materials

Article
Dissolution of Metals in Different Bromide-Based
Systems: Electrochemical Measurements and
Spectroscopic Investigations
Simona Varvara 1, *, Sorin-Aurel Dorneanu 2 , Alexandru Okos 3 , Liana Maria Muresan 2 ,
Roxana Bostan 1 , Maria Popa 1 , Daniel Marconi 4 and Petru Ilea 2
1 Department of Cadastre, Civil Engineering and Environmental Engineering, “1 Decembrie 1918” University
of Alba Iulia, 15-17 Unirii Street, 510009 Alba Iulia, Romania; [email protected] (R.B.);
[email protected] (M.P.)
2 Department of Chemical Engineering, Babes-Bolyai University, 11 Arany Janos Street, 400028 Cluj-Napoca,
Romania; [email protected] (S.-A.D.); [email protected] (L.M.M.);
[email protected] (P.I.)
3 Physical and Chemical Analysis Department (RBRO/EQV-A), SC Robert Bosch SRL, Tetarom III Industrial
Park, 1, Robert Bosch Street, Jucu Herghelie, 407350 Cluj, Romania; [email protected]
4 Department of Molecular and Biomolecular Physics, National Institute for Research and Development of
Isotopic and Molecular Technologies, 400293 Cluj-Napoca, Romania; [email protected]
* Correspondence: [email protected]; Tel.: +40-258-806263; Fax: +40-258-806329




Received: 7 July 2020; Accepted: 12 August 2020; Published: 17 August 2020


Abstract: The dissolution of the main metals (Cu, Zn, Sn, Pb and Fe) found in waste printed circuit
boards (WPCBs) was investigated by electrochemical corrosion measurements (potentiodynamic
polarization and electrochemical impedance spectroscopy (EIS)) in different bromide-based systems
that could be used as lixiviants in hydrometallurgical route of metals recovery. The analysis of the
corrosion products was carried out by X-ray diffraction (XRD) and X-ray photoelectron spectroscopy
(XPS) measurements. All measurements showed that the addition of bromine in the electrolyte favors
to great extents the dissolution process of all studied metals as compared to bromine-free electrolytes.
In the investigated experimental conditions, the highest dissolution rates of the metals were obtained
in acidic KBr solution containing 0.01 mol/L bromine and they decreased in the following order:
Zn >> Sn > Pb > Fe > Cu. The XRD and XPS chemical assessment allowed the identification of the
dissolution products formed on the metallic surfaces after exposure to the electrolytes. They consisted
mainly of oxides in the case of Cu, Zn, Sn and Fe, while the presence of PbBr2 was also noticed on the
lead surface. Based on the results of EIS and surface investigations, several models explaining the
corrosion behavior of the metals were proposed and discussed. The obtained results demonstrate
that all studied metals could be successfully leached using brominated solutions, providing a viable
alternative for the selective and efficient recovery of the base metals from WPCBs through a multi-step
hydrometallurgical processing route.

Keywords: dissolution of metals; metals recovery; bromide-based electrolytes; potentiodynamic

polarization; electrochemical impedance spectroscopy; XPS; XRD

1. Introduction
The hydrometallurgical route is a selective way toward metal recovery from waste printed
circuit boards (WPCBs), being easier to control and creating less environmental hazards than the
pyrometallurgical approach [1]. Various leaching reagents were used to ensure a fast kinetics of the
dissolution process of both noble and base metals: cyanide [2], thiosulfate [3,4], thiourea [5,6], H2 SO4

Materials 2020, 13, 3630; doi:10.3390/ma13163630 www.mdpi.com/journal/materials

Materials 2020, 13, 3630 2 of 28

(alone or in combination with H2 O2 , nitric acid or other oxidants) [7,8], potassium persulfate [9],
etc. Much less often, halides-based systems such as I− /I2 [10–12], I2 -H2 O2 [13] and Cl− /Cl2 [14] were
used. To the best of our knowledge, Br− /Br2 system was used before only for extraction of gold from
gold ores [15]. Fast leaching rate, low-toxicity and applicability over a wide range of pH values
(from acidic to neutral) are important characteristics of the bromide leaching system [16]. On the
other side, because bromine is an efficient lixiviant, but because elemental bromine is hazardous and
extremely difficult to store or transport in safe conditions, it is preferable to be produced in-situ, in the
leaching reactor [15,17].
In a previous research [18], we investigated the ability of the Br− /Br2 leaching system to remove
the exposed metallic parts from different models of computer motherboards (CMB), simultaneous
with the electrochemical lixiviants regeneration and the partial electrodeposition of the dissolved
metals. Although the operating parameters were not optimized, the results indicate that CMB could be
completely dismantled in about 18 h, with cathodic and anodic mean current efficiencies of 43.6% and
58.4%, respectively. Taking into consideration that leaching plays a key role in the electrochemical
recycling of WPCBs, a better understanding of the metals’ dissolution mechanisms in different lixiviants
might be an important step in the processes’ optimization.
In the present work, a fundamental study on the electrochemical dissolution behavior of the main
base metals (Cu, Zn, Sn, Pb and Fe) found in WPCBs was carried out by corrosion measurements
(potentiodynamic polarization and electrochemical impedance spectroscopy) in different bromide-based
systems that could be used as lixiviants in hydrometallurgical route of metals recovery. The paper
aims at assessing the electrochemical dissolution rates of the metals and establishing their dissolution
mechanisms in different bromide-based electrolytes.
To gain a deeper insight into the dissolution behavior of the investigated metals, the electrochemical
results were corroborated with XRD and XPS analysis of the metallic surface, after the attack of the
bromide-based lixiviants. These techniques allowed the identification of the surface compounds formed
during the dissolution of the metals, which might validate the corrosion mechanisms proposed based
on EIS measurements. This study is part of a more complex research project aiming to elaborate an
innovative and pollutants-free electrohydrometallurgical technology for metals recovery from WPCBs.

2. Materials and Methods

2.1. Materials
The working electrodes were cylindrical probes made of different pure metals embedded in
epoxy resin (EpoxycureTM, Buhler, Uzwil, Switzerland): Cu (0.28 cm2 ), Zn (0.27 cm2 ), Sn (0.28 cm2 ),
Pb (0.27 cm2 ) and Fe (0.28 cm2 ). Note that the working electrodes were made of pure metals instead
of waste PCBs materials, and therefore they should be defined as model electrodes, simulating the
metals from circuits. The exposed metallic surfaces were prepared via an abrading procedure, using
successive grade of silicon carbide paper grit (from 1200 to 4000), washed thoroughly with distilled
water and ethanol.
Four different bromide-based electrolytes were used as corrosive solutions:

- Sol. A: 2 M KBr (pH = 6);

- Sol. B: 2 M KBr + 0.5 M HBr (pH = 0.3);
- Sol. C: 2 M KBr + 0.5 M HBr (pH = 0.3) + 0.01 M Br2 ; and
- Sol. D: 2 M KBr + 0.5 M HBr (pH = 0.3) + 0.001 M Br2 .

