Corrosion Science 115 (2017) 78–92

Contents lists available at ScienceDirect

Corrosion Science
journal homepage: www.elsevier.com/locate/corsci

Exploring corrosion protection properties of solvent based

epoxy-graphene oxide nanocomposite coatings on mild steel
Sepideh Pourhashem a , Mohammad Reza Vaezi a,∗ , Alimorad Rashidi b ,
Mohammad Reza Bagherzadeh b
a
Division of Nanotechnology and Advanced Materials, Materials and Energy Research Center, Karaj, Iran
b
Nanotechnology Research Center, Research Institute of Petroleum Industry, Tehran, Iran

a r t i c l e i n f o a b s t r a c t

Article history: Solvent-based epoxy coatings filled with graphene oxide nanosheets (GO) are developed to enhance the
Received 28 May 2016 corrosion protection of mild steel substrates. Results reveal that GO distribution in polymer matrix is the
Received in revised form most important parameter influencing coating performance. Viscosity of polymer matrix and GO weight
19 September 2016
percent (wt.%) are exceptional aspects for GO dispersion quality. Among nanocomposites prepared with
Accepted 14 November 2016
different methods and GO wt.% (0.05, 0.1, 0.3, and 0.5), coatings prepared via direct addition of 0.1 wt.%
Available online 15 November 2016
GO to polymer matrix with lower viscosity show desirable GO dispersion quality, better adhesion to
substrate, outstanding barrier properties and superior corrosion protection under NaCl electrolyte.
Keywords:
Epoxy © 2016 Elsevier Ltd. All rights reserved.
Graphene oxide
Nanocomposite coating
A. Mild steel
Corrosion protection

1. Introduction between graphene sheets due to their high specific surface area
[16,17]. Since graphene has high tendency to agglomerate and the
Organic coatings are widely used for corrosion protection stacked graphene sheets behave like graphite with low surface area,
of metallic structures in modern industry [1–4]. Meanwhile, the homogenous dispersion and efficient interfacial interaction of
solvent-based epoxy coatings are amongst the most common graphene in organic coatings are the main challenges to obtain a
industrial polymers used as barrier coatings due to their out- nanocomposite with optimal properties [16–19].
standing mechanical properties, superior adhesion to substrate, One of the approaches to improve the dispersion of graphene
dimensional stability, good chemical resistance, and low cost [5–7]. sheets in nanocomposites is utilizing graphene oxide (GO) which
However, epoxy coatings suffer from brittleness and poor resis- has functional groups including hydroxyls, epoxides and carboxyls
tance to crack propagation [8,9]. Moreover, they have hydrophilic on its structure in order to alter the Van der Waals interactions,
hydroxyl groups and free volumes in their structure which make making them easily dispersed in the polymer matrix. Besides, these
them permeable to aggressive agents (i.e. water, oxygen, Cl− ); functional groups are more compatible with polymer structure
therefore, corrosion of the metallic substrate accelerates over time [16,20]. In this regard, Li et al. [20] reported that the functional
[10–12]. groups of GO play an important role in GO dispersion in polymer
In this regard, nanocomposite coatings are promising method matrix and transfer of interfacial stresses in the composite. Fur-
for improving corrosion resistance of epoxy coatings [12,13]. ther, the processing technique influences the dispersion level of
Recent studies show that graphene improves the corrosion resis- nanofillers in the epoxy matrix and the nanocomposite property
tance of organic coatings by increasing the path of corroding [19]; because strong hydrogen bonds can be formed between GO
agents in the coating to reach to the metal/coating interface and sheets dispersed in the polymer matrix during drying [21].
enhancing the barrier properties of polymer matrix [14,15]. Despite Although several studies have been previously conducted on the
that, there are strong van der Waals forces and ␲-␲ interactions mechanical application of epoxy-GO nanocomposites [6,22–24],
investigations on the corrosion resistance of epoxy coatings
containing GO nanosheets without further functionalization are
limited. Therefore, the main aim of this research is to improve
∗ Corresponding author.
the corrosion protection efficiency of solvent-based epoxy coatings
E-mail address: m r [email protected] (M.R. Vaezi).

http://dx.doi.org/10.1016/j.corsci.2016.11.008
0010-938X/© 2016 Elsevier Ltd. All rights reserved.
 S. Pourhashem et al. / Corrosion Science 115 (2017) 78–92 79

via using GO without any functionalization. The way of GO addi- out whether it is better to add directly GO suspension to epoxy
tion to epoxy coating and wt.% of GO in nanocomposite coatings resin or polyamide hardener to prepare a homogeneous and stable
have been considered to achieve maximum corrosion protection. nanocomposite with superior corrosion resistance property:
To the best of our knowledge, for the first time, this study demon-
strates that the way of GO addition whether to epoxy resin or (1) GO suspension was directly added to epoxy resin (R); fol-
polyamide hardener can alter the corrosion protection behavior of lowed by probe sonication for 10 min and then, heating at
nanocomposite coatings via its effect on the rheology and bondings 50 ◦ C to evaporate the solvent slowly. Finally, the stoichiometric
formed between GO and polymer matrix. Additionally, epoxy-GO amount of hardener (H) was added and mixed with a labora-
nanocomposite coatings containing different wt.% of GO provide tory mixer. These nanocomposite coatings are abbreviated by
different protection efficiency. The effect of GO on corrosion per- (R + GO) + H.
formance of epoxy coating is investigated by means of polarization (2) GO suspension was directly added to polyamide hardener;
resistance, electrochemical impedance spectroscopy, and salt spray followed by probe sonication for 10 min and then, heat-
test. ing at 50 ◦ C to evaporate the solvent slowly. Finally, the
polyamide hardener-GO mixture (H + GO) was poured into the
2. Experimental stoichiometric amount of epoxy resin and mixed with a labora-
tory mixer. These nanocomposite coatings are abbreviated by
2.1. Materials R + (H + GO).

