Droplet impact on anisotropic superhydrophobic surfaces - Langmuir


Droplet impact on anisotropic superhydrophobic surfaces - Langmuir...

3 downloads 128 Views 2MB Size

Subscriber access provided by UNIV OF DURHAM

Article

Droplet impact on anisotropic superhydrophobic surfaces Chunfang Guo, Danyang Zhao, Yanjun Sun, Minjie Wang, and Yahua Liu Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b03752 • Publication Date (Web): 13 Feb 2018 Downloaded from http://pubs.acs.org on February 14, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Droplet Impact on Anisotropic Superhydrophobic Surfaces† Chunfang Guo,‡ Danyang Zhao,∗,‡ Yanjun Sun,‡ Minjie Wang,‡ and Yahua Liu∗,‡,¶ ‡Key Laboratory for Precision & Non-traditional Machining Technology of Ministry of Education, Dalian University of Technology, Dalian 116024, China ¶Key Laboratory for Micro/Nano Technology and System of Liaoning Province, Dalian University of Technology, Dalian 116024, China E-mail: [email protected]; [email protected]

Abstract A droplet impacting on superhydrophobic surface exhibits complete bouncing. The impacting process usually consists of spreading and retracting stages, during which the droplet contacts the underlying substrate. Recent research has been devoted to reducing the contact time using textured surfaces with different morphologies or flexibility. Here we design submillimeter superhydrophobic ridges and show that impacting droplets bounce off the surface immediately after capillary empty in an petal-like shape at certain Weber number range. The absence of horizontal retraction process in two directions leads to ∼ 70% reduction in contact time. We demonstrate that the petal bouncing is attributed to the synergistic cooperation of the hierarchical structures and anisotropic property which endows an effective energy storage and release. When touching the bottom of the grooves, obvious flying wings appear along the ridges with a velocity component in the vertical direction, which promote the energy releasing process to achieve a fast droplet detachment. At higher Weber number, the anisotropic †

Correspondence should be addressed to D.Z. ([email protected]) or Y.L. ([email protected])

1

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

surface distorts the mass distribution and promotes the uniform fragmentation of the droplet, and therefore the overall contact time is dramatically reduced. Simple analyses are proposed to explain these phenomena, showing a good agreement with the experimental results. The contact time reduction on anisotropic superhydrophobic surfaces is expected to have great importance for the design and fabrication of anti-icing and self-cleaning surfaces.

Introduction Liquid droplet impacting on solid surfaces particularly on superhydrophobic surfaces, is an important natural phenomenon, which has been studied intensively in the past decade. 1–8 During the impact process, the droplet spreads to reach a maximum spreading diameter, with its kinetic energy being partly transformed into interfacial energy. Subsequently, the droplet retracts to reduce its interfacial energy and if dissipation within the droplet and friction with the surface are sufficiently small, it will have enough energy to rebound. The contact time between the impacting droplet and the superhydrophobic surface is reported to be independent of impact velocity. 1,2 Since the contact period affects the momentum, heat and mass transfer, reducing the contact time between bouncing droplets and underlying surfaces is considered to be crucial for a wide range of applications, such as self-cleaning, anti-icing and dropwise condensation. 9–16 Approaches reported to reduce the contact time are to use surfaces with hierarchical superhydrophobic features, symmetry breaking surfaces or flexible surfaces. 17–26 Bird et al 17 demonstrated the possibility to reduce the contact time by using superhydrophobic surfaces with macroscopic ridges which break the droplet into two. The contact time was reduced by ∼ 37%. Gauthier et al. 18 used repellent macrotextures to discrete the contact time by changing the droplet impact point relative to the texture under different impacting velocities. Recently, Liu et al. 15,24 demonstrated pancake bouncing on submillimeter-scale post arrays in which the contact time was reduced by ∼ 80%. Pancake bouncing occurs when the gap 2

