate is controlled at 2 mm. After five to eleven scanning repetitions, uniform coatings with stalagmite-like protrusions were synthesized on the surfaces of stainless steel meshes and flat silicon substrates.

Material characterizations. The microstructure and morphology of coatings were characterized by field emission scanning electron microscopy (FE-SEM) (HITACHI SU8010, Japan), and transmission electron microscopy (TEM) (JOEL, JEM-2010, Japan). As for the chemical composition analyses of surface coating, X-ray photoelectron spectroscopy (XPS) (PHI 5000 VersaProbe II, USA) was performed using the C 1s peak energy (284.6 eV) as calibrated energy standard. The optical microscopy was utilized to examine the filtered liquids for remaining free oil droplets or emulsions.

Contact angle measurements (in air and underwater). The static contact angles (SCA) on the as-prepared sample in air were measured from sessile drops by a goniometer (First Ten Angstrom1000, USA) equipped with a tilting stage. The underwater SCA (UWSCA) of soybean oil, hexane and n-dodecane were measured in a glass cuvette. The as-synthesized stainless steel mesh was attached on a Teflon plate and put upside down to immerse the substrate into liquid solutions. The oil and organic solvent droplets (approximately 6 μL, 2.2 mm in diameter) were placed slowly by a J-shaped needle (0.914 mm in diameter). At least three individual measurements were performed on each tested sample.

Gravity oil-water separation. The as-synthesized stainless steel mesh coated with tungsten oxide stalagmite-like protrusions was placed in a glass filtration apparatus (Advantec, USA). The filter holder was 25 mm in diameter. In order to prevent leakage, the contact area between the glass funnel and the edge of support glass was covered with a layer of high vacuum grease (Dow Corning®, USA). The mixture of oil-water, oil-sulfuric-acid and oil-saline-water were prepared by mixing under 50% v/v ratio. The total weight of tested liquid is controlled under approximately 18 g. The sulfuric acid solution contained 20 vol% of acid, and the saline water was composed of 3.5 wt% sodium chloride in deionized water.

Results and discussion

Superhydrophilic surfaces for underwater superoleophobicity

According to the Young's equation,²⁴ the UWSCA of a flat surface can be expressed by eqn (1): where θ_ow is the UWSCA of oil on a flat surface. γ_sw, γ_so, γ_oa, γ_wa and γ_ow represent the surface tensions of the solid/water, solid/oil, oil/air, water/air and oil/water interfaces, respectively. θ_oa and θ_wa are the SCA of oil and water in air on a flat surface, respectively. To enhance the underwater oleophobicity (i.e. the value of θ_ow is increased), the value of γ_oa cos θ_oa has to be larger than that of γ_wa cos θ_wa. Therefore, a surface with stronger affinity toward water should be expected to obtain better underwater superoleophobicity.

Fig. 1a illustrates that two types of superhydrophilic surfaces can be rapidly synthesized through APP technique. In general, the plasma jet is known to effectively introduced hydroxyl groups upon the surface, and renders the sample superhydrophilic after plasma treatment. As shown in our previous research,²³ a hierarchically structured superhydrophilic coating can be directly grown on the substrate (SCA < 10°) by tuning the composition of working gas. Consider oil and water are naturally immiscible, here, we developed an underwater superoleophobic surface by creating a thin water layer upon the two types of superhydrophilic surfaces. After deposited a few sessile water droplets, the surfaces would be wetted immediately and formed a smooth and homogeneous water layer. As depicted in Fig. 1b, when the oil droplet is placed on the water film, the oil droplet will adjust the shape to minimize the surface energy afterwards.

Fig. 1 Schematics of (a) the fabrication processes for superhydrophilic silicon surfaces through the atmospheric pressure plasma (APP) technique, and (b) the mechanism for underwater selective superantiwettability.

We began by demonstrating this concept on the flat silicon surface. The silicon surface with hierarchical nanostructures was synthesized by introducing the oxygen mixed argon gas as working gas. The gas flow rates of argon and oxygen were measured to be 15 standard liter per minute (slm) and 40 standard cubic centimeters per minute (sccm), respectively. Argon and oxygen mixed before supplying to the plasma jet were fed by using separate pipes and flow meters to control the optimal flow rate. In the plasma atmosphere, frequent collisions between the tungsten electrode and oxygen radicals in the generated plasma jet formed excited tungsten ions. Those excited tungsten ions transferred to tungsten through spontaneous redox reaction and self-assembled into stalagmite-like tungsten protrusions. The as-synthesized coating was further oxidized from the outer to the inner layers due to the subsequent supplement of oxygen flow and the elevated temperature of plasma jet. After treating the surface for 55 s, which is the optimized condition in terms of underwater superoleophobicity (Fig. S1†), a uniform tungsten oxide coating was synthesized.

