Abstract
Polyaniline (Pani) and exfoliated polyaniline graphite (EPaniG) nanocomposites were used to modify the protective properties of an epoxy zinc-rich primer (ZRP). The corrosion resistance properties of primers were evaluated in 3.5% sodium chloride solution for a period of 120 days via electrochemical noise (EN) and electrochemical impedance spectroscopy (EIS). EN data were analysed via Wavelet and Hilbert spectra analysis to evaluate the protection mechanism of different ZRP coatings during immersion. Coating resistance and charge transfer resistance of the primers were evaluated by EIS measurements. Results showed good compatibility between |Z|0.01Hz from EIS measurements and mean noise resistance which results from EN data and these evaluations allowed the examination of coating performances during immersion. Free corrosion potential (E corr) measurements and salt spray test revealed that both the cathodic protection and barrier properties of the ZRP primer were improved after addition of EPaniG nanocomposite to the ZRP sample.
Introduction
Since the 1930s, zinc-rich primers (ZRPs) have been used to protect materials against corrosion especially in marines and harsh industrial environments [1,2]. ZRPs protect the steel substrate through sacrificial protection mechanism [3], and only zinc particles in galvanic contacts with the substrate are contributing to this effect [4]. In these coatings, zinc particles in the primer should electrically be connected to the steel substrate and they should contain high levels of metallic zinc powder (over 80% by weight) which dispersed in a binder with high saponification resistant [5]. Protective performances of ZRPs strongly depend on the pigment volume concentration (PVC), the shape and the size of zinc powder [6]. High level of zinc powder used in these primers encounter problems of increasing porosity of the resulted dry film. Lamellar zinc pigments exhibit a higher surface area than spherical zinc pigments [7] and the use of lamellar zinc could reduce the PVC of the coating. Meroufel and Touzain have reported the addition of carbon black [8] and polyaniline [9] into the zinc-rich powder coatings in order to enhance the electronic conduction paths between zinc particles inside the coating and the steel substrate. They found that by adding polyaniline into the ZRPs the corrosion potential remained stable but cathodic during 100 days of immersing in 3% sodium chloride solution. Jalili et al. reported enhancing the galvanic action and corrosion protection properties of zinc-rich paint by the replacement of 2% by weight of zinc dust in the standard ZRP coating with aluminium nanoparticle [10]. Hayatdavoudi and Rahsepar studied the addition of graphene nanosheets with superior percolation action in ZRP coating to provide an enhanced barrier effect against aggressive species and thereby provide a less aggressive environment for corrosion of zinc particles [11]. Park and Shon studied on the conductivity and adhesion strength and hence on the corrosion protectiveness of ZRP coated carbon steel [12]. They found that epoxy zinc coatings containing multiwall carbon nanotubes and higher zinc contents showed higher corrosion protectiveness which correlated with an increase in the cathodic protection of coated carbon steel [12]. Cheng et al. reported that the addition of graphene into the waterborne ZRP improved the effective zinc content and the cathodic protection of the coating due to the electrical conductivity of graphene [13]. The coating with 2 wt-% graphene maintained cathodic protection for 40 days [13]. Recently, Hayatdavoudi and Rahsepar reported superior protection properties for ZRPs containing inhibitor-intercalated layered double hydroxide (LDH) nanocontainers due to the combination of barrier effect, sacrificial cathodic protection and smart inhibition action [14]. Teng et al. introduced Zinc-reduced graphene oxide into ZRPs to enhance anticorrosion performance. They claimed generating improved barrier properties against aggressive species and long-term cathodic protection ability for modified zinc-rich coatings [15]. Ramezanzadeh et al. claimed on a new strategy for providing ZRPs with enhanced cathodic and barrier protection mechanisms simultaneously using conductive polyaniline/graphene oxide composites [16]. In the previous works, we studied the influence of polyaniline and polyaniline nanocomposites on the protective behaviour of ZRPs [17–20]. Results revealed that the conductive form of polyaniline nanocomposites can improve electrical contact between zinc particles themselves and the steel substrate. These nanocomposites can act as a barrier and electrochemically active inhibitors to decrease the porosity of the paint. Therefore, it could be predicted that the modified primers had higher barrier properties than the original unmodified primer. In this paper, we reported the results of electrochemical noise (EN) measurements to evaluate protective performance of polyaniline (Pani) and exfoliated polyaniline graphite (EPaniG) nanocomposite modified ZRPs. Attempts were made to use different methods for analysing EN data and mechanistic study of the enhanced cathodic protection performance of modified ZRPs. Finally, the EN results were approved with EIS and salt spray test evaluations.
