Article pubs.acs.org/Langmuir
Directional Self-Cleaning Superoleophobic Surface Hong Zhao and Kock-Yee Law*,† Xerox Corporation, Xerox Research Center Webster, 800 Phillips Road, 147-59B, Webster, New York 14580, United States ABSTRACT: In this work, we report the creation of a grooved surface comprising 3 μm grooves (height ∼4 μm) separated by 3 μm from each other on a silicon wafer by photolithography. The grooved surface was then modified chemically with a fluorosilane layer (FOTS). The surface property was studied by both static and dynamic contact angle measurements using water, hexadecane, and a polyethylene wax ink as the probing liquids. Results show that the grooved surface is both superhydrophobic and superoleophobic. Its observed contact angles agree well with the calculated Cassie−Baxter angles. More importantly, we are able to make a replica of the composite wax ink−air interface and study it by SEM. Microscopy results not only show that the droplet of the wax ink “sits” on air in the composite interface but also further reveal that the ink drop actually pins underneath the re-entrant structure in the side wall of the grooved structure. Contact angle measurement results indicate that wetting on the grooved surface is anisotropic. Although liquid drops are found to have lower static and advancing contact angles in the parallel direction, the drops are found to be more mobile, showing smaller hysteresis and lower sliding angles (as compared to the FOTS wafer surface and a comparable 3-μm-diameter pillar array FOTS surface). The enhanced mobility is attributable to the lowering of the resistance against an advancing liquid because 50% of the advancing area is made of a solid strip where the liquid likes to wet. This also implies that the contact line for advancing is no longer smooth but rather is ragged, having the solid strip area leading the wetting and the air strip area trailing behind. This interpretation is supported by imaging the geometry of the contact lines using molten ink drops recovered from the sliding angle experiments in both the parallel and orthogonal directions. Because the grooved surface is mechanically stronger against mechanical abrasion, the self-cleaning effect exhibited in the parallel direction suggests that groove texturing is a viable approach to create mechanically robust, self-cleaning, superoleophobic surfaces.
■
INTRODUCTION Digital color printers and offset presses are complex electromechanical devices that put marks of imaging materials, such as toner and ink, on paper. Regardless of the type of print engine and process, printing basically involves moving the images of toner or ink from one surface to another. A fundamental understanding of the forces that govern the attraction or repulsion of toner or ink toward different print surfaces is crucial to the development of high-quality printing engines. One of the key failures in printing is surface contamination due to high toner or ink adhesion on the print surface. Inspired by the self-cleaning effect displayed by lotus leaves in nature1−4 and also the realization that ink and toner are organic materials, one of our research goals is to develop superoleophobic surfaces that repel organic materials enabling self-cleaning in print surfaces. Earlier, we5 reported the successful fabrication of a textured surface comprising 3-μm-diameter pillar arrays (∼7 μm in height with a 6 μm pitch) on a silicon wafer, and the surface was found to be both superhydrophobic and superoleophobic with water and hexadecane. Superhydrophobicity and superoleophobicity are defined as a surface exhibiting water and hexadecane contact angles larger than 150°. The texture was created by photolithography using the Bosch etching process, and the surface was then chemically modified with a fluorinated self-assembled layer (FOTS). A mechanistic study showed that both surface fluorination and the re-entrant © 2012 American Chemical Society
structure in the side wall of the pillar are key enablers for the surface to become superoleophobic in addition to being superhydrophobic. This conclusion was further supported by studying the morphology of the wax ink−air−solid interface of the wax ink drop. Specifically, after the contact angle measurement of the molten wax ink, the sessile drop was cooled to room temperature carefully. The bottom of the sessile drop, which was essentially the replica of the composite interface, was studied by SEM microscopy. The result clearly showed that the ink drop actually penetrates the pillar structure with the contact line pinning underneath the re-entrant structure in the side wall of the pillar. More recent work6 revealed that the superoleophobic surface indeed exhibits super-repellency toward toner and ink, indicating that the surface can be cleaned off easily if it is contaminated by toner and ink. One of the concerns in implementing the pillar array superoleophobic surface is its poor mechanical property where pillars may bend or buckle upon abrasion. In this work, we report an attempt to enhance the mechanical property of the FOTS-textured surface using a (3 μm) grooved structure on a silicon wafer while maintaining the superoleophobicity. Experimentally, directional wetting is Received: May 9, 2012 Revised: July 5, 2012 Published: July 18, 2012 11812
dx.doi.org/10.1021/la301894e | Langmuir 2012, 28, 11812−11818
Langmuir
Article
Figure 1. SEM micrographs of the grooved textured surface. purity, Aldrich) were used as test liquids. The solid wax ink (yellow) was obtained internally, and the material is currently used in the Xerox ColorQube product and is available commercially.9 Contact angle measurements were conducted on an OCA20 goniometer from Dataphysics, which consists of a computer-controlled automatic liquid deposition system and a computer-based imageprocessing system. The measurements with water and hexadecane were carried out at room temperature. In a typical static contact angle measurement, ∼5 μL of the test liquid droplet was gently deposited on the testing surface using a microsyringe and the static contact angle was determined by the computer software (SCA20), and each reported data point is an average of >5 independent measurements. A typical contact angle measurement error is ∼2°. For dynamic measurements, the advancing contact angle was measured by adding a small volume of the test liquid to the surface, and the receding contact angle was measured by slowly removing the test liquid from the drop (0.15 μL/s). The tilting angle measurement was made by tilting the base unit at a rate of 1°/s with a ∼10 μL droplet using a TBU90E tilting base unit. All measurements were averaged from five to eight measurements using a pristine area of the substrate for each measurement. The tilted angle is defined as the angle where the test liquid droplet starts to slide (or move). As for solid wax ink, the contact angle and sliding angle were determined inside a heating chamber at 105 °C. The size of the ink drop (∼1 to 2 μL) was controlled by the careful sieving of the ink pallets with a screen.
