1. Introduction
Metal is an indispensable material in modern industry, but it is often plagued by corrosion. Research on the anti-corrosion technology of metals has received increasing attention, with environmentally friendly anti-corrosion coatings being the focus [
1]. However, it is an undeniable fact that chromium-containing anti-corrosion coatings are still widely used in industry, despite their toxicity. The insulating coating for oriented silicon steel faces this dilemma. Due to the requirements of the rolling process, the coating needs to be both high-temperature- and corrosion-resistant. There has never been a commercially available chromium-free anti-corrosion coating in this field. The reason behind this is that the anti-corrosion technology of chromium-free coatings is less effective than that of chromium-containing coatings.
Anti-corrosion coatings are typically composed of film-forming substances and functional fillers. Defects can occur in the coating due to the release of small molecular substances during the curing process and cross-linking condensation. Although there have been studies using monomers to optimize the curing process and reduce defects [
2], the curing mechanism of water-based coatings determines that coating defects are difficult to avoid. The use of functional fillers has become a common choice for researchers. The unique anti-corrosion mechanism of chromium salt fillers makes them effective in addressing the problem of coating defects, even though hexavalent chromium salts (Cr
6+) used in chromium-containing coatings are highly toxic [
3,
4]. Cr
6+ forms a three-dimensional anti-corrosion mechanism by intervening in cathodic and anodic reactions and generating passivating substances through valence changes. The anti-corrosion mechanism is based on the mixed oxides containing trivalent chromium (Cr
3+) and Cr
6+. During the curing of coating and at corrosion sites, Cr
3+ rapidly forms a passivation layer with ferrous ions (Fe
2+), iron ions (Fe
3+), and O
2−. When the coating is damaged, soluble Cr
6+ is released, coupled with a reduction reaction at the cathode and a metal dissolution reaction at the anode, generating new Cr
3+ and passivation substances to repair coating defects [
5,
6,
7,
8,
9]. It is worth noting that although it is considered to have self-healing properties, the conversion of Cr
6+ to Cr
3+ is an irreversible behavior in the anti-corrosion mechanism. When Cr
6+ is completely released, the coating will lose its self-healing performance [
10]. It can be seen that the key to achieving the anti-corrosion effect of chromium-like salt is to use valence conversion to block both the cathodic and anodic reactions of electrochemical corrosion and form multiple anti-corrosion mechanisms.
Chromium-free functional fillers can also improve corrosion resistance by filling defects and preventing electrochemical reactions [
5]. Based on this, functional fillers can be classified into filling and consumption types. The filling type can delay the penetration of corrosive media and usually has a low density and high stability, such as micro/nano-materials like silica [
11], fiber material [
12], and graphene [
13,
14,
15]. However, the compatibility between functional fillers and the coating is limited, and excessive addition can lead to a deterioration in corrosion resistance. The consumption type can intervene in corrosion reactions, interrupting the original electrochemical corrosion process, such as by replacing anodes [
16,
17,
18] or forming passivation layers through reactions with fFe
2+ and Fe
3+ [
19,
20]. The typical representative of the former is zinc (Zn) in zinc-rich coatings. Zn in zinc-rich coatings not only acts as a sacrificial anode but can also be converted into zinc oxide (ZnO) through reactions to fill coating defects. The Zn content is very important for the anti-corrosion performance of the zinc-rich coating. If the amount added is too small, a conductive path cannot be formed between the fillers to connect with the substrate, and it is impossible to ensure that the metal substrate is at the cathode. At this point, Zn will become the starting point of the pitting area. In reports related to fillers with passivation effects, the molybdate (MoO
42−) and phosphomolybdate (PMo
12O
403−) are particularly noteworthy, as Mo is a congener of Cr. Nonetheless, current research suggests that their mechanism is different from that of Cr. The action mode of MoO
42− is to compete with chloride ions (Cl
