Corrosion Resistance of Mild Steel Coated with Phthalimide-Functionalized Polybenzoxazines

Abstract

Herein, we synthesized two new phthalimide-functionalized benzoxazine monomers, pPP-BZ and oPP-BZ, through Mannich reactions of 2-(4-hydroxyphenyl)isoindoline-1,3-dione (pPP) and 2-(2-hydroxyphenyl)isoindoline-1,3-dione (oPP), respectively, with p-toluidine and paraformaldehyde. The structures of these two monomers were confirmed using Fourier transform infrared (FTIR) and nuclear magnetic resonance spectroscopy. We used differential scanning calorimetry, FTIR spectroscopy, and thermogravimetric analysis to study the polymerization behavior and thermal stability of the monomers and their corresponding polybenzoxazines. Poly(pPP-BZ) and poly(oPP-BZ) were formed on mild steel (MS) through spin-coating and subsequent thermal curing polymerization. We used various corrosion testing methods to examine the effect of the curing temperature on the corrosion resistance of the coated MS samples in 3.5 wt.% aqueous solution of NaCl. Among our tested systems, the corrosion rate reached a low of 2.78 µm·Y⁻¹ for the MS coated with poly(pPP-BZ)₁₈₀ (i.e., the coating that had been cured at 180 °C); this value is much lower than that (4.8 µm·Y⁻¹) reported for a maleimide-based benzoxazine compound (MI-Bz)/33 wt.% ACAT (amine-capped aniline trimer) blend. Thus, the incorporation of the imide functional group into the PBZ coatings is an effective strategy for affording high-performance corrosion resistance.

1. Introduction

Intra and intermolecular hydrogen bonds among their heterocyclic polymer chains impart polybenzoxazines (PBZs) with many useful properties, including high thermal stability, high glass transition temperatures, high char yields, near-zero shrinkage upon polymerization, low water absorption, low flammability, excellent mechanical and dielectric properties, low surface free energies, and high water contact angles (WCAs) [1,2,3,4,5,6,7,8,9,10]. Benzoxazine monomers, which are generally prepared through Mannich reactions of phenols, primary amines, and formaldehyde, can polymerize simply through thermal curing without any catalyst and without releasing any by-products during their ring-opening polymerization (ROP) [11,12,13,14,15,16,17,18,19,20]. Aromatic polyimides are also high-performance polymers that display high thermal stability, excellent mechanical properties, and high chemical resistance [21]; their inert imide rings and strong interchain interactions make them useful in, for example, plastics, films, laminating resins, insulating coatings, and high-temperature structural adhesives [22]. The high degrees of molecular design flexibility and functionalization of benzoxazine monomers can be exploited to improve the performance of these monomers and their resulting PBZs, expanding their opportunities for potential applications. For example, the introduction of a sulfonic acid unit into the benzoxazine backbone has led to corresponding PBZs displaying good thermal stability, low methanol permeability, and excellent acid resistance, making them promising materials for proton exchange membranes in direct methanol fuel cells [23]. Furthermore, PBZ/polyimide-blend films have exhibited good toughness and thermal stability [24]. Organic coatings are often employed to protect iron and mild steel (MS) against corrosion by preventing corrosive agents from reaching the substrate and by acting as barriers against a direct current [25]. The protective barrier properties of a coating are typically characterized in terms of the rates of transport of water and ions through its polymer network [26]. The incorporation of particular functional groups into PBZ coatings can enhance their adhesion to metal substrates and improve the corrosion resistance; for example, the charge transfer resistance (R_ct) of MS coated with a silane-based PBZ (1.2891 MΩ·cm²) was higher than that of the uncoated MS (0.0396 MΩ·cm²) [27]. Recently, PBZ-based electroactive species have been used to form protective passive oxide layers on steel substrates to prevent their corrosion [28,29,30]. For example, coating MS with PBA-ddm has resulted in excellent corrosion resistance and a decrease in the corrosion current by two orders of magnitude relative to that of the uncoated MS [31]. Phthalimide materials have been evaluated as anticorrosive coatings for MS in 1 mol·L⁻¹ H₂SO₄; they displayed anticorrosion activity arising from their π-delocalized phthalimide systems and lone pairs of electrons on their N and O atoms [32]. Herein, we report the synthesis of two novel phthalimide-functionalized benzoxazine monomers, pPP-BZ and oPP-BZ. First, we prepared the phthalimide-functionalized phenols pPP and oPP by reacting phthalic anhydride with para- and ortho-aminophenol, respectively, in acetic acid. We then obtained pPP-BZ and oPP-BZ from the reactions of pPP and oPP, respectively, with p-toluidine and paraformaldehyde in 1,4-dioxane. Fourier transform infrared (FTIR) and nuclear magnetic resonance (NMR) spectroscopy confirmed the chemical structures of these benzoxazine monomers. We used differential scanning calorimetry (DSC), FTIR spectroscopy, and thermogravimetric analysis (TGA) to study the polymerization behavior of the oxazine units in these benzoxazine monomers and the thermal stability of their corresponding polymers after curing at various temperatures. Furthermore, we measured open-circuit potentials (E_ocp), polarization, and WCAs and employed electrochemical impedance spectroscopy (EIS) to investigate the effect of the curing temperature on the corrosion resistance of poly(pPP-BZ) and poly(oPP-BZ) coatings on MS in 3.5 wt.% aqueous solution of NaCl.

