The effect of sol–gel surface modified silver nanoparticles on the protective properties of the epoxy coating

Abstract

In this study, the effect of surface modified silver nanoparticles on the corrosion protection of an epoxy coating on mild steel was studied. An organosilane (3-methoxy silyl propyl metacrylate) was used as a surface modifier to increase the dispersability of the inorganic nanoparticles in the organic epoxy coating matrix. Fourier transform infrared spectroscopy (FTIR) and thermogravimetric analysis (TGA) were used to characterize the surface modified nanoparticles. Differential scanning colorimetry (DSC) was employed to study the effects of modified and unmodified nano-silver on the curing heat and glass transition temperature of the epoxy coatings. Salt spray and electrochemical impedance spectroscopy (EIS) were used to evaluate the anticorrosion properties of the epoxy coatings. The dispersability of the modified and unmodified nano-silver in the epoxy coating was examined by scanning electron microscopy (SEM). The results showed improved anticorrosion properties in the presence of the surface modified silver nanoparticles. Pull-off strengths and hardness measurements of the coatings were also found to be improved in the presence of modified nanoparticles.

---

1. Introduction

Nowadays, mild steel is widely used in construction due to its mechanical properties, machine ability and low price.^1,2 However, its resistance to corrosive environments is low. Organic coatings such as paint and polymer coatings are effective methods to protect mild steel.^3–6 Organic resins such as epoxy, urethane, phenolic and vinyl ester can provide high corrosion protection for mild steel.^7–9 However, the polymer coatings usually include defects due to improper surface preparation, application and curing, which provide pathways for corrosive agents such as oxygen, water and ions such as chloride to the metallic substrate leading to coating disbonding and substrate corrosion.^10,11 It is observed that incorporation of corrosion inhibitors^12–14 and anticorrosion pigments,^6,15,16 adhesion promoters and surface modifiers^17–21 into the polymeric resin can increase the anticorrosion properties of the coatings.

Composite coatings with nano-sized fillers have found great interest to develop high performance protective coatings.²² It has been found that composite coatings filled with nano-fillers enhance corrosion protection due to the high surface area to volume ratio compared to micro-sized fillers. In this regard nano-fillers such as clay,^22–26 titanium oxide (TiO₂),^27–29 silica (SiO₂),^13,30–34 silver oxide (Ag₂O),^35–37 iron oxide (Fe₂O₃),^32,38,39 zinc oxide (ZnO),⁴⁰ cerium oxide (Cr₂O₃),^18,41 alumina (Al₂O₃)⁴² and zirconia (ZrO₂)^43,44 have been employed in polymeric coatings.

However, application of inorganic nanoparticles in the polymer coatings needs care due to the weak interfacial interaction between nano particles and polymer matrix and their tendency to form aggregates and agglomerates in the polymer coatings.^18,45 Surface modifiers are widely used to increase compatibility between nanoparticles and polymer matrix in composite coatings.^29,41,46 Mechanical properties and corrosion protection properties of the composite coatings are significantly affected by the compatibility of the filler and polymer matrix.

Organosilanes can be employed as an effective surface modifiers with a general chemical structure of (RO)₃SiX for inorganic nanoparticles.^46,47 These chemicals include hydrolysable alkoxy (RO) such as methoxy and ethoxy and an organofunctional group (X) such as methyl, vinyl and methacrylate groups. Silanization of the inorganic fillers usually occurs through sol–gel route in two steps, i.e. hydrolysis and condensation.

It has been illustrated that surface modification of iron oxide, chromium oxide and zirconium oxide nanoparticles with organosilanes improves dispersability of nanoparticles in polymeric matrix. It has been shown that improve in dispersability leads to enhancement of anticorrosion and mechanical properties of the organic coatings.^18,38,44

Silver nanoparticles have excellent antibacterial properties. Therefore, they can be used in polymer coatings to extend the life time of the coatings especially in the humid atmosphere suitable for microorganism to grow on the coatings surface. Akbarian et al.³⁷ studied the effect of silver nanoparticles on corrosion protection of polyurethane coatings. It has been shown that silver nanoparticles can increase corrosion protection to some extent, but they are vulnerable to the chloride ions diffused to the coatings.³⁷ Bulky AgCl can be formed on the particles surface which in turn results in expansion and degradation of the coating. Application of silver nanoparticle for corrosion inhibition of steel in acid solution has been reported by Atta et al.⁴⁸ It has been shown that modified silver nanoparticles with polymerizable surfactants can inhibit mild steel corrosion. This work is novel over the previous published works as it reports application of silanized silver nanoparticles in epoxy coating. Silver nanoparticles, usually stabilized with organic capping agents, have different surface chemistry compared to nano-zirconia,⁴⁴ nano-iron oxide³⁸ and nano-chromium oxide.¹⁸ In this study not only corrosion protection properties of the epoxy coating has been evaluated, but also the side effects of the use of silanized nanoparticles on the curing reaction of the coating has been studied.

2. Experimental

2.1. Materials

Epoxy resin (Epikote 828) and polyamine hardener (Epikure F205) were obtained to prepare epoxy coating. Nano-silver and organosilane (3-methoxy silyl propyl metacrylate) was purchased from Sigma-Aldrich with the properties tabulated in Table 1. The particle size of purchased Ag nanoparticles reported in Table 1 was measured by field emission scanning electron microscope (FESEM, Mira3, Tescan, Czech Republic).

Table 1 Properties of Ag nanoparticles and organosilane

Component	Molecular weight/g mol^−1	Specific area/m^2 g^−1	Purity/%	Particle size
a Diameter of most of the particles was below 100 nm; however, some agglomerates above 100 nm were also detected. According the manufacturer datasheet, the Ag nanoparticles were stabilized by PVP (polyvinylpyrrolidone).
Nano silver	107.86	5	99	<100 nm^a
Silane agent	248.35	—	98	—

---

The mild steel plates (St12) were used as substrate. Elemental composition of the used mild steel is shown in Table 2.