ACS Paragon Plus Environment

Page 2 of 20

Page 3 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

between the posts on the superhydrophobic surface is large enough that liquid can penetrate as the droplet hits the substrate. The so-called pancake bouncing of millimetric water droplets on surfaces patterned with hydrophobic posts can be reproduced by centimetric water-filled balloons impacting on a bed of nails. 20 Liu et al. 21 fabricated bioinspired curved surfaces with macrotextures comparable to the droplet size, on which an impinning droplet exhibits symmetry-breaking bouncing dynamics and the contact time was reduced by ∼ 40%. The contact time can also be reduced by increasing the droplet impact velocities on surfaces with hierarchical textures due to the prompt splash and fragmentation of liquid lamellae. 22 On oblique superhydrophobic surfaces, the stretched rebound of droplets reduces the contact time compared with conventional sliding rebound. 23 Weisensee et al. 25 observed that droplets impacting on an elastic superhydrophobic surface can lead to a two-fold reduction in contact time. The collaborative effect of substrate flexibility and surface micro/nanotexture was also demonstrated to enhance both icephobicity and the repellency of viscous droplets. 26 In nature, many plant leaf surfaces have directional groove structures with specific wettability, as exemplified by the rice leaves. 27 The anisotropic wetting property means that liquid droplets roll off more easily along the direction parallel to the grooves. Droplet impact experiments have been conducted on grooved structures surfaces. 28,29 Results show that an impacting droplet spreads further along the grooves than in the perpendicular direction. However, few studies have been conducted to investigate the contact time variation on anisotropic superhydrophobic surfaces. In this work, we consider water droplet impacting on superhydrophobic grooves and find that the contact time can be considerably reduced compared with that on flat surfaces.

Experimental In the experiments, directional ridge arrays were prepared to form groove structures, based on copper substrates with a size of 3.0 × 3.0 cm2 and a thickness of 4.0 mm. Schematics

3

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 20

of the ridge surfaces and geometrical parameters are shown in Figure 1(a) and (b). The geometrical parameters of the fabricated ridge surfaces are illustrated in Table 1, including ridge top width b, ridge bottom width d, ridge height h and center-to-center spacing w. For sample preparation, we began by fabricating straight ridges using wire-cutting machining. After cutting, the samples were thoroughly cleaned in ethanol and deionized water for 10 min and washed in hydrochloric acid (1 M) for 10 s to remove the native oxide layer. The asfabricated surfaces were then immersed in a mixed solution of 2.5 mol l−1 sodium hydroxide and 0.1 mol l−1 ammonium persulphate, to generate microstructures. Multiple etching cycles were conducted to form tapered ridges on the samples, because the etching rate at the ridge top is several folds of that at the bottom. Specifically, the newly formed surfaces were washed with diluted hydrochloric acid (1 M) for 10 s to remove the oxide layer produced during the previous etching cycle. After etching cycles, the surfaces were modified in 1 mM n-hexane solution of trichloro(1H,1H,2H,2H-perfluorooctyl)silane for ∼60 min, followed by heat treatment at 150◦ C in air for 1 h to enhance the superhydrophobicity. b

(b)

(a)

h

w d

(c)

(e)

(d)

(f) 200 μm

5 μm

Figure 1. (a) Schematic drawing of the ridge surfaces. (b) Geometrical parameters. (c) SEM image. (d) Micro-spheres and nano-needles. (e) Water contact angle perpendicular to the ridges. (f) Water contact angle parallel to the ridges. Figure 1(c) and (d) show scanning electron microscope (SEM) images of the fabricated ridge surfaces. All the surfaces have uniformly deposited micro-flowers and nano-needles. The apparent water contact angles of water droplets on the superhydrophobic surfaces were 4

ACS Paragon Plus Environment

Page 5 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Table 1: Geometrical parameters and wettability measurements of the anisotropic superhydrophobic ridge surfaces Ridge surface

Ridge top width b (μm)

Ridge bottom width d (μm)

Ridge height h (μm)

Center-to-center spacing w (μm)

θe ( °)

θa ( ° )

H400W250

100

150

400

250

156 ± 1.0

160 ± 1.1

H400W300

100

150

400

300

157 ± 1.1

160 ± 1.2

H400W350

100

150

400

350

158 ± 1.2

160 ± 1.2

H800W300

100

150

800

300

157 ± 1.1

160 ± 1.1

measured with a standard contact angle goniometer. The intrinsic water contact angle on the superhydrophobic flat surface was measured to be 155◦ ± 1.0◦ . The contact angle parallel the ridges are close to that on flat surfaces. The contact angle perpendicular to the ridges are slightly larger and regarded as the equilibrium contact angle (θe ), shown in Figure 1(e) and (f). Table 1 also illustrates the advancing contact angles (θa ) in the perpendicular direction of the ridge surfaces. The contact angle and contact angle hysteresis on all the surfaces are larger than 150◦ and less than 10◦ respectively, indicating the superhydrophobicity of the surfaces. Note that the static contact angles on the ridge surfaces were measured from deionized water droplets of ∼ 5.0 µL. At least five individual measurements were performed on each sample. Liquid droplet impact experiments were conducted at room temperature with 60% relative humidity. Deionized water droplets of radius r0 =1.0 mm, 1.3 mm and 1.45 mm were released from a fine needle using a syringe pump at a volume rate of 2 µl s−1 . The height between the droplets and the samples was from 5 mm to 400 mm, corresponding to the impact velocity v0 from 0.31 m s−1 to 2.80 m s−1 . The Weber number is defined as We = ρv02 D0 /γ , with ρ and γ being the density and surface tension of water droplets, D0 being the droplet diameter. The impact dynamics of the droplets were captured from the side view and the top view using a high-speed camera at a frame rate of 10,000 fps.

