Novel anticorrosive coating of silicone acrylic resin modified by graphene oxide and polyaniline

2.2.1 Preparation of coatings modified by GO

Graphene oxide was weighed 0, 0.1, 0.3, 0.5, 0.7, and 1 wt% of the mass of silicone-acrylic resin, respectively, and added to the silicone-acrylic resin, which was evenly dispersed by ultrasound. Adding appropriate amount of defoamer and stirring until the bubbles completely disappeared. Finally curing agent was added to obtain GO-modified coating, noted as GO 0%, GO 0.3%, GO 0.5%, GO 0.7%, and GO 1%, respectively.

2.2.2 Preparation of coatings modified by PANI/GO

An amount of 100 mL of perchloric acid solution with concentration of 1 mol L⁻¹ was configured and divided into two parts. GO was weighted on bases of 1% mass fraction of aniline monomer into the prepared perchloric acid solution, and then 1.86 g of aniline was added with ultrasonic dispersion during above process until completely mixed, which was labeled as solution A. According to the molar ratio of aniline to ammonium persulfate of 1:1.2 and 5.7 g of ammonium persulfate was add into another prepared perchloric acid solution with stirring for full dissolution, which was labeled as solution B. The prepared solution B was slowly added dropwise to the prepared solution A by peristaltic pump and stirring at the same time, and the mixture was left for 12 h at room temperature after drop finished. Then the suspension was filtered, and filter residue was washed with deionized water and anhydrous ethanol, and dried later. The obtained products were the primary polyaniline/graphene oxide composite with GO mass fraction of 1%. The PANI/GO composites with GO content of 0, 5, 10, 20, 30, 40, and 50 wt% were prepared by the same steps.

An amount of 0.7 wt% PANI/GO with GO content of 0, 1, 5, 10, 20, 30, 40, and 50 wt% was added to as-prepared silicone acrylic resin with ultrasonic dispersion, respectively. Moderate dose of defoamer was added, and curing agent was finally putted when the bubbles disappear completely to obtain composite coatings modified by PANI/GO, which were recorded as PANI/GO 0%, PANI/GO 1%, PANI/GO 5%, PANI/GO 10%, PANI/GO 20%, PANI/GO 30%, PANI/GO 40%, and PANI/GO 50%, respectively.

2.2.3 Coating on metal surface

The surface of tinplate was sanded with 800-grit sandpaper and wiped with deionized water and ethanol. After cut into a suitable large spare, the glass sheet was wiped with deionized water and ethanol again and dried for coating. Coating was brushed to a layer evenly on the tinplate sheet and the glass sheet, respectively, and set aside at room temperature for curing.

3.1.1 Thermo-gravimetric analysis

TG analysis was applied to measure the mass of substances in relation to temperature under programmed temperature. During the rise of temperature, some behaviors may occur for substances, such as vaporize, sublimate, lose crystal water, and decompose, resulting in changes in mass. Meanwhile, the mass variation was recorded in the form of thermo-gravimetric curve to understand the rate of weight loss, the total mass change of the substance, and eventually evaluate the thermal stability. It is generally believed that the mass loss is due to the evaporation of the solvent before T_5% at which the mass of coating loses 5%, and the coating starts to degrade after temperature of T_5%. T_max is the temperature corresponding to the maximum weight loss rate of the coating during the temperature-rising test. Figure 1a shows the thermogram of the unmodified silicone acrylic resin coating with T_5% of 227 °C and T_max of 389 °C, while the composite coating modified by graphene oxide (Figure 1b) showed T_5% of 260 °C and T_max of 398 °C. Compared with the unmodified coating, the T_5% and T_max of the GO composite coating was increased by 33 and 9 °C, respectively. This is because GO is a two-dimensional layered structure with good thermal stability itself, and the surface is rich in oxygen-containing groups, which can combine with silicone resin to form a dense protective layer and improve the thermal stability of the coating.

