Orientation-Induced Effects of Water Harvesting on Humps-on-Strings of Bioinspired Fibers

Smart water-collecting functions are naturally endowed on biological surfaces with unique wettable microstructures, e.g., beetle back with “alternate hydrophobic, hydrophilic micro-regions”, and spider silk with wet-rebuilt “spindle-knot, joint” structures. Enlightened by the creature features, design of bio-inspired surfaces becomes the active issue in need of human beings for fresh water resource. Recently, as observed from spider web in nature, the net of spider silk is usually set in different situations and slopes in air, thus spider silks can be placed in all kinds of orientations as capturing water. Here, we show the styles and orientations of hump-on-string to control the ability of water collection as bioinspired silks are fabricated successfully. As different strings, sizes (height, length, pitch) of humps can become the controlling on volumes of extreme water drops. It is related to the different solid/liquid contact regions resulting in the as-modulated wet adhesion due to orientations of humps-on-strings. The conversion of high-low adhesion can be achieved to rely on orientations for the effect of capturing water drops. These studies offer an insight into enhancement of water collection efficiency and are helpful to design smart materials for controlled water drop capture and release via conversions of high-low adhesion.

in the PVDF/DMF solution with a drawing out velocity. A polymer solution film would be uniformly coated on this fiber. Then the film is quickly broken up into multi-level humps via Rayleigh instability. In our study, we use the main fibers with different diameters (e.g., d = 20 μ m, 45 μ m, 80 μ m, and 98 μ m, where d is diameter of fiber) to be immersed into poly(vinylidene fluoride, PVDF, M w = 530000)/DMF (N,N-dimethylformamide, 9%, 10%, 11% or 12% (wt)) solutions. After being drawn out of the PVDF/DMF solution with a velocity 200 mm/min, the as-coated polymer solution film is quickly broken up into periodic droplets. Dried after 10 min in the 60 °C, the humps-on-strings form on fibers. Figure 1a shows the optic images of a series of bioinspired fibers with humps-on-strings. The diameters of strings are near to diameter of fiber due to formation of periodic polymer droplets by Rayleigh instability. As for the fabrication of bioinspired fiber 9,17 , the polymer solution is one of crucial importance influence factor, e.g., given fiber (d = 20 μ m), the size (e.g., L, H) of humps with higher concentration of solution (10% and 11%) are bigger than that with lower one (9%). This is because there is more PVDF solution over on the fiber after drawing out the solution with the higher concentration. But it doesn't mean that the size of hump is directly proportional with solution concentration. For an excessive high concentration, it would generate a high gravity to deform the formation of humps. The red sphere between the dotted lines in Fig. 1a has vividly shown that the parameters present a parabola with the concentration increasing. Figure 1b  , where H is the semiminor axis of hump (the height of hump); L is the semimajor axis long the surface of normal fiber and d is the diameter of string (estimated by diameter of main fiber); x is the one side of solid-liquid contact line along axis of fiber; dx is the integrating unit line of solid-liquid contact at x position. Thus the solid-liquid contact areas for larger hanging-droplets can be predicted by optimizing the concentration of polymer and diameter of main fiber. As for diameters for main fibers with d = 20 μ m, 45 μ m, 80 μ m and 98 μ m, the proper concentration can be 10%, 10%, 11% and 12% for excellent formation of humps, respectively.
