Transparent, abrasion-insensitive superhydrophobic coatings for real-world applications

Superhydrophobic surfaces and surface coatings are of high interest for many applications in everyday life including non-wetting and low-friction coatings as well as functional clothing. Manufacturing of these surfaces is intricate since superhydrophobicity requires structuring of surfaces on a nano- to microscale. This delicate surface structuring makes most superhydrophobic surfaces very sensitive to abrasion and renders them impractical for real-life applications. In this paper we present a transparent fluorinated polymer foam that is synthesized by a simple one-step photoinitiated radical polymerization. We term this material “Fluoropor”. It possesses an inherent nano-/microstructure throughout the whole bulk material and is thus insensitive to abrasion as its superhydrophobic properties are not merely due to a thin-layer surface-effect. Due to its foam-like structure with pore sizes below the wavelength of visible light Fluoropor appears optically transparent. We determined contact angles, surface energy, wear resistance and Vickers hardness to highlight Fluoropor’s applicability for real-word applications.

ScIenTIfIc REPORTS | 7: 15078 | DOI: 10.1038/s41598-017-15287- 8 We achieved the formation of superhydrophobic transparent polymer foam "Fluoropor" through light-induced polymerization of fluorinated perfluoropolyether methacrylates in a non-solvent (cyclohexanol) and an emulsifying agent (fluorinated alcohol). Fluoropor possesses a bulk nano-/microstructure (see Fig. 1a and Supplementary Fig. S1) and can be polymerized in varying thicknesses (see Fig. 1b). Due to the polymerization process Fluoropor samples possess a non-structured layer of polymer on top ("lid" see Supplementary Fig. S2) which can be readily removed by abrasion. The optical transparency of Fluoropor is mainly dependent on the degree of light scattering of the surface exposed by the removal of the lid. Thus there is no linear correlation between layer thickness and optical transparency. All thin-layer glass coatings between 55 µm and 170 µm show high optical transparency within the visible light region (see Fig. 1c). Bulk samples of 800 µm and 3500 µm thickness show decreased light transmission but are still see-through upon background contact (see Fig. 1b,c). The transparency of foams is dependent on light scattering at the pore surfaces via reflection and refraction 22 . When the pores are small enough, i.e. when their diameter is well below the wavelength of visible light, the material appears transparent. At 400 nm, non-foamed Fluoropor possesses an optical transmission of 99%, foamed Fluoropor with a smooth surface (lid structure intact) possesses an optical transmission of 86% and the lowest transmission is displayed by foamed Fluoropor with a rough surface (lid structure removed) with 66% at 400 nm (100 µm films, see Supplementary Fig. S3). Therefore, as expected, the pore size in the bulk as well as the surface roughness are responsible for the loss in transparency. Still the pores are small enough to make Fluoropor appear transparent. Optical transmissions were measured against a blank of the supporting glass and therefore display the values for the coating only. The transmission of the supporting glass is displayed Supplementary Fig. S4. Functionalization of glass surfaces relies on exposing the surface silanol groups by treatment with methanol/ hydrochloric acid (1:1 v/v) 23 and consecutive attachment of 3-(dimethylchlorosilyl)propyl methacrylate to the surface. The methacrylate groups are covalently attached to the glass surface and take part in the polymerization reaction of Fluoropor thus covalently attaching the polymer.