Figure 1:

TG-DTG curves of (a) unmodified silicone acrylic resin coating, (b) coating modified by GO.

3.1.2 Tafel curve

Tafel polarization curves and electrochemical impedance spectroscopy are common methods to characterize the corrosion behavior of metals. As displayed in Figure 2, the Tafel curves of coatings modified by different contents of GO were measured after immersion in 3.5% NaCl solution for one day. The tendency of corrosion reaction can be reflected by the value of corrosion potential (E_corr, V vs. Hg/Hg₂Cl₂) at some extent, and corrosion current density (I_corr) is in direct proportion to the CR (corrosion rate). Therefore, the higher E_corr and the lower I_corr in the polarization curve, the better the corrosion protection performance of the coating. It is observed that the corrosion potentials of the coatings modified by GO were all higher than those unmodified coating, in which the coating with GO mass fraction of 0.7 wt% having the best performance with a corrosion potential of −0.3 V.

Figure 2:

Tafel curves of coated samples modified by GO with doping amounts of 0, 0.1, 0.3, 0.5, 0.7, and 1 wt%, after 1 days and bare electrode (in 30 min) exposure to 3.5 wt% NaCl solution.

The cathodic and anodic Tafel coefficients were obtained by linear fitting based on the Tafel curve, and the polarization resistance R_p (Ω cm²) was calculated according to the Stern–Geary equation, as shown in equation (1).

where R_p was the polarization resistance (Ω cm²), I_corr was the corrosion current density (A cm⁻²), β_a and β_b were the Tafel coefficients of the cathode and anode, respectively.

CR was the corrosion rate (mm year⁻¹) calculated by equation (2) (Ramirez-Soria et al. 2020).

where k was a constant (3268.5), M_m is the equivalent weight mass of metal substrate (56 g/mol) (Li et al. 2019a,b; Rodríguez-Díaz et al. 2015), and ρ_m was the density of iron (7.85 g/cm³).

As shown in Table 1, with the amount of GO added from 0 to 0.7 wt%, the corrosion potential and polarization resistance of coating increases, while corrosion current density and corrosion rate decreases, indicating that addition of GO obviously enhances the performance of coating in corrosion resistance. At GO mass fraction of 0.7 wt%, the coating presents the best performance on corrosion resistance, which could result from the fact that the lamellar structures from GO in the coating stacking reach the most compact structure, and are enough to make the corrosion medium with the good corrosion resistance. The main mechanism of the positive effect and chemical inertia by GO in coating on corrosion resistance is the barrier effect, which can maintain the stability of structure in coating and increase difficulty for contact between reactive component (like O₂, H₂O) and metal (Kulyk et al. 2021). Compared with the unmodified coating (GO 0%), the corrosion potential of GO 0.7% increased from −0.8 to −0.3 V, and the corrosion current density decreased from 3.75 × 10⁻⁷ to 1.17 × 10⁻¹⁰ A cm⁻² (with a decrease of 3 orders of magnitude). The polarization resistance was maximum of 1.12 × 10⁹ Ω cm², and the corrosion rate was 2.73 × 10⁻⁶ mm year⁻¹ at the minimum. Additionally, when the addition of GO is up to 1 wt%, excessive addition of GO cause the pores and agglomeration of GO in the coating, leading to the decline of coating density and corrosion resistance of coating with smaller corrosion potential and polarization resistance, and increased corrosion current density.

3.1.3 Electrochemical impedance

The corrosion resistance of coating relies on the effect of isolating the substrate from the corrosive medium by attaching the coating to the substrate. Due to the defects in the formulation, the coating is not dense enough or swollen during the immersion process, resulting in the corrosion medium continuously penetrating into the coating. Generally, the stage which the corrosive medium penetrates in the coating and does not reach the substrate is called the early stage of immersion. In order to test the corrosion resistance of the coatings with different graphene oxide additions, electrochemical impedance spectroscopy was used to further examine the performance and the measured data were fitted with the fitting software ZVIEW.