Table 2 Chemical composition of the mild steel substrate

Element	Fe	C	Mn	P	S	Others
Wt%	99.09	0.12	0.6	0.045	0.045	≤0.1

---

2.2. Silanization of silver nanoparticles

The procedure was exactly the same as the one we reported for silanization of nano-zirconia.⁴⁴ However, a brief is provided here. The organosilane (5 ml) was added into the aqueous solution of ethanol (95 ml ethanol and 5 ml deionized water) under stirring at 250 rpm. The pH of the solution was adjusted at 2 with drop-by-drop addition of acetic acid solution (1 M). This solution was stirred for 24 h at room temperature. Then, 8 g nano-Ag was added into the solution. The pH was adjusted at 5.5 with drop-by-drop addition of 1 M ammonia solution and the mixture was further stirred at 350 rpm for 2.5 h to perform silanization through condensation reaction in sol–gel route at 70 °C. Then, the mixture was centrifuged and residue was washed five times with deionized water and dried in vacuum oven at 40 °C for 72 h.

2.3. Preparation of coating samples

The procedure was exactly the same as the one we reported for preparation of epoxy coatings containing nano-zirconia.⁴⁴ However, a brief is provided here. Unmodified and modified Ag nanoparticles (1 wt%) were separately added to epoxy resin. Dispersion was performed by high speed mixer (3000 rpm) for 30 min. Then, 45 cm³ of the mixtures were ultrasonicated for 5 min. Ice-water bath was used to cool down the mixture during ultrasonication.

The mild steel samples were abraded by emery papers (up to no. 1000) followed by acetone cleaning. Then, the polyamine hardener was added to the epoxy components containing unmodified and modified Ag nanoparticles at stoichiometric ratio. In addition, as a reference for comparison, a blank epoxy coating containing no Ag nanoparticle was prepared. As ultrasonication may partially destroy the functional groups of epoxy resin, in order to compare the coatings in the same conditions, the blank epoxy resin was also ultrasonicated through above mentioned procedure. Then, the coating materials were applied on the mild steel substrates by drawdown method using a film applicator (Elcometer 3520 baker, UK) with the wet film thickness of 60 μm. Coated steel plates remained at ambient condition (25 ± 5 °C and 30 ± 5 RH) for 12 h. Then, the coated samples were kept at 80 °C for 2 h for complete curing. An area of 75 ± 3 cm² of the coatings with a dry film thickness around 25 ± 3 μm was considered for salt spray exposure. A hot melt beeswax-colophony mixture was used to seal edge and back of the coated mild steel plates. Coatings were X-scribed before salt spray exposure. After salt spray exposure, an area of 1 cm² near the X-scribes was chosen for EIS measurements. EIS measurements and salt spray exposure were performed on triplicates to ensure repeatability.

2.4. Techniques and analyses

2.4.1. FTIR spectroscopy. Fourier transform infrared spectroscopy (FTIR) in transmission mode was used to investigate grafting of silane on the surface of Ag nanoparticles. Also, the effect of the modified and unmodified nanoparticles on the chemical structure of the epoxy coatings was evaluated using FTIR in reflectance mode. This spectroscopy was measured employing Tensor 27 (Bruker, Germany) within wavenumber range of 400–4000 cm⁻¹.

2.4.2. TGA. To evaluate the extent of silane grafting onto the nanoparticles, thermogravimetric analysis (TGa) was used under nitrogen atmosphere. TG analyzer L-801I (LINSEIS, Germany) was employed within the temperature range of 25 to 600 °C and heating rate of 10 °C min⁻¹.

2.4.3. Contact angle measurement. Contact angle test was used to evaluate grafting of silane on the surface of Ag nanoparticles and the effect of surface modification on surface energy of the nanoparticles. For this purpose, sessile drop method was used. Water and formamide were used as the test liquids. Nanoparticles were compressed to build a disc of 1.4 mm diameter by a hydraulic press. After 15 s placement of droplets on the surface of discs, the shapes of droplets were recorded by an optical camera (Dino Lite China). The images were transmitted to a personal computer and the contact angles were calculated by using an image analysis (Dino Lite Software).

2.4.4. DSC measurement. Differential scanning calorimetry (DSC) was used to evaluate the effect of unmodified and modified Ag nanoparticles on curing behavior of the epoxy coating by using Netzsch-DSC 200 F3 at 100 °C under purging of pure nitrogen. DSC technique was also used to determine glass transition temperature of the cured epoxy coatings with a 10 °C min⁻¹ thermal ramp from 10 to 120 °C.

2.4.5. Salt spray exposure. Anticorrosion performance of the X-scribed epoxy coatings containing modified and unmodified Ag nanoparticles was studied by salt spray exposure according to ASTM B 117 for 240 h.

2.4.6. EIS measurement. Electrochemical impedance spectroscopy of the coated specimens after 240 h exposure to salt spray was measured on an area of 1 cm² near the X-scribes where the coatings disbonded from the substrate. The rest areas where sealed with a hot melt mixture of beeswax-colophony. A potentiostat (Ivium Compactstat, Netherland) was employed to measure electrochemical impedance spectroscopy (EIS). The coated mild steel specimens immersed in 3.5 wt% NaCl solution for 1 day before EIS measurements. The EIS measurements were conducted at open circuit potential, with 10 mV peak to peak perturbation within the frequency range of 10 mHz to 10 kHz. All the measurements were conducted in a three-electrode cell including Ag/AgCl (3 M KCl, with potential of 0.21 vs. SHE: standard hydrogen electrode) as a reference electrode, platinum as a counter electrode and coated specimen as a working electrode. In order to analysis the EIS data Iviumsoft was used.

