Corrosion resistance of Fe–Zn surfaces of galvannealed steel revealed after coulometric stripping in acid,alkaline and saline media

Abstract

The corrosion of galvannealed steel revealed three different Fe–Zn surfaces on the steel substrate, and each surface showed a different corrosion behaviour in a given environment. In order to evaluate the corrosion behaviour of each revealed surface, the electrochemical impedance spectroscopy (EIS) technique was performed in acid, alkaline and saline media containing chlorides and sulphates. The Mossbauer spectroscopy was also used to elucidate the iron phases on each surface, and these results identified the delta phase on all surfaces obtained after coulometric dissolution of galvannealed steels. In the hydrochloric acid and sodium chloride solution, the outer surface richest in zinc showed the highest polarisation resistance among the surfaces obtained using coulometric stripping of galvannealed steel, and the corrosive process occurred on the electrode surface. In the alkaline solution, the inner layer showed the highest polarisation resistance among the surfaces obtained after dissolution of galvannealed steel, and the equivalent circuit fitted to the EIS data is characteristic of a corrosion of a porous layer.

Keywords

Mossbauer spectroscopy,Galvannealed steel Electrochemical impedance spectroscopy Fe–Zn phases

Introduction

Electrogalvanised and hot dip galvanised steels without heat treatment show the phase η in the zinc layer. The processes of stamping, welding and painting of these steels were developed in order to prevent operational problems.

Hot dip galvanised steels with heat treatment, or galvannealed (GA) steels, are used in automotive industries because of their properties of weldability, workability and corrosion resistance.^1–3 The quality of the zinc layer is related to the chemical composition, microstructure and surface condition of steel as well as the chemical composition and temperature of the zinc bath, and the operational conditions such as immersion time and heat treatment of the galvanised steel.

The GA steels show iron–zinc phases such as ζ; (5-6 wt-%Fe), δ (10-13 wt-%Fe) and ┌(20-22 wt-% Fe).⁴ The presence of these phases can produce stamping problems such as powdering, flaking and welding.^5–7 Powdering as well as flaking are phenomena related to the loss of adherence of the coating. When the GA steel strip is submitted to compression strength, fracture of the gamma phase occurs, producing powdering. The thickness of gamma phase is decisive in the occurrence of powdering. In the case of flaking, a layer of zeta phase is gripped in the stamping matrix during the pressing process. Spalling occurs when the friction forces overcome the adhesion forces on the material surface and produce flaking.

The characterisation of the zinc–iron layer is important to optimise the industrial process in order to control the formation and growth of the Fe–Zn phases.

Techniques such as electron probe microanalyser, X-ray diffraction (XRD), transmission electron microscopy (TEM), secondary ion mass spectrometry, inductively coupled plasma optical emission spectrometry (ICP-OES) and Mossbauer spectroscopy have been used in order to evaluate the iron content, the thickness of the galvanneal coating and the depth dependence of each Fe–Zn phase of galvanneal coating.^8–14

Electrochemical stripping has been used to obtain the microstructural profile of the galvanneal layer of GA steels.^3–5 The basic principle of this technique is that the different phases have different potentials. At a specific current density, each phase dissolves at one potential. The electrochemical stripping provides an indirect compositional profile of the galvanneal coating. The parameters of electrochemical stripping of GA steels were optimised in a previous work.¹⁵ The results obtained using electrochemical stripping are similar to the glow discharge optical emission spectroscopy and Auger electron spectroscopy results.¹⁵ Each plateau of the electrochemical stripping curve corresponds to a compositional range and to one predominant Fe–Zn phase.¹⁵

The surfaces obtained after coulometric dissolution of the GA steel were characterised using XRD, scanning electron microscopy (SEM), energy dispersive spectroscopy, and X-ray photoelectron spectroscopy in a previous work.¹⁵ In the present work, the surfaces obtained after coulometric stripping were characterised using Mossbauer and electrochemical impedance spectroscopy (EIS). The GA steel is an important industrial product that requires a compositional and chemical characterisation regarding application in acid, alkaline and a medium containing chloride ions. Knowledge of GA steel and its structure is relevant to project its application, mechanical conformation and heat treatment.

Literature on galvanised steel corrosion mainly describes corrosion performance of galvanised steel on the basis of zinc layer corrosion without mentioning the role of intermetallic layers, which are formed during the hot dip galvanising process.^16,17 Yadav et al. ³ have investigated the effect of Fe–Zn alloy layers, which are formed during hot dip galvanising process on the corrosion resistance of galvanised steel under atmospheric marine environment using an accelerated wet–dry cyclic corrosion test, potentiodynamic polarisation, surface potential measurement and AC impedance techniques.

Yadav et al. ^3,16 and Silva et al. ¹⁸ have described the corrosion resistance of GA steel under marine atmosphere. Yadav et al. ³ discussed what role such layers play in enhancing the protective ability of zinc corrosion products considering atmospheric corrosion. However, in solution, the role that the Fe–Zn phases play in corrosion process was not clarified in literature.

