Electrochemical Behavior of Carbon Steel ASTM A36 in Diluted Pregnant Leach Solutions from Electrowinning of Copper

Abstract

In Northern Chile, large amounts of highly corrosive solutions are currently generated in the process of cathode washing after completing the electrowinning or electrorefining process of copper. This study investigates the electrochemical behavior of ASTM A36 carbon steel in pregnant-leach-solution (PLS) wash water. Measurements of electrochemical impedance spectroscopy and linear sweep voltammetry, complemented with weight loss measurements, were performed. Four ratios of PLS containing reverse osmosis (RO) water are evaluated, considering both quiescent and rotating conditions of the steel specimen. The results indicate that oxygen reduction, hydrogen evolution, and iron oxidation reactions are all involved during the corrosion of carbon steel in pure RO water, with the corrosion rate increasing up to 4 times under rotating conditions. In the case of corrosion in RO wash water containing PLS, a galvanic process occurs whereby copper is reduced at the expense of iron oxidation, superimposed on former partial reactions. The deposited copper induces notable corrosion inhibition of steel, observed as a significant drop in corrosion rate from high initial to constant residual values. Morphological and X-ray analyses support that corrosion is affected by oxide layer formation and galvanic copper deposition, confirming the results obtained from electrochemical analysis and weight loss measurements.

1. Introduction

Most of the copper is obtained electrolytically, after pyrometallurgical or hydrometallurgical copper ore processing.

In copper hydrometallurgy processing, copper in an aqueous form named as pregnant-leach-solution (PLS) is obtained from a leaching stage of oxide mineral using a sulfuric acid solution. In a further step, unwanted ions are removed to a large extent by a solvent extraction process following which copper is recovered as solid in a so-called copper electrowinning process. In this electrochemical process the main half-cell reactions are [1]:

$${2H_{2}O\to O}_2+4H^{+}+4e^{-}\hspace{1em}\hspace{1em}\hspace{1em}{E}^o=+1.23V_{SHE}$$

(1)

$${Cu}^{2+}+2e^{-}\to Cu\hspace{1em}\hspace{1em}\hspace{1em}{E}^o=+0.34V_{SHE}$$

(2)

Four main relevant aspects concerning this process are:

(a)

This cell requires an external power supply having a potential value larger than 0.89 V to function as expected under standard conditions. Typically, the cell potential, which includes the potentials determined from Nernst Equation, the anodic and cathodic overpotentials, and the potential drop caused by the solution resistance, is close to 2 V at 300 A/m².

(b)

The presence of some ions in the PLS solution like ${Fe}^{2+},{Cl}^{-}$ and others originate additional half-cells that compete with mainly cathodic reaction of copper causing both, energy losses and detrimental quality of the copper deposited in the cathode electrode. Minimizing the concentration of iron reduces the consumption of current by this reaction. Therefore, an efficient purification process to remove impurities from PLS solution is essential [2,3,4,5,6].

(c)

Proper anode and cathode material selection is essential to fulfill main requirements of lowest overpotential, good quality of the deposited copper in the cathode [7] and minimum electrode corrosion rate [8]. The conventional anode and cathode materials are lead alloys and 316L stainless steel, respectively [9,10]. Alternative anode materials have been already developed but its high comparative cost with lead is a constraint for widespread industrial use [11,12,13].

(d)

In order to meet quality requirements for commercialization the final copper cathodes sheets are finally rinsed with good quality water in order to remove remnants of electrolyte solution adhered on its surface [14]. In the Chilean mining copper facilities this is done in a rinsing booth either manually or automated by using reverse osmosis treated seawater [15]. The average amount of water consumption for this operation is 0.008 L of water/kg of produced $Cu$.

In the pyrometallurgy of copper the main products are impure copper anodes containing about 8% of impurities that are removed by electrorefining [16,17]. This process differs from an electrowinning cell in that the anode electrodes are replaced by impure copper anodes. Under this condition the main anodic half-cell is the dissolution of copper from the impure anode, according to:

$$Cu\to {Cu}^{2+}+2\overline{e}\hspace{1em}\hspace{1em}\hspace{1em}{E}^o=+0.34V_{SHE}$$

(3)

This half-cell will suppress the oxygen evolution reaction because of its significantly more positive potential than reaction (1).

To remove the impurities released from the anode and maintain an adequate chemical profile, the electrolyte in the electrorefining cell must be treated by adding specific chemicals and filtrated in a closed-circuit processing system [15].

Once the cathodic copper plates are ready, the post treatment to clean them is like that of electrowinning.

The residual water that is generated from electrorefining and electrowinning facilities comes not only from cathode washing but also from periodic wall and floor washing to remove acid films from wetted surfaces exposed to acid mist that is formed in the electrolytic cells [6,14]. The amount of residual water is estimated as 0.01 L of water/kg $Cu$. Normally, this water is recycled to either the leaching heaps and/or PLS tanks. The flow of this water through the processing facilities represents a corrosion risk for metal infrastructure that generally consists of carbon steel. Despite the convenience to assess the extent of this risk for corrosion prevention reasons, no reference was found concerning metal corrosion under these conditions.

In this work the corrosivity of the residual water generated in electrolytic processing of copper in carbon steel A36 were investigated assuming PLS dilution in RO water in a range between 0 and 0.02 L PLS/L RO water. The factors considered are the prevalent corrosion mechanisms, the influence of the hydrodynamic conditions, and the concentration of PLS in the residual water. A metal specimen consisting of a rotating cylinder is exposed to a sample of residual water while the corrosion evolution is monitored by weight loss and electrochemical measurements. Also, selected metal specimens are characterized by scanning electron microscopy (SEM) and X-ray diffraction (XRD). In considering that the electrolytic processing is the best technological option for copper production from the point of view of the environmental impact, the concept of green metallurgy in this context is the rational planning of this process to promote environmental protection awareness, by minimizing the impact in the ecological environment, resources, and energy consumption. The options are choosing efficient chemicals for solvent extraction, minimizing impurities that affect current efficiency in the electrolytic cell and minimizing acid mist generation, among others. Accordingly, this research that intends to assess corrosion infrastructure in contact with RO wash solutions is a contribution to green metallurgy.

2. Materials and Methods

2.1. Material and Test Solution

Carbon steel type ASTM A36 with a nominal chemical composition of (%wt.): 0.27 $C$, 1.03 $Mn$, 0.2 $Cu$, 98 $Fe$, 0.04 $P$, 0.28 $Si$, 0.05 $S$, were used in all experiments.