2. Materials and Methods

2.1. Materials

p-Toluidine (97%), o-aminophenol (98%), phthalic anhydride [C₆H₄(CO)₂O], paraformaldehyde ((CH₂O)_n, 99%), 1,4-dioxane, ethyl acetate (EtOAc), magnesium sulfate (MgSO₄), acetic acid (CH₃COOH), p-toluidine, sodium carbonate (Na₂CO₃), and chloroform (CHCl₃) were purchased from Acros. All chemicals were used as received without further purification.

2.2. Synthesis of 2-(4-Hydoxyphenyl)isoindoline-1,3-dione (pPP)

pPP was prepared according to a previously reported method [5]. Yield: 90%. M.p.: 299 °C. ¹H NMR (400 MHz, DMSO-d₆, δ, ppm, Figure S1): 9.70 (OH), 6.75–7.90 (8H, ArH). FTIR (KBr, cm⁻¹): 3413 (OH stretching), 1778, 1717 (imide I), 1397 (imide II, CN stretching), 722 (C=O bending).

2.3. Synthesis of 2-(2-Hydoxyphenyl)isoindoline-1,3-dione (oPP)

oPP was prepared according to a previously reported method [5]. Yield: 95%. M.p.: 220 °C. ¹H NMR (400 MHz, DMSO-d₆, δ, ppm, Figure S2): 6.75–7.9 (8H, ArH), 9.70 (OH). FTIR (KBr, cm⁻¹): 3379 (OH stretching), 1772, 1720 (imide I), 1389 (imide II, CN stretching), 722 (C=O bending).

2.4. Synthesis of 2-(3-(p-Tolyl)-3,4-dihydro-2H-benzo[e][1,3]oxazin-6-yl)isoindoline-1,3-dione (pPP-BZ)

A solution of p-toluidine (1.0 g, 0.0093 mol) and paraformaldehyde (0.80 g, 0.030 mol) in 1,4-dioxane (30 mL) was stirred under reflux at 70 °C for 30 min in a 100-mL round-bottom flask. pPP (2.00 g, 0.0083 mol) was added and the mixture was then heated at 110 °C for 24 h. The solution was cooled to room temperature and the solvent evaporated using a rotary evaporator. The residue was dissolved in EtOAc and washed three times with 1 N NaOH and twice with water. The organic phase EA was dried (MgSO₄) and concentrated using a rotary evaporator. Finally, the residue was precipitated in hexane (100 mL) to give a white powder: yield: 55%. FTIR (KBr, cm⁻¹): 1778, 1717 (imide I), 1494 (stretching of trisubstituted benzene ring), 1385 (imide II), 1227 (COC symmetric stretching), 1014 (CN asymmetric stretching), 925 (out-of-plane CH bending). ¹H NMR (CDCl₃, δ, ppm): 4.5 (s, ArCH₂N), 5.2 (s, OCH₂N), 6.75–7.9 (ArH, 14H). ¹³C NMR (CDCl₃, δ, ppm): 20 (CH₃), 50 (NCH₂Ar oxazine ring), 80 (OCH₂N), 116–152 (aromatic carbon nuclei), 167 (C=O).

2.5. Synthesis of 2-(3-(p-Tolyl)-3,4-dihydro-2H-benzo[e][1,3]oxazin-8-yl)isoindoline-1,3-dione (oPP-BZ)

A solution of p-toluidine (1.0 g, 0.0093 mol) and paraformaldehyde (0.80 g, 0.030 mol) in 1,4-dioxane (30 mL) was stirred under reflux at 70 °C for 2 h in a 100-mL round-bottom flask. oPP (2.0 g, 0.0083 mol) was added and then the mixture was heated at 110 °C for 6 h. The solution was cooled to room temperature and the solvent evaporated using a rotary evaporator. The residue was dissolved in EtOAc and washed three times with 1 N NaOH. The organic phase was dried (MgSO₄) and concentrated using a rotary evaporator. Finally, the residue was precipitated in hexane (100 mL) to give brown crystals. Yield: 90%. FTIR (KBr, cm⁻¹): 1772, 1720 (imide I group) 1507 (stretching of trisubstituted benzene ring), 1386 (imide II, CN stretching), 1209 (COC antisymmetric stretching), 929 (out-of-plane CH bending). ¹H NMR (400 MHz, CDCl₃, δ, ppm): 4.5 (s, ArCH₂N, oxazine), 5.2 (s, OCH₂N, oxazine), 6.8–7.8 (ArH, 14H). ¹³C NMR (400 MHz, CDCl₃, δ, ppm): 20 (Me group in p-toluidine), 50 (NCH₂Ar), 80 (OCH₂N), 119–150 (aromatic carbon nuclei), 166 (C=O).