Queiroz and Costa⁴ studied the coulometric stripping and characterised the Fe–Zn phases using potentiodynamic polarisation and potentiostatic polarisation only in the stripping solution. The corrosion resistance of each Fe–Zn surface revealed using coulometric stripping in acid, alkaline and saline aqueous media is a gap in literature. As the steel corrosion progresses, new steel surfaces are exposed to the environment, and the evaluation of corrosion resistance of these materials, which are distinct from the original, is of technological and scientific relevance. The EIS is a tool that was used to elucidate the mechanisms of corrosive processes and has sensitivity to differentiate the iron–zinc surfaces regarding chemical properties. The Mossbauer spectroscopy was also used to elucidate the iron phases of each surface revealed using stripping, and the iron phases present on surfaces affect their corrosion resistance in a given medium. The aim of this work is to investigate the corrosion resistance of each iron–zinc layer revealed using coulometric stripping of GA steel in an acid, alkaline and saline aqueous media.

Methodology

The GA steel was obtained from industrial reels, degreased in laboratory, and the steel samples were submitted to the electrochemical stripping.

The chemical composition of the steels was determined: carbon and sulphur contents of the steels were analysed using combustion with detection by infrared, using LECO 444 LS equipment, Leco Corporation. Aluminium concentration was evaluated using ICP-OES, Spectroflame Modula of the Spectro GMBH. Manganese, silicon, phosphorus, niobium, titanium and nitrogen were analysed using the SRS 3000 Sequential X-Ray Spectrometer, Siemens.

The mass and the chemical composition of the zinc layer of the GA steels were evaluated in five samples, 40 mm × 80 mm in dimension. The samples were degreased, cleaned, weighed and immersed in a solution for removal of the zinc layer. The removal solution used was an aqueous solution of hydrochloric acid (1:1 v/v) with an addition of hexamethylenetetramine (3.5 g L^{− 1}) as a corrosion inhibitor. The immersion time was considered until the end of hydrogen evolution. The samples without the zinc layer were cleaned with distilled water, dried with nitrogen and weighed. The mass of the galvanised layer was determined by calculating the mass difference between the galvanised sample and the same sample after the removal of the zinc layer.

The chemical composition of the zinc–iron layer was evaluated using ICP-OES, Spectroflame Modula Spectro GMBH. The contents of zinc, iron, aluminium and silicon were determined by analysing the removal solution.

For the electrochemical stripping, the 100 mm × 75 mm samples were degreased using an aqueous solution of 20 g L^{− 1} of a commercial alkaline degreasing at 60°C, cleaned in distilled water and dried with nitrogen.¹⁵ The electrochemical cell has a capacity of 300 mL, and the sample has an exposed area to the electrolyte of 19.63 cm². The galvanostatic measurement of the electrochemical potential of each phase was carried out using three solutions reported in literature^12,19,20: solution 1 is an aqueous solution of 2.73M sodium chloride+0.27M zinc chloride, pH = 5.0 ± 0.1; solution 2 is an aqueous solution of 1M sodium sulphate+0.35M zinc sulphate heptahydrate, pH = 5.5 ± 0.1; solution 3 is an aqueous solution of 3.42M sodium chloride+0.35M zinc sulphate heptahydrate, pH = 4.5 ± 0.1. The values of 1, 2, 4 and 7.5 mA cm^{− 2} of current density were studied using the three solutions.^9,10 The best result of electrochemical dissolution of GA steel was obtained for the 0.35 mol L^{− 1} zinc sulphate heptahydrate and 3.42 mol L^{− 1} sodium chloride aqueous solution. A 2 mA cm^{− 2} current density at pH 4.5 produced a better definition of the potential levels and a reduced testing time.¹⁴

The equipment used was EG & G Princeton Applied Research Potentiostat/Galvanostat, model 273.

Electrochemical measurements were performed with a potentiostat brand Princeton Applied Research; model Versa Stat 3, Versa Studio software. The samples analysed were GA steels, and each layer revealed using coulometric dissolution.¹⁵ The electrochemical measurements were performed at room temperature (25 ± 0.05°C) using an electrochemical cell of three electrodes with a capacity of 200 mL. The media used were electrolytic solutions of 3.42 mol L^{− 1} NaCl+0.35 mol L^{− 1} ZnSO₄.7H2O, NaCl 0.35 mol L^{− 1}, NaOH 0.1 mol L^{− 1} and HCl 0.1 mol L^{− 1}. The exposed area of the steel samples in contact with the electrolyte was 1.0 cm². The samples were washed with deionised water and 70% ethanol after the electrochemical measurements. The reference electrode used was Ag/AgCl, KCl (sat) and the counter electrode used was a platinum plate. After the open circuit potential stabilised, the steel sample was analysed using an EIS technique in the frequency range of 10 kHz to 1 mHz, and an amplitude of 10 mV in relation to the open circuit potential. For data treatment, the ZviewTM software was used. An analysis in terms of the corrosion mechanism was also performed. The impedance measurements were performed in quintuplicate.

Mossbauer spectra were obtained at room temperature using a source of iron isotope 57 (Eo = 14.37 keV, RT = 1.96 meV) immersed in the cobalt matrix. Measurements were made in the configuration of backscattering electrons. The Conversion Electron Mossbauer Spectroscopy analysis is a technique that results in a maximum depth of ∼2000 Å of material from the surface and is therefore sensitive to phases containing iron present on the exposed surfaces of the steel.

The morphology of the zinc–iron layer was determined before and after each stage of electrochemical stripping using the scanning electron microscopy, EVO 50 Zeiss equipment, 20 kV of acceleration tension, coupled with the energy dispersive spectrometer, INCA 350 model, and the wave dispersion spectrometer, model INCA 500i, Oxford Instruments.

