PH-Sensitive Self-Healing Waterborne Epoxy Coating: Polydopamine Grafted Graphene Oxide with 2-Mercaptobenzimidazole Inhibitors and Cerium Cation Towards Anti-Corrosion Performance

Abstract

The development of smart coatings with active protection is a promising approach to prolonging the service life in extreme environments. Herein, the corrosion inhibitors 2-mercaptobenzimidazole (MBI) and CeO₂ were in situ loaded onto the surface of graphene oxide (GO) by dopamine (DA) polymerization, and we ultimately obtained the multifunctional composite MBI@CeO₂@PDA@GO (MCPG). The electrochemical impedance spectroscopy (EIS) results revealed that after 30 days of immersion in the corrosive media, the |Z|_0.01 Hz value of MCPG/WEP coating remained at 3.7 × 10⁹ Ω/cm², which displayed four orders of magnitude higher than that of pure WEP coating (1.4 × 10⁵ Ω/cm²). In a 200 h salt spray test, the MCPG/WEP coating also demonstrated minimal corrosion products and bubbles, affirming the exceptional corrosion-inhibiting effect and excellent self-healing performance. Consequently, the synergistic combination of pH-sensitive properties and outstanding barrier effect imparted dual active/passive anti-corrosion capabilities to the coating, resulting in long-lasting metal protection.

1. Introduction

Applying epoxy resin (EP) coatings is an effective method of corrosion protection for metals [1,2]. Driven by the goals of environmental protection and green production, waterborne epoxy coatings (WEPs) have increasingly supplanted the organic epoxy resin in anti-corrosion coatings [3,4]. However, WEP coatings offer merely passive protection and the inevitable formation of micro-pores and defects during the curing process, which in turn reduce its overall protective efficacy for long-term services [5,6].

To address these issues, the incorporation of corrosion inhibitors is a promising strategy to provide active corrosion protection and extend the service life of WEP coatings [7,8]. For example, both metallic ions (such as calcium [9], zinc [10] and rare earth cerium ions [11]) and imidazole derivatives (including benzotriazole, BTA [12]; 2-mercaptobenzimidazole, MBI [13], etc.) have all demonstrated effective corrosion inhibition. Thereinto, cerium-based inhibitors exhibited high oxygen affinity to reduce the dissolved oxygen concentration [14], whereas imidazole derivatives form a protective film by coordinating directly with the metal surface [15]. Lin et al. synthesized imidazoline inhibitors with different carbon chain lengths and doped them directly into cellulose-based gel coatings. The results demonstrated that composite coatings exhibited excellent anti-corrosion and self-healing performance [16]. However, directly importing inhibitors into resin introduces several challenges, such as uncontrollable inhibitor release, coating internal structure damage, and reduced service life [17,18]. To mitigate these adverse interactions and enable a controlled release, the encapsulation of corrosion inhibitors is the most promising strategy for active coatings [19].

Inspired by the composition of adhesive proteins in mussels [20,21], dopamine (DA) as a hormone and neurotransmitter is widely distributed in the natural environment and can easily self-polymerize on substrate surfaces to form the polydopamine (PDA) layer with excellent adhesion [22]. In addition, PDA exhibits anti-corrosion performance through forming catechol-Fe³⁺ complexes and preventing the diffusion of corrosive media, further highlighting its significant potential as eco-friendly inhibitor [23]. Li et al. developed a PDA-GO-CeO₂-EP coating by anchoring CeO₂ to GO surface via PDA polymerization. This composite coating provided active-passive corrosion protection through a synergy between the physical barrier of GO sheets and the inhibition effect from CeO₂ [24]. Ji et al. designed a smart carrier BMPDA-MBI@PDA (M-M@P) and incorporated into a waterborne epoxy (WEP) coating. The WEP/M-M@P composite coating exhibited strong barrier inhibition protection in acidic conditions duo to its high inhibitor capacity and pH responsiveness [25]. However, most studies targeted individual material properties rather than pursuing an integrated strategy to enhance overall corrosion resistance. In particular, few researchers focused on the potential synergistic protective effects of combining cerium with imidazole derivatives in coatings remain.

