1 Introduction

Wind power has garnered significant attention in recent years due to the substantial environmental pollution resulting from the utilization of fossil fuels and the escalating concerns regarding climate change [1]. As an alternative solution to fossil energy, there has been a substantial surge in new wind power installations between 2020 and 2022. This remarkable global growth serves as a testament to wind power’s status as one of the most sustainable and economically viable energy sources [2]. However, the persistent issue of icing on wind turbine blades has posed a longstanding challenge, leading to substantial energy losses and critical situations annually. In cold regions, wind turbines often necessitate shutdowns due to the accumulation of ice on their surfaces [3]. According to a study by L. Gao et al., the presence of ice accretion on the blades can result in a drastic 90% reduction in power generation, while also influencing the frequency and aerodynamic behavior of the turbine blades. This detrimental effect may cause damage and give rise to additional safety concerns associated with the turbines [4]. The current state of anti-/de-icing technologies can be categorized into two distinct approaches: active anti-/de-icing and passive anti-/de-icing [5]. Active techniques encompass energy conversion heating [6], physical de-icing [7], chemical de-icing [8], etc. In contrast, passive methods involve superhydrophobic anti-/de-icing [9], slippery surface anti-/de-icing [10], self-replenishing anti-/de-icing [11], etc. Among these, active heating anti-/de-icing, such as hot air heating [12] and electric heating [13] employing carbon fiber layers, has been widely employed in commercial applications.

Currently, some high anti-/de-icing performance materials have been reported, such as Jiaying Wang et al. reported a high photo-thermal anti-/de-icing performance coating through the deposition of fluorinated silica nanoparticles on the MXene@Au-waterborne polyurethane coating. They obtained a photo-thermal superhydrophobic coating, which exhibited 73% de-icing efficiency [14]. Huiying Xiang reported a very low ice adhesion anti-/de-icing material through liquid infusion surface with pores structure on the surface [15]. The above reports have potential environmental issues and durability issues. The function was very easy to be influenced by the environmental conditions, especially on the wind turbines.

Olivier Parent et al. summarized the technologies of anti-/de-icing on wind turbines. There is no coating that is truly ice-phobic based on many tests on the wind turbines in the past many years. The designed coating materials cannot meet the complex environment. Active heating is the only useful method for de-icing purposes on the blades [16]. Electric heating was considered a more efficient approach than hot air heating and photo-heating [17]. Yubo Liu et al. developed a superhydrophobic electric heating coating through fluorine resin and carbon nanotube mixing. The coating showed good mechanical properties and anti-/de-icing behavior [18]. However, the using of fluorine elements will cause significant environmental problems. In our previous work, we reported a graphene-based electric heating and photo-thermal green hydrophobic coating. The field tests proved that it can achieve high-efficiency anti-/de-icing performance in the arctic region in the winter [19]. Although the above coating showed good performance and great potential applications on wind turbines. The durability and feasibility of the coating were not identified. Once the naked conductive heating layer is damaged, it will cause anti-/de-icing failure. Therefore, an anti-/de-icing protection coating with high thermal conductivity is still needed. The coating also can meet the wear resistance request of the turbine coating. Insufficient wear resistance can result in material failure and the loss of anti-/de-icing functionality [20]. There are some researchers introduced how to improve the wear resistance of the anti-/de-icing coating. Yibin Liu et al. reported a poly 2-perfluoroethoethyl methacrylate/carbon nanotubes polydimethylsiloxane material capable of maintaining anti-/de-icing performance after 10 wear cycles [21]. Peng Wang et al. introduced hydrophobic carbon powders and carbon nanotubes into the polyurea-urethane resin. Both thermal conductivity and wear resistance were improved [22]. However, only a few investigations involved wind turbine coatings. Therefore, it is extremely urgent to further solve the problem of ice accumulation on blades in winter.