Results

The chemical composition of the commercial steels studied is shown in Table 1.

Table 1

Chemical composition of substrate of GA steels

Steel	Chemical composition/wt-%
	C	Mn	S	P	Si	Ti	Nb	Alsol
GA	0.0019	0.12	0.009	0.013	0.01	0.022	0.019	0.034

The mass per area and the chemical composition of the zinc layer are shown in Table 2.

Table 2

Chemical composition and mass of galvanised layer

		Chemical composition of the galvanised layer/mass-%
Steel	Zinc mass/g m^{− 2}	Zn	Fe	Al	Si
GA	40.2	Balance	12.74	0.31	< 0.002

The surface of the GA steel, the surfaces of outer, intermediate and inner layer obtained performing coulometric stripping are shown in Fig. 1a–d respectively. Columnar, hexagonal, stick and platelet crystals were identified on the surface of the outer layer obtained using the electrochemical dissolution technique (Fig. 1b). Figure 1c shows the surface of the intermediate layer, and some columnar crystals were identified on this surface. On the surface of the inner layer, there were no crystals, and the surface seemed to be melted as shown in Fig. 1d.

Image (SEM) of a GA steel and surfaces associated of b outer, c intermediate and d inner layer obtained from coulometric stripping

Mossbauer spectrometry

Figure 2 shows the Mossbauer spectra of samples of GA steel, the outer, intermediate and inner layers, and the substrate from top to bottom in the sequence.

Mossbauer spectra of samples of GA steel, outer, intermediate and inner layers, and substrate from top to bottom in sequence

Electrochemical impedance spectroscopy

Table 3 shows the values of corrosion potential of GA steel and surfaces, which were revealed using coulometric stripping, in an acid, alkaline and saline media. In acid and saline solutions, the GA steel showed the lowest corrosion potential among the samples studied. In the alkaline medium, the outer layer rich in zinc showed the lowest corrosion potential among the samples analysed.

Table 3

Corrosion potential of GA steel and surfaces obtained using coulometric stripping

Electrolyte	Corrosion potential /mV versus Ag/AgCl
	Inner layer	Intermediate	Outer layer	GA
HCl 0.1 mol L^{− 1}	− 504 ± 35	− 497 ± 49	− 516 ± 31	− 776 ± 54
NaOH 0.1 mol L^{− 1}	− 400 ± 32	− 608 ± 54	− 1156 ± 89	− 670 ± 60
NaCl 0.35 mol L^{− 1}	− 736 ± 44	− 776 ± 62	− 801 ± 64	− 908 ± 73
NaCl 3.42 mol L^{− 1} + ZnSO₄.H2O 0.35 mol L^{− 1}	− 613 ± 55	− 615 ± 37	− 626 ± 44	− 858 ± 61

In the acid medium, the inner, intermediate and the outer layer showed one maximum peak at intermediate frequencies (10-100 Hz) in the Bode diagram of phase angle versus frequency (Fig. 3a), and the data were adjusted to the equivalent circuit shown in Table 4. The outer layer, richest in zinc, showed the highest polarisation resistance in the media of HCl solution (Fig. 3b and Table 4), and the highest phase angle of the maximum peak in Bode diagram (close to 40°), demonstrating a more capacitive behaviour than the intermediate and inner layer of the galvanneal coating. The surface richest in iron (inner layer) showed the lowest polarisation resistance in an aqueous solution of hydrochloric acid (Table 4). The GA steel showed more than one time constant, and the equivalent circuit adjusted to the data is shown in Table 4. The surface of GA steel showed the highest phase angle (65°) at the maximum peak, indicating a more capacitive behaviour than did the surfaces obtained after electrochemical stripping (Fig. 3a).

Bode diagrams of phase angle versus a frequency and b impedance modulus of iron–zinc phases in aqueous solution of hydrochloric acid 0.1 mol L^{− 1}

Table 4

Electrochemical parameters of GA steel and surfaces obtained after electrochemical stripping in acid, alkaline and saline aqueous solutions

		R1Ω cm²	CPE1/F sⁿ cm^{− 2}	R 2/Ω m²	CPE2/F sⁿ cm^{− 2}	R 3/Ω cm²	Equivalent circuit
HCl	GA	239 ± 14	4.33 10^{− 6} ± 3.03 10^{− 7}	7979 ± 718	6.31 10^{− 8} ± 4.41 10^{− 9}	4.81 10⁴ ± 2.88 10³

HCl	Outer	224	4.14 10^{− 5}	1719	…	…
		± 15	± 2.64 10^{− 6}	± 69
	Intermediate	221 ± 11	4.16 10^{− 5} ± 1.66 10^{− 6}	662 ± 39	…	…

	Inner	224	1.46 10^{− 4}	275	…	…
		± 11	± 5.84 10^{− 6}	± 22
NaCl	GA	502 ± 35	8.85 10^{− 7} ± 2.65 10^{− 8}	6144 ± 491	5.01 10^{− 6} ± 3.00 10^{− 7}	7541 ± 603

NaCl	Outer	480	1.22 10^{− 3}	2531	…	…
		± 38	± 9.76 10^{− 5}	± 101
	Intermediate	494 ± 34	1.92 10^{− 3} ± 9.60 10^{− 5}	770 ± 46	…	…

