The contact angles of the samples fabricated and studied in this work are listed in Table 1. It should be noted that the measured sliding angles are very low (<1.5°) for all textured samples 1–4, which indicates a heterogeneous wetting regime of the obtained coatings. The presented data demonstrate that the selected laser processing conditions make it possible to achieve a superhydrophobic state with extremely high contact angles. Despite a fivefold difference in laser scanning speed, the contact angles of samples 1 and 2 are close to each other. This is due to the similarity of the hierarchical morphologies of the two coatings (cf. Figs. 1a, 1b and 1c, 1d), which differ only slightly in the size of microaggregates formed on the surface during the deposition of particles from the laser plume (Fig. 1).

SEM images of the surface of samples 1 (a, b), 2 (c, d), 3 (e, f), 4 (g, h), 5 (i, j), and 6 (k, l) at different magnifications.

It was of interest to compare the effect of temperature on the wettability of samples produced at a relatively low laser scanning speed (200 mm/s) and at a fivefold higher speed (1000 mm/s). In Fig. 2, red diamonds correspond to sample 1 and to samples reported in [16], obtained using the same laser processing parameters as for sample 1 but on substrates preheated to 85, 240, or 360°C, whereas blue circles correspond to samples 2, 3, and 4 produced in the present work (Table 1).

Effect of sample heating temperature during texturing on the contact angle of the resulting superhydrophobic coating: red diamonds correspond to a group of samples produced at a laser scanning speed of 200 mm/s and a pulse repetition rate of 20 kHz [16]; blue circles show the results obtained in the present work at a laser scanning speed of 1000 mm/s and a pulse repetition rate of 100 kHz (samples 2–4).

According to the data reported in [16], at a laser processing speed of 200 mm/s, increasing the sample temperature to 240°C leads to a slight increase in the contact angle, followed by a decrease at T = 360°C. In contrast, at a high processing speed (1000 mm/s), the combination of heating and laser texturing is accompanied by a gradual, slight decrease in the contact angles (Fig. 2). Although laser texturing of a preheated sample preserves the hierarchical nature of the surface texture, its features change. Analysis of SEM images (Fig. 1) of samples 2–4 shows that laser processing of a preheated sample results in a significant coarsening of nanoparticle aggregates forming the coating texture and an increase in the microporosity of the surface layer, which is reflected in an increase in all parameters characterizing surface roughness (see Table 1). The most pronounced changes in both porosity and aggregate size are observed already at a heating temperature of 240°C. At the same time, for all superhydrophobic samples produced in this work, the contact angles of freshly prepared coatings remained no lower than 170°. The observed slight decrease in contact angles when combining surface preheating with accelerated laser scanning, at first glance, appears to contradict literature data. According to these data, an increase in coating roughness, and primarily an increase in the Sratio parameter, should lead to an increase in the contact angle. However, this contradiction is only apparent, and below we provide an explanation of the observed phenomenon.

As follows from the theoretical analysis presented in the seminal work by Marmur [17], the dependence of the contact angle on surface roughness is determined by the wetting regime established upon contact of a water droplet with the surface. In the homogeneous wetting regime, where the liquid penetrates into the depressions of a hydrophobic surface relief, an increase in the surface roughness parameter Sratio leads to a monotonic increase in the contact angle on hydrophobic surfaces, in accordance with the Wenzel equation [18]. In the case of a heterogeneous wetting regime, where the liquid contacts only the tops of surface asperities while the depressions are filled with air, the roughness of the upper parts of the surface textures wetted by the liquid may differ substantially from the overall surface roughness. Under these conditions, an increase in the overall roughness of the surface layer that is not accompanied by an increase in the roughness of the wetted area does not lead to an increase in the contact angle.

Furthermore, another factor influencing wettability should be considered. As theoretically shown in [19], the curvature of surface texture elements plays a decisive role in the magnitude of the local contact angle, which has the meaning of the Young contact angle. For example, the contact angle on hydrophobic nanoparticles is higher than the corresponding angle on a flat smooth surface, and the greater the curvature of the particle surface, the more pronounced the increase in the contact angle at a constant surface composition. Conversely, for pores with a concave surface of negative curvature, the higher the curvature, the smaller the contact angle of the pore wall. The observed effect of heating the textured surface on the decrease in the water contact angle correlates well with an increase in the size of both individual convex particles and their aggregates forming the coating texture (cf. Figs. 1c, 1d with Figs. 1e, 1f and Figs. 1g, 1h). According to the theoretical analysis in [19], an increase in particle size and, consequently, a decrease in particle curvature should lead to a decrease in the contact angle of a hydrophobic textured surface.

