Wet-jet milling exfoliated hexagonal boron nitride as industrial anticorrosive pigment for polymeric coatings

PIB was provided in the form of Oppanol® N80 (average molecular weight—Mv—800 000) by BASF Italia S.p.A. Toluene (purity >99.7%) was supplied by Merck. N-methyl-2-pyrrolidone (NMP) was provided by ThermoFisher Scientific. Few-layer h-BN flakes were produced in the industrial plant of BeDimensional S.p.A. as previously described in our works [62, 63].

The production procedure of the h-BN/PIB composites was carried out as follow. Firstly, solid PIB was dissolved in toluene with a PIB/toluene weight ratio of 1:8 and stirred for 12 h at 800 rpm and 80 °C until a homogeneous solution was obtained. Then, different mass loadings of WJM-produced h-BN flakes were added into the PIB solution followed by mixing in a Thinky ARE-250 Mixing and Degassing Machine (planetary centrifugal mixer) at 1000 rpm for 5 min to produce h-BN/PIB composite resins with h-BN weight percentage (wt.%) of 2.5, 5, 10 and 20. The as-produced resins were deposited onto cylindrical substrates of structural steel (S355 grade) by doctor blading. Finally, the resulting coatings were dried at room temperature for 1 h, followed by 15 h at 60 °C to evaporate the residual solvent. The thickness of the resulting pristine PIB or h-BN/PIB coatings onto the S355 steel substrates was measured with a Trotec BB20 thickness magnetic induction-based measurement system.

The morphological properties of the WJM-produced h-BN flakes were studied through transmission electron microscopy (TEM) measurements, which were carried out with a JEOL Jem-1011 (Jeol) transmission electron microscope operated at an acceleration voltage of 100 kV. The samples were prepared by drop-casting the produced h-BN dispersions in NMP at a 1:50 weight ratio onto ultrathin C-film on holey carbon 400 mesh Cu grids (Ted Pella Inc). The grids were stored under vacuum at room temperature to remove the solvent residues. The lateral size of the h-BN flakes was measured from TEM images, and the statistical data analysis was performed using OriginPro 2020 on 100 flakes, fitting the data with a log-normal distribution. The thickness of the h-BN flakes and the roughness of the coatings with different h-BN contents were analysed by atomic force microscopy (AFM) measurements performed with an NX10 AFM (Park System) in non-contact mode using a non-Contact Cantilever PPP-NCHR 10 M (resonance frequency ∼330 kHz, force constant 42 N m⁻¹). The samples were prepared by drop-casting the WJM-produced h-BN dispersions diluted 1:100 in isopropyl alcohol onto mica sheets (Ted Pella, Inc.), followed by heating at 250 °C for 20 min to remove the solvent residues. The images were collected on different areas of 35 × 35 µm² (512 × 512 data points), with a scan rate of 0.30 Hz and keeping the working set point at ∼6 nm. Gwyddion (64 bit) software was used to process the images and the height profiles, while the statistical data analysis was carried out on 100 flakes using OriginPro 2020. Thermogravimetric analysis (TGA) measurements were carried out using a TGA Instruments STD650 under an N₂ atmosphere (flow rate: 100 ml min⁻¹) and a temperature ramp of 10 °C min⁻¹ from 30 °C to 700 °C. The x-ray diffraction (XRD) patterns were recorded on a PANalytical Empyrean x-ray diffractometer equipped with a 1.8 kW CuKα ceramic x-ray tube, PIXcel3D 2 × 2 area detector and operating at 45 kV and 40 mA. The diffraction patterns were collected in air at room temperature using parallel-beam geometry and symmetric reflection mode. The scanning electron microscopy (SEM) measurements were acquired using a JEOL JSM-6490LA SEM analytical (low-vacuum) instrument with a thermionic electron gun equipped with a tungsten source after coating the samples with 10 nm of gold layer. The hydrophobicity of the composites was investigated by contact angle (CA) measurements, imaging a 10 μl water drop deposited on the different composites with an OSSILA L2004A1 CA goniometer. The molecular structure of the coatings was evaluated by Fourier-transform infrared (FTIR) analysis. Thin films (thickness of ∼4 μm) of the different composites were deposited by the doctor blading technique on CaF₂ substrates (Crystran). The films were characterised in transmission mode using a PerkinElmer Frontier FTIR spectrometer equipped with an N₂ flow for purging the chamber during the measurements. All the spectra were recorded from 4000 to 600 cm⁻¹ with a 4 cm⁻¹ resolution, accumulating 128 scans to improve the signal-to-noise ratio. A clean CaF₂ substrate was used as a blank. Raman measurements were carried out using a Renishaw InVia micro-Raman spectrometer with a 100× objective (numerical aperture of 0.85). An excitation wavelength of 514.5 nm was used with an incident power of less than 1 mW to avoid sample overheating. Optical microscopy images of the steel substrate before and after the long-term immersion test were obtained using a Leica optical microscope with a 100× objective (N.A. 0.85).

