Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Multiscale Superhydrophobic Zeolitic Imidazolate Framework Coating for Static and Dynamic Anti‐Icing Purposes

Multiscale Superhydrophobic Zeolitic Imidazolate Framework Coating for Static and Dynamic... IntroductionIce formation is a natural phenomenon resulting in the malfunction or breakdown of a variety of engineering systems. The development of superhydrophobic surfaces is one of the extensively used approaches for ice mitigation. Inspired by lotus leaves, superhydrophobic surfaces are referred to the surfaces having apparent contact angles (CA) higher than 150°. These bionic surfaces have a contact angle hysteresis (CAH) smaller than 10° or 5° according to various studies.[1–4] Superhydrophobic surfaces can be fabricated by the combination of a hierarchy of micro‐ and nano‐scale surface structures and modification of low surface energy chemistry.[5–7] Various techniques such as anodic oxidation, laser texturing,[7–9] and deposition[10] have been used to develop superhydrophobic surfaces. However, in addition to high costs and multiple‐step procedures, these methods usually require special devices and specific conditions.[3,11,12] Limited by complex fabrication procedures, a facile, low‐cost, and scalable coating method is of great importance.[13–16] In this regard, the development of superhydrophobic coatings containing nanoparticles can be a good substitution for existing conventional time‐consuming, and complex processes. For instance, titanium oxide (TiO2) and silicon oxide (SiO2) nanoparticles have been used to provide superhydrophobic coatings by producing a hierarchical surface structure as an addition to a low surface energy material.[9,17–20] Due to the inherent hydrophilicity of SiO2/TiO2 nanoparticles, polymeric materials such as polydimethylsiloxane (PDMS) and resins are generally required to be used as a matrix to create low surface energy in such coatings.[21–25] Uneven dispersion or aggregation of particles is one of the common problems with these composite coatings and needs to be dealt with sensitivity.[26] Moreover, contact angles reported for superhydrophobic SiO2/TiO2 nanoparticle based coatings are ≈150°, and complex processes are required to enhance their performance.[27–29]Recently, metal–organic frameworks (MOFs) have attracted much due to their significant properties, such as ordered porosity, high internal surface area, low density, and adjustable structure. Zeolitic imidazolate frameworks (ZIFs), a subclass of MOFs, are highly porous and crystalline materials having zeolite‐like structures, which are formed by divalent metal ions nodes and imidazolate ligands. Among this family, ZIF‐8 consists of tetrahedrally coordinated Zn+2 bridged by 2‐methylimidazolate linkers and exhibits sodalite (SOD) topology. ZIF‐8 is comprised of micropores located in the center, which comprises micropores located in the center accessible through six‐membered windows.[30,31] This distinctive porous and cage‐like structure of ZIF‐8 not only has an important role in the superhydrophobic behavior of the coating in terms of creating more air packets and superhydrophobicity but also causes higher durability against moisture penetration compared to conventional coatings based on SiO2 and TiO2 nanoparticles. Furthermore, ZIF‐8 is an intrinsically highly hydrophobic material, mainly due to the presence of the imidazolate rings in its structure.[32,33] This significantly contributes to the water repellency of ZIF‐8‐based coatings in terms of creating low surface energy. Therefore, ZIF‐8 nanoparticles can solely create significant superhydrophobic effects due to their extraordinary structure and nature.Moreover, excellent chemical, and thermal stability, good flexibility for structure and surface modification, adjustable particle size, modifiable pore size, and environmentally friendly synthesis process offer the opportunity to develop coatings with additional functionalities for critical applications, especially for anti‐icing applications.[34,35]Numerous studies have been conducted on static icing of superhydrophobic coatings.[36–38] These studies mainly focus on static icing conditions such as (sub)millimetric droplets[39] and film[40] icing and utilized simple hydrothermal analysis on the anti‐icing nature of the superhydrophobic coating.[41] A detailed analysis of the effect of surface chemistry and texture on mechanisms behind the superhydrophobicity of the coating such as Laplace pressure within the structure, non‐wetting mechanism, condensation‐induced icing mitigation approaches, and ice growth direction is necessary.[42–44] Most importantly, ice formation is dictated by impacting droplets on supercooled surfaces in many industrial applications.[45–47] At high surface supercooling temperatures, the droplet contact time is longer than the freezing timescale. The freezing‐governed hydrothermal interaction between the droplet and supercooled surface and the dynamics of the three‐phase contact line (droplet/surface/air) are still poorly understood at supercooled surface temperatures.[4,48–50] For instance, although advancing and receding contact angles do not vary with the contact line speed on non‐wetting surfaces followed by complete rebound under isothermal conditions,[51] three‐phase contact line dynamics and contact angle hysteresis on supercooled conditions require more analysis. The effective design of anti‐icing surfaces requires an understanding of transient thermal transport through the droplet/surface interface, which is governed by hydrothermal characteristics of the contact line during the impact process.In this study, we report a functional ZIF‐8‐based superhydrophobic multiscale coating (SHMC) with CA > 172°, rolling angle <5°, and CAH < 3° and applied it to metallic using the practical spray coating method. The non‐wetting mechanism of the coating was explained with a fractal theory‐based model of water contact angle. For static icing tests, the effect of SHMC on ice nucleation, ice growth mode, total surface icing time, and possible condensation‐induced freezing was investigated. In dynamic icing experiments, the three‐phase contact line characteristics including contact times, contact diameters, and interfacial heat transfer during the spreading and retraction stages of the impacting droplet on SHMC were covered. A numerical model was developed to further investigate the three‐phase contact line dynamics during the impact process to provide the transient temperature distribution within the droplet. An order of magnitude reduction in heat transfer rate during the total droplet contact time was obtained on the coating.The flow and heat transfer interaction between the droplet and supercooled surface and the contact line dynamics during the impact are still poorly understood. This study not only addresses the shortcomings of the current superhydrophobic surface preparation methods by proposing a MOF‐based multiscale coating but it also presents a fundamental analysis of static and dynamic icing using both computational and experimental methods. Compared to the available coatings which exhibit loss of superhydrophobicity after repetitive icing‐deicing cycles and droplet impact,[42,52] the SHMC maintained its water repellency performance for considerably longer cycle numbers and high impact velocities.Results and DiscussionCharacterization of ZIF‐8 NanoparticlesScanning electron microscope (SEM) images in Figure 1a show nanosized crystals of ZIF‐8. The images indicate the uniform size distribution of the nano spherical particles within the range of 50–100 nm. Figure 1b shows the X‐ray diffraction (XRD) pattern of the as‐synthesized ZIF‐8 nanoparticles which is in good agreement with the literature.[53,54] The presence of well‐defined peaks with high intensity is indicative of good crystallinity of ZIF‐8 nanoparticles. XRD pattern indicates characteristic diffractions at two thetas of 6°, 10.1°, 13.2°, 15.2°, 17°, 18.5°, 19.9°, 23°, 24.9°, 25°, 26°, 27° and 30° degrees corresponding to (110), (200), (211), (220), (310), (222), (321), (411), (332), (431), (440) and (334) planes of the crystalline structure of ZIF‐8, respectively. No peak associated with impurities was spotted. The Fourier transform infrared (FTIR) spectrum of ZIF‐8 nanoparticles is shown in Figure S1 (Supporting Information). The FTIR results are in agreement with the literature and confirm the structure of the nanoparticles along with XRD results.1Figurea) SEM image, b) XRD pattern, c) isotherm linear plot, and d) Horvath‐Kawazoe differential pore volume plot of ZIF‐8 nanoparticles.The N2 adsorption measurement results are shown in Figure 2c,d. The N2 adsorption isotherm in Figure 1b indicates type 1 adsorption behavior. The increase in the adsorbed volume at low pressures is linked with the existence of micropores, while the rapid increase at high pressures is associated with meso or macroporosity due to the stacking of nanoparticles. The micropore volume and Brunauer‐Emmett‐Teller (BET) surface area are obtained as 0.4641 cm3 g −1 and 1173 m2 g−1, respectively. The results are in good agreement with the reported values in the literature.[35,55]The distribution of micropore size of ZIF‐8 nanoparticles was measured using the Horvath‐Kawazoe (HK) method. As shown in Figure 1c, three narrow pore size distributions through the pore width range of 0.7–1.8 nm centered ≈0.75 nm, 1.125 nm, and 1.45 can be observed.2Figure(a‐d) SEM images of ZIF‐8‐based coating in different magnifications, e,f) appearance and sliding WCA values on the developed multiscale coating.Characterization of the Superhydrophobic ZIF‐8 Base CoatingThe morphology and surface texture of the superhydrophobic ZIF‐8‐based coating were characterized by the SEM technique (Figure 2). Observations in low magnifications (Figure 2a) are indicative of the uniform distribution of ZIF‐8 nanoparticles. Images at higher magnifications (Figure 2c,d) and in‐Lens detector images (Figure 2b) exhibit nanoscale cavities through coatings, which result from the packing of nanoparticles. SEM observations along the surface profile of the coatings confirm a micro/nano surface structure for the coating. On the other hand, BET results of ZIF‐8 nanoparticles indicate the existence of a subnano texture on particles, i.e., coating. Taking all these into account, it can be concluded that micro to sub‐nano topology could be formed by this approach.The water contact angles (WCA) were measured by an optical contact angle meter with 3 µL drops of distilled water at ambient temperature. The average value was determined by measuring the same sample at three different positions. The rolling angle (RA) was assessed by increasing the tilting angle with a 1° interval of 10s until the droplet started to roll from the surface. The contact angle hysteresis (CAH) was calculated by subtracting the receding water contact angle from the advancing water contact angle for each captured image at the rolling angle.Non‐Wetting Mechanisms of the Superhydrophobic ZIF‐8 Base CoatingThe superhydrophobic surfaces can be realized by using a combination of low surface energy material and surface texture. Generally, materials with nonpolar chemistries and closely packed stable atomic structures have low surface energy and exhibit water repellency. Polysiloxanes (−Si−O−Si− groups), fluorocarbons (CF2/CF3), nonpolar materials (with bulky CH2/CH3 groups), or polymers with combined chemistry are some examples of low surface energy materials. The methyl functionalized Im linkers as well as the coordinative saturation of the metal sites in ZIF‐8 have a significant contribution to its hydrophobic nature. On the other hand, silane modification of nanoparticles with 1H,1H,2H,2H perfluorooctyltriethoxysilane (PFOTES) results in superhydrophobic behavior of the coating. The partially perfluorinated silane having three hydrolyzable functional ethoxy groups bonds to ZIF‐8 nanoparticles and leads to a structure morphology with strongly bonded perfluoroalkyl functionalities. Low‐energy perfluorinated alkyl chains on a surface with a hierarchical texture generate a very high water repellency. Water molecules cannot form a hydrogen bond with the superhydrophobic ZIF‐8‐based coating and form hydrogen bonds with themselves in order to decrease the energy of the system. Therefore, the water molecules have very low interaction with the coating and cannot wet the surfaces.In terms of the surface structure, the classical Cassie‐Baxter (CB) theory assumes that the coating structure consists of two different materials: MOF particles with WCA of θs and a surface fraction area of φs, and air pockets with WCA of θv and a surface fraction area of 1‐φs.