Preparation of Zn₅Mo₂O₁₁·5H₂O@Sulfonated Graphene by Template Method and Its Anticorrosion Mechanism in Polyurethane Coatings

Abstract

Currently, waterborne coatings are widely used, but their poor anticorrosion performance compared with solvent-based coatings hinders their further application. In this study, a series of Zn₅Mo₂O₁₁·5H₂O@sulfonated graphene (SZMO@SG) composites are prepared using the template method. SZMO@SG is a type of two-dimensional material with a large diameter and low thickness compared to traditional zinc molybdate. Combining SZMO@SG with waterborne hydroxy acrylic latex and isocyanate, the polyurethane coatings with 3 wt% SZMO@SG achieve the best anticorrosion performance in the 3.5 wt% NaCl immersion test. In electrochemical impedance spectroscopy (EIS), the charge transfer resistance (R_ct) of polyurethane coatings with 3 wt% SZMO@SG reaches 283,100 Ω·cm² after soaking in a 3.5 wt% NaCl solution for 48 h. Its corrosion potential and corrosion current are only −0.453 V and 1.71 × 10⁻³ mA/cm², respectively. After metal substrate corrosion, the metal substrate of polyurethane coatings with SZMO@SG contain more Mo, Zn and low-state Fe atoms. It confirms that a passivation film was deposited on the metal substrate and prevents Fe from oxidation. The elemental content and distribution of the coatings after the NaCl solution immersion test indicate that SZMO@SG may have the Na⁺ adsorption function.

1. Introduction

Carbon steel is widely used in bridges, ships, pipelines, automobiles and other fields [1], but the economic loss caused by steel corrosion is inestimable [2]. Conversion coatings, especially phosphate conversion coatings, are widely applied to carbon steel, and improve its anticorrosion performance [3,4]. In order to broaden the application field of phosphate conversion coatings, researchers have explored various modification methods and deposition processes of conversion coatings [5,6]. For some waterborne anticorrosive coatings, their anticorrosive performance also depends on the phosphate deposition on the metal substrate [7].

Among the various resins available for organic coatings, polyurethane is a type of crosslinking curing resin [8,9], which is widely used and which dominates the coating industry. After adding a crosslinker such as isocyanate, a network structure can be formed, which improves its water resistance [10]. Waterborne polyurethane coatings, having the advantages of high gloss, high adhesion, and high hardness, are widely used in various fields [11,12,13]. Waterborne polyurethane coatings also have the advantages of low VOCs and environmental friendliness [14]. The size, shape, and chemical composition of the fillers are the important factors that affect the anticorrosion performance of coatings. The physical barrier performance of sheet fillers is significantly better than that of spherical fillers and linear fillers [15]. The reports also show that the chemical composition of the filler has a great impact on the self-healing and corrosion responses of the coatings [16,17].

Pollution problems have prompted scientists to look for novel environment-friendly corrosion inhibitors [18,19]. Molybdate, which has low toxicity in the environment compared to chromate, is better than phosphate in its anticorrosion performance [20,21,22]. It is expected to be an alternative high-performance anticorrosive filler. Early studies on molybdate mainly focused on passivation treatment on metal surfaces [23,24] to protect metal substrates, or on directly adding molybdate into coatings as functional fillers [25,26], so as to improve the anticorrosion performance of organic coatings. In recent years, many studies [27,28,29,30] have indicated that the performance of anticorrosion fillers can be further improved if multiple anticorrosion mechanisms are applied. Research on the new structures of molybdate fillers has emerged, such as ionic molybdate doping fillers [31] and molybdate intercalation modified hydrotalcite [32,33].

The template method is a very common way of synthesizing nano-materials such as nanoflakes, nanotubes, and nanoparticles [34]. It is easy to control the size and shape of nano-materials using the template method. Graphene is a type of two-dimensional sheet material with high conductivity and mechanical strength, which could be employed to modify the molybdate filler to improve its comprehensive performance. Sulfonic groups enhance the surface charge and chemical activity of the graphene [35]. Due to its sheet structure, graphene is often used as a template to prepare thin sheet materials [36,37]. Fillers with a sheet structure can form physical barriers, improving the anticorrosive performance of coatings.

