Introduction

Self-healing materials inspired by living tissues have attracted considerable attention owing to their prolonged lifetime, enhanced reliability, and durability1,2,3, and have been employed in many fields, such as anticorrosion coatings4,5,6. In contrast to conventional coatings, self-healing anticorrosion coatings can repair damage and recover their anticorrosion function7. Their main function is the autonomous protection of the metal surface from corrosion via reconstruction of the barrier or release of corrosion inhibitors to passivate defects in the metal substrates8,9,10. This strategy does not involve any human intervention, which is critical for the long-term performance of metallic materials in corrosive environments.

Self-healing anticorrosion coatings generally utilize the extrinsic strategy of imbedding microcapsules or nanocontainers containing self-healing agents or corrosion inhibitors4,11. In the microcapsule systems, healing agents—such as tung oil12, silyl ester13 and isophorone diisocyanate14—encapsulated by a polymer shell—such as polyurethane/poly(urea-formaldehyde)15—are released and polymerized to form a protective layer and seal the scratch. For nanocontainer systems, corrosion inhibitors have been preloaded in halloysite nanotubes16,17,18, mesoporous silica nanoparticles19,20, zeolitic imidazolate framework21,22, polymer microspheres23,24 or nanofibers25. Upon the onset of corrosion triggered by the mechanical pressure, pH modification, light irradiation, etc., the corrosion inhibitors are released into the defect areas and autonomously protect the metal surface26. Recently, an improvement in self-healing anticorrosion coatings was related to the efficient release of corrosion inhibitors and the formation of a durable protective film that seals the damaged area by constructing composites27,28,29,30,31,32. Since incorporating additives into coatings generally compromises their integrity and destructs their barrier and mechanical properties7, the intrinsic self-healing process relies on the recombination of dynamic interactions, which makes the use of additives unnecessary. Non-covalent (such as hydrogen bonds33,34, metal-ligand coordination bonds35, host-guest interactions36, and ionic bonds37) or covalent bonds (such as disulfide bond38, Diels-Alder reaction39, and transesterification reaction40) are utilized to construct intrinsic self-healing polymer coatings. However, external intervention, such as heating, is generally required for intrinsic self-healing to proceed41,42, which poses a challenge to the development of an autonomous self-healing anticorrosion coating using intrinsic polymers that can restore anticorrosion properties without human intervention.

Herein, in-situ self-crosslinking strategy of intrinsic polymers for preparing autonomous self-healing anticorrosion coating is proposed. When the prepared polymers was damaged, the metal substrate corroded and generated Fe3+. Subsequently, dynamic metal coordination crosslinks were formed between the Fe3+ and catechol groups in the polymer chain, thus restoring the anticorrosion properties of the coating. Compared to extrinsic self-healing anticorrosion coatings, the intrinsic polymer coating in this study was capable of restoring anticorrosion performance without any additives or external human intervention.

Results and discussion

Synthesis and characterization of intrinsic self-healing anticorrosion polymers (ISAP)

Previous self-healing anticorrosion coatings consist of a polymer matrix and embedded self-healing microcapsules. When the coating is damaged, the released self-healing agents or corrosion inhibitors from capsules accumulate at the breakage and inhibit corrosion. Supplementary Fig. 1 illustrates the ISAP coating in this work. The designed ISAP repairs the breakage through the cross-linking of catechol in the polymer chain with Fe3+ produced from the damaged region of the metal. The molecular structure and synthesis process of the ISAP are shown in Fig. 1 and Supplementary Fig. 2. A catechol-containing chain extender was synthesized and characterized. Its chemical structure was confirmed using 1H NMR and 13C NMR spectroscopy (Supplementary Figs. 3 and 4)43. The polymer was synthesized using α,ω-aminopropyl-terminated poly(dimethylsiloxane) (PDMS), isophorone diisocyanate (IPDI) and the synthesized chain extender (Supplementary Fig. 2).

Fig. 1
figure 1

Molecular structure of the intrinsic self-healing anticorrosion polymer (ISAP).