The solutions were prepared using analytical grade reagents (Sigma Aldrich, Taufkirchen, Germany
and Merck, Darmstadt, Germany) and distillated water.
Materials 2020, 13, 3630 3 of 28

2.2. Electrochemical Measurements

A three-electrode cell was used for the electrochemical experiments. The potentials were measured
against Ag/AgCl/KClsat as a reference electrode (Ref), while a twisted platinum wire (Φ = 0.5 mm,
L = 10 cm) was used as counter-electrode.
The electrochemical measurements were carried out using a Princeton Applied Research (PAR)
model 2273 potentiostat (Princeton Applied Research, Oak Ridge, TN, USA, 2008). Before each
measurement, the working electrodes were left for at least 1 h in the corrosive solution to obtain
stable open circuit potentials (OCP). In each experiment, a fresh working electrode was used, and two
measurements were performed to ensure the data reproducibility.
Potentiodynamic polarization curves were recorded at a constant sweep rate of 10 mV min−1 ,
in a wide potential range of ±250 mV vs. OCP, from the cathodic to the anodic direction.
Impedance spectra were recorded at OCP, in the frequency range of 100 kHz–10 MHz, at 5 points
per hertz decade, using an alternating voltage amplitude of ±10 mV. The experimental impedance data
were modeled using ZSimpWin 3.21 software (EChem Software, MI, USA).
The electrochemical measurements were performed at room temperature, in the electrolytes under
non-stirred and naturally aerated conditions.

2.3. X-ray Photoelectron Spectroscopy Measurements

X-ray photoelectron spectroscopy (XPS) data were collected using a custom built Thermofisher
ESCALAB 250 Xi XPS spectrometer (Thermofisher, East Grinstead, UK) equipped with a scanning
electron microscope and an X-ray detector for the acquisition of EDX data. XPS spectra were acquired
using the Al Kα radiation (1486.6 eV). The data were collected from a circular surface with a radius
of 650 µm. The sample’s surface was kept perpendicular to the analyzer axis for all experiments.
The equipment was calibrated in the binding energy (B.E.) scale using an Au sample with the B.E. of
the main Au peak observed at 84 eV. The C1s peak attributed to adventitious C was also used as a
B.E. reference point for each sample. The B.E. for this peak was observed to vary between 284.6 and
285.1 eV between the samples. Survey scans were acquired in the constant analyzer energy mode
(CAE) using a pass energy of 200 eV. High resolution narrow scans spectra were also recorded in the
CAE mode using a pass energy of 15 eV. Depth profile experiments were performed using the lowest
possible ion beam energy (typically 500 eV) to minimize the reduction of the chemical species. The XPS
spectra were recorded and fitted using Avantage software (version 5.9915.0.6619, Thermofisher, East
Grinstead, UK). The background was modeled by a “Smart” function type for all samples. The peak
shape function was described by the product between a Lorentzian function and a Gaussian function
with a mixing parameter of 30%. The mixing parameter was usually constrained to remain fixed.
XPS data were collected after different immersion times of the investigated metals in Sol. C, i.e.,
60 min in the case of Cu, Sn and Zn and 120 min for Fe and Pb.

2.4. X-ray Diffraction Measurements

X-ray diffraction measurements (XRD) were carried out at room temperature on a Bruker D8
Advance powder diffractometer (Bruker, Karlsruhe, Germany) using Cu Kα1 radiation (λ = 0.154056 nm)
at 45 kV and 40 mA. The θ–2θ Bragg–Brentano configuration geometry and incident-beam Ge (111)
monochromator were used to investigate the structural properties of the samples. The 2θ range of
10◦ –85◦ was recorded at the rate of 0.02◦ and 2θ/0.5 s. XRD has a detection limit of ~5% by volume,
if a compound is present below this level, no peaks above noise will be detected in the diffraction
pattern. The exact detection limit depends on the density, Z number and crystal structure of the
compounds in the sample. The crystal phases were identified comparing the 2θ values and intensities
of reflections on X-ray diffractograms with JCP data base using a Diffraction AT-Brucker program.
Given the limitations of our structural detection, XRD diffractograms were collected after 120 h
immersion of Cu, Zn, Sn, Pb and Fe electrodes in Sol. A, Sol. B and Sol. C, respectively.
Materials 2020, 13, 3630 4 of 28

3. Results and Discussions

Materials 2020, 13, x FOR PEER REVIEW 4 of 29
3.1. Potentiodynamic Polarization Measurements
To3.1. Potentiodynamic
investigate Polarization Measurements
the dissolution behavior of the metals in different bromide-based electrolytes,
To investigate
potentiodynamic the dissolution
polarization behaviorwere
measurements of the metals in different
performed at room bromide-based
temperature. electrolytes,
Figure 1 illustrates
potentiodynamic
the polarization curvespolarization measurements
obtained after were for
1 h immersion performed at room
the anodic temperature.
and cathodic Figure 1
scans.
illustrates the polarization curves obtained after 1 h immersion for the anodic and cathodic scans.

-1 0

-2 -1

-3 -2
lg i / A cm-2

-2
lg i / A cm
-4 -3

-5 -4

-6 -5

-7 Cu -6 Zn
-0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 -1.3 -1.2 -1.1 -1.0 -0.9 -0.8 -0.7
E / V vs. Ag/AgCl/KClsat E / V vs. Ag/AgCl/KClsat

(a) (b)

-2
-2
-3
lg i / A cm-2
-2
lg i / A cm

-4
-4

-6 -5

-6
-8
Sn Pb
-7
-0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3
E / V vs. Ag/AgCl/KClsat E / V vs. Ag/AgCl/KClsat

(c) (d)
0

-1

-2
lg i / A cm-2

-3

-4

-5

-6
Fe
-7
-0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0
E / V vs. Ag/AgCl/KClsat
(e)

Figure Figure 1. Polarization

1. Polarization curves
curves collectedafter
collected after 1-h
1-h immersion
immersion of of
electrodes in different
electrodes bromide-based
in different bromide-based
electrolytes: Cu (a); Zn (b); Sn (c); Pb (d); and Fe (e). (—) Sol. A, (—) Sol. B, (—) Sol. C, (—) Sol. D.
electrolytes: Cu (a); Zn (b); Sn (c); Pb (d); and Fe (e). (—) Sol. A, (—) Sol. B, (—) Sol. C, (—) Sol. D.

The strong acidification of KBr solution, from pH = 6 to pH = 0.3, and the supplementary addition
of Br2 affect the electrochemical behavior of the investigated metals in different ways, as discussed below.
Materials 2020, 13, x FOR PEER REVIEW 5 of 29

Materials
The2020, 13, 3630
strong 5 of 28
acidification of KBr solution, from pH = 6 to pH = 0.3, and the supplementary addition
of Br2 affect the electrochemical behavior of the investigated metals in different ways, as discussed
below.
Figure 1 shows that, in most cases, the cathodic polarization curves exhibit limiting diffusion
Figure
plateaus, 1 shows
while that, in
the anodic most cases,
dissolution the cathodic
branches presentpolarization
either a Tafelcurves
behaviorexhibit limiting
or surface diffusion
passivation.
plateaus, while the
Consequently, the anodic dissolution
evaluation of the branches
corrosionpresent
kineticeither a Tafel behavior
parameters or surface
(i.e., corrosion passivation.
current density,
Consequently,
corrosion potential, anodic and cathodic Tafel slopes) by Tafel extrapolation method overdensity,
the evaluation of the corrosion kinetic parameters (i.e., corrosion current a wide
corrosionrange
potential potential,
would anodic and
lead to cathodic Tafel
uncertainties slopes)
and error by Tafel extrapolation method over a wide
sources.
potential range
Instead, inwould leadvicinity
the close to uncertainties and error sources.
of the open-circuit corrosion potential, it is expected that the
Instead, in the close vicinity of the open-circuit
Stern–Geary theory [19] could be applied. It considers corrosion potential,
that both it is expected
the anodic that the
and cathodic Stern–
processes
Geary theory [19] could be applied. It considers that both the anodic and cathodic
follow the Tafel law and the global current density i could be expressed by the following Equation: processes follow
the Tafel law and the global current density i could be expressed by the following Equation:
i = ia + ic = icorr {ebaba (E−Ecorr) - ebc (E−Ecorr) } (1)
i = ia + ic = icorr{e (E − Ecorr) – ebc (E − Ecorr)} (1)
where icorr is the
corr is corrosion current density, EEcorr
the corrosion the corrosion potential and ba and bcc are
is the
corr is are the anodic
and cathodic activation coefficients (V−1 ),respectively.
−1), respectively.
To yield
yield quantitative
quantitative approach,
approach,the
theelectrochemical
electrochemicalkinetic kineticparameters
parameters(i(i
corr , ,EEcorr
corr corr, baa and bc) were

evaluated from the experimental

then evaluated experimental data by a non-linear regression calculation near zero overall
current. A A user-defined
user-defined function
function of
of “Non-linear
“Non-linear least least squares
squares curve
curve fitting”
fitting” of Origin 9.3 software
(OriginLab Corporation, Northampton,
Northampton, MA,MA, USA)USA) based on Equation (1) was applied in the potential
domain limited
limited to ± 50 mV with respect to EEcorr
to ± [20]. In
corr [20]. Inall
allcases,
cases,the correlation factor R22 varied
thecorrelation varied within
0.9828–0.9998, indicating good fitting results.
As an example, Figure 2 presents the polarization curves obtained by regression calculation in
Sol. C overlaid on the experimental data represented by symbols.