observed. The contact angles parallel to the direction of the groove are found to be smaller than those in the orthogonal direction where superhydrophobic-/superoleophobic-like contact angles at >150° are observed with water, hexadecane, and solid wax ink. However, although the repellency in the parallel direction is reduced, both sliding angles and contact angle hysteresis are found to be smaller, indicative of higher drop mobility and lower surface adhesion. Evidence is provided that drops of water, hexadecane, and solid wax ink are all in the Cassie−Baxter state on the grooved FOTS surface. The use of the directional self-cleaning effect in printing application is discussed.7
■
EXPERIMENTAL SECTION
Fabrication of Superoleophobic and Superhydrophobic Grooved Surfaces. The textured surface, consisting of ∼3-μm-wide grooves (∼4 μm in height with a 6 μm center-to-center spacing), was fabricated on 4 in. test-grade silicon wafers (Montco Silicon Technologies, Inc.) by the conventional photolithographic technique followed by surface modification with a conformal nanocoating. The lithographic mask was constructed by Photronics Inc. The surface texture was created by etching the Si wafer using the Bosch deep reactive ion etching (DRIE) process. After the grooved structure was created, the surface was chemically modified with an ∼1.5-nm-thick fluorosilane layer (FOTS), which was obtained by the molecular vapor deposition of tridecafluoro-1,1,2,2-tetrahydrooctyltrichlorosilane onto the bare grooved silicon surface in an MVD100 reactor from Applied Microstructures, Inc.8 The fluorosilane-treated grooved surface was then heat cured in an oven at ∼150 °C for ∼30 min prior to the contact angle measurement. Details of the etching and stripping steps as well as the chemical modification procedure, including the schematic, have been reported earlier.5 Materials and Surface Measurements. DI water (18 MΩ·cm, purified by a reverse osmosis process) and hexadecane (certified 99.4%
■
RESULTS AND DISCUSSION Fabrication and Microscopy of the FOTS-Modified Grooved Surface. The grooved surface comprising ∼3-μmwide grooves (∼4 μm in height) separated by 3 μm from each other was fabricated on a silicon wafer using a conventional photolithographic technique. The grooved surface was then chemically modified by a self-assembled FOTS fluorosilane
11813
dx.doi.org/10.1021/la301894e | Langmuir 2012, 28, 11812−11818
Langmuir
Article
also given in Figure 2, and anisotropic wetting is observed. The static and dynamic contact angle data are summarized in Table 1 along with the two controls, the smooth FOTS silicon wafer surface, and the 3-μm-diameter pillar array FOTS surface on a silicon wafer. Although the smooth FOTS surface is very hydrophobic and slightly oleophilic, with water and hexadecane contact angles of 107 and 73°, respectively, the FOTS pillar surface is both superhydrophobic and superoleophobic, with water and hexadecane contact angles exceeding 150° and sliding angles of ∼10°. With the grooved FOTS surface, we observe anisotropic wetting. The water and hexadecane static contact angles are 131 and 113°, respectively, in the parallel direction and 154 and 162°, respectively, in the orthogonal direction. The Cassie− Baxter state and the fully wetted Wenzel state are commonly used to describe the state of a liquid drop on a rough surface. The static contact angles of the liquid drop in the Cassie− Baxter state (θCB) and the Wenzel state (θW) are given by eqs 1 and 2.10,11