Graphite powder (particle size <20 ␮m) was purchased from Moreover, pure epoxy coatings (R + H) without any GO nanofiller
Sigma-Aldrich. Potassium permanganate (KMNO4 ), sulphuric acid were prepared via mechanical mixing epoxy resin and polyamide
(H2 SO4 ), H2 O2 , and HCl used for GO synthesize were purchased hardener in stoichiometric amounts of 1:1. Fig. 1 shows the
from Merck. Epoxy resin (diglycidyl ether of bisphenol A, Epon schematic of nanocomposite coating preparation.
1001) and polyamide hardener (Versamid 115) were used to pre- Secondly, since the results of corrosion tests showed that
pare epoxy coatings. The epoxy equivalent mass of epoxy resin is nanocomposite coatings prepared via R + (H + GO) method have
480–550 and the amine value of polyamide hardener is 230–246 mg superior corrosion resistance, nanocomposite coatings containing
KOH per gram resin. The solid content of the epoxy resin and the different weight percentage of GO nanofiller (0.05, 0.1, 0.3, and
hardener are 75%, and 50%, respectively. 0.5 wt.%) were prepared by the mentioned method to evaluate the
Mild carbon steel plates (ST-12) were purchased from Foolad effect of GO wt.% on the corrosion protection performance of coat-
Mobarakeh Co. of Iran and the chemical composition of the steel ings.
substrates is presented in Table 1. Mild steel panels were sand- The prepared coatings were applied by air spray technique on
blasted and then, they were degreased with acetone before spray sandblasted mild steel plates. The coated substrates cured at room
coating on them. temperature for a week and post-cured at 90 ◦ C for 1 h. The thick-
ness of coatings was in the range of 150 ± 10 ␮m.
2.2. Synthesis of GO
2.4. Characterization
GO was synthesized according to modified Hummers method
[25–27]. In a stepwise preparation, 1 g natural graphite powder The crystal structure of the synthesized GO was characterized by
was added into 50 mL H2 SO4 . After that, 6 g KMnO4 was added X-ray diffraction technique (XRD, PW-1840, Philips, Cu-K␣, 40 kV,
very slowly to the suspension cooled with an ice-bath to keep the 30 mA). Chemical grafting of GO in polymeric matrix (i.e. mixture
reaction temperature below 15 ◦ C. Subsequently, the ice bath was of GO with either epoxy resin (R + GO) or hardener (H + GO)) was
removed and the mixture was stirred continuously in water bath characterized by Fourier transform infrared (FTIR, Perkin Elmer
at 35 ◦ C for 1 day. It is then diluted with slow addition of deionized Spectrum-GX) spectroscopy. The morphology of GO and the dis-
(DI) water and the oxidation reaction was terminated by pouring persion quality of GO in nanocomposite coatings were considered
H2 O2 solution. For purification, the mixture was washed by HCl and using field emission scanning electron microscopy (FE-SEM, MIRA3
DI water several times to remove residual acids and salts. The oxi- TESCAN). In order to evaluate the GO dispersion in epoxy coating,
dized solid was subsequently suspended in solvent and exfoliated the coated substrates were broken under liquid nitrogen to study
by high-energy sonication to yield GO. the fracture surface. Also, adhesion of nanocomposite coating to
mild steel substrate was measured by pull-off method (Elcometer,
2.3. Preparation of epoxy-GO nanocomposite coatings HATA 108) according to ASTM D4541-09 [28] and the measure-
ments were repeated 3 times to ensure the repeatability of data.
In this research, the effect of two important parameters includ-
ing (I) nanocomposite coating preparation method and (II) GO wt.% 2.5. Electrochemical and corrosion tests
on the corrosion resistance properties of solvent-based epoxy coat-
ings were considered. The electrochemical properties of coatings were considered by
Hence, first of all, epoxy-GO nanocomposite coatings contain- potentiodynamic polarization test and electrochemical impedance
ing 0.1 wt.% GO were prepared via solution processing technique. It spectroscopy (EIS) using an Autolab (potentiostat/galvanostat,
should be mentioned that for nanocomposite preparation, GO was PGSTAT 30). A three electrode cell was used for electrochemical
dispersed in organic solvent (0.1 g GO per 100 mL solvent) with measurement which consists of coated substrate, platinum rod and
ice bath ultrasonic treatment for 1 h to prepare GO suspension. saturated calomel electrode (SCE) as working electrode, counter
In this regard, two procedures were considered in order to find electrode, and reference electrode, respectively and 3.5 wt.% NaCl
solution was used as corrosive electrolyte. For potentiodynamic
Table 1 polarization tests, open circuit potential (OCP) of system at equi-
Chemical composition (wt.%) of mild carbon steel (ST-12). librium state after immersion for 2 h in 3.5 wt.% NaCl solution
Element Fe Si C S P Mn Al
was recorded as corrosion potential (Ecorr ). Then, the polarization
curve was plotted by sweeping the applied potential from −0.25 to
Wt.% Balance <0.1 0.1 <0.025 <0.025 0.5 0.07
+0.25 V with respect to OCP at scan rate of 1 mV/s on 1 cm2 anode.
80 S. Pourhashem et al. / Corrosion Science 115 (2017) 78–92

Fig. 1. Schematic of nanocomposite coating preparation method.

Corrosion current density (icorr ) was determined from Tafel plot by after oxidation and the presence of functional oxygen groups on
extrapolating the linear portion of the curve to Ecorr . Tafel constants GO basal planes [27]. Also, the morphology of GO is identified by
including anodic (␤a ) and cathodic (␤c ) slopes were calculated for FE-SEM as shown in Fig. 3, revealing the sheet-like structure of
anodic and cathodic parts of Tafel plot, respectively. The EIS tests exfoliated GO.
were carried out in frequency range of 10−2 to 105 Hz with AC
amplitude of 10 mV at OCP. The EIS measurements were performed 3.2. The effect of preparation method on corrosion resistance of
on coatings immersed in 3.5 wt.% NaCl solution at different immer- nanocomposite coatings
sion times of 1, 7, and 14 days. Then, the impedance spectra were
analyzed by fitting the experimental data to electrical equivalent 3.2.1. FTIR analysis
circuit with the help of Nova 1.8 software. The quality of fitting FTIR spectroscopy is used to understand the bonding interac-
was evaluated by reducing the chi-square (␹2 ) which is the sum tions between the epoxy resin and polyamide hardener with GO
of square of the differences between theoretical and experimental sheets. FTIR spectra of GO, epoxy resin (R), epoxy resin-GO mix-
points and also, by limiting the relative error of each element in the ture (R + GO), polyamide hardener (H), and polyamide hardener-GO
electrical equivalent circuit to 5%. mixture (H + GO) is represented in Fig. 4.
Further, corrosion studies were carried out by salt spray test on FTIR spectra of GO (Fig. 4 (a)) indicates the presence of character-
scratched samples for 300 h exposure to accelerated environment istic bands at 1625 cm−1 (C C, skeletal vibrations from unoxidized
of salt fog cabin according to ASTM B117 [29]. graphitic domains), 1390 and 1721 cm−1 (C O, carbonyl group),
1068 cm−1 (C O, alkoxy group), 1227 cm−1 (C O C, epoxy group),
3424 cm−1 (C OH, hydroxyl group), and the doublet bands at 2850
3. Results and 2917 cm−1 (symmetric and asymmetric CH2 ) [30].These func-
tional groups are produced on the internal and external surfaces
3.1. GO characterization of GO which can increase the surface polarity and further alter the
surface charges [31–34].
Fig. 2 represents the XRD pattern of GO which shows a strong As shown in Fig. 4(b), the characteristic absorption bands for
peak at 2␪ = 10.5◦ , corresponding to an interlayer spacing of pure epoxy resin appeared at 3459 cm−1 (stretching vibration
0.84 nm. The appearance of diffraction peak at 10.5◦ and the large of hydroxyl groups of free and hydrogen bonded OH groups),
interlayer spacing of GO demonstrates the exfoliation of graphite 3035, 1582, and 750 cm−1 (stretching of aromatic rings), 2967 and
 S. Pourhashem et al. / Corrosion Science 115 (2017) 78–92 81

Fig. 2. XRD pattern of the synthesized GO according to modified Hummers method.

Absorption bonds of polyamide hardener (Fig. 4(c)) are evident

at 3294, 3061, 1549, and 1496 cm−1 (N H stretching vibration),
1891, 1274, 1148, and 1054 cm−1 (C N stretching), 3022 and
1377 cm−1 (CH3 stretching), 2925 and 2854 cm−1 (CH2 symmet-
rical and asymmetrical stretching), 2345 cm−1 (CO2 group), 1939,
1461, and 1122 cm−1 (C O stretching), 1647 cm−1 (C O stretch-
ing phenyl ring amino acid), and 1081 cm−1 (C C stretching).
In comparison, FTIR spectra of H + GO sample overlaid with the
FTIR spectra of polyamide hardener. However, the intensity of
N H stretching vibration bands (3294,3061, 3022, 1647, 1549,
and 1461 cm−1 ) increases, the intensity of CH2 stretching bands
at 2925 and 2854 cm−1 decreases and new peaks at 3076 and
3009 cm−1 appear attributed to C H stretching, suggesting the
chemical hydrogen bonding between hardener and GO is formed.