2.6. Preparation of poly(pPP-BZ) and poly(oPP-BZ)

The monomers pPP-BZ and oPP-BZ were polymerized through thermally induced ROP to form phenolic structures featuring a Mannich base bridge (CH₂NRCH). The cross-linking of each sample was performed using the following curing program: 150 °C (2 h), 180 °C (2 h), 210 °C (2 h); slow cooling to ambient temperature.

2.7. Preparation of MS Samples Coated with poly(pPP-BZ) and poly(oPP-BZ)

The low-carbon MS used in this study had the following chemical composition: 0.19% C, 0.94% Mn, 0.05% Si, 0.022% Cu, 0.009% P, 0.004% S, 0.014% Ni, 0.034% Al, 0.009% Cr, 0.016% V, 0.003% Ti, and 98.709% Fe [33]. The MS surface was polished with emery paper to remove its surface oxide layer. The MS surface was washed with EtOH and acetone and then dried. The monomers pPP-BZ and oPP-BZ were dissolved separately in CHCl₃ at 100 g·L⁻¹. Each solution was spin-coated onto the MS surface and then dried at room temperature for 24 h. The coated samples were cured at temperatures of 150, 180, and 210 °C for 2 h for each curing stage. The thickness (ca. 5 µm) of the coating was determined by measuring each MS sample with a micrometer caliper before and after coating. The coated samples are named herein as poly(pPP-BZ)₁₅₀, poly(pPP-BZ)₁₈₀, poly(pPP-BZ)₂₁₀, poly(oPP-BZ)₁₅₀, poly(oPP-BZ)₁₈₀, and poly(oPP-BZ)₂₁₀.

2.8. Corrosion Tests

The corrosion resistance of each of the prepared coatings (cured at 150, 180, or 210 °C) was examined in terms of open-circuit potentials (E_ocp), potentiodynamic polarization, and EIS, performed using a Voltalab 40 PGZ 301 instrument controlled by VoltaMaster 4 software (Radiometer Analytical SAS, Lyon, France). All samples were immersed in 3.5 wt.% aqueous solution of NaCl at ambient temperature for 30 min. The electrochemical cells included a saturated calomel electrode (SCE) as the reference electrode, a platinum wire as the counter electrode, and an uncoated or coated MS sample as the working electrode. The scanning potential of polarization measurements was ±250 mV of the open-circuit potential at a scan rate of 0.2 mV·s⁻¹. EIS was performed in the frequency range between 100 kHz and 0.01 Hz; EIS analysis was performed using ZSimpwin software. The electrochemical experiments were performed three times for each tested sample to obtain greater accuracy and reproducibility of the experimental results, and mean values were recorded.

3. Results and Discussion

3.1. Synthesis of pPP-BZ and oPP-BZ

The para- and ortho-phthalimide-functionalized benzoxazine monomers were synthesized through reactions of the phthalimide-functionalized phenols pPP and oPP, respectively, p-toluidine, and paraformaldehyde in 1,4-dioxane, as presented in Figure 1c,d. The ring-closing reaction of oPP-BZ monomer was much easier than pPP-BZ, because of the existence of intermolecular hydrogen bond in oPP, which accelerates the ionization OH group during stages forming of oPP-BZ monomer, this behavior was reported elsewhere [5]. The ROP of the monomers pPP-BZ and oPP-BZ as shown in Figure 2. The structures of pPP-BZ and oPP-BZ were confirmed using ¹H NMR spectroscopy (Figure 3a and Figure 4a). The signals of the aromatic protons were located at 6.75–7.9 ppm. The signals of the ArCH₂N and OCH₂N units of the oxazine rings were located at 4.5 and 5.2 ppm, respectively [34,35,36]. The signal of the methyl group appeared at 2.01 ppm. Figure 3b and Figure 4b present the ¹³C NMR spectra of the monomers pPP-BZ and oPP-BZ. The signals of the carbon nuclei of the ArCH₂N and OCH₂N units of the oxazine rings were located near 50 and 80 ppm, respectively. The signals for the symmetric C=O groups of the imide moieties were centered at 166.55 ppm. The signals of the aromatic rings appeared in the range of 119–150 ppm.

Furthermore, the structures of pPP-BZ and oPP-BZ monomers were confirmed using FTIR spectroscopy (Figure 5 and Figure 6). The signals at 1778 and 1717 cm⁻¹ for pPP-BZ and at 1772 and 1720 cm⁻¹ for oPP-BZ represent asymmetric and symmetric stretching, respectively, of the C–C(=O)–C units of their imide groups [37]. For both benzoxazine monomers, the signals at 1227 and 1209 cm⁻¹ represent the aromatic ether units (C–O–C asymmetric stretching), while the bands at 925 and 929 cm⁻¹ represent out-of-plane bending of the benzene rings attached to oxazine rings [38]. All these results suggested that we successfully synthesized benzoxazine monomers pPP-BZ and oPP-BZ.