Full size image

Table 1 lists the compositions, molecular weights and polydispersity index (PDI) of the ISAP, where their chemical compositions were adjusted by varying the feeding ratios of PDMS, chain extender, and IPDI. Their molecular weights range from 2.48 to 3.18 × 104 g mol−1 and PDI range from 1.42 to 2.22. Their molecular structures were further confirmed by UV–vis and Fourier-transform infrared (FT-IR) spectroscopy (Fig. 2, Supplementary Figs. 5 and 6, Supplementary Table 1). In Fig. 2a, ISAP-0 shows a flat line with no absorption peaks, while the samples containing chain extender show absorption peaks at 285 nm, corresponding to the characteristic absorbance peak of catechol44. The intensity of the absorption peaks gradually increased with increasing chain extender content. These results indicate that the catechol group was successfully introduced into the polymer, and the group content was controllable. Furthermore, in Fig. 2b, an absorption peak between 500 and 700 nm is observed for ISAP-2 with added Fe3+ at pH of 10, which indicated the formation of a bis-/tris-complex.

Table 1 Summary of the formula and molecular weight of intrinsic self-healing anticorrosion polymers (ISAP).
Full size table
Fig. 2: The chemical component and coordination of the intrinsic self-healing anticorrosion polymers (ISAP) with Fe3+.
figure 2

a Ultraviolet–visible (UV–vis) absorbance of the intrinsic self-healing anticorrosion polymers (ISAP), b UV–vis absorbance of the ISAP-2 and ISAP-2 with the addition of Fe3+ at pH = 10.

Full size image

Raman spectroscopy was performed to demonstrate the catechol-Fe3+ coordination. As shown in Fig. 3, a clear peak at 644 cm−1 was observed for ISAP-2, which represents the Fe–O bond vibrations of the catechol-metal complex45,46. In addition, resonance peaks at 1266, 1322, 1423, and 1428 cm−1 were observed, resulting from catechol ring vibrations, according to previous reports45,46,47. It clearly confirmed the polymer was able to form cross-links via the coordination of catechol with Fe3+.

Fig. 3: The formation of catechol-Fe3+ coordination.
figure 3

The Raman spectra of the intrinsic self-healing anticorrosion polymers ISAP-2 with Fe3+ at pH = 10.

Full size image

Mechanical property of the intrinsic self-healing anticorrosion polymers

The mechanical properties of the ISAP were further studied using tensile tests, the results for which are shown in Fig. 4. For ISAP-0, the tensile strength of the sample was 2.75 MPa, the Young’s modulus was 4.12 MPa, and the toughness was 9.03 MJ cm-3. With the increase in chain extender content, the tensile strength gradually increased from 3.25 to 5.16 MPa, and Young’s modulus increased from 6.06 to 10.40 MPa. The toughness of ISAP-2 and ISAP-3 was comparable at ~11 MJ cm−3, which is higher than that of ISAP-0. These results indicate that the introduced urea-hydrogen bonds and the rigid benzene ring structure in the polymer chain contribute to a gradual increase in the tensile strength, modulus, and toughness of the materials.

Fig. 4: Mechanical properties of the intrinsic self-healing anticorrosion polymers (ISAP).
figure 4

a Stress-strain curves, b tensile strength, c Young’s modulus, and (d) toughness of the intrinsic self-healing anticorrosion polymers (ISAP) (error bars stand for the standard deviations from three independent samples).

Full size image

The self-healing properties of the ISAP samples were further investigated. Dumbbell-shaped specimens were cut off, realigned, and soaked in an FeCl3 solution (0.01 mol L-1) for different durations. As seen in Fig. 5, the stress-strain curve of ISAP-1 gradually coincided with that of the original sample, and the tensile strength was gradually restored with increasing immersion time. Additionally, the self-healing efficiency of ISAP-0 was 45.9% after five days of immersion, while ISAP-1 reached as high as 67.9%. This is because the catechol group at the fracture interface of ISAP-1 was coordinated and crosslinked with Fe3+, and the fracture interface was repaired. The above results indicate that the as-prepared ISAP samples can be repaired by the dynamic crosslinking of the polymer chains with Fe3+.

Fig. 5: The self-healing performance of the intrinsic self-healing anticorrosion polymers (ISAP).
figure 5

a Strain-stress curves of the original and self-healing ISAP-1 specimens after 24, 72, and 120 h immersed in an FeCl3 solution (0.01 mol L−1). b The self-healing efficiency of the ISAP-0 and ISAP-1 (error bars stand for the standard deviations from three independent samples).

