1. Introduction
The marine environment is complex and harsh, with highly corrosive seawater and alternating dry/wet conditions; thus, it can cause severe corrosion of marine equipment. In addition, marine biological settlement can seriously damage marine equipment. Therefore, marine corrosion and biofouling have emerged as the most serious issues affecting the safety, lifetime, and reliability of ships and other marine infrastructures, potentially resulting in equipment destruction and severe economic losses [
1], [
2], [
3]. As such, the protection of marine industrial equipment has gained worldwide attention [
4], [
5].
At present, the use of anti-corrosion and anti-fouling coatings is the most effective method to address the abovementioned issues [
6], [
7], [
8], [
9], [
10]. However, most marine protective coatings are composed of an anti-corrosion primer and an anti-fouling topcoat [
5], [
11], [
12], [
13]. It is challenging to combine both anti-corrosion and anti-fouling properties in a single coating, as doing so can cause issues such as the need for multiple coats of paint, difficult construction, and high labor costs. Anti-corrosion coatings, in particular, must prevent the entry of corrosive media, so the resins of such coatings must strongly adhere to the substrate [
14], [
15]. In comparison, anti-fouling coatings are responsible for inhibiting the adhesion of marine organisms and usually comprise low-surface-energy resins [
16], [
17]. Therefore, the requirements for the resins used in anti-corrosion and anti-fouling coatings are different or even contradictory, making it extremely challenging to design a marine protective coating with both excellent anti-corrosion properties and superior anti-fouling properties [
5], [
18], [
19], [
20].
Graphene nanosheets are often used as functional fillers to obtain high-efficiency anti-corrosion coatings by forming a physical barrier and preventing the diffusion and penetration of the corrosive medium; however, the graphene can easily agglomerate [
21], [
22], [
23], [
24], [
25], [
26]. Nanosilver-loaded graphene is helpful in preventing agglomeration by reducing the van der Waals interactions between adjacent layers [
27], [
28]. Furthermore, silver nanoparticles (AgNPs) are considered to hold promise as an antifoulant due to their efficient antibacterial activity and relatively lower environmental risks, compared with other high additive (> 20 wt%), toxic, and environmentally hazardous antifoulants such as tributyltin and cuprous oxide. Hence, AgNPs on graphene can improve the anti-fouling performance of a coating [
29], [
30], [
31], [
32].
Amphiphilic anti-fouling coatings are generally fabricated by adding hydrophilic polymers to low-surface-energy resins [
33], [
34], [
35], [
36], [
37]. This is because hydrophilic polymers can form a hydration layer on the coating surface to resist the attachment of marine organisms, while hydrophobic resins can reduce the adhesion strength of the organisms [
38], [
39], [
40]. However, there are two major problems in the preparation of an amphiphilic coating: ① It is difficult to add a hydrophilic polymer to a hydrophobic resin; and ② hydrophilic polymers are prone to agglomerate and create obvious phase separation in the coating, which can reduce the protective anti-fouling and anti-corrosion performance [
32], [
41].
In this work, a marine anti-corrosion and anti-fouling coating was developed (
Fig. 1). We introduced reduced graphene oxide (rGO)/AgNPs and a hydrophilic polymer into a bio-based silicone-epoxy resin to prepare a coating with an interpenetrating network (IPN) structure. The silicone-epoxy resin, rGO/AgNPs, and hydrophilic polymer endow the coating with excellent anti-fouling performance through a ternary synergistic mechanism involving fouling release, contact inhibition, and a hydration effect. Moreover, an outstanding anti-corrosion performance is provided by a ternary synergistic anti-corrosion mechanism that includes a dense structure, barrier effect, and passivation [
42]. Furthermore, a dynamic second network achieves defect repair when the coating is damaged, thereby preventing its rapid failure. This work provides a marine protective coating for engineering equipment that combines anti-corrosion and anti-fouling capabilities and possesses the advantages of excellent functionality, easy preparation, simple construction, and large-scale application.
2. Experimental section
2.1. Materials
Graphene oxide (GO) was obtained from Suzhou Tanfeng Technology Co., Ltd. (China). Glycidyl methacrylate (GMA), thiolacetic acid (TA), triethylamine (TEA),
N-isopropylacrylamide (NIPAM), benzoin dimethyl ether (DMPA), silver acetate (AgAc), methyltriacetoxysilane (MTAS), 1,1,3,3-tetramethyldisiloxane (HMM), dibutyltin dilaurate (DBTDL) catalyst (purity > 99%), tetrabutylammonium bromide (TBAB), and isophoronediamine (IPDA) were supplied by Aladdin (China). Sodium hydroxide, acetone, petroleum ether, 1,4-dioxane, dichloromethane, and chloroform were supplied by Sinopharm Chemical Reagent Co., Ltd. (China). The silicone-epoxy resin (hydrophobic tetramethyldisiloxane eugenol-epoxy resin, HMME-EP) and the hydrophilic polymer (poly(
N-isopropylacrylamide)-thiol, PNIPAM-SH) were synthesized according to previous works [
30], [
32], [
43]; the synthetic routes and the
1H nuclear magnetic resonance (NMR) spectra (400 MHz, Avance III Bruker NMR spectrometer, Bruker, Switzerland) of their chemical structures are provided in Figs. S1-S4 in Appendix A.