−) in the corrosive medium through stronger cation adsorption, react with Fe
2+ and Fe
3+, form a passivation layer, and exhibit a certain degree of self-healing function [
10,
21,
22]. However, its oxidation is weak, and the passivation layer cannot efficiently modify defects in the same manner as the chromium passivation layer. MoO
42− can form both homopolyacid salts and heteropolyacid salts [
23], among which PMo
12O
403− has attracted more attention [
24,
25]. Its mechanism is identical to that of MoO
42− [
26], and its corrosion resistance is better than that of MoO
42−. This is because PMo
12O
403− has stronger oxidability than MoO
42−, and the passivation layer forms more rapidly under acidic conditions [
27]. Related studies also indicate that the formation of PMo
12O
403− requires specific reaction conditions [
28,
29]. To ensure that the structure of PMo
12O
403− is retained in the final coating, other researchers have first prepared PMo
12O
403− and then added it to the coating [
30,
31]. When comparing the mechanisms of molybdenum and chromium salts, the singularity of the molybdenum salt mechanism is the fundamental reason for the difference in effectiveness. However, considering oxidation as a key factor affecting the anti-corrosion performance of molybdenum salts is also one of the reasons limiting their performance improvement. The Mo in MoO
42− and PMo
12O
403− is hexavalent (Mo
6+), and this oxidation limit restricts the design of functional fillers. For this reason, previous research has adopted composite fillers to enhance the release of molybdenum salts [
32,
33,
34]; however, the application system is limited, and the anti-corrosion performance still needs improvement. In short, compared with other chromium-free functional fillers with similar passivation mechanisms, the advantage of molybdenum salt fillers is that MoO
42− and PMo
12O
403− have stronger cation adsorption, which can form a denser passivation layer and weaken the negative impact of Cl
− on the anti-corrosion process. However, compared with chromium salt fillers, molybdenum salt fillers have a more passive anti-corrosion mechanism; they enter the passivation layer formation process after the formation of Fe
2+ or Fe
3+. The multiple anti-corrosion mechanisms of chromium salt fillers make them more active and able to actively intervene in corrosion electrochemical reactions after the penetration of corrosive media. Therefore, making the mechanism of molybdenum salt fillers more active by introducing low-valance Mo to consume the corrosive media would be an effective method to improve the anti-corrosion performance. In fact, multivalent molybdenum materials have been applied in fields such as energy [
35,
36], biological medicine [
37,
38], and information technology [
39], and their mechanisms utilize the conversion between high and low valances in multivalent molybdenum systems. Some researchers have also observed that the presence of low-valance molybdenum contributes to the anti-corrosion performance of the molybdenum-based conversion coatings [
40]. The low-valence molybdenum in their coatings is not the original intention of formulation design but rather formed in small amounts during passivation treatment. Unlike passivation treatment, the curing process of water-based coatings does not spontaneously produce low-priced molybdenum, so it is necessary to actively form multivalent molybdenum salt fillers.
Referring to the mechanism of chromium salts, the reductant was introduced during the preparation of the molybdenum salt to construct a multivalent molybdenum salt system in this manuscript. By introducing low-priced Mo, the mechanism of action of molybdenum salt fillers can be shifted from passive corrosion prevention by reacting with Fe2+ or Fe3+ to a more active mechanism that is similar to chromium salts, which simultaneously consumes corrosive media and forms a passive layer. The influence of the valence composition of molybdenum in multivalent MoO42− systems on the corrosion resistance of chromium-free insulating coatings on oriented silicon steel was studied. The valence composition of molybdenum in the multivalent MoO42− and multivalent PMo12O403− systems and their corrosion resistance were compared, which aids in attaining an in-depth understanding of the anti-corrosion mechanism of molybdenum.
3. Results and Discussion
To investigate the corrosion resistance of the PMZ coating on oriented silicon steel sheets, the NSS test was conducted. The time required for the coating to reach a 5% corrosion area was recorded.