5

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Results and discussion Figure 2(a) shows selected snapshots of a droplet (r0 = 1.3mm) impacting on the ridge surface (H800W300) at We=10.6. The impacting droplet exhibits an approximately circular symmetry during the bouncing, as evidenced by the images from both the side views parallel and perpendicular to the ridges, as well as the top view. The contact time is ∼ 14.2 ms (= p 2.57 ρr03 /γ), which is in good agreement with conventional rebound in previous research. 1,30 However, at higher W e, the droplet bounces off the surface in a petal-like shape when we see it from the top view and the contact time decreases dramatically, as exemplified by an impact at We=28.3, as shown in Figure 2(b). In this case, the liquid penetrates into the p grooves and the droplet detaches from the surface at ∼ 4.3 ms (= 0.78 ρr03 /γ) immediately after the capillary emptying, without obvious horizontal retraction in two directions. The contact time was reduced by ∼ 70%. For droplet impact on the ridge surface (H400W300) at lower W e, the bouncing behavior of the droplet is similar to the ridge surface with h=0.8 mm. However, when We increases to certain value such as We=35.3, the droplet touches the bottom of the ridges, and then detaches in a shorter time ∼ 3.5 ms, shown in Figure 2(c). We quantify the difference in bouncing dynamics between conventional and petal bouncing by the ratio between the diameter of the droplet when it detaches from the surface from the surface djump and its maximum spreading dmax , defined as Q = djump /dmax . It contains Q⊥ and Qk for the anisotropic ridge surface, which refer to the ratios in the perpendicular and parallel directions, respectively. It is also useful to analyze the timescales tcontact , tmax and t↑ , where tcontact is the time period from the drop first touches the surface to that when it bounces off the surface, tmax is the time when the droplet reaches its maximum lateral spreading including tmax,⊥ for the perpendicular direction and tmax,k for the parallel direction and t↑ is the time interval between the moments when the droplet first touches the surface and the substrate is completely emptied, during which the fluid undergoes downward penetrating and upward capillary emptying processes.

6

ACS Paragon Plus Environment

Page 6 of 20

Page 7 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(a) 0 ms

t↑= 1.6 ms

3.3 ms

3.5 ms

tmax,⊥= 4.0 ms

tcontact=14.2 ms

0 ms

t↑= 1.6 ms

3.3 ms

3.5 ms

tmax,‖= 4.0 ms

tcontact=14.2 ms

0 ms

1.6 ms

3.3 ms

3.5 ms

4.0 ms

14.2 ms

0 ms

1.6 ms

3.3 ms

tmax,⊥= 4.0 ms

t↑= tcontact =4.3 ms

5.0 ms

0 ms

1.6 ms

3.3 ms

tmax,‖= 4.0 ms

t↑= tcontact=4.3 ms

5.0 ms

0 ms

1.6 ms

3.3 ms

4.0 ms

4.3 ms

5.0 ms

0 ms

1.6 ms

t↑=3.3 ms

tcontact=3.5 ms

tmax,⊥=4.1 ms

5.5 ms

0 ms

1.6 ms

t↑=3.3 ms

tcontact=3.5 ms

tmax,‖=4.3 ms

5.5 ms

0 ms

1.6 ms

3.3 ms

3.5 ms

4.2 ms

5.5 ms

(b)

(c)

Figure 2. (a) Selected sequential images showing a droplet (r0 = 1.3mm) impacting on the ridge surface (H800W300) at W e = 10.6. After hitting the top of the ridge surface at t=0 ms, the droplet slightly penetrates the ridges and recoils back to the top of the surface at t↑ ∼ 1.6 ms. After its maximum lateral extension in perpendicular and parallel direction at 7

