Research Article www.acsami.org
Octagonal Wetting Interface Evolution of Evaporating Saline Droplets on a Micropyramid Patterned Surface Xin Zhong,† Junheng Ren,† Mingfeng Lin,† Karen Siew Ling Chong,‡ Kian-Soo Ong,‡ and Fei Duan*,† †
School of Mechanical and Aerospace Engineering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore ‡ Institute of Materials Research and Engineering, A*Star, 2 Fusionopolis Way, Innovis, Level 9, Singapore 138634, Singapore S Supporting Information *
ABSTRACT: Textured surfaces have been extensively employed to investigate the dynamics, wetting phenomena, and shape of liquid droplets. Droplet shape can be controlled via the manipulation of topographic or chemical heterogeneity of a solid surface by anchoring the three-phase line at specific sites. In this study, we demonstrate that droplet shape on a topographically patterned surface can be modified by varying the concentration of salt potassium chloride (KCl) in the droplet solution. It is found that at the beginning of evaporation the octagonal shape of the solid−liquid interface is changed to a rectangle with corners cut upon increasing the salt concentration. Such a variation in the solid−liquid interface versus the salt concentration is explained by the analysis of free energy difference. It indicates that the increases in solid−liquid and liquid−vapor surface tensions by raising the salt concentration result in a favored extension of the three-phase line intersecting the micropyramid bottom sides than the counterpart intersecting the micropyramid diagonal edges. The saline droplets experience a pinning stage at first and a depinning one afterward. The onset of depinning is delayed, and at which the instantaneous contact angle is larger upon raising the salt concentration. The three-phase line which intersects the micropyramid diagonal edges recedes ahead of the one along the micropyramid bottom sides, making the octagonal wetting interface evolve toward a circle. A close view at the droplet edge indicates that the three-phase line repeats “slow slip−rapid slip” across row by row of micropyramids during the depinning stage. KEYWORDS: droplet shape control, octagonal wetting interface, saline droplet, micropyramid surface, evaporation
1. INTRODUCTION Superficial structures on a solid surface are of vital importance to determine the surface wettability. In nature, purely smooth surfaces are rare, but the surfaces exhibit different degrees of roughness which have great impacts on solid−liquid interactions. Such interactions can be exemplified by the “lotus effect” or certain insects such as microvelia which can skate at the water surface due to the air trapping between the legs of the insects and water.1 Artificial patterning on solid surfaces in scientific and industrial fields has been extensively explored and is normally realized through creating topographic or chemical heterogeneities. Surface engineering, mainly through micropatterning, nanoprinting, coating, etching, etc., is commonly employed to study the wetting phenomenon, to control the solid−liquid interface, and to even induce self-driven motion of a liquid droplet in contact with a textured surface. Xu et al. improved the understanding on how a water droplet receding is affected by the roughness of regular patterned substrates. The pinning− depinning transition and the bottom area at the Cassie−Wenzel transition are sensitive to the surface roughness.2 The Cassie state indicates that the droplet sits above the patterned surface where air remains trapped below the droplet, and thus “airpockets” are formed. As it transits to the Wenzel state, the trapped “air-pockets” are replaced by liquid, so the droplet is fully in contact with the solid surface. Evaporative dynamics of © 2017 American Chemical Society
multicomponent droplets have been also investigated on the patterned surfaces, particularly the microscale stepwise motion of the three-phase line across island rows.3 Besides, by examining the dynamic contact line on the textured surface at microscale, Paxson et al. found that droplet adhesion was governed by capillary bridges at the receding three-phase line.4 Adhesion/release of droplets to/from a solid surface can be controlled by employing an anisotropic surface engineered with arrays of poly(p-xylylene) nanostructures. The retention force difference reached as high as 80 μN in the pinning and releasing directions.5 The micropatterned surface is capable of manipulating droplet shape as well. It was found that the droplet solid−liquid area was able to exhibit various polygons from squares to dodecagons on different substrates treated either topographically or chemically.6,7 In addition, spontaneous motion of a droplet could be manipulated on specific-designed patterned surfaces.8−13 Bliznyuk et al. devised a lithographically treated pattern featuring stripes of alternating high and low wettability. Such an anisotropic feature made the liquid propagate parallel and transport perpendicular to the stripes.16 Chu et al. created a unidirectional spread of a sessile droplet on an asymmetric Received: May 27, 2017 Accepted: August 1, 2017 Published: August 1, 2017 28055
DOI: 10.1021/acsami.7b07533 ACS Appl. Mater. Interfaces 2017, 9, 28055−28063
Research Article
ACS Applied Materials & Interfaces
side length d, diagonal length l, and roughness r denoting the ratio of the top surface area of a micropyramid to its bottom surface area. The geometric parameters are indicated in Figure 1c. The structure of the fabricated micropyramid was visualized by a confocal microscope (DCM8, Leica Microsystems) and presented in Figure 1b. The homogeneous saline solution samples were prepared by dissolving potassium chloride powders (KCl, Sigma-Aldrich, >99%) in nanofiltered water with resistivity at 18.2 MΩ cm. The initial concentrations CKCl of the KCl solutions were 0%, 5%, 10%, and 20%, which were lower than the saturation concentration at roughly Csat = 23.7%.21 During droplet evaporation, the loss of water led by evaporation resulted in an increase in salt concentration, and the droplet could reach a state of supersaturation, under which the crystalline formation could be initiated. The moment when crystalline starts to appear is defined as the precipitation time tp. At the late stage, the crystalline can grow to a degree that it defects the liquid−vapor interface or the three-phase line. However, we report contact angle and baseline length both prior to and after the deformation of the droplet spherical profile for the sake of providing full-spectrum information. 