Experimental
Pani and EPaniG nanocomposites were synthesised using a procedure reported elsewhere [18]. A commercial three component ZRP (Zinc dust: Epoxy vehicle: hardener = 36:6:3) with practical data which is presented in Table 1 was selected and 1 wt-% of its zinc content was replaced by the synthesised Pani and EPaniG nanocomposites, separately. Carbon steel panels with elements compositions presented in Table 2 were used as substrate. The carbon steel panels were blasted to SSPC-SP 5 (according to SSPC-VIS 1-89) with a profile of 30-40 µm (according to ASTM D4417-14 method B), and then decreased with toluene before painting. Commercial and modified ZRPs have been applied on the carbon steel panels separately with a dry film thickness of 100 ± 10 µm. Painting was done by air spray equipment. Coated panels were stored in a standard atmosphere having a temperature of 23 ± 2°C and a relative humidity of 50 ± 5%. Measurements were started in 10 days. For EN measurements, two nominally identical coated panels were used as working electrodes and a saturated calomel electrode (SCE) was used as a reference. An area of 2 cm2 for each working electrode was selected and an H type glass tube was attached to this area for evaluations. Glass tube was filled with 3.5 wt-% NaCl aqueous solution and the electrochemical measurements were performed up to 120 days of immersion. Immersion levels of the test specimens were marked and maintained by additions of deionised water as required (see Figure 1). EN measurements were done by ACM Field Machine instrument and electrochemical potential and current noise were measured simultaneously. Data acquisition was performed within a period of 512 s at 0.5 s intervals with a frequency range from 1 to 0.5 Hz, based on the expressions fmax = 1/2Dt and fmin = 1/NDt where Dt and N are the data acquisition time interval and the total number of data recorded, respectively. Before statistical data analysis, DC trend removal was done through moving average removal (MAR) method using MATLAB software. EIS measurements were carried out in a three-electrode-cell system. One of coated steel specimen, (as working electrode), a graphite rod (as the counter electrode) and a SCE (as reference electrode) were used in EIS evaluations (see Figure 1). EIS measurements were performed using a Potentiostat-Galvanostat model 273A and SI 1255 frequency response analyzer at the perturbation and frequency range of 10 mV and 10 kHz to 10 mHz, respectively. ZView2 software with 100 iterations fitting setup was used to analyse the impedance data. A digital multimeter (Philips PM 2519) was used to determine the free corrosion potential (E corr.) of the coated samples during the immersion period. Different coated panels were evaluated in corrotherm 616 salt spray cabinet in accordance with standard ASTM B117-16 up to 2000 h. Water vapour transmission rate passes through free films of different coatings were determined based on ASTM D1653-13.
Schematic representation of the electrochemical cell. Practical properties of selected epoxy-based ZRP. Chemical compositions of used carbon steel panels (wt-%).
Results and discussion
EN measurements
For the analysis of EN, potential (E) and current (I) noise data must be collected simultaneously [21]. Figure 2 shows typical time records of filtered electrochemical current noise plots for ZRPs coated steel panels at initial and after 120 days immersion in 3.5% NaCl solutions. It can be seen in this figure that the EPaniG modified ZRP has the highest noise at initial and after 120 days of immersing. The second order statistics, the ratio of the standard deviations of voltage fluctuation divided by the standard deviations of current, is used to determine the noise resistance values of different samples, Equation (1).