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

tmax,⊥ = tmax,k ∼ 4.0 ms, the droplet retracts on the surface and finally detaches from the surface at tcontact ∼ 14.2 ms. (b) Selected sequential images showing a droplet impacting on the ridge surface (H800W300) at W e = 28.3. The droplet partly penetrates into the ridges and recoils back to the top of the surface and bounces off in a petal-like shape at tcontact ∼ 4.3 ms. (c) Selected sequential images showing a droplet impacting on the ridge surface (H400W300) at W e = 35.3. The droplet partly penetrates into the ridges and touches the bottom of the grooves then recoils back to the top of the ridges at t↑ ∼ 3.3 ms. It detaches from the surface with obvious flying wings along the ridges at tcontact ∼ 3.5 ms. Note that the droplet reaches its maximum lateral extension in perpendicular direction at tmax,⊥ ∼ 4.1 ms and parallel direction at tmax,k ∼ 4.3 ms after bouncing off. Timescale analysis of a droplet (r0 = 1.3mm) impacting on the ridge surface (H400W300) is illustrated in Figure 3. When W e 31.8, Q in both directions are close to 1 and tcontact is approximately equal to t↑ indicating that the petal bouncing is driven by the upward p motion of the penetrated liquid. Over the W e range, tmax ∼ ρr03 /γ is nearly constant and independent of the impact velocity. 30,31 But tmax,k is slightly larger than tmax,⊥ , because the spreading in the parallel direction has not finished when the droplet starts to retract in the perpendicular direction. At lower W e, the droplet detaches with djump,k smaller than djump,⊥ , but dmax,k approximately equals to dmax,⊥ , thus Qk is much smaller. At larger W e, after petal bouncing off the surface with djump,k slightly larger than djump,⊥ , the droplet continues to spread and dmax,k is much larger than dmax,⊥ , leading to a relatively smaller Qk . We further performed droplet impact experiments using droplets with different radius on ridge surfaces with varied center-to-center spacing. The contact time variations with increasing W e of these scenerios are shown in Figure 4. 8

ACS Paragon Plus Environment

Page 8 of 20

Page 9 of 20

1.2

20 t

18

tcontact



tmax,

tmax,||

Q

Q 1.0

16 14

Time (ms)

12

0.8

10

Q

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

8

0.6

6 4

0.4

2 0 6

12

18

24

30

36

42

0.2 48

We

Figure 3. Timescale analysis of a droplet (r0 = 1.3 mm) impacting on the ridge surface (H400W300). The droplet exhibits conventional bouncing with tcontact much larger than t↑ , and Q below 0.6 when W e < 31.8. However, the droplet detaches from the surface in a petal-like shape with tcontact ≈ t↑ and Q close to 1.0 when W e > 31.8. To interpret the mechanism behind the droplet bouncing observed in the experiments, the two timescales tmax and t↑ are analyzed. The droplet penetrating and emptying between the ridges is mainly subject to the capillary force. Here, we neglect the viscous dissipation, because the Reynolds number in the impact process is ∼ 100. The capillary force can be approximated by Fc ∼ −nr0 γ cos θ, where n is the number of wetted ridges, and θ ≈ 155◦ is the apparent contact angle of water on superhydrophobic flat surface. 32–34 During droplet impacting process, the penetrated liquid is confined to a localized region with a lateral extension approximately equivalent to the initial droplet diameter and hence n ∼ r0 /w. 35,36 The acceleration of the penetrated liquid moving between the ridges scales as a↑ ∼ −γ cos θ/ρr0 w, where the droplet mass ∼ ρr03 . Thus, when the droplet does not touch the surface bottom, t↑ is scaled as

t↑ ∼ v0 /a↑ = v0 ρr0 w/(−γ cos θ),

9

ACS Paragon Plus Environment

(1)

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 20

which increases with impact velocity v0 , as shown in Figure 3. However, when the droplet touches the groove bottom, the total reactive force acting on the droplet contains the capillary force Fc and the force from the bottom substrate Fs . According to the momentum conservap tion law, Fs can be calculated by Fs = mv0 /tmax ∼ mv0 / ρr03 /γ. The total force when the p droplet touches the bottom can be written as Fr = Fc + Fs ∼ −r02 γ cos θ/w + mvh / ρr03 /γ, p where vh = v02 − 2a↑ h is the velocity of the impacting droplet when touching the bottom. Then the duration between the droplet touching and resting on the surface bottom can be p p scaled as ttouch = mvh /Fr ∼ v02 − 2a↑ h/(a↑ + (v02 − 2a↑ h)γ/ρr03 ). Therefore, t↑ can be p approximately expressed as t↑ = tdown + ttouch + tup , where tdown ∼ (v0 − v02 − 2a↑ h)/a↑ p is the duration for the droplet impaling the ridges to groove bottom, and tup ∼ 2h/a↑ is the period for the droplet upward emptying the ridges. As the impact velocity v0 becomes p p large enough, t↑ tends to be t↑lim ≈ ρr03 /γ + 2hρr0 w/(−γ cos θ) , which is independent of impact velocity. The experimental t↑ and the theoretical result are in the same order, and the variation trends are consistent. Besides, tcontact approximately equals to t↑ over the petal bouncing range, and is larger for the ridge surfaces with larger w and r0 , shown in Figure 4. The effect of droplet size and surface parameters especially the center-to-center spacing on the contact time variation are explored. Considering the touching groove bottom scenario, the penetration depth is the ridge height h. The interfacial energy per unit area associated with the ridge-water and water-air interfaces are −γ cos θ and γ , respectively. The change of the interfacial energy due to transversal water penetration is