3.2.2. FE-SEM results

The effect of nanocomposite preparation method on dispersion
quality of GO sheets in polymer coating is investigated by FE-SEM.
Fig. 5 shows the cross-section images of pure epoxy, (R + GO) + H,
and R + (H + GO) coatings. The low magnified SEM image repre-
sents that pure epoxy coating has smooth and uniform fracture
surface without obvious defects, suggesting brittle fracture and
low fracture toughness [16,23]. However, nanocomposite coatings
show rough fracture surface and crack deviation, indicating tough-
Fig. 3. FE-SEM image of the synthesized GO.
ening mechanism of GO nanosheets in nanocomposites. Indeed,
GO with its two dimensional structure can disturb and deflect the
1384 cm−1 ( CH3 asymmetric and symmetric stretching), 2930, crack propagation [38,39]. Moreover, apparent differences between
2873, and 1460 cm−1 (CH2 symmetric stretching, CH2 asymmetric (R + GO) + H and R + (H + GO) nanocomposite coatings are observed,
stretching), 1362 cm−1 ( CH3 bending vibration), 915 cm−1 (C O since the cracks have propagated and expanded more in R + (H + GO)
stretching of epoxide ring vibration), 2731, and 1247 cm−1 ( CH than (R + GO) + H coating.
epoxy ring and ether bonds in epoxide groups), 1608, 1582, 1510 According to high magnified SEM images in Fig. 5, GO sheets
and 830 cm−1 (benzene ring of epoxy), 1297–1184, and 1087 cm−1 have better exfoliation in nanocomposite coating prepared via
(C O C asymmetric and symmetric stretching bands) [35,36]. R + (H + GO) method, due to thin GO layers, no aggregation of GO
Although all the functional groups of epoxy resin are preserved sheets, and good adhesion to polymer matrix with no obvious gaps
in the FTIR spectra of R + GO mixture, there is not any evidence in R + (H + GO) sample. It can be observed that the GO in (R + GO) + H
of GO bands in the R + GO spectra which is most likely due to the sample is pulled outside from the fracture surface, which shows a
low amounts of GO [2]. However, new bands appeared at 2759, weak interfacial bonding [29]. While, GO in R + (H + GO) sample is
2488, 2429, and 2295 cm−1 for R + GO spectra; the first absorption embedded in the coating, and no obvious GO pull-out is seen, as a
band is due to C H stretching group and the others are attributed result of stronger covalent interfacial bonding between GO and the
to C OH stretching. Moreover, the intensity of absorption band coating [16].
around 3439 cm−1 has increased due to overlap of C OH bond of It should be mentioned that the rheology of dispersion plays an
epoxy resin and graphene oxide [37]. important role for GO distribution in polymer matrix [40]. In this
82 S. Pourhashem et al. / Corrosion Science 115 (2017) 78–92

Table 2
Polarization parameters for R + H, (R + GO) + H, and R + (H + GO) samples immersed
in 3.5 wt.% NaCl solution.

Sample Ecorr icorr (␮A/cm2 ) ␤a (V/dec) ␤c (V/dec)

(V vs. SCE)

R+H −0.543 14.6 0.076 −0.350

(R + GO) + H −0.562 0.88 0.199 −0.337
R + (H + GO) −0.455 0.22 0.188 −0.086

3.2.3. Adhesion measurement

Since one of the main parameters for corrosion protection of an
organic coating is its adhesion strength to metallic substrate [43],
the effect of GO on epoxy coating adhesion is considered. Fig. 6
shows the average results of pull-off adhesion measurements for
pure and nanocomposite epoxy coatings. Accordingly, the adhe-
sion of epoxy coating increases by incorporation of small amount
of GO (0.1 wt.%) and both nanocomposite coatings strongly adhere
to metallic substrate, leading to efficient corrosion protective coat-
ing. Pure epoxy coating adheres to mild steel substrate via chemical
adhesion aided by mechanical interlocking adhesion. The reactive
functional groups of epoxy coating such as hydroxyl groups tend to
adhere to metal substrate via formation of hydrogen bonding with
the hydroxyl groups on the mild steel substrate. As FTIR results
show, the functional groups of epoxy coating increase and the adhe-
sion promotion is achieved via adding GO to epoxy coatings [44,45].
Meanwhile, well-dispersed nanosheets of GO in R + (H + GO) sam-
ple (based on FE-SEM images) cause better adhesion of R + (H + GO)
coating than (R + GO) + H sample and better protective performance
under corrosive environment will be achieved.

3.2.4. Potentiodynamic measurement

Fig. 7 shows the potentiodynamic polarization curves of R + H,
(R + GO) + H, and R + (H + GO) coatings in 3.5 wt.% NaCl solution after
stabilization for 2 h and the electrochemical parameters obtained
from polarization curves are presented in Table 2.
The Ecorr of pure epoxy coating is about −0.543 V vs. SCE
with a corrosion current density of about 14.6 ␮A/cm2 , confirm-
ing that it is susceptible to corrosion. According to polarization
curves, nanocomposite coatings display quite different behavior
from that of neat epoxy coating and the way of nanocomposite
coating preparation has a great effect on its corrosion resistance.
In the case of (R + GO) + H nanocomposite coating, Ecorr shifts to
more negative potentials (−0.562 V vs. SCE) and icorr significantly
decreases to 0.88 ␮A/cm2 . On the other hand, R + (H + GO) coated
substrate exhibits more positive Ecorr (−0.455 V vs. SCE) with a
remarkable decrease of icorr (0.22 ␮A/cm2 ); the noble shift of Ecorr
accompanying with decrease of icorr , suggest that the R + (H + GO)
nanocomposite coating acts as a protective layer which improves
the corrosion resistance of the mild carbon steel substrate remark-
ably. Besides, both anodic and cathodic Tafel slopes have changed
by incorporation of GO in coating, revealing the corrosion protec-
Fig. 4. FTIR spectra of (a) GO; (b) pure epoxy resin (R), and epoxy resin-GO mixture tion properties of coatings has changed via GO addition and based
(R + GO); (c) polyamide hardener (H), and polyamide hardener-GO mixture (H + GO). on preparation method. Accordingly, GO distribution in polymer
matrix enhances the barrier properties of coating and corrosion
protection property.

research, the polyamide hardener used has lower viscosity than 3.2.5. EIS measurements
the epoxy resin; therefore, it is difficult for GO nanosheets to be dis- In this paper, EIS has been used to measure the dielectric prop-
persed in resin epoxy with high viscosity compared with polyamide erties and consider the degradation of nanocomposite coatings at
hardener with low viscosity [8,41]. Similar results have been pre- OCP in 3.5 wt.% NaCl solution for different immersion times. The
sented by Dong et al. [42] about clay dispersion in polymer matrix variation of OCP values of samples as a function of immersion time
and they also found that organoclay dispersion in low viscosity is shown in Fig. 8. According to Fig. 8, the OCP values of R + (H + GO)
polymer matrix is better than that with high viscosity. The appro- sample is more positive than the other samples during immersion
priate dispersion and exfoliation of GO in coatings enhances the time, indicating better corrosion protection performance due to
corrosion resistance properties which is verified in further studies. barrier effects of GO in polymer coating. However, the OCP val-
 S. Pourhashem et al. / Corrosion Science 115 (2017) 78–92 83

Fig. 5. FE-SEM topographies from the cross-section of (a, b) neat epoxy, (c, d) (R + GO) + H, (e, f) R + (H + GO) at different magnifications.

ues of samples decrease to more negative potentials by increasing Figs. 9–11 show the Bode and Nyquist plots of pure epoxy
immersion time which is related to continuous diffusion of corro- and nanocomposite coatings after 1, 7, and 14 days of immersion.
sive agents to coating/metal interface, leading to corrosion process The Bode plots exhibit high frequency and low frequency regions
at metal surface. due to coating capacitance and charge transfer process, respec-
tively [46]. It can be seen from Fig. 9 (a) that the total impedance
84 S. Pourhashem et al. / Corrosion Science 115 (2017) 78–92

Fig. 8. OCP values of R + H, (R + GO) + H, and R + (H + GO) samples after immersion in

Fig. 6. Pull-off adhesion results of coated steel substrates. 3.5 wt.% NaCl solution for various times.