Electrochemical measurements were carried out using a BioLogic VMP3 Multichannel Potentiostat in a three-electrode 1L electrochemical cell at room temperature in a 3.5 wt.% NaCl aqueous solution, following the procedure described in ASTM G5-14 standard. A KCl-saturated Ag/AgCl radiometer Analytical REF201 Red Rod Reference Electrode (Biologic) was used as the reference electrode, whereas a graphite rod was used as the counter electrode. The standard working electrode assembly consisted of a cylindrical sample of steel substrate (0.785 cm² area) coated by PIB or the h-BN/PIB composites, drilled and tapped with a 3-48 UNC thread, and screwed onto the support rod. A polytetrafluoroethylene compression gasket ensures a leak-free seal. The open-circuit voltage (OCV) was monitored for 30 min, after which potentiodynamic anodic polarisation measurements were acquired with a scan rate of 10 mV min⁻¹. The corrosion performance of the coatings was investigated by potentiodynamic anodic polarisation measurements and their Tafel analysis, as described in the ASTM G5-14 standard, for the determination of the corrosion current (i_corr) and the corrosion potential (E_corr) of metal [64, 65]. The linear values of polarization resistance (R_p) were calculated following the ASTM G59-97 protocol from the slope of the polarisation curve at the E_corr point [66], i.e.:

$$\begin{equation*}{R_{\text{p}}} = {\left( {\frac{{{\text{d}}E}}{{{\text{d}}I}}} \right)_{{E_{{\text{corr}}}}}}.\end{equation*}$$

The corrosion rate (CR) value was calculated from i_corr by the Faraday law:

$$\begin{equation*}{\text{CR}} = \frac{{K{W_{{\text{eq }}}}{{\text{i}}_{{\text{corr}}}}}}{{\text{D}}}\end{equation*}$$

in which CR is the corrosion rate (in mm yr⁻¹), K is a constant with a value of 3.27 × 10⁻³, W_eq is the equivalent weight of iron in ferrous compounds (27.9 g eq⁻¹), i_corr is the corrosion current density (in μA cm⁻²) and D is the density of steel (7.85 g cm⁻³) [67]. The inhibition efficiency (η_p) values of the composites are calculated from i_corr by the following equation:

$$\begin{equation*}{\eta _{\text{p}}}\% = \frac{{i_{{\text{corr}}}^0 - {i_{{\text{corr}}}}}}{{i_{{\text{corr}}}^0}} \times 100\end{equation*}$$

in which i⁰ _corr and i_corr are the corrosion current densities in the absence and presence of inhibitors, respectively [68].

Electrochemical impedance spectroscopy (EIS) measurements were carried out in potentiostatic mode at OCV with an AC sinusoidal amplitude of 10 mV over a 0.1 Hz–200 kHz frequency range, following the ASTM G106-89 standard. Cyclic potentiodynamic polarization measurements were conducted at 0.6 V h⁻¹, following the ASTM G61-86 standard guidelines. A long-term immersion test in 3.5 wt% NaCl aqueous solution was conducted for continuous 1000 h following the ASTM G31-72 protocol.

The exfoliated h-BN flakes were produced through WJM exfoliation of bulk h-BN crystals, as described in our previous work [62, 63]. This technique allows the massive production of defect-free and high-quality 2D crystal dispersions, showing a material production rate of up to 1 kg d⁻¹ on a single WJM apparatus [62, 63]. The as-produced h-BN flakes were characterised by TEM and AFM measurements to investigate their morphological properties. The inset to figure 1(a) shows a representative TEM image of h-BN flakes (additional TEM images are shown in supporting information, figure S1) exhibiting rounded edges and having lateral sizes ranging from 25 to 1000 nm approximately. The size distribution follows a log-normal shape, peaking at 105.2 nm (figure 1(a)). The inset to figure 1(b) displays an AFM image of representative h-BN flakes (additional AFM images are shown in supporting information, figure S2).