To address this concern, a commercial gelcoat coating which was widely used on turbine blades was employed, aiming to mitigate the issue [23]. gelcoat coating contained polymer resin, magnesium aluminum silicate, and titanium dioxide. In order to enhance thermal transfer properties and promote wear resistance, boron nitride and graphene which are considered excellent additives for improving the thermal conductivity of resin materials were incorporated as thermal transfer mediums and wear-resistant additives, respectively, within the gelcoat coating [24,25,26]. At the same time, the reinforcement effect of nanoparticles can also increase the strength of the material, thereby improving the wear resistance of the material [27, 28]. The mechanisms underlying thermal conductivity and wear resistance were comprehensively characterized. The introduction of graphene significantly improved the wear-resistance performance of the coating due to its unique layered structure, whereas the inclusion of boron nitride resulted in pronounced wear [29]. The improved gelcoat composite coating exhibited exceptional heat transfer efficiency, thereby augmenting the anti-/de-icing effectiveness of existing active heating technologies while simultaneously reducing energy consumption. Furthermore, the modified gelcoat composite coating demonstrated superior anti-/de-icing performance compared to the pure gelcoat coating. This work is improved based on commercial wind turbine blade coatings. In terms of cost and feasibility, it is more advantageous than the work that has already been reported. In addition, graphene composite materials usually exhibit electromagnetic interference shielding [30], and radio frequency properties [31]. The developed composite coating has a great potential application in outdoor wind turbine coating, icing detection sensors, nanogenerator materials, and other relevant fields.

2 Experiment and materials

2.1 Materials

The gelcoat coating, hardener, and grinding paper (400 mesh) were purchased from Biltema AB, Sweden. The graphene was bought from Tanfeng Tech. Inc., China. The boron nitride nanoparticle was obtained from Zhengzhou Fengyue Chemical Tech. Inc., China. Other experiment materials were purchased from VWR. AB, Sweden.

2.2 Experiment

2.2.1 Preparation of the gelcoat composite coating

Ten grams gelcoat coating was mixed with 0.2 g, 0.4 g, 0.6 g, 0.8 g, and 1 g graphene under mechanical blending with 100 rpm/min for 30 min. Then, 0.05 g hardener was added to the mixture, stirring at 300 rpm/min for 1 min. Then, the gelcoat graphene composite coating was obtained. Ten grams gelcoat coating was mixed with 0.2 g, 0.4 g, 0.6 g, 0.8 g, and 1 g boron nitride under mechanical blending with 100 rpm/min for 30 min. Then, 0.05 g hardener was added to the mixture during stirring at 300 rpm/min for 1 min. Then, the gelcoat boron nitride composite coating was obtained.

2.2.2 Preparation of the gelcoat composite coating samples

Two grams gelcoat composite coating was put on one 50 × 50 mm polycarbonate plate. A squeegee (Fig. S1) was used three times to flatten the coating and then swipe off the residual coating. The composite coating samples were then kept at room temperature for 24 h.

2.3 Characterization method of gelcoat samples

2.3.1 Structure and mechanical properties

The FTIR spectra which showed the structure of the gelcoat composite coating were taken by infrared spectroscopy (Thermo Fisher MAS60). An insert the specific model of our Nano-Indenter here Nano indentor was employed to characterize the mechanical properties of the composite coating samples with 0.1 mN load, repeated 10 times.

2.3.2 Wear resistance test

The composite coating’s nano-micro scale wear properties were characterized using the Nano-Indenter with a load of 20 mN and a scratch length of 500 µm, which was repeated 2 times [32]. The micro–macro UMT test rig (Fig. S2) using the pin-on-disc configuration was employed to evaluate the large-scale wear resistance [33]. As a generic and qualitative way of evaluating the wear performance of the coating, a pin surface of the grinding paper (400 mesh) which contained many abrasive particles was used to simulate the wear situation of a turbine blade. The movement type was reciprocation with a scratch distance of 10 mm. The abrasive paper size was 19.5 mm × 0.95 mm, and the load was 0.026 MPa. The test duration was 10 min, and the frequency was 1 Hz. The 3D Optical surface topography was recorded through Zygo (New Way 7300) 3D optical microscope with × 2.75 ×  × 0.5 magnification [34]. The scanning electron microscope (SEM) images and energy dispersive X-ray spectroscopy (EDS) of the gelcoat boron nitride coating before and after wear were collected under 20.0 kV, WD15.1 mm, × 1000, 10-µm resolution conditions.