Full size image

Self-healing anticorrosion property of the polymer coating

The self-crosslinking of polymer chains with Fe3+ renders polymer application prospects for self-healing and anti-corrosion. We tested ISAP-0 and ISAP-2 polymer coatings on carbon steel. All the samples exhibited hydrophobicity with a water contact angle of ~105° and ~90° after immersion in 3.5 wt% NaCl solution for 2 months owing to the PDMS hydrophobic segments (Supplementary Figs. 6 and 7). This also indicated that the introduction of hydrophilic catechol did not change the surface properties of the coating.

Salt immersion experiments were conducted to illustrate the self-healing anticorrosion properties of the ISAP coatings. Figure 6 shows the topographies of the scratched ISAP-0 and ISAP-2 coatings and the corresponding metal substrates after 7 days of immersion in salt water. For ISAP-0, the coating material exhibited severe scratch expansion and a large amount of metal rust was attached (Fig. 6a–c), indicating that severe corrosion occurred at scratches on the metal substrate. For ISAP-2, there were no noticeable changes in the polymer scratches, and the metal substrate was only slightly rusted (Fig. 6d–f). These results indicate that the ISAP-2 coating has a good self-healing anticorrosion effect.

Fig. 6: The morphology of the scratched coatings after immersion in salt water.
figure 6

Photos (a, d) and Optical microscopy (b, c, e, f) of the scratched area after peeling coating from the metal substrates (a, b, d, e) and the peeled intrinsic self-healing anticorrosion polymer coating (c, f) after immersion in salt water (0.1 M) for 7 days. (a–c) ISAP-0, (d–f) ISAP-2. The length of the scratch: 0.5 cm. Scale bar: 500 μm.

Full size image

The corrosion protection properties of the two coatings were further investigated using electrochemical impedance spectroscopy (EIS), of which results are shown in Fig. 7. The electrochemical impedance of the two coatings was tested by immersing them in a brine environment with NaCl (0.1 mol L−1 NaCl for 4, 24, 48, 72, and 96 h), and measurements were illustrated in Bode plots (Fig. 7a, b). For the scratched ISAP-0 coating, shown in Fig. 7a, the impedance values at low frequencies decreased rapidly after soaking for 4 to 48 h. In the Bode phase plots, with prolonged immersion time, two-time constants were clearly observed, which were related to the coating properties and the charge transfer process at the interface of metal and coating. These results suggest that the barrier properties of the scratched ISAP-0 coatings were severely deteriorated. For the ISAP-2 coating shown in Fig. 7b, the impedance values increased at both low and high frequencies after 24 h and 48 h of immersion. The impedance at low frequencies (|Z | 0.01 Hz) is often used to semi-quantitatively describe the corrosion resistance of coatings. As shown in Fig. 7c, the |Z | 0.01 Hz value of ISAP-0 coating drops continuously from 3.58 × 104 to 0.93 × 104 ohm cm2 in the extended immersion periods. For ISAP-2 coatings, the |Z | 0.01 Hz value was 2.35 × 104 ohm cm2 at 4 h, 3.84 × 104 ohm cm2 at 24 h and 7.29 × 104 ohm cm2 at 48 h. After 96 h of immersion, the impedance value remained at a high value of 6.26 × 104 ohm cm2 which is comparable to those of the intact ISAP-0 sample, even higher than that at 4 h (Fig. 7c and Supplementary Fig. 8). This indicates that the corrosion resistance of scratched ISAP-2 rapidly increases after soaking in salt-water. Therefore, the ISAP-2 coating exhibited self-healing anticorrosion performance.

Fig. 7: The self-healing anticorrosion performance of the scratched intrinsic self-healing anticorrosion polymer (ISAP) coatings.
figure 7

EIS Bode plots of the scratched intrinsic self-healing anticorrosion polymer (ISAP) coatings ISAP-0 (a) and ISAP-2 (b) at different immersion times. Solid symbol represents |Z|. Hollow symbol represents -phase. c The impedance at low frequencies (|Z|0.01 Hz) of ISAP-0 and ISAP-2 with different immersion times.

Full size image

Based on the above results, a self-healing anticorrosion mechanism for the ISAP-2 coating was proposed, as shown in Fig. 8. When the coating was damaged, corrosive media, such as water, O2, and Cl- penetrated the coating and a corrosion electrical reaction occurred on the metal substrate, leading to the production of Fe3+ and a weakly alkaline microenvironment as a result of the electrochemical corrosion process48. The polymer chains were self-crosslinked by the coordination of catechol and diffused Fe3+ ions in a weakly alkaline environment. The damaged area was further repaired, and the anticorrosion performance was recovered.