2.4.7. Pull-off adhesion strength. The adhesion strength of the blank and filled epoxy coatings to the mild steel substrate after salt spray exposure were estimated by using pull-off adhesion tester (Positest, USA). The test was conducted according to ASTM D 4541. Dollies of the 25 mm diameter were adhered to the coated mild steel using a cyanoacrylate adhesive, Loctite 401 (Hankel). After 3 days drying at ambient temperature, the dollies were pulled-off vertically at a speed of 10 mm min⁻¹ until the epoxy coatings were detached from the mild steel substrates. All tests were carried out on three samples and the average values were reported.

2.4.8. Micro-Vickers hardness. The effect of the Ag nanoparticles on the coatings hardness at room temperature was evaluated using micro-Vickers hardness (MDPEL M400) under load of 10 g. Micro-Vickers hardness measurements were performed in 5 different places on the coatings and the mean values were reported.

2.4.9. SEM. A scanning electron microscope (SEM Cam Scan MV2200, Vega Tescan, Czech Republic) under voltage of 5 kV, equipped with tungsten heated cathode gun and SE detector was employed to study the dispersion of unmodified and modified nano-Ag in epoxy coatings. For this purpose, the coating materials were applied on the PTFE sheet to prepare the free films. The cured free films were stretched to break using universal tensile machine Roell-Z010 (Zwick, Germany) at room condition (24 °C and 30% RH) with a loading rate of 1 mm min⁻¹. The cross-section of the coating at rupture area was examined by SEM at 10 000× magnification.

3. Result and discussion

3.1. Characterization of surface modified Ag nanoparticles

3.1.1. FTIR spectroscopy. Silane grafting onto the surface of nano-Ag was evaluated by using FTIR spectroscopy. The unmodified and silanized Ag nanoparticles were placed in a vacuum oven at 60 °C for 48 h before FTIR spectroscopy to ensure of dehydration. The FTIR spectra of the both silanized and unmodified nano-Ag have been illustrated in Fig. 1. The main FTIR absorption bands are listed in Table 3.

Fig. 1 FTIR absorbance spectrum of (a) unmodified and (b) silanized Ag nanoparticles.

Table 3 Characteristic absorption peaks observed in the FTIR spectrum of unmodified and silanized Ag nanoparticles

No.	Functionality	Wavenumber/cm^−1 (unmodified Ag)	Wavenumber/cm^−1 (silanized Ag)
1	O–H stretching	3386	3431
2	Ag–O–Si stretching	—	1110
3	Si–O–Si asymmetric stretching	—	1180
4	Si–O–Si symmetric stretching	—	800
5	Si–O–Si bending	—	450
6	C–H stretching	2814, 2931	2854, 2923
7	C=O stretching	1650	1650
8	Deformation mode of pyrrolidone ring	1450	1450

---

According to the manufacturer datasheet, Ag particles were stabilized with PVP (polyvinylpyrrolidone). That is why some peaks related to organics can be seen in the FTIR spectrum of unmodified nano-Ag. The stretching vibration of O–H bond observed at 3341 cm⁻¹ has been depressed after silanization of nanoparticles which indicates consumption of surface hydroxides due to condensation reaction with silanol groups reflecting grafting of organosilane onto the Ag nanoparticle surface. The observed peaks at 2854 and 2923 cm⁻¹ in spectrum were assigned to asymmetric stretching of C–H in methacrylate segment of the organosilane⁴⁹ and also vinyl and pyrrolidone segments of PVP. Intensification of these absorption bands after silanization, again, reflects grafting of organosilane onto the Ag nanoparticle surface. Also, appearance of the absorption bands after silanization at about 1110 and 1180 cm⁻¹ respectively assigned to Ag–O–Si³⁵ and Si–O–Si⁵⁰ asymmetric stretching and also at around 800 and 450 respectively connected to Si–O–Si symmetric stretching⁵⁰ and bending vibration⁵⁰ of Si–O–Si indicates successful silane agent grafting onto the surface of the nano-Ag. The peak around 1650 cm⁻¹ in Fig. 1a is attributed to C=O (carbonyl on PVP) stretching.⁵¹ Intensification of this peak after silanization is an indication of increased carbonyl bond concentration on the Ag surface due to presence of carbonyl on metacrylate segment of the organosilane. None of the above mentioned evidences does not help to rule out desorption of the PVP due to silanization as the absorption bands for the PVP and organic segment of organosilane overlap. Depression of bands around 1450 cm⁻¹ related to deformation mode of pyrrolidone ring⁵¹ indicates partial desorption of the PVP from the surface where silanization has been take place.

Considering all evidences obtained from the FTIR results, the silanization procedure for Ag nanoparticles is schematically given in Fig. 2.

Fig. 2 The scheme of silanization of a silver nanoparticle.

3.1.2. TGA. TGA thermograms of unmodified and silanized Ag nanoparticles are shown in Fig. 3.

Fig. 3 TGA thermograms of unmodified and silanized Ag nanoparticles.

According to Fig. 3, the weight loss was occurred in two steps. In the first stage which occurred in temperature range of 20–180 °C, the physically adsorbed water on the surface of nanoparticles released that was about 0.7 and 0.4 wt% for unmodified and silanized Ag nanoparticles, respectively. This result shows that silanized Ag nanoparticles have less tendency to adsorb water molecules because of grafting of organosilane on the particles surface which is in good agreement with the lower OH absorption bands in FTIR results.^18,52 The second stage of weight loss was occurred in the temperature range of 220–500 °C. Weight loss values in this stage for unmodified powder is about 1.5 wt% which attribute to releasing water molecules due to condensation reaction of hydroxyl groups on the surface of particles (Ag–OH). The weight loss for silanized nanoparticles was 2.4 wt% which was due to condensation reaction of hydroxyl groups and thermal degradation organic segment of grafted silane on the surface of the Ag nanoparticles.