We used DSC and FTIR spectroscopy to study the ROP of the monomers pPP-BZ and oPP-BZ (Figure 2). Figure 7a and Figure 8b display DSC thermograms of pPP-BZ and oPP-BZ, respectively: uncured and after curing at 150, 180, and 210 °C. For the uncured pPP-BZ, the endothermic peak (melting point), exothermic peak (ROP), and heat of polymerization were 175 °C, 234 °C, and 49.42 J·g⁻¹, respectively. The sharp and high melting peak indicated good purity for the phthalimide-modified benzoxazine monomer and its high rigidity (cf. other traditional benzoxazine monomers) as a result of the attachment of the imide group to the aromatic structure. For the uncured oPP-BZ, the endothermic peak was not evident because it partially overlapped the exothermic peak, which had shifted to 205 °C [5]. In this study, the exothermic peaks appeared at temperatures lower than those of some other reported imide-functionalized benzoxazines that have been used in the synthesis of high-performance materials [39]. The exothermic peaks of pPP-BZ and oPP-BZ disappeared completely after curing at 180 °C, indicating their complete ROP. Furthermore, the glass transition temperatures (T_g) of poly(pPP-BZ) and poly(oPP-BZ) after curing at 180 and 210 °C were obtained as mean values of the three repeated measurements and the standard deviation (SD) was calculated in

Table 1. DSC and TGA results for pPP-BZ and oPP-BZ before and after thermal treatment.

Samples	Curing Temperature (°C)	Td5 (°C)	Td10 (°C)	Char Yield (%)	Tg/DSC (°C)
pPP-BZ	RT	179	215	13	-
	150	244	286	26	-
	180	259	286	33	214 ± 1
	210	296	321	37	219 ± 1
oPP-BZ	RT	219	245	40	-
	150	234	256	42	-
	180	255	279	44	212 ± 1
	210	264	289	48	217 ± 1

. As shown in Table 1, the T_g values of poly(pPP-BZ) and poly(oPP-BZ) after curing at 180 and 210 °C were 214, 219 °C, and 212 and 217 °C, respectively. Thus, our new PBZs had values of T_g higher than those previously reported for the cross-linked material NDOPoda Bz (192 and 205 °C after curing at 180 and 210 °C, respectively) [40]. We attribute the high values of T_g of our polymers to their high densities of intra- and intermolecular hydrogen bonds between the phenolic OH groups and the N atoms in the Mannich bridges or the C=O groups in the imide structures.

We also used FTIR spectroscopy to study the thermal ROP behavior of the monomers pPP-BZ and oPP-BZ (Figure 7b and Figure 8b, respectively). For pPP-BZ and oPP-BZ, bands appeared at 1227 and 1209 cm⁻¹, respectively, for C–O–C antisymmetric stretching and at 925 and 929 cm⁻¹, respectively, for out-of-plane bending of the benzene rings attached to oxazine rings. After curing at temperatures from 180 to 210 °C, the signals representing the oxazine rings of the monomers disappeared progressively and broad bands representing phenolic OH groups appeared in the range 3500–3300 cm⁻¹, consistent with the formation of poly(pPP-BZ) and poly(oPP-BZ) (Figure 7b and Figure 8b).

We used TGA to investigate the thermal stability of the uncured pPP-BZ and oPP-BZ and their corresponding polymers obtained after curing at 150, 180, and 210 °C (Figure 9a,b, Table 1). We considered the temperatures of 5 and 10% weight loss (T_d5 and T_d10, respectively) to characterize the thermal stabilities of the monomers and corresponding PBZs. The values of T_d5 and T_d10 and the char yields increased progressively upon increasing the curing temperature, due to the formation of highly cross-linked thermosets. For poly(pPP-BZ) after curing at 210 °C, the values of T_d5 and T_d10 and the char yield at 800 °C were 296 and 321 °C and 37 wt.%, respectively; for poly(oPP-BZ) after curing at 210 °C, these values were 264 and 281 °C, and 48 wt.% respectively. Thus, the thermal stabilities of our new polymers poly(pPP-BZ) and poly(oPP-BZ) cured at 210 °C were higher than that of Pp-IB, a monofunctional benzoxazine featuring an imide group (286 and 322 °C and 32 wt.%, respectively) [41]. The thermal stability and char yields of poly(oPP-BZ) (formed from the ortho-monomer) were much higher than those of poly(pPP-BZ) (formed from the para-monomer), because of a higher degree of cross-linking and a higher density of intramolecular hydrogen bonds.