	Inner	470	4.02 10^{− 3}	1103	…	…
		± 23	± 3.21 10^{− 4}	± 77
NaOH	GA	165 ± 8	7.70 10^{− 5} ± 5.39 10^{− 6}	23755 ± 1425	1.90 10^{− 4} ± 7.60 10^{− 6}	32715 ± 2290

NaOH	Outer	143 ± 10	4.35 10^{− 4} ± 3.48 10^{− 5}	4101 ± 246	…	…

	Intermediate	142	1.11 10^{− 3}	1834	…	…
		± 11	± 6.66 10^{− 5}	± 146
NaOH	Inner	110 ± 9	4.28 10^{− 5} ± 3.42 10^{− 6}	146 ± 9	9.01 10^{− 5} ± 7.20 10^{− 6}	2.45 10⁵ ± 1.71 10⁴

ZnSO₄ + NaCl	GA	22 ± 1	2.38 10^{− 7} ± 1.42 10^{− 8}	5271 ± 421	2.78 10^{− 5} ± 2.50 10^{− 6}	2.54 10⁴ ± 1.77 10³

ZnSO₄ + NaCl	Outer	21 ± 1	2.78 10^{− 4} ± 2.22 10^{− 5}	1420 ± 127	…	…

	Intermediate	23	2.46 10^{− 4}	1377	…	…
		± 1	± 2.03 10^{− 5}	± 96
ZnSO₄ + NaCl	Inner	23 ± 2	1.04 10^{− 3} ± 8.32 10^{− 5}	243 ± 14	4.21 10^{− 4} ± 2.94 10^{− 5}	980 ± 59

The Bode diagram of phase angle of surfaces obtained after electrochemical stripping (Fig. 4a) showed a onetime constant in an NaCl solution, and the equivalent circuit fitted to data is shown in Table 4. The outer layer showed the highest angle (28°) at the maximum peak, and the intermediate layer showed the lowest angle (17°) and a more resistive behaviour in an aqueous sodium chloride solution among the surfaces obtained after stripping. The maximum peak of the inner layer is at a lower frequency than the peak of the outer and intermediate layer. The GA steel showed two time constants and peaks at higher frequencies than the surfaces obtained after coulometric stripping. The equivalent circuit fitted to GA steel EIS data is shown in Table 4. In the Bode diagram of impedance modulus (Fig. 4b), the GA steel showed the highest impedance and the intermediate layer showed the lowest impedance in saline solution as shown in Table 4.

Bode diagrams of phase angle versus a frequency and b impedance modulus of iron–zinc phases in aqueous solution of sodium chloride 0.35 mol L^{− 1}

In alkaline solutions, the inner layer and the GA steel showed a higher corrosion resistance than the outer and intermediate layers in alkaline solutions (Fig. 5a). The outer and intermediate layers obtained after stripping showed one peak in the Bode diagram of phase angle (Fig. 5b) corresponding to a one time constant, and the equivalent circuit fitted to data is shown in Table 4. The inner layer showed the highest phase angle (close to 80°) at the maximum peak demonstrating a capacitive behaviour. The inner layer and the GA steel showed broader peaks in the Bode diagram of phase angle (Fig. 5b), and circuits with two time constants were adjusted to EIS data.

Bode diagrams of a impedance modulus and phase angle versus b frequency of iron–zinc phases in aqueous solution of 3.42 mol L^{− 1} NaCl+0.35 mol L^{− 1} ZnSO₄.7H2O

In the electrolyte of sodium chloride and zinc sulphate, the GA steel showed a higher corrosion resistance than the surfaces revealed using dissolution of GA steel (Fig. 6a). The Bode diagram of phase angle of intermediate and outer layers showed one time constant (Fig. 6b), and the surface of the inner layer and the GA steel showed two time constants.

Bode diagrams of a impedance modulus and phase angle versus b frequency of iron–zinc phases in aqueous solution of sodium hydroxide 0.1 mol L^{− 1}

Discussion

The substrate of GA steels has a typical chemical composition of ultralow carbon steel and interstitial free (IF) atoms. The stabilising agents are Ti and Nb for the GA steels. The IF-Nb and IF-Ti+Nb steels are most commonly used as a substrate for the GA steel industrially produced having a better adhesion of the coating layer, increasing the resistance to mass loss during the forming processes. The content of phosphorus is low (0.013 wt-%). Literature reported that phosphorus in steel can delay or suppress the growth of phase and promote the growth of ζ; phase.²¹ The higher the content of phosphorus in steel, the faster the growth of the ζ; phase.

The galvanneal layer contains Zn, Fe, Al and Si. The steel substrate contains 0.01 wt-% of silicon, and the content of silicon in the galvanneal layer is < 0.002 wt-%. Aluminium is added in the galvanising bath to produce an inhibitor layer to prevent the formation of iron zinc phases. Aluminium additions of 0.135 mass-% at 460°C are sufficient to precipitate intermetallic compounds of Fe₂Al₅ on the surface of the metal plate.^19,22 In practice, the aluminium content in the zinc bath flux is >0.135 wt-% to ensure precipitation of all the iron dissolved in the intermetallic compound Fe₂Al₅. Precipitation of Fe₂Al₅ is limited by the availability of aluminium from the zinc bath flux.^19,22 If the solidification of the coating occurs before completion of the precipitation of dissolved iron as Fe₂Al₅, the remaining free iron is not precipitated and is dissolved in the liquid phase. Zinc from the galvanising bath precipitated very rapidly as crystalline zeta phase rods. These crystals are responsible for the embrittlement during the stamping operations.^19,22