2.3.3 Thermal conductivity test

A Hot-Disk TPS2200 thermal coefficient instrument was employed to evaluate the thermal conductivity of the composite coating. Photos of the heat distribution were taken with an IR camera (OPTRIS PI 4000). A DU401_BV Raman instrument was used to analyze the structure of the graphene and boron nitride and the thermal conductive mechanism. The wavelength of the laser source was 532 nm. Acquisition mode was 5 times accumulations.

2.3.4 Anti-/de-icing test

The ice adhesive strength was recorded through a purpose-built ice shear force measurement instrument, as shown in Fig. S3 [35]. It is a push force sensor that was fixed on a frequency motor. The test was set with 400 mm/s of pushing speed, and the push force curve with time was recorded by the force sensor. The ice cube had a size of 20 mm × 20 mm × 20 mm, and the composite coating sample size was 50 mm × 50 mm. Firstly, wet the surface with natural water and put the ice cube on the wet area, and then, press it tight and keep it in the fridge to freeze. Every sample was measured 25 times of ice adhesion strength, and the average value was used in Fig. 8b. The icing delay time and water contact angle of the composite coating were evaluated by a Biolin scientific contact angle measurement device (Attention Theta). A water drop with a 4 μl volume was put on the composite coating sample surface and then kept inside a freezer. Contact angle measurements were done with the Attention Theta instrument every 2 min by taking out the samples from the freezer. The fridge temperature was − 20 ± 1 ℃. To get a good understanding of the anti-/de-icing behavior, the surface topography and surface structure were characterized. The surface topography was measured using a 3D profile meter (Zygo, New Way 7300), with a × 2.75 ×  × 0.5 magnification.

3 Result and discussion

3.1 Molecular structure and mechanical property

The FTIR spectra show the structure of the composite coatings, as shown in Fig. 1a, b. In this work, the boron nitride and graphene were introduced through physical blending with different concentrations. Therefore, the additives have no influence on the molecular structure, and there is no significant change in the FTIR spectra. The graphene will absorb some light which will cause the baseline of the FTIR spectra to have a small change [36]. The mechanical properties were evaluated using a Nano-Indenter [37], as shown in Fig. 1c, d. With the increase of concentration of boron nitride and graphene, the hardness of the composite coating increased until the mass concentration reached 8% and then decreased at 10% of mass concentration. It means the strength of the composite coating can reach a maximum with an additive concentration of 8%. On the other hand, the penetration depth showed a decrease at first and then an increase. It reached the minimum value when the concentration of the additive reached 8%. Compared with the pure coating samples, the strength of the composite coating at 8% additive concentration can reach 5.59 MPa, which is higher than pure gelcoat coating. This result shows that for both boron nitride and graphene, the optimal concentration is 8%.

Fig. 1
figure 1

The FTIR spectra of the gelcoat boron nitride coating (a) and gelcoat graphene coating (b). c The hardness of the gelcoat boron nitride coating and gelcoat graphene coating. d The penetration depth of the gelcoat boron nitride coating and gelcoat graphene coating

Full size image

3.2 Thermal conductivity and mechanism

Boron nitride and graphene are common thermal conductive additives [38], and graphene is currently even a hot topic material due to its many high-performance properties [39]. In this work, boron nitride and graphene are used to improve thermal conductivity. As shown in Fig. 2a, the pure gelcoat coating showed a 0.20 W/mK thermal conductive coefficient. The gelcoat boron nitride coating showed very little change in the thermal conductivity. The maximum coefficient reached 0.34 W/mK. Moreover, there is no clear trend with the concentration increasing. On the other hand, the gelcoat graphene coating has a clear trend and the coefficient of the thermal conductive is three times that of pure gelcoat coating. It can reach 0.61 W/mK when the concentration of the graphene rises to 10%. As shown in Fig. 2b, the heating distribution and speed were evaluated through the IR camera; the gelcoat graphene coating can transfer heating faster which results in the surface temperature can reach the same temperature as the heating pad in 30 s. Pure gelcoat coating and gelcoat boron nitride coating photos still showed the blurry shadow part which means the surface temperature is not uniform.