3.2. Surface Properties of MS Coated with poly(pPP-BZ) and poly(oPP-BZ)

The surface hydrophobicity and adhesion of a polymer coating are characteristics that affect its long-term use and corrosion resistance performance [32]. Figure 10a,b displays the surface hydrophobicities of the uncoated MS and of the MS coated with poly(pPP-BZ)₁₅₀, poly(pPP-BZ)₁₈₀, poly(pPP-BZ)₂₁₀, poly(oPP-BZ)₁₅₀, poly(oPP-BZ)₁₈₀, and poly(oPP-BZ)₂₁₀ before exposure to a corrosive medium, as determined using WCA measurements. The results as mean values and their SD are summarized in Table S1. The uncoated MS had a hydrophilic surface with a WCA of 70° ± 1°, suggesting that it would readily be attacked by corrosive agents. The curing temperature had a slight influence on the WCA of the prepared coating samples, but all of the coated samples had hydrophobic surfaces (WCAs: >90°), which we attribute to the low surface free energies of PBZ coatings [25,42,43,44]. Hydrophobic phthalimide-functionalized PBZ coatings have been demonstrated previously to decrease the wettability of corrosive media and provide strong barrier effects [34]. Among our tested materials, the poly(pPP-BZ)₁₈₀ and poly(oPP-BZ)₁₈₀ coatings provided the highest WCAs of 102° and 98°, respectively, because of their high degrees of hydrogen bonding and high densities of cross-linking decreasing the permeability of water to the MS surface [45]. Our poly(pPP-BZ)₂₁₀ and poly(oPP-BZ)₂₁₀ coatings had slightly lower WCAs of 95° and 93°, respectively, presumably, because their hydrogen bonding was ruptured at the high curing temperature [46]. Furthermore, we used the scalpel-cutting test to evaluate the adhesion of the poly(pPP-BZ) and poly(oPP-BZ) coatings cured on the MS surfaces at 150, 180, and 210 °C (Figure S3) [47]. In general, the adhesion of poly(pPP-BZ) on MS was better than that of poly(oPP-BZ), because of the greater degree of the hydrogen bonding between the OH groups of the MS surface and the cured coating. The cut edges of the poly(pPP-BZ)-coated samples were smooth and free from any peeling arising from the cutting (Figure S3a–c). In contrast, in the case of the poly(oPP-BZ) coatings, some peeling occurred along with the cutting after being slicing with the scalpel (Figure S3d–f). Nevertheless, the adhesion between each coating and the MS surface was strong, presumably because of electrostatic interactions between the positively charged N atoms (imide and Mannich bridge) and the negatively charged MS surface as well as hydrogen bonding between the cured PBZ coating and the OH groups of the MS surface.

3.3. Corrosion Resistance of MS: Uncoated and Coated with poly(pPP-BZ) and poly(oPP-BZ)

We used the open-circuit potentials (E_ocp), thermodynamic polarization and electrochemical impedance spectroscopy (EIS) measurements to evaluate the corrosion resistance of uncoated and MS coated with poly(pPP-BZ) and poly(oPP-BZ) that cured at 150, 180, and 210 °C in a 3.5 wt.% NaCl solution at room temperature. Figure 11a,b represents the E_ocp-t plots of uncoated and coated MS with poly(pPP-BZ) and poly(oPP-BZ) that cured at 150, 180, and 210 °C. From this figure, it was observed that the E_ocp value of the uncoated MS shifted to a more negative potential after immersion time from about −0.530 to −0.718 V. A sharp shift toward much more negative potential is due to the formation of the corrosion on the MS electrode [48]. In contrast, the E_ocp values for poly(pPP-BZ) coatings cured at 150, 180, and 210 °C were stable with immersion time in which their start values were −0.460, −0.520, and −0.530 V then becomes −0.521, −0.527, and −0.568 V, respectively and same behavior was observed for poly(oPP-BZ) coatings, indicating the formation of protective film on MS surface. In general, high values of E_ocp clearly indicate that the coatings provide high corrosion resistance [49], and it was observed that the E_ocp values of coatings that cured at 210 °C decreased a little, which can be attributed to the increase in their porosity with a higher permeability for corrosive types. We used potentiodynamic polarization to evaluate the corrosion resistance and inhibition of the uncoated and coated MS samples in 3.5 wt.% NaCl aqueous solutions at room temperature.

Figure 12a,b presents the anodic and cathodic polarization curves of the uncoated and coated MS at various curing temperatures in 3.5 wt.% NaCl.

Table 2. Potentiodynamic polarization data (± standard deviation (SD)) of uncoated MS and of MS coated with poly(pPP-BZ)₁₅₀, poly(pPP-BZ)₁₈₀, poly(pPP-BZ)₂₁₀, poly(oPP-BZ)₁₅₀, poly(oPP-BZ)₁₈₀, and poly(oPP-BZ)₂₁₀ in 3.5% NaCl.