The GA layer before the electrochemical dissolution showed the presence of stick crystal, which is a characteristic of the zeta phase, and columnar crystal, which is typical of the delta phase.¹⁸ According to XRD results, which were obtained in a previous work, the delta phase is predominant in galvanneal layer (62.3% of counts per second), followed by the gamma (27.6%) and zeta (10.1%) phases.¹⁵

In the coulometric dissolution, each phase dissolves at a characteristic potential, which was in a range between − 860 and − 855 mV(SCE) for the zeta phase, − 808 and − 798 mV(SCE) for the delta phase and between − 721 and − 716 mV(SCE) for the gamma phase.^18,23 The dissolution time was 2290 s for the zeta phase, 810 s for the delta phase and 368 s for the gamma phase.¹⁵ In this work, the surface obtained at a lower potential was named as the outer layer, the surface obtained at intermediate potential was the intermediate layer and the surface obtained at a higher potential was named as the inner layer. It is not possible to dissolve one phase at each time, as there is an interpenetration of iron–zinc phases on each layer.¹⁵ The predominant phase in the galvanneal coating was the delta phase, representing 62.3% of the galvanneal layer as obtained using XRD analysis.¹⁵ When the stripping started, the zeta and delta phases dissolved, the delta phase was dissolved preferentially and the fraction of gamma phase increased from 27.6 wt-% on the surface of GA steel to 66.8% on the outer layer.¹⁵ After the second stage of electrochemical stripping, more delta phase was dissolved, and the fraction of gamma phase increased until 73.9% on the intermediate layer.¹⁵ After the third stage of electrochemical stripping, only the gamma phase was identified on the inner layer.¹⁵

Mossbauer spectroscopy

In all samples analysed using Mossbauer spectroscopy, it is possible to identify the subspectrum corresponding to the substrate, which is present in Fig. 2 as the sextet. The steel base (sextet) is detectable before any step of the removal of the Fe–Zn layers. This may indicate non-homogeneous coating with a small total thickness < 2000 Angstroms. The delta phase appears in the spectrum of all surfaces analysed.

The samples with the outer layer exposed, after the first step of anodic dissolution, always showed a poor signal/noise ratio, resulting in measurements that do not allow identification of phases even after action times of over 140 h. This result can be interpreted supposing that the surface with the zeta phase results in exposed components with a large emission of secondary electrons (not Mossbauer), which contribute only to the background of the spectra.

According to the Mossbauer results, the delta phase appears in all samples and in a higher content than the zeta phase on the outer and intermediate layers. Chakraborty et al. ⁸ characterises the coated layer on industrially produced GA IF steel and also found that the coating contained mostly the delta phase. The delta and zeta phases were consisted of three and one subspectrum respectively as reported in literature.¹⁰ On the surface of the inner layer, the delta phase was identified in a low concentration (Fig. 2). According to the XRD results obtained in a previous work, the delta phase was predominant on the surface of GA steel.¹⁵ The delta phase was not identified using XRD on the surface of the inner layer and on the substrate after electrochemical stripping.¹⁵

Corrosion potential

The inner layer constituting the gamma phase showed a more positive corrosion potential than did the intermediate and outer layers in the alkaline solution. This phase contains the highest iron content, and at pH >9, the passivation of iron occurs.²⁴ In the alkaline solution, the open circuit potential of the outer layer, which contains a high fraction of zeta phase, is the lowest. In electrolytes of an aqueous hydrochloric acid solution, an aqueous sodium chloride solution and a solution of NaCl 3.42 mol L^{− 1} and ZnSO₄.H2O 0.35 mol L^{− 1}, the values of corrosion potential of the three intermetallic phases were similar, and the GA steel showed the lowest corrosion potential. The GA steel surface shows a heterogeneous distribution of Zn–Fe phases, presenting a selective dissolution of zinc, the presence of cracks arising in the manufactured process and the existence of paths with different impedances for the charge transfer.⁴

The corrosion potential values of all three Fe–Zn alloy phases (gamma, delta and zeta phases) were found to be lower than that of Fe (0.60 V versus SSE), but higher than the value of Zn (1.00 V versus SSE).²⁵ Recently, Rout et al. ²⁶ have reported on the electrochemical as well as corrosion behaviour of GA and galvanised coatings, which supported the finding of Lee et al. ²⁵ The corrosion potential of the surfaces analysed was close to the potential range between the potential of iron and of zinc.

Electrochemical impedance results

Hydrochloric acid aqueous solution

After the first step of GA steel dissolution, the outer layer, richest in zinc, showed the highest polarisation resistance in the media of hydrochloric acid solution (Fig. 3b), and the highest phase angle of the maximum peak in the Bode diagram (Fig. 3a), demonstrating its more capacitive behaviour than the intermediate and inner layer of the galvanneal coating. In acid media, the cathodic reaction of hydrogen reduction is inhibited on zinc surface. Activation polarisation occurs on the zinc surface in the acid media due to the high hydrogen overvoltage.²⁴ The surface richest in iron (inner layer) showed the lowest polarisation resistance in an aqueous solution of hydrochloric acid. Independent of the polarisation resistance value, the outer, intermediate and inner layers exhibited one time constant characteristic of the corrosion process occurring on the electrode surface.