Another issue that requires discussion concerns the contact angles of the hydrophobic polished sample 5 and the hydrophobic non-textured sample 6. According to the data in Table 1, the roughness parameters Sa, Sq, and Sz for sample 6 are almost three times higher than the corresponding parameters for sample 5, while remaining at the level of fractions of a micrometer or several micrometers. This is associated with the formation of microscopic rolling marks and scratches. At the same time, the contact angle is mainly influenced by nanoscale roughness, the presence of which would be reflected in the value of Sratio. However, this parameter for both non-textured samples is close to 1 because of the microscopic size of surface defects, which should result in similar contact angles for the polished and unpolished samples. Indeed, the measured contact angle values for sample 6 only slightly exceed the corresponding values measured for sample 5.

As noted above, abrasive wear of surfaces operated under open-atmosphere conditions is one of the main factors leading to the loss of functional properties and, primarily, the superhydrophobic state of textured coatings. Let us consider in more detail how coatings produced at increased laser surface processing speeds, or by combining an increased texturing speed with preliminary substrate heating, behave under conditions of intensive abrasive loading generated by the oscillating sand method.

The main factors affecting the state of superhydrophobic coatings under abrasion are associated with removal of the hydrophobic agent layer, damage to the micro- and nanotexture, and loss of the nanotexture. These factors should lead to a decrease in contact angles and an increase in roll-off angles of water droplets and, as a consequence, may result in the loss of the heterogeneous wetting regime. The coating test results shown in Fig. 3 indicate similar behavior for coatings produced at room and elevated temperatures. A pronounced decrease in the contact angle and an increase in the roll-off angle are observed during the first 15 min of exposure to oscillating sand (Fig. 3a). Further abrasive action is accompanied by a slight gradual decrease in the contact angle and an increase in the roll-off angle. For the roll-off angles of coatings obtained under additional heating conditions, a tendency toward reaching a plateau is even observed (Fig. 3b). It should be noted that the wear resistance of samples textured under additional heating conditions is higher than that of samples processed by laser at room temperature. At the same time, the contact angles measured for all studied samples after 3 h of continuous abrasive loading, which represents a very severe wear test, remained well above 150°. It should also be noted that water droplets on the coatings after the abrasive test freely roll over the surfaces and are removed even at a small tilt angle. The obtained results indicate high stability of the micro- and nanotexture and preservation of the superhydrophobic state of the coatings produced in this work under severe abrasive wear.

Effect of abrasive treatment time on the contact angles (a) and roll-off angles (b).

The increase in the abrasive resistance of textured coatings observed in this study upon additional surface heating followed by laser processing is in good agreement with literature data [16, 20, 21]. As noted in the literature, the enhancement of mechanical properties is associated with several factors, including additional surface oxidation, redistribution of defects in the surface layer, changes in phase composition, and healing of cracks that form under certain laser texturing regimes during additional heating of the material.

Another factor that is of particular importance under the considered additional substrate heating is related to the adhesion of nanoparticles and their aggregates deposited from the laser plume to the substrate. In the case of weak adhesion, when these particles interact with the substrate only via van der Waals forces, various mechanical loads applied to the laser-processed substrate lead to particle removal, destruction of the hierarchical structure, and ultimately loss of the superhydrophobic state. In contrast, at high laser beam energy, for example, at high peak energy flux densities, or under preliminary substrate heating, deposition of nanoparticles onto a hot surface is accompanied by diffusion welding of the particles to the surface. This markedly enhances the adhesion of the hierarchical texture to the substrate and, consequently, makes the coating more resistant to mechanical damage. The results of our experiments, in which an increase in the preliminary substrate heating temperature improves the resistance of superhydrophobic coatings to abrasive wear, directly confirm the above analysis.

Let us now consider the effect of accelerated laser processing or of combining an increased texturing speed with preliminary substrate heating on ice adhesion to the resulting superhydrophobic surfaces. As discussed in the literature [1, 22–25], superhydrophobic coatings are known for their anti-icing properties, which are associated, in particular, with a reduction in the adhesion strength of ice to micro-/nanostructured superhydrophobic surfaces and with an increase in the freezing delay of water deposits on such surfaces. Therefore, a detailed investigation of ice adhesion properties is required to assess the functional potential of new types of laser-fabricated coatings.

In our recent studies [16, 25], it was shown that the shear adhesion strength of ice to rough surfaces decreases sharply with increasing holding time at a given subzero temperature from the moment of ice formation to its detachment. The mechanism of this phenomenon is associated with the formation of metastable ice menisci/bridges inside the depressions of a rough surface during the quasi-adiabatic stage of water crystallization [25–27]. At this stage, the water vapor pressure increases to values exceeding the saturation pressure at the substrate temperature. Prolonged contact of ice with the substrate at a constant subzero temperature, accompanied by relaxation of vapor supersaturation as the system approaches equilibrium, is associated with continuous sublimation of the metastable menisci formed at the quasi-adiabatic crystallization stage. Since metastable menisci form spontaneously in the hierarchical texture of the coating under nonequilibrium conditions, their structure and position vary from experiment to experiment. As a result, measurements of adhesion strength performed shortly after ice formation show a large scatter of values for the same substrate. Moreover, the higher the substrate roughness, the larger the scatter in adhesion strength for substrates of this type (Fig. 4). Therefore, to improve the statistical reliability of ice adhesion strength measurements, for each type of fabricated coating we performed measurements on 20–30 samples of the same coating type at the same holding time after ice crystallization.