To confirm the chemical composition of the synthesized surface under oxygen mixed argon gas plasma, the XPS full spectrum as illustrated in Fig. 2a (blue) revealed the characteristic peaks of W 4f and W 4d. These two peaks were observed at 35.8 and 38.1 eV, corresponding to the valence of W⁶⁺ (ref. 25) based on the XPS spectrum in Fig. 2b. The composition of surface was identified as tungsten oxide (WO₃). The loss feature of WO₃ was also detected at 41.8 eV. According to the TEM image (Fig. 2c), the as-synthesized tungsten oxide surface is made of nanoparticles with approximately 5 to 10 nm in diameter. On the other hand, to further explore the role of nanostructures in oil-water separation, we prepared a superhydrophilic surface with flat morphology by treating the surface of silicon with pure argon plasma under the flow rate of 15 slm. Comparing to the XPS full spectrum of cleaned silicon substrate (Fig. 2a, black), the characteristic peak of O 1s for the silicon treated by pure argon plasma (Fig. 2a, red) obtained higher intensity. The difference in the intensities is believed to be caused by the hydroxyl groups under plasma treatment. Additionally, for the silicon substrate treated by pure argon, there was no appearance for the characteristic peaks of W 4f and W 4d in the XPS spectrum.

Fig. 2 (a) XPS full spectrum of the three different types of surfaces. (b) XPS spectra of W 4f for the silicon surface treated by oxygen mixed argon plasma. (c) A TEM image of the tungsten oxide nanoparticles deposited by APP technique under the working gas of oxygen mixed argon gas (O₂/Ar).

The corresponding morphologies and UWSCA toward soybean oil of the three aforementioned surfaces are shown in Fig. 3. The cleaned silicon with flat surface (Fig. 3a) exhibited approximately 20% of UWSCA lower than the value for both of the surfaces treated by plasma. Fig. 3b and c revealed that both of the samples obtained high UWSCA, and we did not observe any significant difference between the two samples. Moreover, except the sample of cleaned silicon, the strong underwater superoleophobicity was successfully preserved when the oil sessile droplets were compressed by J-shaped needle on the two plasma-treated samples (Fig. 3d). The oil droplets left no residue when removed by J-shaped needle. We attributed the low adhesion between substrates and oil droplets to the thin water layer trapped upon the surface, which contributes low roll-off angles (<2°) towards soybean oil droplets (ESI Video S1†).

Fig. 3 SEM image of (a) a cleaned silicon surface (underwater static contact angle, UWSCA = 127° ± 0.8°), (b) a silicon surface treated by pure argon plasma (UWSCA = 160° ± 0.5°), and (c) a hierarchically structured silicon surface treated by O₂/Ar plasma (UWSCA = 164° ± 0.8°). (d) The photos for an attachment testing between soybean oil sessile drops upon the three types of substrates underwater.

Robust underwater superoleophobicity

To evaluate the long-term stability for the superhydrophilicity in air, which is considered to correlate well to the capability of entrapping the thin water layer, three kinds of samples, cleaned silicon substrate, silicon treated by pure argon plasma, and silicon treated by oxygen mixed argon plasma, were placed inside the chamber under approximately 50% relative humidity atmosphere at room temperature. Fig. 4 illustrates the effect of aging on the SCA toward water. The SCA were measured for 7 days, 14 days, and 28 days with a 5 μL water droplet to evaluate the effect of aging test. When the cleaned silicon surface was exposed to the ambient condition, a thin native silica layer, which is intrinsically hydrophilic, was easily formed on the surface of silicon at room temperature, and the SCA stably stayed at 75° ± 1.4° for 28 days (Fig. 4, blue). The SCA was slightly decreased due to the formation of native oxide of silicon on the surface. However, the silicon treated by pure argon exhibited obviously increasing SCA after aging process (Fig. 4, red). More than 30° of SCA was increased due to the interaction between reactive oxygen species and air.²⁶ The adsorption of water molecules or organic contaminants induced the reduction of surface energy and consequently an increase of SCA occurred after a sufficient aging time.