The test solution was a mixture of PLS and RO water with composition varying between 0 and 0.02 L PLS per 1 L of RO. The PLS was an industrial solution obtained from Lomas Bayas Copper Facilities in Northern Chile with composition as follows; copper: 42 g/L, sulphate: 150 g/L, iron: 0.01 g/L, chloride: 0.08 g/L, and other components lower than 0.01 g/L. The RO water was an artificial solution prepared from distilled water, dosed with measured amounts of sodium chloride, sodium sulphate, sodium carbonate, sodium borate, and magnesium chloride so as to obtain the following water composition; sodium: 480 mg/L, chloride: 630 mg/L, sulphate: 200 mg/L, boron: 2 mg/L, carbonate and bicarbonate: 30 mg/L, and magnesium: 50 mg/L. The final pH of the RO water was adjusted to 7.5 by adding a few drops of a 1 M sulfuric acid solution. The final measured conductivity of the RO water was 1800 µS/cm. All chemicals used were of analytical grade (Merck, Chile).

2.2. Electrochemical Measurements

The electrochemical measurements were performed using a conventional three-electrode cell mounted in a home-made rotating electrode accessory and connected to an SP1 Zive Potentiostat (ZiveLab, WonA Tech, Seoul, Korea) as is shown in Figure 1. The working electrode (WE) was a carbon steel type ASTM A36 specimen, the reference electrode (RE) an Ag/AgCl (KCl Sat.) half-cell, and the counter electrode (CE) a graphite rod.

The WE specimens were manufactured from a carbon steel cylinder with a diameter of 12 mm and a length of 15 mm. This electrode was concentrically inserted in the rotating shaft using a ¼” threaded extension sealed with an O-ring to keep the threaded section in dried condition while the exposed surface of the cylinder is immersed in the test solution. Thus, the total exposed area was 7.25 × 10⁻⁴ m². The counter electrode was a 10 mm diameter graphite cylinder positioned at the vertical central axis form the bottom of the cell reactor. Air gas was bubbled continuously during the experiments to maintain a uniform oxygen concentration in solution. All potentials are referred to the standard hydrogen electrode (SHE). Before each measurement, the WE were successfully polished with SiC paper from 800 to 1200 grit, cleaned in an ultrasonic bath with distilled water then washed with isopropyl alcohol and acetone, dried in air, and finally weighed in an analytical scale. One aspect of practical interest to favor reproducibility is the abraded method that was made in a spindle machine under a constant speed.

Electrochemical measurements of linear sweep voltammetry (LSV) and electrochemical impedance spectroscopy (EIS) for the rotating cylinder electrode immersed in the mixed PLS-RO solutions were sequentially performed many times along extended periods up to 70 h at room temperature of 22 ± 0.5 °C. The LSV scan range was in anodic direction from −1000 to 0 mV/SHE at a rate of 2 mV/s. The EIS measurements were conducted using an AC signal with a signal amplitude perturbation of 10 mV over a frequency range of 100 kHz to 10 mHz.

Using this WE design, LSV and EIS measurements were made at different immersion times, combined with a global corrosion rate determination by the weight-loss method at the end of the run. For that purpose, the corroded WE were cleaned according to ASTM G1-03(2017) [18] and weighed on an analytical scale.

The electrochemical behaviors of carbon steel from LSV measurements were analyzed assuming that the measured total current density can be linearly decomposed in terms of partial electrochemical reactions of hydrogen evolution (HE), dissolved oxygen reduction reaction (ORR) and iron oxidation (IO), according to the follow expressions:

$$i=i_{{\mathrm{O}}_2}+i_{{\mathrm{H}}_2}+i_{\mathrm{F}\mathrm{e}}$$

(4)

$$i_{O_2}=\frac{P}{2·i_{l,O_2}}·\left(\sqrt{P^2+4·i_{l,O_2}^2}-P\right)$$

(5)

$$P=a_{O_2}·e^{\left(-\frac{2.303·E}{t_{O_2}}\right)}$$

(6)

$$i_{H_2}=a_{H_2}·e^{\left(-\frac{2.303·E}{t_{H_2}}\right)}$$

(7)

$$i_{Fe}=a_{Fe}·e^{\left(\frac{2.303·E}{t_{Fe}}\right)}$$

(8)

where $i$ is the total current density, $i_{O_2}$ and $i_{H_2}$ are the partial reduction current densities for ORR and HE respectively, $i_{Fe}$ is the partial oxidation current density for IO, $a_{{\mathrm{O}}_2}$, $a_{{\mathrm{H}}_2}$ and $a_{\mathrm{F}\mathrm{e}}$ are the charge transfer reaction coefficient for ORR, HE and IO reactions respectively, which are directly related to the exchange current densities, $i_{l,O_2}$ is the limiting current density for ORR, $t_{{\mathrm{H}}_2},t_{{\mathrm{O}}_2}$ and$t_{\mathrm{F}\mathrm{e}}$ are the Tafel slopes for HE, ORR and IO, and $P$ is a simplification factor in Equation (5), which corresponds to the mathematical model for a charge transfer mechanism as is given in Equation (6). The kinetic expressions developed in terms of the potential mixed theory and its manipulation are described elsewhere [19,20]. Additionally, the corrosion current density ($i_{corr}$) was determined from the anodic and cathodic partial current densities at the corrosion potential ($E_{corr}$) under an $i=0$ A/m² condition.

The EIS measurements were analyzed by the experimental data fitting to an appropriate equivalent electrical circuit to obtain electrochemical parameters at the interface of the carbon steel.

2.3. Morphological and Phase Analysis

A Zeiss EVO MA 10 scanning electron microscope (SEM) (Zeiss, Oberkochen, Germany) equipped with an energy-dispersive X-ray (EDS) analyzer was used for morphological characterizations. A Shimadzu XRD-600 diffractometer (Shimadzu Corp., Kyoto, Japan) using Cu Kα radiation at an angular step of 0.02° (2θ) and a counting time per step of 4 s was used for the determination of the main corrosion products.

