Fouling Inhibition by Replenishable Plastrons on Microstructured, Superhydrophobic Carbon‐Silicone Composite Coatings

Superhydrophobic surfaces are known to resist diatom and bacteria adhesion if stable air layers are formed underwater (known as a plastron). However, most preparation techniques to obtain superhydrophobic surfaces need sophisticated chemical treatments and/or complicated chemical procedures. Here a 3D printing technique is used to create different molds for polymer casting. A conductive graphite‐carbon black‐silicone composite mixture is developed to fabricate different polymer surface casts from these molds. The obtained surfaces exhibited contact angles >145°, which led to a plastron formation on the surfaces underwater, and areas with intact plastrons protected the samples from diatom attachment. Due to the conductivity of the coatings, it is possible to replenish the plastrons by heating the surfaces.


Introduction
Biofouling, known as the undesired accumulation of biological matter [1] negatively affects applications in the marine industry, [2] medical implants, [3] water treatment technologies, [4] and food processing units. [5]Membrane fouling leads to a decrease in water purification as well as high costs for cleaning. [6]Biomass deposition on ship hulls leads to higher fuel consumption due to an increased drag. [7]All these challenges desire solutions, such as powerful antifouling coatings.With the first report on the lotus effect in 1997 by Barthlott and Neinhuis [8] superhydrophobic surfaces, defined by a contact angle larger than 150° [ 9] have gained substantial attention.The phenomenon of superhydrophobicity is well-known in nature and has been studied by many researchers.Duck feathers, [10] lotus leaves, [8] strider legs, [11] and butterfly wings [12] are common examples.Especially the lotus effect and its self-cleaning feature are well known.The DOI: 10.1002/admi.202300964almost perfectly formed spheres of water can slide easily on the hierarchically structured lotus leaves. [8][15] This difference can be explained by the fact, that the microstructures of the rose petal exhibit higher pitch values resulting in a partially wetted nanostructure. [14]While Young's equation can be used for the calculation of the water contact angle (WCA) on homogenous flat surfaces, two other models have been developed to describe the wetting of a rough surface, the Wenzel state and the Cassie-Baxter state.A liquid droplet is in the Wenzel state when the liquid touches the base of the bulges.A droplet that rests on top of the rough surface features forms air pockets underneath, which is termed the Cassie-Baxter state. [16]uperhydrophobic surfaces can be used for different applications, such as anti-icing, [17] self-cleaning, [18] anti-corrosion, water-oil separation, water collection, drag reduction, and liquid transport, [19] as they feature the ability to stay dry, clean, or icefree. [20]Furthermore, their ability to reduce fouling [21][22][23] as well as to reduce drag [24,25] due to the formation of an air layer makes them great candidates for antifouling coatings. [26]Reduction of biofouling through superhydrophobic surfaces was reported and the reduction in settlement of marine organisms directly correlated with the plastron lifetime underwater. [26]Surfaces exhibiting a stable plastron reduced the adhesion of organisms to a greater extent compared to surfaces with plastrons of limited longevity.In particular, the surface roughness was found to be important for air stabilization as the adhesion of E.coli was suppressed due to the presence of an air layer. [27]While superhydrophobic surfaces underwater were frequently reported to reduce the attachment of bacteria, a depletion of the air layer could result in high fouling as the rough surfaces provide a high surface area that can be contacted by the adhesives of the fouling organisms. [28]Thus, the stabilization of plastrons underwater is essential for their performance.Several different methods have been reported, such as the usage of electrolysis for in-situ gas formation to regenerate the plastron, [29] or the usage of local, gentle heating by only 2-3 °C of the surfaces underwater. [22]The latter results in a temperature increase of the water near the surface, leading to a lowering of the air solubility in water and thus the nucleation of the air at the water-air (plastron) interface.The heating can be achieved through the usage of a heating pad in contact with the surface.Once the plastron on the heated surface was stabilized, it showed better antifouling performance over time than the same surface without heating and a slowly decaying plastron.This method of heat-induced plastron regeneration was also applied to superhydrophobic candle soot coatings with inherently stable plastrons. [23]For conductive coatings, plastron regeneration can even be achieved by Joule heating, i.e. running a heating current across the conductive coating. [30]The temperature increase leads as described above to a growth of the plastron over time.
[15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31] Rough surfaces can be produced by different methods like chemical etching, plasma etching, laser ablation, sol-gel processes, and recently 3D printing. [32]While most of these methods require chemical treatments and complicated preparation protocols, 3D printing is a cost-effective, rapid, and easy method to produce superhydrophobic surfaces. [32,33]Especially fused deposition modeling (FDM) as the most frequently used 3D printing technique offers easily accessible and ecofriendly materials. [34]Furthermore, a safe, simple, and inexpensive fabrication [35] is possible with fast printing speed, [36] making it a great method for designing molds to produce rough and textured surfaces.While for several applications, the layers created during the printing process and thus the roughness of the printed objects might be a disadvantage, it is perfectly suited to produce superhydrophobic surfaces by casting a PDMS surface from a 3D printed mold. [32]Different WCAs were measured on polymer slabs cast from molds printed at different angles. [32]Those surfaces showed interesting surface properties but haven't yet been tested for their antifouling properties.
In this study, we fabricated different 3D-printed molds using FDM to challenge the resulting surfaces against diatom adhesion.Using a graphite-carbon black silicone composite coating, polymer films with a superhydrophobic surface were obtained by casting on the different molds.The obtained surfaces were characterized and tested against attachment of the diatom Navicula perminuta, to evaluate their antifouling potential.Furthermore, the possibility of restoring plastrons by Joule heating to increase the longevity of the coatings underwater was tested.