Ice adhesion strength to the surfaces of superhydrophobic samples (1, 2, 3), a polished hydrophobic sample (5), and an unpolished hydrophobic sample (6) after 2 and 20 h of exposure at a constant temperature of –10°C.

The values of the shear ice adhesion strength, averaged over all results obtained for the same substrate at two exposure times (2 and 20 h), are presented in Fig. 4.

Analysis of the obtained data shows that for superhydrophobic coatings produced by laser texturing at room temperature and at 240°C, the transition of metastable ice menisci to a near-equilibrium contact between ice and the hierarchically rough substrate is accompanied by a significant decrease in the adhesion strength. At the same time, comparison of the adhesion strength values for samples 1 and 2 produced at room temperature and sample 3 textured at 240°C shows that the adhesion strength after 2 h of exposure increases with increasing surface roughness. This result is consistent with the mechanism discussed above, according to which adhesion increases due to an increase in the real contact area with a rough surface associated with the formation of bridges in surface depressions. It is important to note that after 20 h of exposure, following sublimation of the metastable menisci, the ice adhesion values for samples (1), (2), and (3) are nearly identical and amount to approximately 12 kPa, which is determined by the similar water contact angles of the investigated coatings.

For the non-textured polished hydrophobic sample (5) and the microscopically rough hydrophobic sample (6), the trend is opposite. For these samples, the approach of the system to equilibrium strengthens the ice–substrate contact. Such an effect was mentioned previously in one study [28], where the authors associated it with relaxation of interfacial defects and aging of ice at the interface with a solid substrate. It should also be recalled that smooth hydrophobic surfaces and hydrophobic coatings with microtexture are wetted by aqueous media in the homogeneous regime; that is, ice bridges formed in microdepressions of the surface relief are stable immediately after their formation, and therefore sublimation of such menisci does not occur during prolonged exposure at subzero temperatures.

It is important to emphasize that, in the literature, materials are commonly classified as having anti-icing properties if the ice adhesion strength does not exceed 100 kPa. Among the samples studied here, only the superhydrophobic coatings can be classified as anti-icing, since the characteristic average ice adhesion strength after 2 h of exposure for samples 1, 2, and 3 is 28, 33, and 50 kPa, respectively.

Since the formation and detachment of ice from surfaces are accompanied by significant stresses in the surface layer, which can lead to degradation of the hierarchical roughness, not only the absolute value of the ice adhesion strength but also its variation with an increasing number of ice formation and detachment cycles is important for evaluating the adhesion properties of coatings. To assess the mechanical durability of the coatings fabricated in this work, the dependence of the ice adhesion strength on the number of ice detachment cycles was investigated for coatings 2 and 3. The obtained data are shown in Fig. 5.

Variation of the ice adhesion strength to coatings 2 and 3 (measurements after 20 h of ice exposure at –10°C) during multiple cycles of ice formation and detachment at the same location on the coating surface. The blue and green lines show linear trends of the ice adhesion strength obtained by the least-squares method for coatings 2 produced by texturing at room temperature and coatings 3 processed by laser at T = 240°C, respectively.

The experiments were carried out simultaneously on four samples of each coating; therefore, four data points are presented in the figure for each detachment cycle number. The blue dashed and green dotted lines represent linear trends of the ice adhesion strength obtained by the least-squares method using all measured values for the selected coatings, corresponding to coating 2 produced by texturing at room temperature and coating 3 processed by laser at T = 240°C, respectively. For coating 2, the trend line shows a very small increase in adhesion strength with increasing number of cycles, whereas the slope of the trend line for coating 3 is practically zero. It should be noted that the coefficients of determination for both trend lines (0.08 for coating 2 and 0.0002 for coating 3) are close to zero, which also confirms the absence of coating degradation over the investigated number of ice detachment cycles.

The obtained results allow the conclusion that the coatings produced under accelerated laser surface processing exhibit very high mechanical durability under repeated ice formation and detachment cycles. An important feature of laser processing under accelerated texturing regimes is the higher resistance to degradation during ice detachment for coatings produced at elevated temperatures. It should be noted that data reported for coatings fabricated at higher laser fluences and lower laser scanning speeds [16] indicate the opposite effect, which is apparently associated with the formation of cracks in the surface layer of the material at high irradiation energies and low laser beam scanning speeds.