-1

-2 Zn
Fe
lg i / A cm-2

-3 Pb Sn

-4 Cu

-5

-6
-1.0 -0.8 -0.6 -0.4 -0.2 0.0
E / V vs. Ag/AgCl/KClsat
Polarization curves
Figure 2. Polarization curves of various
various metals in Sol. C.
C. Symbols,
Symbols, experimental
experimental data; lines, calculated
Stern–Geary.
data by Stern–Geary.

The activation
The activation coefficients
coefficients were
were linked
linked to
to Tafel slopes, β,𝛽,bybythe
Tafel slopes, thefollowing
followingEquation:
Equation:
2.303
2.303
b𝑏 =
= (2)
(2)
β𝛽
The
The calculated
calculated corrosion
corrosion kinetic
kinetic parameters
parameters corresponding
corresponding to
to metal
metal corrosion
corrosion in
in different
different
bromide-based electrolytes are illustrated in Table 1.
bromide-based electrolytes are illustrated in Table 1.
Materials 2020, 13, 3630 6 of 28

Table 1. Corrosion kinetic parameters for metals in different bromide-based electrolytes.

Ecorr Icorr βa |βc |

Solution
(mV vs. Ref.) (µA cm−2 ) (mV dec−1 ) (mV dec−1 )
Cu
A −316.7 10.7 70.8 879.0
B −321.3 19.5 71.7 428.1
C −251.1 146.2 72.3 198.0
D −279.9 43.4 64.6 -
Zn
A −972.0 8.4 12.6 479.1
B −987.1 26580 145.6 104.9
C −978.2 27960 143.8 123.6
D −971.2 27790 159.3 97.1
Sn
A −409.2 1.4 37.7 79.3
B −503.5 13.4 23.4 187.1
C −483.7 399.1 39.1 148.2
D −481.3 92.2 29.2 164.5
Pb
A −557.9 37.8 32.0 336.2
B −557.9 37.8 32.0 336.2
C −525.9 164.4 24.6 67.9
D −527.3 50.4 25.2 108.3
Fe
A −383.8 2.07 149.8 nd
B −300.2 4.06 27.8 76.0
C −254.2 157.4 23.5 164.5

In the case of Cu, the electrochemical behavior was found to be rather similar in all investigated
electrolytes, as shown in Figure 1a. A very slight increase of the anodic and cathodic current densities
could be noticed at high potentials in strong acidic KBr solution.
The addition of increasing Br2 concentrations shifts the Ecorr towards more positive values and
markedly increases the corrosion current density (Table 1). For instance, the icorr value calculated in
Sol. C is almost 7.5 times higher than that obtained in Sol. B. This corelates with an increased corrosion
rate of Cu in the solution containing 0.01 M Br2 . However, as shown in Figure 1a, the addition of Br2
mainly accelerates the cathodic reduction of the dissolved oxygen, which is under diffusion control,
while the anodic dissolution of Cu is less affected. The increasing of the cathodic current density
observed in the presence of bromide could be due to a reduction of the molecular Br2 that takes place
in parallel with the oxygen reduction [21].
The anodic behavior of Cu in bromide-based electrolytes agrees with the results previously
reported by other authors [22–24]. Thus, an apparent Tafel region with a slope of around 70 mV dec−1
could be observed on the anodic polarization curves obtained in the absence and presence of Br2 .
The anodic process occurring in this region might be related to the fast transformation of Cu into Cu+
ions, followed by the formation of soluble CuBr2 − complexes [22,23]. As the potential becomes more
positive, an anodic current peak appears at about −60 mV vs. Ref on the polarization curves obtained
in brominated solutions. The subsequent decrease of the current density noticed in Figure 1a might be
associated to the formation of a porous and insoluble layer of CuBr on the electrode surface [24]. Finally,
a region of sudden increase in the current density appears to be due to the presence of the aggressive
Br− ions reducing the adhesion properties of the adsorbed CuBr layer and leading to formation of
soluble CuBr2 − complexes, which might suffer dissolution into the bulk solution. It has also been
reported that, at high potential values (>0.2 V vs. Ref), the presence of CuO on the metallic surface is
rather common, among the predominant Cu2+ compounds [24].
Materials 2020, 13, 3630 7 of 28

As expected, Zn presents the most negative potential (−0.97 V vs. Ref) and the highest anodic
activity in all investigated electrolytes (Figure 1b). In near neutral KBr solution, a well-defined cathodic
diffusion plateau was observed on the polarization curves due to the oxygen reduction reaction,
while the anodic current density increases exponentially for about four orders of magnitude, within a
potential range of +50 mV from Ecorr , with an apparent anodic Tafel slope of 12.6 mV dec−1 , slightly
lower as compared to a previously reported value [25].
A marked shift in both cathodic and anodic branches of the polarization curves towards higher
current densities could be noticed in the strong acidic electrolytes (Figure 1b), suggesting an important
enhancement of Zn corrosion. Although the calculated Ecorr values in different electrolytes are rather
similar (Table 1), the corrosion current density significantly increases as the solution’s pH decrease.
The changes in the anodic Tafel slopes suggest that the dissolution mechanism of Zn might
occur by a different pathway in strong acidic KBr solution as compared to neutral one. Nevertheless,
the addition of Br2 slightly accelerates the electrochemical reactions, but does not modify the mechanism
of Zn corrosion. The large estimated cathodic Tafel slopes are consistent with the hypothesis of a
diffusion limited reduction reaction.
In Sol. A, the region of active Sn dissolution observed at low potentials is followed by a narrow
semi-passive region [26], characterized by lower anodic current density values (Figure 1c), due to the
presence of a thin film of tin oxide on the electrode surface [27]. The passivity breakdown observed at
−0.32 V vs. Ref could be related to the pitting induced by Br− ions at the oxide–electrolyte interface [28].
As revealed in Figure 1c, the Sn dissolution process is significantly favored by the electrolyte
acidification. Moreover, the cathodic oxygen reduction reaction appears to be greatly enhanced by Br2
addition. In Table 1, an increase of icorr values can be noticed in all investigated acidic solutions, while
the Ecorr values are shifted towards less noble values. The highest negative shift of Ecorr was observed
in the presence of 0.01 M Br2 . At this concentration, the corrosion current density presents its greatest
value, which is about 29 and 285 times higher than icorr values obtained in Sol. B and Sol. A, respectively.
The acceleration of Sn dissolution in strong acidic electrolytes might be due to adsorption of aggressive
Br− ions at active sites of Sn surface, followed by the formation of soluble complexes transferred into
the solution. As previously stated by Johnson and Liu [29], such complexing processes decrease the
metallic ions concentration retarding the formation of the oxide passive layer. However, at around
−0.35 V vs. Ref, a minimum peak current could be observed on the polarization curves obtained in
strong acidic KBr solution, both in the absence and in the presence of Br2 . Its interpretation is rather
uncertain; some authors attributed it to the formation of Sn(IV) species [26], while others to some
complex compounds formed on the metallic surface [28]. Nevertheless, the increased aggressiveness
of Sol. C induces pitting corrosion, which explains the sharp increase of the current density values at
high potentials (Figure 1c).
The anodic Tafel slopes present similar values regardless of the electrolyte’s nature, suggesting
that the Sn dissolution mechanism is rather similar.
As shown in Figure 1d, the polarization curves of Pb were shifted towards more positive values
in strong acidic KBr solutions, and even more in the presence of Br2 . Although not very clearly seen
in Figure 1d, a small transition region appears at around −0.58 V vs. Ref on the polarization curves
obtained in Sol. A; it may correspond to the formation of a passivating layer on the metallic surface,
which further dissolves as the potential become more positive. It may be inferred that the reactivation
region is associated with the formation of soluble lead species, subsequent to the passivating step [30].
In strong acidic solutions, regardless the addition of bromine, the superficial corrosion products
layer cannot impart passivity because its solubility is relatively high [30] and Pb surface dissolves
readily leading to Pb2+ formation [31]. A similar behavior was previously noticed for Pb corrosion in
chloride electrolytes [30].
At about −0.47 V vs. Ref, a sharp increase of the current density could be observed on the anodic
part of the polarization curves in all investigated electrolytes. This decrease of the current density may
be due to the formation of a layer corrosion products on Pb surface.
Materials 2020, 13, 3630 8 of 28