3D Printed Molds
Using computer-aided design software (Autodesk Inventor) three different molds were designed.The 3D model of the hierarchical structure is shown exemplarily in Figure S1 (Supporting Information).These molds were 3D printed using a polylactic acid (PLA) filament with a diameter of 1.75 mm.For printing, an Anycubic i3 mega was used, with a printing speed of 60 mm s −1 , a temperature of 200 °C, a layer thickness of 0.2 mm, and a print fill of 15%.Mold A was printed at a 90°angle while mold B and C were printed flat on the printing table.Table 1 shows the structures with schematical descriptions.

Surface Preparation
For the polymer surface, a thin layer of Sylgard 184 (DOW, Germany) (prepolymer: curing agent = 10: 1) was filled into the mold and cured for 2 h at 65 °C.Next, 1.5 g graphite and 0.5 g carbon black was mixed with 5 g Sylgard 184 (prepolymer) at 1500 rpm using a speed mixer (Hauschild Engineering; DAC 150.1 FVZ) twice for 20 s to obtain the silicone composite with reinforced mechanical properties [37,38] as well as conductive properties.The viscous mixtures were diluted with 8 ml n-hexane and finally, 0.5 g of the Sylgard 184 curing agent was added.The mixture was spread into the mold onto the already cured pure Sylgard layer.
The surface was allowed to cure overnight at 65 °C.The so-casted surface was carefully removed from the mold.For Joule heating experiments, the surfaces were prepared as follows: A small layer of Sylgard 184 was prepared as before.Next, a small amount of the silicone composite was spread at the edges of the mold and was allowed to settle for 5 min.Afterward, two copper tapes were placed on the silicone composite, and the mold was filled with silicone composite, integrating the copper tape into the surface.The so-prepared surfaces were allowed to cure overnight at 65 °C.Afterward, a small layer of Sylgard was spread onto the cured surfaces, letting it cure for another 2 h at 65 °C.Finally, the surfaces were carefully removed from the molds.The parts of the copper tape, not incorporated into the surface were isolated using nail polish.

Surface Characterization
For surface characterization, contact angle measurements as well as scanning electron microscopy (SEM) were used.The WCA was measured using a custom-built goniometer.Using a charge-coupled device camera, a picture of the droplet was taken, and the WCA was measured using Image-J.Contact angle hysteresis (CAH) was measured at an angle of 40°using a 30 μl droplet of tridestilled water.Images of the surface topography as well as diatom settlement were obtained using SEM(10 kV, 4.5 spot size, Quanta 3D FEG, FEI Company, Hillsboro, OR, USA).Using sputter coating deposition (Cressington Scientific Instruments Ltd, Watford, UK), the samples were coated with a ≈10 nm thick layer of gold to avoid charging effects during the measurement.