3.1.3. Contact angle measurements. The surface chemistry of the unmodified and silanized Ag nanoparticles was evaluated by using contact angle measurement. Water and formamide were used to calculate the both polar and disperse component of the surface free energy of the discs. Water and formamide droplets on the surface of the unmodified and silanized Ag discs are illustrated in Fig. 4.

Fig. 4 Contact angle of (a₁) water droplet on the surface of unmodified nano-Ag disc, (a₂) water droplet on the surface of silanized nano-Ag disc, (b₁) formamide droplet on the surface of unmodified nano-Ag disc and (b₂) formamide droplet on the surface of silanized nano-Ag disc.

The work of adhesion (W_a) and surface free energy (γ_S) values were determined for unmodified and modified Ag discs through Young's (eqn (1)) and Owens–Wendt (eqn (2) and (3)) equations, respectively.^53,54

W_a = γ_lv(1 + cos θ) (1)

γ_S = γ^D_S + γ^P_S (3)

where, γ_lv is surface tension of water, γ^D_L and γ^P_L are the dispersive and polar parts of the surface tension of the liquids and γ^D_S and γ^P_S are the dispersive and polar parts of the surface free energy of the discs, θ is the contact angle and γ_S is total surface free energy of the discs.⁵⁴ The values of the dispersive and polar parts of the water and formamide surface tension are presented in Table 4.

Table 4 The values of the dispersive and polar parts of the surface tension of water and formamide

Liquid	γL/mN m^−1	γ^DL/mN m^−1	γ^PL/mN m^−1
Water	72.8	21.8	51
Formamide	58	19	34

---

The work of adhesion and surface free energy were calculated by eqn (1)–(3). The values of W_a and γ_S for the unmodified and silanized Ag discs are tabulated in Table 5.

Table 5 The values of work of adhesion and surface free energy of the unmodified and silanized nano-Ag discs

Sample	Wa/mJ	γ^DS/mN m^−1	γ^PS/mN m^−1	γS/mN m^−1
Unmodified Ag disc	135.81	6.16	62.17	68.34
Silanized Ag disc	118.4	12.41	35.79	48.20

---

It can be observed from Fig. 4 that the contact angle of the unmodified Ag disc is around 30°. The result shows that silanization of Ag nanoparticles led to an increase in contact angles (see Fig. 4) and decrement in W_a and γ_S values (see Table 5). This evidence confirms successful silanization of the Ag nanoparticles providing a hydrophobic surface on the particles surface. The silane condensation reaction on the surface of the nanoparticles results in the formation of –Ag–O–Si– covalent bonds and –Si–O–Si– cross-links.⁵³ The formed –Ag–O–Si– covalent bonds and –Si–O–Si– cross-links and organic segment of organosilane provides more hydrophobic surface for the silanized Ag-nanoparticles compared with unmodified nanoparticles with hydroxide functionalities on the particle surface covered with PVP.

3.2. Characterization of the nanocomposite coatings

3.2.1. SEM. The morphology of fracture surface of the epoxy coatings containing both unmodified and silanized Ag nanoparticles was evaluated by SEM. Fig. 5 shows the SEM image of fracture surface of the epoxy coatings.

Fig. 5 Fracture surface of the epoxy coatings containing (a) unmodified and (b) silanized Ag nanoparticles.

Fig. 5a and b show the dispersion of unmodified and silanized Ag nanoparticles in the epoxy coatings, respectively. It can be seen that surface modification of the nanoparticles improved the dispersion of nanoparticles in epoxy matrix due to increase in compatibility between nanoparticles and epoxy matrix. Increasing compatibility between nanoparticles and matrix leads to uniform dispersion of the particles in coating matrix. Dispersion of unmodified nanoparticles is not perfect due to agglomeration of the nanoparticles in coating matrix. The average size of the agglomerates is around 1 μm (see Fig. 5a). The increase in hydrophobicity of silanized nanoparticles compared with unmodified one (see Section 3.1.) can be attributed to better wettability and dispersability of the silanized nanoparticles with organic coating material.

3.2.2. FTIR spectroscopy. FTIR spectroscopy also employed to investigate the effect of the unmodified and silanized Ag nanoparticles on the chemical structure of the epoxy coatings. Blank and filled epoxy coatings were coated on the glass substrates. After curing, FTIR analysis in the reflectance mode was performed on the surface of the coatings. The FTIR spectra of the epoxy coatings are shown in Fig. 6.

Fig. 6 FTIR reflectance spectra of epoxy coatings containing (a) no nanoparticle, (b) unmodified Ag nanoparticles and (c) silanized Ag nanoparticles.

FTIR spectra were normalized according to the asymmetric stretching of C–H in CH₂ at 2890 cm⁻¹. The wide peak observed around 3400 cm⁻¹ was assigned to stretching vibration of O–H bonds on the surface of the epoxy coatings and adsorbed water molecules on the coating surface.⁴⁴ The lower intensity of O–H stretching vibration (around 3400 cm⁻¹) and OH bending vibration (around 1600 cm⁻¹) in epoxy coating containing silanized Ag nanoparticles indicates higher hydrophobicity of coating compared to blank coating and the coating containing unmodified Ag nanoparticles.

Additionally, reflectance bands at around 1500 cm⁻¹ are assigned to bending vibration of N–H bonds in polyamine hardener.⁴⁰ The curing reaction of the epoxy coatings proceeds with ring opening reaction of the epoxide group through deprotonation of amine groups. Therefore, the higher intensity of bending vibration of N–H bond in the filled epoxy coatings compared to blank epoxy coating shows that in the presence of nanoparticles the cross-linking reactions are inhibited to some extent.⁴⁰ Increase in viscosity of the epoxy coating by inclusion of nanoparticles can inhibit effective collision of reactive groups.