As reported 15 , the hanging water droplet ability is related to the three-phase contact line, which is in proportion to the adhesion force at solid-liquid interface. Figure 1c shows the photos of droplets that are hung on humps of fibers (i.e., d = 20 μ m and 98 μ m), where the volume of water droplet can be expressed with 2 . It indicates that larger hump increases the solid-liquid contact area to hang the larger hanging-droplet effectively. 2 . Generally, the tendency is that larger contact area induces larger hanging droplet in volume (e.g., d = 98 μ m, V max = 14 μ L). Figure 1d shows the relationship between volumes of hanging-drops and solid-liquid contact areas. The statistical data of the solid/liquid contact area of one hump versus hanging-drop volume is fitted by the testing data. Given the same 10% PVDF solution coated on different main fibers (i.e., d = 20 μ m, 45 μ m, 80 μ m and 98 μ m), the solid-liquid contact area can be different for volumes of hanging-droplet. The volumes of water droplets are increasing with the solid/liquid contact area increasing. The solid-liquid contact areas are determined by geometries of humps formed finally on fiber. We define the curvature of the hump κ = α = ( )/ tan H d L, where α is the semi-axis angle; H is the hump of height; L is the length of hump, κ is directly related to the Laplace pressure in difference. Figure S1 shows the κ value versus concentration of PVDF (9%, 10%, 11% and 12%) for main fibers with different diameters. The average of κ value obviously decreases with the diameter increasing, which means the diameters of main fibers result in the differences of hump geographies, in addition to polymer solution concentrations and drawing velocities. Figure S2 shows the obvious tendency with diameter increasing at 9%. We also prepared different categories of fiber including nylon (d = 20 μ m), silkworm silk (d = 23 μ m), copper wire (d = 65 μ m) and fishing line (d = 163 μ m) in Figure S3. The change of κ value is induced by materials property and diameter of main fiber. When κ value is bigger, namely, the diameter of main fiber is less than others. The semi-axis angle is too big which results that the water droplet is easy to slip down from the string part to hump surface. The solid/liquid contact area is sharply decreasing, which results that the experimentally area is less than the calculative value. The volume of hanging water droplet is decreasing accordingly. When κ value is too small with a bigger diameter of fiber string, even the solid/liquid contact area is not big enough, the volume of water droplet is more than that bigger solid/liquid contact area humps with thin diameter. It results from the actual extensive contact area on string between humps. It is visible that the topography directly affects the solid/liquid contact area, which influences the water hanging ability 29 . Specifically, environmental SEM images of periodic humps of spider silk with the half apex angle of 9.5° ± 2.5°, κ is 0.172 ± 0.02. For the humps with middle κ value, the relation of water droplet volume and solid/liquid contact area can be depicted by : e V 9 79651 6 74178 9 79651 1 x 21060 3 1286 27 , where the volumes of hanging droplets can be evaluated effectively according to the sizes of humps.
Bioinspired fibers weaving net would be expected to have a promising application for water collection, e.g., fog collectors [30][31][32] . In the process of water collection, droplet is collected usually on multi-humps. Especially, the efficiency of water collection is determined by structures of humps-on-strings on a fiber. The wet adhesion to droplet would be determined by multi-humps. For instance, a droplet is stably hung on two humps with two κ values. Two fibers (i.e., d = 20 μ m and 98 μ m) have the same pitch of both humps (P = 865 μ m), the humps-on-strings are obliquely placed at tilt angles (β = 0°, 5°, 10°, 20°, 30°, 40°) in humidity of ~95%. Figure 2a shows the photos of hanging-drops pinned by humps-on-strings of both fibers (defined as d 20 Figure S4. Water collecting time on the second fiber declines from ~139.2 s to ~49.2 s. The d 98 fiber with more string diameter presents more drop volume and more collecting time in the once water collecting process. We define efficiency in single-cycle ϕ = V max /t max , where V max and t max are the most volume of water drop and collecting time once water droplet hanging process at the same title angle, respectively. In multiple repeated statistical analysis, not only for the d 20 fiber with more κ value, but also for the d 98 one with less κ value, both of the two ϕ values are appearing a peak value between 5° and 10° (Fig. 2b). It implies that a high adhesion state can bear a bigger water drop in single-pass water collecting process in the horizontal situation, but it does not mean the highest collecting efficiency. The adhesion force is adjusted by a title angle. Then, water drop is hanging on a low adhesion state. They can quickly slip down in a transient time. It can promptly circle into next water collecting process. The efficiency achieves the extremum at humps-on-strings with β = 10°. We introduce the circulation collecting water rate with different tilt angle to illustrate the collection efficiency. The cycle-index is stipulated for five times. The water collection efficiency in multi-circles: Ψ = V total /t total , which the V total and t total are respective the total water collection volume and collecting time during water hanging in multi-cycles. As shown in Fig. 2c, multiple circulation water collecting also is generated an extreme value at β = 10°. Notably, the water drop collecting rate is obvious higher than single-pass process. This results from the coalescence with peripheral tiny water droplet on the string. It follows that the adhesion station decides the most volume of hanging water drops, but the highest collecting efficiency appears on the lower adhesion station. The significance of this work is that we can attempt to use different adhesion station to complete water collection. Based on the efficiency of circulation, suitable adhesion station should be chosen in our application.
