1. Introduction
Engineering surfaces that control the movement of liquid droplet is of importance to a wide range of applications including self-cleaning [
1,
2], anti-icing [
3,
4], anti-fogging [
5,
6], electricity generation [
7,
8], heat transfer [
9,
10], and droplet-based microfluidics [
11,
12]. A drop striking a non-wetting surface will spread out to maximum diameter and then recoil to rebound in the vertical direction [
13]. The dynamics of droplet impact is altered when such surface has a morphology. This controlled droplet impact is critical to many technologies.
Droplet impact on hydrophobic surfaces decorated with submillimeter superhydrophobic textures presents pancake bouncing or petal bouncing, which allowed rapid drop detachment [
14,
15]. Superhydrophobic ridges could break drop impact symmetry and reduce the contact time by modified droplet deformation near the miniature features [
16,
17]. The springboard effect was generated for a droplet impacting an elastic superhydrophobic surface, which led to a twofold reduction in contact time [
18]. Droplet impacting on a curved surface displayed asymmetric bouncing with a ~40% contact time reduction when the curvature was on the order of the droplet diameter [
19]. Also, drops impacting on a convex surface with a dome shape could rapidly bounce off by evolving into an annulus shape where the inner and external rims had a high retracting velocity [
20]. Oblique drop impact on superhydrophobic surface with two-tier roughness could bounce off the surface rapidly in an elongated shape, allowing a 10%∼30% reduction of contact time [
21]. An ellipsoidal drop impacted on a superhydrophobic surface is demonstrated to reduce the contact time and suppress the bounce magnitude by breaking the symmetry [
22]. Moreover, translational motion of an impacting droplet was converted to gyration, with a maximum rotational speed exceeding 7300 revolutions per minute, through heterogeneous surface wettability regulation [
23]. Numerous researches of droplet impact are focused on minimizing contact time [
24]. In contrast, drop dynamics after the restitution process, which is often unpredictable, attracts little attention. The result is that droplets can move in uncontrolled and potentially undesirable ways after impacting the surface, which is a major problem in many technologies [
25]. One approach to droplet rebounce control is engineering nonuniform textures (i.e., roughness gradients) on the surface. Droplets impacting on a rough surface with a wettability gradient was shown to rebound obliquely or migrate following the wettability gradient due to the unbalanced interfacial forces created by wetting difference across the droplets [
26]. However, the result is that droplets did not always move following the direction of decreasing static CA due to the energy barrier caused by the Cassie-to-Wenzel transition [
27]. The wetting symmetry of a droplet can be broken at high temperature by creating two concurrent thermal states (Leidenfrost and contact-boiling) on a surface with structural roughness gradient, which also induce a preferential motion of a droplet towards the region with a higher heat transfer co-efficiency [
28,
29].
In contrast to complex surface roughness design, this work shows that the dynamic behavior of impacting droplets was precisely controlled by well-designed surface chemistry. The non-wetting surfaces with patterned “defect” of extreme wettability was synthesized in a simple and straightforward way. The “defect” of extreme wettability on non-wetting surfaces introduced a net surface tension force to impinging droplets, allowing patterned droplet deposition, controlled droplet splitting, and directed droplet rebound. The work also shows how defect’s size, impact position, and impact velocity alter the dynamic behavior of impacting droplets, achieving controlled vertical rebound of impinging droplets on inclined surfaces. This research shows attaining sophisticated droplet motions by pinning forces originated from chemically engineered surfaces, which opens an avenue in future explorations, such as matter transportation, energy transformation, and object actuation.