2.2. Preparation of rGO/AgNPs
The rGO/AgNPs were synthesized from GO and AgAc (Fig. S5 in Appendix A). In this process, 5 g of GO and 5 g of AgAc were mixed using an agate mortar at room temperature for 30 min. Then, the mixture was calcined at 300 °C for 3 h under a nitrogen atmosphere, and rGO/AgNPs were obtained after cooling.
2.3. Preparation of integrated anti-corrosion and anti-fouling coatings
The bio-based silicone-epoxy resin HMME-EP and the curing agent IPDA, with a mass ratio of 1:0.09, were placed in a 300 mL beaker and pre-polymerized with stirring at a stirring rate of 200 r∙min
−1 for 30 min. The rGO/AgNPs and the hydrophilic polymer PNIPAM-SH were dispersed in anhydrous ethanol, added into an epoxy system prepared as described above, and further stirred for 30 min. After being degassed in a vacuum oven, the mixture was coated on a steel substrate by means of a standardized bar coater and cured at room temperature in a ventilated place to obtain an integrated anti-corrosion and anti-fouling coating. The thickness of the coating was (75 ± 5) μm (measured by an FY2050 high-precision coating-thickness gauge from Beijing Time Yuanfeng Technology Co., Ltd. (China)). As shown in
Table 1, four HE-
X-
Y coatings with different contents of PNIPAM-SH and rGO/AgNPs (where HE refers to HMME-EP,
X is the content of rGO/AgNPs, and
Y is the content of PNIPAM-SH) were prepared.
2.4. Characterizations
An AVANCE III NMR spectrometer was used to examine the 1H NMR spectra of the silicone-epoxy resin and the hydrophilic polymer. A D8 Advance X-ray diffractometer (Bruker, Germany), a Nicolet 6700 Fourier-transform infrared (FTIR) spectrometer (Thermo Fisher Scientific, USA), a Renishaw inVia Reflex Raman spectrometer (Renishaw, UK), a Lambda 950 ultraviolet (UV)-visible (vis)-near infrared (NIR) spectrometer (PerkinElmer, USA), an S4800 scanning electron microscope (SEM; Hitachi, Japan), a Tecnai F20 transmission electron microscope (TEM; FEI, USA), and a Dimension Icon scanning probe microscope (SPM; Bruker, USA) were utilized to characterize the composition and structure of the rGO/AgNPs. The morphologies and roughnesses of the coatings were observed by means of the S4800 SEM and a UP-Lambda confocal laser scanning microscope (Rtec, USA). The coating structure was characterized via the Tecnai F20 TEM. The coating samples were packaged using epoxy resin and then cut into ultra-thin slices to load onto copper grids for the TEM examination. A DSC 214 differential scanning calorimeter (DSC; NETZSCH, Germany) and a Z1.0 electric universal testing machine (Zwick, Germany) were used to investigate the glass transition temperatures and the tensile properties of the coatings, respectively. The adhesion was evaluated according to American Society for Testing and Materials (ASTM) D3359 and ASTM D4541-09, and the dynamic contact angles of the coatings were measured on an OCA20 contact angle meter (DataPhysics Instruments, Germany).
2.5. Anti-fouling properties
2.5.1. Anti-protein measurement
A bovine serum albumen (BSA)-fluorescein isothiocyanate (FITC) solution of 2 mg∙mL−1 was prepared in phosphate-buffered saline (PBS). The coating samples were immersed in the BSA-FITC solution, put in a light-resistant place for 4 h, and then washed with deionized water. A fluorescent inverted microscope was used to observe the treated coatings, and ImageJ software was used to compute the protein attachment area based on the obtained images.
2.5.2. Anti-bacterial measurement
The shaking flask method was used to investigate the anti-bacterial performance of the coatings toward Escherichia coli (E. coli), Vibrio parahaemolyticus (V. parahaemolyticus), and Staphylococcus aureus (S. aureus). The coatings were immersed in 105 colony-forming units (CFU)∙mL−1 of bacteria solution and shaken at 37 °C for 3, 6, 12, or 24 h. A 0.1 mL sample of the solution was diluted by ten-fold, coated on nutrient agar plates in triplicate, and incubated at 37 °C for 24 h. The bacterial colonies on each plate were counted. The bacterial killing rate (R) was computed via the equation R = (Nc − Ns)/Nc, where Ns is the number of microorganism colonies on the plate, and Nc is the number of microorganism colonies for the control test (without coating).