As for the water capturing application, we utilize the statistical data to prove that high efficiency water collecting is derived from fleetly water droplet condensation and drop out. Changing the adhesion station not only can improve the water efficiency, but also can release the hanging water droplets. In our research, we further provide four adhesion stations, including horizontal-humps, slope-humps, uniclinal-humps and double-incline-humps, as illustrated in the optical images of d 20 and d 98 fiber in Fig. 3a,b. Different adhesion states represent discriminative hanging water drop ability. Here, we depict the most volume of hanging water drops between two humps under four adhesion states. As shown in Fig. 3a, the d 20 fiber can be H = 120 ± 10 μ m, L = 280 ± 20 μ m, P = 700 μ m. P is pitch between two humps. The red line explicitly reveals the outline changing of solid/liquid contact line under different adhesion states. The most hanging water drop volume also assumes sharply decreasing with the changing from high adhesion to low adhesion state. There are V = 7.73 μ L, 6.30 μ L, 4.25 μ L and 3.15 μ L under four adhesion states, respectively. Water drops will release quickly once the volume exceeds the most value. Meanwhile, for the most hanging water drop volume, the humps on d 98 fiber also show the similar regularity, as demostrated in Fig. 3b. We choose the same P = 700 μ m, d = 20 μ m, H = 200 ± 10 μ m, L = 640 ± 20 μ m, respectively. From Scientific RepoRts | 6:19978 | DOI: 10.1038/srep19978 the optical images, the red solid/liquid contact line shows the same changeable feature. For horizontal-humps, slope-humps, uniclinal-humps and double-incline-humps adhesion states, there are V = 13.11 μ L, 10.88 μ L, 9.02 μ L and 5.74 μ L, respectively. However, for the uniclinal-humps and double-incline-humps adhesion states, the volume is the most volume hanging on two humps, but not the most growing volume before dropping down instantaneously. Namely, the drop hanging between two humps is not the most volume. It is different from the thinner bioinspired fiber, e.g., d = 20 μ m. As for horizontal-humps and slope-humps adhesion states, the most volume of water drop hangs on two humps. They are dropping along the contact line and slipping out from the humps. But for the uniclinal-humps and double-incline-humps adhesion states, when the droplet drops from the hump to the joint-part, the right end of left hump or the left end of right hump, the water drops are sequentially growing until the adhesion force unable to compete the gravity, which results in the most hanging volume appearing on one hump and uniform fiber string, and not between two humps. This implies that the adhesion state is not only related to the contact area of hump and water drops, but also to the fiber string diameter. Detailed discussion about relationship of adhesion state and hanging water drop volume will be given in the next part. Figure 3c illustrates the difference of hanging water drops condition on four adhesion situations. The scheme of horizontal-humps (H h ), slope-humps (S h ), uniclinal-humps (U h ) and double-incline-humps (D h ) shows the change on the apparent receding angle θ , the semi-apex-angle α of humps and solid/liquid contact surface (the red dotted line). As known, a large water drop hangs stably on a thin fiber, which is attributed to the solid/liquid adhesion by the wetted feature 28 . Generally, the transformation of high and low adhesion states is decided by adhesion force, which is balanced by the gravity of hanging-drop. However, the adhesion force is provided by two humps and the uniform fiber between them. The capillary adhesion for a larger pearly hanging-drop on a rough-curve fiber can be thereby described as: = + F F F 2 hump fiber , where F hump is the combination of the two forces, including the capillary adhesion of knots and the compensatory factor on the difference in Laplace pressure resulting from the geometry of the humps. The description of knots force is introduced as follows: , here, the capillary adhesion force is described as , where x is the coordinate variable along the fiber axially, x a and is the initial sites located at the end of left hump for the integration and x b is the terminate one at the right end in one hump, θ( ) x is the crossing angle between the liquid surface and horizontal reference plane, r is the roughness of geometry of bioinspired humps, γ is the surface tension of water. And Δ P can be described as: F knot total in four state conditions. Another force ( ) F fiber is also taken into account. In the previous reports 24 , the capillary force in wettability fiber is exceeding πγ 2 D resulted from