In Table 1, it is clear that the overall corrosion process of Pb is enhanced by the aggressiveness of
bromine. Thus, the calculated icorr value reached 144.4 µA cm−2 in the presence of 0.01 M Br2 , which is
almost four time greater than the value obtained in Br2 -free acidic solution (37.8 µA cm−2 ).
As previously reported, Pb corrosion in aqueous solutions might be represented by a two-electron
transfer reaction [31]. Depending on the solution’s nature, this reaction may occur in a sequence of
steps, in which the formation of monovalent intermediate is possible [31], as suggested by the low
values of βa calculated in strong acidic solutions, in the absence and in the presence of Br2 (Table 1).
In the case of iron, Br− ion might adsorb on the electrode surface promoting the metal dissolution [32].
Consequently, no passivation of Fe surface was observed in the studied experimental conditions
(Figure 1e). Particularly, the Br2 addition significantly enhances both the active dissolution of Fe and
cathodic oxygen reduction, which is under diffusion control. Thus, the corrosion current density in the
presence of 0.01 M Br2 is more than 38 and 76 times higher as compared to icorr values calculated in
Sol. B and Sol. A, respectively. Moreover, a displacement of Ecorr towards more noble values takes place
in Br2 -containing solutions (Table 1). The low values of the anodic Tafel slopes calculated in all acidic
electrolytes advises that some adsorbed intermediates might be involved in Fe corrosion mechanism.

3.2. Electrochemical Impedance Spectroscopy

The electrochemical dissolution behavior of the metals in different bromide-based systems was
further investigated by electrochemical impedance spectroscopy (EIS). The impedance measurements
were conducted at the open-circuit potentials and the obtained results are displayed as Nyquist
diagrams and Bode plots in Figures 3–7. The insets in Figures 3–7 present some of the Nyquist diagrams
in enlarged scale.
As shown in Figures 3–7, disregarding the metal, an important decrease of the impedance magnitude
could be observed in strong acidic KBr solution as compared to the near neutral one, and even more in
the presence of Br2 , which is consistent with an increased dissolution rate of all metals in brominated
electrolytes, in agreement with the potentiodynamic polarization results.
Before attempting to fit the experimental impedance data with electrical equivalent circuits,
an evaluation of the EIS data validity with respect to linearity, causality and stability was carried out.
For this purpose, Kramers–Kronig relations were applied to several experimental impedance data by
transforming the real axis into the imaginary axis and the imaginary axis into the real axis and then
comparing the transformed quantities with the corresponding experimental data [24]. The satisfactory
agreement between the measured and transformed values (within the range of experimental errors)
demonstrates that the experimental impedance data are consistent with K-K transformations and
therefore stable, as shown in Section S1 (Figure S1).
EIS results obtained in Sol. A, Sol. B and Sol. C were further analyzed by numerical simulation, using
different equivalent electrical circuits that allow estimation of the R-Q parameters corresponding to the
electrochemical systems under investigation and might validate the proposed reaction mechanisms [31].
The capacitive contributions of the impedance plots were simulated using constant phase elements
(CPE), represented by the terms Q and n, instead of a pure capacitors (C) due to the non-ideal behavior
of the metallic surfaces. The semicircle depression is often explained by the surface heterogeneity,
caused by surface roughness, formation of porous layers, defects in the crystal lattice, variations in the
properties or compositions of the surface layers, adsorption and presence of impurities [33].
The impedance of CPE is given by [34]:

Q = ZCPE(ω) = [C(jω)n ]−1 (3)

where Q represents a pre-exponential factor, which is a frequency-independent parameter with

dimensions of Ω−1 cm−2 sn ; j is an imaginary number; ω = 2πf is angular frequency in rad·s−1 ; and n is
the exponent which defines the character of frequency-dependence (−1 ≤ n ≤ 1).
 heterogeneity, caused by surface roughness, formation of porous layers, defects in the crystal lattice,
variations in the properties or compositions of the surface layers, adsorption and presence of
impurities [33].
The impedance of CPE is given by [34]:
Materials 2020, 13, 3630 Q = ZCPE(ω) = [C(jω)n]−1 (3) 9 of 28
where Q represents a pre-exponential factor, which is a frequency-independent parameter with
dimensions
The ofthe
values of Ω−1pseudo-capacitances
cm−2 sn; j is an imaginary
(C) number; ω with
associated = 2πf CPEs
is angular
werefrequency in rad·
recalculated s−1; the
using andEquation:
n is
the exponent which defines the character of frequency-dependence (−1 ≤ n ≤ 1).
The values of the pseudo-capacitances (C) 1/n with CPEs were recalculated using the
 associated
C = R1−n Q (4)
Equation:

 R1 nQ
The equivalent electrical circuits used forCexperimental 1/ n
EIS data simulation are illustrated (4) together 
with the impedance diagrams in Figures 3–7.
The equivalent electrical circuits used for experimental EIS data simulation are illustrated
The used equivalent electrical circuits reproduce properly all experimental impedance, as could
together with the impedance diagrams in Figures 3–7.
be seen inTheFigures 3–7, whereelectrical
used equivalent good overlapping between
circuits reproduce the measured
properly data (represented
all experimental impedance, as bycould
scattered
symbols) and
be seen incalculated
Figures 3–7,ones (denoted
where by —+—)between
good overlapping were obtained for alldata
the measured studied systems.
(represented byThe quality of
scattered
the fitting procedures was also 2 ) values situated in 10−4 –10−3 range
symbols) and calculated onesevaluated
(denoted byby—+—)
the Chi-squared
were obtained (χfor all studied systems. The quality
and by thefitting
of the error procedures
percentageswas corresponding
also evaluated byto the
each parameter(χof
Chi-squared the used
2) values electrical
situated circuits,
in 10−4–10 which
−3 range

and by the error

were mostly below 10%. percentages corresponding to each parameter of the used electrical circuits, which
were mostly
Since below 10%.
the physical meaning of the time constants used for impedance simulations was different,
Since the physical
depending on the studied meaning
metalofand
the time constantscharacteristics,
electrolyte used for impedance simulations
a detailed was different,
discussion on their
depending on the studied metal and electrolyte characteristics, a detailed discussion on their
interpretation will be further provided.
interpretation will be further provided.
3.2.1. Copper
3.2.1. Copper
The experimental EIS spectra obtained from Cu surface exposed to the bromide-based electrolytes
The experimental EIS spectra obtained from Cu surface exposed to the bromide-based
are presented in Figure 3. in Figure 3.
electrolytes are presented