In this work, a novel smart coating was fabricated by incorporating MBI-CeO₂@PDA@GO (MCPG) into the WEP, which was then applied to Q235 steel. PDA was used to combine the MBI, CeO₂ with GO sheets, thereby improving the dispersion of nanosheets in aqueous epoxy resin. Meanwhile, MBI and cerium ion were employed to provide the coating with active self-healing capability. The structure, morphology, loading capacity and sustained release behavior of the fabricated materials, anti-corrosion and self-healing performance of composite coatings were comprehensively investigated. Consequently, the MCPG/WEP coating exhibited significantly enhanced anti-corrosion performance.

2. Experimental Section

2.1. Raw Materials

Graphene oxide (GO-325) was provided by Institute of Coal Chemistry (Taiyuan, China). 2-Mercaptobenzimidazole (2-MBI, 98 wt.%), tris-hydrochloride buffer (pH 8.5), dopamine hydrochloride (C₈H₁₁NO₂·HCl, 98 wt.%), Cerium (III) nitrate hexahydrate (Ce(NO₃)₃·6H₂O, 99.95 wt.%) and methanol (CH₃OH, AR) were obtained from Shanghai Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). Waterborne epoxy resin emulsion (MU-618) and its matched curing agent (CU-600) were sourced from Shanghai Runtan New Material Technology Co., Ltd. (Shanghai, China). Q235 steels were obtained from Jixing Stainless Steels Factory (Beijing, China). Before coating, they were pretreated by a sandblasting machine, then washed with acetone, and finally dried at 60 °C.

2.2. Synthesis of MCPG

2-Mercaptobenzimidazole (MBI) and CeO₂ co-modified graphene oxide (MBI@CeO₂@PDA@GO, MCPG) was prepared by the following method (as shown in Figure S1): Firstly, 100 mg of GO was dissolved in 25 mL Tris-HCl buffer and 25 mL methanol solution, the mixture solution was subjected to ultrasonication for 30 min. Secondly, 100 mg of DA, 50 mg of Ce(NO₃)₃·6H₂O, and 50 mg of MBI were added sequentially and stirred at 50 °C for 24 h. Thirdly, the resulting MCPG composites were centrifuged, washed and ultimately obtained after freezing drying. The same procedures were employed to prepare the PDA@GO (PDGO) composites.

2.3. Fabrication of Composite Coatings

Composite coating with MCPG (0.5 wt.%, based on solid weight of resin and curing agent) was fabricated as follows: Firstly, 40.8 mg of MCPG was dispersed in 10 mL DI water and sonicated for 10 min. The dispersion was combined with 10 g of waterborne epoxy emulsion (MU-618) and 5 g of curing agent (CU-600) and then stirred for 30 min. Finally, the prepared coatings were sprayed onto the pretreated Q235 steel and cured at 40 °C for 48 h. The resultant coating thickness was controlled at 40 ± 5 μm, and these coatings were noted as MCPG/WEP. For comparison, pure WEP, GO/WEP (0.1 wt.%, based on our previously reported [26]) and PDGO/WEP (0.5 wt.%) coatings were prepared by using a similar method.

2.4. Characterizations

X-ray diffraction (XRD, Bruker D8 Advance, Karlsruhe, Germany) was used to determine the crystal structures of the samples. The chemical composition was further obtained by X-ray photoelectron spectrometer (XPS, PHI5300, Waltham, MA, USA) with an Al Kα radiation at an energy of 150 W, and in a vacuum of 10⁻⁷ Pa. Thermogravimetric analysis (TGA, Mettler TGA2, Zurich, Switzerland) was performed from 30 to 900 °C at a heating rate of 10 °C/min under a nitrogen atmosphere. The morphology and microstructure of GO, PDGO and MCPG sheets were investigated using scanning electron microscopy (SEM, ZEISS GEMINISEM 500, Oberkochen, Germany) and transmission electron microscopy (TEM, JEOL JEM-2100F, Akishima, Japan), equipped with energy dispersive spectroscopy (EDS). The release behavior of MBI was investigated using ultraviolet–visible spectrophotometer (UV-vis, Shimadzu, UV-2450, Kyoto, Japan) and Zeta potential analyzer (Zetasizer Nano ZS90, Malvern, UK).