2.2. Droplet Characterization. The experimental setup is shown in Figure 1a. An optical microscope (Nikon Eclipse LV100ND) with the brightfield schema was employed to record droplet evaporation process from top view. Droplet images were captured every 1 s over the full-spectrum of evaporation. Simultaneously, two HiSpec-2 highspeed cameras were used to capture droplet side profile every 1 s from two side views. One view is parallel to the micropyramid bottom side, and the other is along the diagonal line of the bottom surface. The “parallel” and “diagonal” axes of viewing are therefore with a 45° interval as demonstrated in Figure 1a,c and are indicated as line-ofsight “∥” and line-of-sight “∠”, respectively. The droplets with the same initial volume controlled at 0.4 μL were dispensed by a micropipette (Thermo Fisher Scientific). The base diameters along the two lines-of-sight, and the wetting interface, namely the solid−liquid interface, were analyzed by the software of NIS Elements. Snapshots of droplet profile from side views were postprocessed by the program, ImageJ, to extract information on contact angles.22 To ensure experimental reliability, three tests were done for every CKCl. In the following figures of this article, the error bars of the parameters relevant to contact angle, depinning time, and precipitation time indicate standard deviations from the three repeated results along each line-of-sight. The error bars of the parameters relevant to baseline length along each line-of-sight indicate standard deviations from six results since along each line-of-sight there are two baseline length values for one droplet. The droplets were evaporating in an open condition with the surrounding temperature and humidity maintained at 22 ± 1 °C and 55 ± 2%.
nanostructured surface. Such an automatic spreading highly depended on the asymmetry of the nanostructure.17 The other motions in a variety, including droplet dynamic expansion/ contraction,13,14 spontaneous separation of droplet from a solid surface,15 actuation of streams of droplet,10 and selective transport of droplet11 correlated with surface anisotropy, can be found in ref 18. On the other hand, droplet wettability and its shape on patterned surfaces can be controlled by simply modifying the droplet solution composition, although the related studies are far rarer than the ones devoted to solid surface intervention. Courbin et al. initiated the control of droplet shape via using different liquid mixtures characterized by various equilibrium three-phase angles on the same textured surface. The droplet imbibition on the patterned surface had a square profile for a binary solution consists of 25 vol % ethanol and 75 vol % isopropanol, an octagonal shape for pure isopropanol, a rounded octagon for hexane liquid, and a circle for silicon oil.19 Droplet composition was also investigated by varying the water−ethanol ratio to control the shape of the solid−liquid interface on the substrate with micropyramid cavities. The bottom area of the binary droplet evolved from an octagon to a rectangle upon increasing the ethanol component.20 These approaches utilized liquid mixtures that consist of components with low surface tensions like alcohol or silicon oil. Droplets could barely form but evolve toward films at high concentrations of the low surface tension liquids. In our study, we employ solutions by dissolving potassium chloride (KCl) in pure water with different concentrations to obtain various high liquid−vapor surface tensions of the saline droplets. At first, the influence of KCl is analyzed on the wettability and shape of saline droplets at the beginning of evaporation in section 3.1. Afterward, the lifetime evaporation of droplets on various salt concentrations is discussed in section 3.2. Herein droplet depinning, an interesting phenomenon occurring in the middle of evaporation, is emphasized with respect to its dependence on the salt concentration and droplet shape evolution during the depinning from an entire view of the droplet. During the depinning phase, the contact line exhibits a unique behavior as across the micropyramids island from a close view, which is emphasized at last in section 3.3. These findings improve our understanding of the influence of droplet composition on evaporative dynamics on textured surfaces from a different angle through adopting liquid mixtures of high surface tensions.
3. RESULTS AND DISCUSSION 3.1. Droplet Initial Wettability. The influence of dissolved KCl salt on droplet wettability on the textured surface can be reflected by the wetting interface, contact angle, and baseline length of the droplet at the initial moment of evaporation. The initial contact angle and the initial baseline length, obtained at time 0 s of droplet evaporation, are denoted as θ0 and L0, respectively. The wetting interface of the pure water and saline droplets on the micropyramid surface exhibits an octagonal shape, despite the fact that it is partially blocked by the upper droplet body, as demonstrated in Figure 2a. The snapshots taken from the top view in Figure 2a also show that the octagonal wetting interface has two baselines with each one along line-of-sight “∥” and line-of-sight “∠”, respectively. The side profiles of the droplets in Figure 2a show that the initial contact angle along line-of-sight “∥” is larger than the counterpart along line-of-sight “∠”. This is consistent with the larger initial baseline length along line-of-sight “∠” than that along line-of-sight “∥”. The initial baseline length, L0, normalized to the averaged initial baseline line length of the
2. MATERIALS AND METHODS 2.1. Surfaces and Solution Samples. The poly(methyl methacrylate) (PMMA) substrates were patterned with micropyramid islands using nanoimprint lithography. The PMMA polymer was imprinted with a nickel shim mold at a temperature of 140 °C and a pressure of 20 bar for 10 min using an Eitre 6 nanoimprinter (Obducat) so that the polymer could fill up the cavities of the nickel shim mold. The demolding was done at 50 °C. Each substrate was cleaned to remove contaminants by rinsing under deionized water (Milli-Q) and drying by compressed nitrogen gas. Micropyramid geometries are summarized in Table 1, including the central height h,
Table 1. Geometric Parameters of the PMMA Substrate Engineered with Micropyramids h (μm)
d (μm)
l (μm)
r
11
30
23.89
1.24 28056
DOI: 10.1021/acsami.7b07533 ACS Appl. Mater. Interfaces 2017, 9, 28055−28063
Research Article
ACS Applied Materials & Interfaces
Figure 1. (a) Schematics of the experimental configuration. The droplet is simultaneously visualized by the microscope from top view and the fast cameras from side views along the two lines-of-sight. (b) The structure of micropyramids at a PMMA substrate. (c) Schematics of the micropyramids geometry characterized by the central height h, bottom side length d, and diagonal edge length l. The two lines-of-sight are along the side and the diagonal line of micropyramid bottom area, respectively.