Zoom In Zoom Out Reset image size

The statistical analysis of the flake thickness indicates that these data follow a log-normal distribution peaked at 4.3 nm, with more than 75% of the flakes showing thickness lower than 20 nm (figure 1(b)). Considering the experimental thickness of an h-BN monolayer (0.09 nm) [69] and an interlayer distance of 0.33 nm [70], the AFM data analysis suggests that the exfoliated h-BN sample mainly consists of few-layer (⩽10) h-BN flakes. The solvent residues in the produced h-BN flakes were determined by TGA analysis. As shown in figure 1(c), a negligible amount of water and NMP (0.6% and 0.8 in wt.%, respectively) remained in the dried exfoliated h-BN powder. The comparison of the TGA analysis of exfoliated h-BN with bulk h-BN demonstrates that the solvent used for the WJM process is effectively removed in the exfoliated h-BN powder. The crystalline structure of the exfoliated h-BN and the starting bulk material was determined by XRD, and the results are shown in figure 1(d). The XRD pattern of exfoliated h-BN reveals a highly crystalline structure, with a characteristic intense peak at 26.7° corresponding to the crystallographic (002) plane of BN [71]. If this peak is compared to the peak of bulk h-BN, it can be seen that both have a similar diffraction angle and shape, indicating that the procedure of exfoliation did not change the crystalline structure of the starting material. The weak peaks observed in both samples at 2θ values of 41.6°, 43.9°, 55.1° and 76.0° correspond to the crystallographic planes of (100), (101), (004) and (110), respectively [72].

The h-BN flakes were embedded into PIB by varying their content from 2.5 to 20 wt.% with respect to PIB, following the method described in section 2.2. The resulting resins were cast onto structural S355 steel substrates, as depicted in figure 2, yielding coatings with a measured average thickness of 58.4 ± 6.8 μm. Based on their h-BN content, the resulting composite coatings are named hereinafter h-BN/PIB 2.5%, h-BN/PIB 5%, h-BN/PIB 10% and h-BN/PIB 20%. An h-BN-free pristine PIB coating was also produced as a reference sample.

Zoom In Zoom Out Reset image size

The morphology of the as-produced h-BN/PIB composites with different h-BN contents was analysed by SEM imaging and compared with the one of pristine PIB. The SEM image of pristine PIB (figure 3(a)) shows a homogenous polymeric structure free of fillers, whereas the SEM images of the h-BN/PIB composites (figures 3(b)–(e)) reveal that the incorporated h-BN flakes are homogeneously distributed within the polymeric matrices in all the composites with different h-BN contents. More specifically, no agglomeration of h-BN flakes is observed anywhere in the polymeric structure, reflecting the successful dispersion of the 2D h-BN nanofillers obtained with the method described in figure 2. The thermal stability of pristine PIB and h-BN/PIB composites was investigated by TGA. Figure 3(f) displays the TGA curves of the different coatings. All of them exhibited similar thermal behaviour, undergoing an abrupt weight loss at around 340 °C due to the decomposition of the PIB matrix. After 430 °C, the weight stabilizes at a constant value, which corresponds to the final residual weight.

Zoom In Zoom Out Reset image size

Table 1 shows the values of the temperature at which the weight loss is 10% (T_−10%) and 50% (T_−50%), which are similar for all the samples, suggesting that the thermal stability of the composites is not affected by the presence of h-BN flakes. The residual weight values approximately correspond to the h-BN weight content in the different composites.