Fluoropor possesses a low Vickers hardness of 1.19 HV at 100 mN/20 s -which is in the appropriate range of rigid polymer foams 24 . For further comparison, polyethylene possesses a Vickers hardness of 5-8 HV 24 . Due to its softness Fluoropor is very flexible at layer thicknesses below ~1000 µm (see Fig. 2a). Polymers and fluoropolymers in particular are of high interest for low-friction coatings, i. e. coatings with friction coefficients of 0.4 and lower 25 . Wear resistance measurements of Fluoropor gave a coefficient of friction of µ = 0.2 sliding against a brass sphere (see Fig. 2b) which is comparable to polytetrafluoroethylene (PTFE) sliding against stainless steel (µ = 0.15-0.2) 26 . The wear tracks of the tribometer were below 1 µm in depth thus the material is sufficiently robust for prolonged wear-abrasion (see Supplementary Fig. S5). Fluoropor was further characterized in terms of its superhydrophobic properties. A droplet of water readily bounces off a Fluoropor surface (see Fig. 2c). The superhydrophobicity is independent on the drop volume when gravitational effects on the drop-shape are minor, i.e. between 1 µL and 15 µL and the average static contact angle of Fluoropor was calculated to be 163.7 ± 6.8° (see Fig. 2d). The static contact angle is a common but not an ideal characterization method for rough surfaces. The maximum (advancing) and minimum (receding) contact angles are better suited for producing reliable results on inhomogeneous surfaces. The advancing and receding contact angles were determined to be 158.0 ± 7.9° and 151.9 ± 1.9°, respectively (see Fig. 2e) with an average contact angle hysteresis of 6.1°. The variations in the advancing contact angle are due to vibrations of the enlarging droplet during the measurement (see Supplementary Fig. S6), which lead to variations in the contact angles determined by the instrument automatically by evaluating the high-speed video data image-by-image. The decrease in receding contact angles over time is caused by decreasing contact radius of the droplet due to evaporation over time 27 . The surface possesses a very low free energy of 7.2 mN/m. In accordance with the low hysteresis a water roll-off-angle of 6.6 ± 1.7° was determined.
Abrasion is a threat to any functional structured coating. Fluoropor with its high wear-resistance is not easily abraded but if abrasion occurs the inherent bulk nano-/microstructure makes its superhydrophobic properties insensitive to the abrasion (see Fig. 3). Upon treating Fluoropor with sandpaper and abrading a total of 20 µm of the ~80 µm coating the superhydrophobic properties of the surface are fully retained with a minor change in the static contact angle from 165 ± 2° to 161 ± 4°. Scratching the surface leads to slightly decreased transparency (see Fig. 3b), however the transparency of the material can be recovered by polishing the surface with a microfiber cloth which has no influence on the contact angle which was determined to be 161 ± 2° after polishing (see Fig. 3c). SEM images of Fluoropor after abrasion and after abrasion and consecutive polishing (see Supplementary Fig. S7) show the unaltered nano-/microstructure of Fluoropor.
Besides glass, Fluoropor was also coated on other practically relevant surfaces such as metals, polymers and textiles to show its applicability as a superhydrophobic coating (see Fig. 4). Fluoropor films in the range of ~100 µm were polymerized onto gold, copper and epoxy surfaces using a supporting layer of silane. The metal surfaces were functionalized with a protocol adapted from Wu et al. 28 where a self-assembled monolayer of 3-(mercaptopropyl) triethoxysilane is formed on the metal surface, followed by hydrolysis and condensation of the triethoxysilane groups by corona discharge to form a siloxane network with exposed silanol groups. In a second step, the silanol groups react with 3-(dimethylchlorosilyl)propyl methacrylate to give surface-bound methacrylate groups that covalently attach Fluoropor during polymerization. Fluoropor can also be used as a coating for textiles (see Fig. 4e) coating the fibers with a layer of polymer (see Supplementary Fig. S8). Under water the coated substrates form an air-retaining Salvinia layer (see Fig. 4f). Fluoropor on copper was further employed to  We demonstrated the formation of nano-/microstructured polymer foam "Fluoropor" with adjustable thickness and high optical transparency. The abrasion-insensitivity of the superhydrophobic properties of Fluoropor, as well as its ease of manufacturing and its compatibility with a wide range of substrate materials makes Fluoropor a transparent coating suitable for real-world applications.
Synthesis of nano-/microstructured transparent fluoropolymer. 50 wt% Fluorolink MD700 was mixed with 15 wt% cyclohexanol and 35 wt% 1 H,1 H,2 H,2H-perfluorooctanol. The mixture was blended with 0.5 wt% DMPAP dissolved in acetone (1 mg/µL). Custom-made polymerization chambers were constructed from regular glass slides, glue and spacers of different height (made from cyclic olefin copolymer, COC). The mixture was polymerized under UV light (370 nm, Lumatec Superlite SUV mercury arc lamp, Lumatech, Germany) for 2-5 min depending on the layer thickness. The polymer was placed in 2-propanol overnight and consecutively dried in a vacuum furnace at 100 °C for 1 h at 30 mbar. The top layer of the Fluoropor samples was removed after drying to expose the nanostructure due to the polymerization of a non-structured layer on top of the polymerization chamber (see Supplementary Fig. S2). This layer can be removed by regular laboratory cotton wipes in case of thin-layer coatings or by abrasion with sandpaper for thicker substrates.