3.2.3. DSC analysis. The results of DSC analysis of the epoxy coatings are shown in Fig. 7. The curing enthalpy (area under exothermic curve) of the coatings is tabulated in Table 6.

Fig. 7 Isothermal DSC thermograms obtained at 100 °C for the epoxy coatings containing (a) no nanoparticle, (b) unmodified Ag nanoparticles and (c) silanized Ag nanoparticles.

Table 6 Curing enthalpy of the epoxy coatings (obtained from isothermal DSC thermograms shown in Fig. 7)

Sample	Blank epoxy coating	Epoxy coating containing unmodified nano-Ag	Epoxy coating containing modified nano-Ag
ΔHT (J g^−1)	250.8	244.4	226.7

---

The results clearly show that the total curing enthalpy of epoxy coating decreases by adding Ag nanoparticles. Decrease in curing enthalpy can be attributed to lower reaction of epoxy resin and polyamine hardener due to the steric hindrance of the Ag nanoparticles in epoxy coating matrix.⁴⁰ Inclusion of Ag nanoparticles in epoxy coating results in increase of viscosity of the coating which in turn restricts effective collision of reactive groups. This is a negative side effect of utilizing Ag nanoparticles in the epoxy coatings. The lower curing enthalpy of the epoxy coating containing silanized nanoparticles compared with the epoxy coating containing unmodified nanoparticles can be attributed to the formation of the Ag agglomerates for the unmodified one. Agglomeration of nanoparticles decreases resin and particles interface which results in diminished steric hindrance and therefore less negative effect on curing.

The non-isothermal DSC was also performed on the cured epoxy coatings to determine the glass transition temperature (T_g). The non-isothermal plots are shown in Fig. 8.

Fig. 8 Dynamic DSC thermograms obtained at 10 °C min⁻¹ for epoxy coatings containing (a) no nanoparticle, (b) unmodified Ag nanoparticles and (c) silanized Ag nanoparticles.

The transition from glassy to rubbery state occurred in temperature range of 78–84 °C for all coatings. The value of glass transition temperature (T_g) of the coatings is presented in Table 7.

Table 7 The glass transition temperature of the coatings (obtained from dynamic DSC thermograms shown in Fig. 8)

Sample	Blank epoxy coating	Epoxy coating containing unmodified Ag nanoparticles	Epoxy coating containing silanized Ag nanoparticles
Tg (°C)	83.1 ± 0.8	79.7 ± 0.7	81.2 ± 0.7

---

According to the curing enthalpy results, Ag nanoparticles in epoxy resin have negative effect on curing. Therefore, one may conclude from curing enthalpies that free volume of coating matrix and segmental motion of resin chain will be increased and the glass transition temperature will be decreased in the presence of the Ag nanoparticles.^10,40 However, the Ag nanoparticles can occupy the free volumes and limit the segmental motion of the polymer chains. This is why we see increase of T_g for the coatings in the presence of silanized nanoparticles compared to the unmodified nanoparticles. The agglomerated particles are too large to occupy the free volumes and restrict nano-metric segmental motions. In other words, occupation of Ag nanoparticles in the free volumes compensates the negative increase of free volume due to incomplete curing.

3.2.4. Micro-Vickers hardness. The micro-Vickers hardness test was performed to determine the hardness values of the epoxy coatings. The micro-Vickers hardness results are shown in Fig. 9.

Fig. 9 The micro-Vickers hardness values of epoxy coatings containing (a) no nanoparticle, (b) unmodified Ag nanoparticles and (c) silanized Ag nanoparticles.

In micro-Vickers test, only the mechanical properties of coating surface can be obtained. The results show that the epoxy coating containing silanized Ag nanoparticles has highest micro-Vickers value. According to the DSC results, the epoxy resin containing silanized nanoparticles has the lower curing enthalpy and cross-linking density compared to the blank epoxy coating. The higher micro-Vickers value of the epoxy coating containing Ag nanoparticles compared to the blank epoxy coating can be attributed to higher elastic modulus of Ag nanoparticles in comparison with the epoxy matrix. The higher micro-Vickers value of the epoxy resin containing silanized Ag nanoparticles compared to the one containing unmodified nanoparticles can be connected to the better dispersion and occupation of silanized nanoparticles in the epoxy coating free volume spaces.

3.3. Corrosion protection performance of the coatings

3.3.1. Salt spray exposure. The effect of the unmodified and silanized Ag nanoparticles on the anticorrosion properties of the epoxy coatings was studied by salt spray test. The appearance of the epoxy coating containing 1 wt% unmodified and silanized nano-Ag after 240 h exposure to 3.5% NaCl salt spray are given in Fig. 10.

Fig. 10 The appearance of epoxy coatings containing (a) no nanoparticle, (b) unmodified Ag nanoparticles and (c) silanized Ag nanoparticles, after 240 h of exposure to (5% NaCl) salt spray.

It can be clearly observed that the corrosion protection of epoxy coatings increases due to addition of nano-Ag to the epoxy coatings. The surface modification of nanoparticles improves the anticorrosion properties of the epoxy coating compared with unmodified nanoparticles.⁴⁴ The percentage of the coatings' delaminated area after 240 h salt spray exposure is presented in Table 8.

Table 8 The percentage of the delaminated area of the coatings after 240 h salt spray exposure

Coating	Blank epoxy coating	Epoxy coating containing unmodified Ag nanoparticles	Epoxy coating containing silanized Ag nanoparticles
Delamination area%	81	43	37

---

It is obvious from Table 8 and Fig. 10 that the delamination area of the coatings from mild steel surface has been decreased significantly in the presence of Ag nanoparticles. The epoxy coating containing silanized nano-Ag particles has the lowest delamination from mild steel. Although X-scribes on the coatings provide good conductive pathways for the corrosive electrolyte to reach the coating metal interface, the permeability of the coatings to water and oxygen is crucial to proceed the cathodic delamination in disbonding front. Depletion of water and oxygen in disbonding front can prevent further delamination at cathodic area and oxidation at anodic area. Ag nanoparticles, and more specifically silanized nanoparticles, can occupy the free volumes inside the coatings (see Section 3.2.3.). This results in increase in the coatings barrier effect and depletion of water and oxygen concentration at coating–metal interface.