E. coli was chosen to investigate the bacterial attachment resistances of the coatings. The coatings were put in 24-well plates and seeded with 100 μL of 106 CFU∙mL−1 bacterial solution. After culturing at 37 °C for 4 h, the bacteria that were not adhered to the surface of the samples were washed away with PBS, and the samples were stained with a calcein acetoxymethyl ester (Calcein-AM)/propidium iodide (PI) kit and observed using a fluorescent inverted microscope. The samples were then fixed with glutaraldehyde and dewatered. After Pt treatment, the samples were observed by SEM, and the bacteria attachment was calculated via the software ImageJ, based on the obtained images.
2.5.3. Anti-algae measurement
Phaeodactylum tricornutum (
P. tricornutum) was chosen to study the algae attachments on the coatings. The coatings were put in 24-well plates and seeded with 1 mL of algae suspension (10
8 cells per milliliter). The growth conditions used for
P. tricornutum were to keep it under a sunlight lamp for 12 h, followed by darkness for 12 h, in an artificial climate chamber at 22 °C [
44], [
45]. After 24 h, artificial seawater was used to wash away the unadhered algae, and the obtained samples were observed by means of a fluorescent inverted microscope. ImageJ software was used to compute the algae attachment area based on the obtained images.
2.5.4. Marine field measurement
The integrated anti-corrosion and anti-fouling coatings (around 100 μm thick) were coated on stainless-steel panels (70 mm × 100 mm) after treatment by sand blasting. The samples were then immersed in 0.2-2.0 m-depth seawater from June 15, 2021, to July 15, 2021, at Meishan Island in Ningbo, China (29°49′N, 122°0′E). Scheduled recording of the adhesion situation of marine organisms on the coatings was performed.
2.6. Anti-corrosion measurement
2.6.1. Electrochemical examination
The electrochemical behavior of the integrated anti-corrosion and anti-fouling coatings was studied in a CHI-660E electrochemical workstation (Shanghai CH Instrument Co., Ltd., China) using the traditional three-electrode system (a saturated calomel electrode as the reference electrode, a platinum plate with an area of 2.5 cm2 as the counter electrode, and coated mild steel with an exposed area of 1 cm2 as the working electrode). NaCl solution (3.5 wt%) was used as the corrosive medium. An open-circuit potential test was performed for 30 min to achieve an equilibrium state, and electrochemical impedance spectroscopy (EIS) curves were measured under a sinusoidal perturbation with an amplitude of 10 mV and a frequency range from 100 kHz to 10 mHz. ZSimpWin software was used to fit the electrochemical impedance. In addition, SEM, X-ray diffraction (XRD; Bruker, Germany), and Raman spectroscopy were utilized to examine the micromorphology and components of rust layers on the steel.
2.6.2. Self-healing anti-corrosion measurement
A VersaSCAN micro-scanning electrochemical workstation (AMETEK, USA) was used to examine the self-healing and anti-corrosion performance of the coatings using the local electrochemical impedance spectroscopy (LEIS) mode. The electrode, with a diameter of 10 μm, was set to vibrate at a speed of 200 μm∙s−1 with an amplitude of 10 mV. Impedance values were collected at the frequency of 10 Hz on an area of 2.5 mm × 2.5 mm with 26 × 26 scanning points. The surface of the electrode was scratched with a scalpel to make a crack with a width of about 1 mm in order to reveal the metal substrate.
2.6.3. Salt spray test
The coating samples were exposed in a spray chamber that continuously sprayed 5 wt% NaCl solution at (35 ± 2) °C, according to ASTM B117.
3. Results and discussion
3.1. Characterization of rGO/AgNPs
Figs. 2(a)-(e) present the XRD, FTIR, Raman, and UV-vis absorption spectra of the GO before and after modification with AgNPs. As shown in
Fig. 2(a), there is a (001) diffraction peak at 2
θ = 9.8° and a (020) diffraction peak at 2
θ = 42.5°. After loading the nanosilver, these two peaks disappeared, while two new weak peaks appeared at 28.1° and 31.1° for the (032) and (320) crystal planes of the graphene, respectively. In addition, the rGO/AgNPs exhibited new peaks at 38.1°, 44.3°, 64.4°, 77.4°, and 81.5° for the (111), (200), (220), (311), and (222) crystal planes of silver, respectively, indicating successful preparation of the rGO/AgNPs [
46]. As shown in the FTIR spectra (
Fig. 2(b)), for GO, there are apparent characteristic peaks at 3400, 1720, and 1051 cm
−1 for the -OH, C=O, and C-O-C stretching vibration, respectively. For the rGO/AgNPs, the peak at 3440 cm
−1 for -OH weakens, and the peaks for C=O almost disappear. These results demonstrate that GO was transformed to rGO after calcination at 300 °C [
25]. Both the rGO/AgNPs and GO have a D peak and a G peak in the Raman spectra (
Fig. 2(c)). The intensity ratio of the D and G peaks (
ID/
IG) is usually used to reflect the ratio of sp
2 to sp
3 carbons, which is one of the strongest pieces of evidence confirming the structure variation of graphene. After loading the nanosilver, the D peak shifted from 1335 to 1352 cm
−1, and the G peak shifted from 1583 to 1585 cm
−1. In addition, the strength of both peaks increased, indicating that the silver improved the Raman response of the graphene. The
ID/
IG ratio decreased slightly from 1.032 for GO to 0.955 for rGO/AgNPs, suggesting that the nanosilver repaired the defects in the graphene. The peaks at 644 and 971 cm
−1 are the characteristic peaks of nanosilver particles. The UV-vis spectra of the GO and rGO/AgNPs in
Fig. 2(d) show an absorption band at 343 nm for the carboxyl group of GO and an absorption band at 412 nm for the nanosilver particles of the rGO/AgNPs, indicating successful nanosilver loading on the surface of the GO [
28].