Fig. 9. (a) Bode and phase diagrams and (b) Nyquist diagrams of R + H, (R + GO) + H,
and R + (H + GO) samples immersed in 3.5 wt.% NaCl solution for 1 day; marker points
Fig. 7. (a) Potentiodynamic polarization curves for R + H, (R + GO) + H, and
and solid lines show the experimental and fitted data, respectively.
R + (H + GO) coatings on mild carbon steel after immersion in 3.5 wt.% NaCl solu-
tion for 2 h; (b) Tafel extrapolation on the polarization plot of (R + GO) + H sample
for determination of corrosion current and Tafel slopes.
neat epoxy after 1 day of immersion; the |Z|0.01Hz value for pure
epoxy, (R + GO) + H, and R + (H + GO) coatings is 1.48 × 106 , 5 × 106 ,
of both (R + GO) + H and R + (H + GO) nanocomposite coatings in and 19.6 × 106 .cm2 , respectively. This shows the good barrier
high frequency region (104 –105 Hz) after 1 day of immersion are property of nanocomposite coatings at initial of immersion. Mean-
equal and slightly greater than the pure epoxy coating. Further, while, the higher |Z|0.01Hz value of R + (H + GO) sample compared to
the impedance modulus at low frequency of 0.01 Hz (|Z|0.01Hz ) (R + GO) + H, shows that nanocomposites prepared via direct addi-
for both nanocomposite coatings is significantly higher than the tion of GO to low viscosity hardener have higher barrier property
 S. Pourhashem et al. / Corrosion Science 115 (2017) 78–92 85

Fig. 11. (a) Bode and phase diagrams and (b) Nyquist diagrams of R + H, (R + GO) + H,
Fig. 10. (a) Bode and phase diagrams and (b) Nyquist diagrams of R + H, (R + GO) + H,
and R + (H + GO) samples immersed in 3.5 wt.% NaCl solution for 14 days; marker
and R + (H + GO) samples immersed in 3.5 wt.% NaCl solution for 7 days; marker
points and solid lines show the experimental and fitted data, respectively.
points and solid lines show the experimental and fitted data, respectively.

against electrolyte diffusion through coating [47]. By increasing the for nanocomposite coatings is about −90◦ (Fig. 11(a)) even after
immersion time to 7 days (Fig. 10(a)), degradation of pure epoxy immersion for 14 days, indicating strong interfacial interaction of
coating occurs and the |Z|0.01Hz value decreases to 1.46 × 104 .cm2 nanocomposite coatings with metallic substrates.
due to permeation of corrosive solution into the coating. On The above mentioned observations represent that the corrosion
the other side, although the |Z|0.01Hz value of (R + H) + GO and protection performance of epoxy coating significantly improves via
R + (H + GO) nanocomposite coatings decreases slightly by increas- addition of GO as nanofiller. GO enhances the barrier properties
ing immersion time to 7 days (which is about 2.41 × 106 and of epoxy coating against diffusion of corrosive agents to reach the
7.42 × 106 . cm2 , respectively), it is still much higher than pure metal/coating interface. However, the results clearly show that the
epoxy coating. Further, as shown in Fig. 11(a), the |Z|0.01Hz value for preparation method of nanocomposite coating remarkably influ-
R + H, (R + GO) + H, and R + (H + GO) samples after 14 days of immer- ences the degree of corrosion protection performance of coatings. In
sion is 6.03 × 104 , 5 × 106 , and 8.55 × 106 .cm2 , respectively. fact, direct addition of GO to low viscosity hardener, i.e. R + (H + GO)
Moreover, the Z module at high frequency for nanocomposite coat- sample, leads to the increase of corrosion resistance of epoxy
ings is significantly higher than neat epoxy coating after 7 and coatings compared to direct GO addition to high viscosity resin
14 days of immersion. In addition, R + (H + GO) sample shows higher ((R + GO) + H sample). This phenomenon can be attributed to bet-
impedance than (R + GO) + H coating at low frequency, indicating ter distribution of GO nanosheets in R + (H + GO) sample based on
that R + (H + GO) sample provides higher protective performance. FE-SEM observations.
These results signify the excellent barrier properties of nanocom- For further investigations, the EIS measurements are fitted via
posite coatings after 14 days of immersion in corrosive electrolyte an electrical equivalent circuit (Fig. 12). The equivalent circle con-
and the effect of nanocomposite preparation method on corrosion sists of Rs , Qcoat , Rpore , Qdl , and RCT which are solution resistance,
protection performance of coatings. coating constant phase element, coating pore resistance, double
Moreover, phase angle at high frequency is another parameter layer constant phase, and charge transfer resistance, respectively.
for evaluating the corrosion resistance of coatings. Phase angle at Constant phase element (CPE) represents the deviation from ideal
high frequency of pure epoxy and nanocomposite coatings after capacitive behavior and is used in the equivalent circuit to minimize
immersion in NaCl solution for 1 day is about −90◦ (Fig. 9(a)). the systematical error and obtain more precious fitting results. The
However, the phase angle at high frequency of pure epoxy coat- CPE is associated with an exponent (0 ≤ N ≤ 1) which indicates the
ing decreases to about −40◦ after immersion for 7 days (Fig. 10(a)), deviation from an ideal dielectric behavior; if N = 1, it represents an
which is another evident sign for complete delamination of epoxy ideal capacitor behavior and if N = 0, it behaves as an ideal resistor
coating. On the other hand, the phase angle at high frequency [36,48,49]. The electrochemical parameters are given in Table 3.
86 S. Pourhashem et al. / Corrosion Science 115 (2017) 78–92

Table 3
The electrochemical parameters extracted from EIS data for R + H, (R + GO) + H, and R + (H + GO) samples immersed in 3.5 wt.% NaCl solution for different times.

Sample time CPEcoat Rpore CPEdl RCT 2

( cm2 ) ( cm2 )

Y0 n Y0 n
(−1 cm−2 sn ) (−1 cm−2 sn )

R+H 1 day 34.3 × 10−10 0.784 360 × 103 0.58 × 10−6 0.542 807 × 103 0.027
7 days 4.98 × 10−10 0.969 4.4 × 103 80 × 10−6 0.463 11.4 × 103 0.019
14 days 4.75 × 10−10 0.957 7.2 × 103 39.1 × 10−6 0.544 14.2 × 103 0.040

(R + GO) + H 1 day 5.12 × 10−10 0.924 3.48 × 106 9.01 × 10−7 0.571 1.96 × 106 0.052
7 days 11.4 × 10−10 0.878 0.75 × 106 1.50 × 10−7 0.456 1.77 × 106 0.044
14 days 5.90 × 10−10 0.921 1.28 × 106 0.34 × 10−7 0.410 5.23 × 106 0.009

R + (H + GO) 1 day 4.45 × 10−10 0.975 5.86 × 106 0.08 × 10−7 0.612 14.0 × 106 0.050
7 days 7.87 × 10−10 0.943 3.24 × 106 0.32 × 10−7 0.485 3.94 × 106 0.009
14 days 7.46 × 10−10 0.947 3.60 × 106 0.71 × 10−7 0.352 5.52 × 106 0.028

Further, the Qdl which is related to distribution of ionic charge

at metal/coatings interface [36] is significantly lower for nanocom-
posite coatings in the whole period of immersion. Besides, the Qdl
of R + (H + GO) sample is lower than (R + GO) + H sample. The Qdl
value increases by extending immersion time due to expansion of
active sites [51] and the fluctuations observed for Qdl value during
immersion is the result of formation and dissolution of corrosion
products at coating/metal interface.
Moreover, the Qcoat for nanocomposite coatings is lower than
pure epoxy coating after 1 day of immersion in NaCl solution;
because GO fills the micropores in the coating and decreases the
water uptake into the coating. The Qcoat of pure epoxy coating
decreases to a stable value by extending immersion time due to
delamination of coating and filling the pores of coating by cor-
rosion products [46,52]. However, the Qcoat for nanocomposite
coatings increases by increasing immersion time to 7 days and then,
a decrease in Qcoat value is observed after immersion in electrolyte
for 14 days, which can be ascribed to saturation of GO hydrophilic
structure and the decrease of hydrophilic tendency of nanocom-
posite coatings.
Based on the above mentioned results, the addition of GO to
epoxy coating significantly enhances the barrier properties of coat-
ing and reduces the defects and micro-pores of coating. Meanwhile,
nanocomposite coatings prepared via R + (H + GO) method show
higher corrosion protection ability compared to coatings prepared
Fig. 12. Proposed equivalent circuit for numerical simulation of the EIS measure- via (R + GO) + H method, due to better dispersion quality of GO
ments. nanosheets in coating matrix.