Fig. 2
figure 2

a The thermal conductivity coefficient of gelcoat boron nitride coating and gelcoat graphene coating. b The comparison of the heat distribution between gelcoat boron nitride coating, gelcoat graphene coating, and pure gelcoat coating

Full size image

As some references reported, thermal conductivity depends on the phonons which are the segments of the molecular and crystal cube. The heat can be transferred by the vibration of the segments of the molecular and crystal cubes, which can be measured by Raman shifted [40, 41]. As shown in Fig. 3, boron nitride and graphene have characteristic Raman-shifted peaks. In Fig. 3a, the boron nitride showed sharp 676 cm−1 and 3600 cm−1 peaks, both presenting the B-N vibration, but the 3600 cm−1 peak presented the B-N inside the crystal cube. When the boron nitride was mixed with the gelcoat coating, the crystal regions were divided and destroyed by the gelcoat coating. This is why there was no 3600 cm−1 peak appearing in the gelcoat boron nitride coating samples. From 2 to 6% concentration gelcoat boron nitride coating, the peaks are flat and weak. This means that there was no significant vibration of the segments and crystal cube in gelcoat boron nitride coating. From 8 to 10%, the Raman peaks are sharper than the lower concentration, but many weak peaks appeared. The unnecessary vibration will reduce the heat transfer. This is the reason the thermal conductivity of gelcoat boron nitride coating has not increased significantly. Figure 3b shows the graphene and gelcoat graphene coating Raman shifts. Graphene showed 1580 cm−1 (G peak) and 2700 cm−1 (2D peak) peaks. G peaks present the variation of the crystal cubes. 2D peak presents the structure of the crystal cubes. At the same time, graphene has a 2D carbon layer structure, and there exists an Π-Π interaction and Van der Waals force between different layers [42]. The heat can be transferred through the crystal cubes and layer structure. After being mixed with the gelcoat coating, the G peak and 2D peak occurred blue shifted and splitting into two peaks. This indicates that the vibration of the crystal cubes was influenced by the gelcoat coating.

Fig. 3
figure 3

The Raman shifts spectra of gelcoat boron nitride coating (a) and gelcoat graphene coating (b)

Full size image

Furthermore, with the concentration of graphene increasing, the G peak and 2D peak red-shifted to 1580 cm−1 and 2700 cm−1. This is because the high concentration of graphene formed a continuous phase inside the gelcoat graphene coating. The continuous graphene phase builds up the thermal conductive network inside the gelcoat graphene coating. The heat can be transferred through the continuous graphene phases. Consequently, the coefficient of thermal conductive will increase with the increased concentration of graphene. However, a high concentration of additives may cause the materials to lose their mechanical strength. As some reports mentioned, the composite materials would lose their elastic properties and cause severe wear [43]. This may cause dangers and even damage the turbines. Therefore, in this work, we choose the suitable concentration value based on the mechanical strength.