Figure 1 illustrates the effect of the Zn mass fraction on the corrosion resistance duration of the PMZ coating. The corrosion resistance duration exhibits an oscillatory trend concerning the Zn mass fraction. In this experiment, the Na
2MoO
4 was consistently maintained at 0.15 wt%. When the mass of Zn added was 3 mg, the corrosion resistance of the coating was weaker than that of the coating without Zn. When the mass of Zn added was 6 mg, the best corrosion resistance was observed. Subsequently, the further increase in zinc content reduced the corrosion resistance of the coating. When the mass of Zn added exceeded 12 mg, the effect of zinc on corrosion resistance tended to stabilize without significant changes. When the mass of Na
2MoO
4 used was consistent, as designed, with the increase in Zn dosage, the amount of low-valence molybdenum in the coating would gradually increase. It can be hypothesized that when 3 mg of Zn was added, the small amount of low-valence molybdenum acted as a localized point of corrosion initiation, thus weakening the coating’s corrosion resistance. This phenomenon is similar to that observed in zinc-rich coatings, where insufficient Zn content fails to provide adequate corrosion protection. Obviously, this analysis has limitations. It can be postulated that, similar to zinc-rich coatings, insufficient amounts of low-valence molybdenum acting as sacrificial anodes may reduce the coating’s corrosion resistance. However, the observation that increasing the Zn dosage, which should increase the low-valence molybdenum, also detracts from corrosion resistance suggests that the initial hypothesis does not fully explain the experimental results.
To better understand the change in the corrosion resistance of this system, a more detailed analysis of the molybdenum species and valence states in the PMZ system is necessary. XPS spectra and Mo 3d high-resolution spectra were measured to investigate the molybdenum species and valence states in the PMZ system.
Figure 2a,c display the XPS full spectra of the PMZ
0 and PMNZ
0.06 systems. It can be observed that the peak shapes for PMZ
0 and PMNZ
0.06 are similar, with prominent peaks for O 1s, C 1s, and P 2p. The Al 2p peak is also noticeable. However, the Mo 3d and Na 1s peaks are relatively weak, and a very weak Zn 2p peak is also detected in the PMNZ
0.06 sample. These findings align with the composition of the PMZ system. The PMZ system was prepared using Al(H
2PO
4)
3 solution, Na
2MoO
4, and water. Since Na
2MoO
4 was added in smaller amounts, the signal from Al(H
2PO
4)
3 elements is stronger, while the signal from Na
2MoO
4 elements is weaker. Additionally, because the proportion of Al in the Al(H
2PO
4)
3 solution is lower than that of P, the intensity of the Al 2p peak is weaker than that of the P 2p peak.
The fitting results are shown in
Figure 2b,d. Despite the high spectral noise and large fluctuations due to the small amount of Na
2MoO
4− added to the PMZ system, the peak shapes of the PMZ
0 and PMZ
0.06 samples are clearly different. The binding energy range of the accompanying peaks in the PMZ
0.06 spectrum is lower than that of the accompanying peaks in the PMZ
0 spectrum. This indicates that the molybdenum species in PMZ
0.06 differ from those in the PMZ
0 system. In
Figure 2b, the molybdenum species are identified as MoO
42− and Na
2MoO
4·2H
2O. In the PMZ
0 system, molybdenum exists solely as MoO
42−. During the vacuum drying process, some Na
2MoO
4 adsorbed crystallization water that was not fully expelled. This is reasonable considering that anhydrous Na
2MoO
4 is more difficult to prepare, and Na
2MoO
4·2H
2O is a more commonly used reagent. It is plausible that crystallization water remains in the sample during the short-duration vacuum drying process. After calculation, the relative content of MoO
42− in the PMZ
0 sample is 41.13%, while the content of Na
2MoO
4·2H
2O is 58.87%. Based on
Figure 2d, the PMZ
0.06 sample contains both MoO
42− and Mo
3+, indicating that the molybdenum in the PMZ
0.06 system exists in both the MoO
42− and Mo
3+ forms. This also confirms that the addition of Zn causes the MoO
42− in the PMZ system to be reduced to Mo
3+, as shown in the following reaction equation [
43](1):
The source of H
+ in the reaction is the free H
3PO
4 present in the Al(H
2PO
4)
3 solution. To analyze the effect of Zn addition on the composition of the PMZ system, the Mo 3d high-resolution spectra of all PMZ samples were fitted. The results are summarized in
Table 1. As shown in
Table 1, there is Mo
3+ in all PMZ samples except PMZ
0, and Na
2MoO
4·2H
2O is no longer present. Based on the amount of Zn added to the PMZ system, the effect of Zn addition on the relative proportion of Mo
3+ is illustrated in
Figure 3. In
Figure 3, the relative proportion of Mo
3+ increases with the addition of Zn up to a certain point, then decreases. When the Zn addition exceeds 0.012 g, the relative proportion of Mo
3+ stabilizes. The inflection point of the curve occurs at a Zn addition of 0.006 g, corresponding to the PMZ
0.06 sample.