Fig. 4 The aging stability of superhydrophilicity in air for the three samples: cleaned silicon substrate (blue), silicon treated by pure argon plasma (red) and silicon treated by argon gas mixed oxygen plasma (black).

As for the silicon treated by oxygen mixed plasma, the SCA increased from approximately 0° to 5° ± 0.2° for 14 days, and maintained the SCA below 8° even for 28 days. As a result of the electron-deficient of tungsten atoms (W: [Xe] 6s² 4f¹⁴ 5d⁴) and polar covalent bonding between tungsten and oxygen atoms which increases the surface polarity, the surface exhibited strong affinity toward water. The roughness (r) of the nanostructured surface further enhances the intrinsic property of surface based on the theory of Wenzel state (cos θ* = r cos θ).^27,28 Consequently, the surface with WO₃ nanostructures is rendered superhydrophilic with satisfactory long-term stability.

To explore the potential applications of the stalagmite-like tungsten oxide coatings, we evaluated the underwater superoleophobicity toward different organic solvents and in harsh environments. The stainless steel meshes (26 μm aperture size) coated with superhydrophilic stalagmite-like tungsten oxide protrusions were prewetted by water before immersing into the various aqueous solutions, such as deionized water, acidic solution and saline water. After keeping the meshes underwater for 2 hours, 6 μL of organic solvents was deposited by J-shaped needle. Fig. 5 reveals that the stalagmite-like tungsten oxide coating provided satisfactory underwater superoleophobicity against soybean oil (γ = 31.4 mN m⁻¹), hexane (surface tension γ = 18.43 mN m⁻¹) and n-dodecane (γ = 25.35 mN m⁻¹). The corresponding UWSCA were 153° ± 2°, 164° ± 4°, and 161° ± 7°, respectively. Additionally, comparing to the UWSCA in water, the values of UWSCA in both sulfuric acid solution and saline water did not show any degradation. The corresponding UWSCA were 168° ± 1.3° and 162° ± 5°, respectively. Because of the extremely low adhesion between tested oil droplets and coated meshes, all the UWSCA values were measured from the pendant oil drop (i.e. the J-shaped needle was in the drop). The results indicated that the underwater superoleophobicity can be well-preserved in such harsh environment.

Fig. 5 The evaluation of underwater superoleophobicity of synthesized tungsten oxide coating against soybean oil and two kinds of organic solvents in different aqueous solutions.

Moreover, the coated meshes were applied for separating the oil from harsh environments. The corresponding separation efficiencies (η) for various oils and solvents were calculated based on eqn (2) by:

η = (m₁/m₀) × 100% (2)

where m₀ and m₁ are the mass of the oil before and after separation process, respectively. High separation efficiency ranged from 96% to 99% can be achieved (Table S1†). The superior performance of oil-water separation was also demonstrated by optical microscopy as shown in Fig. S2.† No oil or organic solvent droplets appeared in the filtered solution. The flux (F_W) of water permeation through the coated stainless steel mesh was also calculated to be 346 mL m⁻² s⁻² by F_W = V/At, where V is the volume that passed through the mesh. A represents the effective area of mesh and t is the time for the permeation of the entire volume of water.

Furthermore, to evaluate the mechanical stability, we characterized the morphologies of the coated meshes after separation process. Before characterization, the coated mesh was washed by water stream (approximately 20 mL s⁻¹) for 30 s after oil-water separation process for one cycle. As shown in Fig. S3,† the fine structures of asperities of stalagmite-like tungsten oxide protrusions on the coated mesh were partially damaged after the separation and washing processes. However, because of the wavy structures of stainless steel mesh, the hierarchically structured tungsten oxide protrusions can still maintain on the side walls of the metal wires. Additionally, the rough layer between the metal wire and the stalagmite-like tungsten oxide protrusions seems to maintain satisfactory adhesion which provides the superhydrophilicity and underwater superoleophobicity for the mesh.