In this study, the sheet molybdate anticorrosion filler with a high diameter-to-thickness ratio was synthesized. A novel sheet zinc molybdate composite filler (SZMO@SG) was prepared by depositing Zn(OH)₂ on the surface of the sulfonated graphene and transforming to zinc molybdate. Sheet zinc molybdate can be obtained when the addition of sulfonated graphene reaches 1.5 wt%. Combining SZMO@SG with waterborne hydroxy acrylic latex and isocyanate, the polyurethane coatings with 3 wt% SZMO@SG achieve the best anticorrosion performance. After immersion in 3.5 wt% NaCl solution for 48 h, the R_ct of polyurethane coatings with 3 wt% SZMO@SG reaches 283,100 Ω·cm². Moreover, the anticorrosion mechanism of Zn₅Mo₂O₁₁·5H₂O was confirmed using EIS and an X-ray photoelectron spectrometer (XPS), which was lacking in in previous studies. SZMO@SG has several anticorrosion functions, including forming a physical barrier and a passivation film, and absorbing Na⁺. The sheet molybdate filler by template method can provide a long-term anticorrosion performance for waterborne polyurethane coatings.

2. Materials

2.1. Materials

The materials used in this study were shown below: zinc nitrate hexahydrate (Zn(NO₃)₂·6H₂O, 99%, General-Reagent, Shanghai, China), ammonium molybdate ((NH₄)₆Mo₇O₂₄·4H₂O, 99%, Sinopharm, Shanghai, China), ammonium hydroxide (NH₃·H₂O, 25%, Sinopharm, Shanghai, China), sodium dodecyl sulfate (SDS, 88%, Sinopharm, Shanghai, China), sulfonated graphene water dispersion (1%, Graphene-Tech Co., Ltd., Suzhou, China), waterborne hydroxy acrylic latex (Wanhua Chemical Group Co., Ltd., Yantai, China), isocyanate (Wanhua Chemical Group Co., Ltd., Yantai, China), and anhydrous ethanol (99.7%, General-Reagent, Shanghai, China). All the above reagents were used as received.

2.2. Synthesis of SZMO@SG Composite

SZMO@SG composites were prepared using the template method. The synthesis of Zn(OH)₂@SG is referred to previous study [38]. Figure 1 is the schematic diagram of the preparation process of SZMO@SG. In the first step, a certain amount of SDS was added to 100 mL distilled water. A quantity of sulfonated graphene (SG) water dispersion was then added. The dispersion was treated by ultrasonication with 150 W power for 30 m, and then transferred into a 250 mL three-necked flask. A 25 mL 1.2 mol/L zinc nitrate solution (prepared with zinc nitrate hexahydrate) was added after 20 min, followed by 50 mL 1.25 mol/L ammonia solution (prepared with 25% ammonia) after 30 min. After reaction for 1 h, the solution was washed and filtered with distilled water three times.

In the second step, the product obtained in the previous step was placed in a 500 mL three-necked flask, and then 200 mL distilled water was added, followed by ultrasonication with 150 W power after 30 min. A 50 mL 0.087 mol/L ammonium molybdate solution (prepared with ammonium molybdate tetrahydrate) was added and reacted for 1 h under 80 °C with 500 rpm. Finally, the product was washed with distilled water and ethanol, and then dried in 50 °C.

2.3. Preparation of Polyurethane Coatings Based on SZMO@SG

The formula of coatings and dosage of composite fillers were based on our previous work [7]. An amount of SZMO@SG (weight ratio of 1 wt%, 3 wt%, or 5 wt%) was added to the waterborne hydroxy acrylic latex, and stirred for 30 m. A quantity of isocyanate was then added. Coatings were prepared on carbon steel for a water contact angle test, an electrochemical impedance spectroscopy test, and a 3.5 wt% NaCl solution immersion test. The film thickness was set as 50 ± 5 μm. Blank waterborne hydroxy acrylic latex was used as a control.

2.4. Characterization

The surface morphology, particle size, and lamellar thickness of Zn₅Mo₂O₁₁·5H₂O (ZMO) and SZMO@SG fillers were obtained using a scanning electron microscope (Zeiss Sigma 300, Zeiss, Oberkochen, Germany) at 3 kV. X-ray diffraction (XRD, PANalytical, Almelo, Netherlands) patterns of SG, ZMO, and SZMO@SG were recorded using an X’Pert PRO MPD and CuK_α as the radiation source (λ = 0.15406 nm). A Fourier transform infrared spectrometer (Nicolet, Thermo Fisher Scientific, Massachusetts, America) was used to obtain the functional group information of the SG, ZMO, and SZMO@SG. The surface elements of SG, ZMO, and SZMO@SG were tested using an X-ray photoelectron spectrometer (Axis Ultra DLD Kratos Axis SUPRA, KRATOS, Manchester, Britain) with monochromatic AlK_α sources. The C 1s signal was calibrated to 284.8 eV. The peak fitting method depends on the atomic spin splitting. The single peak fitting was applied to C 1s and O 1s. The double peaks fitting was applied to Zn 2p, Fe 2p, S 2p, and Mo 3d. Before the electrochemical test, the carbon steel coated with coatings was cut, and the edge and back surface of the sample was sealed with a mixture of rosin: paraffin = 1:1. The coatings with a retention area of 1 cm × 1 cm were tested using the electrochemical method. After soaking in 3.5 wt% NaCl solution for a period of time, the EIS and Tafel curves of coatings were obtained using an electrochemical workstation CHI660E (Shanghai Chenhua Device Co., Shanghai, China). Both Tafel curves and EIS testing were carried out with respect to SCE. Before the water contact angle test, coated carbon steel was soaked in deionized water for a certain period of time, then dried at room temperature for 5 min. A water contact angle tester (Shanghai Pushen Testing instrument Co., Ltd., Shanghai, China) was used to obtain the contact angle data of coatings.