[56] The apparent CB WCA (θCB) on the composite interface is then expressed as:1cosθCB=φscosθS+(1−φs)cosθv=φscosθS+(φs−1)\[\begin{array}{*{20}{c}}{\cos {\theta _{{\rm{CB}}}} = {\varphi _s}\cos {\theta _{\rm{S}}} + \left( {1 - {\varphi _{\rm{s}}}} \right)\cos \theta v = {\varphi _s}\cos {\theta _{\rm{S}}} + \left( {{\varphi _s} - 1} \right)}\end{array}\]As can be seen in Figure 3a, there is a difference between CB WCA values and those obtained experimentally. The difference is due to the CB model being proposed assuming the morphology of single‐scale roughness over the surface. Considering the multiscale nature of the proposed coating, the WCA of the coated surface can be modeled using the Fractal theory.[13,56] The Fractal theory was introduced to evaluate complex geometries, in which a fractal geometry can show excellent superhydrophobicity by mimicking nature.[2] The Koch curve is one of the most used fractal geometries and is constructed by a recursive procedure (Figure 3b). Starting from a straight line, each geometry is made from the preceding geometry by splitting each line into three identical fragments. The fractal structure becomes more complex with the number of series and micro‐ and nanostructures over a lotus surface.[57] Therefore, Equation (1) can be used after introducing a roughness factor (L/l)D‐2:2cosθFr=φs(Ll)D−2cosθS−φv\[\begin{array}{*{20}{c}}{\cos {\theta _{{\rm{Fr}}}} = {\varphi _{\rm{s}}}{{\left( {\frac{L}{l}} \right)}^{D - 2}}\cos {\theta _{\rm{S}}} - {\varphi _{\rm{v}}}}\end{array}\]Here, D is the Hausdorff dimension, and L and l are the upper and lower limit scales of the fractal structure surface, respectively. Figure 3c shows the Atomic Force Microscopy (AFM) results taken from the sample surface area of 1 × 1 µm2 (top) and 20 × 20 µm2 (bottom). Multiscale randomness and self‐similarity features of the surface texture imply that the coating has fractal nature (as discussed in Section 3.2 and shown in Figure 3c,d), and fractal theory can be utilized to analyze the wetting mechanism.[58] More information about the fractal analysis and contact angle calculation is provided in the Supporting Information S2 Section. Based on the fractal analysis, θf of the nanostructure in the fractal wetting model can be directly calculated by extracting the morphology of the multiscale coating. The calculated CB (θCB), fractal (θFr), intrinsic (θs), and experimental (θexp) contact angles are shown in Figure 3a. As seen, a good agreement between the obtained results and experimental ones is achieved.3Figurea) Comparison between the predicted static WCAs by the classical Cassie‐Baxter and fractal methods and experimental results on different samples, b) an example of random fractal Koch structure generated using the MATLAB software with seven iterations, c) AFM results of the surface area of 1 × 1 µm2 (top) and 20 × 20 µm2 (bottom), d) surface profile of the copper coated surface.Phase Change DynamicsFreezing experiments were performed on the Superhydrophobic Multiscale Coating (SHMC) and plain surfaces to investigate the effect of surface energy and surface texture. According to the icing behavior observations, ice formation and accumulation on SHMC samples require a larger supercooling, and ice formation starts at a much lower supercooling temperature on plain surfaces. Equations (3) and (4) express the critical Gibbs Free Energy (nucleation barrier) for homogeneous and heterogeneous nucleation.[59]3ΔGHom∗=16πσ33ΔGv2\[\begin{array}{*{20}{c}}{\Delta G_{{\rm{Hom}}}^ * = \frac{{16\pi {\sigma ^3}}}{{3\Delta G_v^2}}}\end{array}\]4ΔGHet∗=f(θ,R)ΔGHom∗\[\begin{array}{*{20}{c}}{\Delta G_{{\rm{Het}}}^ * = f\left( {\theta ,R} \right)\Delta G_{{\rm{Hom}}}^ * }\end{array}\]Here, f(θ,R) is the shape function that indicates the deviation of the heterogeneous nucleation process from the homogeneous nucleation process. Figure 4a shows the variation in shape function with apparent contact angle and fractal dimensions for hydrophilic and hydrophobic fractal surfaces.[60] As seen, the shape function increases with the wettability of the fractal surface. According to the Classical Nucleation Theory, ice nuclei need to overcome a larger energy barrier (Critical Gibbs Free Energy) to become stable, grow and continue to decrease the energy of the system on the coated surface compared to a plain sample. The plain samples have a higher surface energy and act as a preferable surface for ice nucleation and facilitate freezing.4Figurea) Plot of shape function for different hydrophilic and hydrophobic fractal surfaces (redrawn from,[60] b) surface freezing time on SHMC and bare samples on different surfaces and under ambient conditions, c) Icing activity on SHMC and plain area (the coating at the edges of the sample was cleaned to show the anti‐icing effectiveness of the coated area), d) schematic and microscopic images of coalescence induced droplet jumping on SHMC, e) icing and icing mode on the bare aluminum surface, f) icing and icing mode on the coated aluminum surfaceThe total surface icing times on the SHMC and bare surfaces under different conditions are shown in Figure 4b. The results show that SHMC considerably prolongs the icing time under different surface and ambient conditions. This implies that the critical radius of stable nuclei may be larger on the coating compared to plain surfaces in heterogeneous nucleation, which means that the embryos need to be larger to be stabilized (more molecules aggregate), which requires more time. On the other hand, the rate of ice nucleation can be calculated as:5J=KAint(θ,t)exp(−ΔGc(θ,R)kT)\[\begin{array}{*{20}{c}}{J = K{A_{{\mathop{\rm int}} }}\left( {\theta ,t} \right)\exp \left( { - \frac{{\Delta {G_c}\left( {\theta ,R} \right)}}{{kT}}} \right)}\end{array}\]where K, k, Aint are the kinetic constant, Boltzmann's constant, and the geometric substrate‐liquid “apparent” contact area, respectively. From Equation (5), it is apparent that SHMC increases the critical Gibbs Free Energy of heterogeneous nucleation and decreases the nucleation rate (Figure 4c). The freezing delay is not only related to the higher ice nuclear barrier but high WCA, low CAH, and lower solid‐liquid contact area of the condensate droplets also contribute to icing hindrance on the coated surfaces. Dropwise condensation due to high static WCA and enhanced mobility because of low CAH on SHMC remarkably reduce the risk of condensation freezing. The low solid‐liquid contact area limits the heat transfer from the coating to droplets and extends the icing time. Coalescence induced droplet jumping is the other parameter contributing to the prevention of icing on SHMC (Figure 4d).[61] The jumping of a droplet is a process where condensed water jumps from a surface as the excess surface energy is converted into upward kinetic energy when condensed water droplets merge.Ice growth is thermodynamically driven by the minimization of the Gibbs surface energy, which is ΔG = ΔH − TΔS. The ice growth mode on the SuperHydrophobic Multiscale Coating is majorly different from that of the plain surface. Figure 4e shows the ice growth mode on the bare surface. The ice growth is along the surface, and ice continues to grow in the same focal plane as the substrate. As time passes, ice layers grow and form a thick ice layer. The superhydrophobic nature of the coating imposes a large positive value of TΔS, which overcomes a small positive value of ΔH. The entropic contribution to the Gibbs energy, TΔS, dominates over the enthalpic contribution, ΔH, making it more energetically feasible for the ice crystal to grow off the surface rather than along the surface. As a result, ice growth on SHMC undergoes an off‐surface growth mode, as demonstrated in Figure 4f.[62] Since the contact area of the ice and SHMC is much lower than the bare surface, the ice layer thickness is also lighter compared to the plain samples.The trapped air within SHMC cavities induces a capillary force on the droplet. The stability of these air pockets is critical for the durability of the superhydrophobic coating and for preventing CB to Wenzel transition during condensation as well as repeating icing/deicing cycles. The droplet can penetrate into i) the passages formed due to superimposed MOF nanoparticles and ii) the nanoparticle pores. As shown in Figure 5a, liquid penetration depends on the droplet size (Rd), passage or pore size (d), and droplet and meniscus angles (θd and θm). The net force acting on a droplet on a hydrophobic pore or passage is given as:[63]6Fnet=πσd22Rd+ρgRd(1−cosθd)πd24+πσdcos(θm)\[\begin{array}{*{20}{c}}{{F_{{\rm{net}}}} = \frac{{\pi \sigma {d^2}}}{{2{R_{\rm{d}}}}} + \rho g{R_{\rm{d}}}\left( {1 - \cos {\theta _{\rm{d}}}} \right)\frac{{\pi {d^2}}}{4} + \pi \sigma d\cos \left( {{\theta _{\rm{m}}}} \right)}\end{array}\]Here, σ and ρ are the surface tension (0.075N m−1) and density of water (998 kg m−3) at zero degrees, and g is the gravitational acceleration (9.81 m s−2). The first, second, and third terms on the right‐hand side equation are the Laplace force acting on a drop, gravitational force, and capillary force by a meniscus in a passage, respectively. When a surface is hydrophobic (θd, θm > 90°), the meniscus exerts a force in the upward direction, preventing water penetration. However, when a drop is small enough to make Laplace pressure larger than the capillary pressure, drop penetration can take place due to the positive net force in the downward direction.5Figurea) Schematic and SEM image of the passages formed by MOF nanoparticle superposition and the mechanisms responsible for coating stability against droplets, b) Net forces acting on droplets at different pore/passage and droplet sizes (Here the green region indicates the area corresponding to the calculated maximum passage sizes on SHMC surfaces), c) water contact angle variation with icing/deicing cycles at different surface temperatures.Figure 5b shows the calculated net forces for different droplet and passage/pore sizes. The green area shows the region corresponding to the calculated maximum passage size formed by MOF nanoparticle superimposition. Passages with sizes <1 µm provide a stable condition by preventing the penetration of droplets with radii >100 nm. The liquid penetration within the MOF nanoparticle pores (d < 10 nm) also requires droplet radii <10 nm. The microstructures prevent the droplet from touching the valleys, while hydrophobic nanoparticles pin liquid droplets and thus prevent liquid from filling the valleys between asperities (Figure 5a). The dense microscale structure, hydrophobic nature, high surface area, and nanoporous structure of ZIF8 nanoparticles provide air pocket stability for a higher number of icing/deicing cycles. As can be seen in Figure 5c, the static WCA decreases with the number of cycles. Although a 3% decrease in WCA is observed on tested specimens at Ts = −5 °C, a further decrease in the surface temperature results in hindering the superhydrophobicity of the coating (6% decrease in WCA at Ts = −15 °C).Dynamic Water/Ice Repellency and Contact Line BehaviorAn impacting droplet experiences three stages: spreading, retraction, and bouncing. Figure 6a shows the obtained numerical results for an impacting droplet for different surfaces and under impact conditions. Detailed information about the numerical analysis is provided in Supporting information S3. The droplet dynamics during these stages strongly depend on the relative magnitude of inertia, surface tension, and viscous forces as well as the surface wetting state. Weber (We = ρV2D0/σ), Reynolds (Re = ρVD0/µ), Capillary (Ca = We/Re = µV/σ), and Ohnesorge (Oh = µ/(ρσD0)0.5) numbers are the non‐dimensional numbers used to characterize the impact process, where We, Re, Ca, and Oh numbers stand for the relative magnitude of inertia to surface tension forces, the inertia to viscous forces, viscous to interfacial forces, and viscous to inertia and surface tension forces, respectively. The impact experiments were performed at different surface supercooling temperatures from −5 °C to −20 °C, impact velocities from 1 to 2.5 m s−1, and droplet diameters of 1, 2, and 3 mm. The early stages of droplet impact and spreading for a droplet with We = 80 on plain and coated surfaces at room temperature are illustrated in Figure 6b. The droplet edge upon spreading on the coated surfaces stays untouched by the substrate, resulting in a reduction in the contact area between the liquid and solid phases (Figure 6c). This reduces the viscous dissipation and contact line friction and assists the droplet to maintain its energy and to spread over much larger diameters with larger three‐phase contact line velocities. During the later phases of the spreading stage, fingering of the droplets is evident on coated surfaces even at Weber numbers as low as 25. The Rayleigh‐Taylor instability at the liquid/air interface is the main reason for the formation of fingers and break up at the spreading front of the decelerating interface during the later phases of spreading (Figure 6d).