2.5. Release Behavior of Inhibitors from MCPG Composites

To examine the release behavior of MBI in different conditions, 5 mg of MCPG was dispersed in 50 mL of 3.5 wt.% NaCl solution. The pH of solution was adjusted to 3, 5, 7, 9 and 11 using H₂SO₄ or NaOH. After soaking for different periods, 200 µL of the solution was drawn up and diluted tenfold with ethanol for measurement using the UV-vis spectrophotometer. The release of MBI was measured by the absorbance values at a wavelength of 247 nm. MBI concentration-absorbance curves were established by measuring the absorbance values of standard solutions (with a series of concentrations 4, 6, 8, 10, 15, 20 ppm), with the concentration (ppm) of MBI as X-axis and Y as the absorbance values at a wavelength of 247 nm.

2.6. Corrosion and Self-Healing Performance of MCPG/WEP

The anticorrosion performance of composite coatings was investigated using electrochemical impedance spectroscopy (EIS) and potentiodynamic polarization curves. These tests were performed on an electrochemical workstation (Biologic, VSP-300, Grenoble, France) equipped with a standard three-electrode configuration. In this configuration, the Q235 steel with coating (exposing a circular area of 1 cm²) was the working electrode. Platinum (2 × 2 cm) and saturated calomel (SCE) were the counter electrode and reference electrode, respectively. The electrolyte was 3.5 wt.% NaCl neutral solution (pH = 7), which simulated a seawater environment. The coatings were first subjected to an open circuit potential (OCP) method lasting 600 s to obtain stabilized potential. Subsequently, EIS tests were conducted with an amplitude of 20 mV at frequency ranged from 10⁵ to 0.01 Hz. The ZView software (Version 3.1) was used to fit the equivalent circuit of the experimental data. Similarly, EIS serves as an effective method for evaluating the self-healing properties of coatings. To simulate the mechanical damage, MCPG/WEP samples were scratched with dimensions of 1 cm using a scalpel as the working electrode, and impedance evolution was measured over time to evaluate the self-healing behavior of coatings.

Potentiodynamic polarization curves were plotted from OCP of −250–250 mV with a sweep rate of 1 mV/s. Anodic Tafel slope (βa), cathodic Tafel slope (βc), corrosion potential (E_corr) and corrosion current density (I_corr) were acquired via Tafel extrapolation fitting. The inhibition efficiency (η) [27] was used as follows:

$$\eta ={\displaystyle \frac{I_{corr}^0-I_{corr}}{I_{corr}^0}}$$

(1)

where $I_{corr}^0$ is the corrosion current density of Q235 steel coated with pure WEP resin, $I_{corr}$ is the corrosion current density of coatings composed of nanoparticles.

The mechanical properties of composite coatings were studied by adhesion tests. The adhesion strength between Q235 steel substrate and coating was obtained via the pull-off adhesion test (SH-M, Jingkelian Materials Testing Machine Co., Ltd., Tianjin, China). The dry and wet adhesion strengths were measured after drying and after immersion in 3.5 wt.% NaCl solution for 7 days, respectively.

Salt spray tests were conducted according to the standard ASTM B117-09. The scratched specimens were placed in a salt spray chamber (YWX-015, Changzhou Guoli Test Equipment Co., Ltd., Changzhou, China) with 50 g/L atomized NaCl solution for 200 h, and the test temperature was maintained at 35 °C.