3.3.2. Pull-off adhesion test. After 240 h salt spray exposure, the wet adhesion strength of the coatings was examined by pull-off method. The extent of adhesion strength for the epoxy coatings is shown in Fig. 11. It should be noted that type of failure for all the coatings was adhesive.

Fig. 11 Pull-off adhesion strength carried out on the mild steel substrates coated with epoxy coatings containing (a) no nanoparticle, (b) unmodified Ag nanoparticles and (c) silanized Ag nanoparticles.

It can be seen in Fig. 11 that the pull-off strength of the epoxy coating increases in the presence of silanized Ag nanoparticles. According to the previous results, silanized nanoparticles have good wettability with the epoxy matrix.⁵³ Therefore, nanoparticles can be dispersed uniformly, without forming agglomerates, in the coating material and occupy the free volumes between the polymer chains. This resulted in lower diffusion rate of water and oxygen through the epoxy coating which are responsible for loss of adhesion due to hydrolysis of the epoxy coating and anodic and cathodic reactions on the mild steel surface. As a result, the epoxy coating containing silanized nanoparticles has the higher pull-off strength compared to the rest of coatings.

3.3.3. EIS measurements. After 240 h salt spray exposure, EIS of the coated specimens was measured on an area of 1 cm² near the X-scribes where the coatings disbonded from the substrate. The typical Nyquist and Bode plots obtained from EIS measurements for coated mild steel specimens immersed in 3.5 wt% NaCl solution for 1 day have been displayed in Fig. 12 and 13, respectively.

Fig. 12 The Nyquist plots for () blank epoxy coating; () epoxy coating containing unmodified Ag nanoparticles and () epoxy coating containing silanized Ag nanoparticles. These spectra were measured on an area of 1 cm² near the X-scribes where the coatings disbonded from the substrate. The measured data points are displayed by symbols and the fitted curves by the equivalent electrical circuit shown in Fig. 14 are displayed by solid lines.

Fig. 13 The phase angle (a) and magnitude (b) Bode plots for () blank epoxy coating; () epoxy coating containing unmodified Ag nanoparticles and () epoxy coating containing silanized Ag nanoparticles. These spectra were measured on an area of 1 cm² near the X-scribes where the coatings disbonded from the substrate. The measured data points are displayed by symbols and the fitted curves by the equivalent electrical circuit shown in Fig. 14 are displayed by solid lines.

Nyquist and Bode plots shown in Fig. 12 and 13 reveal two-time constant electrochemical process for the coated samples. The high frequency capacitive loop is connected to the coating and the one at low frequency is attributed to the corrosion on the mild steel surface. The diameter of the semi-circles in Fig. 12 at high and low frequencies in the Nyquist diagrams are assigned to the coating resistance and charge transfer resistance, respectively. It is obvious that in the presence of silanized Ag nanoparticles the extent of charge transfer resistance and coating resistance increased. However, in the presence of unmodified Ag nanoparticles, there is no significant increase for the resistance elements which could be connected to the agglomeration of the nanoparticles.

Phase angle values at high frequencies could be an indication of the coating capacitive or resistive behavior. High coating resistance facilitates current to pass through the capacitance element instead of resistance element resulting in capacitive behavior which can be identified in the Bode diagrams at high frequencies by the shift of phase angle to −90°.⁵⁵ According to Fig. 13a, the most capacitive behavior is obtained in the case of silanized Ag nanoparticles again confirming the best performance of this coating among the examined coated samples. In the presence of unmodified Ag nanoparticles, the phase angle value at high frequency has less negative shift compared to the blank coating which can be attributed to the agglomeration of Ag nanoparticles.

The magnitude of impedance at low frequencies (Fig. 13b) can give good estimation of the corrosion resistance of specimens. The highest magnitude is obtained in the presence of silanized Ag nanoparticles, which again confirms improved protection of the nanocomposite coating.

To extract the electrochemical parameters involved in the corrosion of the coated specimens, several electrical circuits were used to fit the EIS data. The electrical circuit providing the best fit was considered as equivalent electrical circuit. Fig. 14 shows the equivalent electrical circuit used to extract the electrochemical parameters. In this figure, R_s, R_f, R_ct, CPE_dl and CPE_f represents solution resistance, coating film resistance, charge transfer resistance, constant phase element of double layer and constant phase element of coating film, respectively.

Fig. 14 The equivalent electrical circuit used to fit the EIS measurements.

The fitting curves are shown in Fig. 12 and 13 as solid lines. The electrochemical parameters extracted from the fitting of the EIS measurements are listed in Table 9. In this table, instead of the Y₀ (the CPE admittance) and n parameters (the CPE exponent), capacitance of the corresponding CPE parameters are reported. Considering the normal time-constant distribution, the effective coating capacitance (C_f) and double layer capacitance (C_dl) was calculated from the corresponding CPE and resistance parameters according to eqn (4)⁵⁶

Table 9 The electrochemical parameters extracted from the fitting of the EIS measurements on an area of 1 cm² near the X-scribes where the coatings disbonded from the substrate

Coating type	Rf (kΩ cm^2)	Cf (nF cm^−2)	Rct (kΩ cm^2)	Cdl (μF cm^−2)
Blank epoxy	1.5 ± 0.2	15.2 ± 1.7	21.0 ± 3.1	516.4 ± 44.7
Epoxy containing unmodified Ag	5.0 ± 0.4	4.7 ± 0.8	29.8 ± 3.7	245.1 ± 33.1
Epoxy containing silanized Ag	70.7 ± 4.1	3.3 ± 0.8	151.1 ± 25.2	49.1 ± 5.6

---

According to the results listed in Table 9, in the presence of silanized Ag nanoparticles the extent of charge transfer resistance and coating resistance show a considerable increase. However, no significant increase for the resistance elements can be observed for unmodified nanoparticles due to the agglomeration of the nanoparticles in the coatings which diminishes their performance. The increase of both charge transfer resistance and coating resistance in the presence of the silanized Ag nanoparticles indicates barrier effect of these nanoparticles.