condition that a droplet is placed on a smooth and homogeneous surface without any fluid motion, its contact angle is the equilibrium contact angle, θE, which satisfies Young’s equation: cos θE =
γSV − γSL γLV
(1)
where γSV, γLV, and γSL are the solid−vapor, liquid−vapor, and solid−liquid surface tensions, respectively. Note that Young’s equation describes an ideal situation with a smooth, rigid, and homogeneous surface. Therefore, the contact angle hysteresis is neglected and the equilibrium contact angle has a unique value for a certain droplet. In nonideal situations, however, solid surfaces are inevitably with topographical or chemical heterogeneities. So the observed contact angle on a solid surface is within a range which is confined between the receding contact angle, θR, and the advancing contact angle, θA. The equilibrium contact angle θE in reality can be evaluated by θR and θA:23
Figure 2. (a) Snapshots taken simultaneously from top view and side views along the diagonal and parallel lines-of-sight at time 0 s. (b) The initial baseline length, L0, normalized to the averaged initial baseline length of the pure water droplet, Lw0, and (c) the initial contact angle, θ0, normalized to the averaged initial contact angle of the pure water droplet, θw0, along each line-of-sight with respect to salt concentration CKCl.
⎛ Γ cos θA + ΓR cos θR ⎞ θE = arccos⎜ A ⎟ ΓA + ΓR ⎝ ⎠
(2)
where ⎞1/3 ⎛ sin 3θR ⎟ ΓR = ⎜ 3 ⎝ 2 − 3 cosθR + cos θR ⎠
pure water droplet, Lw0, along each line-of-sight, as demonstrated in Figure 2b, is found to decrease with CKCl, revealing that the wetting interface appears to be smaller at a higher salt concentration. Since the droplets were produced with the same initial volume, a smaller initial baseline corresponds to a larger initial contact angle. It can be seen from the side profiles in Figure 2a that the droplet has a larger contact angle at a higher CKCl. The initial contact angle, θ0, normalized to the averaged initial contact angle of the pure water droplet, θw0, along each line-of-sight is shown in Figure 2c. θ0/θw0 taken from both lines-of-sight is enlarged at a higher CKCl. The normalized initial baseline length L0/Lw0 and the normalized contact angle θ0/θw0 exhibit linear variations with CKCl, and such relations are fitted by the best fitting lines. The corresponding fitting equations are indicated in Figure 2b,c. The initial wettability reflects the equilibrium state of the droplet with different CKCl to some extent. Under an ideal
(3)
and ⎞1/3 ⎛ sin 3θA ⎟ ΓA = ⎜ 3 ⎝ 2 − 3 cosθA + cos θA ⎠
(4)
The receding and advancing contact angles were measured by the approach of adding and removing the volume of a droplet. The contact angle right before advancement of the droplet when its volume is added is referred to as the advancing contact angle. Contrary to it, the contact angle right before receding when the droplet volume is reduced is referred to as the receding contact angle. The averaged initial contact angle θ0, the receding and advancing contact angles θR and θA measured along the line-of-sight “∠” in our experiment, 28057
DOI: 10.1021/acsami.7b07533 ACS Appl. Mater. Interfaces 2017, 9, 28055−28063
Research Article
ACS Applied Materials & Interfaces together with the estimated equilibrium contact angle θE based on eq 2 are listed in Table 2.