Etrans =

i πr02 γ h p 2 ( 4h + (d − b)2 + w − d)(− cos θ) − (w − b) , w

(2)

where these terms are due to the appearing composite interface between the penetrated water and the groove surface, and the disappearing water-air interface. At the end of lateral spreading, the droplet exhibits a maximum spreading radius, which scales as rmax ∼ r0 We1/4 . 31 The impacting droplet at maximum spreading is approximately a cylinder with a radius rmax .

10

ACS Paragon Plus Environment

Page 11 of 20

24 r0=1.00 mm H400W300

22

Contact time (ms)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

r0=1.30 mm H400W250

20

r0=1.30 mm H400W300

18

r0=1.30 mm H400W350 r0=1.45 mm H400W300

16 14 12 10 8 6 4 2 4

8

12

16

20

24

28

32

36

40

44

48

52

We

Figure 4. Timescale analyses of droplets impacting on the ridge surfaces. The droplets detach from the surfaces in a petal-like shape with tcontact approximately equals to t↑ when W e reaches certain number, and is larger for the ridge surfaces with larger w and r0 . When W e is larger enough, the limit of tcontact as well as t↑ tends to be independent of impact velocity. Thus the change of surface energy during lateral spreading is

Elat = We1/2 πr02 γ(1 − cos θCB ) − 4πr02 γ,

(3)

where θCB ≈ 160◦ is the apparent Cassie-Baxter contact angle of a droplet on the ridge surfaces. Here we consider the effect of contact line pinning during droplet spreading on the groove structures. During the spreading stage, the contact angel hysteresis force per unit length of contact line can be measured as γ(cos θe − cos θa ). The energy lost to overcome the pinning force at n ≈ 2r0 /w number of ridge edges can be estimated as 28

Wpin =

3 2πrmax γ (cos θe − cos θa ). w

(4)

The kinetic energy of the impacting droplet Ek for petal bouncing occurrence should be greater than the sum of the three energies. Thus the critical W e for petal bouncing on the 11

ACS Paragon Plus Environment

Langmuir

ridge surfaces can be expressed as i 6r 3hp 2 0 ( 4h + (d − b)2 + w − d)(− cos θ) − (w − b) + We3/4 c (cos θe − cos θa ) w w (5) 1/2 + 3Wec (1 − cos θCB ) − 12.

W ec =

Figure 5 plots the experimental and theoretical W ec with center-to-center spacing w and droplet radius r0 , respectively. The experimental results are in good agreement with those predicted by the model. Note the results of the model are slightly larger than the experimental ones, because we consider a uniform penetration and the effect of contact line pinning is relatively overestimated.

(a)

60

60

(b)

Model Experimenetal

50

40

40

30

30

20

20

10

10

0 200

250

300

350

Model Experimenetal

50

Wec

Wec

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 20

400

0 0.75

w (m)

1.00

1.25

1.50

1.75

r0 (mm)

Figure 5. Comparison of experimental (black squares) and theoretical (red circles) critical weber number W ec for petal bouncing occurrence.