In the presence of silanized Ag nanoparticles the extent of double layer capacitance and coating film capacitance showed a considerable decrease. Again, no significant decrease for the resistance elements can be observed for unmodified nanoparticles which can be attributed to the agglomeration of the nanoparticles in the coatings. As water has higher dielectric constant than the polymeric coatings, penetration of water in the coatings is usually associated with the increase of coating capacitance.⁵⁷ Therefore, decrease of coating film capacitance in the presence of silanized Ag nanoparticles confirms the barrier effect of these nanoparticles.

It was shown that silver nanoparticles are vulnerable to the chloride ions diffused to the coatings.³⁷ It was reported that AgCl can be formed on the particles surface which in turn results in expansion and degradation of the coating. The increase in corrosion protection properties of epoxy coatings in the presence of silanized Ag nanoparticles not only originates from improved dispersability but also can be the result of higher resistance of the silanized nanoparticles against chloride attack.

4. Conclusion

This work presented the effect of 1 wt% unmodified and silanized Ag nanoparticles with 3-methoxy silyl propyl methacrylate silane on anticorrosion performance of the epoxy coating on the mild steel substrate. The obtained results can be summarized as follow:

• FTIR and TGA results illustrated that silanization of Ag nanoparticles was successfully carried out.

• Contact angle measurement revealed increase in hydrophobicity of the Ag nanoparticles by silanization.

• SEM results showed improved dispersion of Ag nanoparticles in the epoxy matrix due to increase in hydrophobicity and therefore, better wettability by silanization.

• DSC results from the reacting coating material showed that the Ag nanoparticles could decrease the total curing enthalpy of the epoxy resin, which is negative side effect of these nanoparticles. Silanization strengthened this negative side effect.

• DSC results from the cured coating showed higher glass transition temperature for the coating containing silanized Ag nanoparticles than the one containing unmodified nanoparticles. This indicates that occupation of silanized Ag nanoparticles in the free volumes compensates their negative side effect, i.e. incomplete curing.

• The results of the salt spray exposure showed that the epoxy coating containing silanized Ag nanoparticle provided better corrosion protection performance in comparison with the blank epoxy coating and the one containing unmodified nanoparticles.

• The epoxy coating containing silanized Ag nanoparticles illustrated superior adhesion strength after salt spray exposure compared to the other coating samples reflecting diminished diffusion of water and oxygen through this coating sample.

• EIS measurements on the coatings exposed to the salt spray showed barrier protection mechanism of the Ag nanoparticles and confirmed better corrosion protection performance of silanized Ag nanoparticle in comparison with the blank epoxy coating and the one containing unmodified nanoparticles.