means that the wetting interface evolves toward a rectangle with corners cut, and contrarily it alters to be a more regularized octagon. Once the depinning is initiated, the wetting interface loses its octagonal configuration and the ratio τ is unable to be obtained. The ratio, τ, is plotted as a function of CKCl in Figure 3. τ is increased with an increase in CKCl, suggesting that L∥ at the parallel sides is more prolonged as compared to L∠ at the diagonal sides, and hence the wetting interface evolves toward a rectangle with corners cut. The variation in τ, however, is not remarkable although CKCl is greatly increased from 0% to 20%. The configuration of the wetting interface and its variation with CKCl can be explained by using surface energy analysis. Herein we modify the free energy difference, Fpill, originally for the case of a vertically uniform cylindrical micropillar, to address our case of a vertically heterogeneous micropyramid. As shown in Figure 4a, the schematic diagram at the left side
Table 2. Averaged Initial Contact Angle θ0, the Receding and Advancing Contact Angles θR and θA, and the Equilibrium Contact Angle θE in Degrees along the Diagonal Line-ofSight at Various Salt Concentrations CKCl CKCl
θ0
θR
θA
θE
0% 5% 10% 20%
97.87 99.95 101.51 102.40
10.33 18.95 22.45 29.17
107.55 110.30 113.45 116.30
49.13 55.3 57.8 62.15
It can be seen from Table 2 that the initial contact angle θ0 is close to the advancing contact angle θA, and it is greatly larger than the equilibrium contact angle θE. It is attributed to the intrusive micropyramids which act like obstacles that prevent the liquid from spreading to the position if there were no such micropyramids. Such a confinement of the droplet makes θ0 analogous to θA. The equilibrium contact angle θE increases upon raising CKCl, and it reflects the role of the dissolved salt in varying the droplet wettability. The enhancement of θA by CKCl is primarily attributed to the increase in γLV led by the dissociation of KCl into potassium cations, K+, and chloride anions, Cl−, in an aqueous solution. The strong interactions between potassium cations and the partial negative oxygen atoms of water molecules, and between chloride anions and the partial positives hydrogen atoms of water molecules take place and hence strengthen the surface tension of water. In addition to that, introducing KCl in the water droplet also leads to an increase in γSL. The strong polarity of KCl molecules would enhance the overall polarity of the saline solution, making its affinity with the nonpolar PMMA substrate weaker. As a result, the interfacial tension, γSL, would be increased, and it further reduces the droplet wettability. Apart from droplet wettability, KCl also affects the shape of the wetting interface. The octagonal shape, as depicted in the inset of Figure 3, can be quantified by the side ratio τ =
L L∠
Figure 4. (a) Left: a schematic top view of a three-phase line intersecting a vertically uniform cylindrical pillar. Right: a side view of the pillar along the red arrow shown in the left schematics. (b) Upper: a part of the three-phase line intersects the diagonal edges of the micropyramids. Below: schematics of a micropyramid with its surface ADE and ABE in contact with solution. (c) Upper: a part of the threephase line intersects the bottom sides of the micropyramids. Below: schematics of a micropyramid with its side DE coinciding with the three-phase line.
,
where L∥ is the side length along line-of-sight “∥” and L∠ is the side length along line-of-sight “∠”. For each droplet, the ratio τ is calculated by dividing the summed lengths of four sides L∥ by those of the four sides L∠. It is found that L∥ stays longer than L∠ for all of the droplets during the initial pinning stage, so τ remains larger than unity in the present study. The rise of τ
shows a uniform cylindrical pillar from a top view. The schematic illustration at the right side indicates the side view of the pillar from the direction indicated by the red arrow in the left schematic diagram. A partial side wall of the cylindrical pillar is in contact with the solution, and the length of the solid−liquid interface is denoted as B. On the other hand, it would be a straight liquid−vapor interface given that there were no pillars obstructing the solution. So the length of the liquid− vapor interface is the one linking the two contacting threephase points at the pillar surface, and it is denoted as b. As a result, the areas of the solid−liquid and liquid−vapor interfaces are expressed as Bhp and bhp, respectively, where hp is the height of the pillar section immersed in solution. Therefore, Fpill, indicating the free energy difference between the straight liquid−vapor interface and the curved solid−liquid interface intersecting the pillar, is expressed as24
Figure 3. Initial side ratio τ at various salt concentrations CKCl. The inset indicates the two side lengths of the octagonal wetting interface along the two lines-of-sight, respectively.
Fpill = γSLBhp − γLVbhp 28058
(5) DOI: 10.1021/acsami.7b07533 ACS Appl. Mater. Interfaces 2017, 9, 28055−28063
Research Article
ACS Applied Materials & Interfaces
Figure 5. (a1) Normalized contact angle θ/θ0 and (a2) normalized baseline D/D0 at intermediate CKCl along the diagonal line-of-sight versus normalized evaporation time t/tf. (b1) Normalized contact angle θ/θ0 and (b2) normalized baseline D/D0 at intermediate CKCl along the parallel line-of-sight versus normalized evaporation time t/tf. The inset of (a2) is the reducing rate of contact angle kθ during the initial pinning stage with respect to CKCl. The inset of (b2) is normalized contact angle θ/θ0 versus normalized time t/tf for the pure water droplet and the one with CKCl at 20% in this study and the pure water droplet and the one with ethanol concentration at 15 vol % in ref 20.