3.3. The effect of GO wt.% on corrosion resistance of

The results presented in Table 3 show that the Rpore and RCT
nanocomposite coatings
values of coatings increase significantly by incorporating small
amount of GO, which is supportive for the enhanced barrier prop-
Since nanocomposite coatings prepared via direct addition of GO
erties of nanocomposite coatings. Meanwhile, the higher value of
to hardener showed superior corrosion resistance, nanocompos-
Rpore and RCT for R + (H + GO) coating than (R + GO) + H sample can
ite coatings containing different wt.% of GO were prepared based
be explained by the good dispersibility of GO in coating and improv-
on R + (H + GO) method. The effect of GO wt.% on corrosion perfor-
ing coating electrical resistance against transfer of corrosive species
mance of nanocomposite coatings is discussed in order to optimize
through the pores of coating. Rpore and RCT values of coatings
the wt.% of GO in nanocomposite coatings.
decrease with prolonging immersion time which is attributed to
uptake of electrolyte into coating. The corrosive electrolyte reaches
at coating/metal interface by extending immersion time; there-
fore, the electrochemical reactions including oxidation of active 3.3.1. FE-SEM results
sites on metal surface and reduction of solution species (O2 and H+ Fig. 13 shows the cross-section images of the prepared
ions) take place at coating/metal interface, leading to formation of nanocomposite coatings containing different wt.% of GO. It can be
corrosion products and delamination of coating. It is implied that seen that GO aggregation occurs by increasing GO wt.% to 0.3 and
pure epoxy coating becomes completely defective after 7 days of 0.5 wt.% and some obvious gaps are observed between the matrix
immersion due to remarkable descending of Rpore and RCT values. and the aggregates, indicating relatively weak interfacial bonding
However, both nanocomposite coatings show slight changes in val- between GO and polymer coating [9]. When the content of GO
ues of Rpore and RCT , suggesting the incorporation of GO retards the increases, good dispersion and exfoliation of GO is not obtained
corrosion reactions on the coated metal surface [50]. in nanocomposite coatings due to high specific area of GO [5].
 S. Pourhashem et al. / Corrosion Science 115 (2017) 78–92 87

Fig. 13. FE-SEM images from the cross-section of nanocomposite coatings prepared via R + (H + GO) method containing (a) 0.05, (b) 0.1, (c) 0.3, and (d) 0.5 wt.% GO.

3.3.2. Potentiodynamic polarization measurement Table 4

Polarization parameters for nanocomposite coatings containing 0, 0.05, 0.1, 0.3, and
Fig. 14 shows the polarization curves of mild steel substrates
0.5 wt.% GO immersed in 3.5 wt.% NaCl solution.
coated by pure epoxy (0 wt.% GO) and nanocomposite epoxy coat-
ing containing 0.05, 0.1, 0.3, and 0.5 wt.% GO. The parameters Sample Ecorr Icorr (␮A/cm2 ) ␤a (V/dec) ␤c (V/dec)
(V vs. SCE)
obtained from polarization curves are summarized in Table 4.
The results reveal that corrosion current density decreases *
0 −0.543 14.6 0.076 −0.350
significantly by incorporating small amount of GO and all nanocom- 0.05 −0.538 3.91 0.137 −0.322
**
0.1 −0.455 0.22 0.188 −0.086
posite coatings show superior corrosion resistance (more positive
0.3 −0.474 0.83 0.235 −0.496
Ecorr , and lower icorr ) compared to pure epoxy coating. Mean- 0.5 −0.541 2.93 0.130 −0.347
while, Ecorr of nanocomposite coatings containing 0.1 wt.% GO shifts *
This sample is R + H (reference coat).
towards the positive directions more than the other samples and **
This sample is R + (H + GO) coating.
shows the least icorr . Therefore, it can be seen that epoxy-0.1 wt.%
GO nanocomposite coating can perform much better corrosion pro-
tection than other nanocomposite coatings. The better performance
of this sample can originate from good GO distribution in epoxy
88 S. Pourhashem et al. / Corrosion Science 115 (2017) 78–92

Fig. 14. (a) Potentiodynamic polarization curves for nanocomposite coatings containing 0, 0.05, 0.1, 0.3, and 0.5 wt.% GO after immersion in 3.5 wt.% NaCl solution for 2 h;
Tafel extrapolation on the polarization plots of nanocomposite coatings containing (b) 0.05, (c) 0.3, and (d) 0.5 wt.% GO for determination of corrosion current and Tafel
slopes.

It can be seen from Fig. 15 that all nanocomposite coatings show

higher (more positive) OCP values compared to pure sample (0 wt.%
GO) during immersion time. Further, the results show that the OCP
tends to more positive values by increasing the amount of GO in
polymer coating at the beginning of immersion (1 day). However,
the OCP values decrease to more negative values after 7 days of
immersion.
The Bode and Nyquist plots of samples after immersion in
3.5 wt.% NaCl solution for 1 and 7 days are presented in Figs. 16–18.
Further, the impedance data is modeled via electrical equivalent
circuit shown in Fig. 12 and the extracted electrochemical param-
eters are presented in Table 5.
Fig. 16 shows that the |Z|0.01Hz value of nanocomposite coatings
significantly depends on GO wt.%. The |Z|0.01Hz value of nanocom-
posite coatings decreases in the order of 0.1 wt.% GO > 0.3 wt.%
GO > 0.5 wt.% GO > 0.05 wt.% GO during immersion time. Further,
the |Z|0.01Hz values of coatings decrease by increasing immersion
Fig. 15. OCP values of nanocomposite coatings containing different wt.% of GO (0,
0.05, 0.1, 0.3, and 0.5 wt.% GO) after immersion in 3.5 wt.% NaCl solution for 1 and time. This behavior of coating ensures the diffusion of corrosive
7 days. electrolyte into the coating and decreasing the barrier proper-
ties of coating. Also, the results show that by incorporating small
matrix, enhancing GO barrier effects against corrosive electrolyte amount of GO (0.05 wt.%) to epoxy coating, the corrosion protec-
[53]. tion performance of coatings increases. Although the |Z|0.01Hz value
of pure epoxy coating and nanocomposite coating with 0.05 wt.%
GO is almost similar after 1 day immersion, the differences in the
3.3.3. Electrochemical impedance measurements
behavior of these coatings can be clearly seen after prolonging
EIS measurement is used to evaluate the corrosion protection
the immersion time. Indeed, although the |Z|0.01Hz value of pure
performance of nanocomposite coatings with different GO wt.%.
epoxy coating decreases significantly after 7 days of immersion, the
The OCP values for different nanocomposite coatings after immer-
|Z|0.01Hz value of epoxy-0.05 wt.% GO decreases slightly.
sion in 3.5 wt.% NaCl solution for 1 and 7 days are shown in Fig. 15.
 S. Pourhashem et al. / Corrosion Science 115 (2017) 78–92 89

Table 5
The electrochemical parameters extracted from EIS data for nanocomposite epoxy coatings containing 0, 0.05, 0.1, 0.3, and 0.5 wt.% GO after immersion in 3.5 wt.% NaCl
solution for different times.

Sample time CPEcoat Rpore CPEdl RCT ␹2

( cm2 ) ( cm2 )

Y0 n Y0 n
(−1 cm−2 sn ) (−1 cm−2 sn )
*
0 1 day 34.3 × 10−10 0.784 0.36 × 106 0.58 × 10−6 0.542 0.81 × 106 0.027
7 days 4.98 × 10−10 0.969 4.38 × 103 80 × 10−6 0.463 11.4 × 103 0.019

0.05 1 day 11.2 × 10−10 0.905 0.51 × 106 0.9 × 10−6 0.748 0.58 × 106 0.025
7 days 9.21 × 10−10 0.915 0.14 × 106 2.05 × 10−6 0.737 0.49 × 106 0.020
**
0.1 1 day 4.45 × 10−10 0.975 5.86 × 106 0.08 × 10−7 0.612 14.0 × 106 0.050
7 days 7.87 × 10−10 0.943 3.24 × 106 0.32 × 10−7 0.485 3.94 × 106 0.009

0.3 1 day 8.70 × 10−10 0.914 2.70 × 106 0.48 × 10−7 0.352 13.8 × 106 0.017
1 day 8.92 × 10−10 0.933 1.35 × 106 1.11 × 10−7 0.177 2.50 × 106 0.015

0.5 24 h 9.17 × 10−10 0.932 3.63 × 106 5.01 × 10−7 0.679 8.66 × 106 0.052
7 days 13.5 × 10−10 0.905 0.85 × 106 1.67 × 10−7 0.741 1.72 × 106 0.008
*
This sample is R + H (pure epoxy coating).
**
This sample is R + (H + GO) coating.