[64,65] Figure 6e shows the Dirac delta function (δ [1/m]) at the droplet/air interface on hydrophilic (HPhi) and superhydrophobic (SHPho) surfaces during the retraction stage. Here, the images are not at the same scale in the x and y directions to show a clear distribution of the Dirac delta function on the interface. The Dirac delta function represents the surface tension force (f = σκnδ) (more information is included in Supporting Information S4).[66] Larger δ values for the droplet on the superhydrophobic (SHPho) surface suggest a stronger interfacial surface tension force, which results in a larger pressure difference across the elongated droplet/air interface. The unbalanced pressure distribution in the droplet also causes the droplet to contract toward its center with larger retraction velocities on the SHPho surface. The retraction velocity has a direct effect on the contact time of a bouncing droplet (τ ∼ (ρD3/σ)1/2). During the later phases of the retraction stage, where the retraction velocity decreases to zero, the droplet bouncing is observed on coated surfaces (Figure 6f).6Figurea) Numerical results on the effect of surface wetting state (HPhi: Hydrophilic, SHPho: Superhydrophobic) and impact condition (We: Weber number defined as the ratio of inertia to surface tension forces); b) High‐speed images taken from the early stages of droplet impact on bare (uncoated) and SHMC aluminum surfaces; c) Droplet edge upon spreading on the bare (uncoated) and SHMC aluminum surfaces; d) The Dirac delta function at the droplet/air interface during the retraction stage; e) The pressure distribution within retracting droplet on superhydrophobic and hydrophilic surfaces; f) High‐speed images taken from the early later stages of droplet spreading on bare (uncoated) and SHMC aluminum surfaces.The performance of the SHMC surface for anti‐icing applications was further examined by investigating the impacting dynamics of droplets with diameters of 1, 2, and 4 mm at substrate temperatures of −5 °C, −10 °C, and −20 °C. The heat transfer between the supercooled surface and droplet is a function of contact time and can be characterized using the heat diffusion rate, defined as the ratio of heat transfer at the droplet/surface interface (qL×τ) over the total sensible and latent heat (q′V).[67] The heat transfer rate at the droplet/surface can be estimated as:[68]7q˙L=hA(Td−Ts)=hπDCL24ΔT=Nu.kwLπDCL24ΔT\[\begin{array}{*{20}{c}}{{{\dot{q}}_L} = hA\left( {{T_{\rm{d}}} - {T_{\rm{s}}}} \right) = h\frac{{\pi D_{CL}^2}}{4}\Delta T = \frac{{Nu.{k_{\rm{w}}}}}{L}\pi \frac{{D_{CL}^2}}{4}\Delta T}\end{array}\]8Nu=0.664Pr0.6Re0.8, Re=ρVCLDCL/μ\[Nu = 0.664{\Pr ^{0.6}}{{\mathop{\rm Re}\nolimits} ^{0.8}},\;{\mathop{\rm Re}\nolimits} = \rho {V_{{\rm{CL}}}}{D_{{\rm{CL}}}}{\rm{/}}\mu \]9q˙L=0.166Pr0.6Re0.8kwπDCLΔT\[\begin{array}{*{20}{c}}{{{\dot{q}}_L} = 0.166{{\Pr }^{0.6}}{{{\mathop{\rm Re}\nolimits} }^{0.8}}{k_{\rm{w}}}\pi {D_{{\rm{CL}}}}\Delta T}\end{array}\]where qL is the heat transfer rate [W], h is the convective heat transfer [W m−2 K−1], Nu is the Nusselt number [‐], kw is the thermal conductivity of water [W m−1 K−1], L is the characteristic length [m], A is the surface area between the droplet and substrate (πD2CL/4, [m2]), and Td and Ts are droplet and surface temperatures (K).The following equation can be used to estimate the total sensible and latent heat of the droplet:[67]10q′V=(c(Td−Ts)+hs) ρVdroplet, Vdroplet=16πD03\[q{\prime _V} = (c({T_{\rm{d}}} - {T_{\rm{s}}}) + {h_{\rm{s}}})\;\rho {V_{{\rm{droplet}}}},\;{V_{{\rm{droplet}}}} = \frac{1}{6}\pi D_0^3\]11q′V=(cΔT+hs)ρπ6D03\[\begin{array}{*{20}{c}}{{{q'}_V} = \left( {c\Delta T + {h_{\rm{s}}}} \right)\frac{{\rho \pi }}{6}D_0^3}\end{array}\]Here, q′V is the heat transfer rate [J], c is the specific heat [J kg−1 K−1], hs is the latent heat of solidification [J kg−1], ρ is the density [kg m−3], and Vdroplet is the volume of the droplet [m3].The heat diffusion rate is expressed as:12RQ=q˙L×τtotqV′=0.166Pr0.6Re0.8kwπDCLΔT×τtot(cΔT+hs)ρπD03/6=0.996Pr0.6kwρ0.2D03VCL0.8DCL1.8×(τspread+τretract)(c+hs/ΔT)\[\begin{array}{c}{R_Q} = \frac{{{{\dot{q}}_L} \times {\tau _{{\rm{tot}}}}}}{{q_V^\prime }} = \frac{{0.166{{\Pr }^{0.6}}{{{\mathop{\rm Re}\nolimits} }^{0.8}}{k_{\rm{w}}}\pi {D_{{\rm{CL}}}}\Delta T \times {\tau _{{\rm{tot}}}}}}{{\left( {c\Delta T + {h_s}} \right)\rho \pi D_0^3{\rm{/}}6}}\\ = \frac{{0.996{{\Pr }^{0.6}}{k_{\rm{w}}}}}{{{\rho ^{0.2}}D_0^3}}\frac{{V_{CL}^{0.8}D_{CL}^{1.8} \times \left( {{\tau _{{\rm{spread}}}} + {\tau _{{\rm{retract}}}}} \right)}}{{\left( {c + {h_s}{\rm{/}}\Delta T} \right)}}\end{array}\]Here, the contact line diameter (DCL), droplet and substrate temperature difference (ΔT), contact line velocity (VCL), and contact time (τtot = τspread+ τretract) dictate the heat transfer rate and RQ ratio. The spreading (τspread/τtot) and retraction (τspread/τtot) ratios at different surface temperatures and droplet conditions are shown in Figure 7a. Although spreading and retraction times increase with the surface supercooling temperature, prolongation of the total droplet contact time reduces the spreading time ratio and increases the τspread/τtot in the spreading stage. The sharp decrease in τspread/τtot with the surface temperature at We = 23 turns into a gradual reduction at We = 230 (as indicated by pink arrows), indicating the major effect of the temperature on regions where surface tension and capillary forces are dominant. The effect of surface temperature on the contact line velocity (VCL) and contact line diameter (DCL) on the SHMC surface is shown in Figure 7b for an impacting droplet with a diameter of 2 mm and We = 23. While the contact line velocity and diameter decrease during the spreading stage (red area), a prominent change in DCL and VCL at the retraction stage (blue area) is observed. The capillary numbers are much lower than unity in the retraction stage, indicating the domination of the surface tension over viscous force. Although the impacting droplet on the supercooled surface has a higher viscosity, a low droplet velocity results in an almost 40% reduction in droplet capillary number. The initial stages of the retraction stage on SHPho and HPhi surfaces were numerically analyzed to show the effect of surface wettability on the heat transfer mechanism and temperature distribution. As shown in Figure 7c, a larger DCL during the retraction stage on the HPhi surfaces results in enhanced convective heat transfer relative to SHPho surfaces. A sharp decrease in droplet temperature is evident, which results in contact line freezing and droplet icing on HPhi surfaces. Lower temperature and temperature gradient within the droplet on SHPho surface, especially at droplet edges, highlight the role of contact line diameter and droplet mobility in limiting the heat transfer between the droplet/surface contact area. The heat transfer rate on a superhydrophobic surface is almost one order of magnitude lower than that of a hydrophilic surface. Figure 7 compares the QR ratio on SHMC and plain samples for different surface temperatures. On a plain sample, the sensible/latent heat transfer is the dominant heat transfer mechanism, especially in the spreading stage. On the coated surfaces, while the temperature difference between the droplet and substrate is high during the spreading stage, the heat transfer is limited by low τspread and either DCL or VCL. The heat transfer rate in the retraction stage is also limited by a small contact area and low contact velocity.7Figurea) Spreading and retraction contact time ratios for different surface supercooling temperatures and Weber numbers; b) Transient contact line velocity and contact line diameter for a droplet impact with a diameter of 2 mm and We = 23; c) High‐speed images taken from the last stages of droplet bouncing from the coated surface at surface temperatures of 20 °C and −10 °C; d) numerical results of the droplet temperature distribution, temperature gradient, and heat transfer rate on superhydrophobic and hydrophilic surfaces; d) IR thermography images of an impact on bare and coated aluminum surfaces.The droplet dynamics upon bouncing from the coated surface are shown in Figure 7e at room temperature (20 °C) and surface temperatures of 10 and −10 °C for We = 80. The effect of a further decrease in surface supercooling on droplet dynamics and resultant heat transfer is negligible, while the surface supercooling gradually begins to have an effect as the inertia force decreases. Compared to bare surfaces, the proposed coating provides outstanding performance in terms of droplet mobility and reduction in the heat transfer rate. The IR thermography images of a droplet impact with We = 80 on the bare and SHMC surfaces shown in Figure 7e confirm the heat transfer deterioration during and after impact on the coated sample. Instant droplet deposition and freezing on a supercooled aluminum surface are evident. The non‐uniform temperature distribution on the coated surface is the result of droplet splashing, breakup, and bouncing, which decreases the droplet contact time and eliminates the risk of icing on the proposed coating. The gradual decrease in the surface temperature of the coated surface after impact (t > 60 ms) also indicates the liquid stage of the secondary droplets remained on the coated surface.ConclusionIn this study, we developed a MOF‐based functionalized superhydrophobic multiscale coating (SHMC) with an apparent CA value larger than 171°, rolling angle of <5°, and contact angle hysteresis of <3°. The developed coating was prepared using the practical spray coating method on aluminum and copper surfaces, two of the most used materials in the relevant industries. A comprehensive examination of the static and dynamic icing performance of the SHMC was made using microscopic and macroscopic analyses. The microscale peaks of the coating texture were achieved by NP superposition, MOF NPs formed nanoscale structures, and nanoparticle pores, and their effect on NP morphology led to subnano features of the structure. Multiscale, randomness, and self‐similarity features of the SHMC texture suggested the fractal nature of the coating, and a fractal theory‐based model of water contact angle was adapted to reveal the non‐wetting mechanism on SHMC. Different modes of icing were observed on plain (on‐surface) and coated surfaces (off‐surface), resulting in different ice morphology due to the presence of the coating on the substrate. The multiscale texture of the SHMC extended the icing time by at least 300% and maintained its superhydrophobicity for >30 icing/deicing cycles. The capillary pressure generated within the multiscale coating prevented the droplet from penetrating into the structure and reduced the risk of condensation‐induced freezing under different variant supercooling and relative humidity conditions and within continuous icing/deicing cycles. Compared to the plain sample, which exhibited instant icing at 60 ms after impact, no icing was observed on the SuperHydrophobic Multiscale Coating. The three‐phase contact line characteristics including the contact times, contact diameters, and interfacial heat transfer during the spreading and retraction stages of the impacting droplet on SHMC were assessed for different surfaces and ambient conditions. The high speed and IR thermography results proved that at least an order of magnitude reduction in heat transfer rate during the total droplet contact time could be obtained on the SuperHydrophobic Multiscale Coating.Experimental SectionMaterials2‐methylimidazole (Hmim, 99%), 1H,1H,2H,2H‐perfluorooctyltriethoxysilane (PFOTES, 97%) and zinc nitrate hexahydrate (Zn(NO3)2.6H2O,99% were purchased from Sigma Aldrich. The other chemicals such as absolute ethanol, n‐hexadecane, and acetone were purchased from Sigma Aldrich and were used without any further purification. Deionized water (DI water) purified by a water purification system was used for all the aqueous solution preparations. Copper (purity ≥ 99.5%) and Aluminum were used for the metallic substrates.Synthesis of ZIF‐8 NanoparticlesZIF‐8 nanoparticles were synthesized in an aqueous system at room temperature. 