100
1200 (a) (a')
- ZImg /  cm2

1000 70 (a'')
- Phase angle / degree

50 60
- ZImg /  cm2

900
100 50
| z | /  cm2

0 40
0 50 100
600
ZRe /  cm 2
30
10
20
300
10
1
0 0
0 300 600 900 1200 10-2 10-1 100 101 102 103 104 105
10-2 10-1 100 101 102 103 104 105
ZRe /  cm2 f / Hz
f / Hz

(b) (c)

Figure 3. EIS results of Cu after immersion in different bromide-based electrolytes: Nyquist diagrams (a);
Bode plot of |Z| vs. frequency (a’); and Bode plot of phase angle vs. frequency (a”). The equivalent
electrical circuits used to fit the EIS data (b,c). Symbols represent the experimental data and lines with
cross (—+—) correspond to the simulated spectra. Electrolytes: () Sol. A; (O) Sol. B; (♦) Sol. C; and
(∆) Sol. D.

As shown in Figure 3a, the impedance diagrams of Cu in the investigated electrolytes are
characterized by a semicircular appearance with badly separately capacitive loops. Although not
clearly seen in Figure 3a, two-time constants are necessary to suitably reproduce the experimental
data obtained in Sol. A and Sol. B. The equivalent electrical circuit [35,36] depicted in Figure 3b was
adopted for these data simulations. In the equivalent circuit in Figure 3b, Re represents the electrolyte
resistance, the high-frequency parameters (Rct −Qdl ) correspond to the charge transfer resistance and
double layer capacitance and the low frequency circuit represented by RF −QF was ascribed to a redox
process involving the corrosion products (i.e., Cu2 O and CuO) accumulated at the interface, based on
the following reactions [20]:
Cu ↔ Cu+ ad + e−
Materials 2020, 13, 3630 10 of 28

Cu+ ad → Cu2+ + e−

The impedance spectra obtained in the presence of Br2 were fitted to the equivalent circuit in
Figure 3b. It contains an additional Warburg impedance to account for the appearance of the diffusion
tail observed at lowest frequency region in the EIS diagrams collected from Br2 -containing electrolytes.
The R-Q parameters for Cu dissolution in different bromide-based electrolytes obtained by fitting
the equivalent electrical circuits are given in Table 2.

Table 2. Electrochemical parameters of Cu dissolution in different bromide-containing systems.

Re Rct Qdl Cdl RF QF CF W

Sol. ndl nF
(Ω cm2 ) (Ω cm2 ) (µF sn−1 cm−2 ) (µF cm−2 ) (Ω cm2 ) (mF sn−1 cm−2 ) (mF cm−2 ) (S s1/2 cm−2 )
A 0.79 191.5 9.05 0.91 4.82 3783 0.44 0.46 0.71 -
B 0.53 48.78 161.1 0.80 47.93 488.3 1.80 0.51 1.58 -
C 0.56 21.54 212 0.79 50.64 38 0.23 0.95 0.18 0.19

As shown in Table 2, the charge transfer resistance, Rct , values significantly decrease as the
electrolyte’s pH becomes acidic and even more in the presence of 0.01 mM Br2 . This decay of Rct is
accompanied by an increase of the double layer capacitance Cdl . The corrosion process is faster in
strong acidic electrolytes and Cu surface becomes coarser, leading to higher Cdl values. At the same
time, the faradaic resistance calculated in Sol. C is more than 9 and 75 times lower than the RF values
obtained in Sol. B and Sol. A, respectively. These results advise for an increased electrochemical
reactivity of Cu in brominated electrolyte, in agreement with the polarization measurements.

3.2.2. Zinc
The experimental EIS spectra obtained from Zn electrode exposed to the bromide-based electrolytes
are presented in Figure 4.
As shown in Figure 4, the impedance plots obtained during Zn dissolution in near neutral and
strong acidic KBr electrolytes present different features, which indicates that some changes in the
reaction mechanisms with the solution pH might occur.
Although the corrosion behavior of Zn in bromide [37] or bromine-containing solutions is scarcely
reported [38], its dissolution mechanism in chloride-based electrolytes was subject to numerous
investigations. The previously reported reaction models [39–42] involve several parallel steps of
dissolution and three adsorbed species (Znad + , ZnOHad and Znad 2+ ), depending on Zn surface
preparation and the experimental conditions (pH, oxygen concentration and the presence of different
oxidized species, i.e., oxide or hydroxide, on Zn surface) [43]. For example, Cachet and Wiart [39]
proposed a reaction
Materials 2020, 13, x FORscheme for the dissolution of Zn in aerated ZnCl2 and NH4 Cl which involves
PEER REVIEW 12 of 29two
parallel steps, stimulated by the presence of chloride ions:

𝐂𝐥−
(2)
𝐞− + 𝐙𝐧𝟐+ → 𝐙𝐧𝟐+
𝐚𝐝 (3) 𝐬𝐨𝐥
⍬2
𝐂𝐥−
𝐙𝐧 𝐞− + 𝐙𝐧+
𝐚𝐝 ⍬𝟏
(1)

𝐙𝐧,𝐂𝐥−
(5)
𝐙𝐧𝟐+ +
𝐬𝐨𝐥 + 𝐙𝐧𝐚𝐝 + 𝟐𝐞
−

𝐂𝐥− ,𝐇 +
+
𝐙𝐧𝐎𝐇𝐚𝐝 𝐙𝐧𝐎𝐇𝐬𝐨𝐥 + 𝐞−
(4)

The major path (2) is catalyzed by Zn+ads. The minor path (1), is much more dependent on the
diffusion of the chloro-zinc species than the major one. The formation of ZnOHads is considered a side
reaction, which is caused by the chemical oxidation of Zn by the electrolyte. Close to the rest potential,
additional difficulties come from the competition between the slow chemical formation of the
Materials 2020, 13, 3630 11 of 28

The major path (2) is catalyzed by Zn+ ads . The minor path (1), is much more dependent on the
diffusion of the chloro-zinc species than the major one. The formation of ZnOHads is considered a side
reaction, which is caused by the chemical oxidation of Zn by the electrolyte. Close to the rest potential,
additional difficulties come from the competition between the slow chemical formation of the oxidized
protective layer and its elimination under the effect of Zn dissolution [41].
This study found that the impedance corresponding to Zn corrosion in Sol. A could be suitably
represented using two time constants under capacitive relaxation and an inductive contribution at low
frequency, according to the electrical circuit in Figure 4b. The two capacitive loops were ascribed to
the charge transfer process (Rct −Qdl ) and the coverage relaxation of an adsorbed Znad + intermediate
(R1 −Q1 ), respectively. The appearance of an inductive loop (RL −L) in the low frequency-end was
assumed to be due to the relaxation of an oxidized intermediate species (i.e., ZnOHad ); its formation
might express the chemical oxidation of Zn by the electrolyte [41]. This very thin layer of oxidized Zn
species might present some protective properties on the metallic surface [41,42,44]. As discussed below,
the low value of Cdl (5.5 µF cm−2 ) obtained in Sol. A is consistent with the presence of a protective
layer on Zn surface. Furthermore, the presence of a thin layer of ZnO on the metallic surface
Materials 2020, 13, x FOR PEER REVIEW
was
11 of 29
identified in the XRD pattern of Zn exposed for 120 h to Sol. A, according to the data in Figure 8b.