Pendulum Hardness
Characterization of the mechanical properties was conducted by measuring the pendulum hardness (byk, byko-swing König/Persoz).The König pendulum was used, and its oscillation time was calibrated using a glass slide (250 ± 10 s).

Chemical Stability
The chemical stability of the three different superhydrophobic coatings was determined gravimetrically by immersing them in Milli-Q water, seawater, and ethanol as well as in an acidic environment (pH 6) for 24 h.The samples were dried at 65 °C for 30 min and their weight determined gravimetrically.The remaining mass fraction (RMF) of the coatings was calculated by dividing the weight after immersion (m t ) by the coating weight before immersion m 0 with the equation RMF = m t m 0 .

Plastron Characterization
Plastron stability on different 3D surfaces was tested by immersion of the surfaces in Milli-Q water overnight.Photographs of the samples were recorded at the beginning and after 24 h.

Resistance Measurements
Resistance of the conductive coatings was assessed by measuring the resistance along the longitudinal axis (l = 8 cm) of the surface with a multimeter (Voltcraft M-3850D).

Heating Measurements
The Joule heating rate was measured using a digital temperature sensor (Voltcraft 302 K J −1 ).The surfaces were connected to a power supply (Eventek KPS305D) and the temperature increase was measured in the middle of the surfaces in defined time slots.

Plastron Replenishment
Plastron replenishment experiments were performed as previously reported. [22]Therefore, the surfaces were connected to a power supply and immersed in Milli-Q water in a small container with an outlet.Temperate increase near the surface was measured using a digital temperature sensor (Voltcraft 302 K J −1 ).
Next to the small container, a larger water reservoir was used to supply cold, fresh water to the small container using a variable flow pump (Aquarius Universal 440, Oase Living-water, Germany).A series of photographs was acquired to visually analyze the growth of the plastron over time.

Dynamic Diatom Assay
To assess the colonization by diatoms, one of the structured surfaces was immersed in artificial seawater (ASW) and the plastron was removed by a high-pressure flow of ASW across the coating using a 5 ml syringe.A suspension of diatoms from a continuous culture was diluted to a cell concentration of 2 × 10 6 cells ml −1 in sterile ASW.Accumulation assays under dynamic conditions were performed as previously reported. [23]In short, the surfaces were placed into individual chambers onto an orbital shaker (neolab DOS-10L) under constant dynamic agitation (1 h, 65 rpm) with 34 ml of the diatom suspension in ASW.Afterward, the diatoms were fixed with glutaraldehyde solution (2.5% in Milli-Q water) for 10 min.The surfaces were then washed with Milli-Q water two times for 10 min and finally dried overnight in the dark.For each structure, two replicates were subjected to two individual bioassays, and the settlement was visually examined at several points on the surface using the scanning electron microscope.The attachment of cells was quantified using the program Image-J.For each structure, the individual algae were counted on four images and the average diatom density was calculated for each structure relative to the same structure without the plastron.Error bars are the standard errors (n = 4).

Manufacturing of the Coatings
PDMS-carbon composite coatings were successfully prepared by a 2-step fabrication using various 3D printed molds (Figure 1).First, a layer of pure PDMS was filled in the molds and cured for 2 h at 60 °C.In the second step, the carbon-PDMS composite mixture was added on top of the first layer, followed by a third layer of pure PDMS.After filling the molds with the composite material and the pure PDMS, the coatings were allowed to cure for 24 h at 60 °C and subsequently removed from the molds.Molds for three differently structured conductive coatings were designed.The casted surfaces were easily and without residue removed from the molds and exhibited the inverse structure of the different molds.All polymer slabs were flexible and could be bent back and forth without permanent alteration of the surface structures.