layer synthesized from tridecafluoro-1,1,2,2-tetrahydrooctyltrichlorosilane by the molecular vapor deposition technique. Details of the schematic, the Bosch etching process, and the surface-modification procedure have been given earlier.5 Figure 1 shows the SEM micrographs of the grooved surface. As shown in the highly magnified micrograph in the inset, the side wall of the groove is not straight but rather comprises a ∼300 nm wavy structure from top to bottom as a result of the Bosch etching process. Surface Property. Static Contact Angle Data. The surface property of the grooved FOTS surface is characterized by measuring its static and dynamic contact angles with water, hexadecane, and solid wax ink. Figure 2 summarizes the sessile
cos θCB = R f f cos θY + f − 1
(1)
cos θ W = r cos θY
(2)
where f is the area fraction of the projected wet area, Rf is the roughness ratio on the wet area, Rf f is the solid area fraction, r is the roughness ratio, and θY is the contact angle of the smooth surface. For the grooved surface in Figure 1, f is 50% and θY is 73°. The θCB and θW values can be calculated using eqs 1 and 2 and are given in Table 1. The calculated θCB values for water and hexadecane are 130 and 111°, respectively, and are in agreement with the values observed in the parallel direction. Similar agreement is also obtained with the solid wax ink. As shown later in this article, the contact line of the grooved surface in the parallel direction is ragged because of the difference in wettability between the solid strip and the “air” areas. In other words, the contact line of the liquid droplet is 50% on the solid strip and 50% on air in the parallel direction. The agreement between the calculated θCB values, which are the cosine average of the solid strip and air contact angles, and the experimental data suggests that the liquid droplets studied in this work are all in the Cassie−Baxter state on the grooved FOTS surface. This conclusion is supported by the SEM
Figure 2. (Top) Schematic for the directional contact angle measurements. (Bottom) Sessile drop data of water and hexadecane on the grooved FOTS surface.
drop data of water and hexadecane on the grooved FOTS surface. The orientation of the contact angle measurement is
Table 1. Water, Hexadecane, and Solid Wax Ink Contact Angle Data on the Grooved FOTS Surface surface grooves parallel
grooves orthogonal
controls 3 μm pillar arrays
smooth FOTS
test liquid
θa
θCB/θWb
θA/θRc
θA−θR(deg)d
α (deg)e
water hexadecane solid wax ink water hexadecane solid wax ink
131 113 120 154 162 156
130°/134° 111°/48° 114°/64° 130°/134° 111°/48° 114°/64°
138°/123° 119°/100°
15 19
159°/119° 164°/98°
40 66
8 4 25 23 34 not slide
waterf hexadecanef solid wax ink waterg hexadecaneg solid wax ink
156 158 155 107 73 79
153°/142° 143°/40° 144°/60°
161°/143° 165°/121°
18 44
116°/96° 74°/65°
20 9
10 10 33−58 14 9 not slide
Static contact angle, ±2°. bCalculated from eqs 1 and 2. cθA, advancing contact angle; θR, receding contact angle, ±2°. dContact angle hysteresis. eα, sliding angle, ±3°. fData taken from ref 5. gData taken from ref 12. a
11814
dx.doi.org/10.1021/la301894e | Langmuir 2012, 28, 11812−11818
Langmuir
Article
Figure 3. Contact angle and sliding angle data of wax ink on the grooved FOTS surface and the controlled pillar array FOTS surface.
Figure 4. (a) Optical micrographs of the sessile drops of wax ink on the grooved surface and the pillar array surface. (b) SEM micrograph of the bottom of the wax ink drop showing the geometry of the composite interface between the ink drop and the grooved surface. (c) Schematic of the Cassie−Baxter state.