2.4. Reproducibility and Data Handling Considerations

In this research, selected runs were replicated, for example, for pure RO under quiescent and at 1200 rpm conditions. The main parameters such as $i_{l,O_2}$, $i_{corr}$, and $E_{corr}$ exhibited very good reproducibility in a margin of 30% uncertainty which is considered very good. For example, $i_{l,O_2}$ have values in a range of 0.5–0.7 A/m² and 3–5 A/m² for quiescent and 1200 rpm, respectively, were obtained. Unfortunately, a few of the model parameters produced unreasonable values in that they do show chaotic tendencies; in our point of view the reason lies in the nonlinearity of the models where in a few cases, dissimilar parameters can give the same results. Curiously, despite these difficulties most of the results give clear tendencies that are consistent between different examination methods.

From a practical point of view, three different actions were considered to favor reproducibility, (a) all solutions were taken from a single preparation stocks, (b) the abrading process for carbon steel were made in a spindle machine under a constant speed, and (c) the first LSV and EIS measurements for all runs were started after 2 min of WE immersion in the electrolyte.

3. Results and Discussion

3.1. Steel Corrosion in Pure RO Water

The corrosion behavior of carbon steel A36 was characterized in terms of a complete set of electrochemical parameters extracted from LSV and EIS measurements for carbon steel in artificial RO water without rotation and rotating at 1200 rpm. The polarization curves obtained from LSV are shown in Figure 2. For the results obtained under null rotation rate, the polarization curves reveal a shape shifting in the anodic branch toward more positive potentials as the immersion time increase. In contrast, the cathodic branch plateau corresponding to $\left|i_{{l,O}_2}\right|$ remains almost unchanged with the immersion time between 0.5 to 0.7 A/m². On the other hand, for the results at a rotation rate of 1200 rpm, $\left|i_{{l,O}_2}\right|$ shows significantly higher values ranging between 3 to 5 A/m²; because of this, significantly higher $i_{corr}$ values in comparison with no rotation situation will be obtained [21]. The cathodic branch reveals a continuous decrease of the current density as the immersion time increases. This condition was observed up to 32 h of immersion, after that the cathodic current density increases. Contrary behavior was observed for the anodic branch, which is characterized by an increase in the current density as the immersion time increase, reaching a maximum value at 25 h, followed by a decrease in its current density. This variation in the cathodic branch, is attributed to the formation of a rust layer that inhibits the dissolved oxygen mass transfer from the bulk to the metallic surface. In fact, this rust layer that gradually grows with immersion time should be detached under the wall shear stress action generated by WE rotation when a critical thickness is reached.

In addition, the influence of the hydrodynamic condition on the corrosion behavior is clearly observed as an increase in the cathodic current density with the rotation rate, which is associated to the mass transfer of dissolved oxygen from bulk to the metallic surface [22]. The corrosion potential shift to anodic direction and the significant increase of $i_{corr}$ are other consequences of the rotation rate influence.

The electrochemical behaviors were examined over time in terms of the kinetic for the ORR, HE, and IO reactions, and corrosion parameters, which were determined by a fitting of the set of Equations (4)–(8), and tabulated in

Table 1. Parameters for carbon steel immersed in RO water from LSV measurements.

Time (h)	${\mathit{a}}_{{\mathit{O}}_2}$ (A/m2)	${\mathit{a}}_{{\mathit{H}}_2}$ (A/m2)	${\mathit{a}}_{\mathit{F}\mathit{e}}$ (A/m2)	${\mathit{i}}_{{\mathit{l},\mathit{O}}_2}$ (A/m2)	${\mathit{t}}_{{\mathit{O}}_2}$ (mV/dec)	${\mathit{t}}_{{\mathit{H}}_2}$ (mV/dec)	${\mathit{t}}_{\mathit{F}\mathit{e}}$ (mV/dec)	${\mathit{E}}_{\mathit{c}\mathit{o}\mathit{r}\mathit{r}}$ (mV/SHE)	${\mathit{i}}_{\mathit{c}\mathit{o}\mathit{r}\mathit{r}}$ (A/m2)
	0 rpm
0	2.20 × 10−3	−1.04 × 10−4	50.39	−0.55	−121	−200	132	−287	0.33
2	2.81 × 10−2	−7.61 × 10−7	1305.0	−0.68	−334	−116	105	−382	0.30
16	2.30 × 10−3	−2.26 × 10−5	22.79	−0.59	−145	−171	188	−342	0.35
24	5.20 × 10−3	−2.27 × 10−5	22.95	−0.54	−162	−167	195	−346	0.39
44	8.00 × 10−3	−6.35 × 10−4	15.67	−0.54	−182	−239	190	−321	0.32
49	3.40 × 10−3	−5.92 × 10−4	17.40	−0.49	−158	−254	189	−334	0.30
	1200 rpm
0	0.38	−6.51 × 10−7	48.90	−4.95	−404	−142	109	−185	0.97
2	0.19	−2.29 × 10−4	71.49	−3.97	−340	−215	140	−261	0.98
8	0.22	−4.41 × 10−4	51.87	−2.95	−379	−230	206	−333	1.26
25	0.23	−7.79 × 10−4	54.63	−3.88	−361	−240	214	−332	1.52
32	0.31	−2.30 × 10−3	20.27	−2.62	−434	−276	284	−336	1.33
49	0.37	−7.90 × 10−3	21.04	−4.62	−444	−286	253	−292	1.47

.

All the parameters listed in Table 1 for the initial time, where the metal surface is in a bare condition that is free from surface layers after the abrading process, are in the order of magnitude of published data for carbon steel in saline solutions [19,23,24,25]. In general, from the bare condition, the oxide layer thickness on carbon steel increases with immersion time, and in this context some parameters show values and tendencies described as, (i) A very high $a_{Fe}$ value observed for 2 h under quiescent condition which is visually appreciated as a highly anodic slope for the 2 h curve in Figure 2. (ii) An increasing value of $t_{Fe}$ with immersion time that is more pronounced for carbon steel under rotation. Referring to Equation (8), this is corroborated by the lower slopes observed at $E_{corr}$ values in the polarization curves at immersion times higher than 24 h (marked as red color in Figure 2). (iii) The quiescent $\left|i_{l,O_2}\right|$ values are significantly lower that non quiescent $\left|i_{l,O_2}\right|$. This indicates that agitation promotes oxygen diffusion to the carbon steel surface thereby increasing the corrosion rate. (iv) An increasing trend of the $\left|a_{H_2}\right|$ values indicating that under a constant $t_{H_2}$ value, the partial $i_{{\mathrm{H}}_2}$ values are shifted to positive potentials with increasing immersion times. This tendency is more pronounced for the agitation condition. All these features are linked to the iron oxide precipitation on the metal surface that alters the surface morphology and the hydrodynamic condition.