Functionalization of glass slides. Cleanroom glass slides were immersed in a mixture of concentrated hydrochloric acid and methanol (50:50) for 30 min. Afterwards the slides were rinsed with 2-propanol and deionized water and dried with compressed air. To functionalize the surface, the slides were incubated in a 100 mM solution of (3-methacryloyloxypropyl)dimethylchlorosilane in toluene for 1 h. The slides were rinsed with 2-propanol and deionized water and dried with compressed air.
Functionalization of copper, gold and epoxy substrates. Gold slides and copper slides were etched to show the KIT logo by etching with iodine solution and ammonium persulfate solution, respectively, using AZ1514H photoresist as a mask. The metal substrates were cleaned with 2-propanol and dried with compressed air. The substrates were placed in 2 vol% solution of 3-(mercaptopropyl)triethoxysilane in ethanol for 8 min. The substrates were rinsed with deionized water and dried with compressed air. After cleaning the metal substrate were activated with a corona discharger (type BD 20, purchased from ETP, USA) for about 1 min. During activation the corona discharger was placed at a distance of approximately 0.5 cm above the sample surface. The substrates were then immersed in a 100 mM solution of 3-(dimethylchlorosilyl)propyl methacrylate in toluene for 1 h. The slides were rinsed with 2-propanol and deionized water and dried with compressed air.
Textile functionalization. The microfiber cloth was used without prior treatment. Contact angle measurement. Contact angles were measured with a OCA15 Pro (Data Physics, USA). All droplet-shapes were evaluated by using the Young-Laplace fit if not stated otherwise. For advancing and receding contact angle measurements by the ARCA program of the instrument, a 5 µL droplet of deionized water was dispensed onto the surface and held in place by the dispenser needle. Additional 5 µL of deionized water were dispensed to a final droplet volume of 10 µL (advancing contact angle). After 1 min 5 µL of the droplet volume were removed (receding contact angle). After 1 min the droplet was expanded to 10 µL and consecutively after 1 min reduced to 5 µL again. In this way five receding and five advancing contact angles were measured by evaluation of the video data, in this case the contact angles were determined by the tangent-fit to reduce computing time. Left-and right-sided contact angles were determined for each frame and the arithmetic middle of the two angles was plotted. The roll-off angle was characterized by dispensing a 10 µL droplet onto the surface and consecutive tilting of the instrument at a given angle per time. Static contact angles and the surface energy of the material were determined using 5 µL droplets. For the determination of the free surface energy, the static contact angles of 5 µL droplets of deionized water and diiodomethane were determined and the OWRK method was used to calculate the surface energy.
UV/VIS measurements. The optical transmission of the nano-/microstructured Fluoropor samples was determined using a UV/VIS spectrometer (Evolution 201, Thermo Scientific, Germany). Coatings on glass were measured against a glass blank, free polymer substrates were measured against air.
Vickers Hardness. The Vickers hardness was determined using a Vickers hardness tester (Fischerscope HCU, Fischer, Germany) at a load of 100 mN and a loading time of 20 s. Wear Resistance. The wear resistance was determined using a custom-built tribometer equipped with a brass test body of 8 mm diameter at 50% relative humidity (RH). The test body moves across the sample thereby causing abrasion. Samples were tested with a load of 400 g, a glide path of 10-12 mm, v = 0.5 mm/s and 1000 experimental cycles.

Sandpaper abrasion test.
Fluoropor coatings on glass were glued to the table using double-sided adhesive tape. 2500 grain sandpaper was glued onto the bottom of a flat 0.74 kg Teflon block. The block was moved across the sample approximately 100 times to give an even abrasion across the whole sample. The sample was cleaned by careful strokes with a cotton wipe, microfiber cloth and/or usage of a pressurized air gun to ensure the full removal of any fluorinated particles.
Layer thickness measurements. The thicknesses of thin-layer coatings on glass were determined using a Heidenhain MT 60 M length gauge. Thicker unsupported samples were measured using a caliper.
Triboelectric measurement. A ~ 100 µm layer of Fluoropor was coated onto a copper plate. A second copper plate was rubbed against the coated plate. Both plates were connected to a multimeter to determine the voltage produced by rubbing.
Data availability statement. The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.