3. Results and Discussion

3.1. Characterization of SZMO@SG

The FT-IR spectra of SG, ZMO, and SZMO@SG was shown in Figure 2. In the FT-IR spectra of SG, the infrared absorption at 1705 cm⁻¹, 1615 cm⁻¹, 1141 cm⁻¹, and 1033 cm⁻¹ corresponds to the stretching vibration absorption of C=O, C=C, O=S=O, and C–O [7,39], respectively. The wide peak in the range of 3300~3450 cm⁻¹ is the stretching vibration absorption of sulfonated graphene with –OH [40]. The stretching vibration and bending vibration of molybdate ion corresponding to 800~930 cm⁻¹ are shown in the spectrum of ZMO [41,42]. The wide peak in 3200~3250 cm⁻¹ proves that ZMO has crystal water [43]. In addition, the absorption peak of 1610 cm⁻¹ is also related to the bending vibration of crystal water [44]. The absorption peak at 480 cm⁻¹ is ascribed to the Zn–O. In the spectrum of SZMO@SG, the absorption peaks of C=O, O=S=O, and C–O of sulfonated graphene are located at 1725 cm⁻¹, 1143~1233 cm⁻¹, and 1030 cm⁻¹, respectively. It can be concluded that the prepared composite filler was a compound of ZMO and sulfonated graphene.

To discuss the surface chemical composition of SG, ZMO, and SZMO@SG, the elemental composition and surface elemental valence states of SG, ZMO, and SZMO@SG are characterized by XPS, and the results are shown in Figure 3. In the spectrum of SG C 1s, the three peaks at 284.8 eV, 286.2 eV, and 288.8 eV are attributed to C–C, C–O, and C=O in the sulfonated graphene [45]. In the O 1s spectrum, the peaks at 532.7 eV, 533.5 eV, and 532.0 eV are ascribed to O=S=O, C–O, and carboxyl –OH, respectively [46]. The peaks of 164.3 eV and 169.3 eV in S 2p correspond to C–S and S–O, which proves the existence of sulfonic acid groups in the graphene surface. In the O 1s spectrum of ZMO, the 530.5 eV and 532.1 eV peaks are attributed to molybdate and binding water [47], and the 531.2 eV peak is considered to be the surface hydroxyls. There are two strong peaks of 1021.5 eV and 1044.5 eV in the spectrum of Zn 2p, which are related to the binding energies of Zn 2p_3/2 and Zn 2p_1/2, respectively. The difference between the two peaks’ binding energies δ is 23.0 eV. It can be inferred from the value of δ that Zn element exists in a bivalent state in ZMO [48]. In the Mo 3d spectrum, both 232.7 eV and 235.8 eV are characteristic peaks of Mo⁶⁺, which are associated with the binding energies of Mo 3d_5/2 and Mo 3d_1/2 [49].

The XPS spectrum of SZMO@SG illustrates that O 1s has three peak values of 530.6 eV, 531.3 eV, and 532.3 eV, which are consistent with molybdate, surface hydroxyls, and binding water. The binding energies of Zn 2p_3/2 and Zn 2p_1/2 of Zn 2p are located at 1021.7 eV and 1044.7 eV. Compared with ZMO, Mo 3d of SZMO@SG only has Mo 3d_5/2 and Mo 3d_3/2 signals at 232.7 eV and 235.9 eV. The C1s spectra of SZMO@SG have C–C, C–O–C, and C=O peaks of SG at 284.8 eV, 286.3 eV, and 288.8 eV. On the other hand, in the S 2p spectrum of SZMO@SG, the signal is weak at the positions of 162.9 eV and 169.0 eV. This is because the sampling depth of XPS is only 2~5 nm, while the outer layer of SZMO@SG is mostly composed of molybdate with less S element content on the surface. The XPS analysis reflects the fact that ZMO and SZMO@SG have similar atomic ratios of Zn:Mo, which are 1.78:1 and 1.74:1, respectively. We were able to confirm from the XPS spectra that the samples prepared are composites of ZMO encapsulated SG.