Furthermore, the corrosion of carbon steel evaluation performed by EIS measurements is shown in Figure 3. The analyses of these curves were done by fitting the data to an equivalent circuit as is shown in the inset of Figure 3a,d, composed by a solution resistance $R_s$, resistance and capacitance for the $R_1$ and $Q_1$ respectively corresponding to the double layer, and resistance and capacitance for the $R_2$ and $Q_2$ respectively corresponding to the corrosion products layer [26,27,28]. The kinetic parameters obtained from fitting are tabulated in

Table 2. EIS parameters for an equivalent circuit shown in Figure 3.

Time (h)	${\mathit{R}}_{\mathit{s}}$ (Ohm)	${\mathit{R}}_1$ (Ohm)	${\mathit{Q}}_1$ (sα/Ohm)	${\mathit{\alpha }}_1$	${\mathit{R}}_2$ (Ohm)	${\mathit{Q}}_2$ (sα/Ohm)	${\mathit{\alpha }}_2$
0	6.37 × 10−3	76.47	4.24 × 10−3	0.74	22.47	9.90 × 10−9	1.0
2	9.57 × 10−4	169.83	7.49 × 10−3	0.70	20.34	1.01 × 10−8	1.0
16	7.40 × 10−3	254.18	7.46 × 10−3	0.62	20.98	4.96 × 10−9	1.0
24	1.40 × 10−2	232.64	7.52 × 10−3	0.64	23.94	8.85 × 10−9	1.0
44	1.40 × 10−1	195.44	7.58 × 10−3	0.59	27.33	7.26 × 10−9	1.0
49	5.38 × 10−2	197.13	7.21 × 10−3	0.61	19.98	5.28 × 10−9	1.0
0	1 × 10−10	28.95	5.28 × 10−2	0.73	17.36	1.43 × 10−8	0.97
2	1 × 10−10	31.12	4.88 × 10−2	0.70	20.29	1.02 × 10−8	1
8	2.2 × 10−13	39.72	4.65 × 10−2	0.65	30.48	8.81 × 10−9	1
25	1.48 × 10−10	29.97	2.70 × 10−2	0.61	24.05	8.75 × 10−9	0.99
49	9.54 × 10−10	23.33	2.23 × 10−2	0.60	39.22	4.42 × 10−9	1

.

Significant differences are noted between EIS measurements at quiescent and rotating conditions (Figure 3). These are: lower noisy reading observed for no rotation at some immersion times (Figure 3a–c), two arcs for rotation (Figure 3d) as compared with one arc for no rotation (Figure 3a), and significantly lower impedance variation for no rotation (Figure 3c). A correlation between the real component of the impedance and $i_{corr}$ will be used as a validation criterion of $i_{corr}$ obtained from LSV measurements.

The $R_s$ values for quiescent condition, which represents the solution resistance, changes between 20 and 28 ohms except for the almost null value at the initial immersion time (Table 2). The EIS measurement at initial time is subjected to the most unstable condition since a rapid oxide layer forms over the bare metal surface. In contrast, $R_s$ values for rotation presents almost null values, which indicates that ohmic resistance decreases with rotation [29].

Concerning the parallel elements $R_1/Q_1$ and $R_2/Q_2$ of the equivalent circuit (Figure 3) for no rotation when $Q_1$ and $Q_2$ exhibit significantly different values, such as at 24 and 49 h immersion time (Table 2), then the equivalent circuit $R_s-(R/Q)$ fits very well to the EIS data. This indicates that a second element $R/Q$ for one Nyquist shape arc improves the curve fitting obtained using the proposed model. For a rotation case that exhibits two Nyquist shape arcs, the two elements $R_1/Q_1$ and $R_2/Q_2$ (or some similar model) are necessary.

As a means of validation of the parameters shown in Table 1 and Table 2, the $i_{corr}$ values are recalculated using EIS results under the following considerations:

(1)

It is assumed that the open circuit condition (OCP) under which the EIS measurements were done, is nearly equivalent to the corrosion potential and also the controlling corrosion mechanism is a charge transfer process. For this condition, the Stern–Geary equation applies [30]:

$$i_{corr}=\frac{{\beta }_a\times {\beta }_c}{2.303\times R_{ct}({\beta }_a+{\beta }_c)}$$

(9)

where, ${\beta }_a$ and ${\beta }_c$ are the anodic and cathodic slopes which are equivalent to $t_{Fe}$ and $t_{O2}$ shown in Table 1, respectively.

$R_{ct}$ values have been determined using EIS data from some equivalent circuits [31].

(2)

The mathematic model for the equivalent circuit can be expressed by:

$$Z_{Rs-R1/Q1-R2/Q2}=R_s+\frac{R_1}{1+R_1{Q}_1{\left(wj\right)}^{{\alpha }_1}}+\frac{R_2}{1+R_2{Q}_2{\left(wj\right)}^{{\alpha }_2}}$$

(10)

For $R_{ct}$ determination, a criterion that is valid for this case is [32]:

$$R_{ct}={\left|Z\left(jw\right)\right|}_{w\to 0}-{\left|Z\left(jw\right)\right|}_{w\to \infty }$$

(11)

Applying Equation (11) to Equation (10) gives:

$$R_{ct}=R_1+R_2$$

(12)

Finally, the relationship to calculate $i_{corr}$ is:

$$i_{corr}=\frac{t_{Fe\times }{t}_{O2}}{2.303\times \left(t_{Fe}+t_{O2}\right)\times \left(R_1+R_2\right)\times A}$$

(13)

where, $A$ is the active area of the working electrode in m².

The corrosion rate determined from Equation (13) is tabulated in

Table 3. Instantaneous $i_{corr}$ values calculated from LSV and EIS data and global $i_{corr}$ from weight loss for pure RO water.