Fig. 16. Bode and phase diagrams of nanocomposite coating containing different
wt.% of GO (0, 0.05, 0.1, 0.3, and 0.5 wt.% GO) after immersion in 3.5 wt.% NaCl solu- Fig. 17. Nyquist plots of nanocomposite coating containing different wt.% of GO (0,
tion for (a) 1 day and (b) 7 days; marker points and solid lines show the experimental 0.05, 0.1, 0.3, and 0.5 wt.% GO) after immersion in 3.5 wt.% NaCl solution for 1 day;
and fitted data, respectively. marker points and solid lines show the experimental and fitted data, respectively.

According to Table 5, all nanocomposite coatings show higher

Rcoat and RCT than pure epoxy coating; meanwhile, nanocompos- GO nanosheets dispersed in nanocomposite coatings act as solid
ite coatings containing 0.1 and then, 0.3 wt.% GO show significantly barrier against corrosive species. In spite of that, the functional
greater Rcoat and RCT values. The increase of Rcoat and RCT indicates groups on GO make it hydrophilic and by increasing the GO, more
the effect of GO nanosheets on decreasing the penetration of cor- electrolytes permeate into coating [44,55]. Moreover, GO aggre-
rosive species to the coating/metal interface through decreasing gation occurs by increasing the amount of GO in nanocomposite
coating porosity [54]. coatings (based on FE-SEM observations), leading to decrease of GO
90 S. Pourhashem et al. / Corrosion Science 115 (2017) 78–92

in the order of pure epoxy > 0.05 wt.% GO > 0.5 wt.% GO > 0.3 wt.%
GO > 0.1 wt.% GO. The salt spray test confirms that GO wt.% has
great effect on corrosion resistance of nanocomposite coatings and
nanocomposite with 0.1 wt.% GO shows the best corrosion protec-
tion performance.

4. Discussion

The delamination of coating on metallic substrates is the result

of corrosive species diffusion into the coating/metal interface via
defects and micro-pores of coating [54]. It should be mentioned
that the following oxidation and reduction reactions take place at
the interface of metal/coating via diffusion of electrolyte beneath
the organic coating, leading to corrosion phenomenon on metal
substrate [1,57,58]:

Fe → Fe2+ + 2e− (1)

2+ 3+ −
Fe → Fe + 1e (2)
− −
O2(g) + 2H2 O + 4e → 4OH (3)

2Fe2+ (aq) + O2(g) + 2H2 O → 2FeOOH + 2H+ (4)

The reactions show that H2 O and O2 are required for corro-

sion of mild steel substrates and the corrosion phenomenon can
be prevented by inhibiting H2 O and O2 from accessing the metal
surface [1,56]. In this regard, GO nanosheets reduce the micro-
scopic defects and pores in the coating and increase the pathways
for electrolyte diffusion to coating/metal interface; therefore, the
defects of coating reduce and the local anode and cathode directions
on metal surface are blocked. Besides, GO nanosheets significantly
influence the barrier properties of epoxy coating due to their sheet-
like structure and make the diffusion pathway of corrosive species
more tortuous. Indeed, the diffusion path of water and oxygen from
coating to substrate is increased due to barrier effect of GO against
corrosive species (as shown schematically in Fig. 20) and therefore,
Fig. 18. Nyquist plots of nanocomposite coating containing different wt.% of GO (0, lower amount of corrosion products, rust and blisters are formed
0.05, 0.1, 0.3, and 0.5 wt.% GO) after immersion in 3.5 wt.% NaCl solution for 7 days; [12,59].
marker points and solid lines show the experimental and fitted data, respectively. The above mentioned properties of GO can be obtained in a
coating system, if well-dispersed GO nanosheets are obtained in
aspect ratio and therefore, accelerates the corrosion phenomenon a polymer coating. Well-dispersed GO nanosheets in epoxy coating
[56]. increase the resistance of coating against the diffusion of corrosive
All of these observations demonstrate that GO wt.% and there- species (H2 O, O2 , and Cl− ), leading to prolong life time of coating.
fore, GO distribution in polymer matrix have an important role The results show that GO nanosheets disperse well in polymers
on coating protective performance. According to FE-SEM results, with low viscosity and also, GO sheets tend to agglomerate by
GO nanosheets tend to agglomerate in polymer matrix by increas- increasing GO wt.% in a coating system. This effect is because of
ing GO wt.% to more than 0.1 wt.% and nanocomposites containing the tendency of oxygen functionalities to agglomerate and form
0.1 wt.% GO have the best GO distribution in polymer matrix; by highly oxidized domains surrounded by areas of pristine graphene.
increasing GO wt.%, GO nanosheets tend to agglomerate due to their It has been shown that GO addition to polymer matrix with
high specific surface area [16,20], reducing the corrosion resistance lower viscosity (either resin or curing agent) enhances the corro-
of nanocomposite coatings. sion resistance of epoxy coatings in 3.5 wt.% NaCl solution. Since
the hardener used in this research has lower viscosity compared
3.3.4. Salt spray test to epoxy resin, nanocomposite coatings prepared via direct GO
The corrosion protection performance of the prepared coatings addition to polyamide hardener (R + (H + GO) sample) show higher
is visually inspected by salt spray test. Fig. 19 shows the images corrosion resistance as a result of good dispersion of GO in coating
of samples after 300 h exposure to accelerated conditions of salt matrix. This situation leads to the increase of coating cross-link
spray chamber. Severe corrosion products, rusts, and blisters have density and the decrease of micro-pores in coating; reducing
appeared on pure epoxy coating and the corrosion products are the electrolyte pathways through coating matrix and providing
extended around the scratch, indicating poor corrosion resistance excellent barrier behavior against corrosive agents. Further, the
of epoxy coatings. Nanocomposite coatings with different wt.% of adhesion of coating to metallic substrate increases via GO addi-
GO show different behavior; although rust and blisters (with small tion to polymer coating and meanwhile, R + (H + GO) sample reveals
size) are observed on coatings with 0.05 wt.% GO, lower extent of higher pull-off adhesion. High adhesion of nanocomposite coat-
rust and corrosion products appear near the scribe of coating with ings accompanying with their excellent barrier property provide
0.5 wt.% GO. Meanwhile, nanocomposite coatings including 0.1 and restricted delamination at coating/metal interface in the presence
0.3 wt.% GO show no blister and only very few rust spots can be of corrosive electrolyte.
seen over these samples, and the corrosion around the scratch is The amount of GO in coating matrix is another important param-
not extended. Therefore, the delamination of coatings decreases eter for improving corrosion resistance of epoxy coatings. Although
 S. Pourhashem et al. / Corrosion Science 115 (2017) 78–92 91

Fig. 19. Images of epoxy coatings containing 0, 0.05, 0.1, 0.3, and 0.5 wt.% GO after 300 h exposure to salt spray test.

Fig. 20. Schematic representation of corrosion protection mechanism by epoxy-GO nanocomposite coatings on mild steel substrates.

the corrosion protection ability of epoxy coating becomes better Nanocomposites prepared via direct addition of GO to polyamide
even by incorporation of 0.05 wt.% GO, this nanocomposite coating hardener show better dispersion quality and superior corrosion
cannot provide long term corrosion resistance. The best corro- protection compared with nanocomposites synthesized by direct
sion protection is obtained in the term of nanocomposites with addition of GO to epoxy matrix due to lower viscosity of polyamide
0.1 wt.% GO and by increasing the wt.% of GO to 0.3 and 0.5 wt.%, the hardener vs. epoxy resin used in this research. Further, based on
corrosion protection performance of coatings decreases. Indeed, FE-SEM results and electrochemical measurements, 0.1 wt.% GO
the corrosion protection ability of coatings decreases in the is chosen as an appropriate GO content to prepare nanocompos-
order of epoxy-0.1 wt.% GO > epoxy-0.3 wt.% GO > epoxy-0.5 wt.% ites with the best corrosion resistance; because by increasing GO
GO > epoxy-0.05 wt.% GO » pure epoxy. The aggregation of GO amount to 0.3 and 0.5 wt.% in nanocomposite coatings, GO aggrega-
nanosheets occurs via increasing the amount of GO in coating tion happens which reduces the barrier properties of coatings. We
matrix; hence, the protective performance of nanocomposite coat- believe that the proposed methods for controlling GO dispersion
ing declines. Nanocomposite coatings with optimum GO content are versatile plans to enhance the corrosion promotion activity of
show good chemical bonding at polymer/GO interface and fewer epoxy-GO nanocomposite coatings.
defects. Consequently, the compactness of coating increases and
the amount of micro-pores in coating decreases, limiting the per-
meation of corrosive electrolyte through coating. Acknowledgements