On the ridge surfaces, obvious splash and fragmentation were observed in the experiments at large droplet impact velocity. For instance, when a droplet (r0 = 1.3 mm) impacts the ridge surface H400W300, satellite droplets were sputtered from the periphery of the spreading droplet in the parallel direction when W e>56.5. Dramatic fragmentation of the droplet was observed in the parallel direction after petal bouncing from the surface when W e>63.6. Detaching from the ridge surface in a petal-like shape, the droplet behaves like an ellipse film with a perpendicular axis a and parallel axis b, and b is larger than a due to the easy spreading along the ridges. The surface energy of the droplet can be approximately

12

ACS Paragon Plus Environment

Page 13 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

expressed as Es ≈ 2πabγ, and the horizontal retraction force is Fa ≈ 2πbγ and Fb ≈ 2πaγ in the respective directions. 21 Thus the droplet retracts faster in the perpendicular direction than the parallel direction, which leads to droplet breakup. When W e>106.0, petal bouncing disappears because the droplet contacts the surface during the latter retracting stage. In this case, the integral contact time of the impacting droplet is still shortened compared with conventional bouncing due to the subsequent fragmentation, shown in Figure 6. (a)

0 ms

3.3 ms

6.0 ms

tcontact=13.8 ms

0 ms

3.3 ms

6.0 ms

tcontact=10.2 ms

(b)

Figure 6.

Selected sequential images showing a droplet (r0 = 1.3 mm) impacting on

superhydrophobic flat surface and the ridge surface (H400W300) at W e = 106.0. (a) The droplet spreads and retracts and wholly detaches from the flat surface at ∼ 13.8 ms. (b) The droplet splashes and separates into several main fragments along the parallel direction and detaches from the ridge surface at ∼ 10.2 ms.

The ridges hinder the spreading of penetrated droplet, but accelerate the recoil of droplet during the retracting stage because of the inward Laplace force. When the penetrated portion begins to retract, the droplet periphery parallel to the ridges are still expanding outward. This spreading and retracting difference in two directions creates more obvious flying wings. The change of mass distribution promotes the splash and fragmentation of the droplet in the parallel direction. Since the contact time of each fragmented droplets also p scales as tcontact ∼ ρr03 /γ, the small droplets detach from the surface more quickly, leading to the reduction of the overall contact time. The contact time of the impacting droplet p √ can be approximated as tcontact ∼ ρr03 /γ/ N ∗ , where N ∗ is a nominal factor indicating the number of the fragmented droplets. 22 Due to the hierarchical water-repellent structures 13

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 20

of the surfaces, no obvious liquid pinning occurs between the ridges at large W e. We can image that the fragmentation effect will be more dominant if the impact velocity continues to increase, and the contact time will be further decreased. For instance, the impacting droplet splashes and breaks both on the flat and ridge surface at W e = 282.6. The fragmentation on flat surface occurs on the outside rim and the main part of the droplet also experiences retraction process, as shown in Figure 7(a). However, the ridge surface breaks the droplet into more uniform smaller ones and the contact time is shorter, as shown in Figure 7(b). (a)

0 ms

4.0 ms

6.6 ms

tcontact=13.6 ms

0 ms

4.0 ms

tcontact=6.6 ms

11.0 ms

(b)

Figure 7.

Selected sequential images showing a droplet (r0 = 1.3 mm) impacting on

superhydrophobic flat surface and the ridge surface (H400W300) at W e = 282.6. (a) The droplet splashes on the rims and the main part of the droplet retracts and detaches from the flat surface at ∼ 13.6 ms. (b) The droplet splashes and separates into several small parts and detaches from the ridge surface at ∼ 6.6 ms.

The relationship between the contact time of impacting droplets (r0 = 1.3 mm) and Weber number on superhydrophobic flat surface and the ridge surface (H400W300) in the experiments is revealed in Figure 8. The contact time on the flat surface is about 14.2 ms, which is independent of W e. On the ridge surface, the contact time of impacting droplet is first reduced through petal bouncing if W e reaches a certain number. When W e increases and exceeds the petal bouncing region, the contact time is still reduced under the effect of splash and fragmentation.

14

ACS Paragon Plus Environment

Page 15 of 20

20

Ridge surface (H400W300)

Flat surface

18

Petal bouncing

16

Splash and fragmentation

14

Contact time (ms)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

12 10 8 6 4 2 0 0

30

60

90

120

150

180

210

240

270

300

We

Figure 8. Contact time variations of droplets (r0 = 1.3 mm) impacting on superhydrophobic flat surface and the ridge surface (H400W300). The contact time on the flat surface is nearly constant, while the contact time on the ridge surface is shortened through petal bouncing at certain W e range and then reduced under the effect of splash and fragmentation of the droplets.