	From LSV			From EIS	From Equation (13)	From Weight Loss
Time (h)	${\mathit{t}}_{{\mathit{O}}_2}$ (mV/dec)	${\mathit{t}}_{\mathit{F}\mathit{e}}$ (mV/dec)	${\mathit{i}}_{\mathit{c}\mathit{o}\mathit{r}\mathit{r}}$ (A/m2)	${\mathit{R}}_{\mathit{c}\mathit{t}}$ (Ohm)	${\mathit{i}}_{\mathit{c}\mathit{o}\mathit{r}\mathit{r}}$ (A/m2)	${\mathit{i}}_{\mathit{c}\mathit{o}\mathit{r}\mathit{r}}$ (A/m2)
0 rpm
0	−121	132	0.33	98.94	0.42	0.23
2	−334	105	0.30	190.17	0.26
16	−145	188	0.35	275.16	0.21
24	−162	195	0.39	256.58	0.25
44	−182	190	0.32	222.77	0.28
49	−158	189	0.30	217.10	0.27
1200 rpm
0	−404	109	0.97	46.31	1.23	1.15
2	−340	140	0.98	51.40	1.28
8	−379	206	1.26	70.20	1.24
25	−361	214	1.52	54.01	1.65
32	−434	284	1.33	-	-
49	−444	253	1.47	62.55	1.46

. It is interesting to note that predicted values for instantaneous $i_{corr}$ from polarization curves and Equation (13) are in good agreement in having both, similar comparative values and the negligible change with immersion time. Also, the notable influence of rotation in corrosion is evident from average $i_{corr}$ values under rotation that are up to six times higher for those corresponding under quiescent corrosion [33]. Some small differences between $i_{corr}$ values determined from LSV, EIS and Equation (13) can be attributed to either experimental errors or variational uncertainties in numerical data manipulation. Furthermore, the reliable average $i_{corr}$ value determined by weight loss (applying Faraday’s law) indicates about a 10% overvalued prediction.

3.2. Steel Corrosion in RO Wash Water

LSV and EIS measurements were performed in three wash RO samples characterized in

Table 4. Chemical characterization water samples.

Dilution (L PLS/L RO)	${\mathit{C}\mathit{u}}^{+2}$ (g/L)	${\mathit{H}}_2\mathit{S}{\mathit{O}}_4$ (g/L)	${\mathit{C}\mathit{l}}^{-}$ (g/L)	pH	Conductivity (mS/cm)
Pure RO	0	0.2	0.62	7.5	2.33
PLS	44	150	0.03	-	-
0.005	0.22	0.95	0.62	3.51	2.99
0.01	0.44	1.71	0.62	3.13	3.18
0.02	0.88	3.22	0.62	2.70	4.01

.

For the conditions indicated in Table 4, LSV and EIS measurements were carried out to know the corrosion behavior of carbon steel. Representative results for LSV and EIS measurements are shown in Figure 4 and Figure 5, respectively, with and without rotation of the working electrode.

The electrochemical parameters for carbon steel determined from LSV and EIS measurements are shown in

Table 5. Electrochemical parameters calculated from LSV measurements for carbon steel in RO wash water at different dilutions and agitation.

Dilution (L PLS/L RO)	Time (h)	${\mathit{a}}_{{\mathit{O}}_2}$ (A/m2)	${\mathit{a}}_{{\mathit{H}}_2}$ (A/m2)	${\mathit{a}}_{\mathit{F}\mathit{e}}$ (A/m2)	${\mathit{i}}_{{\mathit{l},\mathit{O}}_2}$ (A/m2)	${\mathit{t}}_{{\mathit{O}}_2}$ (mV/dec)	${\mathit{t}}_{{\mathit{H}}_2}$ (mV/dec)	${\mathit{t}}_{\mathit{F}\mathit{e}}$ (mV/dec)	${\mathit{E}}_{\mathit{c}\mathit{o}\mathit{r}\mathit{r}}$ (mV/SHE)	${\mathit{i}}_{\mathit{c}\mathit{o}\mathit{r}\mathit{r}}$ (A/m2)
	0 rpm
0.005	0	1.02 × 10−1	−5.32 × 10−5	943.50	−0.91	−467	−198	137	−441	0.57
	7	6.15 × 10−2	−3.37 × 10−7	133.96	−0.70	−344	−128	128	−335	0.39
0.01	0	1.05 × 10−1	−3.40 × 10−5	253.73	−2.97	−314	−208	157	−365	1.19
	7	2.16 × 10−1	−4.60 × 10−3	306.36	−2.84	–464	−504	146	−357	1.10
	22	5.87 × 10−8	−2.19 × 10−2	5095.60	−0.20	−3	−413	101	−436	0.25
	31	5.06 × 10−8	−5.30 × 10−3	2521.50	−0.44	−64	−336	119	−450	0.41
	45	1.18 × 10−6	−3.30 × 10−3	524.62	−0.70	−92	−305	134	−456	0.20
	73	8.52 × 10−5	−3.60 × 10−3	240.63	−0.69	−133	−309	155	−452	0.29
0.02	0	7.74 × 10−2	−9.70 × 10−3	107.65	−3.92	−278	−237	214	−393	1.56
	2	5.70 × 10−2	−5.69 × 10−5	208.20	−2.73	−302	−208	165	−388	0.91
	5	3.28 × 10−2	−5.15 × 10−9	977.70	−2.80	−264	−109	122	−380	0.77
	9	3.06 × 10−2	−9.22 × 10−9	1017.10	−2.60	−295	−112	123	−397	0.60
	22	9.09 × 10−6	−5.66 × 10−7	503.49	−0.64	−125	−122	142	−515	0.12
	26	2.36 × 10−5	−1.94 × 10−5	65.97	−0.35	−155	−165	198	−548	0.11
	29	7.39 × 10−6	−5.89 × 10−5	777.26	−0.27	−142	−185	138	−540	0.09
	44	7.12 × 10−24	−9.56 × 10−5	166.23	−0.27	−23	−174	176	−509	0.21
	54	3.35 × 10−17	−9.68 × 10−4	111.59	−0.26	−33	−236	195	−513	0.26
	68	3.39 × 10−11	−2.30 × 10−3	13.42	−0.25	−77	−285	360	−597	0.29
	1200 rpm
0.005	0	5.62 × 10−1	−2.90 × 10−3	10.75	−4.66	−164	−249	465	−203	3.93
	2	1.38 × 10−2	−4.50 × 10−3	12.54	−2.21	−219	−279	260	−350	0.57
	4	8.82 × 10−2	−1.40 × 10−3	15.36	−1.63	−322	−249	355	−410	1.07
	16	4.11 × 10−2	−8.90 × 10−3	19.41	−1.15	−363	−327	344	−472	0.82
	20	1.52 × 10−2	−9.80 × 10−3	20.10	−1.08	−307	−333	330	−480	0.70
	26	8.60 × 10−3	−1.06 × 10−2	16.45	−0.99	−286	−337	344	−486	0.61
	46	1.39 × 10−2	−1.81 × 10−2	15.26	−0.92	−357	−376	344	−481	0.61
0.01	0	1.57 × 10−1	−1.30 × 10−3	31.00	−9.42	−233	−1732	670	−373	8.62
	2	1.13 × 10−1	−6.26 × 10−2	54.69	−4.60	−310	−414	208	−326	1.49
	5	6.84 × 10−2	−1.04 × 10−1	32.99	−3.81	−291	−497	275	−362	1.58
	9	8.53 × 10−2	−4.33 × 10−2	73.43	−10.52	−256	−450	190	−316	1.59
	25	5.43 × 10−2	−1.22 × 10−2	7.13	−1.16	−224	−343	390	−336	0.98
	30	1.00 × 10−2	−2.03 × 10−2	0.01	−0.83	−152	−367	398	−377	0.99
	46	3.50 × 10−3	−3.76 × 10−2	6.51	−0.57	−131	−414	390	−356	0.80
0.02	0	2.37 × 10−1	−3.93 × 10−1	26.39	−8.07	−202	−496	481	−373	8.62
	1	3.26	−2.51	19.85	−14.08	−536	−1134	1384	−291	12.22
	5	1.18 × 10−1	−8.61 × 10−2	39.59	−3.03	−348	−513	245	−354	1.43
	7	1.71 × 10−1	−5.04 × 10−2	17.90	−2.12	−409	−458	333	−374	1.35