We gratefully acknowledge the financial support of Materi-

5. Conclusion als and Energy Research Center (MERC) and Research Institute of
Petroleum Industry (RIPI).
In conclusion, we have presented solvent based epoxy-GO
nanocomposite coatings for corrosion protection of mild steel sub-
References
strates. The nanocomposite coatings were shown to efficiently
protect mild steel substrate due to their enhanced barrier proper- [1] C.H. Chang, T.C. Huang, C.W. Peng, T.C. Yeh, H.I. Lu, W.I. Hung, C.J. Weng, T.I.
ties compared to pure epoxy coatings. The quality of GO dispersion Yang, J.M. Yeh, Novel anticorrosion coatings prepared from
in polymer coating is the main concept for achieving desirable bar- polyaniline/graphene composites, Carbon 50 (2012) 5044–5051.
[2] K.C. Chang, W.F. Ji, M.C. Lai, Y.R. Hsiao, C.H. Hsu, T.L. Chuang, Y. Wei, J.M. Yeh,
rier properties and corrosion protection in nanocomposite coatings. W.R. Liu, Synergistic effects of hydrophobicity and gas barrier properties on
The viscosity of polymer matrix and the GO wt.% are two impor- the anticorrosion property of PMMA nanocomposite coatings embedded with
tant variables to prepare well-dispersed GO in polymer coatings. graphene nanosheets, Polym. Chem. 5 (2014) 1049–1056.
92 S. Pourhashem et al. / Corrosion Science 115 (2017) 78–92