The GA steel showed the highest corrosion resistance in all media, except in the alkaline solution, where the iron content on the surface played the crucial role in determining the corrosion resistance of the metal surface. The GA steel showed two time constants in acid solution, and the corrosive process occurs on the electrode surface and on an inner layer in the galvanneal layer. The galvanneal coating is heterogeneous, with preferential paths of selective dissolution of iron or zinc depending on the electrolyte. In hydrochloric acid solution, the hydrogen atoms adsorb on the zinc surface preventing the hydrogen evolution, inhibiting the cathodic reaction. In acid medium, zinc dissolution is inhibited and iron dissolution is favoured. In a previous work, the characterisation of the galvanneal coating was performed. The eta, zeta, delta and gamma phases were identified. The zeta and delta phases interpenetrated each other, and a regular and horizontal layer of each phase was not observed. Depending on the corrosive medium, one phase dissolves preferentially and the corrosion process progresses toward the substrate, and the equivalent circuit showed in Table 4 for the GA steel in HCl solution is a physical description of this system. The GA steel surface shows a heterogeneous distribution of Zn–Fe phases, a selective dissolution of zinc or iron depending on the electrolyte, the presence of cracks arising in the manufactured process and the existence of paths with different impedances generating more than one time constant and more than one surface where the corrosive process occurs.⁴

The electrode surface cannot be considered as an ideal capacitor according the data obtained and the constant phase element (CPE) is used as a substitute for the ideal capacitor in equivalent circuits given in Table 4. This fact can be explained by the heterogeneous nature of the solid surface or by the dispersion of some physical property values of the system.²⁷ Therefore, the impedance ZCPE is defined as 1 where Q is the CPE (Ω^{− 1} cm^{− 1} sⁿ) and j ² = − 1. For n = 1, the CPE can be identified as a capacitance.^27,28

Saline aqueous solution

In a saline solution of NaCl, and a solution of ZnSO₄ and NaCl, GA steel showed the highest corrosion resistance among the samples studied (Figs. 4 and 5). The highest coating thickness and the barrier effect seemed to play a major role in the corrosion behaviour of GA steel. The electrolyte of sodium chloride and zinc sulphate was selected from literature data²⁰ for the coulometric dissolution of the GA steel and has a greater conductivity among the electrolytes studied, being the most aggressive medium to the steel surface. Lee et al. ⁹ have studied the electrochemical behaviour of pure and homogeneous Fe–Zn intermetallic phases in a NaCl + ZnSO₄ solution at pH 5. The authors have shown that the polarisation resistance value of an iron–zinc alloy increases with the increase in Fe content. In this study, the polarisation resistance values of the outer and intermediate layers richest in zinc in NaCl and ZnSO₄ solution were of the same magnitude order, and were higher than the polarisation resistance of the inner layer richest in iron. This result differs from the result obtained by Lee et al. ⁹ and can be explained by the presence of zinc ions in the electrolyte, which inhibits the dissolution of zinc. The production of corrosion products such as zinc hydroxide [Zn(OH)₂] and zinc hydroxychloride [Zn₅(OH)₈Cl₂.H₂O] in saline neutral media can block the coating pores and the coating acts as a corrosion resistant layer.²⁸ The outer and intermediate layers in presence of sodium chloride and zinc sulphate showed higher impedance values than the inner layer, one time constant and the process of charge transfer is limited on the electrode surface. The blocking of pores by zinc corrosion products can explain this behaviour. The values of the CPE obtained for the outer layer richest in zinc of 10^{− 5} and 10^{− 4} F sⁿ cm^{− 2} are characteristic of zinc layers.¹⁸ The inner layer showed lower corrosion resistance than the outer and intermediate layers and two time constants in sodium chloride and zinc sulphate, and the corrosive process occurs on the surface and at the steel/layer interface.

The polarisation resistance of the GA steel in solution of sodium chloride and zinc sulphate was one magnitude order higher than the polarisation resistance of the surfaces revealed using stripping. The surface of GA steel is richest in zinc, and the blocking of coating pores and cracks by zinc corrosion products can explain this behaviour. For the GA steel, a corrosion front reaching the intermetallic layers will have different corrosion behaviours in terms of kinetics and galvanic effect.³ The GA steel showed two time constants and a diffusion control of corrosive process at low frequencies. The criterion for a coating system to take part in both charge transfer as well as diffusion control process is that the ratio of Rct/Rcp ranges between 0.2 and 5.²⁸ It is found that the ratio of charge transfer resistance (Rct) to coating pore resistance (Rcp) is one which falls in that range (Rct/Rcp = 0.2075). The ratio is calculated using the data shown in Table 4. Nyquist diagram also showed the diffusive control in the corrosion process of GA steel.

For diffusion controlled process the Warburg impedance is given by 2 where B = l/(D)^1/2, D is the diffusion coefficient, l is the diffusion layer thickness, Y0 = [σ(2)^1/2]^1/2 and σ is the Warburg coefficient.

Little is known regarding the electrochemical behaviour of Fe–Zn alloy layers on the corrosion behaviour of galvanised steel. Yadav et al. ³ discussed what role such layers play in enhancing the protective ability of zinc corrosion products considering atmospheric corrosion. However, in solution, the role that the Fe–Zn phases play in corrosion process was not clarified in literature.