During the preparation of the PMZ system, it was observed that gas was generated after adding Zn. Due to the high viscosity of the Al(H
2PO
4)
3 solution, the addition of a small amount of Zn tends to accumulate at the bottom of the container, and the color change is more easily observed than the gas generation. Although gas production is slower compared to the color change, it can be observed throughout the reaction. The endpoint of the reaction is marked not only by the cessation of the color change but also by the absence of further gas evolution. This indicates that Zn is involved in reactions beyond that mentioned in Reaction (1). Possible additional reactions in the system may include the following reaction:
Reactions (1) and (2) represent a competitive relationship. In the PMZ series systems, the amount of Na
2MoO
4 added is small. When the amount of Zn added exceeds a certain point, more Zn participates in the reaction described by Reaction (2), consuming H
3PO
4 and H
+ in the system. This reduces the available H
+, thereby slowing down Reaction (1). Consequently, as the amount of Zn increases, the Mo
3+ content in the PMZ system decreases. It is noteworthy that the characteristics presented in
Figure 3 explain the changes in the corrosion resistance of the PMZ series coatings. The corrosion resistance of the coating decreases with increasing Zn content when the Zn content exceeds 0.06 wt‰ due to the decrease in the Mo
3+ content in the PMZ system and, consequently, in the PMZ coating. The reduction in the functional component (Mo
3+) leads to a loss in the corrosion resistance of the coating.
Therefore, the use of Zn can partially reduce MoO42− in acidic systems, where Mo6+ is reduced to Mo3+. As the mass of Zn added to the PMZ0 system increases, the content of Mo3+ first increases and then decreases. The anti-corrosion performance of the coating increases with the increase in Mo3+ content. When the mass of Zn is 6 mg, the relative proportion of Mo3+ in the system reaches 20.52%, and the corrosion resistance time in NSS is as long as 248 h.
In the anti-corrosion coatings industry, PMo
12O
403− is also a typical representative of molybdenum salt fillers. For comparison, a PMNZ
1 coating was prepared using the obtained multivalent PMo
12O
403− system. To ensure consistency in the amounts of Al(H
2PO
4)
3 and Mo in the comparison system and facilitate the comparison of molybdenum salt species and valence states, the composition of PMZ was adjusted and labeled as PMZ’. The prepared PMZ’
0.1 system was light brown, while the PMNZ
1 system was dark blue. The resulting PMZ’
0.1 coating was light yellow, while the PMNZ
1 coating was light blue. Using an XPS analysis, the Mo composition in PMZ’
0.1 was found to be 87.34% MoO
42− and 12.62% Mo
3+, which is consistent with the composition characteristics of the PMZ system (
Figure S1). The primary components in both coatings, such as Al(H
2PO
4)
3, SiO
2·nH
2O, and additives, were used in the same amounts, but the Zn content differed slightly. The Zn mass fraction in PMZ’
0.1 was 0.0094%, while in PMNZ
1, it was 0.0038%. Compared to other factors, the effect of Zn on the corrosion resistance of the coatings was relatively small. At this point, the comparison focuses on the impact of the form of molybdenum salt on the coating’s corrosion resistance.
Figure 4 shows the morphology of the oriented silicon steel sheets coated with base, PMNZ
1, and PMZ’
0.1 coatings and their morphologies after 192 h of the NSS test.
Figure 4a–c present the photos before the NSS test. It can be observed that although both the PMNZ
1 and PMZ’
0.1 coatings show some coloration, their steel sheets are similar to those with the base coating. The coatings are relatively thin, approximately 1 μm, and the markings from the rolling process are still visible. After 192 h of the NSS test, the base coating exhibited the largest corrosion area, reaching 16%. The corrosion area of the PMNZ
1 coating is smaller but still exceeds 5%, reaching 8.5%. Most of the corrosion points are light yellow, which is characteristic of the early stages of corrosion. The PMZ’
0.1 coating, however, shows a corrosion area of less than 5%, reaching 2.5%, indicating that the coating still possesses excellent corrosion resistance. Among the three coatings, the PMZ’
0.1 coating demonstrated the best corrosion resistance. It is noteworthy that the areas of corrosion on both the PMNZ
1 and PMZ’
0.1 coatings did not develop along the laser-engraved marks, which were introduced during the rolling process to reduce iron loss. Typically, when the corrosion resistance of a coating is poor, these marks expose part of the metal substrate, and rust often propagates along the exposed substrate. However, in the case of the PMZ’
0.1 and PMNZ
1 coatings, the rust did not spread along the etched marks, indicating that, despite the metal substrate experiencing some exposure, the coatings still provided effective corrosion protection.