1.17 gr Zn(NO3)2.6H2O was dissolved in 8 ml DI water while 22.70 gr 2‐methylimidazole was dissolved in 80 ml DI water. The zinc nitrate solution was poured into the 2‐methylimidazole solution and was mixed with a magnetic stirrer to prepare the synthesis solution with the molar ratio of Zn2+ 2‐methylimidazole: H2O = 1:70:1238. After two hours of stirring for 2 hours at room temperature, the product was collected by centrifuge. The nanoparticles were then washed with DI water and dried in a vacuum oven at 85 °C for 24 h.Preparation of superhydrophobic ZIF‐8‐based CoatingIn order to develop a uniform and durable superhydrophobic coating, ZIF8 nanoparticles were first modified by 1H,1H,2H,2H‐perfluorooctyltriethoxysilane (PFOTES). 0.5 ml of 1H,1H,2H,2H‐perfluorooctyltriethoxysilane (PFOTES) was mixed ultrasonically in 50 ml absolute ethanol for 1 h. Consequently, 1.5 gr ZIF‐8 nanoparticles were dissolved ultrasonically in the as‐prepared solution for 30 mins to obtain the coating solution. Metallic Substrates were cut into 30 mm × 30 mm pieces and polished with sandpaper (up to 2000 grade). Consequently, the substrates were degreased by acetone and cleaned with DI water ultrasonically (for 10 min each). The coating solution was spray‐coated on the substrates using a spray gun. Air pressure of 0.1 MPa and spray distance of 20 cm were the spray coating parameters.Materials CharacterizationScanning electron microscopy (SEM) Zeiss Leo Supra 35VP equipped with energy dispersive X‐ray spectroscopy (EDX) was employed to characterize the morphology of the nanoparticles, surface structure of the coating, and elemental composition of the materials. X‐ray diffraction (XRD) analysis was done by recruiting a Bruker D2 Phaser (Bruker AXS GmbH) by CuKα radiation (wavelength of 1.54 A) to characterize the crystal structure of ZIF‐8 nanoparticles. Infrared spectra were recorded using a Nicolet iS50 FT‐IR (Fourier Transform Infrared Spectroscopy) spectrophotometer to confirm the structure of ZIF‐8 nanoparticles (FTIR spectra are provided in Supporting Information S1 Section). The specific surface area and pore volume of the crystals were measured using a Micromeritics gas adsorption analyzer instrument equipped with commercial software for calculation and analysis. The BET surface area was calculated from the adsorption isotherms using the standard Brunauer–Emmett–Teller (BET) equation.Icing TestsIcing experiments were performed in a lab‐made temperature‐ and humidity‐controlled environmental chamber at various relative humidity (RH) settings. Figure 8 shows a schematic of the experimental setup. More information about the icing analysis is provided in the Supporting Information S1 Section. Care was taken to preclude possible contamination such as that due to hydrocarbons. The sample was mounted using thermally conductive adhesive tape (McMaster Carr, 6838A11) on the cold plate of a Peltier cooler (Peltier Module TEG High Temperature). The subcooling of the sample surface was maintained by setting the Peltier cooler at different temperatures. A cooling loop was designed to reduce the temperature of the hot plate of the thermoelectric device for thermal stability in long icing experiments. The sample surfaces were monitored and measured simultaneously using an IR thermal camera (FLIR T1020). A standard DSLR camera (Canon EOS) was used to visualize the icing experiments from the top. The real‐time experiments were recorded from top and side using IR, hush speed, and Digital Single‐Lens Reflex (DSLR) camera. Image processing was performed using MATLAB software.8FigureSchematic of the experimental setup for static and dynamic icing. The green region shows the equipment required for dynamic icing tests and orange region shows the surface cooling apparatus.Droplet Impact TestsThe droplet impact experiments were performed using deionized water as the working fluid with droplet volumes of ≈0.5, 4.2, and 33.5 µl (diameters of 1, 2, and 4 mm). The corresponding apparatus was illustrated schematically in Figure 8b. Different needles (different gauge sizes) were connected to a syringe pump (LEGATO 200, KD Scientific, Holliston, MA, USA). Droplets detached due to gravity from needles mounted at heights between 5 and 32 cm, leading to impact velocities ranging from 1 to 2.5 m s−1. The impact dynamics of droplets were captured using a speed camera (Phantom v9.1 vision research high‐speed camera) operated at different frames per second with long‐distance lenses and a workstation with visualization software (Phantom PCC 3.7 software). A cold light source was used to backlight the impacting droplet on the target surface. Depending on the required frame per second (fps), the resolution of the recorded videos ranged from 960 × 240 (7648 fps) to 1632 × 1200 (1000 fps). The experiments were performed on a single droplet impact, where each experiment was repeated for at least five times.The measuring errors in this study mainly involve the surface temperature and impact conditions measurements. The surface temperatures were measured using a T‐type thermocouple (±1 °C) and IR Thermal Camera FLIR T1020 (±2 °C, with 1024 × 768 resolution). Uncertainty in the ImageJ image software analysis was 1 pixel (±0.01 mm). The impact velocity was calculated by examining the last 20 consecutive frames with a maximum uncertainty of ±0.03 m s−1. The maximum uncertainty in the impact Weber number was 7.58%, which was calculated using the following equation:[69]13δWeWe=(δD0D0)2+(2δu0u0)2\[\begin{array}{*{20}{c}}{\frac{{{\delta _{We}}}}{{We}} = \sqrt {{{\left( {\frac{{{\delta _{{D_0}}}}}{{{D_0}}}} \right)}^2} + {{\left( {2\frac{{{\delta _{{u_0}}}}}{{{u_0}}}} \right)}^2}} }\end{array}\]AcknowledgementsThe authors would like to thank Dr. Mohammad Sajad Sorayani Bafqi for his help in FTIR analysis. The authors also thank Sabanci University Faculty of Engineering and Natural Science (FENS) for graduate student support and appreciate the support from TÜBİTAK (The Scientific and Technological Research Council of Turkey) Support Program for Scientific and Technological Research Project Grant No. 120M659 and Turkish Academy of Sciences (TUBA).Conflict of InterestThe authors declare no conflict of interest.Author ContributionsThe manuscript was prepared through the contributions of all authors. All authors provided their approval to the final version of the manuscript.Data Availability StatementThe data that support the findings of this study are available from the corresponding author upon reasonable request.M. J. Kreder, J. Alvarenga, P. Kim, J. Aizenberg, Nat. Rev. Mater. 2016, 1, 15003.T. Onda, S. Shibuichi, N. Satoh, K. Tsujii, Langmuir 1996, 12, 2125.W. Barthlott, C. Neinhuis, Planta 1997, 202, 1.Q. Li, Z. Guo, J. Mater. Chem. A 2018, 6, 13549.X. Deng, L. Mammen, Y. Zhao, P. Lellig, K. Müllen, C. Li, H.‐J. Butt, D. Vollmer, Adv. Mater. 2011, 23, 2962.J.‐P. Lee, S. Choi, S. Park, Langmuir 2011, 27, 809.G. Kwak, M. Seol, Y. Tak, K. Yong, J. Phys. Chem. C 2009, 113, 12085.B. Qian, Z. Shen, Langmuir 2005, 21, 9007.Y. Lu, S. Sathasivam, J. Song, C. R. Crick, C. J. Carmalt, I. P. Parkin, Science 2015, 347, 1132.J. Song, W. Xu, Y. Lu, J. Mater. Sci. 2012, 47, 162.C. Yeom, Y. Kim, J Ind Eng Chem 2016, 40, 47.C. Wu, Q. Liu, R. Chen, J. Liu, H. Zhang, R. Li, K. Takahashi, P. Liu, J. Wang, ACS Appl. Mater. Interfaces 2017, 9, 11106.S. Parvate, P. Dixit, S. Chattopadhyay, J. Phys. Chem. B 2020, 124, 1323.E. K. Sam, D. K. Sam, X. Lv, B. Liu, X. Xiao, S. Gong, W. Yu, J. Chen, J. Liu, Chem. Eng. J. 2019, 373, 531.A. Hooda, M. S. Goyat, J. K. Pandey, A. Kumar, R. Gupta, Prog. Org. Coat. 2020, 142, 105557.W. Li, Y. Zhan, S. Yu, Prog. Org. Coat. 2021, 152, 106117.S. Zheng, D. A. Bellido‐Aguilar, X. Wu, X. Zhan, Y. Huang, X. Zeng, Q. Zhang, Z. Chen, ACS Sustainable Chem. Eng. 2019, 7, 641.S. Zheng, D. A. Bellido‐Aguilar, Y. Huang, X. Zeng, Q. Zhang, Z. Chen, Surf. Coat. Technol. 2019, 363, 43.L. Zhao, Q. Liu, R. Gao, J. Wang, W. Yang, L. Liu, Corros. Sci. 2014, 80, 177.Y. Qing, C. Yang, N. Yu, Y. Shang, Y. Sun, L. Wang, C. Liu, Chem. Eng. J. 2016, 290, 37.Z. Zhang, B. Ge, X. Men, Y. Li, Colloids Surf. A 2016, 490, 182.K. A. Saharudin, S. Sreekantan, N. Basiron, L. K. Chun, V. Kumaravel, T. K. Abdullah, Z. A. Ahmad, Surf. Coat. Technol. 2018, 337, 126.T. Kamegawa, Y. Shimizu, H. Yamashita, Adv. Mater. 2012, 24, 3697.N. Sharifi, M. Pugh, C. Moreau, A. Dolatabadi, Surf. Coat. Technol. 2016, 289, 29.X. Zhang, T. Geng, Y. Guo, Z. Zhang, P. Zhang, Chem. Eng. J. 2013, 231, 414.S. Sreekantan, A. X. Yong, N. Basiron, F. Ahmad, F. De'nan, Polymers (Basel) 2022, 14, 2406.H. Chen, X. Zhang, P. Zhang, Z. Zhang, Appl. Surf. Sci. 2012, 261, 628.S. S. Latthe, A. V. Rao, Surf. Coat. Technol. 2012, 207, 489.Y.‐L. Yan, Y.‐X. Cai, X.‐C. Liu, G.‐W. Ma, W. Lv, M.‐X. Wang, Langmuir 2020, 36, 14924.Y.‐R. Lee, M.‐S. Jang, H.‐Y. Cho, H.‐J. Kwon, S. Kim, W.‐S. Ahn, Chem. Eng. J. 2015, 271, 276.S. Feng, X. Zhang, D. Shi, Z. Wang, Front Chem Sci Eng 2021, 15, 221.S. Chen, X. Li, E. Dong, H. Lv, X. Yang, R. Liu, B. Liu, J. Phys. Chem. C 2019, 123, 29693.E. E. Sann, Y. Pan, Z. Gao, S. Zhan, F. Xia, Sep. Purif. Technol. 2018, 206, 186.N. Nordin, A. Ismail, A. Mustafa, P. Goh, D. Rana, T. Matsuura, RSC Adv. 2014, 4, 33292.Y. Pan, Y. Liu, G. Zeng, L. Zhao, Z. Lai, Chem. Commun. 2011, 47, 2071.C. Chen, Z. Tian, X. Luo, G. Jiang, X. Hu, L. Wang, R. Peng, H. Zhang, M. Zhong, ACS Appl. Mater. Interf. 2022, 14, 23973.Y. Zhao, C. Yan, T. Hou, H. Dou, H. Shen, ACS Appl. Mater. Interfaces 2022, 14, 26077.H. Wang, M. He, H. Liu, Y. Guan, ACS Appl. Mater. Interfaces 2019, 11, 25586.M. J. Nine, T. T. Tung, F. Alotaibi, D. N. H. Tran, D. Losic, ACS Appl. Mater. Interf. 2017, 9, 8393.N. Wang, D. Xiong, Y. Deng, Y. Shi, K. Wang, ACS Appl. Mater. Interf. 2015, 7, 6260.Y. Li, B. Li, X. Zhao, N. Tian, J. Zhang, ACS Appl. Mater. Interf. 2018, 10, 39391.R. Pan, H. Zhang, M. Zhong, ACS Appl. Mater. Interf. 2021, 13, 1743.A. Kim, C. Lee, H. Kim, J. Kim, ACS Appl. Mater. Interf. 2015, 7, 7206.G. S. Watson, D. W. Green, B. W. Cribb, C. L. Brown, C. R. Meritt, M. J. Tobin, J. Vongsvivut, M. Sun, A. Liang, J. A. Watson, ACS Appl. Mater. Interf. 2017, 9, 24381.D. Song, Y. Jiang, T. Chou, K. Asawa, C.‐H. Choi, Langmuir 2020, 36, 11245.M. Schremb, I. V. Roisman, C. Tropea, J Fluid Mech 2018, 835, 1087.P. Jin, X. Yan, M. J. Hoque, K. F. Rabbi, S. Sett, J. Ma, J. Li, X. Fang, J. Carpenter, S. Cai, W. Tao, N. Miljkovic, Cell Rep. Phys. Sci. 2022, 3, 100894.Y. Zhao, Q. Guo, T. Lin, P. Cheng, Int. J. Heat Mass Transfer 2020, 159, 120074.T. Wu, W.‐h. Xu, K. Guo, H. Xie, J.‐p. Qu, Chem. Eng. J. 2021, 407, 127100.Y. Wang, J. Xue, Q. Wang, Q. Chen, J. Ding, ACS Appl. Mater. Interfaces 2013, 5, 3370.I. S. Bayer, C. M. Megaridis, J Fluid Mech 2006, 558, 415.H. Cheng, G. Yang, D. Li, M. Li, Y. Cao, Q. Fu, Y. Sun, Langmuir 2021, 37, 12016.A. Awadallah‐F, F. Hillman, S. A. Al‐Muhtaseb, H.‐K. Jeong, J. Mater. Sci. 2019, 54, 5513.N. A. H. M. Nordin, A. F. Ismail, N. Misdan, N. A. M. Nazri, in AIP Conference Proceedings, AIP Publishing LLC,  2017, 1891, 020091.X. Yao, S. Wu, L. Chen, J. Ju, Z. Gu, M. Liu, J. Wang, L. Jiang, Angew. Chem., Int. Ed. 2015, 54, 8975.P. Roach, N. J. Shirtcliffe, M. I. Newton, Soft Matter 2008, 4, 224.B. Bhushan, Y. C. Jung, K. Koch, Philos. Trans. R. Soc., A 2009, 367, 1631.G. Jiang, J. Hu, L. Chen, Langmuir 2020, 36, 8435.Z. Zhang, X.‐Y. Liu, Chem. Soc. Rev. 2018, 47, 7116.Q. Zeng, S. Xu, J. Phys. Chem. C 2015, 119, 27426.X. Chen, R. S. Patel, J. A. Weibel, S. V. Garimella, Sci. Rep. 2016, 6, 18649.J. Liu, C. Zhu, K. Liu, Y. Jiang, Y. Song, J. S. Francisco, X. C. Zeng, J. Wang, Proc. Natl. Acad. Sci. USA 2017, 114, 11285.H. Choi, H. Liang, J. Colloid Interface Sci. 2016, 477, 176.D. H. Sharp, Physica D 1984, 12, 3.B. Bhushan, Y. C. Jung, K. Koch, Philos Trans A Math Phys Eng Sci 2009, 367, 1631.J. U. Brackbill, D. B. Kothe, C. Zemach, J Comput Phys 1992, 100, 335.C. Zhang, H. Liu, Phys. Fluids 2016, 28, 062107.T. L. Bergman, T. L. Bergman, F. P. Incropera, D. P. DeWitt, A. S. Lavine, Fundamentals of Heat and Mass Transfer., Wiley, xxxx 2011.A. Gultekin, N. Erkan, E. Ozdemir, U. Colak, S. Suzuki, Exp Therm Fluid Sci 2021, 120, 110255. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Advanced Materials Interfaces Wiley

Multiscale Superhydrophobic Zeolitic Imidazolate Framework Coating for Static and Dynamic Anti‐Icing Purposes

Download PDF
13 pages

Loading next page...