Figure 3a–c shows the Nyquist plots of the coatings with different GO additions after 1, 30, and 60 day of immersion, respectively. Equivalent circuit in Figure 3d was used to fit ESI results, and detailed fitting parameters are shown in Supplementary Tables S1–S3. The coating resistance modified by 0.7 wt% GO is decreased from about 10⁹ Ω cm² to 7.28 × 10⁸ Ω cm², to 1.02 × 10⁸ Ω cm², and to 1.97 × 10⁵ Ω cm² after immersing for 1 day, 30, and 60 days, respectively. The electrochemical impedance spectra of each coating at the beginning of immersion for 1 day (Figure 3a) all show a capacitive resistance arc, implying that the coating has the high resistance value and good anticorrosion, and the corrosive medium just starts to penetrate in the GO coating at the initial stage. The slope of the capacitive anti-arc line of GO 0.7% is the largest, indicating that the coating has the best corrosion resistance. On contrary, the slope of the unmodified coating (GO 0%) is the smallest, corresponding to the worst corrosion resistance. The corrosion resistance of each coating changed greatly after 30 days of immersion. The rapid reduction of the semicircular arc was observed from the coating impedance spectrum (Figure 3b), demonstrating that the corrosive medium was saturated in the coating, and further corrosion may cause local coating off and more serious corrosion. At this point, the corrosive medium has reached the substrate through the coating, but visible corrosion or peeling defects has not formed, and the coating still has a certain protective effect, which can be called for the immersion of the middle during period. As shown in Figure 3c, when immersed for 60 days, corrosion medium in the coating was completely saturated and completely penetrated to reach the substrate with the occurrence of the corrosion reaction in the substrate. Some of the coated specimens displayed the apparent rust stains, but no peeling and local loss of corrosion protection, while other parts were intact. Additionally, complete capacitive resistance arc and warp tail has appeared in the electrochemical impedance spectrum, proving the existence of diffusion impedance as well as coating impedance.

Figure 3:

EIS graphs of coated samples after immersion in 3.5 wt% NaCl solution for (a) 1 day, (b) 30 days, (c) 60 days; (d) equivalent circuit models used to perform EIS data (the coatings were modified by GO with doping amounts of 0, 0.1, 0.3, 0.5, 0.7, and 1 wt%).

The larger the radius of the capacitive reactance arc of the coating, the greater the electrochemical impedance. After soaking for 30 and 60 days, the maximum radius of the capacitive reactance arc was appeared at same coating with 0.7% GO, confirming that the GO addition of 0.7 wt% was the optimum content in coating with the maximal increment of corrosion resistance. Accordingly, the corrosion resistance of silicone-acrylic resin modified by GO was improved effectively, because GO has a unique two-dimensional sheet structure, which can change the corrosion path and prolong the time of the corrosive medium to the substrate and thus slow down the corrosion rate. When the GO amount is 0.1, 0.3, and 0.5 wt%, it is difficult to achieve the optimal effect of corrosion resistance owing to the parallel distribution of large gaps of insufficient GO in coating. At the optimal amount of GO of 0.7 wt%, the coating has the best corrosion resistance, which can be ascribed to the maximum corrosion path suffer from uniform distribution of two-dimensional flakes of GO in the coating. Excessive GO like 1 wt%, may bring about agglomeration, which reduces the density of the coating and slightly worsen the anticorrosive effect compared with the 0.7% GO.