The PMo
12O
403− is generally considered to exhibit superior corrosion resistance [
30]. However, in the comparison between the PMNZ
1 and PMZ’
0.1 coatings, the PMZ’
0.1 coating demonstrated better corrosion resistance. To further understand the reason behind this, a compositional analysis of the PMNZ and PMZ’ systems was conducted. Initially, the FT-IR was used for a preliminary analysis of the PMNZ
0 and PMZ’
0 samples. The results are shown in
Figure 5. The dark curve represents the PMZ’
0 system, and the red curve represents the PMNZ
0 system. Both spectra exhibit peaks at 3400 cm
−1 and 1640 cm
−1, corresponding to the stretching and bending vibrations of –OH groups. In the PMNZ
0 spectrum, peaks at 1380 cm
−1, 1059 cm
−1, and 752 cm
−1 are observed, which correspond to the asymmetric stretching, symmetric stretching, and in-plane bending vibrations of the nitrate ion (NO
3−). Additionally, peaks at 929 cm
−1 and 862 cm
−1 represent Mo-O vibrations in the PMo
12O
403− structure [
44,
45,
46,
47,
48]. The peak at 800 cm
−1 is also believed to correspond to Mo-P-Mo bonding [
49]. In contrast, in the PMZ’
0 spectrum, peaks at 1127 cm
−1, 987 cm
−1, and 509 cm
−1 are observed, which are attributed to the symmetric stretching of P-O
2, symmetric stretching of P-O
3, and asymmetric bending vibration of P-O, respectively [
50]. Notably, there are no peaks corresponding to the PMo
12O
403− structure in the PMZ’
0 spectrum, while these characteristic peaks are present in the PMNZ
0 spectrum. This suggests that the forms of Mo in the two systems are different.
Based on these observations, XPS was used to further analyze the PMNZ system, as shown in
Figure 6. By comparing
Figure 6a with
Figure 2a, it is observed that the main elements in the two spectra differ. In the PMNZ
0 spectrum, strong Mo 3d, Na 1s, N 1s, C 1s, and O 1s peaks are evident, while the intensity of the P 2p peak is very low, and the Al 2p peak is not detected. This corresponds well with the composition characteristics of the PMNZ
0 system, which is prepared from Na
2MoO
4, H
3PO
4, HNO
3, and water. Since H
3PO
4 is used in small amounts, the signals from Na
2MoO
4 and HNO
3 are stronger, while the signal from H
3PO
4 is weak, and the Al signal cannot be detected. In
Figure 6b, the orange dot-dashed line representing the original data is smoother and less fluctuating compared to
Figure 2b. This is because, in the PMZ system, Na
2MoO
4 is added in small amounts, resulting in weak detection signals and higher spectral noise, while in PMNZ
0, Na
2MoO
4 is added in larger quantities, leading to stronger signals and smoother spectra. Additionally, in
Figure 6b, the forms of Mo are MoO
42− and PMo
12O
403−, with the peak area representing PMo
12O
403− being larger. This indicates that in the PMNZ
0 system, Na
3PMo
12O
40 was successfully synthesized, with PMo
12O
403− being the predominant species in the system, although MoO
42− is also present.
Figure 6c,d presents the XPS full spectrum and Mo 3d high-resolution spectrum for the PMNZ
1 system. Comparing
Figure 6a with
Figure 6c, it can be observed that the elemental characteristics of PMNZ
1 are consistent with those in
Figure 6a. Since the amount of Zn added is relatively small compared to other components, the Zn 2p peak signal is weak. When comparing
Figure 6c with
Figure 2c, we notice that the Mo 3d original spectrum for PMNZ
1 is smoother, whereas the Mo 3d spectrum for PMZ
0.06 has more noise. This difference is influenced by the lower addition of Na
2MoO
4 in the PMZ
0.06 system. Based on the fitted spectra, MoO
42−, PMo
12O
403−, and Mo
5+ are present in PMNZ
1.