Surface Analysis
Figure 2 A1-C1 and A2-C2 show SEM images of the structures on the surfaces and A3-C3 photographs of the different surfaces after immersion underwater.A1 and A2 show SEM images of the striped surfaces.It can be noticed especially in A1 that the surface consists of alternating grooves with different depths and two deeper grooves (≈320 μm deep and 300 μm wide) are followed by two shallower grooves (≈80 μm deep and 15 μm wide).A3 shows a photograph of the surface underwater revealing that a plastron formation occurs preferentially within the deeper grooves.The second pattern is intersected striped surfaces (SEM image B1, B2) with shallower lines that form a wavy, ribbon-like structure, where the pitches are ≈250 μm distant from each other and are 150 μm high.The wavy structures are oriented perpendicular to the deeper trenches that intersect the structures with a width of 250 μm and a depth of 200 μm spaced 1.6 mm apart.After water immersion, plastron formation was again preferably observed in the deeper, intersecting stripes (B3).The third hierarchical surface (SEM images C1 and C2) shows "pyramids" on the surface with a height of ≈1.5 mm and a width of 2 mm.Furthermore, it can be noticed, that the pyramids consist of several layers and exhibit a rough surface structure.The image of the surfaces underwater (C3) reveals a plastron that covers nearly the entire surface.However, the top part of the tips of the pyramids is in contact with the surrounding water.While the hierarchical structure showed a plastron that was stable for at least 24 h, the striped and intersected striped plastron decayed over 24 h.Drying of the surfaces and subsequent re-immersion resulted in a reformation of the plastron.The mechanical properties of the different materials were investigated by measuring the pendulum hardness.
For the pure PDMS, the oscillation time was 39 1 s, whereas the graphite-carbon-black PDMS composite material revealed a similar oscillation time of 36 ± 4 s.Furthermore, the three superhydrophobic coatings exhibited a constant mass when being immersed in Milli-Q water, seawater, and in an acidic environment for 24 h.Even in ethanol, only a mass loss of less than 2% was observed for the coatings.
Figure 3A shows the sessile WCA determined on the different structures.The image of the water droplet on the surface reveals, that the droplet rests on the structures of the striped pattern and thus indicates the droplet to be in the Cassie-Baxter state.On the intersected striped pattern, the water droplet does not fill the deeper stripes but seems to rest only on the highest protrusions.The images of the hierarchical structures show that the water droplet fits into the space between two pyramids and gets pinned in between.Therefore, the calculation of the WCA is not as straightforward as for flat surfaces.WCAs measured under these sessile conditions were 145 ± 2°for the striped surface, 144 ± 2°for the intersected striped, and 151 ± 2°for the hierarchical structures.Thus, the WCAs on all structured coatings were larger than on the smooth composite material (110 ± 2°) and on smooth PDMS (93 ± 1°).As the size of the droplet determines which part of the structures was contacted, a second set of sessile droplet measurements was used with a droplet volume of 30 μl, which led in all cases to smaller contact angles.Using a large droplet, and a tilting angle of 40°, additionally the CAH of all surfaces was measured as a difference between the advancing and the receding contact angle (Figure 3B).The values of the CAH ranged between 42°and 73°.Tilting the surfaces parallel to the structures led to a smaller CAH for the striped surface while the CAH of the intersected striped and hierarchical structures was smaller compared to the CAH measured by tilting the surfaces against the structures.For the striped surfaces, the CAH was smaller parallel to the grooves (58.8 ± 2.8°compared to 70.4 ± 8.5°for the perpendicular case).The intersected striped surface revealed CAH of 73.5 ± 3.4°parallel to the structures and 66.5 ± 4.9°perpendicular to the structures.The lowest CAHs were obtained for the hierarchical structures and 59.7 ± 11.5°were measured in the direction of the shorter distance between the pyramids, where 42.4 ± 7.6°were measured in the direction along the diagonal.Flipping the surfaces upside down with a water droplet on top of the surface structures revealed a strong attachment of the droplet toward the surfaces, which is shown in Figure 3C, indicating a petal effect.

Dynamic Diatom Attachment Assay
Bioassays with the marine diatom N. perminuta were used to investigate the influence of the surface-bound plastron on the diatom attachment on the coatings under dynamic conditions.Figure 4 shows the diatom density of the structured surfaces with intact plastron (P) relative to the same structure without the plastron.Each coating was exposed to a total of two independent bioassays and the settlement was examined at several points using SEM.A significant decrease (p < 0.05) in settlement was observed for all coatings when the plastron was present on the surface as compared to the surfaces with decayed plastrons.On the hierarchically-shaped surfaces, the plastrons exhibited the strongest influence with a reduction in diatom attachment of 84% compared to the surfaces without the protecting plastron.Compared to this, the striped surface showed a decrease of about 74 % in settlement and for the intersected striped surface, colonization decreased by about 58% compared to the same surface structures without plastron.The differences in coverage and the preferred attachment sites can be seen in the SEM images in A1-C4.A1/2-C1/2 show the surfaces which were tested with a completely decayed plastron and diatoms were observed across the entire surface.A3/4-C3/4 shows the surfaces tested with an intact plastron.It can be seen that no diatoms attached to the parts of the surface, where the plastron was present, which was the case mostly for recessed parts closer to the substrate.