XRD can be used to further study the crystal structure of the fillers. As shown in Figure 4, strong characteristic peaks appear at 11.7° and 20.7° in the XRD patterns of the SG. The peak of 11.7° is commonly regarded as the layer spacing of graphene materials. According to the Bragg equation, the layer spacing of sulfonated graphene can be calculated as 0.76 nm, which is close to the data in the literature [50]. The 20.7° peak corresponds to the 0.43 nm layer spacing of graphite, indicating that the sulfonated graphene used has an incomplete oxidation structure [51]. There are three characteristic peaks of 12.3°, 26.6°, and 34.0° in the XRD pattern of ZMO, which correspond to the crystal faces of (003), (015), and (021), respectively. The lattice constants a = 6.1423 Å, b = 6.1423 Å, and c = 21.6347 Å were calculated from the data of the XRD patterns, which were very similar to JCPDS No. 30-1486 Zn₅Mo₂O₁₁·5H₂O (a = 6.136 Å, b = 6.136 Å, and c = 21.6 Å) [52]. The XRD patterns of SZMO@SG show a great similarity with Zn₅Mo₂O₁₁·5H₂O, and the signals at 12.1°, 26.5°, and 34.0° correspond to the crystal planes of (003), (015), and (021). The calculated lattice constants are a = 6.1271 Å, b = 6.1271 Å, and c = 21.5753 Å. The disappearance of SG characteristic peaks in the XRD patterns of SZMO@SG indicates that the lamellar layers of SG are filled with deposited Zn₅Mo₂O₁₁·5H₂O, and the spacing between SG layers of SZMO@SG is much larger than that of pure SG. The weak peaks at 2 θ = 20.2°, 20.9°, 27.2°, and 27.8° correspond to (011), (101), (111), and (102) crystal planes, indicating that the filler contains a small amount of Zn(OH)₂. However, the Zn₅Mo₂O₁₁·5H₂O dominates in the composite filler. The XRD spectra proved that the main components of ZMO and SZMO@SG are Zn₅Mo₂O₁₁·5H₂O, and also indirectly proved that SZMO@SG is a composite filler from ZMO and SG.

SEM images of ZMO and SZMO@SG are shown in Figure 5. It shows that the particle size of ZMO is in the range of 15~40 μm in Figure 5a. In the ZMO crystal growth process, the small irregular ZMO particles are generated first, and then the small ZMO particles are assembled to form large ZMO particles with rough surfaces. Therefore, the grooves formed by the splicing of the particles can be clearly seen on the ZMO surface. In addition, due to the hexagonal structure of Zn₅Mo₂O₁₁·5H₂O, ZMO has numerous ~100 nm sheet folded structures. SEM images of SZMO@SG with moderate SG are shown in Figure 5b, which show a disk-like morphology. The diameter of SZMO@SG is ~30 μm, and the thickness of the lamellar is ~100 nm. Due to the existence of the template, small SZMO@SG particles were formed in the growth process. However, the concentration of SZMO@SG particles is appropriate to stack in parallel and grow radially. Finally, large SZMO@SG particles are formed. The grooves formed by splicing can still be seen in Figure 5b, but there is almost no sheet folded structure compared with that in Figure 5a. As can be seen in Figure 5c, after the addition of 0.6 wt% SG, the size of SZMO@SG particles prepared was in the range of 1~10 μm, and the thickness of the lamella was ~200 nm. When the amount of SG exceeded a certain amount, the size of initially formed SZMO@SG particles was smaller, and the number of particles per unit volume of solution increased sharply. The different size and orientation of SZMO@SG particles can be spliced in different directions, and, finally, SZMO@SG particles with smooth surface but folded structure were synthesized.

3.2. Corrosion Resistance of Coatings with SZMO@SG

The change in the coating contact angle with the soaking time reflects the transformation of hydrophilic or hydrophobic property of the surface of the coatings. The curves of the contact angles of the coatings are shown in Figure 6.

It can be seen that the contact angle of the coatings tends to decrease with the increase in immersion time. During immersion, water molecules were absorbed on the surface of the coatings, which led to the permeation of the corrosive medium on the surface of the coatings more easily. Moreover, the coating was constantly damaged in the corrosion process, and the increasing of defects and cracks on the surface of the coating reduces the contact angle and anticorrosion performance. The contact angle of the control and the sample with 1 wt% SZMO@SG decreased the most, while the contact angle of samples with 3 wt% and 5 wt% fillers did not change significantly. Therefore, high-content SZMO@SG coatings (3 wt% and 5 wt%) may be beneficial for improving the anticorrosion performance.