In our case, in order to obtain Fpill, the manner of wetting on the micropyramids at three-phase line needs to be examined to determine the solid−liquid and liquid−vapor interfaces. The images of local contact line intersecting micropyramids along line-of-sight “∥” and line-of-sight “∠” are presented in Figure 4b,c. For the diagonal side L∠, the three-phase line intersects the upper two diagonal edges l of the micropyramid. It is illustrated by the lengths AD and AB in the schematics of Figure 4b. So the liquid−solid interface includes two identical triangles, SABE and SADE, and the planar liquid−vapor interface is the triangular area, SABD. Therefore, the surface energy difference, F∠, for one micropyramid along line-of-sight “∠” is F∠ =
1 2 γ d d 2 + 4h2 − γ hd 2 SL 2 LV
∂F∠ >0 ∂C KCl
(7)
⎞1/2 ∂γSL ⎛ d2 ∂γLV > ⎜ 2 + 2⎟ ∂C KCl ⎠ ∂C KCl ⎝ 2h
(8)
or
L∥ would always be more extended than L∠ with an increase of CKCl. By introducing the values of d and h in Table 1, it can be acquired that τ would be increased when ∂γSL ∂C KCl
> 0.42
∂γLV ∂C KCl
(9)
The degree that the wetting interface varies toward a rectangular profile with raising CKCl, although not as significant, manifests in the case of water−ethanol binary droplets on the PMMA substrate engineered with the microcavities as well.20 The side ratio was reported to increase from less than 1.5 for the pure water droplet to 3.5 for the one with 30 vol % ethanol. Different from KCl, ethanol introduced in water reduces both the solid−liquid and liquid−vapor surface tensions. Although the manner of contact line intersecting a microcavity is not provided, the variation in the free energy difference, resulting from the decreases in both the solid−liquid and liquid−vapor surface tensions, is presumably to make the extension of the parallel sides favored upon increasing the ethanol concentration. The unexpected conclusions can be drawn that ethanol and KCl, with opposite effects on liquid−vapor and solid− liquid surface tensions, lead to the same varying tendencies of the wetting interface profile.
(6)
On the other hand, the three-phase line along line-of-sight “∥” intersects the bottom sides of micropyramids (see Figure 4c). Therefore, the three-phase line is not deformed but remains straight, and thus the surface free energy difference F∥ = 0. As a result, F∠ should be larger than zero, so L∥ is longer than L∠ as observed in the experiments. It is worth noticing that the intersecting manner of the threephase contact line remains the same along the two lines-of-sight regardless of CKCl. The contact line length along each line-ofsight, however, varies with CKCl. The increasing ratio of τ suggests that the prolongation of L∥ is preferred over that of L∠ upon raising CKCl, so F∠ should be increased with a higher loading of salt KCl. Because both γLV and γSL are enhanced as CKCl is increased, according to the formula of F∠, as long as 28059
DOI: 10.1021/acsami.7b07533 ACS Appl. Mater. Interfaces 2017, 9, 28055−28063
Research Article
ACS Applied Materials & Interfaces 3.2. Evaporation Dynamics across Droplet Lifetime. Dissolved KCl in the droplet also influences its lifetime evaporation dynamics, reflected by the evolutions of the contact angle and baseline length at various salt concentrations. The contact angle and baseline length from each line-of-sight, normalized to its respective initial value obtained at 0 s of evaporation, are plotted in Figure 5. It can be seen that for the saline droplets with intermediate CKCl the contact angle reduces and the baseline keeps nearly constant at first, suggesting that the droplets experienced the constant contact radius mode (CCR) which features the decreasing contact angle and constant radius of a droplet during its evaporation. Afterward the contact angle remains unchanged, while the baseline is reduced, indicating the droplets underwent the constant contact angle mode (CCA) which is characterized by the constant contact angle and decreasing radius of a droplet in the proceeding of evaporation. Different from the saline droplets, the pure water droplet experienced the CCR, CCA, and at last the mixed mode characterized by simultaneously decreasing contact angle and baseline. During the initial CCR stage, the contact angle decreases while the baseline remains nearly unchanged. By taking the best fitting line of the contact angle curve during the CCR stage, the averaged reducing rates of contact angles at various CKCl are obtained and plotted in the inset of Figure 5a2. The contact angle decreases more slowly at a higher CKCl, suggesting the suppression on evaporation by salt. As for the latter CCA stage, the baseline is reduced and contact angle remains roughly constant at various CKCl. The reducing rate of the baseline during the CCA stage seems to be attenuated upon increasing the CKCl as well. To compare the evaporation dynamics of saline droplets to those of water−ethanol binary droplets on textured surfaces, we plot the contact angle normalized to its initial value versus the evaporation time normalized to the droplet lifetime in the inset of Figure 5b2 for the pure water droplet and the one with CKCl at 20% in our study and for the pure water droplet and the one with ethanol concentration at 15 vol % in ref 20. Surprisingly, the two water droplets used as control groups and the water− ethanol binary droplet have nearly overlapped contact angle curves. It suggests that ethanol has little effect in varying the evaporation regimes of droplets on the patterned surface. The saline droplet at 20% of CKCl, however, undergoes CCR and CCA modes as suggested by the contact angle evolution, and the contact angle decreases much slower than the other droplets during the CCR stages. The lack of a mixed mode at the late stage of evaporation is due to the salt crystallization which pulls the solution to cover its side. As a result, the solution is prevented from descending toward the substrate, and thus the contact angle is maintained at a relatively high and constant value until the end of drying. In the middle of droplet evaporation, it is interesting to find that the depinning occurs nonsimultaneously for the threephase line along line-of-sight “∥” and line-of-sight “∠”, and such an unsynchronized depinning makes the octagonal wetting interface evolve to a circular shape. It is found that the threephase line along line-of-sight “∠” recedes ahead of the counterpart along line-of-sight “∥”. The receding starts to occur from the two ends of the three-phase line along line-ofsight “∠”, as shown in Figure 6a2,b2 for the pure water and the one at 5% of CKCl, and consequently the distinct angles highlighted in Figure 6a1,b1 of the octagonal shape become rounded as shown in Figure 6a2,b2. Afterward, the rest section of the three-phase line along line-of-sight “∠” starts to recede
Figure 6. Depinning process of the droplet with CKCl at (a) 0%, (b) 5%, and (c) 20%. (a1 and b1) The octagonal wetting interface right before depinning taking place for the droplets with CKCl at 0% and 5% respectively; (a2 and b2) depinning begins from the two ends of the contact lines along diagonal line-of-sight, as indicated by the white arrows; (a3 and b3) the rest parts of the contact lines along diagonal line-of-sight start to recede along the direction as indicated by the white arrows; (a4 and b4) the wetting interface becomes a circular shape. (c1) for the droplet with CKCl at 20%, crystalline is formed prior to depinning; (c2) depinning occurs from the two ends of the contact lines along diagonal line-of-sight as indicated by the white arrows; (c3) the droplet is partially pinned and the rest part is pulled toward the crystalline; (c4) the octagon evolves to an irregular round shape enclosing the crystalline.