Alkaline medium

Figure 6 shows the highest corrosion resistance of the inner layer in an alkaline environment. The impedance modulus (Fig. 6a) and the phase angle at the maximum peak (Fig. 6b) were the highest for the inner layer in an aqueous solution of NaOH, indicating a more capacitive behaviour of the inner layer in this medium. The inner layer in alkaline medium showed two time constants and an equivalent circuit characteristic of a localised corrosion through a porous layer. According to Pourbaix diagram for the iron–water system, iron exhibits a passive behaviour in alkaline environment.²⁴ The stable phase of iron for the pH studied and the corrosion potential of inner layer is Fe₂O₃, and this region is of passivity in iron–water Pourbaix diagram at 25°C.²⁴

Considering the pH value and the corrosion potential for the GA steel, the stable phase is ZnO₂ ^{2 −}, and the region is of corrosion.²⁴ In this case, the corrosion mechanism is the selective dissolution of zinc, and the steel is protected by the galvanic effect. The highest thickness of coating before the coulometric stripping also contributes to improve the corrosion resistance of GA steel in all media.

The polarisation resistance of the inner layer, which is richest in iron in an alkaline medium, is two magnitude orders higher than the polarisation resistance of the intermediate and outer layers. In this case, the highest content of iron on the surface plays a critical role in the corrosion resistance of metal, in spite of the smaller thickness in relation to the outer and intermediate layer. The outer and intermediate layers richest in zinc showed one time constant, and the equivalent circuit fitted to data indicates that the charge transfer process occurs mainly on the electrode surface. The GA steel also showed a high corrosion resistance in alkaline solution, two time constants and the corrosive process occurs on the electrode surface and on an inner interface in the galvanneal layer.

The Kramers–Kroning (K-K) relationships were applied to check the validity of experimental results. Any system that satisfies the a priori conditions of linearity, stability and causality must satisfy the K-K relationships. The K-K technique transforms the real component into the imaginary component and vice versa, so that the transformed quantities may be compared directly with their corresponding experimental values for the same parameters. A good agreement between the experimental and transformed impedance data for both real and imaginary components validates the EIS data.²⁹

The GA steel showed the highest corrosion resistance in all media, except in the alkaline solution, where the iron content on the surface plays the crucial role in determining the corrosion resistance of the steel. According to our results, the presence of internal interfaces in galvanneal coating coupled to the interpenetration of Fe–Zn phases in the coating can be crucial in conferring a high corrosion resistance of GA steel in several media. Corrosion process progresses toward the substrate preferentially attacking iron or zinc depending on the corrosive medium, but finds different resistances due to the distribution of unordered phases interpenetrating.

Conclusions

The corrosion of GA steel generates three new surfaces with different chemical behaviour in a given environment.

In the acid and saline media, the outer phase richest in zinc showed the highest polarisation resistance among the surfaces revealed after GA steel dissolution. In acid and sodium chloride solutions, the layers obtained after coulometric stripping of steel showed one time constant with the charge transfer process occurring on the electrode surface.

In the alkaline solution, the inner layer richest in iron showed the highest corrosion resistance among the surfaces obtained after dissolution of GA steel, two time constants and an equivalent circuit characteristic of a corrosion of porous layer.

According to the Mossbauer spectroscopy results, the delta phase was identified on all surfaces obtained before and after coulometric dissolution of GA steel, and in a higher content than zeta phase on the outer and intermediate layers.

References

Bandyopadhyay

Jha

Singh

A. K.

Rout

T. K.

and Rani

: ‘Corrosion behaviour of galvannealed steel sheet’ Surf. Coat. Technol., 2006 200 4312–4319. doi: 10.1016/j.surfcoat.2005.02.153

Pistofidis

Vourlias

Konidaris

Pavlidou

Stergiou

and Stergioudis

: ‘Corrosion behaviour of galvannealed steel sheet’ Mater. Lett., 2006 60 786–789. doi: 10.1016/j.matlet.2005.10.013

Yadav

A. P.

Katayama

Noda

Masuda

Nishikata

and Tsuru

: ‘Effect of Fe-Zn alloy layer on the corrosion resistance of galvanized steel in chloride containing environments’ Corros. Sci., 2007 49 3716–3731. doi: 10.1016/j.corsci.2007.03.039

Queiroz

F. M.

and Costa

: ‘Electrochemical, chemical and morphological characterization of galvannealed steel coating, Surf’ Coat. Technol., 2007 201 7024–7035. doi: 10.1016/j.surfcoat.2007.01.005

Xhoffer

Dillen

and Cooman

B. C.

: ‘Quantitative phase analysis of galvannealed coatings by coulometric stripping’ J. Appl. Electrochem., 1999 29 209–219. doi: 10.1023/A:1003435824284

Okamoto

N. L.

Kashioka

Inomoto

Inui

Takebayashi

and Yamaguchi

: ‘Compression deformability of Γ and ζ; Fe–Zn intermetallics to mitigate detachment of brittle intermetallic coating of galvannealed steels’ Scr. Mater., 2013 69, (4), 307–310. doi: 10.1016/j.scriptamat.2013.05.003

Lee

K. K.

Lee

I. H.

Lee

C. R.

and Ahn

H. K.

: ‘In-situ observation in a scanning electron microscope on the exfoliation behavior of galvannealed Zn–Fe coating layers’ Surf. Coat. Technol., 2007 201, (14), 6261–6266. doi: 10.1016/j.surfcoat.2006.11.021

Chakraborty

Saha

and Ray

R. K.

: ‘Characterization of industrially produced galvannealed coating using cross-sectional specimen in TEM’ Mater. Charact., 2009 60 882–887. doi: 10.1016/j.matchar.2009.02.006

Cook

Tuszynski

and Townsend

: ‘Compression deformability of Γ and ζ; Fe–Zn intermetallics to mitigate detachment of brittle intermetallic coating of galvannealed steels’ Hyperfine Interact 1990 54, (1), 781–785. doi: 10.1007/BF02396128

10.