The variation of element distribution in the cross-section of the coatings with high magnification is shown in Figure 7 and the corresponding data is shown in

Table 1. The change in element contents of different coatings with different immersion times.

Samples	O (%)	Na (%)	Cl (%)	Fe (%)	Zn (%)	Mo (%)
control-0 h	38.07	0.29	0.20	61.39	0.06	0.00
control-1008 h	39.69	9.03	0.21	51.07	0.00	0.00
3 wt% SZMO@SG-0 h	23.86	0.90	0.12	67.94	5.59	0.90
3 wt% SZMO@SG-1008 h	40.12	10.71	2.44	39.86	5.87	0.99

. It can be clearly seen that the signals of the Na and Cl element for the control were enhanced and uniformly distributed in the coatings after immersion. The proportion of Na was higher than that of Cl, proving that the permeation rate of Na⁺ in the coatings is faster than that of Cl⁻ during immersion. As can be seen from Figure 7c,d, Zn and Mo signals appeared in the coatings after the addition of SZMO@SG, and Zn and Mo even overlapped with some O signals. After soaking in NaCl solution, Na signals were still enhanced. In addition, the distribution of Na overlapped with Zn and Mo, which may be due to the adsorption and enrichment of Na⁺ on SZMO@SG surface.

In order to visually measure the corrosion resistance of coatings with different SZMO@SG content, the samples were immersed in 3.5 wt% NaCl for a test. As shown in Figure 8a, it was found that the corrosion resistance of the coatings increased first, and then decreased with the increase in the amount of SZMO@SG. The control without the addition of SZMO@SG filler showed serious corrosion after soaking for 96 h. The corrosion medium penetrated completely into the coatings, and diffused into the interface between the carbon steel and coatings. The control also developed blistering above the soaking line after soaking for 288 h. When the addition of SZMO@SG reached 1 wt%, the coatings showed serious corrosion at the soaking line between 288 h and 480 h, and there were also traces of penetration by the corrosion medium above the soaking line. When the amount of SZMO@SG increased to 3 wt%, the polyurethane coatings only showed slight corrosion at 1 cm near the immersion line after immersing for 1008 h. There was only one rust spot below the immersion line, almost no corrosion trace on the other position of sample, and no obvious traces of penetration by the corrosion medium above the immersion line was observed. However, when the amount of SZMO@SG reached a high level of 5 wt%, the corrosion resistance of the coatings decreased and a large number of rust spots appeared after 480 h immersion. The results indicate that adding 3 wt% SZMO@SG fillers can effectively prevent the penetration of corrosion medium in the coatings, improve the corrosion resistance of the coatings, and effectively protect the metal substrate.

Moreover, rusty spots appeared on the surface of coatings with 3 wt% ZMO after soaking in a 3.5 wt% NaCl solution for 288 h. The corrosion resistance of 3 wt% ZMO coatings is obviously weaker than 3 wt% SZMO@SG coatings. Because the ZMO could not form the layer-by-layer structure in the coatings. In addition, due to the poor compatibility between resin and inorganic fillers, ZMO coatings have more defects compared to the SZMO@SG coatings, leading to the formation of corrosive medium permeation channels in the coatings. Therefore, the corrosive medium is more likely to penetrate into the 3 wt% ZMO coatings during immersion.