Figs. 2(e)-(h) show the X-ray photoelectron spectroscopy (XPS) spectra of the GO and rGO/AgNPs. Compared with the GO, the rGO/AgNPs exhibit a peak for the silver (Ag) 3d (
Fig. 2(e)); moreover, the integration of the peak areas at 284.4, 285.6, 286.4, and 287.8 eV for C-C/C=C, C-OH, C-O-C, and C=O all obviously decrease (
Fig. 2(f)), and the integration of the peak areas for O-C (532.3 eV), O=C (531.2 eV), and O-C=O (530.2 eV) all weaken (
Fig. 2(g)), indicating that the GO was partially reduced. Ag
+ was successfully reduced to Ag and loaded onto the graphene, as evidenced by the peaks for the Ag 3d
5/2 (373.8 eV) and Ag 3d
3/2 (367.8 eV) in the Ag 3d XPS spectra (
Fig. 2(h)) [
47].
Furthermore, the TEM, SEM, and energy dispersive spectroscopy (EDS) images in
Figs. 2(i)-(l) all demonstrate the even loading of the nanosilver particles on the graphene. The crystal lattice distance of the (111) crystal plane for the particles in the rGO/AgNPs is 0.237 nm, as shown in the high-resolution TEM (HRTEM) image (
Fig. 2(j)), demonstrating that the particles are indeed nanosilver particles [
48]. The average particle size of the rGO/AgNPs (
Fig. 2(m)), as analyzed from the TEM and SEM images, is around (8.6 ± 2.1) nm. The thickness of the rGO/AgNP layer (3.1 nm) is lower than that of the GO (3.3 nm). This might be ascribed to the reduction of the GO, which decreased the oxygen-containing groups, and the loaded AgNPs, which can increase the layer distance of the aggregated graphene (
Fig. 2(n)). The particle size of the AgNPs (10.9 nm) is in agreement with the results from the SEM and TEM.
3.2. Characterization of the coatings
3.2.1. Dispersed stability of the rGO/AgNPs
Graphene easily aggregates due to interactions between the graphene layers, which could decrease the homogeneity of the coatings, form a water vapor channel, and accelerate the corrosion of the substrate. Thus, it is essential to improve the dispersion of graphene in the solvent or the uncured coatings. As shown in
Fig. 3(a), compared with the GO, the rGO/AgNPs exhibited better dispersed stability in ethanol and in ethanol solution containing the hydrophilic polymer PNIPAM-SH. After being immersed in ethanol or in ethanol solution containing PNIPAM-SH for 8 h, no aggregation or precipitation occurred in the rGO/AgNPs system. Better dispersity of the rGO/AgNPs was achieved in the ethanol solution containing PNIPAM-SH (no precipitation occurred after 72 h) because PNIPAM-SH can prevent the aggregation and precipitation of rGO/AgNPs through chelation between the -SH and AgNPs in the rGO/AgNPs [
27].
3.2.2. Morphology and structure
Phase separation often occurs when incorporating a hydrophilic polymer into a hydrophobic silicone-epoxy resin, leading to a decrease in both the homogeneity and the performance of the coating (i.e., the anti-corrosion and anti-fouling properties). Thus, in order to improve the homogeneity, we planned to use the chelation of the rGO/AgNPs and the hydrophilic polymer PNIPAM-SH to form a second network in the silicone-epoxy resin network and achieve an IPN. As shown in the SEM images of the fracture section of the coatings (
Fig. 3(b)), the silicone-epoxy resin coating modified by the hydrophilic polymer PNIPAM-SH showed a large amount of undissolved substance, compared with the neat silicone-epoxy resin coating (HE-0-0) and the rGO/AgNP-modified silicone-epoxy resin coating (HE-0.3-0). The EDS results confirmed that this undissolved substance was PNIPAM-SH. The sample with both PNIPAM-SH and rGO/AgNPs (HE-0.3-3) exhibited no obvious undissolved substance on the fracture section, suggesting an improvement in the homogeneity through the formation of an IPN via the chelation of the rGO/AgNPs and PNIPAM-SH. These results are in agreement with the results from the TEM images (
Fig. 3(c)). The roughness results from laser confocal microscopy (
Fig. 3(d)) also indicate an improvement in the compatibility of the hydrophilic polymer PNIPAM-SH and the silicone-epoxy resin after the introduction of rGO/AgNPs. The surface roughness (
Ra) decreased from 0.354 μm for HE-0-3 to 0.105 μm for HE-0.3-3.