3.3 Friction behavior and wear resistance

The friction behavior of the gelcoat graphene coating and gelcoat boron nitride coating was evaluated in Fig. 4a–d. Among them, Fig. 4a, b showed the nano scratch measurement results through a nano indentor, and Fig. 4c, d showed the friction coefficient of the UMT tests. In the nano scratch analysis, the gelcoat boron nitride coating showed a friction coefficient from 0.35 to 0.8. Among these, the gelcoat coating with a boron nitride concentration of 8% showed the lowest friction coefficient compared with other samples. The gelcoat graphene coating showed a friction coefficient from 0.45 to 0.8. During the nano scratch test, the indentor worked as an abrasive particle and penetrated the surface, and the tribology behavior was influenced by the surface hardness under the same vertical load, while particle enhancement impacted surface hardness. In Fig. 1c, the gelcoat boron nitride coating showed higher hardness than the gelcoat graphene coating. Hence, the gelcoat boron nitride coating has lower plastic deformation and lower contact area on the surface, leading to a lower friction coefficient than the gelcoat graphene coating. Next, during the UMT tests, the grinding paper consists of many abrasive particles, which will penetrate the coating surface at different angles and positions. Figure 5 shows the wear mark of the gelcoat samples with boron nitride and graphene. The boron nitride samples showed a larger wear area.

Fig. 4
figure 4

The friction coefficient of nano-scratch wear test of gelcoat boron nitride coating (a) and gelcoat graphene coating (b). The friction coefficient of an abrasive wear test using UMT of gelcoat boron nitride coating (c) and gelcoat graphene coating (d)

Full size image
Fig. 5
figure 5

a The 3D Optical images of the gelcoat boron nitride coating and gelcoat graphene coating after grinding paper wear. b The friction region photos of the gelcoat boron nitride coating and gelcoat graphene coating

Full size image

As shown in Fig. 6b, the grinding paper typically consists of one layer of abrasive particles. During the wear operation, the abrasive particles penetrate the gelcoat composite coating surface under loading. From Fig. 5, the particles caused the removal of the coating through cutting, plowing, and rubbing. Many plastic deformations and penetrations were occurring on the surface [44, 45]. Therefore, the mechanical properties of the gelcoat composite coating will affect its wear behavior. In Fig. 1c, the hardness was measured by a Nano-Indenter. Both the gelcoat boron nitride coating and the gelcoat graphene coating showed the highest hardness around the 8% concentration of the fillers. During the wear process, the removal of materials caused many particles, which are softer than the abrasive particles on the grinding paper. Hence, the additive particles are important and will affect the wear behavior of the gelcoat composite coating. In this work, graphene and cubic boron nitride were used. Graphene has better lubrication properties than boron nitride due to its unique structure [46]. The carbon atom layers of the graphene can slide on each other [47]. When the graphene particles inside the coating get into the interface due to the wear, it will play a role as a lubricant and reduce the wear and friction. However, the boron nitride has a facial center cube structure, but the crystal regions are out of order. Therefore, there will be less sliding between planes in boron nitride than in graphene [48].

Fig. 6
figure 6

a The wear depth of the gelcoat boron nitride coating and gelcoat graphene coating after grinding paper wear. b The diagram of the accelerated grinding paper wear test

Full size image

At the same time, Fig. 6a shows how deep the abrasive particles penetrate the gelcoat composite surface. All the gelcoat graphene coatings showed less wear depth than gelcoat boron nitride coating once the graphene and boron nitride particles get into the interface. The boron nitride particles are harder than graphene and gelcoat resin materials due to the different crystal structures. Therefore, these boron particles would penetrate the coating surface again and cause severe wear but not graphene. The SEM images and EDS analysis gave further analysis.

The scanning electron microscope (SEM) images and energy dispersive X-ray spectroscopy (EDS) of the gelcoat boron nitride coating before and after wear are shown in Fig. 7a, b. The gelcoat graphene coating before and after wear is shown in Fig. 7c, d. There is no significant difference in the initial surface for the gelcoat boron coating and the gelcoat graphene coating. Both have many particles and fragments after surface wear. To further analyze the surface change, we conducted energy dispersive X-ray spectroscopy (EDS) (Fig. S4) to get the element distribution of the white particles and black grooves. The results showed that the boron element has a uniform distribution on the white particles and black groves in Fig. 7. This means that the boron nitride particles have a great compatibility with the gelcoat coating. In contrast, the carbon element distribution is not uniform on the gelcoat graphene coating surface where black grooves have a higher carbon concentration than white particles. During the wear process, the irregular movement of the abrasive particles scratched the surface and resulted in grooves on the surface. High carbon concentration in the grooves means high graphene concentration due to the graphene being the main source of carbon. The above results show that graphene gathers in the grooves and reduces friction and wear between abrasive particles and the coating surface.