Hong

J. -H.

S. -J.

and Kwon

S. J.

: ‘Mossbauer analysis of the iron-zinc intermetallic phases’ Intermetallics 2003 11, (3), 207–213. doi: 10.1016/S0966-9795(02)00191-7

11.

Grant

and Cook

: ‘Mössbauer effect and XRD studies of iron-zinc binary alloys’ Hyperfine Interact 1994 94, (1), 2309–2315. doi: 10.1007/BF02063780

12.

Cook

: ‘In-situ identifications of iron-zinc intermetallics in galvannealed steel coatings and iron oxides on exposed steel’ Hyperfine Interact 1998 111, (1), 71–82. doi: 10.1023/A:1012629028199

13.

Verma

: ‘Mössbauer effect and XRD studies of iron-zinc binary alloys’ Pramana 1984 22, (1), 57–61. doi: 10.1007/BF02875587

14.

Graham

Beaubien

and Sproule

: ‘A Mossbauer study of Fe-Zn phases on galvanized steel’ J Mater. Sci., 1980 15, (3), 626–630. doi: 10.1007/BF00551726

15.

Paranhos

R. M. V.

Lins

V. F. C.

Macedo

W. A.

and Alvarenga

E. A.

: ‘Optimisation of electrochemical stripping of galvannealed interstitial free steels’ Surf. Eng., 2011 27 676–682. doi: 10.1179/1743294410Y.0000000015

16.

Yadav

A. P.

Nishikata

and Tsuru

: ‘Electrochemical impedance study on galvanized steel corrosion under cyclic wet-dry conditions - influence of time of wetness’ Corros. Sci., 2004 46 169–181. doi: 10.1016/S0010-938X(03)00130-6

17.

Graedel

T. E.

: ‘Corrosion mechanisms for zinc exposed to the atmosphere’ J. Electrochem. Soc., 1989 136 193C. doi: 10.1149/1.2096868

18.

Silva

P. S. G.

Costa

A. N. C.

Mattos

O. R.

Correia

A. N.

and Lima-Neto

: ‘Evaluation of the corrosion behavior of galvannealed steel in chloride aqueous solution and in tropical marine environment’ J. Appl. Electrochem., 2006 36 375–383. doi: 10.1007/s10800-005-9083-x

19.

Guttmann

Leprêtre

Aubry

Roch

M. J.

Moreau

Drillet

Mataigne

J. M.

and Baudin

: ‘The Galvatech Proceedings Series (GALVATECH'95)’ Chicago, USA 1995 295.

20.

Gabe

D. R.

: ‘Coulometric techniques for surface coatings’ Trans. Inst. Met. Finish., 1999 77 213–217.

21.

Chen

Zhu

Peng

Liu

Xuping

and Jianhua

: ‘Influence of phosphorus on the growth of Fe-Zn intermetallic compound in Zn/Fe diffusion couples’ Surf. Coat. Technol., 2014 240 63–69. doi: 10.1016/j.surfcoat.2013.12.013

22.

Nakano

Malakhov

D. V.

Yamaguchi

and Purdy

G. R.

: ‘A full thermodynamic optimization of the Zn–Fe–Al system within the 420-500 ∘C temperature range’ Calphad 2007 31, (1), 125–140. doi: 10.1016/j.calphad.2006.09.003

23.

Paranhos

R. M. V.

Lins

V. F. C.

Freitas

M. P.

Alvarenga

E. A.

and Macedo

W. A.

: ‘Caracterização eletroquímica de fases ferro-zinco em aço galvannealed’ Corrosão e Proteção 2011 37 20–22.

24.

McCafferty

: ‘Introduction to corrosion science’ 1st edn77–208; 2010

Berlin

Springer.

25.

Lee

H. H.

and Hiam

: ‘Corrosion resistance of galvannealed steel’ Corrosion 1989 45, (10), 852–856. doi: 10.5006/1.3584993

26.

Rout

T. K.

Banyopadhyay

Venugopalan

and Bhattacharjee

: ‘Corrosion behaviour of galvannealed steel sheet’ Corros. Sci., 2005 47, (11), 2841–2854. doi: 10.1016/j.corsci.2004.11.005

27.

Belkaid

Ladjouzi

M. A.

and Hamdani

: ‘Effect of biofilm on naval steel corrosion in natural seawater’ J. Solid State Electrochem., 2011 15 525–537. doi: 10.1007/s10008-010-1118-5

28.

Rout

T. K.

: ‘Electrochemical impedance spectroscopy study on multi-layered coated steel sheets’ Corros. Sci., 2007 49 794–817. doi: 10.1016/j.corsci.2006.06.008

29.

Zheng

Z. J.

Gao

Gui

and Zhu

: ‘Studying the fine microstructure of the passive film on nanocrystalline 304 stainless steel by EIS, XPS and AFM’ J. Solid State Electrochem., 2014 18 2201–2210. doi: 10.1007/s10008-014-2472-5