row by row, as denoted by the arrows in Figure 6a3,b3. As a result, the octagonal interface progressively becomes a circle (Figure 6a4,b4). Such variations in the wetting interface at depinning are also shown in Videos S1 and S2 for the pure water and the 5% CKCl droplet. At 10% and 20% of CKCl, precipitation emerges before depinning taking place (see Figure 7a and Video S3 for the 20% CKCl droplet). As presented in Figure 6c, the crystalline is formed (Figure 6c1) prior to the onset of receding (Figure 6c2) for the droplet with CKCl at 20%. The formed crystalline pins partial contact line (Figure 6c2,c3) and pulls the remaining solution toward it, during which the octagon evolves to a circular shape as well due to the earlier receding of the diagonal sides L∠. The onset of contact line depinning, whether along line-ofsight “∥” or line-of-sight “∠”, is found to moderately depend on CKCl. As shown in Figure 7a, the depinning time tr normalized to droplet lifetime is delayed at a higher CKCl for the threephase lines along both the lines-of-sight. One reason is that the larger initial contact angle formed at a higher CKCl as shown in Figure 2 devotes to the gap between it and the contact angle θr at the receding moment. Another reason is the occurrence of precipitation prior to depinning at high CKCl. Figure 7a shows that the salt crystalline is initiated ahead of depinning at 10% and 20% of CKCl. As the crystallization begins at the vicinity of the wetting interface (Figure 6c), the local pinning effect could be enhanced, and thus the onset of depinning would be postponed. The contact angles at the onset of receding θr for both the lines-of-sight are plotted in Figure 7b. θr increases moderately 28060
DOI: 10.1021/acsami.7b07533 ACS Appl. Mater. Interfaces 2017, 9, 28055−28063
Research Article
ACS Applied Materials & Interfaces
Figure 8. Droplet volume evolution versus time normalized to droplet lifetime t/tf. Inset indicates two radius along the two lines-of-sight for calculation of the droplet volume.
bars are plotted in Figure 8. It can be seen that the drop volume declines slower at a high CKCl, and it basically reduces linearly. The decline of the evaporation rate with respect to CKCl suggests that salt suppresses sessile droplet evaporation on the textured surface. It is mainly attributed to two reasons. One is the smaller initial wetting interface at a higher CKCl even though its pinning stage is longer. The other one is the higher liquid−vapor surface tension at a higher CKCl which makes it more difficult for water molecules to escape from the liquid− vapor interface. 3.3. Close View of Contact Line Depinning. The droplets regardless of the salt concentration exhibit an initial pinning stage followed by a depinning one. The contact line exhibits an unique depinning behavior as across the micropyramid islands (see Video S4), in contrast to the smooth receding on homogeneous substrates reported in many other studies. Once the depinning is initiated, the contact line shows the repeated “slow slip−rapid slip” behavior. Such a stepwise motion at the microscale is demonstrated in Figure 9 by both the snapshots and corresponding schematics along line-of-sight “∥”. Figure 9 (I) indicates the moment right after a depinning finished. The three-phase line adhering along the ridged surface of several neighboring micropyramids exhibits a zigzag profile. As evaporation continues, the side of the three-phase line jumps to the next micropyramid row earlier than the remaining threephase line does (see Figure 9 (II)). The remaining three-phase line tends to resist depinning such that it is slowly pulled straight along the bottom boundary of the micropyramids, and it is referred to “slow-slip”. Once the remaining three-phase line fails to hold the slow slipping it rapidly jumps and pins at the surface of next micropyramid row (Figure 9 (III)), again presenting a zigzag outline resembling that in Figure 9 (I). Such a moment represents the finishing of depinning for the last row and the beginning of receding for the present row. Then the three-phase line repeats the process that it shrinks row by row until the droplet is dried out. The three-phase line slips each time by approximately 20 μm which is two-thirds of side length d, and it is slowly pulled by roughly one-third of d before next rapid depinning. Such a “slow slip-rapid slip” behavior differs from the “stick−slip” of droplets on substrates engineered with vertically homogeneous islands. The three-phase line can rapidly across over the planar top surfaces of islands before it reaches the boundaries of next island row.3 However, in our study the slightly tilted top
Figure 7. (a) Normalized depinning time tr along the two lines-ofsight and normalized precipitation time tp with respect to CKCl. (b) The contact angle θr at the onset of receding along the two lines-ofsight with respect to CKCl. The inset is the contact angle reduction Δθ during the initial pinning stage at intermediate CKCl.