 
/lp/wiley/multiscale-superhydrophobic-zeolitic-imidazolate-framework-coating-for-NHfyqxkWj5

References (60)

Publisher
Wiley
Copyright
© 2023 Wiley‐VCH GmbH
eISSN
2196-7350
DOI
10.1002/admi.202202510
Publisher site
See Article on Publisher Site

Abstract

IntroductionIce formation is a natural phenomenon resulting in the malfunction or breakdown of a variety of engineering systems. The development of superhydrophobic surfaces is one of the extensively used approaches for ice mitigation. Inspired by lotus leaves, superhydrophobic surfaces are referred to the surfaces having apparent contact angles (CA) higher than 150°. These bionic surfaces have a contact angle hysteresis (CAH) smaller than 10° or 5° according to various studies.[1–4] Superhydrophobic surfaces can be fabricated by the combination of a hierarchy of micro‐ and nano‐scale surface structures and modification of low surface energy chemistry.[5–7] Various techniques such as anodic oxidation, laser texturing,[7–9] and deposition[10] have been used to develop superhydrophobic surfaces. However, in addition to high costs and multiple‐step procedures, these methods usually require special devices and specific conditions.[3,11,12] Limited by complex fabrication procedures, a facile, low‐cost, and scalable coating method is of great importance.[13–16] In this regard, the development of superhydrophobic coatings containing nanoparticles can be a good substitution for existing conventional time‐consuming, and complex processes. For instance, titanium oxide (TiO2) and silicon oxide (SiO2) nanoparticles have been used to provide superhydrophobic coatings by producing a hierarchical surface structure as an addition to a low surface energy material.[9,17–20] Due to the inherent hydrophilicity of SiO2/TiO2 nanoparticles, polymeric materials such as polydimethylsiloxane (PDMS) and resins are generally required to be used as a matrix to create low surface energy in such coatings.[21–25] Uneven dispersion or aggregation of particles is one of the common problems with these composite coatings and needs to be dealt with sensitivity.[26] Moreover, contact angles reported for superhydrophobic SiO2/TiO2 nanoparticle based coatings are ≈150°, and complex processes are required to enhance their performance.[27–29]Recently, metal–organic frameworks (MOFs) have attracted much due to their significant properties, such as ordered porosity, high internal surface area, low density, and adjustable structure. Zeolitic imidazolate frameworks (ZIFs), a subclass of MOFs, are highly porous and crystalline materials having zeolite‐like structures, which are formed by divalent metal ions nodes and imidazolate ligands. Among this family, ZIF‐8 consists of tetrahedrally coordinated Zn+2 bridged by 2‐methylimidazolate linkers and exhibits sodalite (SOD) topology. ZIF‐8 is comprised of micropores located in the center, which comprises micropores located in the center accessible through six‐membered windows.[30,31] This distinctive porous and cage‐like structure of ZIF‐8 not only has an important role in the superhydrophobic behavior of the coating in terms of creating more air packets and superhydrophobicity but also causes higher durability against moisture penetration compared to conventional coatings based on SiO2 and TiO2 nanoparticles. Furthermore, ZIF‐8 is an intrinsically highly hydrophobic material, mainly due to the presence of the imidazolate rings in its structure.[32,33] This significantly contributes to the water repellency of ZIF‐8‐based coatings in terms of creating low surface energy. Therefore, ZIF‐8 nanoparticles can solely create significant superhydrophobic effects due to their extraordinary structure and nature.Moreover, excellent chemical, and thermal stability, good flexibility for structure and surface modification, adjustable particle size, modifiable pore size, and environmentally friendly synthesis process offer the opportunity to develop coatings with additional functionalities for critical applications, especially for anti‐icing applications.[34,35]Numerous studies have been conducted on static icing of superhydrophobic coatings.[36–38] These studies mainly focus on static icing conditions such as (sub)millimetric droplets[39] and film[40] icing and utilized simple hydrothermal analysis on the anti‐icing nature of the superhydrophobic coating.[41] A detailed analysis of the effect of surface chemistry and texture on mechanisms behind the superhydrophobicity of the coating such as Laplace pressure within the structure, non‐wetting mechanism, condensation‐induced icing mitigation approaches, and ice growth direction is necessary.[42–44] Most importantly, ice formation is dictated by impacting droplets on supercooled surfaces in many industrial applications.[45–47] At high surface supercooling temperatures, the droplet contact time is longer than the freezing timescale. The freezing‐governed hydrothermal interaction between the droplet and supercooled surface and the dynamics of the three‐phase contact line (droplet/surface/air) are still poorly understood at supercooled surface temperatures.[4,48–50] For instance, although advancing and receding contact angles do not vary with the contact line speed on non‐wetting surfaces followed by complete rebound under isothermal conditions,[51] three‐phase contact line dynamics and contact angle hysteresis on supercooled conditions require more analysis. The effective design of anti‐icing surfaces requires an understanding of transient thermal transport through the droplet/surface interface, which is governed by hydrothermal characteristics of the contact line during the impact process.In this study, we report a functional ZIF‐8‐based superhydrophobic multiscale coating (SHMC) with CA > 172°, rolling angle <5°, and CAH < 3° and applied it to metallic using the practical spray coating method. The non‐wetting mechanism of the coating was explained with a fractal theory‐based model of water contact angle. For static icing tests, the effect of SHMC on ice nucleation, ice growth mode, total surface icing time, and possible condensation‐induced freezing was investigated. In dynamic icing experiments, the three‐phase contact line characteristics including contact times, contact diameters, and interfacial heat transfer during the spreading and retraction stages of the impacting droplet on SHMC were covered. A numerical model was developed to further investigate the three‐phase contact line dynamics during the impact process to provide the transient temperature distribution within the droplet. An order of magnitude reduction in heat transfer rate during the total droplet contact time was obtained on the coating.The flow and heat transfer interaction between the droplet and supercooled surface and the contact line dynamics during the impact are still poorly understood. This study not only addresses the shortcomings of the current superhydrophobic surface preparation methods by proposing a MOF‐based multiscale coating but it also presents a fundamental analysis of static and dynamic icing using both computational and experimental methods. Compared to the available coatings which exhibit loss of superhydrophobicity after repetitive icing‐deicing cycles and droplet impact,[42,52] the SHMC maintained its water repellency performance for considerably longer cycle numbers and high impact velocities.Results and DiscussionCharacterization of ZIF‐8 NanoparticlesScanning electron microscope (SEM) images in Figure 1a show nanosized crystals of ZIF‐8. The images indicate the uniform size distribution of the nano spherical particles within the range of 50–100 nm. Figure 1b shows the X‐ray diffraction (XRD) pattern of the as‐synthesized ZIF‐8 nanoparticles which is in good agreement with the literature.[53,54] The presence of well‐defined peaks with high intensity is indicative of good crystallinity of ZIF‐8 nanoparticles. XRD pattern indicates characteristic diffractions at two thetas of 6°, 10.1°, 13.2°, 15.2°, 17°, 18.5°, 19.9°, 23°, 24.9°, 25°, 26°, 27° and 30° degrees corresponding to (110), (200), (211), (220), (310), (222), (321), (411), (332), (431), (440) and (334) planes of the crystalline structure of ZIF‐8, respectively. No peak associated with impurities was spotted. The Fourier transform infrared (FTIR) spectrum of ZIF‐8 nanoparticles is shown in Figure S1 (Supporting Information). The FTIR results are in agreement with the literature and confirm the structure of the nanoparticles along with XRD results.1Figurea) SEM image, b) XRD pattern, c) isotherm linear plot, and d) Horvath‐Kawazoe differential pore volume plot of ZIF‐8 nanoparticles.The N2 adsorption measurement results are shown in Figure 2c,d. The N2 adsorption isotherm in Figure 1b indicates type 1 adsorption behavior. The increase in the adsorbed volume at low pressures is linked with the existence of micropores, while the rapid increase at high pressures is associated with meso or macroporosity due to the stacking of nanoparticles. The micropore volume and Brunauer‐Emmett‐Teller (BET) surface area are obtained as 0.4641 cm3 g −1 and 1173 m2 g−1, respectively. The results are in good agreement with the reported values in the literature.[35,55]The distribution of micropore size of ZIF‐8 nanoparticles was measured using the Horvath‐Kawazoe (HK) method. As shown in Figure 1c, three narrow pore size distributions through the pore width range of 0.7–1.8 nm centered ≈0.75 nm, 1.125 nm, and 1.45 can be observed.2Figure(a‐d) SEM images of ZIF‐8‐based coating in different magnifications, e,f) appearance and sliding WCA values on the developed multiscale coating.Characterization of the Superhydrophobic ZIF‐8 Base CoatingThe morphology and surface texture of the superhydrophobic ZIF‐8‐based coating were characterized by the SEM technique (Figure 2). Observations in low magnifications (Figure 2a) are indicative of the uniform distribution of ZIF‐8 nanoparticles. Images at higher magnifications (Figure 2c,d) and in‐Lens detector images (Figure 2b) exhibit nanoscale cavities through coatings, which result from the packing of nanoparticles. SEM observations along the surface profile of the coatings confirm a micro/nano surface structure for the coating. On the other hand, BET results of ZIF‐8 nanoparticles indicate the existence of a subnano texture on particles, i.e., coating. Taking all these into account, it can be concluded that micro to sub‐nano topology could be formed by this approach.The water contact angles (WCA) were measured by an optical contact angle meter with 3 µL drops of distilled water at ambient temperature. The average value was determined by measuring the same sample at three different positions. The rolling angle (RA) was assessed by increasing the tilting angle with a 1° interval of 10s until the droplet started to roll from the surface. The contact angle hysteresis (CAH) was calculated by subtracting the receding water contact angle from the advancing water contact angle for each captured image at the rolling angle.Non‐Wetting Mechanisms of the Superhydrophobic ZIF‐8 Base CoatingThe superhydrophobic surfaces can be realized by using a combination of low surface energy material and surface texture. Generally, materials with nonpolar chemistries and closely packed stable atomic structures have low surface energy and exhibit water repellency. Polysiloxanes (−Si−O−Si− groups), fluorocarbons (CF2/CF3), nonpolar materials (with bulky CH2/CH3 groups), or polymers with combined chemistry are some examples of low surface energy materials. The methyl functionalized Im linkers as well as the coordinative saturation of the metal sites in ZIF‐8 have a significant contribution to its hydrophobic nature. On the other hand, silane modification of nanoparticles with 1H,1H,2H,2H perfluorooctyltriethoxysilane (PFOTES) results in superhydrophobic behavior of the coating. The partially perfluorinated silane having three hydrolyzable functional ethoxy groups bonds to ZIF‐8 nanoparticles and leads to a structure morphology with strongly bonded perfluoroalkyl functionalities. Low‐energy perfluorinated alkyl chains on a surface with a hierarchical texture generate a very high water repellency. Water molecules cannot form a hydrogen bond with the superhydrophobic ZIF‐8‐based coating and form hydrogen bonds with themselves in order to decrease the energy of the system. Therefore, the water molecules have very low interaction with the coating and cannot wet the surfaces.In terms of the surface structure, the classical Cassie‐Baxter (CB) theory assumes that the coating structure consists of two different materials: MOF particles with WCA of θs and a surface fraction area of φs, and air pockets with WCA of θv and a surface fraction area of 1‐φs.