3. Results and Discussion

3.1. Characterization of MCPG

XRD was used to characterize the crystal structure and interlayer spacing of GO, PDGO and MCPG, as shown in Figure 1a. GO exhibited a sharp (001) diffraction peak at 10.97°, attributed to the graphite oxidation [28], corresponding to an interlayer spacing of 0.81 nm. After polymerization of DA, the peak of PDGO shifted towards the left (9.05°), confirming that PDA intercalated between the GO layers [29]. And the (001) peak of the MCPG composite shifted to 8.80° with reduced intensity and a widened peak, corresponding to the interlayer spacing of 1.01 nm, indicating the successful loading of inhibitors and PDA molecules into the GO layers, which could facilitate a higher loading capacity of inhibitors.

Figure 1. (a) XRD patterns and (b) XPS spectra of GO, PDGO and MCPG; (c) S2p and (d) Ce3d deconvoluted spectra of MCPG; (e) TGA curves of GO, PDGO and MCPG.

XPS characterization was used to further analyze the elemental composition and chemical structure. The survey spectra of GO, PDGO and MCPG in Figure 1b revealed that MCPG exhibited additional characteristic peaks for S2p and Ce3d at 163 eV and 885 eV, confirming the successful loading of MBI and CeO₂ [13,30]. Figure 1c displayed the high-resolution S2p spectrum, and the overlapping peaks at 163.8 and 165.0 eV corresponded to the 2p_3/2 and 2p_1/2 orbitals of C-S-C, respectively [31], indicating that the C-SH group of MBI participated in reactions to form a new C-S-C group, consistent with the FT-IR results (as shown in Figure S2a). The high-resolution Ce3d spectrum in Figure 1d showed the peaks corresponding to Ce³⁺ (885.1 and 904.1 eV) and Ce⁴⁺ (882.2, 897.0, 900.9, and 906.7 eV) [32]. Typically, the Ce³⁺/Ce⁴⁺ ratio is a key factor influencing the anti-corrosion performance and stability of the coating [33]. MCPG composites exhibited a higher ratio (Ce³⁺/Ce⁴⁺ = 1.75), which can undergo oxidation reactions. This process generated oxygen vacancies that provide active sites for sequestering corrosive agents, enhancing the protective resistance of the coating.

Thermogravimetric analysis (TGA) was used to evaluate the thermodynamic stability of the composites. Figure 1e illustrated the TGA curves of GO, PDGO and MCPG, respectively. Indicating that the mass of the three samples decreased with the increase of temperature, owing to the evaporation of moisture below 100 °C and the decomposition of oxygen-containing functional groups between 150 and 300 °C [34]. It is worth noting that the MCPG exhibited a flat weight loss curve at 320 °C, which was similar to the TGA curve of MBI. A rapid weight loss (around 87.5%) was observed between 200 °C and 400 °C (as shown in Figure S2b) due to the weak chemical bonds of the imidazole ring structure [35]. The results demonstrated that MBI molecules were chelated by PDA between 0GO layers. In addition, the residual weight of MCPG was 61.2%, which was higher than that of GO (37.2%) and PDGO (54.8%), showing that MCPG exhibited enhanced thermal stability.

The structural morphology of GO, PDGO and MCPG was characterized using SEM and TEM. As shown in Figure 2(a1,b1), the surface of GO sheets presented a relatively smooth texture with characteristic undulating folds. After the polymerization of DA, the resultant PDGO exhibited increased undulating folds (Figure 2(a2,b2)), which was beneficial to improve the dispersion within epoxy resin coatings. In contrast, the MCPG composites displayed a more irregular and coarse surface topography (Figure 2(a3)), with reduced transparency and the presence of visible microparticles (as shown in Figure 2(b3)). Furthermore, elemental mapping by EDS further confirmed the uniform O, N, S, and Ce elemental distribution (Figure 2c), suggesting the successful loading of MBI and CeO₂ in MCPG nanofillers.

Figure 2. (a1–a3) SEM, (b1–b3) TEM images of GO, PDGO and MCPG; (c) EDS mapping of MCPG.