The effect of surface roughness and pore size

Considering the good underwater superoleophobicity of the stalagmite-like tungsten oxide coating, we further investigated the capability of oil-water separation and the effect of surface roughness and pore sizes. The oil-water separation system was developed for this research as shown in Fig. S4.† Three types of stainless steel meshes were prepared, bare mesh, mesh treated by pure argon plasma, and mesh treated by Ar/O₂ plasma. Prior to the oil-water separation, all the tested meshes were pre-wetted by deionized water. A mixture of oil or organic solvents (dyed with Oil Red O) and water or other aqueous solution (dyed with methylene blue trihydrate) was poured in the glass funnel which was contacted with the pre-wetted mesh. The methodology was expected to be energy efficient because of the gravity-driven process.

The surface of bare mesh was contaminated by oil droplet in the water and both the water and oil permeated through the mesh immediately (Fig. S4a†). Even though the mesh was rinsed by deionized water after filtration process, the permeation area on the bare mesh appeared reddish because of the oil residue, which can be seen from the image in Fig. 6a. As for the mesh treated by argon plasma which obtained underwater superoleophobicity but without surface textures, the water first permeated through the mesh, and the oil was left in the glass funnel at the beginning. However, after approximately two minutes, the oil permeated rapidly through the mesh and contaminated the filtered water again (Fig. S4b†). Distinguished from other tested meshes, the mesh with stalagmite-like tungsten oxide protrusions exhibited high oil repellency. It sustained for at least three hours to prevent from the permeation of oil (Fig. S4c†). As shown in Fig. 6b and c, the two meshes with underwater superoleophobicity did not appear the reddish oil residue after rinsed the meshes by deionized water.

Fig. 6 SEM images for the comparison of morphology among the three meshes: (a) bare mesh (blue), (b) mesh treated by pure argon plasma (red) and (c) mesh treated by oxygen mixed argon plasma. And the photos for the appearance of the filtration area circled by white dash lines. The outer circular ring area between the two dash lines is the contact area between the glass funnel and the holder.

Based on our observation, even though underwater oleophobicity plays a critical role in oil-water separation, it is not sufficient to achieve high efficiency for oil-water separation. The water layer entrapped by superhydrophilic meshes has to be stable enough to withstand the hydrostatic pressure (P_int = ρgh) which is created by gravity from the tested solution with density ρ and height h. According to the results reported by Gondal et al.,¹³ when the pore size was decreased to less than 100 μm, the separation efficiency can be improved dramatically. The water film formed between the individual mesh wires caused by capillary force of water²⁹ acted as channels for water permeation, whereas the oil was left behind because of the barrier induced by the intrusion pressure. Based on the Young–Laplace equation,^16,30–33 the balanced Laplace pressure in oil-underwater system can be defined by eqn (3):

where ΔP is pressure difference and r represents the pore radius of mesh. Smaller pore size resulted in larger pressure difference. It contributes more stable contact lines of water, oil and stainless steel mesh. However, eqn (3) cannot explain the phenomena we observed in Fig. S4b and S4c.† We discovered that the two meshes with equal pore sizes but different surface morphologies resulted in different values of ΔP. It seems that the mesh with rougher surface leads to higher value of ΔP.

Considering the effective pore radius should not be constant on a rough surface, the pressure difference in eqn (3) must be modified to eqn (4) by the effective pore radius (r_eff): ³⁴

where r_eff = 2(1 − φ)/φρA, and φ is the volume fraction of solid material. ρ represents the density of solid material, and A is the specific surface area per gram of solid. To calculate the difference between the two values of ΔP_eff for underwater superoleophobic meshes with and without surface textures, we assumed the change of the volume fraction by depositing the tungsten oxide protrusions can be neglected because the thickness of hierarchically structured coating was controlled under approximately 150 nm (Fig. S5†). We estimated the specific surface area of the meshes with and without hierarchically structured tungsten oxide protrusions by atomic force microscopy. The value of A for the coated mesh was measured as approximately four times higher than that of mesh without surface textures (Fig. S6†). It can be considered to be the lower limit because the hierarchically structured tungsten oxide layer contains nano-sized pores within the structure as shown in Fig. S5a.† Consequently, the mesh modified with stalagmite-like tungsten oxide protrusions can withstand higher hydrostatic pressure from oil. On the other hand, the surface roughness contributes the strong affinity toward water. The water layer trapped within the rough textures, therefore, can perfectly isolate the surface of mesh from the oil droplets. Without surface roughness, the mesh surface is easily contaminated by oil droplet, and the oil transfers to the other side of the mesh afterwards.