[3] E. Armelin, R. Oliver, F. Liesa, J.I. Iribarren, F. Estrany, C. Alemı́an, Marine paint [31] S.K. Yadav, J.W. Cho, Functionalized graphene nanoplatelets for enhanced
fomulations: conducting polymers as anticorrosive additives, Prog. Org. Coat. mechanical and thermal properties of polyurethane nanocomposites, Appl.
59 (2007) 46–52. Surf. Sci. 266 (2013) 360–367.
[4] M. Qian, A.M. Soutar, X.H. Tan, X.T. Zeng, S.L. Wijesinghe, Two-part [32] T. Remyamol, H. John, P. Gopinath, Synthesis and nonlinear optical properties
epoxy-siloxane hybrid corrosion protection coatings for carbon steel, Thin of reduced graphene oxide covalently functionalized with polyaniline, Carbon
Solid Films 517 (2009) 5237–5242. 59 (2013) 308–314.
[5] S. Liu, H. Yan, Z. Fang, H. Wang, Effect of graphene nanosheets on morphology, [33] D. Konios, M.M. Stylianakis, E. Stratakis, E. Kymakis, Dispersion behaviour of
thermal stability and flame retardancy of epoxy resin, Compos. Sci. Technol. graphene oxide and reduced graphene oxide, J. Colloid Interface Sci. 430
90 (2014) 40–47. (2014) 108–112.
[6] Y.J. Wan, L.C. Tang, L.X. Gong, D. Yan, Y.B. Li, L.B. Wu, J.X. Jiang, G.Q. Lai, [34] T. Jiang, T. Kuila, N.H. Kim, B.C. Ku, J.H. Lee, Enhanced mechanical properties of
Grafting of epoxy chains onto graphene oxide for epoxy composites with silanized silica nanoparticle attached graphene oxide/epoxy composites,
improved mechanical and thermal properties, Carbon 69 (2014) 467–480. Compos. Sci. Technol. 79 (2013) 115–125.
[7] Z. Zhang, W. Zhang, D. Li, Y. Sun, Z. Wang, C. Hou, L. Chen, Y. Cao, Y. Liu, [35] K. Min, T.H. Han, J. Kim, J. Jung, C. Jung, S.M. Hong, C.M. Koo, A facile route to
Mechanical and anticorrosive properties of graphene/epoxy resin composites fabricate stable reduced graphene oxide dispersions in various media and
coating prepared by in-situ method, Int. J. Mol. Sci. 16 (2015) 2239–2251. their transparent conductive thin films, J. Colloid Interface Sci. 383 (2012)
[8] B. Qi, S.R. Lu, X.E. Xiao, L.L. Pan, F.Z. Tan, J.H. Yu, Enhanced thermal and 36–42.
mechanical properties of epoxy composites by mixing thermotropic liquid [36] Y. Dong, Q. Liu, Q. Zhou, Time- dependent protection of ground and polished
crystalline epoxy grafted graphene oxide, Polym. Lett. 8 (7) (2014) 467–479. Cu using graphene film, Corros. Sci. 90 (2015) 69–75.
[9] Y.J. Wan, L.X. Gong, L.C. Tang, L.B. Wu, J.X. Jiang, Mechanical properties of [37] Z. Li, R. Wang, R.J. Young, L. Deng, F. Yang, L. Hao, W. Jiao, W. Liu, Control of
epoxy composites filled with silane-functionalized graphene oxide, Compos. the functionality of graphene oxide for its application in epoxy
A 64 (2014) 79–89. nanocomposites, Polymer 54 (2013) 6437–6446.
[10] W.G. Ji, J.M. Hu, J.Q. Zhang, C.N. Cao, Reducing the water absorption in epoxy [38] S.I. Abdullah, M.N.M. Ansari, Mechanical properties of graphene oxide
coatings by silane monomer incorporation, Corros. Sci. 48 (2006) 3731–3739. (GO)/epoxy composites, HBRC J. 11 (2) (2015) 151–156.
[11] M.Y. Jiang, L.K. Wu, J.M. Hu, J.Q. Zhang, Silane-incorporated epoxy coatings on [39] S.G. Prolongo, R. Moriche, A. Jimenez-Suarez, M. Sanchez, A. Urena,
aluminum alloy (AA2024), Part 1: improved corrosion performance, Corros. Advantages and disadvantages of the addition of graphene nanoplatelets to
Sci. 92 (2015) 118–126. epoxy resins, Eur. Polym. J. 61 (2014) 206–214.
[12] B. Ramezanzadeh, E. Ghasemi, M. Mahdavian, E. Changizi, M. [40] M. Martin-Gallego, M.M. Bernal, M. Hernandez, R. Verdejo, M.A.
Mohammadzadeh, Covalently-grafted graphene oxide nanosheets to improve Lopez-Manchado, Comparison of filler percolation and mechanical properties
barrier and corrosion protection properties of polyurethane coatings, Carbon in graphene and carbon nanotubes filled epoxy nanocomposites, Eur. Polym. J.
93 (2015) 555–573. 49 (2013) 1347–1353.
[13] X. Wang, W. Xing, L. Song, H. Yang, Y. Hu, G.H. Yeoh, Fabrication and [41] M.J. Jiang, Z.M. Dang, H.P. Xu, Enhanced electrical conductivity in chemical
characterization of graphene-reinforced waterborne polyurethane modified carbon nanotube/methylvinyl silicone rubber nanocomposite, Eur.
nanocomposite coatings by the sol–method, Surf. Coat. Technol. 206 (2012) Polym. J. 43 (2007) 4924–4930.
4778–4784. [42] Y. Dong, D. Bhattacharyya, Effects of clay type, clay/compatibiliser content and
[14] K.C. Chang, M.H. Hsu, H.I. Lu, M.C. Lai, P.J. Liu, C.H. Hsu, W.F. Ji, T.L. Chuang, Y. matrix viscosity on the mechanical properties of polypropylene-organoclay
Wei, J.M. Yeh, W.R. Liu, Room-temperature cured hydrophobic nanocomposites, Compos. A 39 (7) (2008) 1177–1191.
epoxy/graphene composites as corrosion inhibitor for cold-rolled steel, [43] T. Monetta, A. Acquesta, F. Bellucci, Graphene/epoxy coating as
Carbon 66 (2014) 144–153. multifunctional material for aircraft structures, Aerospace 2 (2015) 423–434.
[15] Y.N. Singhbabu, B. Sivakumar, J.K. Singh, H. Bapari, A.K. Pramanick, R.K. Sahu, [44] S. Mohammadi, F.A. Taromi, H. Shariatpanahi, J. Neshati, M. Hemmati,
Efficient anti-corrosive coating of cold-rolled steel in a seawater environment Electrochemical and anticorrosion behavior of functionalized graphite
using an oil-based graphene oxide ink, Nanoscale 7 (2015) 8035–8047. nanoplatelets epoxy coating, J. Ind. Eng. Chem. 20 (6) (2014) 4124–4139.
[16] X. Wang, W. Xing, P. Zhang, L. Song, H. Yang, Y. Hu, Covalent functionalization [45] P. Li, X. He, T.C. Huang, K.L. White, X. Zhang, H. Liang, R. Nishimurae, H.J. Sue,
of graphene with organosilane and its use as a reinforcement in epoxy Highly effective anti-corrosion epoxy spray coatings containing
composites, Compos. Sci. Technol. 72 (2012) 737–743. self-assembled clay in smectic order, J. Mater. Chem. A. 3 (2015) 2669–2676.
[17] S.Y. Yang, W.N. Lin, Y.L. Huang, H.W. Tien, J.Y. Wang, C.M. Ma, S.M. Li, Y.S. [46] G. Ruhi, S.K. Dhawan, Conducting Polymer Nano Composite Epoxy Coatings
Wang, Synergetic effects of graphene platelets and carbon nanotubes on the for Anticorrosive Applications, Intech, 2014.
mechanical and thermal properties of epoxy composites, Carbon 49 (2011) [47] W. Sun, L. Wang, T. Wu, M. Wang, Z. Yang, Y. Pan, G. Liu, Inhibiting the
793–803. corrosion-promotion activity of graphene, Chem. Mater. 27 (2015)
[18] A. Beigbeder, R. Mincheva, M.E. Pettitt, M.E. Callow, J.A. Callow, M. Claes, P. 2367–2373.
Dubois, Marine fouling release silicone/carbon nanotube nanocomposite [48] F. Brusciotti, D.V. Snihirova, H. Xue, M.F. Montemor, S.V. Lamaka, M.G.S.
coatings: on the importance of the nanotube dispersion state, J. Nanosci. Ferreira, Hybrid epoxy–silane coatings for improved corrosion protection of
Nanotechnol. 10 (2010) 2972–2978. Mg alloy, Corros. Sci. 67 (2013) 82–90.
[19] L.C. Tang, Y.J. Wan, D. Yan, Y.B. Pei, L. Zhao, Y.B. Li, L.B. Wu, Y.X. Jiang, G.Q. Lai, [49] J. Mondal, A. Marques, L. Aarik, J. Kozlova, A. Simoes, V. Sammelselg,
The effect of graphene dispersion on the mechanical properties of Development of a thin ceramic-graphene nanolaminate coating for corrosion
graphene/epoxy composites, Carbon 60 (2013) 16–27. protection of stainless steel, Corros. Sci. 105 (2016) 161–169.
[20] Z. Li, R.J. Young, R. Wang, F. Yang, L. Hao, W. Jiao, W. Liu, The role of functional [50] W.G. Ji, J.M. Hu, L. Liu, J.Q. Zhang, C.N. Cao, Water uptake of epoxy coatings
groups on graphene oxide in epoxy nanocomposites, Polymer 54 (2013) modified with ␥-APS silane monomer, Prog. Org. Coat. 57 (2006) 439–443.
5821–5829. [51] L.K. Wu, J.T. Zhang, J.M. Hu, J.Q. Zhang, Improved corrosion performance of
[21] M.M. Gudarzi, F. Sharif, Enhancement of dispersion and bonding of electrophoretic coatings by silane addition, Corros. Sci. 56 (2012) 58–66.
graphene-polymer through wet transfer of functionalized graphene oxide, [52] A. Ghasemi-Kahrizsangi, H. Shariatpanahi, J. Neshati, E. Akbarinezhad,
Polym. Lett. 6 (12) (2012) 1017–1031. Corrosion behavior of modified nano carbon black/epoxy coating in
[22] H. Yang, C. Shan, F. Li, Q. Zhang, D. Han, L. Niu, Convenient preparation of accelerated conditions, Appl. Surf. Sci. 331 (2015) 115–126.
tunably loaded chemically converted graphene oxide/epoxy resin [53] K. Qi, Y. Sun, H. Duan, X. Guo, A corrosion-protective coating based on a
nanocomposites from graphene oxide sheets through two-phase extraction, J. solution-processable polymer-grafted graphene oxide nanocomposite,
Mater. Chem. 19 (46) (2009) 8856–8860. Corros. Sci. 98 (2015) 500–506.
[23] D.R. Bortz, E.G. Heras, I. Martin-Gullon, Impressive fatigue life and fracture [54] B. Ramezanzadeh, A. Ahmadi, M. Mahdavian, Enhancement of the corrosion
toughness improvements in graphene oxide/epoxy composites, protection performance and cathodic delamination resistance of epoxy
Macromolecules 45 (2012) 238–245. coating through treatment of steel substrate by a novel nanometric sol-gel
[24] H. Yang, C. Shan, F. Li, Q. Zhang, D. Han, L. Niu, Convenient preparation of based silane composite film filled with functionalized graphene oxide
tunably loaded chemically converted graphene oxide/epoxy resin nanosheets, Corros. Sci. 109 (2016) 182–205.
nanocomposites from graphene oxide sheets through two-phase extraction, J. [55] Z. Yu, H. Di, Y. Ma, L. Lv, Y. Pan, C. Zhanga, Y. He, Fabrication of graphene
Mater. Chem. 19 (2009) 8856–8860. oxide–alumina hybrids to reinforce the anti-corrosion performance of
[25] H. Koolivand, A. Sharif, M.R. Kashani, M. Karimi, M. Koolivand Salooki, M.A. composite epoxy coatings, Appl. Surf. Sci. 351 (2015) 986–996.
Semsarzadeh, Functionalized graphene oxide/polyimide nanocomposites as [56] H. Wei, D. Ding, S. Wei, Z. Guo, Anticorrosive conductive polyurethane
highly CO2 -selective membranes, J. Polym. Res. 21 (2014) 599–611. multiwalled carbon nanotube nanocomposites, J. Mater. Chem. A 1 (2013)
[26] A. Esfandiar, O. Akhavan, A. Irajizad, Melatonin as a powerful bio-antioxidant 10805–10813.
for reduction of graphene oxide, J. Mater. Chem. 21 (2011) 10907–10914. [57] Y.H. Yu, Y.Y. Lin, C.H. Lin, C.C. Cha, Y.C. Huang, High-performance
[27] B. Paulchamy, G. Arthi, B.D. Lignesh, A simple approach to stepwise synthesis polystyrene/graphene-based nanocomposites with excellent anti-corrosion
of graphene oxide nanomaterials, J. Nanomed. Nanotechnol. 6 (2015) 1–4. properties, Polym. Chem. 5 (2014) 535–550.
[28] ASTM D4541-09, Standard test method for pull-off strength of coatings using [58] B. Ramezanzadeh, S. Niroumandrad, A. Ahmadi, M. Mahdavian, M.H.
portable adhesion testers, ASTM Int. (2009). Mohamadzadeh Moghadam, Enhancement of barrier and corrosion
[29] ASTM B117, Standard practice for operating salt spray (fog) apparatus, ASTM protection performance of an epoxy coating through wet transfer of amino
Int. (2011). functionalized graphene oxide, Corros. Sci. 103 (2016) 283–304.
[30] H. Yang, F. Li, C. Shan, D. Han, Q. Zhang, L. Niu, A. Ivaska, Covalent [59] Z. Yang, W. Sun, L. Wang, S. Li, T. Zhu, G. Liu, Liquid-phase exfoliated
functionalization of chemically converted graphene sheets via silane and its fluorographene as a two dimensional coating filler for enhanced corrosion
reinforcement, J. Mater. Chem. 19 (2009) 4632–4638. protection performance, Corros. Sci. 103 (2016) 312–318.