After soaking for 1008 h in a 3.5 wt% NaCl solution, the control and coatings with 3 wt% SZMO@SG were peeled by soaking in acetone for 24 h. To investigate the chemical changes of coated metal substrates after immersion, a XPS test was carried out on the surface of metal substrate. The change in Fe valence state after corrosion can be determined by Fe 2p spectrum, as shown in Figure 8, where 724.8 eV, 719.4 eV, and 711.6 eV are characteristic peaks of Fe³⁺ [53,54], and the characteristic peaks of 715.7 eV and 710.0 eV are attributed to Fe²⁺ [55,56]. Using software to calculate the peak area, the atomic ratio of the two valence states of Fe³⁺:Fe²⁺ is 1.56:1, which is close to the Fe content in Fe₃O₄. The element valence states on the surface of metal substrate of coatings with 3 wt% SZMO are shown in Figure 8c–e. Due to the deposition of a passivation film on the metal substrate, Zn and Mo signals became stronger, and Fe signal became weaker. The Fe 2p spectrum can further verify the anticorrosion mechanism of SZMO@SG. The characteristic peaks of Fe²⁺ are ascribed to 708.3 eV, 709.8 eV, 711.1 eV, 712.4 eV, and 715.7 eV, while the weak characteristic peak of Fe³⁺ is located at 711.1 eV. In addition, there is also a weak characteristic peak of 0 valence state Fe at 705.9 eV. The ratio of Fe³⁺:Fe²⁺ was calculated as 1:10.08, indicating that the passivation film effectively prevents the oxidation of Fe. In Mo 3d and Zn 2p spectra, Zn²⁺ and molybdate ions were also deposited on the carbon steel, and these ions were involved in the formation of a passivation layer. Compared with the Fe 2p spectrum of the control in Figure 8a, the carbon steel coated with coatings of 3 wt% SZMO@SG contained more Fe²⁺ on its surface after corrosion, and there was even 0 valence Fe on the surface, which proves that the passivation layer formed in the corrosion of the coatings containing SZMO@SG can effectively prevent the oxidation of Fe. Element and valence state analysis indicate that the passivation layer is mainly composed of metal hydroxide and iron molybdate.

The potential dynamic polarization curve is often used to characterize the anticorrosive performance of the coatings [57,58]. The coated carbon steel was soaked in 3.5 wt% NaCl solution for different times for testing. E_corr, i_corr, b_a, and b_c are the corrosion potential, corrosion current, anode slope, and cathode slope in sequence. The protection efficiency (η) can be calculated by Equation (1) [57]:

$$\mathsf{\eta }(\%)=100\%\times \left(1-\left(\frac{i_{corr,i}}{i_{corr,b}}\right)\right)$$

(1)

where i_corr,i and i_corr,b are the corrosion current of coatings with SZMO@SG and controls, respectively.

As shown in Figure 9 and

Table 2. Tafel electrochemical measurement data of different coatings for different times.

Samples	Time	Ecorr (V)	Icorr (×10−3 mA/cm2)	ba	−bc	η (%)
control	48 h	−0.586	6.71	10.166	2.603	-
	144 h	−0.628	10.69	4.489	4.420	-
	840 h	−0.599	109.70	11.200	1.964	-
1 wt% SZMO@SG	48 h	−0.550	6.91	5.130	4.417	−3.0
	144 h	−0.598	8.74	8.904	2.990	18.2
	840 h	−0.583	11.63	8.487	3.397	89.4
3 wt% SZMO@SG	48 h	−0.453	1.71	4.581	2.921	74.5
	144 h	−0.459	3.56	5.330	2.995	66.7
	840 h	−0.682	7.97	4.449	5.984	92.7
5 wt% SZMO@SG	48 h	−0.519	3.76	7.943	3.380	43.9
	144 h	−0.573	10.08	13.861	3.282	4.9
	840 h	−0.646	11.26	9.996	4.642	89.7

, the addition of SZMO@SG had a significant effect on the potentiodynamic polarization curves and corrosion current of the coatings. As the addition of SZMO@SG increased, the corrosion potential increased from −0.586 V for control to −0.453 V for coatings with a 3 wt% composite filler. The corrosion current decreased with the increase in SZMO@SG. When the addition of SZMO@SG was 3 wt%, the 48 h corrosion current reached the lowest value of 1.71 × 10⁻³ mA/cm², while the corrosion current of control reached 6.71 × 10⁻³ mA/cm². The unique lamellar structure of SZMO@SG obstructed the channels of corrosive medium in coatings, blocking the penetration of H₂O, O₂, and ions. Moreover, the molybdate ions reacted with Fe³⁺ and Fe²⁺, forming iron molybdate passivation films on carbon steel [49,59], which effectively reduced the corrosion current. The change in corrosion potential of the SZMO@SG coatings from 48 h to 840 h may reflect the transformation of the corrosion mechanism from mainly anode passivation to both cathode and anode passivation. The metal substrate of coatings with 3 wt% SZMO@SG was confirmed to have metal hydroxide deposited on it, based on the XPS (Figure 8). Moreover, the cathodic part of Tafel curve moved towards negative potential, which indicates that the hydroxide deposited on the metal substrate cathode could hinder the corrosion reaction of cathode. In addition, coatings with 5 wt% SZMO@SG have the larger I_corr value than that of 3 wt%, indicating that the increase in coating defects and ion channels reduces the anticorrosion performance of coatings.