The roughness of the coating surface has a significant influence on the coating hydrophobicity, which is a crucial factor in regulating the final anticorrosion performance [73]. The dependence of the coating roughness on the h-BN content was determined by AFM measurements. As shown in figure 4(a), the pristine PIB sample exhibits a smooth surface with a root mean square (RMS) roughness value of 1.2 nm in an area of 35 × 35 μm². In the case of the h-BN/PIB 2.5% composite, the RMS roughness increases to 10.9 nm due to the occasional presence of h-BN flakes on the coating surface. The surface roughness further increases with raising the h-BN content in the h-BN/PIB 5%, h-BN/PIB 10% and h-BN/PIB 20%, exhibiting RMS roughness values of 41.4, 60.6 and 186.5 nm, respectively. Figure 4(c) shows the plot of the RMS roughness of the coating versus the h-BN content, evidencing a linear correlation between the h-BN content and the surface roughness of the composite coating. In this context, the influence of the h-BN content on the water wettability of the h-BN/PIB composite coatings was studied by water CA measurements. High CA values (>90°) entail superior hydrophobicity, which in turn reduces the penetration of the electrolyte through the coating hindering the diffusion of the electrolyte into the pores or cracks [73]. Figure 4(b) displays images of water drops deposited on the surfaces of the different composite coatings, whereas figure 4(d) shows the measured CA values plotted versus the h-BN content of the coating. Pristine PIB exhibited a CA of 88.3 ± 0.4°, close to the angle corresponding to a hydrophobic material (90°). The incorporation of h-BN flakes into PIB increases the CA beyond 90°. This trend can be ascribed to the presence of hydrophobic h-BN flakes [74, 75]. The h-BN/PIB 10% composite has shown the highest CA (100.5° ± 0.5°), while the increase of h-BN content up to 20 wt.% causes a decrease in the measured CA to 96.6° ± 0.5°. This last behaviour can be attributed to the aggregation of the h-BN flakes in the h-BN/PIB 20% coating, as indicated by the AFM analysis. Consequently, obtaining highly hydrophobic surfaces entails not only a sufficient h-BN content in the composite (e.g. between 2.5 and 15 wt.%) but also a homogenous dispersion of the h-BN flakes on the coating surface. Overall, we observed that h-BN/PIB composites with intermediate h-BN contents (e.g. 5 and 10 wt.%) yield the highest hydrophobicity, suggesting an improvement of the barrier properties against the diffusion of the electrolyte through the protective coating. The integration of h-BN flakes into the PIB matrix was evaluated by FTIR in the wavelength range between 4000 and 600 cm⁻¹. The FTIR spectra measured for the different samples are displayed in figure 4(e). In the PIB spectrum, the characteristic bands appearing at 2961 cm⁻¹ and 2916 cm⁻¹ correspond to asymmetric stretching of CH₃ and CH₂, respectively [76]. The peak at 1471 cm⁻¹ is attributed to the CH₂ groups [76] or –CH bending [77], whereas the double peak at 1389 cm⁻¹ and 1361 cm⁻¹ can be assigned to CH₃ symmetric bending [76]. The peak at 1231 cm⁻¹ is attributed to C–H bending, while the weak peaks at 949 cm⁻¹ and 923 cm⁻¹ correspond to C=C bending due to small amounts of non-polymerised isobutylene [78].

Zoom In Zoom Out Reset image size

The FTIR spectra of h-BN/PIB coatings exhibit differences in the region between the peaks at 1471 cm⁻¹ and 1361 cm⁻¹ compared to the PIB spectrum. In particular, the additional wide band overlapped to PIB peaks can be associated with the in-plane ring B–N stretching vibration (E_u mode) of h-BN at 1396 cm⁻¹ [79]. These results further confirm the effective incorporation of the h-BN flakes into the polymeric matrix. In addition, Raman spectra were also acquired for pristine PIB and all the h-BN/PIB composites to complement the previous measurements, and the results are shown in figure 4(f). The pristine PIB Raman spectra displayed in figure 4(f) reveal a wide peak in the 2810 and 3060 cm⁻¹ region with a maximum intensity at 2914 cm⁻¹ as the main feature, which corresponds to the combination of symmetric and asymmetric carbon–hydrogen modes [80]. In addition, other minor peaks can be observed, especially those appearing at 717 cm⁻¹ and 925 cm⁻¹, corresponding to the methylene rocking mode, γ_r(CH₂) [80], and out-of-plane CH deformation [81], respectively. In the pristine h-BN and h-BN/PIB composites spectra, a sharp peak at 1365 cm⁻¹ can be observed, which is ascribed to the E_2g phonon mode characteristic of h-BN [82]. This peak is analysed in figure 4(g), being absent in the pristine PIB curve, but gradually increases in intensity in the h-BN/PIB composites as the h-BN content rises. The wide peak at 2914 cm⁻¹ characteristic of the PIB spectra previously mentioned, is shown in figure 4(h). It is seen that it reaches the maximum intensity for pristine PIB, whereas it decreases for the h-BN/PIB composites as the h-BN content increases up to 5 wt.%, remaining nearly constant beyond this value.

The characteristic parameters that define the corrosion performance of the coatings can be determined by potentiodynamic anodic polarisation measurements and their Tafel analysis, following the ASTM G5-14 and ASTM G59-97 standards, as described in section 2. Figure 5(a) shows the anodic polarisation curves measured for the investigated h-BN/PIB composite coatings, pristine PIB coating and bare structural steel substrate without any coating. The most relevant metrics calculated from these plots are listed in table 2.