600
1.5
(a) (a')
- ZIm /  cm2

500 1.0 1000

| z | / .cm2

400
- ZIm /  cm2

0.5

0.0
300 1 2 100
ZRe /  cm2
200
10
100

0
1
-100
0 100 200 300 400 500 600 700 10-2 10-1 100 101 102 103 104 105
ZRe /  cm 2
f / Hz

80
- Phase angle / degree

70 (a'')
60
50
40
30
20
10
0
-10
10-2 10-1 100 101 102 103 104 105 (b)
f / Hz

(c)

Figure 4.
Figure 4. EIS
EIS results of Zn
results of Zn after
afterimmersion
immersioninindifferent
differentbromide-based
bromide-based electrolytes:
electrolytes: Nyquist
Nyquist diagrams
diagrams (a);
(a); Bode plot of |Z| vs. frequency (a’); and Bode plot of phase angle vs. frequency (a”).
Bode plot of |Z| vs. frequency (a’); and Bode plot of phase angle vs. frequency (a”). The equivalent The equivalent
electrical circuits
electrical circuits used
used for
for EIS
EIS fitting
fitting (b,c).
(b,c). Symbols
Symbols represent
represent the
the experimental
experimental data
data and
and lines
lines with
with
cross (—+—) correspond to the simulated spectra. Electrolytes: (□) Sol. A; (O) Sol. B;
cross (—+—) correspond to the simulated spectra. Electrolytes: () Sol. A; (O) Sol. B; (♦) Sol. C; and (◊) Sol. C; and
(Δ) Sol. D.
(∆) Sol. D.

As shown in Figure 4, the impedance plots obtained during Zn dissolution in near neutral and
strong acidic KBr electrolytes present different features, which indicates that some changes in the
reaction mechanisms with the solution pH might occur.
Although the corrosion behavior of Zn in bromide [37] or bromine-containing solutions is
 ⍬2
𝐂𝐥−
𝐙𝐧 𝐞− + 𝐙𝐧+
𝐚𝐝 ⍬𝟏
(1)

Materials 2020, 13, 3630 12 of 28

𝐙𝐧,𝐂𝐥−
(5)
𝐙𝐧𝟐+ +
𝐬𝐨𝐥 + 𝐙𝐧𝐚𝐝 + 𝟐𝐞
−

The decrease in the solution pH significantly favors the Zn dissolution process, hindering the
formation of the oxidized species on the metallic surface. This might explain the vanishing of the
inductive loop at low frequency [25]. 𝐂𝐥 − ,𝐇 +
Nevertheless, the appearance of a small inductive loop at
+ −
intermediate frequency (~10 𝐙𝐧𝐎𝐇 Hz) between
𝐚𝐝 two 𝐙𝐧𝐎𝐇
capacitive 𝐬𝐨𝐥 + 𝐞could be noticed in strong acidic
contributions
electrolytes, both in the absence and in the (4)presence of Br2 (Figure 4a) [25,41]. The similar features
of the impedance diagrams obtained in strong acidic electrolytes, regardless of the presence of Br2 ,
confirmThethat a similar
major dissolution
path (2) mechanism
is catalyzed by Zn+ads.ofTheZn minor
is involved, in agreement
path (1), withdependent
is much more the polarization
on the
measurements. Thus, the inductive loop was attributed to the relaxation of the adsorbed Zn +
diffusion of the chloro-zinc species than the major one. The formation of ZnOHads is considered a side ad
intermediate,
reaction, whichwhile the low-frequency
is caused capacitive
by the chemical oxidationsemicircle might
of Zn by the be associated
electrolyte. Close to the rest
to the formation of
potential,
Zn 2+ species during the dissolution reaction [45,46]. However, as shown in Figure 4a, the maximum
ad
additional difficulties come from the competition between the slow chemical formation of the
frequency of the second
oxidized protective layercapacitive loop is very
and its elimination low,the
under which allowed
effect us to assume
of Zn dissolution that the formation
[41].
and consumption of the adsorbed intermediate occurs in parallel with the main dissolution path [46].

80

- Phase angle / degree

15
100 80
- ZIm /  cm2
- ZIm /  cm2

10
50
- ZIm / k cm2

60 5 60
0
0
0 5 10 15
ZRe /  cm2
-50
0 50 100 150
40
40 ZRe /  cm2
20

20
0

(a) (a')
0 -20
0 20 40 60 80 10-2 10-1 100 101 102 103 104 105
2
ZRe / k cm f / Hz

100

10
| Z | / k cm2

0.1

0.01

0.001
(a'')
10-2 10-1 100 101 102 103 104 105
f / Hz
(b)

(c) (d)

Figure 5. EIS results of Sn after immersion in different bromide-based electrolytes: Nyquist diagrams (a);
Bode plot of |Z| vs. frequency (a’); and Bode plot of phase angle vs. frequency (a”). The equivalent
electrical circuits used for EIS fitting (b–d). Symbols represent the experimental data and lines with
cross (—+—) correspond to the simulated spectra. Electrolytes: () Sol. A; (O) Sol. B; (♦) Sol. C; and
(∆) Sol. D.

The impedance diagrams of Zn dissolution in strong acidic KBr solutions were accurately simulated
using the equivalent circuit in Figure 4c and several calculated R-Q parameters are presented in Table 3.
Materials 2020, 13, 3630 13 of 28

Table 3. Electrochemical parameters of Zn dissolution in different bromide-containing electrolytes.

Re Rct Qdl Cdl R1 Q1 C1 RL L R2

Sol. ndl n1
(Ω cm2 ) (Ω cm2 ) (µF sn−1 cm−2 ) (µFcm−2 ) (Ω cm2 ) (µF sn-1 cm−2 ) (µF cm−2 ) Ω cm2 ) (H cm2 ) (Ω cm2 )
A 1.02 322.9 12.3 0.87 5.5 285.5 89.02 0.70 18.4 71.4 39.8 -
B 0.67 0.87 6000 0.55 86.4 - - - - 0.95 0.001 0.065
C 0.59 0.29 7200 0.61 141.5 - - - - 0.28 0.0002 0.046

Materials 2020, 13, x FOR PEER REVIEW

As shown in Table 3, the charge transfer resistance Rct presents a similar pH dependence 13 asofthe
29

icorr values calculated from polarization measurements; it decreases from about 322.9 Ω cm2 in Sol. A
Figure 5. EIS results of Sn after immersion in different bromide-based electrolytes: Nyquist diagrams
to 0.87 Ω cm2 in Sol. B, confirming that Zn suffers severe corrosion in strong acidic KBr electrolyte.
(a); Bode plot of |Z| vs. frequency (a’); and Bode plot of phase angle vs. frequency (a”). The equivalent
The addition of Br2 accelerates to some extent the dissolution process, as proved by the Rct value
electrical circuits used for EIS fitting (b–d). Symbols represent the experimental data and lines with
which is more than three orders of magnitude lower in the presence of 0.01 M Br2 , as compared to its
cross (—+—) correspond to the simulated spectra. Electrolytes: (□) Sol. A; (O) Sol. B; (◊) Sol. C; and (Δ)
absence. Additionally, the significant increase of Cdl values in Br2 -containing electrolytes corroborates
Sol. D.
the formation of coarse surface during Zn dissolution, as revealed by SEM micrograph (results shown
in Figure S2).