3.2. Release Behavior of MCPG

The loading amount of MBI in MCPG was determined using UV–vis spectroscopy tests. Before testing, the standard equation was established by measuring the UV absorbance at 247 nm and the MBI concentration (Figure 3a). MCPG nanoparticles were immersed in 3.5 wt.% NaCl solution and sonicated for 2 h to accelerate the release of MBI. Figure 3b indicated that the loading capacity of MBI was 32.8 wt.%. In addition, PDA could also provide pH-responsive control over MBI release. The efficiency of MBI molecule release from the solution at pH levels of 3, 5, 7, 9 and 11 was evaluated, as shown in Figure 3c. The MCPG corrosion inhibitor carrier exhibits a consistently low release rate at pH of 7 and 9 (about 14.8 and 5.9%, respectively) after immersion for 30 min, illustrating that the adhesive PDA could effectively suppress the release of MBI, indicating that PDA layers effectively retained the inhibitors, thereby ensuring that the MBI molecules remain stored within the MCPG carrier to prevent premature release and subsequent failure. Notably, acidic and strongly alkaline environments promoted higher release rates, and the MCPG corrosion inhibitor carrier demonstrates a release efficiency of 46.3% and 26.9% at pH levels of 3 and 11, respectively, showing efficient and rapid pH responsiveness. These findings suggest that the MCPG corrosion inhibitor is capable of promptly releasing a sufficient quantity of MBI molecules to safeguard the metal substrate against corrosion in acidic and alkaline corrosive conditions.

Figure 3. (a) UV-vis absorption spectra of MBI and concentration-absorption standard equation, (b) UV-vis absorption spectra of MCPG, (c) 10 h release curves of MBI and (d) Zeta potential of MCPG dispersion at different pH levels.

The Zeta potential of MCPG was also monitored under varying pH conditions (Figure 3d). At pH = 7, the partial dissociation of hydroxyl groups conferred a negative charge, resulting in a stable potential of approximately −40 mV. In acidic conditions (pH = 3), protonation of amino groups produced a positive charge, yielding an initial potential of +60.4 mV. Owing to the release of positively charged BTA⁺ ions, which reduced the net positive charge of the system, the potential rapidly decreased to +33.5 mV. Conversely, under alkaline conditions (pH = 11), the dissociation of hydroxyl groups produced an initial potential of −72.1 mV and increased to −60 mV. It was attributed to the instability and decomposition of PDA under strong alkaline conditions [36], resulting in the number of negatively charged hydroxyl groups gradually reduced. These findings were consistent with the UV–vis spectroscopy results. Collectively, these results confirmed the robust pH-responsive release behavior of the inhibitors from MCPG nanofillers, establishing them as a promising material for developing smart active protection coatings.

3.3. Characterization of Waterborne Epoxy Composite Coatings

3.3.1. Fracture Surface Morphology

The cross-sectional morphologies of the composite coating (pure WEP coating, GO/WEP, PDGO/WEP and MCPG/WEP composite coatings, as shown in Figure 4) were evaluated by SEM to analyze the dispersion degree of composite materials within the waterborne epoxy resin matrix. Typically, the fracture surface of WEP showed brittle and ductile fractures and exhibited noticeable pore defects, as shown in Figure 4a, which were caused by rapid solvent evaporation during the curing [37]. These micropores significantly impair the protection properties of coatings [38]. The fracture surface of GO/WEP exhibited smooth, brittle fractures with numerous micropores in Figure 4b, indicating poor dispersion of GO within the resin coating. After the addition of PDGO, the fracture surface became rougher and exhibited a clear reduction in coating defects, but agglomeration of the nanofillers was observed, as shown in Figure 4c. This agglomeration phenomenon of the coating was greatly reduced by the modification of MBI and CeO₂, and the micropores of the epoxy coating were also significantly reduced (Figure 4d), demonstrating superior dispersion within the resin matrix. The above experimental results illustrated the effective compatibility of MCPG with WEP, which can significantly enhance the crosslinking density and barrier properties of the coating, further improving its anti-corrosion performance.

Figure 4. The fracture surfaces of the composite coatings: (a) pure WEP, (b) 0.1 wt.% GO/WEP, (c) 0.5 wt.% PDGP/WEP and (d) 0.5 wt.% MCPG/WEP.