The formation of an IPN in the coating composed of silicone-epoxy resin, PNIPAM-SH, and rGO/AgNPs was further verified by means of DSC. The introduction of just the rGO/AgNPs did not obviously change the glass transition temperature (
Tg) of the coating, while introducing PNIPAM-SH increased the
Tg from 0.5 °C for HE-0-0 to 5.8 °C for HE-0-3 (
Fig. 3(e)).
Tg was further increased to 9.1 °C for HE-0.3-3. This suggests that the second network between the PNIPAM-SH and rGO/AgNPs blocked the chain movement of the silicone-epoxy network. The suggested formation mechanism of the second network is that the AgNPs, supported by the graphene nanosheets, have a large area and act as “anchors” for binding with the sulfhydryl groups of the PNIPAM-SH chains. Thus, the PNIPAM-SH chains are well-dispersed in the silicone-epoxy network to form an IPN structure [
49], [
50].
3.2.3. Mechanical properties and adhesion
Fig. 3(f) presents the mechanical properties of the coatings. It is clear that HE-0-0 exhibited the worst tensile properties, with a tensile strength of 0.42 MPa and an elongation at break of 26.7%. HE-0-3 showed a lower tensile strength of 0.40 MPa and a higher elongation at break of 44.7%. This indicates that introducing the flexible hydrophilic polymer decreased the strength and increased the toughness of the coating. As evidenced by the slightly higher tensile strength and elongation at break of HE-0.3-0, in comparison with those of HE-0-0, only the rGO/AgNPs increased the strength and toughness of the coating. In comparison with HE-0-0, HE-0.3-3 exhibited a much higher tensile strength of 0.74 MPa and elongation at break of 45.7%. This might be attributed to the formation of the IPN increasing the cross-linked density of the coating, corresponding to a higher tensile strength, and the dynamic network of PNIPAM-SH and rGO/AgNPs dissipating the energy, leading to increased toughness.
The adhesion of the coatings on the steel substrate was investigated. All the coatings exhibited the highest level of adhesion (5B, which means the best), based on the crosshatch adhesion method (
Fig. 3(g)). The adhesion-by-pulling force test (
Fig. 3(h)) revealed that HE-0-0, HE-0-3, HE-0.3-0, and HE-0.3-3 also presented excellent adhesions of 12.03, 10.74, 12.11, and 15.94 MPa, respectively.
3.3. Anti-fouling properties of the coatings
3.3.1. Anti-protein adhesion property
Protein adhesion is a prerequisite for the adhesion of marine organisms; thus, evaluating the protein adhesion on the coatings is of great significance in the development of anti-fouling coatings [
51]. BSA-FITC was utilized as the representative protein to investigate the adhesion, and protein absorption was observed using a fluorescence inverted microscope (
Figs. 4(a) and
(b)). After being immersed in BSA-FITC solution for 4 h, 96.1% of the area of the surface of HE-0-0 had protein attached to it; for HE-0-3, only 3.2% of the area had protein attached to it, which was ascribed to a hydration layer caused by the hydrophilic polymer PNIPAM-SH preventing the attachment of protein onto the surface of the coating. For HE-0.3-0 and HE-0.3-3, almost no protein was adsorbed onto the surface, which might be ascribed to denaturation of the protein, triggered by the AgNPs.
3.3.2. Anti-bacterial property
In seawater, bacteria attach to substrates as the primary biofouling organisms; the attached bacteria then provide nutriment and favorable conditions for the further settlement of macro-organisms. Thus, it is paramount to study the anti-bacterial performance of anti-fouling coatings.