Fig. 8
figure 8

Schematic of self-healing anticorrosion mechanism of the intrinsic self-healing anticorrosion polymer (ISAP) coating.

Full size image

In summary, in-situ self-crosslinking strategy for constructing intrinsic self-healing anticorrosion polymers was demonstrated herein. Dynamic coordination crosslinks were formed between the catechol groups in the polymer chains and Fe3+ produced from the metal substrate. The as-synthesized polymers had tunable catechol content and good mechanical properties. In the presence of Fe3+, the healing efficiency of the specimens obtained from the tensile stress was as high as 67.9%. Moreover, the prepared scratched polymer coating showed an increase of impedance values at 0.01 Hz from 2.35 × 104 ohm cm2 at 4 h to 7.29 × 104 ohm cm2 at 48 h according to the EIS experiments. This intrinsic self-healing anticorrosion polymer has potential applications in harsh environments.

Methods

Materials

α,ω-Aminopropyl-terminated poly(dimethylsiloxane) (APT-PDMS) (Mn = 2000 g mol−1) was purchased from Hangzhou Silong Mater. Tech. Co., Ltd. Isophorone diisocyanate (IPDI, 99%, Aladdin) was used as received. Superdry tetrahydrofuran (THF, 99.9%, J&K Scientific), superdry triethylamine (TEA, 99.5%, Meryer, China), and superdry N,N-dimethyl formamide (DMF, 99.8%, J&K Scientific) were used without further purification.

Synthesis of intrinsic self-healing anticorrosion polymers

Intrinsic self-healing anticorrosion polymers with different contents of chain extender were synthesized. The chain extender was synthesized according to a previous literature43. First, 0.978 g (4.4 mmol) of IPDI was placed in a three-necked bottle under an Ar atmosphere. The as-synthesized chain extender (0.142 g, 0.4 mmol) was dissolved in the superdry DMF, followed by the addition of 0.081 g (0.8 mmol) of superdry TEA. The solution was then centrifuged. The supernatant was added dropwise to the IPDI and allowed to react for 4 h under Ar in an ice-water bath. APT-PDMS (8 g, 4 mmol) was heated at 120 °C for 2 h under vacuum to remove the moisture. The dried APT-PDMS was dissolved in 40 mL of superdry THF and added dropwise to the reaction solution. The mixture was allowed to react for 12 h under Ar at room temperature. The product was labeled ISAP-1.

Preparation of polymer anti-corrosion coatings

Carbon steel Q235 was used as the substrate for preparing the anticorrosion coating. Solutions containing 0.15 g ISAP-1 in 1.5 ml of THF were deposited onto clean carbon steel substrates. The coated samples were dried at room temperature for 24 h. The coating thickness was about 150 μm.

Characterization

The 1H-NMR (600 MHz) spectra were recorded on a Bruker AVANCE III spectrometer using CDCl3 and DMSO-d6 as the solvents and tetramethylsilane (TMS) as the internal standard. Attenuated total reflection-Fourier-transform infrared (ATR-FTIR) spectra were obtained using a Thermo Scientific Nicolet 6700 spectrometer. The scanning range was 400–4000 cm−1. The molecular weight and PDI were determined using a gel permeation chromatography (GPC) instrument (Waters 1515) with THF as the mobile phase. Ultraviolet−visible (UV−vis) absorption spectra were acquired using a Thermo Scientific Evolution 220 spectrophotometer at room temperature. Tensile tests were performed on a ZwickRoell instrument at a speed of 80 mm min−1 using a dumbbell-shaped sample. Three parallel samples of each polymer were tested. The water contact angle was measured using an OCA50 Dataphysics. Five different areas of each sample were tested and the average was calculated. Macroscopic healing tests of the coatings were conducted using an optical microscope (Axioskop40 A Pol, Nikon and Olympus BX53M). A scratch (5 mm × 0.02 mm) was made on the coating using a cutter blade. The damaged coating was repaired by soaking in a 0.1 mol L−1 NaCl solution for 4, 24, 48, 72, and 96 h. Then, the polymer coating was peeled off from the metal substrate. EIS was conducted in a 0.1 mol L−1 NaCl solution using an electrochemical workstation (CHI 660E, Shanghai Chenhua Instrument Co., Ltd. China) with a frequency range from 105 to 10-2 Hz. A three-electrode system was used, consisting of a coated steel substrate as the working electrode, a saturated calomel electrode as the reference electrode, and a platinum foil as the counter electrode.