3.3.2. Anti-Corrosion Performance of Composite Coatings

EIS tests are widely used to investigate the anti-corrosion performance of the coatings. Figure 5 presents the corresponding Nyquist and Bode plots of pure WEP coating, GO/WEP, PDGO/WEP and MCPG/WEP composite coatings during immersion in 3.5 wt.% NaCl solutions for 30 days. In general, a larger capacitive arc diameter in the Nyquist plots indicated superior corrosion resistance [38,39]. As shown in Figure 5(a1), the semicircular impedance arc of the WEP coating significantly decreased with the extension of immersion time from 2 to 30 days, suggesting the progressive degradation of the protection ability. Consequently, after 30 days, the WEP coating no longer provided effective protection for the substrate, and the corrosive medium had penetrated to its surface. By comparison, GO/WEP coating exhibited a bigger semicircular impedance arc at the beginning of immersion and then also rapidly decreased (as shown in Figure 5(b1)) during immersion, indicates that the passive barrier properties of GO sheets cannot provide long-term protection capability [20]. In contrast, the MCPG/WEP coating maintained the maximal arc diameter, demonstrating superior protective performance for Q235 steel (Figure 5(d1)).

Figure 5. Nyquist and Bode plots of (a1,a2) pure WEP, (b1,b2) GO/WEP, (c1,c2) PDGO/WEP and (d1,d2) MZPG/WEP after 2, 7, 15, and 30 days immersion in 3.5 wt.% NaCl solution.

The impedance modulus at low frequency (|Z|_0.01 Hz) as a semi-quantitative evaluation index can also represent the protective capacity of the coatings [40]. The corresponding data were presented in Figure S3. The initial |Z|_0.01 Hz value of the WEP coating (1.7 × 10⁷ Ω/cm²) and GO/WEP coating (1.5 × 10⁸ Ω/cm²) significantly decreased to 1.4 × 10⁵ Ω/cm² and 3.1 × 10⁵ Ω/cm² after 30 days of immersion, indicating that the coatings lost corrosion resistance performance. However, the resistance values of the PDGO/WEP coating and MCPG/WEP coating remained at 1.4 × 10⁸ and 3.7 × 10⁹ Ω/cm² after 30 days of immersion. And the resistance values of MCPG/WEP after 15 days (6.3 × 10⁹ Ω/cm²) slightly increased compared to those after 7 days (6.5 × 10⁹ Ω/cm²). The variation of |Z|_0.01 Hz resulted from the healing effects of the inhibitors, suggesting that the MCPG/WEP coating possessed stable and robust anti-corrosion performance.

To further evaluate the differences in protective performance, the EIS data were fitted using representative equivalent circuit models in Zview software (Version 3.1). Two models (Figure 6a) were employed to simulate the coatings’ behavior during initial and long-term immersion stages. For Model I (first stage), R_s, R_c and CPE_coat represent the solution resistance, coating resistance and coating capacitance, respectively. Model II represents the stage when the corrosive medium has penetrated the coating/substrate interface. R_ct and CPE_dl represent the charge transfer resistance and double-layer capacitance [41]. The CPE_coat, CPE_dl and n (dispersion coefficient) were used to replace the pure capacitance, and the coating capacitance (Q_c) could be calculated via the following equation [42]:

$$Q_C={\left({CPE}_{coat}\times R_c^{1-n_1}\right)}^{{\displaystyle \frac{1}{n_1}}}$$

(2)

Figure 6. (a) Corresponding equivalent electrical circuit models for EIS fitting; (b) |Z|_0.01 Hz, (c) R_c and (d) water absorption rate variation of different coatings during immersion.