EIS is a common method for evaluating the corrosion resistance of coatings. Previous studies [60,61] reported that the size of capacitive arc reactance has a great correlation with the anticorrosion performance of coatings. As shown in Figure 10, it can be seen that the capacitance arc radius increases with the increase in SZMO@SG, and reaches its highest value when the addition of SZMO@SG reaches 3 wt%. However, too much SZMO@SG will increase the coating defects and reduce the anticorrosion performance of coatings. The EIS results are consistent with the results of the 3.5 wt% NaCl immersion test. It is easy to confirm that the protective effect of the coatings and the capacitance arc radius constantly decreases with the increase in soaking time, but coatings with 3 wt% SZMO@SG have the largest capacitance arc radius.

As shown in the Bode phase angle diagram (Figure 11), all the coatings have incomplete time constants in the high frequency region and complete time constants in the mid- and low-frequency region after immersion. This is because the corrosive medium has reached the metal substrate. According to Figure 12, in the Bode plots of 48 h and 144 h, the 3 wt% SZMO@SG coating always has the largest impedance modulus in the high-frequency region. It is attributed to the physical barrier of SZMO@SG. However, in the bode plots of 840 h, the impedance modulus of 3 wt% SZMO@SG coatings decreases significantly due to the penetration of electrolyte. In the low-frequency region, the coatings with 3 wt% SZMO@SG have the highest impedance |Z| of more than 10⁵ Ω (48 h), and it still reached more than 6000 Ω after being immersed in a 3.5 wt% NaCl solution for 840 h.

In order to further study the anticorrosion mechanism of coatings, an equivalent circuit diagram (Figure 13) was obtained by fitting the data in Figure 10. The fitting parameters are shown in

Table 3. Electrochemical impedance spectra data of different coatings for different immersion times.

Samples	Time	Rs (Ω·cm2)	CPEf (Ω−1 s−ncm−2)	nf	Rf (Ω·cm2)	CPEdl (Ω−1 s−ncm−2)	ndl	Rct (Ω·cm2)	P (%)
control	48 h	5.0	1.87 × 10−8	1.00	289.1	5.26 × 10−5	0.52	8608	0.75
	144 h	2.9	2.64 × 10−6	0.71	99.5	8.35 × 10−4	0.68	7021	1.41
	840 h	3.6	5.35 × 10−7	0.92	17.9	9.48 × 10−4	0.51	2195	3.43
1 wt% SZMO@SG	48 h	4.0	2.55 × 10−8	0.96	310.9	7.29 × 10−6	0.71	5730	0.78
	144 h	3.4	1.80 × 10−8	1.00	173.8	1.45 × 10−4	0.50	4964	1.46
	840 h	5.2	1.29 × 10−8	1.00	483.6	2.12 × 10−4	0.46	7405	0.82
3 wt% SZMO@SG	48 h	3.7	2.42 × 10−9	1.00	3201.0	4.01 × 10−6	0.56	283,100	0.01
	144 h	6.4	1.38 × 10−8	0.93	819.0	4.08 × 10−5	0.56	104,500	0.02
	840 h	4.9	9.05 × 10−8	0.94	127.5	1.97 × 10−4	0.56	51,910	0.32
5 wt% SZMO@SG	48 h	4.5	3.46 × 10−8	1.00	110.2	2.73 × 10−6	0.82	20,870	0.17
	144 h	5.5	2.02 × 10−7	0.89	49.6	5.02 × 10−4	0.52	7786	0.75
	840 h	5.7	3.38 × 10−9	1.00	1124	6.21 × 10−5	0.64	1421	4.68

. In the equivalent circuit, R_s, R_f, and R_ct are solution resistance, film resistance, and charge transfer resistance, respectively. CPE_f and CPE_dl represent film capacitance and double layer capacitance, respectively. After soaking for 48 h, it can be seen that the R_f values of the SZMO@SG coatings are generally higher than the control, because the flake structure can form better physical barrier effect and prevent the penetration of corrosion medium. However, the R_f of 5 wt% SZMO@SG coatings is significantly lower than 3 wt% SZMO@SG coatings. It can be confirmed that there were more corrosion medium channels in the 5 wt% SZMO@SG coatings. The defects inside the 5 wt% SZMO@SG coatings led to the generation of corrosion medium channels. To further study the porosity of coatings, Equation (2) was used to calculate the porosity of coatings [62,63].

$$P=\frac{R_{ps}}{R_p}{10}^{-\frac{\mathsf{\Delta }{E}_{corr}}{b_a}}$$

(2)

In the equation, P is porosity, R_ps is the polarization resistance of the substrate, R_p is the polarization resistance of the samples, ΔE_corr is the potential difference between the uncoated and the coated substrate, b_a is the anodic Tafel coefficient of the substrate. The electrochemical parameter of carbon steel can be obtained from the related reference [64,65]. In Table 3, the 3 wt% SZMO@SG coatings have the lowest porosity. After immersion for 840 h, the porosities of the control and the 5 wt% SZMO@SG coatings had significantly increased. However, the porosity of 3 wt% coating was only 0.32%. It can be confirmed that the size and number of pores increased inside the control and 5 wt% coatings. It also can be proved that SZMO@SG can form a physical barrier inside the coatings.