the stable solid/ liquid contact line length between two knots. However, the high and low adhesion states present different fiber length, which results in the different force. For the d 20 fiber, the F fiber is insignificant related to F 2 hump . However, for the d 98 fiber, the F fiber can not be ignored. In Fig. 3b, we give the water drop hanging on two knots in four conditions. In fact, the water drops is not dropped out with the time growing in U h and D h conditions. The water drops will creep down from the knot and further grow on the fiber. The F fiber is big enough to balance the gravity of water drop. In conclusion, these investigations indicated that the capillary adhesion for a larger pearly hanging-drop on a rough-curve fiber could be controlled by the cooperation of three main forces: − F hump left , − F hump right and F fiber . According to Fig. 3a,b and sketches in Fig. 3c, the most volume of hanging-drop could be predicted by different wetting condition. Figure 4 depicts the water collection efficiency and the frequency appearing the ideal water drops during one hour under 95% humidity with four adhesion states, as H h , S h , U h and D h . As for thinner fiber (d = 20 μ m), spindles are L = 280 μ m, H = 140 ± 8 μ m and P = 720 μ m. As for thicker fiber (d = 98 μ m), spindles are L = 666 μ m, H = 200 ± 13 μ m and P = 720 μ m, respectively. Both of the bio-inspired fiber is fabricated with 9% PCDF concentration. We calculated the water collection frequency (f) under different adhesion states during one hour. We define the maximum volume of 80% as the ideal drop state. During one hour, the frequency appearing ideal drops is one of measurement criterion on bioinspired fiber working stability. For different main fibers, the water collection efficiency is obviously distinct by the cooperation of F hump and F fiber . As the yellow and green dot shown in Fig. 4, on d 20 fiber, the f is between 42% and 70%. While on d 98 fiber, the f > 94%. It indicates that the less bioinspired fiber is more susceptible to the external environment, such as the surrounding water drops, humidity change and even air movement. The previous research 33 have reported water collecting efficiency with different dip-coated mesh surface, and discussed the influence factors including wind velocity, coating of chemistry, water content. It can be seen that a number of influence factors should be considered to evaluate the water harvesting efficiency. Further, we next work will focus on the influence factors on water collecting efficiency.  This study is significant for developing smart materials to control liquid drop capture and controlled release by changing the adhesion condition. These findings are important in many potential applications, such as the transport of microdroplets 34 , water-harvesting from high humility 35,36 and liquid collection in high efficiency 33 .

Methods
Fabrication of bioinspired fibers. The main (nylon) fibers with diameters (d) were 20 μ m, 45 μ m, 80 μ m and 98 μ m (defined as d 20, d 45, d 80 , and d 98 , respectively). The multi-scale humps were generated via Rayleigh instability technique 36 and the dip-coating method. The main fibers were firstly immersed into poly(vinylidene fluoride, PVDF, M w = 530000)/DMF (N,N-dimethylformamide, 9 wt%) solution. Then, the main fibers were vertically drawn out of the PVDF/DMF solution with a velocity 200 mm/min by a dip-coater machine (DipMaster-50, China). A coated polymer solution film quickly was broke up into periodic droplets and dried to resemble-spider-silk humps after 10 min in drying oven at 60 °C. Thus the bioinspired fibers were fabricated successfully. In detail of process, the dynamic wettability of formation on humps-on-strings was investigated by a using OCAmicro40 meter system (Data-Physics, Germany). And the size parameters, solid/liquid contact area and water hanging droplet volume were observed by analyst of software in Origin.
Characterization of the structures. The structures of humps-on-strings were observed by scanning electron microscope (ESEM, Quanta FEG 250, EFI, America).

Investigations on the water drop hanging under different adhesion situations. The experiments
were carried on an OCAmicro40 meter system (Data-Physics, Germany) at room temperature. The fiber was carefully fixed on the sample frame. Numerous tiny water droplets were generated by an ultrasonic humidifier (YC-E350, Beijing YADU Science and Technology Co., Ltd.) as a fog flow and condensed on the fiber. The critical hanging-drops were recorded as drop hung on fiber with different orientations just the moment that the drops detached off via optical CCD camera in OCAmicro40 meter system.