Fig. 7
figure 7

SEM images of the gelcoat composite coating. a The w = 6% of gelcoat graphene coating before the UMT wear test and b after the UMT wear test. c The w = 6% of gelcoat boron nitride coating before the UMT wear test and d after the UMT wear test

Full size image

Further, the polymer and particles were getting into the gap between the abrasive particles during the wear process. As shown in Figs. 5b and 6b, the materials loss and transfer occurred on the interface, some of them filled between abrasive particles, and the counter side photo showed the significant materials transfer. SEM images in Fig. 7c, d and EDS results in Fig. S4 proved graphene gathered in the grooves and stopped the further penetration of the abrasive particles but not boron nitride. This is the reason the gelcoat graphene coating shows a lower friction coefficient than gelcoat boron nitride coating when the concentration was 6% and 8% under the same grinding load in Figs. 4c, d and 6a. Regarding boron nitride, the rise of the mechanical strength is the main reason to reduce the wear. Gelcoat boron nitride coating showed the highest hardness under 8% of boron nitride concentration. Compared with other boron nitride concentrations, both the friction coefficient of 0.4 and the wear depth of 0.25 mm indicate that the 6% gelcoat boron nitride coating has good wear resistance.

3.4 Anti-/de-icing performance

To analyze the anti-/de-icing behavior and how the wear affects the anti-/de-icing behavior of the gelcoat boron nitride coatings and gelcoat graphene coatings, the icing delay time of the water drops on the gelcoat composite coating surface was measured which can be used to evaluate the anti-/de-icing performance [49, 50]. The icing delay time is a measure of how long a water droplet will remain in a liquid state on the surface at − 20 ℃ before it freezes. As shown in Fig. 8a, the icing delay time showed a similar trend with the concentration of the boron nitride and graphene increasing. The interesting thing is that the after-wear gelcoat boron nitride coating and gelcoat graphene coating showed a shorter icing delay time than the before-wear of gelcoat composite coatings. Compared with the before-wear gelcoat composite coating, the wear would change the surface roughness of the gelcoat composite coating. The surface roughness will affect the surface properties of the gelcoat composite coating. There will be more asperities on the surface which will be as a crystal cube and accelerate the icing procedure. That is why before-wear and after-wear has different icing delay time. The reason for the above phenomena is investigated in Fig. 9.

Fig. 8
figure 8

a The icing delay time of the gelcoat boron nitride coating and gelcoat graphene coating before and after wear. b The ice adhesive strength before wear and after wear of gelcoat boron nitride coating and gelcoat graphene coating

Full size image
Fig. 9
figure 9

a The before-wear and after-wear water contact angle of gelcoat boron nitride coatings. b The before-wear and after-wear water contact angle of gelcoat graphene coatings. c The surface roughness of Ra of the gelcoat boron nitride coating and gelcoat graphene coating. d A schematic diagram of water drops on the smooth and rough surfaces at 20 ℃ and − 20 ℃ (The Cassie-Baxter state denotes the ability of a superhydrophobic surface to maintain its lotus-like self-cleaning effect. Wenzel state presents a liquid droplet that completely wets the substrate and is pinned to the rough surface [52])