Figs. 4(c) and
(d) exhibit the adhesion of
E. coli on the coatings. After the anti-bacterial test for 24 h, numerous live
E. coli were attached on the surface of HE-0-0, with a bacteria attachment area of 38.2%; in comparison, HE-0-3 and HE-0.3-0 exhibited reduced bacteria attachment areas of 2.1% and 18.7%, respectively, and almost no bacteria were attached on the surface of HE-0.3-3. These results might be due to the hydration layer produced by the hydrophilic polymer PNIPAM-SH and the bacteria-killing ability of the rGO/AgNPs. As displayed in the SEM images of HE-0.3-0 (
Fig. 4(c), bottom), the cell membranes of the
E. coli were broken and aggregated, showing that the AgNPs killed the bacteria. A detailed study of the bacteria-killing properties of the coatings based on
E. coli,
V. parahaemolyticus, and
S. aureus was further carried out (
Figs. 4(e)-(h); Fig. S6 in Appendix A). After 6 h, HE-0.3-0 and HE-0.3-3 clearly showed
E. coli elimination ratios of 100% and 81.1%, while the elimination ratios were 95.7% and 27.7% for
V. parahaemolyticus and 37.1% and 54.1% for
S. aureus, respectively. After 24 h, both HE-0.3-0 and HE-0.3-3 exhibited an elimination ratio of 100% toward
E. coli, V. parahaemolyticus, and
S. aureus. These results suggest that the rGO/AgNPs possess excellent bacteria-killing efficiency.
3.3.3. Anti-algae adhesion property
Similar to bacteria, algae can attach in an early stage of biofouling and then provide favorable conditions for other organisms in a marine environment. In this work,
P. tricornutum was chosen to test the algae attachment (
Figs. 4(i)-(k)). After a 24 h test, numerous
P. tricornutum were attached on HE-0-0, while almost no algae were present on the surfaces of HE-0-3, HE-0.3-0, and HE-0.3-3 (
Fig. 4(i)), whose algae coverage reductions reached 99.0%, 99.7%, and 100.0% relative to HE-0-0. This indicates that both the hydrophilic polymer PNIPAM-SH and the rGO/AgNPs can prevent algae attachment. Compared with the appearance of the algae on the surface of HE-0-0 (
Fig. 4(k), left), the structure of the algae on the surface of HE-0.3-0 (
Fig. 4(k), right) was significantly changed, indicating that the rGO/AgNPs damaged the structure of the algae, in line with the enhanced anti-fouling property of the coating.
3.3.4. Marine field test
A marine field test was conducted at Meishan Island in Ningbo, China (29°49′N, 122°0′E) over a period of 30 days. A steel panel and a steel panel with epoxy primer were utilized as the controls. As shown in
Fig. 4(l), the controls accumulated numerous marine organisms, including barnacles. The surface of HE-0-0 was covered with a certain number of barnacles. HE-0-3 and HE-0.3-0 exhibited a small number of fouling organisms, and HE-0.3-3 had almost no attached fouling organisms. These results demonstrate that both the hydrophilic polymer PNIPAM-SH and the rGO/AgNPs can reduce the attachment of marine organisms, and that their synergism grants excellent anti-fouling properties to the coating.
3.4. Anti-corrosion properties of the coatings
3.4.1. Electrochemical measurement
Figs. 5(a)-(c) show the electrochemical properties of the coatings tested in 3.5 wt% NaCl solution.
Fig. 5(a) shows the EIS spectra, including the Nyquist plots (left) and Bode plots (middle and right), of the coatings. The permissive arcs have the following order: HE-0.3-3 > HE-0.3-0 > HE-0-0 > HE-0-3. The impedance values have the same order, with HE-0.3-3, HE-0.3-0, HE-0-0, and HE-0-3 exhibiting impedance values of 1.92 × 10
8, 7.79 × 10
7, 1.03 × 10
7, and 2.53 × 10
6 Ω∙cm
2, respectively. It is clear that the addition of only the hydrophilic polymer PNIPAM-SH decreased the barrier property of the coating as a result of the polymer’s separation in the coating. While the incorporation of only the rGO/AgNPs increased the barrier effect of the coating, the inclusion of the rGO/AgNPs and PNIPAM-SH together further increased the barrier effect. This result can be ascribed to an increased denseness of the coating due to the homogeneously dispersed second network caused by the cross-linking of rGO/AgNPs and PNIPAM-SH, which enhances the anti-corrosion property of the coating. The phase angle result (
Fig. 5(a), right) also illustrated that HE-0-3 was susceptible to fail and that the metal substrate had been corroded.
A further analysis of the EIS data was carried out based on equivalent circuits via ZSimpWin software. In
Fig. 5(b),
Rs,
Rc,
Rct,
Qc, and
Qdl represent the solution resistance, pore resistance, charge-transfer resistance, coating capacitance, and double-layer constant phase, respectively. Among the four coatings, HE-0.3-3 exhibited the largest
Rc, and HE-0-3 exhibited the smallest
Rc. This finding demonstrates more clearly that HE-0.3-3 and HE-0-3 possess the best and worst anti-corrosion properties, respectively.