Plastron Regeneration Through Joule Heating
Due to the conductivity of the carbon-silicone composite coatings, a current can be applied that causes a Joule heating effect on the coatings.The temperature increase when heating at different applied voltages was measured in air (Figure 5A-C).The coatings had a different resistance across the entire slide and the striped coatings revealed 430 Ω, the intersected stripes 215 Ω, and the hierarchically structured coatings 317 Ω.A larger increase in temperature was observed for higher voltages over the same time, especially for the striped and the intersected striped coating.Therefore, the application of 30 V resulted in a temperature difference for the striped and the intersected striped coating of 23.2 and 19.1°C, respectively.In contrast, a lower temperature increase was detected for the hierarchically structured coating at all applied voltages.After 5 min, the application of 30 V increased the temperature at the surface only by 6.7°C.As reported previously [22,30] the increase of the temperature of water close to a heated surface reduces the solubility of air, which leads to a growth and thus a replenishment of the plastron.As shown in Figure 5D-F, the temperature increase underwater is only 1-3°C, and thus much lower than in air, but nonetheless, the plastron grew on the various surfaces over a time span of 2-3 h.Immediately after immersion (0 h) all surfaces exhibited parts with noticeable reflecting areas, indicating the presence of a microplastron.After 1 h of gentle Joule heating, small bubbles started forming on the striped and on the intersected striped surface, which extended in size and number over the next hours.For the intersected striped surface, the bubbles only formed on the stripes, where the plastron was initially present.Plastron growth on the hierarchically structured surface was first observed after 2 h (Figure 5F).After that time, a large bubble formed on the surface, which further enlarged after another hour of incubation.