800 1000
(a) (a')
- ZIm /  cm2

150

100
600
100
- ZIm /  cm2

| Z | /  cm2

50

0
400 0 50 100 150
ZRe /  cm2 10

200

0
0 200 400 600 800 10-2 10-1 100 101 102 103 104 105
ZRe /  cm2 f / Hz

70
(a'')
- Phase angle / degree

60

50

40

30

20

10

0
10-2 10-1 100 101 102 103 104 105
f / Hz
(b)

(c)

Figure 6. EIS
EIS results after immersion
results of Pb after immersioninindifferent
differentbromide-based
bromide-based electrolytes:
electrolytes: Nyquist
Nyquist diagrams
diagrams (a);
Bode plot of |Z| vs. frequency (a’); and Bode plot of phase angle vs. frequency (a”). The
(a); Bode plot of |Z| vs. frequency (a’); and Bode plot of phase angle vs. frequency (a”). The equivalent
electrical
electrical circuits
circuits used
used for
for EIS fitting (b,c). Symbols
Symbols represent
represent the
the experimental
experimental datadata and
and lines
lines with
with
cross
cross (—+—)
(—+—)correspond
correspondtotothethesimulated
simulatedspectra.
spectra.Electrolytes:
Electrolytes:(□)()
Sol. A;A;
Sol. (O)(O)
Sol.Sol.
B; (◊) Sol.Sol.
B; (♦) C; and (Δ)
C; and
Sol. D. D.
(∆) Sol.

35
1.0
- ZIm / k cm2

(a) (a')
30 0.5 10
z | / k cm2
/ k cm2

25 0.0
1
20 -0.5
0.0 0.5 1.0 1.5
2
ZRe / k cm 0.1
15
Im
 Figure 6. EIS results of Pb after immersion in different bromide-based electrolytes: Nyquist diagrams
(a); Bode plot of |Z| vs. frequency (a’); and Bode plot of phase angle vs. frequency (a”). The equivalent
electrical circuits used for EIS fitting (b,c). Symbols represent the experimental data and lines with
cross (—+—) correspond to the simulated spectra. Electrolytes: (□) Sol. A; (O) Sol. B; (◊) Sol. C; and (Δ)
Materials D. 13, 3630
Sol.2020, 14 of 28

35
1.0

- ZIm / k cm2
(a) (a')
30 0.5 10

| z | / k cm2
- ZIm / k cm2
25 0.0
1
20 -0.5
0.0 0.5 1.0 1.5
ZRe / k cm2 0.1
15

10
0.01
5
0.001
Materials 2020, 13,0 x FOR PEER REVIEW 14 of 29
0 5 10 15 20 25 30 35 10-2 10-1 100 101 102 103 104 105
2
ZRe / k cm f / Hz
80
70 (a'')
- Phase angle / degree

60
50
40
30
20
10
0
-10

10-2 10-1 100 101 102 103 104 105

f / Hz
(b)

(c) (d)

Figure
Figure 7. Displays
7. Displays the the impedance
impedance diagrams
diagrams ofimmersed
of Fe Fe immersed in studied
in the the studied electrolytes.
electrolytes. EIS EIS results
results of of
Fe after immersion in different bromide-based electrolytes: Nyquist diagrams (a); Bode plot of |Z| vs.|Z|
Fe after immersion in different bromide-based electrolytes: Nyquist diagrams (a); Bode plot of
vs. frequency
frequency (a’); and(a’);
Bodeand Bode
plot plot ofangle
of phase phasevs.angle vs. frequency
frequency (a”). The(a”). The equivalent
equivalent electricalelectrical circuits
circuits used
used
for EIS for EIS
fitting fitting
(b–d). (b–d).
Symbols Symbolsthe
represent represent the experimental
experimental data and linesdatawithand
crosslines withcorrespond
(—+—) cross (—+—)
correspond
to the simulatedtospectra.
the simulated spectra.
Electrolytes: ()Electrolytes: (□) Sol.
Sol. A; (O) Sol. A;Sol.
B; (♦) (O) C;
Sol.and
B; (◊)
(∆)Sol.
Sol.C;
D.and (Δ) Sol. D.

3.2.3. TinThis study found that the impedance corresponding to Zn corrosion in Sol. A could be suitably
represented
The impedance usingspectra
two timeof Sn constants
exposed under capacitive
to different relaxation and
bromide/bromine an inductive
electrolytes contribution
are illustrated in at
low
Figure 5. frequency, according to the electrical circuit in Figure 4b. The two capacitive loops were ascribed
toAs
theshown
chargein transfer
Figureprocess
5a, the(RNyquist
ct − Qdl) and the coverage relaxation of an adsorbed Znad+ intermediate
diagram obtained for Sn corrosion in near neutral KBr
(R 1 − Q1), respectively. The appearance of an inductive loop (RL − L) in the low frequency-end was
solution consists of a wide capacitive contribution, with a large imaginary part at the lowest frequency.
Theassumed
attemptsto to be due to the
represent the relaxation
EIS data ofof Snan oxidized
exposed intermediate
to Sol. A by one species (i.e., ZnOH
time constant circuitad); its formation
were found
inappropriate and two time constants were required to accurately reproduce the experimentalofresults.
might express the chemical oxidation of Zn by the electrolyte [41]. This very thin layer oxidized
Zn species
As revealed inmight
Figurepresent
5b, the some
modelprotective
adopted forproperties
simulationon the metallic
includes surface
one couple, [41,42,44]. As discussed
Rf −Qf , ascribed to
below, the low value of C dl (5.5 µF cm−2) obtained in Sol. A is consistent with the presence of a
the capacitance and resistance of a thin passive layer composed of tin oxides, formed on the metallic
protective
surface [47] andlayer
the on Zn surface.
Rct −Q Furthermore, the presence of a thin layer of ZnO on the metallic
dl parameters corresponding to the charge transfer resistance and double
surface was identified in the
layer capacitance. The XRD measurements XRD patternperformed
of Zn exposedafterfor
5 h120 h to Sol. A,
immersion according
of Sn in Sol. to the data
A did not in
Figure 8b.
detected the presence of any corrosion products (results not shown here) because their amount was too
small [48].TheInstead,
decrease theinoccurrence
the solution of pHSnOsignificantly favors the Zn dissolution process, hindering the
2 on the surface was evidenced in the XRD diffractogram
formation of the oxidized species
obtained at longer exposure time (Figure 8c). on the metallic surface. This might explain the vanishing of the
inductive loop at low frequency [25]. Nevertheless, the appearance of a small inductive loop at
intermediate frequency (~10 Hz) between two capacitive contributions could be noticed in strong
acidic electrolytes, both in the absence and in the presence of Br 2 (Figure 4a) [25,41]. The similar
features of the impedance diagrams obtained in strong acidic electrolytes, regardless of the presence
of Br2, confirm that a similar dissolution mechanism of Zn is involved, in agreement with the
polarization measurements. Thus, the inductive loop was attributed to the relaxation of the adsorbed
Znad+ intermediate, while the low-frequency capacitive semicircle might be associated to the
 based electrolytes are shown in Figure 8.
Prior to XRD measurements, the above-mentioned metallic samples were mechanically and
chemically cleaned to remove any traces of corrosion from their surface. XRD diffractograms for all
cleaned samples (pure curves in Figure 8) show that they are pure, without traces of oxides or other
compounds
Materials on3630
2020, 13, their surface. 15 of 28

Materials 2020, 13, x FOR PEER REVIEW 18 of 29

(a) (b)

(c) (d)

(e)

Figure8. 8.XRD
Figure XRDdiffractograms
diffractograms collected
collected before
before and
and after
after 120
120hhimmersion
immersionofofelectrodes
electrodesinindifferent
different
bromide-basedelectrolytes:
bromide-based electrolytes:Cu
Cu(a);
(a);Zn
Zn (b);
(b); Sn
Sn (c);
(c); Pb
Pb (d);
(d); and
and Fe
Fe (e).
(e).