Fig. 5(c) presents the potentiodynamic polarization curves of the coatings. HE-0-0, HE-0.3-0, HE-0-3, and HE-0.3-3 showed corrosion potentials of -0.382, -0.649, -0.737, and -0.268 V, respectively, and corrosion current densities of 1.26 × 10
−8, 2.07 × 10
−8, 2.49 × 10
−9, and 5.70 × 10
−12 A∙cm
−2, respectively. Although the rGO/AgNPs increased the impedance of the coating, aggregation of the rGO/AgNPs formed defects, leading to a higher corrosion current density for HE-0.3-0 than for HE-0-0. The lower corrosion current densities of HE-0-3 and HE-0.3-3 might be attributed to the chelation of Fe
3+ and Fe
2+ with PNIPAM-SH. The rGO/AgNPs promoted the homogeneous dispersion of PNIPAM-SH in the coating and further decreased the corrosion current density, shifting the corrosion potential to a positive value.
3.4.2. Salt spray test
The salt-fog test was utilized to study the anti-corrosion performance of the coatings under real service conditions. The coating samples were exposed in a spray chamber, and the corrosion of the steel substrate under the coatings was monitored (
Fig. 5(d)). After undergoing the salt-fog test for 15 days, the scratched places of HE-0-0, HE-0.3-0, and HE-0-3 showed a great deal of red rustiness and corrosion products; in comparison, fewer corrosion products appeared on the surface of HE-0.3-3, suggesting that HE-0.3-3 possessed the best anti-corrosion performance. This is ascribed to the increased denseness of the coating due to the formation of a second cross-linked network from PNIPAM-SH and rGO/AgNPs.
The corrosion products were characterized using XRD, Raman spectroscopy, and SEM. As revealed by the XRD patterns in
Fig. 5(e), the main corrosion product for HE-0-0 is FeOOH. After introducing the rGO/AgNPs and hydrophilic polymer PNIPAM-SH, the main corrosion products become Fe
2O
3 and Fe
3O
4. Raman spectra (
Fig. 5(f)) confirm this result. For HE-0-0, the main peaks are ascribed to β-FeOOH (394, 488, and 675 cm
−1). For HE-0-3 and HE-0.3-0, there are five peaks: at 244, 290, and 411 cm
−1 for α-Fe
2O
3, at 480 cm
−1 for γ-Fe
2O
3, and at 581 cm
−1 for β-FeOOH. For HE-0.3-3, there are three peaks: at 411 cm
−1 for α-Fe
2O
3, at 300 cm
−1 for β-Fe
2O
3, and at 663 cm
−1 for Fe
3O
4 [
52], [
53].
The morphologies of the corrosion products also varied (
Fig. 5(g)). Irregular particles appeared on the surface of the substrate, which is indicative of severe corrosion. Although there were obvious corrosion products on the surfaces of HE-0-3 and HE-0.3-0, the particles of the corrosion products were relatively regular and their sizes were smaller compared with those on HE-0-0. This finding indicated that HE-0-3 and HE-0.3-0 possessed better anti-corrosion properties than HE-0-0. The steel substrate of HE-0.3-3 exhibited a relatively smooth and dense morphology, demonstrating that the steel substrate was well protected.
3.4.3. Self-healing property
LEIS was used to monitor the evolution of corrosion on the defected samples. As shown in
Fig. 5(h), with expansion of the time, the corrosion areas of HE-0-0, HE-0.3-0, and HE-0-3 expanded, and the local impedance modulus increased, which is attributed to the permeation of the corrosive medium into the steel substrate from the defects. In comparison, for HE-0.3-3, the corrosion area did not exhibit obvious expansion, and the impedance value of the defect edges increased. This result demonstrated that HE-0.3-3 possessed a self-healing ability. After damage, the silicone-epoxy network was broken, while the network of PNIPAM-SH and rGO/AgNPs was reformed by the chelation of the thiol group and AgNPs, healing the coating and thereby preventing a rapid loss of effectiveness triggered by the defects [
54].
Microbiologically influenced corrosion (MIC) is a serious matter in complex marine environments. Although protective coatings are inexpensive and effective means for combating MIC, high concentrations of bacteria can attack coatings directly or indirectly through metabolic activity [
55], [
56], [
57]. The results described above illustrate the coating’s anti-fouling and anti-corrosion performance; with its lower microbiological adhesion and excellent bacteria-killing ability, these properties contribute to inhibiting biofilm formation, and the coating’s effective corrosion resistance and self-healing ability provide a physical barrier. Thus, the coating we developed has great potential for MIC prevention.
3.5. Anti-fouling and anti-corrosion mechanisms
3.5.1. Anti-fouling mechanism
To determine the anti-fouling mechanism of the coatings, the coatings’ contact angle toward water was examined. As shown in
Fig. 6(a) and Fig. S7 in Appendix A, all the coatings exhibited an initial contact angle of above 110° and a surface free energy below 30 mN∙m
−1; these findings are attributed to the hydrophobic silicone-epoxy network. According to the relationship between the surface free energy and the fouling adhesion in the Baier curve, coatings with a surface free energy of about 20-30 mN∙m
−1 can reduce interactions between fouling organisms and the surface [
58]. As a result, attached organisms can be easily released with sufficient shear forces, and the coatings exhibit minimum fouling adhesion. Introducing rGO/AgNPs increased the contact angle of the coating due to the increased roughness and the microstructure of the rGO/AgNPs on the surface of the coating; the rGO/AgNPs on the surface can also kill marine fouling organisms.