Coating Samples	Ecorr mV	icorr µA·cm−2	βa mV·dec−1	βc mV·dec−1	CR µm/Y	RP kΩ·cm2	η %
uncoated MS	−803.0	12.44 ± 0.09	282 ± 2	−228 ± 2	145 ± 1	3.68	-
poly(pPP-BZ)150	−738.3	0.46 ± 0.08	218 ± 2	−319 ± 2	5 ± 1	145.09	96.3
poly(pPP-BZ)180	−737.6	0.24 ± 0.05	343 ± 2	−342 ± 2	3 ± 1	222.56	98.1
poly(pPP-BZ)210	−762.7	0.31 ± 0.02	287 ± 2	−289 ± 2	4 ± 1	200.84	97.5
poly(oPP-BZ)150	−699.6	0.91 ± 0.07	227 ± 2	−331 ± 2	11 ± 1	40.74	92.8
poly(oPP-BZ)180	−676.6	0.61 ± 0.06	318 ± 2	−374 ± 2	7 ± 1	100.89	95.1
poly(oPP-BZ)210	−730.4	0.79 ± 0.03	296 ± 2	−358 ± 2	9 ± 1	77.49	93.7

lists the electrochemical parameters determined from the Tafel plots: the corrosion current (i_corr), corrosion potential (E_corr), anodic (β_a), and cathodic (β_c) Tafel slopes, corrosion rate (CR), and polarization resistance (R_p) [50,51].

The CR and the value of R_p were calculated using the following equations [52,53].

$$R_{\mathrm{p}}=\frac{{\mathsf{\beta }}_{\mathrm{a}}\times {\mathsf{\beta }}_{\mathrm{c}}}{2.3\times i_{\mathrm{corr}}\left({\mathsf{\beta }}_{\mathrm{a}}+{\mathsf{\beta }}_{\mathrm{c}}\right)}$$

(1)

$$CR=\frac{i_{\mathrm{corr}}\times K\times M_{\mathrm{wt}}}{\mathsf{\rho }\times A}$$

(2)

where i_corr is the corrosion current density, K is the corrosion rate constant (3.272 µm Y⁻¹), M_wt is the molecular weight of iron, $\mathsf{\rho }$ is the density of MS (7.85 g cm⁻³), and A is the sample area (cm²). We calculated the protection efficiency (ɳ) using the equation [54].

$$\mathsf{\eta }=\frac{i_{c\mathrm{orr}}^{°}-i_{\mathrm{corr}}}{i_{\mathrm{corr}}^{°}}\times 100$$

(3)

where $i_{c\mathrm{orr}}^{°}$ and i_corr are the corrosion currents of the uncoated and coated MS samples, respectively.

Figure 12a displays the Tafel plots of the uncoated MS and of the MS coated with poly(pPP-BZ) and cured at 150, 180, and 210 °C; Figure 12b presents the Tafel plots of the uncoated MS and the MS coated with poly(oPP-BZ). Although the Tafel behavior of both poly(pPP-BZ) and poly(oPP-BZ) was the same, the corrosion protection of poly(pPP-BZ) was superior to that of poly(oPP-BZ). The maximum corrosion protection occurred for both poly(pPP-BZ)₁₈₀ and poly(oPP-BZ)₁₈₀. The anodic and cathodic Tafel lines shifted toward the noble direction for the MS coated with both poly(pPP-BZ) and poly(oPP-BZ), when compared with the uncoated MS. Thus, the poly(pPP-BZ) and poly(oPP-BZ) coatings affected the anodic reaction mechanism, rather than the cathodic reaction mechanism. This behavior was presumably due to the compact cross-linked structures of the poly(pPP-BZ) and poly(oPP-BZ) coatings curbing the anodic reaction on the MS surfaces and, thereby, inhibiting the entire corrosion process in the system [55]. The corrosion current density decreased in the presence of poly(pPP-BZ) or poly(oPP-BZ) when compared with the uncoated MS surface, presumably because of the increase in the corrosion protection efficiency of the poly(pPP-BZ) and poly(oPP-BZ) coatings. The corrosion protection efficiency of the MS coated with poly(pPP-BZ)₁₈₀ was higher than those of the MS coated with poly(pPP-BZ)₁₅₀ and poly(pPP-BZ)₂₁₀; the same behavior was evident in the cases of the poly(oPP-BZ) coatings. The corrosion protection efficiency of the poly(pPP-BZ) coatings on the MS surface was superior to that of the poly(oPP-BZ) coatings in NaCl solution, because of the poly(pPP-BZ) coatings adsorbing more strongly through greater coordinate bonding with the MS substrate [56,57,58,59]. Also, when the curing temperature of coatings increased to 210 °C, the corrosion resistance decreased a little because of decreasing hydrophobic performance [53], and also increase in volume during curing at higher temperatures might be responsible for a decrease in compactness in polybenzoxazine. Table 2 reveals that the CRs of the MS samples coated with poly(pPP-BZ) and poly(oPP-BZ) were about one to two orders of magnitude lower than that of the uncoated MS, indicating that the polymers coatings adsorbed on the MS surfaces inhibited corrosion. Among our tested systems, the CR reached its lowest value of 2.78 µm·Y⁻¹ for the poly(pPP-BZ)₁₈₀ coating; indeed, this value is lower than that of PBA-a-180 °C (5.75 µm·Y⁻¹) [60]. In addition, the polarization resistance (R_p), defined as the resistance of the coatings toward oxidation during the application of an external potential [61], of all of our coatings on MS increased by approximately two to three orders of magnitude when compared with that of the uncoated MS. Among our systems, the values of R_p reached their maxima of 222.56 kΩ·cm² for the poly(pPP-BZ)₁₈₀ coating and 100.89 kΩ·cm² for the poly(oPP-BZ)₁₈₀ coating. The hydrophobic phthalimide-functionalized PBZ coatings having crosslinked polymer networks could decrease the penetration permeability of electrolytes through the coatings and increase the barrier against the corrosion reaction, because of the PBZ coatings possessing low surface free energies resulting from their hydrogen-bonding networks [62,63]. In addition, we suspect that the π-electrons of the aromatic rings and the heteroatoms containing lone pairs of electrons (N and O atoms in the oxazine ring and imide structures) assisted in the strong adsorption of the polymers onto the MS, leading to efficient protection of the metal in corrosive media [64]. EIS is a useful technique for gaining information about a coating–substrate interface because the impedance measurements are very sensitive to the corrosion mechanism [45].