[56] The apparent CB WCA (θCB) on the composite interface is then expressed as:1cosθCB=φscosθS+(1−φs)cosθv=φscosθS+(φs−1)\[\begin{array}{*{20}{c}}{\cos {\theta _{{\rm{CB}}}} = {\varphi _s}\cos {\theta _{\rm{S}}} + \left( {1 - {\varphi _{\rm{s}}}} \right)\cos \theta v = {\varphi _s}\cos {\theta _{\rm{S}}} + \left( {{\varphi _s} - 1} \right)}\end{array}\]As can be seen in Figure 3a, there is a difference between CB WCA values and those obtained experimentally. The difference is due to the CB model being proposed assuming the morphology of single‐scale roughness over the surface. Considering the multiscale nature of the proposed coating, the WCA of the coated surface can be modeled using the Fractal theory.[13,56] The Fractal theory was introduced to evaluate complex geometries, in which a fractal geometry can show excellent superhydrophobicity by mimicking nature.[2] The Koch curve is one of the most used fractal geometries and is constructed by a recursive procedure (Figure 3b). Starting from a straight line, each geometry is made from the preceding geometry by splitting each line into three identical fragments. The fractal structure becomes more complex with the number of series and micro‐ and nanostructures over a lotus surface.[57] Therefore, Equation (1) can be used after introducing a roughness factor (L/l)D‐2:2cosθFr=φs(Ll)D−2cosθS−φv\[\begin{array}{*{20}{c}}{\cos {\theta _{{\rm{Fr}}}} = {\varphi _{\rm{s}}}{{\left( {\frac{L}{l}} \right)}^{D - 2}}\cos {\theta _{\rm{S}}} - {\varphi _{\rm{v}}}}\end{array}\]Here, D is the Hausdorff dimension, and L and l are the upper and lower limit scales of the fractal structure surface, respectively. Figure 3c shows the Atomic Force Microscopy (AFM) results taken from the sample surface area of 1 × 1 µm2 (top) and 20 × 20 µm2 (bottom). Multiscale randomness and self‐similarity features of the surface texture imply that the coating has fractal nature (as discussed in Section 3.2 and shown in Figure 3c,d), and fractal theory can be utilized to analyze the wetting mechanism.[58] More information about the fractal analysis and contact angle calculation is provided in the Supporting Information S2 Section. Based on the fractal analysis, θf of the nanostructure in the fractal wetting model can be directly calculated by extracting the morphology of the multiscale coating. The calculated CB (θCB), fractal (θFr), intrinsic (θs), and experimental (θexp) contact angles are shown in Figure 3a. As seen, a good agreement between the obtained results and experimental ones is achieved.3Figurea) Comparison between the predicted static WCAs by the classical Cassie‐Baxter and fractal methods and experimental results on different samples, b) an example of random fractal Koch structure generated using the MATLAB software with seven iterations, c) AFM results of the surface area of 1 × 1 µm2 (top) and 20 × 20 µm2 (bottom), d) surface profile of the copper coated surface.Phase Change DynamicsFreezing experiments were performed on the Superhydrophobic Multiscale Coating (SHMC) and plain surfaces to investigate the effect of surface energy and surface texture. According to the icing behavior observations, ice formation and accumulation on SHMC samples require a larger supercooling, and ice formation starts at a much lower supercooling temperature on plain surfaces. Equations (3) and (4) express the critical Gibbs Free Energy (nucleation barrier) for homogeneous and heterogeneous nucleation.[59]3ΔGHom∗=16πσ33ΔGv2\[\begin{array}{*{20}{c}}{\Delta G_{{\rm{Hom}}}^ * = \frac{{16\pi {\sigma ^3}}}{{3\Delta G_v^2}}}\end{array}\]4ΔGHet∗=f(θ,R)ΔGHom∗\[\begin{array}{*{20}{c}}{\Delta G_{{\rm{Het}}}^ * = f\left( {\theta ,R} \right)\Delta G_{{\rm{Hom}}}^ * }\end{array}\]Here, f(θ,R) is the shape function that indicates the deviation of the heterogeneous nucleation process from the homogeneous nucleation process. Figure 4a shows the variation in shape function with apparent contact angle and fractal dimensions for hydrophilic and hydrophobic fractal surfaces.[60] As seen, the shape function increases with the wettability of the fractal surface. According to the Classical Nucleation Theory, ice nuclei need to overcome a larger energy barrier (Critical Gibbs Free Energy) to become stable, grow and continue to decrease the energy of the system on the coated surface compared to a plain sample. The plain samples have a higher surface energy and act as a preferable surface for ice nucleation and facilitate freezing.4Figurea) Plot of shape function for different hydrophilic and hydrophobic fractal surfaces (redrawn from,[60] b) surface freezing time on SHMC and bare samples on different surfaces and under ambient conditions, c) Icing activity on SHMC and plain area (the coating at the edges of the sample was cleaned to show the anti‐icing effectiveness of the coated area), d) schematic and microscopic images of coalescence induced droplet jumping on SHMC, e) icing and icing mode on the bare aluminum surface, f) icing and icing mode on the coated aluminum surfaceThe total surface icing times on the SHMC and bare surfaces under different conditions are shown in Figure 4b. The results show that SHMC considerably prolongs the icing time under different surface and ambient conditions. This implies that the critical radius of stable nuclei may be larger on the coating compared to plain surfaces in heterogeneous nucleation, which means that the embryos need to be larger to be stabilized (more molecules aggregate), which requires more time. On the other hand, the rate of ice nucleation can be calculated as:5J=KAint(θ,t)exp(−ΔGc(θ,R)kT)\[\begin{array}{*{20}{c}}{J = K{A_{{\mathop{\rm int}} }}\left( {\theta ,t} \right)\exp \left( { - \frac{{\Delta {G_c}\left( {\theta ,R} \right)}}{{kT}}} \right)}\end{array}\]where K, k, Aint are the kinetic constant, Boltzmann's constant, and the geometric substrate‐liquid “apparent” contact area, respectively. From Equation (5), it is apparent that SHMC increases the critical Gibbs Free Energy of heterogeneous nucleation and decreases the nucleation rate (Figure 4c). The freezing delay is not only related to the higher ice nuclear barrier but high WCA, low CAH, and lower solid‐liquid contact area of the condensate droplets also contribute to icing hindrance on the coated surfaces. Dropwise condensation due to high static WCA and enhanced mobility because of low CAH on SHMC remarkably reduce the risk of condensation freezing. The low solid‐liquid contact area limits the heat transfer from the coating to droplets and extends the icing time. Coalescence induced droplet jumping is the other parameter contributing to the prevention of icing on SHMC (Figure 4d).[61] The jumping of a droplet is a process where condensed water jumps from a surface as the excess surface energy is converted into upward kinetic energy when condensed water droplets merge.Ice growth is thermodynamically driven by the minimization of the Gibbs surface energy, which is ΔG = ΔH − TΔS. The ice growth mode on the SuperHydrophobic Multiscale Coating is majorly different from that of the plain surface. Figure 4e shows the ice growth mode on the bare surface. The ice growth is along the surface, and ice continues to grow in the same focal plane as the substrate. As time passes, ice layers grow and form a thick ice layer. The superhydrophobic nature of the coating imposes a large positive value of TΔS, which overcomes a small positive value of ΔH. The entropic contribution to the Gibbs energy, TΔS, dominates over the enthalpic contribution, ΔH, making it more energetically feasible for the ice crystal to grow off the surface rather than along the surface. As a result, ice growth on SHMC undergoes an off‐surface growth mode, as demonstrated in Figure 4f.[62] Since the contact area of the ice and SHMC is much lower than the bare surface, the ice layer thickness is also lighter compared to the plain samples.The trapped air within SHMC cavities induces a capillary force on the droplet. The stability of these air pockets is critical for the durability of the superhydrophobic coating and for preventing CB to Wenzel transition during condensation as well as repeating icing/deicing cycles. The droplet can penetrate into i) the passages formed due to superimposed MOF nanoparticles and ii) the nanoparticle pores. As shown in Figure 5a, liquid penetration depends on the droplet size (Rd), passage or pore size (d), and droplet and meniscus angles (θd and θm). The net force acting on a droplet on a hydrophobic pore or passage is given as:[63]6Fnet=πσd22Rd+ρgRd(1−cosθd)πd24+πσdcos(θm)\[\begin{array}{*{20}{c}}{{F_{{\rm{net}}}} = \frac{{\pi \sigma {d^2}}}{{2{R_{\rm{d}}}}} + \rho g{R_{\rm{d}}}\left( {1 - \cos {\theta _{\rm{d}}}} \right)\frac{{\pi {d^2}}}{4} + \pi \sigma d\cos \left( {{\theta _{\rm{m}}}} \right)}\end{array}\]Here, σ and ρ are the surface tension (0.075N m−1) and density of water (998 kg m−3) at zero degrees, and g is the gravitational acceleration (9.81 m s−2). The first, second, and third terms on the right‐hand side equation are the Laplace force acting on a drop, gravitational force, and capillary force by a meniscus in a passage, respectively. When a surface is hydrophobic (θd, θm > 90°), the meniscus exerts a force in the upward direction, preventing water penetration. However, when a drop is small enough to make Laplace pressure larger than the capillary pressure, drop penetration can take place due to the positive net force in the downward direction.5Figurea) Schematic and SEM image of the passages formed by MOF nanoparticle superposition and the mechanisms responsible for coating stability against droplets, b) Net forces acting on droplets at different pore/passage and droplet sizes (Here the green region indicates the area corresponding to the calculated maximum passage sizes on SHMC surfaces), c) water contact angle variation with icing/deicing cycles at different surface temperatures.Figure 5b shows the calculated net forces for different droplet and passage/pore sizes. The green area shows the region corresponding to the calculated maximum passage size formed by MOF nanoparticle superimposition. Passages with sizes <1 µm provide a stable condition by preventing the penetration of droplets with radii >100 nm. The liquid penetration within the MOF nanoparticle pores (d < 10 nm) also requires droplet radii <10 nm. The microstructures prevent the droplet from touching the valleys, while hydrophobic nanoparticles pin liquid droplets and thus prevent liquid from filling the valleys between asperities (Figure 5a). The dense microscale structure, hydrophobic nature, high surface area, and nanoporous structure of ZIF8 nanoparticles provide air pocket stability for a higher number of icing/deicing cycles. As can be seen in Figure 5c, the static WCA decreases with the number of cycles. Although a 3% decrease in WCA is observed on tested specimens at Ts = −5 °C, a further decrease in the surface temperature results in hindering the superhydrophobicity of the coating (6% decrease in WCA at Ts = −15 °C).Dynamic Water/Ice Repellency and Contact Line BehaviorAn impacting droplet experiences three stages: spreading, retraction, and bouncing. Figure 6a shows the obtained numerical results for an impacting droplet for different surfaces and under impact conditions. Detailed information about the numerical analysis is provided in Supporting information S3. The droplet dynamics during these stages strongly depend on the relative magnitude of inertia, surface tension, and viscous forces as well as the surface wetting state. Weber (We = ρV2D0/σ), Reynolds (Re = ρVD0/µ), Capillary (Ca = We/Re = µV/σ), and Ohnesorge (Oh = µ/(ρσD0)0.5) numbers are the non‐dimensional numbers used to characterize the impact process, where We, Re, Ca, and Oh numbers stand for the relative magnitude of inertia to surface tension forces, the inertia to viscous forces, viscous to interfacial forces, and viscous to inertia and surface tension forces, respectively. The impact experiments were performed at different surface supercooling temperatures from −5 °C to −20 °C, impact velocities from 1 to 2.5 m s−1, and droplet diameters of 1, 2, and 3 mm. The early stages of droplet impact and spreading for a droplet with We = 80 on plain and coated surfaces at room temperature are illustrated in Figure 6b. The droplet edge upon spreading on the coated surfaces stays untouched by the substrate, resulting in a reduction in the contact area between the liquid and solid phases (Figure 6c). This reduces the viscous dissipation and contact line friction and assists the droplet to maintain its energy and to spread over much larger diameters with larger three‐phase contact line velocities. During the later phases of the spreading stage, fingering of the droplets is evident on coated surfaces even at Weber numbers as low as 25. The Rayleigh‐Taylor instability at the liquid/air interface is the main reason for the formation of fingers and break up at the spreading front of the decelerating interface during the later phases of spreading (Figure 6d).