The dynamic contact angle can also be used to evaluate the transfer of the hydrophilic polymer from the inside to the surface of the coating (
Fig. 6(a)). With an expansion of the time, the contact angles of HE-0-0 and HE-0.3-0 only decreased slightly, while those of the coatings HE-0-3 and HE-0.3-3, both of which contained the hydrophilic polymer PNIPAM-SH, decreased considerably to 84.5°. This result demonstrates that the hydrophilic PNIPAM-SH chain segments can transfer from the inside of the coating to the surface of the coating and form a hydration layer, thereby improving the coating’s anti-fouling performance. Thus, the anti-fouling mechanism of the amphiphilic coating containing both rGO/AgNPs and the hydrophilic polymer PNIPAM-SH can be summarized as follows (
Fig. 6(b)):
(1) Fouling release: The coating’s low surface energy contributes to its dynamic anti-fouling ability; fouling organisms are easily detached from the ship hull by the shearing force of water flow.
(2) Hydration effect: The hydrophilic group in the hydrophilic polymer can transfer onto the coating surface and form a hydration layer to prevent the attachment of fouling organisms.
(3) Contact inhibition: The rGO/AgNPs on the coating surface can kill the fouling organisms and further improve the anti-fouling performance of the coating.
3.5.2. Anti-corrosion mechanism
The barrier effect of the pure silicone-epoxy coating HE-0-0 was the worst; as a result, the corrosion medium easily permeated the coating and quickly corroded the metal substrate. The anti-corrosion mechanism of the amphiphilic coating containing both rGO/AgNPs and the hydrophilic polymer PNIPAM-SH can be summarized as follows (
Fig. 6(b)):
(1) Barrier effect: The rGO/AgNPs increase the integrity of the coating and the physical barrier, which can effectively block and extend the diffusion path of the corrosive medium, improving the effective protection cycle of the coating.
(2) Dense IPN structure: The second network caused by the chelation of the rGO/AgNPs and the hydrophilic polymer PNIPAM-SH increases the denseness of the coating.
(3) Passivation: The thiol group in the PNIPAM-SH absorbs Fe3+ and Fe2+ by chelation, which deactivates and reduces the corrosion of the substrate.
Compared with coatings reported in other works, the integrated anti-corrosion and anti-fouling coating presented in this work not only possesses excellent anti-fouling and anti-corrosion properties but also exhibits superior adhesion to the substrate and self-healing properties (
Fig. 6(c)). In addition, the coating is prepared from a bio-based epoxy monomer, which is more sustainable than petroleum-based alternatives and the coating’s comprehensive properties are better than those of traditional marine protective coatings [
48], [
59], [
60], [
61], [
62], [
63], [
64], [
65], [
66].
4. Conclusions
An integrated anti-corrosion and anti-fouling marine protective coating was successfully prepared by introducing rGO/AgNPs and a hydrophilic polymer into a bio-based silicone-epoxy resin. Excellent anti-fouling performance was achieved through a ternary synergistic mechanism involving fouling release due to a low surface energy, contact inhibition, and a hydration effect. The killing efficiency of the HE-0.3-3 coating toward E. coli, V. parahaemolyticus, and S. aureus was almost 100%, and almost no proteins, bacteria, algae, or other fouling organisms were attached to the coating after laboratory and marine field tests. Outstanding anti-corrosion performance was realized via a ternary synergistic mechanism including the barrier effect of the graphene, a dense IPN structure, and passivation. Compared with the pure silicone-epoxy coating HE-0-0, the integrated coating HE-0.3-3 exhibited higher impedance and lower corrosion current density in a 3.5 wt% NaCl solution and was only slightly corroded after 15 days in a salt-fog test. In addition, the coating possessed superior adhesion on a steel substrate and self-healing ability: The adhesion strength reached 15.94 MPa, and the impedance value of the defect edges increased after self-healing. The coating meets the requirements of anti-fouling and anti-corrosion and avoids the need to paint multiple coats, thereby decreasing the construction difficulty and human cost. Thus, this work provides a new strategy for developing a marine protective coating.
Acknowledgments
This work was financially supported by the Major Project of Ningbo Science and Technology Innovation 2025 (2021Z092) and the Defense Industrial Technology Development Program (JCKY2021513B001).
Compliance with ethics guidelines
Shu Tian, Jinli Zhang, Shuan Liu, Jingyu Li, Jibin Pu, Yugang Hao, Guobing Ying, Qunji Xue, and Guangming Lu declare that they have no conflict of interest or financial conflicts to disclose.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.eng.2022.09.019.
PDF
(5546KB)