To explore the effect of pore size on the oil-water separating process, we selected the stainless steel meshes with three different aperture sizes, 26 μm, 45 μm, and 154 μm, for further testing. After plasma treatments, all the meshes were coated with stalagmite-like tungsten oxide protrusions on the surface, as shown in Fig. 7a to c. The corresponding UWSCA of coated meshes were measured three times for each sample. The coated mesh with 26 μm aperture size showed the highest value of UWSCA among the samples (Fig. 7d). The corresponding separation efficiencies were measured, as shown in Fig. 7e. When the pore size was increased to 154 μm, both water and oil permeated through the mesh. When the pore size was decreased to 45 μm, the efficiency can be improved to approximately 76%. The stainless steel mesh with 26 μm pore size exhibited the highest efficiency which is approximately 99%. According to our observations, when the pore size is larger than the threshold value (i.e. 26 μm), it dominates the separation efficiency because the volume fraction of mesh was obviously decreased so that ΔP_eff was significantly downgraded. Once the pore size reach the value which is closer to the threshold value, the effect of surface roughness becomes more significant. Moreover, the intrusion pressure was measured by P = ρgh_max, where ρ is the density of soybean oil, g is the acceleration of gravity, and h_max represents the maximum height of oil that the coated mesh can sustain. The intrusion pressure for the mesh with 26 μm pore size is approximately 684 Pa, which is 1.6 times higher than that of the mesh with 45 μm pore size. However, once the pore size was increased to 154 μm, the oil directly permeated through the mesh.

Fig. 7 SEM images for coated stainless steel meshes with pore sizes of (a) 154 μm, (b) 45 μm and (c) 26 μm. Inset images are the stalagmite-like tungsten oxide protrusions deposited on the meshes. (d) The corresponding separation efficiencies for the mixture of soybean oil and water of different coated meshes. (e) The corresponding underwater static contact angles for different meshes.

Based on these results, we have evidenced that larger specific surface area and smaller pore sizes result in not only higher intrusion pressure but more satisfactory stability of underwater superoleophobicity.

Conclusions

In summary, we have developed a rapid method for fabricating a superhydrophilic tungsten-oxide coating with hierarchically stalagmite-like protrusions. The strong affinity toward water induced long-term underwater superoleophobicity for oil and organic solvents. Additionally, a uniform hierarchically structured tungsten oxide coating on stainless steel mesh showed high oil-water separation efficiency, which was not only demonstrated in normal water system but in harsh environments, such as strong acid solution and saline water. Compared to the hydrophilic surface without surface texture, the stainless steel mesh coated with stalagmite-like tungsten oxide protrusions obtained higher intrusion pressure against the permeation of oil. We attributed the satisfactory intrusion pressure to the suitable pore size and surface roughness which provide large specific area. The extremely low adhesion between the organic oils and the prewetted mesh contributed the robustness of the underwater superoleophobicity. As a result, we believe that the robust underwater superoleophobic coating in this study will enable a wide range of practical applications in the future. Moreover, the relationship between surface area and intrusion pressure underwater discussed here will shed a light on rational designs for the membranes for oil-water separation.

Acknowledgements

The authors sincerely thank the financial support from the Ministry of Science and Technology, Taiwan (MOST103-2221-E-007-034-MY3 and MOST105-2221-E-007-012-MY4), Precious Instrument Utilization Center sponsored by the MOST, and technical support from Wei-Chen Hung (National Tsing-Hua University, Taiwan), and the Industrial Technology Research Institute (ITRI-MCL-D254W0B100). Ching-Yu Yang especially thanks Dr Chih-Chung Lai (National Synchrotron Radiation Research Center, Taiwan) for the fruitful discussions and constructive suggestions.