3. Results and Discussion
Well-designed surface chemistry is used to produce an accurate control of force transmission to impinging droplets, permitting precise controlled droplet rebounce. The non-wetting surfaces are patterned with “defect” of extreme wettability (hydrophilic) by photolithography. The non-wetting regime (superhydrophobic) are synthesized by a simple and straightforward way using only one inexpensive fluorine-free reagent (methyltrichlorosilane) without complex or costly fabrication procedures. We emphasize that this procedure was carried out in a lab with access to an oxygen plasma cleaner, spin coater, and UV lamp. Photolithography was carried out on the silicon wafer using photoresist spin coating, illumination, and development procedures [
31,
32]. After area-selective development yielding resist-coated patterns and Si/SiO
2 background (θ
A/θ
R = ~56°/~24°), the wafer section was exposed to a 20 min O
2 plasma treatment (Harrick PDC-001). This treatment cleans the background and oxidizes/ablates the outer few nanometers of the ~1.3 μm thick resist. This surface was then treated with CH
3SiCl
3 for introducing silicone nanofilaments that impart superhydrophobicity. After resist removal, the wafer section contained a superhydrophobic background with hydrophilic patterns. As shown in
Figure 1b, the non-wetting regime are covered with a layer of condensed micro-size nanofibers/particles, which mimics the water-repellent mechanism of lotus leaves via micro-to-nanoscale hierarchical morphology. The hydrophilic patterns were determined by what the applied photolithography masks were and could be made in any shape.
Figure 1c demonstrated the resolution of this photolithography/chemistry procedure with SEM micrographs that showed alternating superhydrophobic and hydrophilic rings and a square grid chessboard from 50 to 1000 μm in width.
As shown in
Figure 2a, the non-wetting surfaces containing one hydrophilic cycle are designed to produce an accurate control of force transmission to impinging droplets.
Figure 2b and
Figure S1 illustrates one piece of such patterned surface with images of water droplets confined by three hydrophilic circles (radius = 1.2 mm, one circle of this pattern was image by SEM in
Figure 2c) on the superhydrophobic surfaces. The background surface exhibits contact angles of θ
A/θ
R = 172°/164°, and the hydrophilic circles show contact angles of θ
A/θ
R = ~56°/~24°. The disparity of advancing contact angles between the two areas defined the shape of water droplets acting as three eggs of a table-like construction which supports a piece of PET film (
Figure S1).
For Weber numbers
, where
, R
0, v and γ are the droplet density, initial radius, impact speed, and surface tension, respectively, the maximum extension of droplet is theoretically calculated by
[
30]. Based on this, five water droplet impact positions were designed as shown in
Figure S2. Position P1 is the droplet hitting on the center of the hydrophilic circle (the distance between impact point and center of hydrophilic circle L
dis = 0). P2 is the droplet hitting on the edge of the hydrophilic circle (L
dis = R
defect). Position P3 is the droplet impacting and spreading to just cover the hydrophilic circle (L
dis = R
max − R
defect). On position P4, the droplet impacts and spreads to just touch the edge of the hydrophilic circle (L
dis = R
max + R
defect). For position P5, droplet impacting and spreading does not touch the hydrophilic circle (L
dis > R
max + R
defect). The impact distance from the tip of the needle to the surface of the sample was H = 3–15 cm, with impact velocity v = 0.77–1.71 m/s, We = 9.8–49.0. The outcome of droplet impacting is affected by receding contact angle at the contact line [
33]. For droplet impacting onto superhydrophobic surface, the contact line advances and then recoils (rebounds) (
Figure 2d), while the contact line on hydrophilic surface is pinned without recoil (deposition) (
Figure 2e). The droplet impacting can be tailored with controlled ending by chemically engineering a surface with wetting patterns.
Figure 2f–i illustrates this with two examples. In
Figure 2d, a water droplet impacting onto a non-patterned superhydrophobic surface (treated with CH
3SiCl
3) spreads to a uniform film, retracts, and then lifts off (see
Video S1). For a water droplet impacting onto a superhydrophobic surface patterned with two hydrophilic triangles (2 × 2 × 2 mm) connecting by a line (about 6 × 0.5 mm) (
![Biomimetics 10 00319 i001]()
, one end of this pattern was imaged by SEM in
Figure 2f) in
Figure 2h, a centra assisted spread is promoted. The droplet impacted spreads fast along the pattern with the outward flow generating cups at the ends. The contact line on the hydrophilic pattern gets pinned during recoil; on the superhydrophobic region, the contact line continues to recede until the outward flow impacts the inward flow, leaving a hydrophilic region defined water pattern on the surface with the rest of droplet take-off (see
Video S2). Another example in
Figure 2i shows a superhydrophobic surfaces containing three hydrophilic circles (radius = 2 mm) (⸫, one circle of this pattern was imaged by SEM in
Figure 2g). The droplet impact on the centra of three hydrophilic circles. The droplet spreads and recedes, but the contact line on the hydrophilic circles gets pinned during recoil, which tailors the water droplet into three drops after an impact (see
Video S3).