with an increase in CKCl. The extent that the droplet is away from its equilibrium state can be reflected by the inward force f acting at the unit length of the three-phase line. f is expressed as f = γLV(cos θ − cos θE), where θ is the dynamic contact angle. It is shown in Table 2 that θE is larger on the textured surface at a higher CKCl. Besides, γLV is increased upon raising CKCl due to the interactions between the electrolytes of KCl and water molecules. As a result, a relatively small reduction of the dynamic contact angle θ can make f greatly pronounced at a higher CKCl. The salt-enhanced equilibrium contact angle, θE, further ensures a larger θr with the loading of KCl. In addition, θr along line-of-sight “∠” is smaller than the one along line-ofsight “∥”. It is attributed to the lower initial contact angle θ0 along line-of-sight “∠” which makes f earlier reach the value affordable for depinning than the one along line-of-sight “∥”. In addition, it is notable that the onset of shrinkage is postponed although the dynamic contact angle only needs a small reduction to initiate droplet shrinkage at a higher CKCl. One mentioned reason is the occurrence of salt precipitation which partially pins the three-phase line. The other reason is the slower declining trend of θ during the CCR stage at a higher CKCl (see the inset in Figure 5a2), resulting from the attenuated evaporation rate led by KCl. As the droplet bottom is octagonal, we calculate its volume based on two different circular wetting interface areas and contact angles, respectively. As indicated by the inset in Figure 8, one circular wetting interface has a radius R∠ which is along line-of-sight “∠”. The other circle has a radius R∥ along line-ofsight “∥”. Therefore, two volumes are acquired, with one calculated based on R∠ and the dynamic contact angle taken along line-of-sight “∠”, and the other one based on R∥ and the dynamic contact angle taken along line-of-sight “∥”. The two volumes representing the upper and lower limits of the float 28061
DOI: 10.1021/acsami.7b07533 ACS Appl. Mater. Interfaces 2017, 9, 28055−28063
Research Article
ACS Applied Materials & Interfaces
shape control and evaporation control by manipulating solution composition.
■
ASSOCIATED CONTENT
S Supporting Information *
This material is available free of charge via the Internet at http://pubs.acs.org/. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.7b07533. Video S1 showing the depinning stages of the 0% CKCl droplet. (AVI) Video S2 showing the depinning stages of the 5% CKCl droplet. (AVI) Video S3 showing the depinning stages of the 20% CKCl droplet. (AVI) Video S4 shows the “slow slip−rapid slip” of the threephase line at a close view for the 20% CKCl droplet. (AVI) A description of supporting information provides the details of the above four videos. (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Figure 9. Snapshots and corresponding schematics of the depinning three-phase line along line-of-sight “∥” at (I) t1, (II) t1 + 7 s, (III) t1 + 10 s, and (IV) t1 + 16 s, where t1 = 648 s, for the 20% CKCl droplet at the micropyramid surface.
Fei Duan: 0000-0002-7469-7184 Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS X.Z., J.R., and F.D. are grateful for the financial support from AStar SERC A1783c0006. K.S.L.C. and K.S.O. acknowledge AStar Nanoimprint Foundry under grants (1325307091 and 1525300037). The authors would like to thank Dr. Huicheng Feng for the discussion of the experiments.
surfaces of micropyramids allow the three-phase line to slowly slide upon them until its next jump taking place.
4. CONCLUSIONS We experimentally examined the effect of dissolved KCl salt on the wetting interface shape and evaporative dynamics of drying aqueous droplets on a solid surface patterned with micropyramids. At the beginning of evaporation, upon increasing the salt concentration, the maximum octagonal wetting of the saline droplet was decreased, and the droplet wetting interface moderately evolved toward a rectangle with corners cut. The initial side ratio, defined as the ratio of three-phase line lengths intersecting micropyramid bottom sides to diagonal edges, was slightly increased as the salt concentration was raised from 0% to 20%. It is due to the dissolved KCl salt which enhanced both the liquid−vapor and solid−liquid surface tensions and thus resulted in preferred extension of the three-phase line intersecting micropyramid bottom sides. With respect to the evaporation dynamics across the droplet lifetime, the saline droplet went through CCR and CCA stages, while the pure water droplet experienced CCR, CCA, and a mixed stage at last. The transition from the CCR stage to CCA stage was postponed, and the receding contact angle was larger with increasing the salt concentration. Once depinning was initiated, the octagonal wetting interface altered toward a circle until the end of drying. From a close view at the droplet edge, it was found that the three-phase line repeated “slow slip−rapid slip” row by row of micropyramid islands during the depinning phase. Our study could help a further understanding of the wetting interface behavior and evaporative dynamics of drying sessile droplets with the increased surface tension on the patterned surface and could provide a new path for droplet
■
REFERENCES
(1) Quéré, D. Wetting and Roughness. Annu. Rev. Mater. Res. 2008, 38, 71−99. (2) Chen, Xu.; Ma, R.; Li, J.; Hao, C.; Guo, W.; Luk, B. L.; Li, S. C.; Yao, S.; Wang, Z. Evaporation of Droplets on Superhydrophobic Surfaces: Surface Roughness and Small Droplet Size Effects. Phys. Rev. Lett. 2012, 109, 116101. (3) He, M.; Liao, D.; Qiu, H. Multicomponent Droplet Evaporation on Chemical Micro-Patterned Surfaces. Sci. Rep. 2017, 7, 41897. (4) Paxson, A. T.; Varanasi, K. K. Self-Similarity of Contact Line Depinning from Textured Surfaces. Nat. Commun. 2013, 4, 1492. (5) Malvadkar, N. A.; Hancock, M. J.; Sekeroglu, K.; Dressick, W. J.; Demirel, M. C. An Engineered Anisotropic Nanofilm with Unidirectional Wetting Properties. Nat. Mater. 2010, 9, 1023−1028. (6) Raj, R.; Adera, S.; Enright, R.; Wang, E. N. High-Resolution Liquid Patterns via Three-Dimensional Droplet Shape Control. Nat. Commun. 2014, 5, 4975. (7) Kumar, A.; Raj, R. Droplets on Microdecorated Surfaces: Evolution of the Polygonal Contact Line. Langmuir 2017, 33, 4854− 4862. (8) Tenjimbayashi, M.; Higashi, M.; Yamazaki, T.; Takenaka, I.; Matsubayashi, T.; Moriya, T.; Komine, M.; Yoshikawa, R.; Manabe, K.; Shiratori, S. Droplet Motion Control on Dynamically Hydrophobic Patterned Surfaces as Multifunctional Liquid Manipulators. ACS Appl. Mater. Interfaces 2017, 9, 10371−10377. (9) Lim, H. S.; Han, J. T.; Kwak, D.; Jin, M.; Cho, K. Photoreversibly Switchable Superhydrophobic Surface with Erasable and Rewritable Pattern. J. Am. Chem. Soc. 2006, 128, 14458−14459.