and

Table 6. Parameters from EIS measurements for carbon steel in RO wash water at different dilutions and agitation (equivalent circuit shown in Figure 3).

Dilution (L PLS/L RO)	Time (h)	${\mathit{R}}_{\mathit{s}}$ (Ohm)	${\mathit{R}}_1$ (Ohm)	${\mathit{Q}}_1$ (sα/Ohm)	α1	${\mathit{R}}_2$ (Ohm)	${\mathit{Q}}_2$ (sα/Ohm)	α2	${\mathit{R}}_{\mathit{c}\mathit{t}}$ (Ohm)
	0 rpm
0.005	0	1 × 10−10	22.81	5.42 × 10−7	0.65	131.29	9.8 × 10−3	0.74	154.09
	7	15.66	7.71	4.65 × 10−8	1.07	194.63	3.2 × 10−3	0.78	202.33
0.01	0	2.22 × 10−16	33.94	3.54 × 10−3	0.71	22.04	1.1 × 10−8	0.96	55.98
	7	1 × 10−10	39.40	1.29 × 10−3	1.08	18.38	7.2 × 10−8	0.88	57.79
	22	1 × 10−10	42.29	9.98 × 10−9	0.97	58.72	6.2 × 10−3	0.84	101.01
	31	7.47 × 10−9	32.75	1.84 × 10−8	0.93	79.26	8.2 × 10−3	0.85	112.01
	45	2.28	103.29	1.04 × 10−2	0.72	38.18	2.6 × 10−7	0.90	141.47
	73	1 × 10−10	110.20	1.03 × 10−2	0.78	46.80	6.5 × 10−9	0.97	156.99
0.02	0	1 × 10−10	27.40	2.46 × 10−3	0.84	17.54	4.3 × 10−9	1.03	44.94
	2	1 × 10−10	44.56	2.50 × 10−3	0.90	18.17	7.9 × 10−9	1.01	62.73
	5	1 × 10−10	35.01	3.89 × 10−3	0.91	17.47	5.4 × 10−9	1.03	52.49
	9	1 × 10−10	36.45	5.37 × 10−3	0.88	18.34	3.2 × 10−9	1.03	54.79
	22	1 × 10−10	29.82	8.82 × 10−3	0.88	28.20	6.4 × 10−8	0.86	58.02
	26	2.22 × 10−16	46.22	9.62 × 10−3	0.87	28.80	5.8 × 10−8	0.87	75.02
	29	2.22 × 10−16	52.69	1.47 × 10−7	0.79	67.49	1.0 × 10−2	0.82	120.18
	44	2.22 × 10−16	78.75	9.98 × 10−3	0.79	36.22	1.7 × 10−8	0.90	114.97
	54	1.22 × 10−10	74.68	9.43 × 10−3	0.81	33.56	3.1 × 10−9	0.89	108.24
	68	1 × 10−10	30.69	1.97 × 10−8	0.90	176.80	1.3 × 10−2	0.69	207.49
	1200 rpm
0.005	0	1.77 × 10−9	22.53	3.20 × 10−8	0.92	16.46	1.6 × 10−1	0.68	38.99
	2	1.77 × 10−9	23.43	5.54 × 10−8	0.88	35.57	1.4 × 10−1	0.58	59.00
	4	2.22 × 10−16	55.67	1.12 × 10−1	0.52	22.80	4.8 × 10−6	0.49	78.47
	16	12.00	126.63	7.58 × 10−2	0.40	10.70	6.6 × 10−8	1.00	137.33
	20	11.93	11.18	7.97 × 10−8	0.98	100.78	7.2 × 10−2	0.41	111.96
	26	8.59	14.88	5.61 × 10−8	0.96	94.91	6.3 × 10−2	0.42	109.79
	46	1 × 10−10	30.83	9.82 × 10−8	0.80	24.55	2.3 × 10−2	0.56	55.37
0.01	0	1.30 × 10−10	0.10	5.04 × 10−2	0.85	19.46	1.1 × 10−7	0.84	19.56
	2	4.14	9.64	1.09 × 10−1	1.03	16.28	2.9 × 10−9	1.17	25.92
	5	4.21 × 10−7	24.35	1.06 × 10−1	0.47	20.43	3.3 × 10−9	1.13	44.78
	9	1 × 10−10	51.12	2.65 × 10−3	0.89	14.45	1.9 × 10−8	0.97	65.57
	25	1 × 10−10	33.95	4.92 × 10−7	0.74	63.60	3.4 × 10−2	0.17	97.55
	30	2.22 × 10−16	49.94	9.32 × 10−5	0.40	104.89	1.1 × 10−2	0.56	154.83
	46	1 × 10−10	43.31	4.78 × 10−6	0.64	128.31	1.2 × 10−2	0.41	171.62
0.02	0	1 × 10−10	3.15	3.26 × 10−3	1.04	25.94	8.8 × 10−9	0.98	29.09
	1	1.62 × 10−8	29.21	1.83 × 10−9	1.10	3.75	7.9 × 10−3	0.96	32.96
	5	1 × 10−10	34.20	2.53 × 10−8	0.93	30.76	1.7 × 10−1	0.59	64.96
	7	1 × 10−10	36.21	2.87 × 10−9	1.08	38.75	1.1 × 10−1	0.55	74.95

, respectively.