Full size image

Furthermore, in Fig. 8b, the ice adhesive strength of the gelcoat boron nitride coating decreased from 270 to 220 kPa when the boron nitride was added into the gelcoat coating. The ice adhesive force of gelcoat graphene coating showed a range from 270 to 180 kPa decreasing with the graphene content used. This is because of the roughness difference between gelcoat boron nitride coating and gelcoat graphene coating. The roughness is characterized and analyzed in more detail later. Cubic boron nitride showed a typical cube structure, while graphene showed a typical 2D layer structure [51]. These have different influences on the surface of the gelcoat coatings. The mechanism of anti-/de-icing behavior is characterized in Fig. 9. Gelcoat graphene coating and gelcoat boron nitride coating showed an increase of 50–175 kPa in the ice adhesion after wear. The initial surface of the coating was severely damaged and formed a new surface. Many abrasive particles lead to the formation of large-scale bulges and grooves, which form many mechanical interlocks during the freezing process, resulting in increased ice adhesion strength. Therefore, surface wear and damage are detrimental to anti-/de-icing. In addition, we selected a 6% graphene and boron nitride concentration of gelcoat composite coating to do SEM measurement after ice removal in Fig. S5. The gelcoat boron nitride coating and gelcoat graphene coating showed high modulus due to the particle enhancement. The ice adhesion force is not enough to damage the surface and further affect ice adhesion strength significantly.

3.5 Anti-/de-icing mechanism

The surface properties and structure are very important for anti-/de-icing performance [53]. In Fig. 9a, b, the gelcoat graphene coating showed around 82 ± 3° water contact angle, and the gelcoat boron nitride coating showed around 70 ± 5° water contact angle. This phenomenon responds to the surface roughness difference investigated in Fig. 9c. The surface roughness of the gelcoat boron nitride coating increased 2 times from 2 to 10%. It is worth mentioning that the roughness of the gelcoat graphene coating showed 5 times increase from 2 to 10%. Graphene has a 2D layer structure, which can produce higher roughness more easily than boron nitride. A high roughness can increase the water contact angle of the materials [54].

Figure 9d shows a schematic diagram of the water drop on smooth and rough surfaces and under different temperatures. Under room temperature, rough surface gaps can catch air to form an air pocket. The formation of many air pockets on the surface reduces the surface energy of the material which results in lower wetting of the surface. Therefore, the rough surface usually has a high-water contact angle. Since no chemical changes occur in the coating during the wetting process. Our coating follows the above physical phenomena very well.

Surface roughness increased due to filler use and wear, leading to a larger water contact angle. In addition, when the environment temperature is decreased to − 20 ℃, the water drops will change from the Cassie-Baxter State to the Wenzel state in Fig. 9d [52]. However, high roughness provides many heterogeneous nucleation sites, which accelerates the freezing of water on the surface. This is the reason gelcoat boron nitride coating and gelcoat graphene coating showed shorter icing delay time after wear in Fig. 8a, b. Rapid freezing causes more air pockets to become trapped underneath the ice, which reduces the actual contact area between the ice and the coating. Therefore, the adhesion of ice cubes to the surface decreases compared to the pure gelcoat coating, since gelcoat graphene coating always showed higher roughness than gelcoat boron nitride coating, as shown in Fig. 9c.

4 Conclusion

In this study, we aimed to enhance the wear resistance, thermal conductivity, and anti-/de-icing properties of a gelcoat composite coating by incorporating graphene and boron nitride. The gelcoat graphene coating showed better performance than the gelcoat boron nitride coating and pure gelcoat coating. The improved wear resistance of the gelcoat graphene coating can be attributed to the two-dimensional layer structure of graphene, while the addition of graphene resulted in a threefold increase in the thermal conductivity of the gelcoat composite coating compared to the pure gelcoat coating. The gelcoat composite coatings exhibited a high-water contact angle and low ice adhesive force. It was observed that as the surface roughness increased, the water contact angle also increased. The increase in ice adhesion after abrasion proves that abrasion is always detrimental to de-icing. Despite the extension of icing delay time, the large number of grooves and bumps created by wear results in stronger mechanical interlocking. It is worth mentioning that gelcoat graphene coating still demonstrated lower ice adhesive strength than gelcoat boron nitride coating and pure gelcoat coating. Overall, we successfully developed a gelcoat graphene coating with improved thermal conductivity, wear resistance, and low ice adhesive properties. This novel composite coating has the potential to significantly enhance the efficiency of existing heating technologies for anti-/de-icing applications, thereby reducing energy consumption associated with the turbine blades’ anti-/de-icing system.