The emphasis of this article is controlling the droplet rebounce dynamics using the superhydrophobic surfaces including a hydrophilic defect. In the hydrophilic defect containing surfaces, the disparity of apparent contact angle from the hydrophilic to superhydrophobic region develops a net surface tension force during the receding of droplet impact in one direction potentially allowing directed droplet rebounce. The first experiment shows how the impact position relative to a hydrophilic defect alters droplet rebounce trajectory (
Figure 3a).
Figure 3b illustrates droplet rebounce trajectory with a plot of lateral distance (X) vs. vertical distance (Y) for droplets impacting on five different positions.
Figure 3c–g shows the image sequence as seen from the side.
For a water droplet impacting with defect size R
defect = 0.75 mm, impact distance H = 10 cm (We = 32.67), we see from
Figure 3 that the outcomes of droplet impacting on different positions are quite different. Droplet impacting and spreading that does not touch the hydrophilic defect (P5, L
dis > R
max + R
defect) behaves as if it is in contact with an isotropic superhydrophobic surface, on which the water droplet after an impact spread to a nearly uniform film, retracts symmetrically and lifts off that is orthogonal to the surface. In another case that the water droplet exactly right hits the hydrophilic defect (P1, L
dis = 0), the contact line is free to move on the superhydrophobic region until it recedes reaching the hydrophilic circle, at which point the contact line is pinned. This draw the droplet into two fragments in the end: small lobe leaves on the hydrophilic defect and the rest of the drop flies up vertically to a height that is much lower than that in the usual rebounce. For droplet hits which spreads to touch the hydrophilic defect on one side (e.g., P2, P3, P4), a net surface tension force develops during receding, allowing directed droplet rebounce. As shown in
Figure 3c–g, the symmetry of the droplet is broken during receding. The contact line on the superhydrophobic region is free to move and is pinned on the hydrophilic defect that is located on one side. A net surface tension starts to occur as the contact line recedes to contact with the hydrophilic defect, which recomposes the droplet from the symmetric disk to a shape (e.g., gun) that is ready to launch in a direction that is not orthogonal to the surface. It is important to note that the distance between impact point and the hydrophilic defect is response for different droplet rebounce trajectory. As we see from
Figure 3b, droplet impacting and spreading to just cover the hydrophilic circle (P3, L
dis = R
max − R
defect) allows the longest lateral leap.
A central mechanism for impact position effect is the position of the hydrophilic defect, which starts to meet with the receding contact line.
Figure 4 portrays the typical recoil process of droplet impacting on P2 (L
dis = R
defect) and P3 (L
dis = R
max − R
defect). In the case of P2, the droplet after reaching its maximum spread continues to recede on the superhydrophobic region with high receding contact angle (θ
r = θ
r,pho) until the receding contact line reaches the hydrophilic defect, at which point the liquid gets pinned on the hydrophilic area due to the disparity of receding contact angle from the superhydrophobic to the hydrophilic region. A net surface tension force develops vectoring the droplet up to the instant of detachment. For the case of P3, a net surface tension force develops at the instant of the droplet’s maximum lateral spread and continues to the instant of detachment. As a result, the droplet is directed to a longer lateral leap than that of in the former case.
Video S4 demonstrates the process of droplet impacting on P3.
The second experiment shows the effect of hydrophilic defect size on droplet rebouncing dynamics
Figure 5a.
Figure 5b illustrates droplet rebounce trajectory with a plot of lateral distance (X) vs. vertical distance (Y) for droplets impacting onto different sizes of the hydrophilic circle (R
defect = 0.05, 0.75, 1.00, 2.00 mm) from an impact position where the droplet after am impact spreads to just cover the hydrophilic circle (P3, L
dis = R
max − R
defect).