The first observation from LSV results for carbon steel in wash RO water is a very high initial cathodic activity in comparison with that of pure RO water (Figure 2 and Figure 4). This activity is mainly evidenced at the initial time by a large absolute $i_{l,O_2}$ and $i_{corr}$ values that gradually decrease with immersion time. For instance, while the initial $i_{l,O_2}$ values for wash water without rotation exceeded by a factor of 2 to that of pure RO water, the initial $i_{corr}$ values under rotation exceeds by a factor of 5 to that of pure RO water (Table 1 and Table 5). Another relevant practical observation during a run with wash water under quiescent conditions, is the apparent bubbles emerging from the surface and the copper-like color that intensifies with immersion time (Figure 5). Bubbles under rotation are not observed probably because they are swept away from the shear forces exerted from rotation.

All this evidence suggests the influence of pH and the occurrence of additional partial reaction(s) that interfere with the HE, ORR and IO reactions during the corrosion of iron in RO wash water. In this respect, these factors can be characterized as: (a) The influence of pH, a simple calculation based on the Nernst equation for HE reaction indicates a 230 mV shift for $E_{H_2}^{eq}$ to more negative potentials with respect to RO pure water at neutral condition. Furthermore, it is well known that as the pH decreases, the HER reaction becomes significantly more kinetically easier and requires lower overpotential [34]. All this evidence is compatible with the significant larger $a_{H_2}$ values observed in some cases for corrosion of carbon steel in wash water in comparison with pure RO water. The exception are many values for corrosion in wash water 0.02 under no rotation; in these cases, the significantly lower absolute values of $t_{H_2}$ induce a faster increase in $i_{H_2}$ current (see Equation (7)). With respect to oxygen reduction on carbon steel in aerated solutions its predominance over the HE reaction decreases with a decrease of pH [35]. (b) The influence of Cu⁺² ions, the wash water that contains a Cu⁺² concentration in the range of 0.2–0.8 g/L in contact with the carbon steel surface can undergo a reduction at the expense of iron oxidation due to different standard reduction potentials [36], as is indicated in Equations (14) and (15).

$${Fe}^{2+}+2e=Fe\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}{E}^o=-0.44V_{SHE}$$

(14)

$${Cu}^{2+}+2\overline{e}\to Cu\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}{E}^o=+0.34V_{SHE}$$

(15)

In this context copper ions become readily reduced on the metal iron surface, while an equivalent amount of iron is dissolved. Evidence indicates that the copper recovery by cementation process is diffusion controlled [37].

Thus, in a global prospective for the carbon steel corrosion under RO wash water, the measured current density can be linearly decomposed in terms of partial electrochemical reactions of hydrogen evolution (HE), dissolved oxygen reduction (ORR), iron oxidation (IO) and copper reduction (CR). Considering that ORR and CR are similar in nature as both include a diffusion influence, then the corrosion kinetic model used for pure RO water, can still apply. Then the implicit assumption is that the partial currents for ORR and CR can be represented in a unique kinetic expression. It follows then that $i_{l,O_2}$ involves oxygen and copper reduction. Furthermore, at low pH, ORR is expected to be attenuated [35].

The presence of metallic copper on the WE and hydrogen evolution has been visually confirmed in experiments with wash water, as seen in Figure 6. All this evidence, together with the drastic decrease in $i_{corr}$ with immersion time, suggests an initial corrosion controlled by cementation that is further inhibited by the deposited copper film attached to the iron steel surface. The simultaneous decrease of $\left|i_{l,O_2}\right|$ which is calculated based on total surface area is also a consequence of this, because of the drastic reduction of the available surface for copper diffusion. The adherence of the deposited copper film was rather unstable since occasional detachment of small film sections was observed. When this happened during a potential sweep, this detachment was accompanied by a sudden increase in reduction current. Such an event was marked in red circles in the polarization curve shown in Figure 4. A consequence of this unstable superficial condition is manifested in the LSV and EIS measurements. In general, the fitting of the predicted polarization curves determined from measured LSV data for PLS wash water are of lower quality than that for RO water, and some apparently abnormal values of the electrochemical parameters are observed. Also, Nyquist plots measured at different immersion times shows irregular circular progression with time (Figure 5). Regardless, the equivalent circuit used for pure RO remains valid for RO wash water.

Unlike pure RO the EIS measurements for quiescent and rotation for wash RO are similar (Figure 5) exhibiting 2 arcs Nyquist shape. The difference between them is just a much faster time evolution toward an apparent constant Nyquist shape for rotation condition. The $R_s$ values for all cases are very low such that they can be considered null, except for a few cases that do not surpass 16 ohms (Table 6).

Table 7. Instantaneous $i_{corr}$ values calculated from LSV and EIS data and global $i_{corr}$ from weight loss for RO wash water.

		From LSV			From EIS	From Equation (13)	From Weight Loss
Dilution	Time (h)	${\mathit{t}}_{{\mathit{O}}_2}$ (mV/dec)	${\mathit{t}}_{\mathit{F}\mathit{e}}$ (mV/dec)	${\mathit{i}}_{\mathit{c}\mathit{o}\mathit{r}\mathit{r}}$ (A/m2)	${\mathit{R}}_{\mathit{c}\mathit{t}}$ (Ohm)	${\mathit{i}}_{\mathit{c}\mathit{o}\mathit{r}\mathit{r}}$ (A/m2)	${\mathit{i}}_{\mathit{c}\mathit{o}\mathit{r}\mathit{r}}$ (A/m2)
(L PLS/L RO)	0 rpm
0.005	0	−467	137	0.57	154.09	0.42	0.3
	7	−344	132	0.39	202.33	0.29
0.01	0	−314	157	1.19	55.98	1.21	0.61
	7	−464	146	1.10	57.79	1.24
	22	−3	101	0.25	101.01	0.02
	31	−64	119	0.41	112.01	0.24
	45	−92	134	0.20	141.47	0.25
	73	−133	155	0.29	156.99	0.30
0.02	0	−278	214	1.56	44.94	1.68	0.44
	2	−302	165	0.91	62.73	1.06
	5	−264	122	0.77	52.49	0.99
	8	−295	123	0.60	54.79	0.99
	22	−125	142	0.12	58.02	0.71
	26	−155	198	0.11	75.02	0.72
	29	−142	138	0.09	120.18	0.36
	44	−23	176	0.21	114.97	0.11
	54	−33	195	0.26	108.24	0.16
	68	−77	360	0.29	207.49	0.19
	1200 rpm
0.005	0	−164	465	3.93	38.99	1.79	3.05
	2	−219	260	0.57	59.00	1.16
	2	−322	355	1.07	78.47	1.24
	16	−363	344	0.82	137.33	0.74
	20	−307	330	0.70	111.96	0.82
	26	−286	344	0.61	109.79	0.82
	46	−357	344	0.61	55.37	1.82
0.01	0	−233	670	8.62	19.56	5.53	0.92
	2	−310	208	1.49	25.92	2.98
	5	−291	275	1.58	44.78	2.02
	9	−256	190	1.59	65.57	1.04
	25	−224	390	0.98	97.55	0.91
	30	−152	398	0.99	154.83	0.44
	46	−131	390	0.80	171.62	0.36
0.02	0	−202	481	8.62	29.09	3.05	2.95
	1	−536	1384	12.22	32.96	7.31
	5	−348	245	1.43	64.96	1.38
	7	−409	333	1.35	74.95	1.53