Typical filtered electrochemical current and potential noise of different coatings after 1 h (a, b), and 120 days (c, d) of immersion in saline water. Noise resistance (Rn) and |Z|0.01 Hz. for different ZRP coatings during immersion in saline water.
and 
are potential and current standard deviations, respectively. The values of 
and 
are determined based on Equations. (2) and (3).
, 
are the mean values of the recorded potential and current, respectively, and N is the total number of data points. The statistical treatment shown above applied to the data and Rn calculated for each ZRP coating. The Rn data plotted as a function of exposure time is shown in Figure 3. It is clearly shown in Figure 3 that during 120 days of immersion the lowest Rn values belong to the EPaniG and then Pani modified ZRPs. Based on the literature, it could be assumed that Rn is a measurement of the diffusion barrier properties of the coating: Rn has been equated to the zero frequency limits of impedance modulus (polarisation resistance Rp) and/or coating resistance Rf [22]. This result means that the modified ZRPs have higher electrical conductivity during the 120 days of immersion compared to unmodified ZRP. ZRPs are known as cathodic protective coatings. Electrical contact between zinc particles and the steel substrate is essential for coating performance. These coatings are formulated with a high percentage of metallic zinc dust to ensure electrical conductivity. Therefore, one of the key features of a ZRP is its high content of powdered metallic zinc (typically higher than 65 vol.-%) [23]. High content of zinc powder results in higher porosity of the resulting dry film. In unmodified ZRP, corrosive agents easily reach to the coating/steel interface and then zinc powder acts as a sacrificial anode and sacrifices to protect the metal [24]. As a result of these reactions, nonconductive zinc oxide and other zinc corrosion products are formed which tend to seal the pores between the zinc particles and increasing the coating resistance. In the case of modified primers, the formation of zinc oxide was delayed due to the barrier properties of Pani and fully exfoliated nanocomposite which was added to the primer. Lower noise resistance values in modified ZRPs can be attributed to the increased electrical conductivity of these primers due to the presence of Pani nanoparticles and EPaniG nanocomposites. On the other hand, in the case of the EPaniG sample, the conductivity of nanocomposite increased due to the presence of conductive graphite, and its presence in the coating film would increase the electrical conductivity and, as a result, the noise resistance of the modified primer would be the least.
The fast Fourier transform (FFT) was used to convert the EN data from the time domain to the frequency domain. It has been observed that the slope of current FFT plots changes with increasing coating degradation [25]. Figure 4 shows PSD plots of the current noise obtained from coated panels during immersion. In Figure 4(a) the highest current PSD belongs to the Pani modified ZRP, and all other coatings have the same current PSD. The observed increase in the PSD current results from the rise in the passed current between the working electrodes. After 7 days of immersion (Figure 4(b–f)) the current PSD for blank ZRP is lower than modified ZRPs. Decreasing PSD current resulted from a reduction in the passed current due to the formation of zinc corrosion product which could block the pores presented in the dry film. If q is assumed as the charge pass over the circuit, it could be provided from the voltage and current noise signals based on equation ( Typical PSD current plots of different coatings at an initial time (a), and after 7 days (b), 30 days (c), 60 days (d), 100 days (e), and 120 days (f) of immersion in saline water. Charge (q) variations of different coatings during immersion in saline water. Schematic representation of protection mechanism of modified ZRP (a), and modified ones (b).