[64,65] Figure 6e shows the Dirac delta function (δ [1/m]) at the droplet/air interface on hydrophilic (HPhi) and superhydrophobic (SHPho) surfaces during the retraction stage. Here, the images are not at the same scale in the x and y directions to show a clear distribution of the Dirac delta function on the interface. The Dirac delta function represents the surface tension force (f = σκnδ) (more information is included in Supporting Information S4).[66] Larger δ values for the droplet on the superhydrophobic (SHPho) surface suggest a stronger interfacial surface tension force, which results in a larger pressure difference across the elongated droplet/air interface. The unbalanced pressure distribution in the droplet also causes the droplet to contract toward its center with larger retraction velocities on the SHPho surface. The retraction velocity has a direct effect on the contact time of a bouncing droplet (τ ∼ (ρD3/σ)1/2). During the later phases of the retraction stage, where the retraction velocity decreases to zero, the droplet bouncing is observed on coated surfaces (Figure 6f).6Figurea) Numerical results on the effect of surface wetting state (HPhi: Hydrophilic, SHPho: Superhydrophobic) and impact condition (We: Weber number defined as the ratio of inertia to surface tension forces); b) High‐speed images taken from the early stages of droplet impact on bare (uncoated) and SHMC aluminum surfaces; c) Droplet edge upon spreading on the bare (uncoated) and SHMC aluminum surfaces; d) The Dirac delta function at the droplet/air interface during the retraction stage; e) The pressure distribution within retracting droplet on superhydrophobic and hydrophilic surfaces; f) High‐speed images taken from the early later stages of droplet spreading on bare (uncoated) and SHMC aluminum surfaces.The performance of the SHMC surface for anti‐icing applications was further examined by investigating the impacting dynamics of droplets with diameters of 1, 2, and 4 mm at substrate temperatures of −5 °C, −10 °C, and −20 °C. The heat transfer between the supercooled surface and droplet is a function of contact time and can be characterized using the heat diffusion rate, defined as the ratio of heat transfer at the droplet/surface interface (qL×τ) over the total sensible and latent heat (q′V).[67] The heat transfer rate at the droplet/surface can be estimated as:[68]7q˙L=hA(Td−Ts)=hπDCL24ΔT=Nu.kwLπDCL24ΔT\[\begin{array}{*{20}{c}}{{{\dot{q}}_L} = hA\left( {{T_{\rm{d}}} - {T_{\rm{s}}}} \right) = h\frac{{\pi D_{CL}^2}}{4}\Delta T = \frac{{Nu.{k_{\rm{w}}}}}{L}\pi \frac{{D_{CL}^2}}{4}\Delta T}\end{array}\]8Nu=0.664Pr0.6Re0.8, Re=ρVCLDCL/μ\[Nu = 0.664{\Pr ^{0.6}}{{\mathop{\rm Re}\nolimits} ^{0.8}},\;{\mathop{\rm Re}\nolimits} = \rho {V_{{\rm{CL}}}}{D_{{\rm{CL}}}}{\rm{/}}\mu \]9q˙L=0.166Pr0.6Re0.8kwπDCLΔT\[\begin{array}{*{20}{c}}{{{\dot{q}}_L} = 0.166{{\Pr }^{0.6}}{{{\mathop{\rm Re}\nolimits} }^{0.8}}{k_{\rm{w}}}\pi {D_{{\rm{CL}}}}\Delta T}\end{array}\]where qL is the heat transfer rate [W], h is the convective heat transfer [W m−2 K−1], Nu is the Nusselt number [‐], kw is the thermal conductivity of water [W m−1 K−1], L is the characteristic length [m], A is the surface area between the droplet and substrate (πD2CL/4, [m2]), and Td and Ts are droplet and surface temperatures (K).The following equation can be used to estimate the total sensible and latent heat of the droplet:[67]10q′V=(c(Td−Ts)+hs) ρVdroplet, Vdroplet=16πD03\[q{\prime _V} = (c({T_{\rm{d}}} - {T_{\rm{s}}}) + {h_{\rm{s}}})\;\rho {V_{{\rm{droplet}}}},\;{V_{{\rm{droplet}}}} = \frac{1}{6}\pi D_0^3\]11q′V=(cΔT+hs)ρπ6D03\[\begin{array}{*{20}{c}}{{{q'}_V} = \left( {c\Delta T + {h_{\rm{s}}}} \right)\frac{{\rho \pi }}{6}D_0^3}\end{array}\]Here, q′V is the heat transfer rate [J], c is the specific heat [J kg−1 K−1], hs is the latent heat of solidification [J kg−1], ρ is the density [kg m−3], and Vdroplet is the volume of the droplet [m3].The heat diffusion rate is expressed as:12RQ=q˙L×τtotqV′=0.166Pr0.6Re0.8kwπDCLΔT×τtot(cΔT+hs)ρπD03/6=0.996Pr0.6kwρ0.2D03VCL0.8DCL1.8×(τspread+τretract)(c+hs/ΔT)\[\begin{array}{c}{R_Q} = \frac{{{{\dot{q}}_L} \times {\tau _{{\rm{tot}}}}}}{{q_V^\prime }} = \frac{{0.166{{\Pr }^{0.6}}{{{\mathop{\rm Re}\nolimits} }^{0.8}}{k_{\rm{w}}}\pi {D_{{\rm{CL}}}}\Delta T \times {\tau _{{\rm{tot}}}}}}{{\left( {c\Delta T + {h_s}} \right)\rho \pi D_0^3{\rm{/}}6}}\\ = \frac{{0.996{{\Pr }^{0.6}}{k_{\rm{w}}}}}{{{\rho ^{0.2}}D_0^3}}\frac{{V_{CL}^{0.8}D_{CL}^{1.8} \times \left( {{\tau _{{\rm{spread}}}} + {\tau _{{\rm{retract}}}}} \right)}}{{\left( {c + {h_s}{\rm{/}}\Delta T} \right)}}\end{array}\]Here, the contact line diameter (DCL), droplet and substrate temperature difference (ΔT), contact line velocity (VCL), and contact time (τtot = τspread+ τretract) dictate the heat transfer rate and RQ ratio. The spreading (τspread/τtot) and retraction (τspread/τtot) ratios at different surface temperatures and droplet conditions are shown in Figure 7a. Although spreading and retraction times increase with the surface supercooling temperature, prolongation of the total droplet contact time reduces the spreading time ratio and increases the τspread/τtot in the spreading stage. The sharp decrease in τspread/τtot with the surface temperature at We = 23 turns into a gradual reduction at We = 230 (as indicated by pink arrows), indicating the major effect of the temperature on regions where surface tension and capillary forces are dominant. The effect of surface temperature on the contact line velocity (VCL) and contact line diameter (DCL) on the SHMC surface is shown in Figure 7b for an impacting droplet with a diameter of 2 mm and We = 23. While the contact line velocity and diameter decrease during the spreading stage (red area), a prominent change in DCL and VCL at the retraction stage (blue area) is observed. The capillary numbers are much lower than unity in the retraction stage, indicating the domination of the surface tension over viscous force. Although the impacting droplet on the supercooled surface has a higher viscosity, a low droplet velocity results in an almost 40% reduction in droplet capillary number. The initial stages of the retraction stage on SHPho and HPhi surfaces were numerically analyzed to show the effect of surface wettability on the heat transfer mechanism and temperature distribution. As shown in Figure 7c, a larger DCL during the retraction stage on the HPhi surfaces results in enhanced convective heat transfer relative to SHPho surfaces. A sharp decrease in droplet temperature is evident, which results in contact line freezing and droplet icing on HPhi surfaces. Lower temperature and temperature gradient within the droplet on SHPho surface, especially at droplet edges, highlight the role of contact line diameter and droplet mobility in limiting the heat transfer between the droplet/surface contact area. The heat transfer rate on a superhydrophobic surface is almost one order of magnitude lower than that of a hydrophilic surface. Figure 7 compares the QR ratio on SHMC and plain samples for different surface temperatures. On a plain sample, the sensible/latent heat transfer is the dominant heat transfer mechanism, especially in the spreading stage. On the coated surfaces, while the temperature difference between the droplet and substrate is high during the spreading stage, the heat transfer is limited by low τspread and either DCL or VCL. The heat transfer rate in the retraction stage is also limited by a small contact area and low contact velocity.7Figurea) Spreading and retraction contact time ratios for different surface supercooling temperatures and Weber numbers; b) Transient contact line velocity and contact line diameter for a droplet impact with a diameter of 2 mm and We = 23; c) High‐speed images taken from the last stages of droplet bouncing from the coated surface at surface temperatures of 20 °C and −10 °C; d) numerical results of the droplet temperature distribution, temperature gradient, and heat transfer rate on superhydrophobic and hydrophilic surfaces; d) IR thermography images of an impact on bare and coated aluminum surfaces.The droplet dynamics upon bouncing from the coated surface are shown in Figure 7e at room temperature (20 °C) and surface temperatures of 10 and −10 °C for We = 80. The effect of a further decrease in surface supercooling on droplet dynamics and resultant heat transfer is negligible, while the surface supercooling gradually begins to have an effect as the inertia force decreases. Compared to bare surfaces, the proposed coating provides outstanding performance in terms of droplet mobility and reduction in the heat transfer rate. The IR thermography images of a droplet impact with We = 80 on the bare and SHMC surfaces shown in Figure 7e confirm the heat transfer deterioration during and after impact on the coated sample. Instant droplet deposition and freezing on a supercooled aluminum surface are evident. The non‐uniform temperature distribution on the coated surface is the result of droplet splashing, breakup, and bouncing, which decreases the droplet contact time and eliminates the risk of icing on the proposed coating. The gradual decrease in the surface temperature of the coated surface after impact (t > 60 ms) also indicates the liquid stage of the secondary droplets remained on the coated surface.ConclusionIn this study, we developed a MOF‐based functionalized superhydrophobic multiscale coating (SHMC) with an apparent CA value larger than 171°, rolling angle of <5°, and contact angle hysteresis of <3°. The developed coating was prepared using the practical spray coating method on aluminum and copper surfaces, two of the most used materials in the relevant industries. A comprehensive examination of the static and dynamic icing performance of the SHMC was made using microscopic and macroscopic analyses. The microscale peaks of the coating texture were achieved by NP superposition, MOF NPs formed nanoscale structures, and nanoparticle pores, and their effect on NP morphology led to subnano features of the structure. Multiscale, randomness, and self‐similarity features of the SHMC texture suggested the fractal nature of the coating, and a fractal theory‐based model of water contact angle was adapted to reveal the non‐wetting mechanism on SHMC. Different modes of icing were observed on plain (on‐surface) and coated surfaces (off‐surface), resulting in different ice morphology due to the presence of the coating on the substrate. The multiscale texture of the SHMC extended the icing time by at least 300% and maintained its superhydrophobicity for >30 icing/deicing cycles. The capillary pressure generated within the multiscale coating prevented the droplet from penetrating into the structure and reduced the risk of condensation‐induced freezing under different variant supercooling and relative humidity conditions and within continuous icing/deicing cycles. Compared to the plain sample, which exhibited instant icing at 60 ms after impact, no icing was observed on the SuperHydrophobic Multiscale Coating. The three‐phase contact line characteristics including the contact times, contact diameters, and interfacial heat transfer during the spreading and retraction stages of the impacting droplet on SHMC were assessed for different surfaces and ambient conditions. The high speed and IR thermography results proved that at least an order of magnitude reduction in heat transfer rate during the total droplet contact time could be obtained on the SuperHydrophobic Multiscale Coating.Experimental SectionMaterials2‐methylimidazole (Hmim, 99%), 1H,1H,2H,2H‐perfluorooctyltriethoxysilane (PFOTES, 97%) and zinc nitrate hexahydrate (Zn(NO3)2.6H2O,99% were purchased from Sigma Aldrich. The other chemicals such as absolute ethanol, n‐hexadecane, and acetone were purchased from Sigma Aldrich and were used without any further purification. Deionized water (DI water) purified by a water purification system was used for all the aqueous solution preparations. Copper (purity ≥ 99.5%) and Aluminum were used for the metallic substrates.Synthesis of ZIF‐8 NanoparticlesZIF‐8 nanoparticles were synthesized in an aqueous system at room temperature. 