Discussion
In this work, three different structured silicone-carbon composite surfaces were synthesized using 3D-printed molds.While for most coating applications the mechanical durability of graphene and carbon-containing composites is the key property, this work exploited the conductivity of the materials and applied it for the regeneration of plastrons on structured coatings.A striped pattern, with alternating stripes with different heights, an intersected striped pattern, with stripes on the surface and opposing hills, and a hierarchical structure with rough pyramids were prepared.While all surfaces consisted of the same base material, they showed a different morphology and amount of stabilized air when immersed underwater.The hierarchical structure showed a nearly closed air layer on the surface, which was stable for at least 24 h.Only the tips of the pyramids protruded out of the plastron and showed a wetted surface.For both different striped patterns, an air layer was only observed for the deeper stripes, which however was not stable and decayed over 24 h.Drying of the surfaces and re-immersion resulted in a reformation of the originally observed plastron.Pendulum hardness measurements revealed that the incorporation of graphite and carbon black did not influence the mechanical properties as the oscillation time for the pure PDMS and the carbon-silicone composite material was less than 10% and within the standard deviation.Furthermore, all three coatings exhibited outstanding stability in Milli-Q water, seawater, and ethanol as well as in an acidic environment after a 24 h incubation.WCAs showed superhydrophobic wetting behaviors only for the hierarchical structure with a WCA of 151°.The intersected striped pattern showed a WCA of 144°and reflected a combination of the large structures and the roughness of the composite polymer since the smooth composite film revealed less hydrophobic sessile WCAs of 110°.While the wettability of the base composite material was in part responsible for the plastron stability, the surface structure and the resulting morphology also strongly affected the plastron lifetime.While the pyramids showed a plastron "carpet" in between the pyramids, the surfaces with lines only showed plastron formation within the deeper stripes on the surface.The difference in plastron stability might be due to the fact, that while small defects of the plastron on the pyramid structure might be compensated through the connected air layer, the reduced stability of the plastron of the two other structures might be due to the presence of lines on the surface.Here, a defect of the air layer might induce a chain reaction leading to the complete decay of the plastron within a line.In previous reports, superhydrophobicity was mainly achieved on pyramidal structures that have a nano-rough surface structure. [39]The used concept of hierarchically structured hydrophobic materials that contain a combination of larger structures with nanorough interfaces is frequently observed in nature.Here, Salvinia molesta leaves are a typical example of hierarchically structured surfaces which in addition have a unique combination of a superhydrophobic surface with smaller hydrophilic patches and show a perfect design for long-term plastron stabilization.The stabilization of the air-water interface against oscillation was attributed to the pinning of the air to the interface due to the additional hydrophilic patches. [40]This increased the activation energy needed for air bubble formation causing the increased longevity of the air layer. [40]Though we did not use hydrophilic patches in our work, the inhomogeneous morphology of the plastron on the pyramidal structures could be the reason why the longevity is enhanced compared to the smooth and homogeneous plastron in the grooves of the striped surfaces.The wetted tips of the pyramids on the structured surfaces could have a similar effect as the hydrophilic patches in Salvinia and contribute to the stabilization of the air layer as coalescence is retarded.It has also been reported in the literature that a reproduction of the structures of Salvinia by an immersed surface accumulation-based 3D printing process resulted in superhydrophobic surfaces that showed a petal effect and great potential for energy-efficient oil/water separation. [15]This petal effect was also observed for the pyramidal structures synthesized in this work.Turning the surface upside down showed the pinning of the droplet to the surface.A simple oil/water separation test, where an oil/water mixture was deposited onto the surface and the water droplet was re-collected leaving an oil layer stabilized inside the structures behind (Figure S2, Supporting Information).
Navicula perminuta assays showed a major reduction in diatom attachment on the surfaces with an intact plastron.A decrease of 87% for the pyramid structure was observed.The striped pattern showed a decrease of 70% and the intersected striped pattern a decrease of 60% compared to the same patterns without an air layer on the surface.The number of diatoms on the surfaces was reduced with increasing the area of the surface covered by the plastron.For the pyramid structure, the plastron covered nearly the entire surface, with only the tips of the pyramids protruding from the plastron layer, to which diatoms were capable of attaching.In comparison, for the striped and intersected striped surface, plastron formation was only observed in the macroscopic, wider grooves, so that the plastron did not extend continuously across the surface but was repeatedly interrupted by wetted areas.Thus, due to a smaller area covered by the plastron, these two surfaces did not reduce diatom attachment as efficiently as it was observed for the pyramid-shaped coatings.In previous studies, it was found that a uniformly structured superhydrophobic surface exhibited the lowest bacterial adhesion compared to controls when incubated with a bacterial suspension for one hour. [28]This was attributed to the effect of the surface-bound plastron.However, the plastron completely disappeared from the surface after about 150 min.In this study, significantly reduced colonization of the superhydrophobic coatings by the marine diatom N. perminuta was observed when the plastron was still intact.On the composite-silicone structures used in this work, a much longer plastron lifetime was observed for all three surfaces, with the pyramid-shaped coating showing the highest longevity of more than 24 h.A long plastron lifetime was also observed in literature, where superhydrophobic candle soot coatings exhibited a stable plastron for month which protected the surface against the adhesion of N. perminuta. [23]Furthermore, a superhydrophobic surface (PDMS with silica nanoparticles) was tested in previous studies against the adhesion of N. perminuta.The surface with a plastron on the surface showed no diatom adhesion.A depletion of the plastron within 10-15 h resulted in the loss of the protective layer and consequently a high diatom adhesion on the surface. [22]mong the different approaches to counteract limited plastron longevity is the thermoregeneration of the air layers. [22,30]hile initial studies used a heating pad, to initiate the temperature increase near the surface to initiate the growth of the plastron, conductive coatings allow to use of direct Joule heating, to increase their temperature.This resulted in lower energy consumption and enhanced the versatility of the applications.Among the challenges for this approach are electrolytic reactions near the power connections and the failure of electrical underwater connections.In the current work, the connections were molded into the surfaces, resulting in better connection and easier applicability of the method.Even though all structured surfaces showed a growth of the plastron over time with a current applied to the surface, differences in the plastron could be observed.While the striped structure showed small bubbles all over the surface, the intersected striped surface showed bubbles only on the parts of the surface, where the plastron was initially present.On the hierarchically structured surface, bubbles seemed to be slightly retarded in comparison to the other structures.The heating experiments in air showed a smaller temperature increase of the hierarchically structured surfaces compared to the striped and intersected striped surfaces, which might suggest that such surfaces exhibit a poorer heat distribution over the surface which might be due to its larger and hierarchically shaped structural features.Since the plastron was even more stable on this surface and a larger area was covered by the plastron compared to the other surfaces, it still makes its use appealing for various applications such as anti-biofouling or anti-icing.
While for the antifouling performance of the striped patterns, the alternating wetted and dry parts underwater could be optimized in the future for improved antifouling, this feature might be advantageous for other applications.Coexisting wetting and nonwetting regions, such as hydrophilic and hydrophobic ones, are known to be useful tools for desalination, water collection, and drug delivery. [41]Hydrophilic/hydrophobic surfaces for water harvesting using 3D printing were designed similarly to the structure of the back of the desert beetle and hydrophilic squaresized regions on a structured hydrophobic guide enabled a controlled rolling-off motion that significantly increased the efficiency of water collection. [41]The striped pattern in our work might as well be a good surface to control the direction of the droplet movement.While a uniform superhydrophobic surface will collect droplets that maintain their shape randomly roll across the surface, structured patterns allow directional movement and easier controlled water collection. [42]In order to use our surfaces for such application, the roll-off angle would have to be further reduced.