The fitted parameters were presented in Table S1, and key values were drawn as histograms in Figure 6b,c. R_c serves as a key indicator of the coating’s protective resistance against corrosive media, decreased with the increase of immersion time, indicating the deterioration of the coating [43]. The evolution of the coating resistance was assessed by monitoring the changes of R_c value. As shown in Figure 6b, the MCPG/WEP coating exhibited the highest R_c value, demonstrating excellent barrier properties. Furthermore, the self-healing potential can be further supported by the water absorption rate (X_v%) calculated by the coating capacitance via the Brasher and Kingsbury equation [44]:

$${\mathit{X}}_{\mathit{v}}\%={\displaystyle \frac{\mathit{l}\mathit{o}\mathit{g}\left(\frac{{\mathit{Q}}_{\mathit{C}}\left(\mathit{t}\right)}{{\mathit{Q}}_{\mathit{C}}\left(\mathbf{0}\right)}\right)}{\mathit{l}\mathit{o}\mathit{g}\left(\mathbf{80}\right)}}\times \mathbf{100}\%$$

(3)

The MCPG/WEP composite coating presented the lowest water absorption rate after 30 days of immersion, suggesting its superior barrier properties (Figure 6c). These findings indicated that the MCPG/WEP coating exhibited a stable and enduring anti-corrosion performance toward the metal substrate.

Figure 7a displays the polarization curve of composite coatings, which were immersed in 3.5 wt.% NaCl solution for 2 days. Table 1 displays the relevant electrochemical parameters via Tafel extrapolation fitting. The WEP exhibited an I_corr of 1.51 × 10⁻³ μA/cm², while the I_corr value of GO/WEP, PDGO/WEP, and MCPG/WEP composite coatings decreased by about one, two and three orders compared to that of the WEP coating. Therefore, MCPG/WEP had the lowest corrosion rate (v_corr) of 2.18 × 10⁻⁸ mm/year, which is about three orders of magnitude lower than that of blank WEP coating (1.76 × 10⁻⁵ mm/year). The result of the potentiodynamic polarization test revealed that MCPG/WEP possessed the best anti-corrosion performance, which agreed well with the results of EIS tests.

Figure 7. Comparison of different coatings. (a) Tafel curves after 2 days of immersion and (b) adhesion strength after drying and after 7 days of immersion.

Table 1. Electrochemical parameters via Tafel extrapolation fitting.

Samples	Ecorr (mV)	Icorr (μA/cm2)	βa (mV/Dec)	−βc (mV/Dec)	vcorr (mm/Year)	η (%)
WEP	−474	1.51 × 10−3	152.3	373.4	1.76 × 10−5	/
GO/WEP	−373	1.39 × 10−4	138.5	214.7	1.62 × 10−6	90.79
PDGO/WEP	−229.4	1.38 × 10−5	162.6	419.3	1.61 × 10−7	99.09
MCPD/WEP	−159.6	1.87 × 10−6	309.5	293.5	2.18 × 10−8	99.88

The adhesion strength of coatings on a steel substrate is a critical parameter for evaluating the stability of corrosion protection [45]. As shown in Figure 7b, incorporating nanofillers into the waterborne epoxy resin improved both the dry adhesion and wet adhesion (measured after 7 days immersion) strength. Notably, for MCPG/WEP, both dry and wet adhesion strengths exhibited significant enhancements compared to WEP, increasing from 5.64 to 6.30 MPa and from 3.51 to 6.02 MPa, respectively. Incorporating MCPG into the waterborne epoxy resin matrix could effectively fill pores, thereby enhancing the coating density [46]. The abundant amino, hydroxyl, and catechol functional groups on the nanofillers improve the interfacial bonding between MCPG and the matrix, further enhancing the adhesion strength. This enhanced adhesion inhibited coating delamination from the steel substrate, further improving the corrosion protection of the composite coatings.

3.3.3. Self-Healing Performance of Scratched Composite Coatings

To evaluate the self-healing performance, EIS analyses of scratched MCPG/WEP composite coating were performed in 3.5 wt.% NaCl solutions for 48 h. The Nyquist and Bode plots for various immersion times are shown in Figure 8. The coating-protected metal substrate was directly exposed to the corrosive environment after scratched, leading to accelerated corrosion. With the extension of immersion time, the |Z|_0.01 Hz first improved from 185 Ω/cm² (0 h) to 4.5 kΩ/cm² (24 h), and then decreased to 3.7 kΩ/cm² after 48 h. This effect is attributed to synergistic corrosion inhibition from the release of MBI and CeO₂ [47]. After the removal of the scratched coating, SEM and EDS were further performed to analyze the corrosion morphology and elemental composition of the carbon steel substrate (as shown in Figure S5). In conjunction with the weight loss method and Tafel curves (Figure S6), these inhibitors can form a protective film layer within the scratch, which impeded the electron transfer between anodic and cathodic sites and further promoted coating self-healing performance.