References

1. I. Ahamad, R. Prasad and M. A. Quraishi, Corros. Sci., 2010, 52, 3033–3041.
2. M. Gao, X. Zeng, J. Yan and Q. Hu, Appl. Surf. Sci., 2008, 254, 5715–5721.
3. P. Pedeferri, Construct. Build. Mater., 1996, 10, 391–402.
4. J. Morgan, Cathodic protection, 1987.
5. L. S. Pinchuk, V. A. Goldade, I. M. Vertyachikh, S. Y. Liberman, G. V. Rechits, B. A. Finkelshtein, V. I. Shmurak, J. I. Voronezhtsev, V. V. Snezhkov and G. A. Goncharov, US Pat., US5028479 A, 1991.
6. A. Kalendová, D. Veselý, I. Sapurina and J. Stejskal, Prog. Org. Coat., 2008, 63, 228–237.
7. A. Forsgren, Corrosion control through organic coatings, CRC Press, 2006.
8. M. Alam, E. Sharmin, S. M. Ashraf and S. Ahmad, Prog. Org. Coat., 2004, 50, 224–230.
9. J. B. Bajat, I. Milošev, Ž. Jovanović and V. B. Mišković-Stanković, Appl. Surf. Sci., 2010, 256, 3508–3517.
10. M. Mahdavian, R. Naderi, M. Peighambari, M. Hamdipour and S. A. Haddadi, J. Ind. Eng. Chem., 2014, 21, 1167–1173.
11. M. Pantoja, B. Diaz-Benito, F. Velasco, J. Abenojar and J. C. del Real, Appl. Surf. Sci., 2009, 255, 6386–6390.
12. V. Moutarlier, B. Neveu and M. P. Gigandet, Surf. Coat. Technol., 2008, 202, 2052–2058.
13. A. C. Balaskas, I. A. Kartsonakis, D. Snihirova, M. F. Montemor and G. Kordas, Prog. Org. Coat., 2011, 72, 653–662.
14. H. Wang and R. Akid, Corros. Sci., 2008, 50, 1142–1148.
15. F. Askari, E. Ghasemi, B. Ramezanzadeh and M. Mahdavian, Dyes Pigm., 2014, 109, 189–199.
16. E. Alibakhshi, E. Ghasemi and M. Mahdavian, Prog. Org. Coat., 2013, 76, 224–230.
17. J. P. Matinlinna, M. Ozcan, L. V. J. Lassila and P. K. Vallittu, Dent. Mater., 2004, 20, 804–813.
18. M. J. Palimi, M. Rostami, M. Mahdavian and B. Ramezanzadeh, Prog. Org. Coat., 2014, 77, 1663–1673.
19. R. Zandi Zand, K. Verbeken and A. Adriaens, Prog. Org. Coat., 2011, 72, 709–715.
20. W.-G. Ji, J.-M. Hu, L. Liu, J.-Q. Zhang and C.-N. Cao, Surf. Coat. Technol., 2007, 201, 4789–4795.
21. D. D. Rodrigues and J. G. Broughton, Int. J. Adhes. Adhes., 2013, 46, 62–73.
22. Y. Yu, J. Yeh, S. Liou, C. Chen, D. Liaw and H. Lu, J. Appl. Polym. Sci., 2004, 92, 3573–3582.
23. M. Behzadnasab, S. M. Mirabedini and M. Esfandeh, Corros. Sci., 2013, 75, 134–141.
24. M. R. Bagherzadeh and F. Mahdavi, Prog. Org. Coat., 2007, 60, 117–120.
25. T. T. X. Hang, T. A. Truc, M. G. Olivier, C. Vandermiers, N. Guérit and N. Pébre, Prog. Org. Coat., 2010, 69, 410–416.
26. J. Singh-Beemat and J. O. Iroh, Prog. Org. Coat., 2012, 74, 173–180.
27. T. Tatsuma, S. Saitoh, Y. Ohko and A. Fujishima, Chem. Mater., 2001, 13, 2838–2842.
28. Y. Ohko, S. Saitoh, T. Tatsuma and A. Fujishima, J. Electrochem. Soc., 2001, 148, B24–B28.
29. J. P. Matinlinna, M. Özcan, L. V. J. Lassila and P. K. Vallittu, Dent. Mater., 2004, 20, 804–813.
30. D. A. Kissounko, J. M. Deitzel, S. P. Doherty, A. Shah and J. W. Gillespie, Eur. Polym. J., 2008, 44, 2807–2819.
31. M. R. Yuan, J. T. Lu and G. Kong, Surf. Coat. Technol., 2010, 204, 1229–1235.
32. X. Shi, T. A. Nguyen, Z. Suo, Y. Liu and R. Avci, Surf. Coat. Technol., 2009, 204, 237–245.
33. J.-M. Lin and C.-C. M. Ma, Polym. Degrad. Stab., 2000, 69, 229–235.
34. D. Borisova, H. Möhwald and D. G. Shchukin, ACS Nano, 2011, 5, 1939–1946.
35. G. Suriati, M. Mariatti and A. Azizan, Polym. Bull., 2013, 70, 311–323.
36. D. I. Tee, M. Mariatti, A. Azizan, C. H. See and K. F. Chong, Compos. Sci. Technol., 2007, 67, 2584–2591.
37. M. Akbarian, M. E. Olya, M. Mahdavian and M. Ataeefard, Prog. Org. Coat., 2014, 77, 1233–1240.
38. M. J. Palimi, M. Rostami, M. Mahdavian and B. Ramezanzadeh, Appl. Surf. Sci., 2014, 320, 60–72.
39. X. Liu, Y. Shao, Y. Zhang, G. Meng, T. Zhang and F. Wang, Corros. Sci., 2014, 90, 451–462.
40. B. Ramezanzadeh, M. M. Attar and M. Farzam, J. Therm. Anal. Calorim., 2011, 103, 731–739.
41. M. J. Palimi, M. Rostami, M. Mahdavian and B. Ramezanzadeh, Prog. Org. Coat., 2014, 77, 1935–1945.
42. S. Ahmad, S. M. Ashraf, A. Hasnat, S. Yadav and A. Jamal, J. Appl. Polym. Sci., 2001, 82, 1855–1865.
43. M. Behzadnasab, S. M. Mirabedini, K. Kabiri and S. Jamali, Corros. Sci., 2011, 53, 89–98.
44. S. A. Haddadi, M. Mahdavian and E. Karimi, RSC Adv., 2015, 5, 28769–28777.
45. M. Akbarian, M. E. Olya, M. Ataeefard and M. Mahdavian, Prog. Org. Coat., 2012, 75, 344–348.
46. K. L. Mittal, Silanes and other coupling agents, CRC Press, 2007, vol. 4.
47. A. A. Tracton, Coatings technology handbook, CRC Press, 2010.
48. A. M. Atta, G. A. El-Mahdy and H. A. Allohedan, Int. J. Electrochem. Sci., 2013, 8, 4873–4885.
49. D. L. Pavia, Introduction to spectroscopy, 2009, http://CengageBrain.com.
50. L. Song, J. Chen, Y. Bian, L. Zhu, Y. Zhou, Y. Xiang and D. Xia, J. Porous Mater., 2015, 22, 379–385.
51. K. Buchholz, S. Warn, B. Skeries, S. Wick and E. J. Yaacoub, Carbohydrates as Organic Raw Materials III, 1996, pp. 155–168.
52. M. Behzadnasab, S. M. Mirabedini, K. Kabiri and S. Jamali, Corros. Sci., 2011, 53, 89–98.
53. B. Ramezanzadeh, E. Raeisi and M. Mahdavian, Int. J. Adhes. Adhes., 2015, 63, 166–176.
54. K. L. Mittal, Contact angle, wettability and adhesion, CRC Press, 2006, vol. 4.
55. M. Mahdavian and M. M. Attar, Corros. Sci., 2006, 48, 4152–4157.
56. B. Hirschorn, M. E. Orazem, B. Tribollet, V. Vivier, I. Frateur and M. Musiani, Electrochim. Acta, 2010, 55, 6218–6227.
57. K. A. Thomas, S. Nair, R. Rajeswari, V. Natarajan, T. Mukundan and R. John, Prog. Org. Coat., 2015, 89, 267–270.

---