Figure 5c–f shows the image sequence as seen from the side. The receding contact line meets with the defect as the receding begins, which continuously peels off (de-wetting) the hydrophilic circle up to the instant of detachment (rebound), at which point the droplet is re-shaped to—launch parabolically. The de-wetting contact line on the hydrophilic circle is increased with the increase of the hydrophilic circle’s size. This leads to a general trend that the lateral momentum of droplet rebounce is increased with the increase of hydrophilic defect’s size and that the longitudinal momentum of droplet rebounce is decreased with the increase of the hydrophilic defect’s size.
The third experiment shows the influence of impact height on the directed rebounce (
Figure 6a).
Figure 6b illustrates droplet rebounce trajectory with a plot of lateral distance (X) vs. vertical distance (Y) for droplets impacting from different heights (H = 3, 10, 15 cm,
We = 9.80, 32.67, 49.00, respectively). The droplet hits on position P3 and spreads to cover a hydrophilic defect (R
defect = 0.75 mm).
Figure 5c–e shows the image sequence as seen from the side. It is noted from
Figure 5 that the decrease of impact height decreases the longitudinal momentum of droplet rebounce because of the decrease of initial impact kinetic energy. However, it is important to note that the lateral momentum of droplet rebounce is increased with the decrease of impact height. The reason can be identified from
Figure 6c–e. As the droplet impacts the surface, its kinetic energy is directed to the lateral direction, flattening the droplet with the kinetic energy converting into surface energy. Once all of the kinetic energy has been converted to surface energy of the flattened droplet, the reverse surface-to-kinetic energy conversion initiates, resulting in droplet retraction. The decrease of impact height decreases the initial kinetic energy and thus leads to the reduction of retraction speed (which is theoretically expressed as
, where
is the liquid-air surface tension,
ρ is the liquid density, and
h is the thickness of the flattened droplet). This finally leads to the droplet flying up before totally peeled off from the hydrophilic defect. Therefore, a lateral force continues to work until the main droplet is divorced from the pining contact line by capillary break-up.
In addition to droplet motion on horizontal surfaces, the pinning forces were further exploited to achieve controlled vertical rebound of impinging droplets on inclined surfaces by controlling the defect’s size, impact position, and impact velocity. As indicated in
Figure 7a, compared to droplet bouncing on a horizontal surface, the behavior on a tilted surface is significantly more complex [
34]. When a droplet impacts a tilted superhydrophobic surface, its motion follows a dynamic and asymmetric process. The initial impact causes the droplet to spread rapidly, with greater expansion in the downhill direction due to the surface tilt. The droplet then quickly retracts because of the surface’s low adhesion, and under the influence of gravity’s downhill component, it typically rebounds downward or transitions directly into a rolling motion. If the impact energy is high (indicated by a large Weber number), the droplet may undergo several diminishing oblique bounces before rolling. If the kinetic energy is low or the liquid viscosity is high (high Ohnesorge number), the droplet accelerates rapidly into a rolling motion along the slope. Throughout this process, the droplet’s trajectory remains biased toward the downhill direction, and its speed changes in three stages, including deceleration upon impact, acceleration during retraction/rebound, and stabilization during rolling, before finally detaching from the surface in an almost straight path. Since the pinning forces can provide moments to the liquid droplet, water droplets impacting on a tilted surface should also be controlled. To test this idea, a water droplet was released from H = 10 cm and hit a superhydrophobic surface tilted at 6° containing a hydrophilic circle (R
defect = 0.75 mm), as indicated in
Figure 7b.
Figure 7c illustrates the rebound trajectories of droplets impacting different positions on the inclined surface, plotted as the relationship between horizontal distance (X) and vertical distance (Y).
Figure 7d–f present side-view image sequences of the impact process. When impacting at position P0 (
Figure 7d), the droplet did not spread sufficiently to reach the hydrophilic defect site, resulting in a normal downward rebound along the slope direction. In contrast,
Figure 7e demonstrates the rebound process for droplet impacting at position P1, where the droplet spread far enough to reach the hydrophilic spot, and an upward-directed pinning force along the inclined surface propelled the droplet to ascend along the slope against gravitational forces. For the case in
Figure 7f, where the droplet hit at position P2, the equilibrium between an inclination-parallel upward pinning force and gravitational components enabled vertical rebound of impacting droplets, exhibiting impact dynamics analogous to droplets colliding on horizontal surfaces. The complete impacting process at P2 is recorded in
Video S5.