shows a comparison of the time evolution of corrosion rate for carbon steel determined by LVS, EIS and weight loss measurements. In this case, some Tafel slopes induce very low or high $i_{corr}$ values, which are in concordance with corrosion rates determined from EIS and weight loss.

The corrosion evolution with immersion time shown in Figure 7 indicates that in all cases $i_{corr}$ for RO wash water evolves from very high to nearly constant low values that are maintained all along the running time. The final $i_{corr}$ values for wash water are lower than that for pure RO water at any condition. This clearly indicates an inhibitory effect exerted by a film of reduced copper on the carbon steel. Under the batch condition of the present investigation, the copper precipitation rate on the carbon steel surface is expected to decrease mainly because of the decreasing copper concentration within the cell. In a real situation, where a large volume of wash water is available, the final $i_{corr}$ values could be even lower.

3.3. Morphology and Oxide Composition after 48 h Immersion Time

SEM images for carbon steel in pure RO and RO wash water are shown in Figure 8. The results shown in this figure suggests that steel corrosion immersed in pure RO takes place initially as a typical pitting mechanism that later spread to the sides. Also, there is some influence of the hydrodynamic interaction with solution, as the corroded areas show moderately elongated shape along the apparent flow direction [38,39]. In contrast on the steel immersed in RO wash water, the corroded areas are significantly more affected from hydrodynamic drag forces as is confirmed by corroded areas like continuous grooves along flow direction (Figure 8). The probable mechanism is a combination of pitting and galvanic corrosion from the cementation process exerted from a Cu-Fe couple (Equations (14) and (15)). No reference in the bibliography was found concerning this corrosion case. In a similar published situation consisting of the galvanic effect of magnetite on the corrosion of carbon steel under flow conditions it was reported that generalized corrosion is triggered under the influence of magnetite [40]. Furthermore, Figure 9 shows SEM and EDS analyses of the corrosion products deposited on the steel in RO solution and PLS wash water. The results clearly indicate the absence of Cu in the corrosion products when the steel was exposed to the pure RO solution. However, the corrosion products obtained when the steel was exposed to the PLS wash water reveal that the Cu content in the powders is proportional to the Cu concentration in solution, with values of 0.1, 0.6, and 2% for concentrations of 0.005, 0.01, and 0.02 L PLS/L RO, respectively. These results give more evidence of iron corrosion exerted by the presence of Cu in solution. In addition, the large percentage content of C in all the samples is attributed to the carbon staff used in the SEM analysis, which does not represent in its totality the carbon content in the carbon alloy.

The XRD diffractograms for the corrosion products obtained after corrosion processes in pure RO and RO wash water with addition of PLS are shown in Figure 10. According to the results, the absence of copper phase in pure RO water is confirmed, while for RO wash water, the copper phase was identified [27]. On the other hand, important peaks intensities for iron oxides and hydroxides phases such as lepidocrocite (γ-FeOOH), magnetite (Fe₃O₄), goethite (α-FeOOH) and maghemite (γ-Fe₂O₃), were identified which are characteristic oxide phases for steel corrosion [26,41,42].

4. Conclusions

From the experimental examination of corrosion of carbon steel ASTM A36 in RO water containing PLS from copper electrowinning process the following points can be emphasized:

Corrosion of ASTM A36 steel in pure RO water

The corrosion kinetics of carbon steel in pure RO water can be characterized in terms of partial electrochemical reactions of hydrogen evolution (HE), dissolved oxygen reduction (ORR) and iron oxidation (IO). The main aspects of this corrosion are the range of $\left|i_{l,O_2}\right|$ and $i_{corr}$ values that change between 0.3 to 3 A/m² and 0.3 to 1.5, respectively. These values that do not significantly change with immersion time are up to 4 times higher for corrosion under carbon steel rotation.

The instantaneous corrosion rate estimation in pure RO water using EIS and LSV results, together with the average corrosion determined by weight loss method, are all in very good agreement.

Corrosion of ASTM A36 steel in wash RO water

The partial electrochemical reactions involved in corrosion in wash water are hydrogen evolution (HE), dissolved oxygen reduction (ORR), iron oxidation (IO) and copper ion reduction that comes from PLS solution.

The experimental evidence indicates that iron corrosion in PLS wash water is mainly corroded by a cementation mechanism.

The corrosion kinetic of carbon steel in wash water can also be characterized in terms of the same kinetic expression used for pure RO water but the expression for ORR in this case represents the integrated partial reduction of dissolved oxygen and copper ions.

The main parameters to characterize the corrosion in PLS wash water are the range of $\left|i_{l,O_2}\right|$ and $i_{corr}$ values that change between 0.3 to 9 A/m² and 0.5 to 9, respectively. These values change with immersion time starting at very high values then decreased to a constant lowest value.

The significant $i_{corr}$ decrease with immersion time is attributed to copper ion reduction on the metal surface inducing a corrosion inhibition.

The instantaneous corrosion rate estimation using EIS and LSV results, together with the average corrosion determined using the weight-loss method, are in good agreement. Some deviations observed are attributed to changing conditions in the steel surface because of copper film detachment during the corrosion evolution.