Conclusion
In summary, three different surfaces were prepared from the same material using conventional and oblique 3D printing with subsequent graphite-carbon black composite silicone molding.Due to the different structures on the surface, different plastron morphologies were formed.The surfaces showedWCAs in the range of 144°-151°.Diatom accumulation assays revealed a strong decrease of diatom attachment on the surfaces that decreased with the area covered by a protective plastron.Plastron growth and stabilization were successfully realized by direct Joule heating of the surfaces, where the voltage to be applied and the extent of growth depended on the different surface morphologies.The fabrication of structured coatings via 3D printed molds is a cost-effective, facile, and rapid method for creating superhydrophobic surfaces, which might not only be used for antifouling applications but also show promising features for in anti-icing, water collection, and oil-water separation technologies.

Figure 1 .
Figure 1.Schematic representation of the process to prepare the molds using 3D printing and the subsequent molding process using PDMS-carbon composite polymers.

Figure 2 .
Figure 2. SEM images of the differently structured surfaces (first and second column) and images of the surfaces underwater (third column).The reflection after water immersion was caused by the presence of an air layer on the surface.A1-3: Stripes; B1-3: Intersected stripes, C1-3: Hierarchical morphologies.

Figure 3 .
Figure 3. A) Sessile WCA goniometry on top of the structured composite surfaces.Static WCAs were measured using a 10 and 30 μl droplet.B) CAH on the different structured surfaces determined by tilting of the sample by 40°and measurement of the advancing and receding contact angle.CAH measured parallel to the structures (p) and with the structures (s) as shown in the illustration.C) water droplets on the three different structures resting on top of the surfaces (1-3) and as a hanging droplet (4-6).All CAs were determined using three independent droplets, error bars represent the standard (n = 3).

Figure 4 .
Figure 4. Dynamic diatom attachment assay on the structured graphite composite materials.A) Diatom densities on the differently structured surfaces with intact plastron (P) and without the protective plastron and dynamic attachment assay using the diatom N. perminuta.The statistical significance (p < 0.05) of the observed differences is represented by letters a-c.Bars labeled with different letters are statistically significantly different.A total of four replicates were investigated for each sample and the error bars are the standard error of the mean.B) SEM images of the coatings after the diatom assay.Surfaces A1/2-C1/2 represent the surfaces without a plastron present on the surface.Images A3/4-C3/4 show the attached diatoms and diatom-free areas on coatings with an intact plastron.

Figure 5 .
Figure 5. Plastron regeneration by Joule heating of the conductive graphite-carbon black composite coatings.A-C: Temperature increase of the Jouleheated coatings at different applied voltages.A) Striped coatings (31.3 V, 0.04 A), B) intersected striped coatings (31.0 V, 0.1 A), and C) hierarchically structured coatings (31.4 V, 0.09 A) in air.D-F) Plastron growth in an open flow system was determined over 2-3 h with various voltages applied, resulting in a temperature difference of 0.5-3 °C at the structured surfaces.

Table 1 .
Mold description with the schematical structure of the obtained surfaces.