) [25]. Where; 
represent low-frequency values of current PSD and potential PSD respectively, and B is the Stern–Geary coefficient. Figure 5 shows the characteristic charge (q) variations of coated panels during the immersion. Determination of corrosion event charges 
were achieved via PSD current and potential plots at the lowest frequency (i.e. 0.001 Hz) and the B value of 0.026 V. It can be clearly observed in Figure 5 that after 30 days of immersing the highest q values belong to the EPaniG modified ZRP and unmodified ZRP has the lowest q values. The reduced amounts of charge variations associated with the blank sample can be attributed to the formation of higher amounts of nonconductive zinc oxide in the film. But in the case of modified ZRPs, the presence of conductive nanoparticles acts as a bridge between zinc particles and as a result, the amount of charge variations was much higher (see schematic in Figure 6).
Wavelet transform and Hilbert transmission spectra were used to more precisely analyse the noise data and to study the performance mechanism of different ZRP coatings. Figure 7 shows energy distribution plot of the Electrochemical current noise (ECN) signal for different coatings after 120 days of immersion in saline water and in Figure 8, a schematic representation of the most essential information from an energy distribution plot is provided [26]. Short timescale crystals, typically d2 and d3, are associated with localised (activation control) and large timescale crystals, i.e. d7, and d8 provide information on general corrosion (diffusion control). Medium timescale crystals d4–d6 represent processes under mixed control (regarding localised corrosion, both diffusion and activation control can be of influence) [26]. As seen in Figure 7(a), the contribution of crystals d1 to d3 for the blank sample is greater than the Pani sample and the Pani sample is larger than the EPaniG sample. These results revealed that, for the blank sample the probability of local corrosion has increased due to the inactivation of zinc particles after 120 days of immersion. As Figure 7(b) shows, the relative shares of d7 and d8 crystals are higher than the other crystals. It can be concluded that the ECN signals are provided by the largest transitions that are related to general corrosion for all samples after 120 days of immersion. Figure 9 shows a typical Hilbert spectrum of the ECN signals of different coatings after 120 days of immersion. The higher contribution of low-frequency events for EPaniG samples which are shown in Figure 9 indicates a general corrosion in the EPaniG modified ZRP coating compared to the Pani modified ZRP and blank ones.
Energy distribution plots of the ECN signal (a), and the sum of d1–d6 and d7&d8 (b) for different ZRP primers after 120 days of immersion in saline water. Mechanistic information about corrosion processes that can be obtained from an energy distribution plot [26]. Typical Hilbert spectrum of the ECN signals of Blank ZRP (a), ZRP + Pani (b) and ZRP + EPaniG (c) coatings after 120 days of immersion in saline water.
The results presented in Figures 7 and 9 confirm that in modified coatings, after 120 days of immersion in saline water, there are also active metallic zinc particles on the coatings and they cover their cathodic protection performance, but in the case of unmodified ZRP coating there is a possibility of inactivation of zinc particles in some areas of coating and the occurrence of local corrosion.
EIS measurements
Typical Bode impedance and Nyquist plots of carbon steel panels coated with blank and modified ZRPs during 120 days of immersing in 3.5% sodium chloride solution are shown in Figure 10. Two time-constant data are clearly seen in this figure during all immersion times. From the equivalent electrical circuit (EEC) models which are proposed in the literature for interpretation of EIS data of ZRPs, a porous film model was used in the analysis [8,27–29]. The appropriate EEC model fitted by the measured EIS data is shown in Figure 11. In this circuit Rs, Rc, Cc, Rct and Cdl are the solution resistance, the coating resistance, the coating capacitance, the charge transfer resistance and the double layer capacitance, respectively. The results of coating resistance, charge transfer resistance, and impedance at 10 mHz value obtained from fitting of EIS data for different samples are shown in Table 3. The results reveal that at initial times of immersion the EPaniG modified coating showed the highest resistance, while the Pani modified ones had the lowest resistance. Coating resistances decreased at initial days of immersion and then increased during all days of remaining immersion time. As mentioned before for ZRP coatings, at the first stage of protection zinc behaves as an anode and scarifies itself to protect the metal. As a result of these reactions, zinc corrosion products are formed and seal the pores between the zinc particles, increasing the coating resistance. In the case of modified primers, the formation of zinc oxide was delayed due to the barrier properties of conductive Pani and fully exfoliated nanocomposite which was added to the primers. Therefore, the modified primers showed lower resistance at the end of the immersion. But in the case of unmodified ZRP, the formation of more nonconductive zinc oxide causes to increase the coating resistance and hence decreasing the cathodic protection ability of the coating.