Zoom In Zoom Out Reset image size

All the coatings protect the bare structural steel against corrosion. More in detail, the measured E_corr of structural steel shifts from −0.604 V versus Ag/AgCl in bare steel to values ranging between −0.020 and 0.187 V, indicating the protective role of the polymeric coatings on the metallic surface. Importantly, h-BN/PIB composite coatings exhibit E_corr higher than pristine PIB, reflecting the additional anticorrosion properties of the coating upon the incorporation of h-BN flakes. However, the E_corr of h-BN/PIB composite coatings varies randomly with the h-BN contents. Differently, i_corr decreases as the h-BN content rises until reaching a minimum of 6.4 × 10⁻⁴ μA cm⁻² for the h-BN/PIB 5%. By further increasing the h-BN content above 5 wt.%, i_corr increases. A similar behaviour for i_corr has been previously reported for a PPy-intercalated graphene anti-corrosion system, in which the addition of a proper amount of graphene (0.5 wt.% with respect to polymer) into a PPy matrix reduced the intrusion of corrosive species, whereas an excessive amount of graphene (>0.5 wt.%) caused the formation of aggregates acting as local defects of the coating [83]. As shown in the plot of R_p versus h-BN (figure 5(b)), the h-BN/PIB 5% composite exhibited the maximum R_p value of 6.9 × 10⁷ Ω cm².

Figure 5(c) shows the CR values calculated for each structural steel/coating system plotted versus the h-BN content. As expected from the i_corr analysis, the h-BN/PIB 5% composite coating exhibited the lowest CR of 7.4 × 10⁻⁶ mm yr⁻¹ among the investigated samples, corresponding to an outstanding η_p value of 99.999%. As shown in table 3, the CR value mentioned above is inferior to those obtained by other 2D material-based coatings reported in recent literature with comparable thicknesses. Moreover, cyclic voltammograms, measured following the standard ASTM G61-86 protocol, indicate that pitting corrosion does not take place for all the investigated coatings (figure 5(d)). In practice, EIS measurements are widely used to investigate the interfacial and bulk electrochemical properties of the materials, obtaining relevant physicochemical parameters in the field of corrosion [84]. Therefore, our structural steel/coating systems were evaluated through EIS measurements, following the ASTM G106-89 standard. The EIS data can be represented in Nyquist plots or Bode diagrams. In the first ones, the negative value of the imaginary part of the impedance (−Z'') is plotted versus the real part of the impedance (Z'), yielding characteristic semicircles whose amplitude at high frequencies correlates to the coating barrier properties of each composite [91]. The Bode plots consist of two different graphs, one displaying the logarithm of the impedance modulus (log |Z|) versus the logarithm of the frequency (log (Freq)), and the other representing the phase angle versus log (Freq). The EIS data measured for a system undergoing corrosion can be fitted with the equivalent electrical circuit displayed in the inset to figure 5(e), representing the diffusion impedance of the solution at the interface with the polymeric coating [92]. In the equivalent electric circuit of the coating system, R_s represents the solution resistance, Q_c is the coating capacitance, R_c is the coating resistance towards the ionic species or reactants passing through the coating, Q_dl is a constant phase element representing the double-layer capacitance and R_t is the electron migration resistance [92]. In the Bode plot, the impedance modulus |Z| at low frequency (∼0.1 Hz) is associated with R_c, which can be therefore easily determined [92, 93]. Figure 5(e) shows the Nyquist plots measured for pristine PIB- and h-BN/PIB composite-coated structural steel. The Nyquist plots show the presence of a distorted semicircle in the high-frequency region, which reflects the coating barrier properties. In particular, a bigger radius of the semicircle implies a superior anti-corrosion performance [13]. Thus, the Nyquist plots confirm the trend observed in the Tafel analysis. Namely, the semicircle radius of the Nyquist plot increases when the h-BN content rises from 0 to 5 wt.%, implying the maximum corrosion resistance for the h-BN/PIB is 5%. By increasing the h-BN content from 5 wt.% to 20 wt.%, the semicircle radius dramatically decreases, indicating a reduction of the corrosion-protecting ability of the coating. The enhancement of the anticorrosive properties in the h-BN/PIB composites compared to pristine PIB is associated with the physical barrier expressed by h-BN flakes, which increases the length of the electrolyte diffusion pathways through the polymeric coating, as illustrated in figure 6. Significantly, h-BN does not suffer the galvanic effect reported for graphene, whose barrier properties have been also widely investigated [94].