1.17 gr Zn(NO3)2.6H2O was dissolved in 8 ml DI water while 22.70 gr 2‐methylimidazole was dissolved in 80 ml DI water. The zinc nitrate solution was poured into the 2‐methylimidazole solution and was mixed with a magnetic stirrer to prepare the synthesis solution with the molar ratio of Zn2+ 2‐methylimidazole: H2O = 1:70:1238. After two hours of stirring for 2 hours at room temperature, the product was collected by centrifuge. The nanoparticles were then washed with DI water and dried in a vacuum oven at 85 °C for 24 h.Preparation of superhydrophobic ZIF‐8‐based CoatingIn order to develop a uniform and durable superhydrophobic coating, ZIF8 nanoparticles were first modified by 1H,1H,2H,2H‐perfluorooctyltriethoxysilane (PFOTES). 0.5 ml of 1H,1H,2H,2H‐perfluorooctyltriethoxysilane (PFOTES) was mixed ultrasonically in 50 ml absolute ethanol for 1 h. Consequently, 1.5 gr ZIF‐8 nanoparticles were dissolved ultrasonically in the as‐prepared solution for 30 mins to obtain the coating solution. Metallic Substrates were cut into 30 mm × 30 mm pieces and polished with sandpaper (up to 2000 grade). Consequently, the substrates were degreased by acetone and cleaned with DI water ultrasonically (for 10 min each). The coating solution was spray‐coated on the substrates using a spray gun. Air pressure of 0.1 MPa and spray distance of 20 cm were the spray coating parameters.Materials CharacterizationScanning electron microscopy (SEM) Zeiss Leo Supra 35VP equipped with energy dispersive X‐ray spectroscopy (EDX) was employed to characterize the morphology of the nanoparticles, surface structure of the coating, and elemental composition of the materials. X‐ray diffraction (XRD) analysis was done by recruiting a Bruker D2 Phaser (Bruker AXS GmbH) by CuKα radiation (wavelength of 1.54 A) to characterize the crystal structure of ZIF‐8 nanoparticles. Infrared spectra were recorded using a Nicolet iS50 FT‐IR (Fourier Transform Infrared Spectroscopy) spectrophotometer to confirm the structure of ZIF‐8 nanoparticles (FTIR spectra are provided in Supporting Information S1 Section). The specific surface area and pore volume of the crystals were measured using a Micromeritics gas adsorption analyzer instrument equipped with commercial software for calculation and analysis. The BET surface area was calculated from the adsorption isotherms using the standard Brunauer–Emmett–Teller (BET) equation.Icing TestsIcing experiments were performed in a lab‐made temperature‐ and humidity‐controlled environmental chamber at various relative humidity (RH) settings. Figure 8 shows a schematic of the experimental setup. More information about the icing analysis is provided in the Supporting Information S1 Section. Care was taken to preclude possible contamination such as that due to hydrocarbons. The sample was mounted using thermally conductive adhesive tape (McMaster Carr, 6838A11) on the cold plate of a Peltier cooler (Peltier Module TEG High Temperature). The subcooling of the sample surface was maintained by setting the Peltier cooler at different temperatures. A cooling loop was designed to reduce the temperature of the hot plate of the thermoelectric device for thermal stability in long icing experiments. The sample surfaces were monitored and measured simultaneously using an IR thermal camera (FLIR T1020). A standard DSLR camera (Canon EOS) was used to visualize the icing experiments from the top. The real‐time experiments were recorded from top and side using IR, hush speed, and Digital Single‐Lens Reflex (DSLR) camera. Image processing was performed using MATLAB software.8FigureSchematic of the experimental setup for static and dynamic icing. The green region shows the equipment required for dynamic icing tests and orange region shows the surface cooling apparatus.Droplet Impact TestsThe droplet impact experiments were performed using deionized water as the working fluid with droplet volumes of ≈0.5, 4.2, and 33.5 µl (diameters of 1, 2, and 4 mm). The corresponding apparatus was illustrated schematically in Figure 8b. Different needles (different gauge sizes) were connected to a syringe pump (LEGATO 200, KD Scientific, Holliston, MA, USA). Droplets detached due to gravity from needles mounted at heights between 5 and 32 cm, leading to impact velocities ranging from 1 to 2.5 m s−1. The impact dynamics of droplets were captured using a speed camera (Phantom v9.1 vision research high‐speed camera) operated at different frames per second with long‐distance lenses and a workstation with visualization software (Phantom PCC 3.7 software). A cold light source was used to backlight the impacting droplet on the target surface. Depending on the required frame per second (fps), the resolution of the recorded videos ranged from 960 × 240 (7648 fps) to 1632 × 1200 (1000 fps). The experiments were performed on a single droplet impact, where each experiment was repeated for at least five times.The measuring errors in this study mainly involve the surface temperature and impact conditions measurements. The surface temperatures were measured using a T‐type thermocouple (±1 °C) and IR Thermal Camera FLIR T1020 (±2 °C, with 1024 × 768 resolution). Uncertainty in the ImageJ image software analysis was 1 pixel (±0.01 mm). The impact velocity was calculated by examining the last 20 consecutive frames with a maximum uncertainty of ±0.03 m s−1. The maximum uncertainty in the impact Weber number was 7.58%, which was calculated using the following equation:[69]13δWeWe=(δD0D0)2+(2δu0u0)2\[\begin{array}{*{20}{c}}{\frac{{{\delta _{We}}}}{{We}} = \sqrt {{{\left( {\frac{{{\delta _{{D_0}}}}}{{{D_0}}}} \right)}^2} + {{\left( {2\frac{{{\delta _{{u_0}}}}}{{{u_0}}}} \right)}^2}} }\end{array}\]AcknowledgementsThe authors would like to thank Dr. Mohammad Sajad Sorayani Bafqi for his help in FTIR analysis. The authors also thank Sabanci University Faculty of Engineering and Natural Science (FENS) for graduate student support and appreciate the support from TÜBİTAK (The Scientific and Technological Research Council of Turkey) Support Program for Scientific and Technological Research Project Grant No. 120M659 and Turkish Academy of Sciences (TUBA).Conflict of InterestThe authors declare no conflict of interest.Author ContributionsThe manuscript was prepared through the contributions of all authors. All authors provided their approval to the final version of the manuscript.Data Availability StatementThe data that support the findings of this study are available from the corresponding author upon reasonable request.M. J. Kreder, J. Alvarenga, P. Kim, J. Aizenberg, Nat. Rev. Mater. 2016, 1, 15003.T. Onda, S. Shibuichi, N. Satoh, K. Tsujii, Langmuir 1996, 12, 2125.W. Barthlott, C. Neinhuis, Planta 1997, 202, 1.Q. Li, Z. Guo, J. Mater. Chem. A 2018, 6, 13549.X. Deng, L. Mammen, Y. Zhao, P. Lellig, K. Müllen, C. Li, H.‐J. Butt, D. Vollmer, Adv. Mater. 2011, 23, 2962.J.‐P. Lee, S. Choi, S. Park, Langmuir 2011, 27, 809.G. Kwak, M. Seol, Y. Tak, K. Yong, J. Phys. Chem. C 2009, 113, 12085.B. Qian, Z. Shen, Langmuir 2005, 21, 9007.Y. Lu, S. Sathasivam, J. Song, C. R. Crick, C. J. Carmalt, I. P. Parkin, Science 2015, 347, 1132.J. Song, W. Xu, Y. Lu, J. Mater. Sci. 2012, 47, 162.C. Yeom, Y. Kim, J Ind Eng Chem 2016, 40, 47.C. Wu, Q. Liu, R. Chen, J. Liu, H. Zhang, R. Li, K. Takahashi, P. Liu, J. Wang, ACS Appl. Mater. Interfaces 2017, 9, 11106.S. Parvate, P. Dixit, S. Chattopadhyay, J. Phys. Chem. B 2020, 124, 1323.E. K. Sam, D. K. Sam, X. Lv, B. Liu, X. Xiao, S. Gong, W. Yu, J. Chen, J. Liu, Chem. Eng. J. 2019, 373, 531.A. Hooda, M. S. Goyat, J. K. Pandey, A. Kumar, R. Gupta, Prog. Org. Coat. 2020, 142, 105557.W. Li, Y. Zhan, S. Yu, Prog. Org. Coat. 2021, 152, 106117.S. Zheng, D. A. Bellido‐Aguilar, X. Wu, X. Zhan, Y. Huang, X. Zeng, Q. Zhang, Z. Chen, ACS Sustainable Chem. Eng. 2019, 7, 641.S. Zheng, D. A. Bellido‐Aguilar, Y. Huang, X. Zeng, Q. Zhang, Z. Chen, Surf. Coat. Technol. 2019, 363, 43.L. Zhao, Q. Liu, R. Gao, J. Wang, W. Yang, L. Liu, Corros. Sci. 2014, 80, 177.Y. Qing, C. Yang, N. Yu, Y. Shang, Y. Sun, L. Wang, C. Liu, Chem. Eng. J. 2016, 290, 37.Z. Zhang, B. Ge, X. Men, Y. Li, Colloids Surf. A 2016, 490, 182.K. A. Saharudin, S. Sreekantan, N. Basiron, L. K. Chun, V. Kumaravel, T. K. Abdullah, Z. A. Ahmad, Surf. Coat. Technol. 2018, 337, 126.T. Kamegawa, Y. Shimizu, H. Yamashita, Adv. Mater. 2012, 24, 3697.N. Sharifi, M. Pugh, C. Moreau, A. Dolatabadi, Surf. Coat. Technol. 2016, 289, 29.X. Zhang, T. Geng, Y. Guo, Z. Zhang, P. Zhang, Chem. Eng. J. 2013, 231, 414.S. Sreekantan, A. X. Yong, N. Basiron, F. Ahmad, F. De'nan, Polymers (Basel) 2022, 14, 2406.H. Chen, X. Zhang, P. Zhang, Z. Zhang, Appl. Surf. Sci. 2012, 261, 628.S. S. Latthe, A. V. Rao, Surf. Coat. Technol. 2012, 207, 489.Y.‐L. Yan, Y.‐X. Cai, X.‐C. Liu, G.‐W. Ma, W. Lv, M.‐X. Wang, Langmuir 2020, 36, 14924.Y.‐R. Lee, M.‐S. Jang, H.‐Y. Cho, H.‐J. Kwon, S. Kim, W.‐S. Ahn, Chem. Eng. J. 2015, 271, 276.S. Feng, X. Zhang, D. Shi, Z. Wang, Front Chem Sci Eng 2021, 15, 221.S. Chen, X. Li, E. Dong, H. Lv, X. Yang, R. Liu, B. Liu, J. Phys. Chem. C 2019, 123, 29693.E. E. Sann, Y. Pan, Z. Gao, S. Zhan, F. Xia, Sep. Purif. Technol. 2018, 206, 186.N. Nordin, A. Ismail, A. Mustafa, P. Goh, D. Rana, T. Matsuura, RSC Adv. 2014, 4, 33292.Y. Pan, Y. Liu, G. Zeng, L. Zhao, Z. Lai, Chem. Commun. 2011, 47, 2071.C. Chen, Z. Tian, X. Luo, G. Jiang, X. Hu, L. Wang, R. Peng, H. Zhang, M. Zhong, ACS Appl. Mater. Interf. 2022, 14, 23973.Y. Zhao, C. Yan, T. Hou, H. Dou, H. Shen, ACS Appl. Mater. Interfaces 2022, 14, 26077.H. Wang, M. He, H. Liu, Y. Guan, ACS Appl. Mater. Interfaces 2019, 11, 25586.M. J. Nine, T. T. Tung, F. Alotaibi, D. N. H. Tran, D. Losic, ACS Appl. Mater. Interf. 2017, 9, 8393.N. Wang, D. Xiong, Y. Deng, Y. Shi, K. Wang, ACS Appl. Mater. Interf. 2015, 7, 6260.Y. Li, B. Li, X. Zhao, N. Tian, J. Zhang, ACS Appl. Mater. Interf. 2018, 10, 39391.R. Pan, H. Zhang, M. Zhong, ACS Appl. Mater. Interf. 2021, 13, 1743.A. Kim, C. Lee, H. Kim, J. Kim, ACS Appl. Mater. Interf. 2015, 7, 7206.G. S. Watson, D. W. Green, B. W. Cribb, C. L. Brown, C. R. Meritt, M. J. Tobin, J. Vongsvivut, M. Sun, A. Liang, J. A. Watson, ACS Appl. Mater. Interf. 2017, 9, 24381.D. Song, Y. Jiang, T. Chou, K. Asawa, C.‐H. Choi, Langmuir 2020, 36, 11245.M. Schremb, I. V. Roisman, C. Tropea, J Fluid Mech 2018, 835, 1087.P. Jin, X. Yan, M. J. Hoque, K. F. Rabbi, S. Sett, J. Ma, J. Li, X. Fang, J. Carpenter, S. Cai, W. Tao, N. Miljkovic, Cell Rep. Phys. Sci. 2022, 3, 100894.Y. Zhao, Q. Guo, T. Lin, P. Cheng, Int. J. Heat Mass Transfer 2020, 159, 120074.T. Wu, W.‐h. Xu, K. Guo, H. Xie, J.‐p. Qu, Chem. Eng. J. 2021, 407, 127100.Y. Wang, J. Xue, Q. Wang, Q. Chen, J. Ding, ACS Appl. Mater. Interfaces 2013, 5, 3370.I. S. Bayer, C. M. Megaridis, J Fluid Mech 2006, 558, 415.H. Cheng, G. Yang, D. Li, M. Li, Y. Cao, Q. Fu, Y. Sun, Langmuir 2021, 37, 12016.A. Awadallah‐F, F. Hillman, S. A. Al‐Muhtaseb, H.‐K. Jeong, J. Mater. Sci. 2019, 54, 5513.N. A. H. M. Nordin, A. F. Ismail, N. Misdan, N. A. M. Nazri, in AIP Conference Proceedings, AIP Publishing LLC,  2017, 1891, 020091.X. Yao, S. Wu, L. Chen, J. Ju, Z. Gu, M. Liu, J. Wang, L. Jiang, Angew. Chem., Int. Ed. 2015, 54, 8975.P. Roach, N. J. Shirtcliffe, M. I. Newton, Soft Matter 2008, 4, 224.B. Bhushan, Y. C. Jung, K. Koch, Philos. Trans. R. Soc., A 2009, 367, 1631.G. Jiang, J. Hu, L. Chen, Langmuir 2020, 36, 8435.Z. Zhang, X.‐Y. Liu, Chem. Soc. Rev. 2018, 47, 7116.Q. Zeng, S. Xu, J. Phys. Chem. C 2015, 119, 27426.X. Chen, R. S. Patel, J. A. Weibel, S. V. Garimella, Sci. Rep. 2016, 6, 18649.J. Liu, C. Zhu, K. Liu, Y. Jiang, Y. Song, J. S. Francisco, X. C. Zeng, J. Wang, Proc. Natl. Acad. Sci. USA 2017, 114, 11285.H. Choi, H. Liang, J. Colloid Interface Sci. 2016, 477, 176.D. H. Sharp, Physica D 1984, 12, 3.B. Bhushan, Y. C. Jung, K. Koch, Philos Trans A Math Phys Eng Sci 2009, 367, 1631.J. U. Brackbill, D. B. Kothe, C. Zemach, J Comput Phys 1992, 100, 335.C. Zhang, H. Liu, Phys. Fluids 2016, 28, 062107.T. L. Bergman, T. L. Bergman, F. P. Incropera, D. P. DeWitt, A. S. Lavine, Fundamentals of Heat and Mass Transfer., Wiley, xxxx 2011.A. Gultekin, N. Erkan, E. Ozdemir, U. Colak, S. Suzuki, Exp Therm Fluid Sci 2021, 120, 110255.

Journal

Advanced Materials Interfaces – Wiley

Published: May 1, 2023

Keywords: anti‐icing; contact line dynamics; droplet impact; metal‐organic framework; micro‐nano‐subnano coating; superhydrophobic; ZIF‐8

There are no references for this article.