The phenomenon corresponding to the analysis above can also be observed in the P 2p high-resolution of the PMZ
0.06 and PMNZ
1 systems (
Figure S2). In
Figure S2a, the spectrum of P 2p is relatively smooth, consisting of peaks of P 2p
3/2 and P 2p
1/2 representing the hydrogen phosphate salt (H
2PO
4−). This is consistent with the composition characteristics of the PMZ
0.06 system, where the addition of Al(H
2PO
4)
3 solution is higher and exists in the form of H
2PO
4−. In
Figure S2b, the spectrum of P 2p exhibits significant fluctuations, and two groups of peaks can be identified during fitting. One represents phosphate ions (PO
43−), and the other represents PMo
12O
403−. This further confirms the presence of PMo
12O
403− in the PMNZ system. In addition, the spectral differences among
Figure 6c,
Figure 2c, and
Figure S1b suggest that although the Mo is hexavalent in both MoO
42− and PMo
12O
403−, the reduction of these species leads to different products. For PMo
12O
403−, the reaction is described in Reaction (3).
PMo
12O
403− appears blue after reduction, also known as molybdenum blue. It is a complex mixture composed of PO
43−, Mo
6+, and Mo
5+. PMo
12O
363− is a possible expression of the mixture. To clearly compare the relative proportions of Mo chemical states before and after reduction in the PMNZ system, the fitting results for the PMNZ
0 and PMNZ
1 systems are summarized in
Table 2. In PMNZ
0, the relative proportion of PMo
12O
403− reached 81.56%, which confirms that using the preparation method described in
Section 2.2.2 yields a solution predominantly consisting of Na
3PMo
12O
40. In contrast, according to
Table 1, PMZ
0 contains 41.13% MoO
42− and 58.87% Na
2MoO
4·2H
2O.
It is noteworthy that despite the PMNZ0 system containing more Na2MoO4, the fitting results show that the PMZ0 sample contains Na2MoO4·2H2O, whereas no Na2MoO4·2H2O was detected in the PMNZ0 sample. This difference can be attributed to the high viscosity of the Al(H2PO4)3 solution in the PMZ series, which makes it more difficult for water to evaporate during vacuum drying. The Al(H2PO4)3 in the outer layer of droplets forms a membrane, which further traps moisture inside, leading to the retention of crystallization water. In contrast, the PMNZ series primarily contains Na2MoO4 and water, which have lower viscosity and allow for the faster evaporation of moisture during vacuum drying. Additionally, there is no membrane formation during the drying of the outer layer of droplets in the PMNZ series, and this will not interfere with the escape of inner moisture, thereby preventing the presence of crystallization water. In PMNZ1, there are 37.12% MoO42−, 30.51% PMo12O403−, and 32.37% Mo5+. Compared to PMNZ0, the proportion of MoO42− has increased, the proportion of PMo12O403− has decreased, and Mo5+ has appeared. This indicates that Zn is indeed able to reduce Mo6+ in the PMNZ system to Mo5+. The increase in the proportion of MoO42− can be attributed to two factors: Firstly, the reduction reaction breaks the structure of PMo12O403−, and during this process, part of Mo6+ is reduced to Mo5+, while the remainder remains in the MoO42− form. Secondly, the addition of Zn increases the pH of the solution, which can also cause the structure of PMo12O403− to break down. Under the combined effects of these two factors, the final proportion of MoO42− in PMNZ1 is higher than that in PMNZ0.
Therefore, although Zn can also reduce the Mo in PMo12O403− systems, the obtained low-valence Mo is Mo5+. Under the conditions of the same main composition of the coating and the same amount of Na2MoO4 used, the PMZ system containing Mo3+ has better corrosion resistance. This is different from the common understanding of the anti-corrosion mechanisms of MoO42− and PMo12O403−, indicating that the mechanism of molybdenum salt functional fillers still needs to be improved.