Summary We demonstrate that the contact time of impacting droplets can be significantly reduced using anisotropic superhydrophobic surfaces over a wide Weber number range. With the increased impact velocity, petal bouncing occurs to shorten the contact time remarkably, where the impacting droplet integrally leaves the surface without a horizontal retraction stage. This is attributed to the upward motion driven by the capillary energy stored in the penetrated liquid between the ridges adequate to lift the droplet during capillary emptying process. On the surfaces with relatively lower ridge height, the droplet touches the bottom of the grooves. Obvious flying wings are created along the ridges with a velocity component in the vertical direction, which enhance the detachment of the droplet and further reduce the contact time. The anisotropic surfaces facilitate the spreading in the parallel direction and 15

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

accelerate the retracting in the perpendicular direction. This difference changes the mass distribution and promotes the fragmentation of the impacting droplet when Weber number is large. The impacting droplet breaks into more uniform smaller ones compared with that on flat surfaces, which reduces the overall contact time. Based on these mechanisms, Weber number range for contact time reduction has been greatly broadened. This work may provide insights to promote rapid droplet detachment via fabricating anisotropic surface for practical applications including self-cleaning and anti-icing, etc.

Acknowledgement This work was supported by the Science Fund for Creative Research Groups of NSFC (no. 51621064), the National Natural Science Foundation of China (no. 51605073), the Doctoral Scientific Research Foundation of Science and Technology Commission of Liaoning Province, China (no. 201601047) and the Fundamental Research Funds for the Central Universities (DUT16TD20). We thank Julia M. Yeomans for helpful discussions.

References (1) Richard, D.; Clanet, D., C.and Quéré Surface phenomena: Contact time of a bouncing drop. Nature 2002, 417, 811. (2) Richard, D.; Quéré, D. Bouncing water drops. Europhys. Lett. 2000, 50, 769. (3) Quéré, D. Wetting and roughness. Annu. Rev. Mater. Res. 2008, 38, 71–99. (4) Yarin, A. L. Drop Impact Dynamics: Splashing, Spreading, Receding, Bouncingąŋ. Annu. Rev. Fluid Mech. 2005, 38, 159–192. (5) Josserand, C.; Thoroddsen, S. T. Drop Impact on a Solid Surface. Annu. Rev. Fluid Mech. 2017, 48, 365–391. 16

ACS Paragon Plus Environment

Page 16 of 20

Page 17 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(6) Ruiter, J. D.; Lagraauw, R.; Ende, D. V. D.; Mugele, F. Wettability-independent bouncing on flat surfaces mediated by thin air films. Nat. Phys. 2015, 11, 48–53. (7) Hao, C. L.; Li, J.; Liu, Y.; Zhou, X. F.; Liu, Y. H.; Liu, R.; Che, L. F.; Zhou, W. Z.; Sun, D.; Li, L.; Lei, X.; Wang, Z. K. Superhydrophobic-like tunable droplet bouncing on slippery liquid interfaces. Nat. Commun. 2015, 6, 8986. (8) Li, J.; Hou, Y. M.; Liu, Y. H.; Hao, C. L.; Li, M. F.; Chaudhury, M. K.; Yao, S. H.; Wang, Z. K. Directional transport of high-temperature Janus droplets mediated by structural topography. Nat. Phys. 2016, 12, 606–612. (9) Deng, X.; Mammen, L.; Butt, H. J.; Vollmer, D. Candle soot as a template for a transparent robust superamphiphobic coating. Science 2012, 335, 67–70. (10) Boreyko, J. B.; Chen, C. H. Self-propelled dropwise condensate on superhydrophobic surfaces. Phys. Rev. Lett. 2009, 103, 184501. (11) Mishchenko, B., L.and Hatton; Bahadur, J., V.and Taylor; Krupenkin, T.; Aizenberg, J. Design of ice-free nanostructured surfaces based on repulsion of impacting water droplets. ACS nano 2010, 4, 7699–7707. (12) Khojasteh, D.; Kazerooni, M.; Salarian, S.; Kamali, R. Droplet impact on superhydrophobic surfaces: A review of recent developments. J. Ind. Eng. Chem. 2016, 42, 1–14. (13) Hao, C. L.; Liu, Y. H.; Chen, X. M.; Li, J.; Zhang, M.; Zhao, Y. H.; Wang, Z. K. Bioinspired interfacial materials with enhanced drop mobility: From fundamentals to multifunctional applications. Small 2016, 12, 1825–1839. (14) Vasileiou, T.; Gerber, J.; Prautzsch, J.; Schutzius, T. M.; Poulikakos, D. Superhydrophobicity enhancement through substrate flexibility. Proc. Natl. Acad. Sci. USA. 2016, 113, 13307. 17