References

1. B. Wang, W. Liang, Z. Guo and W. Liu, Chem. Soc. Rev., 2015, 44, 336–361.
2. Z. Chu, Y. Feng and S. Seeger, Angew. Chem., Int. Ed., 2015, 54, 2328–2338.
3. The Huffington Post, http://www.huffingtonpost.com/entry/shell-oil-spill-gulf-mexico_us_57353058e4b060aa7819ee00, accessed Aug, 2016.
4. M. Cheryan and N. Rajagopalan, J. Membr. Sci., 1998, 151, 13–28.
5. L. Feng, Z. Zhang, Z. Mai, Y. Ma, B. Liu, L. Jiang and D. Zhu, Angew. Chem., 2004, 116, 2046–2048.
6. J. Zhang and S. Seeger, Adv. Funct. Mater., 2011, 21, 4699–4704.
7. X. Gui, J. Wei, K. Wang, A. Cao, H. Zhu, Y. Jia, Q. Shu and D. Wu, Adv. Mater., 2010, 22, 617–621.
8. S. Wang, M. Li and Q. Lu, ACS Appl. Mater. Interfaces, 2010, 2, 677–683.
9. J. Li, R. Kang, X. Tang, H. She, Y. Yang and F. Zha, Nanoscale, 2016, 8, 7638–7645.
10. A. K. Kota, G. Kwon, W. Choi, J. M. Mabry and A. Tuteja, Nat. Commun., 2012, 3, 1025.
11. P. S. Brown and B. Bhushan, Sci. Rep., 2015, 5, 8701.
12. H. Yoon, S.-H. Na, J.-Y. Choi, S. S. Latthe, M. T. Swihart, S. S. Al-Deyab and S. S. Yoon, Langmuir, 2014, 30, 11761–11769.
13. M. A. Gondal, M. S. Sadullah, M. A. Dastageer, G. H. McKinley, D. Panchanathan and K. K. Varanasi, ACS Appl. Mater. Interfaces, 2014, 6, 13422–13429.
14. J. Li, H. M. Cheng, C. Y. Chan, P. F. Ng, L. Chen, B. Fei and J. H. Xin, RSC Adv., 2015, 5, 51537–51541.
15. Y. Dong, J. Li, L. Shi, X. Wang, Z. Guo and W. Liu, Chem. Commun., 2014, 50, 5586–5589.
16. Z. Xue, S. Wang, L. Lin, L. Chen, M. Liu, L. Feng and L. Jiang, Adv. Mater., 2011, 23, 4270–4273.
17. J. Li, L. Yan, H. Li, W. Li, F. Zha and Z. Lei, J. Mater. Chem. A, 2015, 3, 14696–14702.
18. J. Li, D. Li, Y. Yang, J. Li, F. Zha and Z. Lei, Green Chem., 2016, 18, 541–549.
19. S. Sinha, K. A. Mahmoud and S. Das, RSC Adv., 2015, 5, 80184–80191.
20. G. D. Bixler and B. Bhushan, Nanoscale, 2013, 5, 7685–7710.
21. Q. Cheng, M. Li, Y. Zheng, B. Su, S. Wang and L. Jiang, Soft Matter, 2011, 7, 5948–5951.
22. P. R. Waghmare, N. S. K. Gunda and S. K. Mitra, Sci. Rep., 2014, 4, 7454.
23. C.-Y. Yang, S.-I. Chuang, Y.-H. Lo, H.-M. Cheng, J.-G. Duh and P.-Y. Chen, J. Mater. Chem. A, 2016, 4, 3406–3414.
24. T. Young, Philos. Trans. R. Soc. London, 1805, 95, 65–87.
25. J. Wang and C. K. Ober, Macromolecules, 1997, 30, 7560–7567.
26. L. Sang-Joon, P. Bu-Geun, K. Guk-Bae and J. Young-Gil, Jpn. J. Appl. Phys., 2006, 45, 912.
27. R. N. Wenzel, Ind. Eng. Chem., 1936, 28, 988–994.
28. A. Tuteja, W. Choi, M. Ma, J. M. Mabry, S. A. Mazzella, G. C. Rutledge, G. H. McKinley and R. E. Cohen, Science, 2007, 318, 1618–1622.
29. R. J. Good and M. Islam, Langmuir, 1991, 7, 3219–3221.
30. C. Journet, S. Moulinet, C. Ybert, S. T. Purcell and L. Bocquet, Europhys. Lett., 2005, 71, 104.
31. Y. Yuan and T. R. Lee, Surface Science Techniques, Springer, Berlin, Heidelberg, 2013.
32. P. M. Heertjes and W. C. Witvoet, Powder Technol., 1969, 3, 339–343.
33. P. G. de Gennes, F. Brochard-Wyart and D. Quere, Capillarity and Wetting Phenomena: Drops, Bubbles, Pearls, Waves, Springer, New York, 2003.
34. L. R. White, J. Colloid Interface Sci., 1982, 90, 536–538.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra20073g

---