The process of establishing and applying the multivalent molybdate system is illustrated in
Figure 7. The addition of Zn to both the PMo
12O
403− and MoO
42− systems indeed reduces Mo
6+ to lower the valence, but the reduction products in the two systems are different. As shown in the figure, there is also a significant visual difference between the two systems. The multivalent PMo
12O
403− system appears dark blue, while the multivalent MoO
42− system appears transparent brown. In the PMo
12O
403− system, Mo
6+ is reduced to Mo
5+, and in the MoO
42− system, Mo
6+ is reduced to Mo
3+. Specifically, PMNZ
1 contains MoO
42−, PMo
12O
403−, and Mo
5+, while the PMZ’
0.1 coating contains MoO
42− and Mo
3+. Among these, PMo
12O
403− is considered to have a stronger oxidizing property than MoO
42−. From the perspective of valence, Mo
5+ is higher than Mo
3+, indicating stronger oxidation. When the same amount of Na
2MoO
4 is used, based on the current mechanisms of MoO
42− and PMo
12O
403−, the PMNZ
1 coating is expected to exhibit better corrosion resistance. However, the aforementioned NSS experimental results suggest that the use of a multivalent molybdenum salt system can optimize the corrosion resistance, and the PMZ’
0.1 coating performs better in terms of corrosion resistance. This further indicates that the oxidizing property should not be regarded as the core influencing factor of the mechanisms of MoO
42− and PMo
12O
403−.
Several possible reasons for this discrepancy can be identified. The base coating used in this manuscript is composed of silica as the main structure, while additives and various condensation products of Al(H2PO4)3 form a network that fills the inter-particle pores of the silica, resembling the continuous phase in the coating. During service, the base coating delays the penetration of corrosive media through the “labyrinth effect”, and the electrochemical process of corrosion remains unchanged. Therefore, after the corrosive media penetrate, corrosion reactions occur and form pitting corrosion, and the corrosion products are usually hydroxides or oxides containing Fe2+. Under the action of chloride ions, this layer is usually relatively loose and unable to serve as a passivation layer, allowing the corrosion to gradually spread and deepen. In contrast, the low-valence Mo in the PMNZ1 and PMZ’1 coatings can react with the corrosive media during the formation of the passivation layer, oxidizing to Mo6+. This process, as shown by the possible reaction equation in the figure, implies that after the corrosive media penetrate, although corrosion occurs in the localized areas where the corrosive liquid accumulates, the corrosive media are also consumed by the oxidation reaction. The corrosion products precipitate with the PMo12O403− or MoO42− to form a passivation layer. As a result, both the anodic and cathodic reactions of the electrochemical corrosion process are hindered, thereby enhancing the corrosion resistance of the coating. Mo3+ has stronger reducibility compared to Mo5+, allowing it to react more effectively with the corrosive media, thus slowing down the corrosion during the passivation layer formation phase. This leads to the formation of a more compact passivation layer. Ultimately, the PMZ series coatings, which contain lower-valence Mo, exhibit better corrosion resistance. It is worth noting that the function of low-valence molybdenum is to consume oxygen, which is similar to the sacrificial anode mechanism of Zn in the zinc-rich coating. It can be found that low-valence molybdenum also has the defect of damaging the anti-corrosion performance of the coating when the addition amount is too small. This is reflected in the poorer anti-corrosion performance of PMZ0.03, PMZ0.12, and PMZ0.15 coatings compared to PMZ0 coating.
When compared with other types of chromium-free fillers and coatings, the PMZ system and coating also have prominent advantages.
Table 3 presents the coatings mentioned in recent reports on chromium-free anti-corrosion coatings. The comparison mainly involves the types of coatings, types of functional fillers, filler content, film thickness, types of metal substrates, and NSS duration. The amount of filler added to the PMZ coating here is expressed in terms of the mass of Na
2MoO
4 and Zn used.
PMZ coating is a water-based inorganic coating. Compared with water-based semi-organic coatings containing resins, water-based inorganic coatings have better heat resistance and avoid the presence of volatile organic compounds (VOCs). Meanwhile, although PMZ coating does not have the longest corrosion resistance time, it uses fewer fillers and has a thinner film thickness, and the corrosion resistance of the oriented silicon steel used is inferior to other metal substrates. This indicates that PMZ coating is lower-cost and has more efficient anti-corrosion performance.