Zoom In Zoom Out Reset image size

More in detail, although graphene was initially proposed as an anticorrosive coating due to its impermeability and chemical stability, the presence of pinholes or scratches in practical situations leads to accelerated local corrosion due to galvanic processes taking place at the junction between the graphene and the metal. In this particular case, graphene behaves as the cathode whereas the steel acts as the anode, being the latter preferentially attacked upon exposure to the corrosive media [32]. In contrast, h-BN flakes behave as a physical barrier while acting as electronic and ionic insulators due to their ionic-free structure and absence of conjugation, resulting in an ideal functional filler for anticorrosion polymeric coatings, even in absence of traditional sacrificial inorganic fillers (e.g. Zn and Mg) [39].

To investigate the behaviour of the protective coatings during long-term operation, structural steel coated with h-BN/PIB 5% composite and pristine PIB were immersed in 3.5 wt% NaCl aqueous solution over 1000 h while monitoring their E_corr values following the protocol described in the ASTM G31-72 standard. As shown in figure 7, the E_corr at the beginning of the immersion test was around 0.09 V versus Ag/AgCl and 0.03 V versus Ag/AgCl for structural steels coated with h-BN/PIB 5% and pristine PIB, respectively. Despite the presence of E_corr fluctuations, both systems remained stable over time. After the 1000 h immersion test, the measured E_corr values were approximately 0.08 V for structural steels coated with h-BN/PIB 5% composite and 0.04 V for pristine PIB.

Zoom In Zoom Out Reset image size

To further assess the effect of the immersion on the structural integrity of the steel substrate, optical microscopy images were taken on the samples after delaminating the coatings (insets, figure 7). After 1000 h of immersion in 3.5 wt.% NaCl aqueous solution, the surface of the structural steel substrate was extremely corroded (figure 7(b)) compared to the same sample before the immersion (figure 7(a)). On the contrary, pristine PIB coating provides clear protection against structural steel substrate oxidation (figure 7(c)). Nevertheless, there are still areas that show the corrosion of the metallic substrate. In contrast, the substrate coated with h-BN/PIB 5% reveals a surface free of oxidation products (figure 7(d)), which is similar to the surface of the bare substrate before the immersion test. This observation further confirms that the composite coating optimally protects the underlying structural steel against corrosion, whereas the inferior anticorrosion performance of PIB coating may still allow some amount of corrosive species to reach the surface of the substrate.

The evolution of the anticorrosive properties of both systems during the immersion test was evaluated by EIS measurements. Figure 8 presents the Bode plots of pristine PIB and h-BN/PIB 5% coated structural steel measured over 26 d of immersion. Figures 8(a) and (c) shows that the bare steel substrate exhibits the lowest |Z|_{0.1 Hz} value of ∼10³ Ω cm². The initial |Z|_{0.1 Hz} values measured for the pristine PIB- and h-BN/PIB 5%-coated structural steel, obtained after a preconditioning time of 2.5 h, were 5.6 × 10⁵ Ω cm² and 3.5 × 10⁶ Ω cm², respectively, reflecting the superior anticorrosion performance of the composite coating. During the first ten days, the evolution of the Bode plots indicates the occurrence of some change related to the anticorrosion coatings. The changes in the Bode plots for anticorrosion systems during the first days of immersion have been often observed in literature for anticorrosion coatings [13, 83, 95, 96]. However, such changes are not necessarily associated with a loss of the protective efficiency of the coating if high |Z|_{0.1 Hz} are still shown. The failure of an anticorrosion coating is instead evidenced by a drastic change of the E_corr of the system towards the E_corr exhibited by the bare substrate (e.g. −0.604 V versus Ag/AgCl in bare steel).

Zoom In Zoom Out Reset image size

In addition, the failure of the coating is typically associated with a continuous change of |Z|_{0.1 Hz} over the subsequent days until approaching the values expressed by the unprotected substrate under corrosion. In this work, the monitored E_corr of the PIB-based coatings remained nearly constant over 1000 h of continuous immersion. After two days of immersion, h-BN/PIB 5%-coated structural steel has still shown a |Z|_{0.1 Hz} as high as 6.2 × 10⁵ Ω cm², which is superior to the values exhibited by the PIB-coated structural steel. After ten days of immersion, the systems based on pristine PIB and composite coatings showed |Z|_{0.1 Hz} values of 1.7 × 10⁴ Ω cm² and 2.9 × 10⁴ Ω cm², respectively. After more than ten days, both systems did not experience any |Z|_{0.1 Hz} change, indicating optimal anticorrosion performance over time.