Figure 13a,b displays the Nyquist plots of the uncoated MS surface and of the MS coated with poly(pPP-BZ) and poly(oPP-BZ) in 3.5 wt.% aqueous NaCl at room temperature. For the MS coated with poly(pPP-BZ) and poly(oPP-BZ), the semicircle was depressed and its diameter—related to the charge transfer resistance (R_ct)—was greater than that in the plot of the uncoated MS. The Nyquist plots had the same shape for the uncoated MS and the coated samples that had been cured at the various temperatures, but with larger diameters indicating increased resistivity. These features reveal the mechanism of corrosion of the MS after coating. According to the theory of EIS, at low frequency, the loops displayed non-ideal semicircles. This variation can be explained by considering the non-ideal behavior of the double layers as a capacitor; this phenomenon was due to the frequency dispersion with relation to roughness and the non-homogeneity of the MS surface [64].

All of the EIS data were fitted using the equivalent circuit models in Figure S4a,b that describe the electrochemical system; a one-time constant model (Figure S4a) was used for the uncoated MS, and a two-time constant model Figure S4b was applied for the MS with the poly(pPP-BZ) and poly(oPP-BZ) coatings. Here, C_dl is the capacitance of the double-layer capacitance (related to the electrolyte–uncoated steel substrate interface); R_p is the corrosion resistance of the coating (related to the electrolyte–coating interface), R_ct is the charge transfer resistance, and Q_dl is the double-layer capacitance (related to the electrolyte–substrate interface) [65]. The experimental data were fitted using the program ZSimpWin.

Table 3. EIS data (±SD) of uncoated MS and of MS coated with poly(pPP-BZ)₁₅₀, poly(pPP-BZ)₁₈₀, poly(pPP-BZ)₂₁₀, poly(oPP-BZ)₁₅₀, poly(oPP-BZ)₁₈₀, and poly(oPP-BZ)₂₁₀ in 3.5% NaCl.

Coating Samples	Qc (µΩ−1·sn·cm−2)	n1	Rp (kΩ·cm2)	Qdl (µΩ−1·sn·cm−2)	n2	Rct (kΩ·cm2)	η1 (%)
uncoated MS	-	-	-	1.238 ± 0.009	0.77	0.35 ± 0.08	-
poly(pPP-BZ)150	2.7 × 10−2	0.38	0.479 ± 0.031	0.084 ± 0.008	0.67	8.22 ± 0.01	95.7
poly (pPP-BZ)180	2.1 × 10−3	0.71	0.084 ± 0.002	0.078 ± 0.005	0.50	29.55 ± 0.03	98.3
poly(pPP-BZ)210	1.4 × 10−2	0.57	0.010 ± 0.005	0.081 ± 0.004	0.60	11.67 ± 0.01	97.0
poly(oPP-BZ)150	6.0 × 10−2	0.38	0.039 ± 0.004	0.118 ± 0.006	0.71	4.98 ± 0.05	92.9
poly(oPP-BZ)180	8.9 × 10−3	0.51	4.381 ± 0.055	0.069 ± 0.003	0.58	9.84 ± 0.04	96.4
poly(oPP-BZ)210	3.9 × 10−2	0.57	0.341 ± 0.061	0.093 ± 0.002	0.43	6.40 ± 0.05	94.5

lists the fitted EIS parameters. The estimated impedance, in this case, Z_CPE, was determined as follows [64]:

$$Z_{\mathrm{CPE}} = \frac{1}{Q_{\mathrm{dl}}{\left(j\mathsf{\omega }\right)}^n}$$

(4)

where the exponent $n$ is the phase shift, j² = −1, ω = 2πf.