28062
DOI: 10.1021/acsami.7b07533 ACS Appl. Mater. Interfaces 2017, 9, 28055−28063
Research Article
ACS Applied Materials & Interfaces (10) Darhuber, A. A.; Valentino, J. P.; Troian, S. M.; Wagner, S. Thermocapillary Actuation of Droplets on Chemically Patterned Surfaces by Programmable Microheater Arrays. J. Microelectromech. Syst. 2003, 12, 873−987. (11) Agapov, R. L.; Boreyko, J. B.; Briggs, D. P.; Srijanto, B. R.; Retterer, S. T.; Collier, C. P.; Lavrik, N. V. Length Scale Selects Directionality of Droplets on Vibrating Pillar Ratchet. Adv. Mater. Interfaces 2014, 1, 1400337. (12) Sekeroglu, K.; Gurkan, U. A.; Demirci, U.; Demirel, M. C. Transport of a Soft Cargo on a Nanoscale Ratchet. Appl. Phys. Lett. 2011, 99, 063703. (13) Duncombe, T. A.; Erdem, E. Y.; Shastry, A.; Baskaran, R.; Böhringer, K. F. Controlling Liquid Drops with Texture Ratchets. Adv. Mater. 2012, 24, 1545−1550. (14) Tadmor, R. Drops That Pull Themselves Up. Surf. Sci. 2014, 628, 17−20. (15) Tadmor, R.; Das, R.; Gulec, S.; Liu, J.; N’guessan, H. E.; Shah, M.; Wasnik, P. S.; Yadav, S. B. Solid-Liquid Work of Adhesion. Langmuir 2017, 33, 3594−3600. (16) Bliznyuk, O.; Jansen, H. P.; Kooij, E. S.; Zandvliet, H. J. W.; Poelsema, B. Smart Design of Stripe-Patterned Gradient Surfaces to Control Droplet Motion. Langmuir 2011, 27, 11238−11245. (17) Chu, K. H.; Xiao, R.; Wang, E. N. Uni-Directional Liquid Spreading on Asymmetric Nanostructured Surfaces. Nat. Mater. 2010, 9, 413−417. (18) Holmes, H. R.; Böhringer, K. F. Transporting Droplets through Surface Anisotropy. Microsyst. Nanoeng. 2015, 1, 15022. (19) Courbin, L.; Denieul, E.; Dressaire, E.; Roper, M.; Ajdari, A.; Stone, H. A. Imbibition by Polygonal Spreading on Microdecorated Surfaces. Nat. Mater. 2007, 6, 661−664. (20) Feng, H.; Chong, K. S.; Ong, K. S.; Duan, F. Octagon to Square Wetting Area Transition of Water-Ethanol Droplets on a Micropyramid Substrate by Increasing Ethanol Concentration. Langmuir 2017, 33, 1147−1154. (21) Pinho, S. P.; Macedo, E. A. Solubility of NaCl, NaBr, and KCl in Water, Methanol, Ethanol, and Their Mixed Solvents. J. Chem. Eng. Data 2005, 50, 29−32. (22) Stalder, A. F.; Melchior, T.; Muller, M.; Sage, D.; Blu, T.; Unser, M. Low-Bond Axisymmetric Drop Shape Analysis for Surface Tension and Contact Angle Measurements of Sessile Drops. Colloids Surf., A 2010, 364, 72−81. (23) Tadmor, R. Line Energy and the Relation between Advancing, Receding, and Young Contact Angles. Langmuir 2004, 20, 7659−7664. (24) Robson, S.; Willmott, G. R. Asymmetries in the Spread of Drops Impacting on Hydrophobic Micropillar Arrays. Soft Matter 2016, 12, 4853−4865.
28063
DOI: 10.1021/acsami.7b07533 ACS Appl. Mater. Interfaces 2017, 9, 28055−28063