Typical plots of ZRP (a), ZRP + Pani (b) and ZRP + EPaniG (c) coatings during immersion in saline water. Equivalent electrical circuits to model coated panels behaviour exposed to 3.5% NaCl solution. Optimum values of coating resistance, charge transfer resistance, and impedance at 10 mHz values which are obtained by Zview software.
Free corrosion potential measurements
Measuring the free corrosion potential (E corr) is a useful method to assess the cathodic protection duration of ZRPs [3,4,30,31]. Variations in E corr values for coated steel panels during the immersion in 3.5% sodium chloride solution up to 120 days are shown in Figure 12.
Variations of free corrosion potential of different ZRPs coatings during immersion in saline water.
For carbon steel which is immersed in 3.5% NaCl solution, cathodic protection will be guaranteed if E corr remains lower than −0.86 V/SCE [28]. It is believed that the electrochemical processes occurring in such systems are the oxidation of zinc particles (
) and the reduction of dissolved oxygen (
). The overall reaction can also be expressed as 
[1,4,9]. Zinc corrosion product (
) is a nonconductive material and its formation within the coating can cause electrical discontinuity of the zinc particles between themselves and with the steel substrate. The E corr. evolution for ZRPs coated samples is in close relationship with the ratio of active area (zinc/steel) and allows defining the cathodic protection duration. In other words, an increase in E corr corresponds to a decrease of the electro-active zinc area which means a reduction in cathodic protection intensity. As seen in Figure 12, for all samples the E corr values are lower than −0.86 V/SCE up to 100 days of immersion. This shows that the zinc particles are electrochemically active up to 100 days. But after 120 days of immersion, E corr values of the original primer reaches to −0.81 V/SCE. This reveals that in the case of original ZRP the zinc pigments were consumed or galvanic contact was lost with time and the cathodic protection performance of carbon steel substrate may not be guaranteed with this primer after 100 days of immersion in saline water, as confirmed by EN and EIS measurements.
Salt spray test
In order to evaluate the practical performance of modified coatings, salt spray test was chosen as an accelerated test because of its widespread usage in the coating industry. The coated panels with 150 cm2 area were tested in salt fog cabinet according to ASTM B117-16 [32]. The panels inspected visually at regular intervals up to 2000 h and digital photographs were taken prepared for different coatings (see Figure 13). Evaluations were done using the method described by ASTM D1654-05, ASTM D610-01 and ASTM D714-02 to determine the extent of corrosion and blistering. Figure 14 shows digital photographs of different coatings after 2000 h salt sprat test with higher magnification compared to Figure 13. It can be seen in Figure 14 that after exposing coatings inside salt spray cabinet for 2000 h, blisters were clearly observed on the surface of unmodified ZRP primer and zinc corrosion products, as white scales, were abundantly appeared on the surface of this primer. In the case of ZRP + EPaniG, no sign of rusts and blisters were observed on the surface of coating and a few white zinc corrosion products were appeared on the coating surface.
Visual appearance of different coated panels at an initial time and during salt spray test up to 2000 h. Visual appearance of different coated panels after 2000 h salt spray test.