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(15) Liu, Y. H.; Moevius, L.; Xu, X. P.; Qian, T. Z.; Yeomans, J. M.; Wang, Z. K. Pancake bouncing on superhydrophobic surfaces. Nat. Phys. 2014, 10, 515–519. (16) Blossey, R. Self-cleaning surfacesŮŮvirtual realities. Nat. Mater. 2003, 2, 301–306. (17) Bird, J. C.; Dhiman, R.; Kwon, H. M.; Varanasi, K. K. Reducing the contact time of a bouncing drop. Nature 2013, 503, 385–388. (18) Gauthier, A.; Symon, S.; Clanet, C.; Quéré, D. Water impacting on superhydrophobic macrotextures. Nat. Commun. 2015, 6, 8001. (19) Liu, Y. H.; Wang, Z. K. Superhydrophobic porous networks for enhanced droplet shedding. Sci. Rep. 2016, 6, 33817. (20) Bro, J. A.; Jensen, K. S. B.; Larsen, A. N.; Yeomans, J. M.; Hecksher, T. The macroscopic pancake bounce. Eur. J. Phys. 2016, 38, 015006. (21) Liu, Y. H.; Andrew, M.; Li, J.; Yeomans, J. M.; Wang, Z. K. Symmetry breaking in drop bouncing on curved surfaces. Nat. Commun. 2015, 6, 10034. (22) Lv, C. J.; Hao, P. F.; Zhang, X. W.; He, F. Drop impact upon superhydrophobic surfaces with regular and hierarchical roughness. Appl. Phys. Lett. 2016, 108, 141602. (23) Zhang, R.; Hao, P. F.; He, F. Drop Impact on Oblique Superhydrophobic Surfaces with Two-tier Roughness. Langmuir 2017, 33, 3556. (24) Liu, Y.; Whyman, G.; Bormashenko, E.; Hao, C.; Wang, Z. Controlling drop bouncing using surfaces with gradient features. Appl. Phys. Lett. 2015, 107, 051604. (25) Weisensee, P. B.; Tian, J. J.; Miljkovic, N.; King, W. P. Water droplet impact on elastic superhydrophobic surfaces. Sci. Rep. 2016, 6, 30328. (26) Vasileiou, T.; Schutzius, T. M.; Poulikakos, D. Imparting Icephobicity with Substrate Flexibility. Langmuir 2017, 33, 6708. 18

ACS Paragon Plus Environment

Page 18 of 20

Page 19 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(27) Feng, L.; Li, S. H.; Li, Y. S.; Li, H. J.; Zhang, L. J.; Zhai, J.; Song, Y. L.; Liu, B. Q.; Jiang, L.; Zhu, D. B. Super-hydrophobic surfaces: from natural to artificial. Adv. Mater. 2002, 14, 1857–1860. (28) Vaikuntanathan, V.; Sivakumar, D. Maximum spreading of liquid drops impacting on groove-textured surfaces: Effect of surface texture. Langmuir 2016, 32, 2399–2409. (29) Pearson, J. T.; Maynes, D.; Webb, B. W. Droplet impact dynamics for two liquids impinging on anisotropic superhydrophobic surfaces. Exp.Fluids 2012, 53, 603–618. (30) Okumura, K.; Chevy, F.; Richard, D.; Quéré, C., D.and Clanet Water spring: A model for bouncing drops. Europhys. Lett. 2003, 62, 237. (31) Clanet, C.; Béguin, C.; Richard, D.; Quéré, D. Maximal deformation of an impacting drop. J. Fluid. Mech. 2004, 517, 199–208. (32) Reyssat, M.; Pépin, A.; Marty, F.; Chen, Y.; Quéré, D. Bouncing transitions on microtextured materials. Europhys. Lett. 2006, 74, 306. (33) Bartolo, D.; Bouamrirene, F.; Verneuil, É.; Buguin, A.; Silberzan, P.; Moulinet, S. Bouncing or sticky droplets: Impalement transitions on superhydrophobic micropatterned surfaces. Europhys. Lett. 2006, 74, 299. (34) Moulinet, S.; Bartolo, D. Life and death of a fakir droplet: Impalement transitions on superhydrophobic surfaces. Eur. Phys. J. E. 2007, 24, 251–260. (35) Xu, L.; Zhang, W. W.; Nagel, S. R. Drop splashing on a dry smooth surface. Phys. Rev. Lett. 2005, 94, 184505. (36) Ryu, S.; Sen, P.; Nam, Y.; Lee, C. Water Penetration through a Superhydrophobic Mesh During a Drop Impact. Phys. Rev. Lett. 2017, 118, 014501.

19

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 20

Graphical TOC Entry Petal bouncing

Fragmentation

200 μm

20

ACS Paragon Plus Environment