3.2.1 Results of FTIR

The FTIR of PANI, GO, and PANI/GO were shown in Figure 4. The peak at 2360 cm⁻¹ is the peak of carbon dioxide in the air background. In the infrared spectrum of GO, the peak at 1650 and 1125 cm⁻¹ was the stretching vibration in carboxyl group C=O and alkoxy group C–O, respectively. The peaks of 1570, 1480, 1300 and 1130 cm⁻¹ in the polyaniline correspond to the stretching vibration of C=C in quinone ring, C=C in benzene ring, C–N on the benzene ring, N atom on the quinone ring, respectively, and peak value at 816 cm⁻¹ was the bending vibration peak of C-H on the benzene ring. Actually, the infrared spectrum of PANI/GO reflects that the characteristic peaks of the composites are completely involved in those of polyaniline, and the absorption peaks move to high frequency in varying degrees, which result from the increase of the distance between GO layers brought by the addition of PANI, and the steric hindrance moves the characteristic peaks, resulting in spatial hindrance and moving the characteristic peak. Therefore, there are no variations in the structure of PANI in composite of PANI/GO, indicating that the PANI/GO composites were successfully prepared.

Figure 4:

FTIR spectra of polyaniline, GO and PANI/GO.

3.2.2 Thermo-gravimetric analysis

It can be seen from Figure 5 that the T_5% and T_max of PANI/GO modified coating is 271 and 394 °C, respectively. Compared with the unmodified silicone acrylic resin, the initial decomposition temperature and the maximum weight loss rate temperature are increased by 44 and 5 °C, respectively. When PANI is compounded with GO, the chemical chain of PANI is inserted into the GO sheet layer, which affects the thermal motion of the whole composite and improves the thermal stability of the composite. Accordingly, the thermal stability of the PANI/GO coating is effectively enhanced.

Figure 5:

TG-DTG curves of the PANI/GO modified coating.

3.2.3 Morphology analysis

Figure 6 shows the SEM images of different materials. It can be observed in Figure 6a that the surface of untreated graphene oxide was smooth and multilayer agglomerated. Figure 6b is the SEM image of perchloric acid primary doped polyaniline, which were of the shapes of short rod and has obvious agglomeration. It can be seen from the SEM image of PANI/GO composite (Figure 6c and d) that graphene oxide is wrapped by polyaniline and maintains the original sheet structure, while the smooth surfaces become rougher and the layer thickness is smaller. This is because the aniline adheres to the surface of GO by adsorption, л–л conjugation, intermolecular interaction, chemical reaction, etc., and polymerization occurs on the surface under the action of oxidant of ammonium persulfate to generate polyaniline (Kinlen et al. 1997; Spinks et al. 2002). Therefore, it is implied that the PANI/GO composite has been successfully synthesized, which is not only a simple physical mixture.

Figure 6:

SEM images of (a) graphene oxide, (b) perchloric acid doped polyaniline, (c and d) PANI/GO composites.

3.2.4 Tafel curve

The Tafel curves of the coatings modified by PANI/GO with different contents of GO were measured after soaking in 3.5% NaCl solution for one day (Figure 7), and the corrosion test results were obtained as listed in Table 2. As the GO content in PANI/GO increases from 0 to 20 wt%, the corrosion potential increases from −0.66 to −0.14 V and the corrosion current density decreases from 2.45 × 10⁻⁷ to 1.8 × 10⁻¹¹ A cm⁻² (with a decrease of 4 orders of magnitude), suggesting that the GO addition of 20% is opportune to be uniformly distributed in the coating, as well as the passivation effect on the metal material from polyaniline in the composite, leading to excellent corrosion resistance for PANI/GO 20% with smallest corrosion rate of 4.18 × 10⁻⁷ mm year⁻¹ and largest polarization resistance of 6.06 × 10¹⁰ Ω cm². When the GO content in PANI/GO increased from 20 to 50 wt%, the corrosion potential decreased from −0.14 to −0.47 V, and the corrosion current density increased to 5.8 × 10⁻⁹ A cm⁻². Apparently, GO agglomerates appeared with the increase of GO content, which results in larger gap, uneven distribution, reduced corrosion potential, and smaller coating corrosion current. Therefore, anti-corrosion performance of coating can be significantly improved by addition of PANI/GO, and the PANI/GO 20% has optimal performance, which is ascribed to the passivating effect of polyaniline on metallic materials (Armelin et al. 2007; Sun et al. 2016) coupled with the physical anti-corrosion effect of graphene oxide, slowing down the corrosion of metal substrate.