orthogonal direction. Because both contact angle hysteresis and the sliding angle are measures of the mobility of the liquid drop on an inclined plane,12−14 the overall results in Table 1 suggest that water and hexadecane drops are more mobile in the parallel direction than in the orthogonal direction. It is also important to point out that the mobility in the parallel direction is even higher than that on the pillar array FOTS surface (sliding angles of ∼10° for water and hexadecane) and the smooth FOTS surface (sliding angles of between 9 and 14° for water and hexadecane). The unexpectedly low hysteresis and low sliding angle are not due to the reduction in the solid area fraction in the parallel direction because if that is the case then a similar effect should be observed in the orthogonal direction too. Recent studies by Nosonovsky15 and Sakai and co-workers16 indicated that surface friction is the main obstacle for drop mobility on rough surfaces. We believe that surface friction also plays a key role in this work. For instance, in the parallel direction, half of the area that the liquid drop advances into is air and half is solid strip. Although air repels any liquid because its surface energy is zero, the solid strip will provide an area where the liquid drop can wet and advance. The solid strip, which has a lower advancing contact angle, would wet preferentially, lowering the barrier for the drop to advance. This reduction in surface friction would result in an enhanced mobility of the liquid drop in the parallel direction. In the orthogonal direction, the liquid drop will encounter full solid 50% of the time and air 50% of the time. In other
microscopy study of the composite interface given later in this article. The static contact angles for water and hexadecane in the orthogonal direction are larger than the calculated θCB values. They are in fact comparable to those observed in the pillar array FOTS surface. Again, from the contact line geometry study we show that the contact line in the orthogonal direction appears as “steps” because of the pinning of the liquid drop underneath the re-entrant structure in the side wall of the groove. Because we established previously that the re-entrant structure in the side wall of the pillar is the key contributor to the superoleophobic-like contact angle for the pillar array FOTS surface,5 the high static contact angle observed in the orthogonal direction can also be attributed to the pinning of the liquid droplet underneath the re-entrant structure. Dynamic Contact Angle Data. Similar to the static contact angle data, anisotropic wetting is also observed for the advancing (θA) and receding (θR) contact angles and sliding angle (α). Although the anisotropy in the advancing contact angle is similar to that of the static contact angle, a large anisotropic effect is observed for the contact angle hysteresis (θA − θR). For example, the contact angle hysteresis values with water and hexadecane are 15 and 19°, respectively, in the parallel direction, and they increase to 40 and 66°, respectively, in the orthogonal direction. Moving in the opposite direction is the sliding angle (α). Both water and hexadecane drops are found to be very mobile, with sliding angles of between 4 and 8° in the parallel direction, increasing to 23−34° in the 11815
dx.doi.org/10.1021/la301894e | Langmuir 2012, 28, 11812−11818
Langmuir
Article
Figure 5. (a) Schematic of a sliding ink drop on an inclined plane, (b) cartoon representations of an ink drop sliding in the parallel (top) and orthogonal (bottom) directions, and (c) SEM images of the contact lines of the sliding ink drops in the parallel (top) and orthogonal (bottom) directions.
words, the liquid drop is basically “hopping” from one solid strip to another. As noted earlier, the contact line in the orthogonal direction appears as steps because of the pinning of the liquid drop underneath the re-entrant structure in the side wall of the groove. For the drop to move or slide, depinning of the contact line from the re-entrant structure is needed. We thus attribute the low drop mobility in the orthogonal direction to the high surface friction resulting from the depinning of the “stepped” contact line as the drop is moving through the solid strips orthogonally. This interpretation is supported by the microscopy data of the contact lines from the solid wax ink experiments given later in this article. Contact Angle Measurements of Wax Ink on the Grooved FOTS Surface. Figure 3 summarizes the sessile drop data for the wax ink on the grooved surface in the parallel and orthogonal directions. The measurement was taken in a heated chamber at 105 °C where the wax ink is in the molten state. The sessile drop data of the same wax ink on the 3-μmdiameter pillar array FOTS surface is also provided for comparison. Similar to those observed with water and hexadecane, anisotropic wetting is observed. The contact angle in the orthogonal direction is 156°, comparable to that observed on the pillar array surface (155°), and is much larger than that observed in the parallel direction, which is only 120°. As for the sliding angle, the ink drop is quite mobile and slides at 25° in the parallel direction despite having a smaller contact angle. Similar to the water and hexadecane drops, the ink drop just pins and sticks on the grooved surface at a 90° tilted angle in the orthogonal direction. The geometry of the composite interface for the wax ink drop on the grooved surface was studied by both optical microscopy and SEM. Figure 4a shows the optical micrographs of the sessile ink drop taken from the top view and the bottom view. Comparative micrographs taken for the sessile drop on the pillar array surface are also shown. The contact line, highlighted in red (Figure 4a), clearly shows that it is rectangular on the grooved surface, whereas it is circular on the pillar array surface. This positively indicates that the molten ink wets the grooved surface anisotropically. In this work, we further acquire direct evidence about the geometry of the composite interface. This is conveniently done by cooling (solidifying) the molten ink drop carefully after the contact
angle measurement. The bottom of the wax drop becomes a replica of the composite interface. The SEM micrograph of the composite interface, shown in Figure 4b, clearly shows the ∼3 μm grooved structure. The most important information comes from the height of the replica of the grooves, which is