CPE_f is related to electrolyte penetration in the coatings. The lowest CPE_f further indicates minimal penetration of the corrosive medium into the 3 wt% SZMO@SG coatings. In Table 3, R_ct of coatings with SZMO@SG increased significantly. After soaking for 48 h, with the increase in SZMO@SG content, R_ct increased from 8608 Ω·cm² (control) to 283,100 Ω·cm² (3 wt% fillers), and then decreased. This proves that the passivation layer is generated at the interface between the coatings and the substrate, which can block ion and charge transfer. The formation of a passivation film is attributed to the reaction of molybdate with Fe²⁺ and Fe^3+, forming FeMoO₄ and Fe₂(MoO₄)₃. R_ct of 3 wt% and 5 wt% SZMO@SG coatings increased significantly compared to other coatings, proving the high reaction activity of Fe ion and molybdate. The passivation film was stable and insoluble; even after being immersed for 840 h, R_ct of 3 wt% SZMO@SG coatings still reached 51,910 Ω·cm². It can be proved that the passivation film plays a dominant role in late corrosion resistance. Due to the corrosion medium channels inside 5 wt% SZMO@SG coatings, most of the molybdate was released into the external solution, leading to a slight reaction between the molybdate and Fe ions. Therefore, R_ct of the 5 wt% SZMO@SG coating was much smaller than the 3 wt% SZMO@SG coating.

3.3. Anticorrosion Mechanism of Polyurethane Coatings with SZMO@SG

From the above discussion it can be concluded that the appropriate dosage of SZMO@SG can effectively improve the corrosion resistance of polyurethane coatings. The anticorrosion mechanism of SZMO@SG coatings was shown in Figure 14. At the initial stage of immersion, the lamellar structure of SZMO@SG forms a physical barrier in the coatings, which can effectively prevent the infiltration of corrosion medium. In the middle and late stages of soaking, MoO₄²⁻ and Zn²⁺ generated by SZMO@SG dissolution can capture Fe²⁺, Fe³⁺, and OH⁻, forming passivation layer. The passivation layer can also prevent carbon steel from contact with the corrosion medium. The reaction for the forming of the passivation layer is as follows [43]:

2Fe³⁺ + 3 MoO₄²⁻ + n H₂O → Fe₂(MoO₄)₃·n H₂O

(3)

Fe²⁺ + MoO₄²⁻ + n H₂O → FeMoO₄·n H₂O

(4)

Zn²⁺ + 2 OH⁻ → Zn(OH)₂

(5)

Zn(OH)₂ generated by the hydrolysis of SZMO@SG can adsorb Na⁺ and Cl⁻ [32,41], which can further improve the anticorrosion performance of the coatings, confirmed in Figure 7. Therefore, SZMO@SG has three anticorrosive functions: forming physical barrier, passivating the metal substrate, and absorbing Na⁺ and Cl⁻. In contrast to traditional anticorrosive fillers, SZMO@SG can significantly improve the anticorrosive performance of waterborne polyurethane coatings.

4. Conclusions

In this study, SZMO@SG was successfully prepared and mixed with hydroxy acrylate latex, significantly improving the anticorrosive performance of waterborne polyurethane coatings. The SZMO@SG fillers have a lamellar structure with the optimized SG addition of 1.5 wt%. The optimized corrosion resistance of coatings was obtained when the dosage of SZMO@SG is 3 wt%, but higher filler addition will diminish the anticorrosion performance of coatings. For the optimized coatings with 3 wt% SZMO@SG, the corrosion current density has the lowest value of 1.71 × 10⁻³ mA/cm², and the maximum charge transfer resistance of the coatings reaches 283,100 Ω·cm². Even after soaking in a 3.5 wt% NaCl solution for 840 h, its R_ct still reached 51,910 Ω·cm². Through the EIS and metal substrate XPS, it can be confirmed that the passivation film was deposited on the interface between coatings and metal substrate. The passivation film blocks the ion and charge transfer, and prevents Fe from oxidation. Furthermore, the cross-section of the coatings EDS confirms the Na⁺ adsorption of SZMO@SG. SZMO@SG is a novel anticorrosion filler with several anticorrosion functions, which could be further applied in other waterborne organic coatings.