Figure 15 display cross-section SEM micrographs corresponding to original ZRP and modified ones before salt spray exposure test. In SEM micrographs the dark region is corresponding to the film porosity and bright ones are shown zinc particles. SEM images show that the distribution of the zinc particles is quite homogeneous. The SEM graphs shown in Figure 15(a,b) revealed frequent pores in the original ZRP and PAni modified ones. It can be seen in these figures that the ZRP has a porous structure which is conductive pathways for corrosive agents. The zinc particles are electrically in contact with each other and steel substrate, providing an electron pathway from the surface of ZRP to the steel substrate. In the case of EPaniG modified coating shown in Figure 15(c) a compact layer can be seen in the top of the film. The formation of this layer is due to the presence of EPaniG nanocomposite which is lighter than zinc particles. These particles migrate to the top of the film and create a compact layer. Figure 16 display cross-section SEM micrographs corresponding to original ZRP and modified ones after 2000 h salt spray test. After exposure to the saline solution the zinc particles attacked with corrosive agents, forming corrosion products within the coating. Figure 16(a,b) shows that the original ZRP and PAni modified ones are extremely attacked in saline solution. Better performance of EPaniG modified primer is a result of the formation of compact layer at the top of dry film which can delay the diffusion of corrosive media so as to improve the barrier properties. It is clearly shown in these figures that zinc content of studied primers are high and the vehicle could not wet them, and micro-pores are frequently formed in dry films. These pores are forms as a network of capillaries which are filled with electrolyte and create ionic conduction paths between the anodic (zinc particles) and cathodic zones (steel). Morcillo et al. [24] illustrated two different conduction processes with ZRP: an ionic conduction and an electronic conduction. According to Morcillo et al. when ionic conduction exists without electronic conduction, both anodic and cathodic reactions take place on the zinc surface, ensuring a rapid accumulation of insulating zinc corrosion products on zinc particle surface. While in the case of electronic conduction only the anodic reactions performed on zinc particles and cathodic reaction in the bare zones on the steel surface.
Cross-section view SEM micrographs of Blank ZRP (a), ZRP + Pani (b), and ZRP + EPaniG (c) before salt spray exposure. Cross-section view SEM micrographs of Blank ZRP (a), ZRP + Pani (b), and ZRP + EPaniG (c) after 2000 h salt spray exposure.
Water vapour transmission rate
Water vapour transmission rate (WVT) from free films of different coatings is shown in Figure 17. The free film sample of unmodified ZRP coating has a high vapour flow rate (WVT = 1.74 g m−2 h−1) due to its high percentage of zinc powder, which produces fine pores in the coating film. A remarkable point in this figure is that the increase of PAni to ZRP coating has increased the amount of WVT from the film. Increasing the WVT can be attributed to the inappropriate distribution of PAni particles in the ZRP matrix and also to the increase in the volume of the pigment present in the coating because the density of the PAni is less than one-third of the density of zinc powder. However, with increasing EPAniG to ZRP coating, the water vapour transmission rate is reduced and reached to 1.22 g m−2 h−1 and these results are in good agreement with SEM observations and electrochemical evaluations.
Average water vapour transmission rate passes through free films of different coatings.
Conclusion
Pani and EPaniG nanocomposites were used to modify cathodic protection properties of epoxy ZRP. Results obtained are listed below:
The lowest Rn and |Z|0.01Hz values were belonged to the EPaniG modified ZRP, while the unmodified ZRP had the highest Rn and |Z|0.01Hz values during all time of immersion. The Rn values were comparable with the and |Z|0.01Hz values which are extracted from EIS measurements. Energy distribution plot of the ECN signal and Hilbert transmission spectrum were successfully used to analyse more precisely the noise data and to study the performance mechanism of different ZRP coatings. In the case of modified primers, the formation of nonconductive zinc oxide was delayed due to the good barrier properties of conductive Pani and EPaniG nanocomposites. Results revealed that partial replacement of zinc particles by conductive Pani or EPaniG nanocomposite enhanced the cathodic protection properties of the ZRP via improving its barrier effects and inhibition of zinc powder consuming. Results showed more negative Ecorr. values of EPaniG modified ZRP after 365 days of immersion in saline water which can guarantee the cathodic protection of carbon steel.
Footnotes
Disclosure statement
No potential conflict of interest was reported by the author.