Figure 8. (a) Nyquist and (b) Bode plots and (c) |Z|_0.01 Hz for scratched MZPG/WEP coating-protected Q235 steel after different immersion times in 3.5 wt.% NaCl solution.

Figure 9 shows the digital photos of coatings after the salt spray resistance experiment. A dense layer of corrosion rust completely filled the scratch on the pure WEP coating (Figure 9a), suggesting rapid penetration of corrosive media (H₂O, O₂ and Cl⁻) to the substrate and subsequent metal oxidation. GO/WEP coating (Figure 9b) illustrated significant corrosion products and bubbles around the scratch, while the extent of corrosion was lowered after incorporating PDGO, which benefited from PDA’s strong adhesion force [48,49]. In contrast, for the MCPG/WEP coating (Figure 9d), rust only appeared in the damaged area. This is because MBI and CeO₂ were released and formed a protective layer on the substrate, effectively blocking electrolyte propagation [24,25]. It was mainly because MBI and cerium cations could form the passive layer on the substrate surface to prevent the propagation of electrolytes. Therefore, the remarkable long-term anti-corrosion performance under aggressive conditions confirmed the promising potential of the MCPG/WEP coating for industrial development.

Figure 9. Surface morphology of (a) WEP, (b) GO/WEP, (c)PDGO/WEP and (d) MCPG/WEP coatings after the salt spray test for 200 h.

3.4. Self-Healing Mechanism of MCPG/WEP Composite Coatings

Figure 10 presents the self-healing corrosion protection mechanism of MCPG/WEP coating in the face of corrosive media (H₂O, O₂ and Cl⁻). These corrosive agents will easily penetrate the micro-pores and defects, initiating the metal corrosion. With the incorporation of well-dispersed MCPG into the coating, the diffusion pathway of corrosive electrolyte was significantly prolonged, thereby enhancing the passive barrier performance of the coating. Over long-term service or damaged surface, the carbon steel substrate will interact with corrosive media, initiating a redox reaction that induces local acidification and alkalization [50]. The pH variation stimulated the release of encapsulated inhibitors from the MCPG composites. The released MBI reacted with exposed Fe²⁺/Fe³⁺ ions at the corrosion site, rapidly forming the protective films. Concurrently, cerium ions migrated to the coating/steel interface and reacted with OH⁻ produced by the cathodic corrosion reaction to form an insoluble cerium-based film, thereby inhibiting the cathodic reaction and reducing the corrosion rate [14,51]. The increased |Z|_0.01 Hz values of scratched coating in EIS tests positively confirmed this mechanism. Therefore, the smart coating could automatically sense when and where corrosion occurred, and accurately heal the corroded area, which contributed to the outstanding long-term anticorrosion performance. This intelligent design enabled the coating to autonomously sense corrosion initiation and deliver targeted healing, resulting in its excellent long-term corrosion protection.

Figure 10. Self-healing mechanism diagram of the scratched MCPG/WEP composite coating.

4. Conclusions

This study fabricated MCPG/WEP coatings doped with MBI and CeO₂ inhibitors and evaluated their self-healing and anti-corrosion performance. The results showed that the steel substrate coated with MCPG demonstrated remarkable and long-lasting corrosion resistance. More importantly, upon the incorporation of MBI and CeO₂, the resulting polymer coating displayed an efficient self-healing function. Meanwhile, the outstanding barrier property provided by the well-dispersed GO sheets effectively impeded the penetration of corrosive agents at defect sites. Combining self-healing capability with enhanced impermeability, the MCPG/WEP coating exhibited significant potential for future corrosion protection applications.