Figure 7:

Tafel curves of the PANI/GO coated samples with different GO contents after 1 day and bare electrode (in 30 min) exposure to 3.5 wt% NaCl solution. The addition of PANI/GO is the 0.7 wt% of the mass of silicone-acrylic resin, and GO content in PANI/GO is 0, 1, 5, 10, 20, 30, 40, and 50 wt%, respectively.

3.2.5 Electrochemical impedance

Figure 8 is the Nyquist diagram of PANI/GO modified coating with different amount of GO after immersion. Detailed fitting parameters are shown in Supplementary Tables S4–S6. The coating resistance modified by PANI/GO 20% is about 10¹⁰ Ω cm² after immersing for 1 day, then it reduced to 2.17 × 10⁹ Ω cm² after immersing for 30 day and continuously decreases to 3.12 × 10⁷ Ω cm² after 60 day. As Figure 8a shown, each coating in the early stage of immersion presented a capacitive resistance arc. At this time, the corrosive medium just began to penetrate in the coating and not reached the substrate, which belongs to the early stage of immersion with good corrosion resistance effect for coating. After 30 days of immersion (Figure 8b), the electrochemical impedance spectrum of each coating specimen were changed greatly with the impedance spectrum semicircular narrowed, where the radius of the arc size was proportional to the impedance value. During this stage of immersion, the corrosive medium may have penetrated the PANI/GO coating, but the corrosion has not occurred on the surface of the metal substrate. When the corrosive medium was in contact with the substrate, it will cause corrosion of the substrate, and even lead to the coating to fall off and further decline of corrosion protection. Actually, the phenomenon of shedding and corrosion was not observed in practice, indicating that the coating still has good corrosion resistance, called medium-term immersion. It can be discovered from Figure 8c that the impedance spectrum semicircular arc further shrined and warped after immersion for 60 days, inferring that coating impedance emerged as well as diffusion impedance emerged after a long period of immersion. The corrosion medium has completely penetrated into the coating and reached saturation, but the coating is still intact and has certain protective ability. In view of the correspondence between the capacitive arc and impedance, it can be revealed that PANI/GO 20% has the best performance on corrosion resistance.

Figure 8:

EIS graphs of PANI/GO coated samples with different GO content after immersion in 3.5 wt% NaCl solution for (a) 1 day, (b) 30 days, and (c) 60 days. The coatings were modified by 0.7 wt% PANI/GO, and GO content in PANI/GO is 0, 1, 5, 10, 20, 30, 40, and 50 wt%, respectively.

Above results of electrochemical impedance spectroscopy showed that the addition of PANI/GO is conducive to the improvement of the corrosion resistance of the coating. When the GO content is 0%, the corrosion resistance of coating is the worst compared to other any amount of GO, and it is difficult to form the maximum corrosion path because PANI has a rod structure and the specific surface area is less than GO. With the increase of GO mass fraction in PANI/GO, the performance of corrosion resistance in coating enhanced and achieved optimal when it was modified by PANI/GO with 20 wt% GO. It can be the explanation that the lamellar structure of GO can be uniformly dispersed in the coating so that the corrosion path was maximized and the arrival of corrosive medium to the substrate was delayed to improve the corrosion resistance. When the amount of GO was insufficient, the barrier cannot be formed completely and the insulating properties was relatively poor. While the content of GO was excessive, it would result in agglomeration and destroy the homogeneity of the composite, which would affect the corrosion resistance. Specially, the impedance of PANI/GO 0% coating was greater than that of unmodified coating, which can be attribute to passivating effect and shielding effect on metal by polyaniline, leading to the better corrosion resistance.