The double-layer capacitance (C_dl) along with a CPE was estimated as reported previously [64]:

$$C_{\mathrm{dl}}=Q_{\mathrm{dl}}{\left({\mathsf{\omega }}_{\max }\right)}^{n-1}$$

(5)

where, ω_max = 2πf_max and f_max is the frequency at the maximal impedance-imaginary component. The MS corrosion protection efficiency (η_I) was obtained as follows [66]:

$${\mathsf{\eta }}_{\mathrm{I}}=\frac{R_{\mathrm{ct}-}{R}_{\mathrm{ct}}^{°}}{R_{\mathrm{ct}}}\times 100$$

(6)

where, R_ct and $R_{\mathrm{ct}}^{°}$ are the charge transfer resistances for the coated and uncoated systems, respectively.

The double-layer capacitances (Q_dl) of the poly(pPP-BZ) and poly(oPP-BZ) coatings were approximately two orders of magnitude lower than that of uncoated MS, indicating the adsorption of the coatings on the MS surface; the charge transfer resistance (R_ct) increased by approximately two orders of magnitude relative to that of the uncoated MS, indicating the formation of an insulating protective layer at the electrolyte–substrate interface [67]. Among our tested samples, poly(pPP-BZ)₁₈₀ had the highest value of R_ct of 29.55 kΩ·cm²; this value is higher than that obtained previously for a P(BA-aptms) coating without epoxy (8.65 ± 3.39 kΩ·cm²) in 3.5% NaCl [25]. In contrast, the poly(oPP-BZ)₁₈₀ coating had the lowest value of Q_dl of 0.069 ± 0.03 µΩ⁻¹·sⁿ·cm². A higher value of R_ct and a lower value of Q_dl always correspond to better corrosion resistance performance, because of the inhibited access of water molecules to the metal surface (through the adsorption of the organic molecules) and a less active dissolution reaction [29,68]. In addition, we suspect that the change in the coating resistance (R_p) was mainly due to the change in the penetration permeability of the electrolyte through the porous coating. From Table 3, it was observed, when the curing temperature of the coating increased to 210 °C, its Q_dl value increases a little. This behavior can be attributed to increasing the internal stress of coating which led to the formation cracks or pores in the coating [69]. Figure 14a,b shows the Bode and phase angles plots, and indicates that the interfacial impedance of MS coated with pPP-BZ and oPP-BZ increased about two orders of magnitude more than those of uncoated MS in the low-frequency range, demonstrating reinforced barrier property of pPP-BZ and oPP-BZ coatings on MS. The Phase angle is also significant to detect the performance of the coating surface. Two-time constants at different frequencies were observed for each coated MS samples cured at 150, 180, and 210 °C. Generally, EIS results for each the tested MS coated were well consistent with the results in Tafel curves. The corrosion resistance performance of organic coatings depends greatly on their surface morphology. Figure S5 shows the SEM surface images of uncoated and MS coated with poly (pPP-BZ)₁₈₀. The surface of coated MS with poly(pPP-BZ)₁₈₀ is uniform, smooth, and not affected by salt solution attack (Figure S5c). On the other hand, the same morphology was not observed for uncoated MS surface (Figure S5b). The scratches on the surface of MS are due to the pre-polishing using the emery papers (Figure S5a).

4. Conclusions

We have synthesized the phthalimide-functionalized benzoxazine monomers pPP-BZ and oPP-BZ from reactions of pPP and oPP, respectively, with p-toluidine and paraformaldehyde in 1,4-dioxane. Their corresponding polymers obtained after thermal curing had high values of T_g (>212 °C), as determined through DSC analysis. The poly(oPP-BZ) series had thermal stabilities and char yields higher than those of the poly(pPP-BZ) series, because of the greater crosslinking density and degree of intramolecular hydrogen bonding in the former after ring-opening of the oxazine units. The corrosion resistance of MS coated by poly(pPP-BZ) coatings in a 3.5 wt.% NaCl solution was better than the MS coated by poly(oPP-BZ) coatings. This was due to the strongly adsorbing of poly(pPP-BZ) coatings through coordination interaction with MS surfaces. Furthermore, the corrosion resistance of the prepared hydrophobic coatings on MS in 3.5 wt.% NaCl was enhanced significantly after curing at temperatures from 150 to 210 °C; the corrosion rate reached its lowest value of 3 µm·Y⁻¹ for the poly(pPP-BZ)₁₈₀ coating (cf. 145.40 µm·Y⁻¹ for the uncoated MS). Finally, incorporating imide structures into benzoxazine coatings appears to be a good strategy for affording high-performance corrosion resistance.