AnThe long immersion

additional inductive of the
loopmetals
at lowinfrequencies
different bromide-based
is observed inelectrolytes
EIS spectrainduces structural to
corresponding
Snmodifications
corrosion in for all electrodes,
strong acidic KBraselectrolytes,
shown in their bothXRD diffractograms.
in the absence andIninthethecase of Cu electrode
presence of 0.001 M
immersed in Sol. A and Sol. B, the XRD diffractograms shows a combination
Br2 (Figure 5c). Several models of equivalent circuits were attempted to fit these experimental of CuO and Cudata.
2O

phases.
The By exposingbetween
best agreement the Cu electrode to Sol.
experiment C, the
and corrosion
fitting productsusing
was obtained changethe
their composition
equivalent and in
circuit
XRD diffractograms show only peaks related to the Cu 2O phase. In the case of Sn, the additional
Figure 5c. It contains an additional (RL −L) couple proposed to account for the low-frequency inductive
diffraction
time constant,peaks
whichappearing
might be in the diffraction
attributed patterncorrosion
to the pitting after 120induced
h were by
assigned
Br− ionsto on
SnO and their
Sn2 surface [49].
intensity increases in Sol. C. In the case of Pb the influence of the Sol. C is more obvious.
However, as the concentration of Br2 increases to 0.01 M, the low-frequency inductive loop becomes The peaks
in the XRD diffractograms were indexed with red arrows (for PbBr2 phase) and black arrows (for
capacitive and the equivalent circuit in Figure 5d was used to simulate the corresponding impedance.
Pb3O4 phase). The predominance of PbBr2 becomes visible.
The XRD diffractograms of Zn and Fe after immersion in Sol. A and Sol. B reveal the presence
of corrosion products, i.e., Fe3O4 and ZnO, respectively. Instead, the XRD diffractograms for Fe
immersion in Sol. C show only the peaks corresponding to KBr.
From the XRD diffractograms obtained on the metal electrodes, a preferential orientation
tendency of the crystallites of the corrosion products was observed according to certain
Materials 2020, 13, 3630 16 of 28

Table 4 illustrates the equivalent circuit parameters calculated for Sn dissolution in the
investigated electrolytes.

Table 4. Electrochemical parameters of Sn dissolution in different bromide-containing solutions.

Re Rf Qf Cf Rct Qdl Cdl R L

Sol. nf ndl
(Ω cm2 ) (Ω cm2 ) (µF sn−1 cm−2 ) (µF cm−2 ) (Ω cm2 ) (mF sn−1 cm−2 ) (mF cm−2 ) (Ω cm2 ) (H cm2 )
A 0.88 61860 34.15 0.91 36.94 41580 0.191 0.85 0.28 - -
B 0.65 0.90 16.84 0.99 16.84 90.13 0.72 0.81 0.39 30.36 18.35
C 0.51 0.27 781 0.80 94.1 5.45 6.03 0.79 0.49 11.31 -

The strong passivation of tin in Sol. A is confirmed by the high values of Rf and Rct (Table 4).
As the electrolyte become more acidic, the resistance of the thin SnO2 layer formed on Sn surface
Rf and the charge transfer resistance of the underlying metal Rct significantly decrease. In parallel,
important increases of Cf and Cdl values were observed in Sol. C. These results correlate with a
passivity breakdown in strong acidic electrolytes [50], leading to higher corrosion rates and much
rougher surfaces with enlarged areas. The presence of 0.01 M Br2 additionally accelerates the active Sn
dissolution, in accordance to the outlined conclusions of the potentiodynamic measurements.

3.2.4. Lead
Figure 6 presents the impedance spectra measured on Pb electrode exposed to different
bromide-based electrolytes and the electrical equivalent circuits used for data simulation.
The development of more than one time constant could be noticed in all impedance spectra
corresponding to Pb corrosion (Figure 6). The Bode plot for Pb dissolution in Sol. A shows two maxima
in the phase, which were interpreted according to the equivalent circuit in Figure 6b. The first time
constant at high frequencies was ascribed to the double layer capacitance Qdl connected in parallel
with the charge transfer resistance Rct , while the second one corresponds to some adsorbed species
that contributes to the formation of the corrosion layer [51]. An additional Warburg tail associated to a
diffusion process at the metallic surface was noticed in the impedance diagram of Pb exposed to Sol. B.
Accordingly, the corresponding EIS data were interpreted in terms of the equivalent circuit depicted in
Figure 6b and the calculated R-Q parameters are shown in Table 5.

Table 5. Electrochemical parameters of Pb dissolution in bromide-containing solutions.

Re Rct Qdl Cdl R1 Q1 C1 W

Sol. ndl n1
(Ω cm2 ) (Ω cm2 ) (µF sn−1 cm−2 ) (µF cm−2 ) (Ω cm2 ) (mF sn−1 cm−2 ) (mF cm−2 ) (S sec1/2 cm−2 )
A 0.94 19.63 210.0 0.84 74.16 1096 1.91 0.50 3.9 -
B 0.58 0.87 285.0 0.88 91.9 125.7 1.58 0.68 0.25 0.06129
D 0.55 0.80 565.0 0.87 170.0 13.39 2.15 0.78 0.79 -

As expected, the fast dissolution of Pb in strong acidic brominated electrolytes was confirmed by
the low values of Rct and R1 (Table 5). The attempts to simulate the EIS data obtained for Pb corrosion
in solutions containing Br2 with different electrical equivalent circuits did not allow a satisfactory
agreement between the used models and experimental results. A possible explanation could be related
to the complexity of the electrochemical processes taking place at Pb surface in the presence of Br2 ,
leading to the formation of various compounds, i.e., Pb3 O4 and PbBr2 , which complicate the corrosion
mechanism. However, the decrease of the overall impedance and phase angle in Sol. C noticed in
Figure 6 correlates with an important increase of Pb corrosion rate in brominated electrolytes.

3.2.5. Iron
The Nyquist diagram of Fe exposed to Sol. A shows a wide depressed capacitive contribution,
which was interpreted according to the Randles circuit model in Figure 7b [52]. Instead, two time
constants are clearly visible in the Bode plots obtained in Sol. B and Sol. D, suggesting that the
mechanism of Fe corrosion in strong acidic electrolytes might involve one adsorbed intermediate
Materials 2020, 13, 3630 17 of 28

(i.e., Fead + ) covering the electrode surface [53]. The relaxation process of the adsorbed intermediate
was represented by R1 −Q1 circuit nested in the typical Rct −Qdl network, as revealed in Figure 7c.
An inductive loop in the low-frequency domain could be observed for Fe corrosion in Sol. C.
It could originate from the relaxation process of some adsorbed intermediate species formed on Fe
surface [54]. The equivalent circuit in Figure 7d, containing the RL −L inductive elements, was used for
these data simulations. It should be noted that similar impedance results were obtained in the case of
Fe dissolution in NaCl solutions of pH = 0, 1, 2, 3, 4 [55].
Table 6 presents the values of R-Q parameters obtained by fitting the experimental data corresponding
to Fe corrosion in different bromide-based electrolytes to the equivalent electrical circuits in Figure 7.

Table 6. Electrochemical impedance parameter values for the corrosion of Fe in various bromide-based
electrolytes.

Re Rct Qdl Cdl R1 Q1 C1 RL L

Sol. ndl n1
(Ω cm2 ) (Ω cm2 ) µF sn−1 cm−2 ) (µF cm−2 ) (Ω cm2 ) µF sn−1 cm−2 ) (µF cm−2 ) (Ω cm2 ) (H cm2 )
A 1.00 42360 116 0.83 155 - - - - - -
B 0.79 59.60 202 0.77 54.0 1594 48.0 0.91 37.7 - -
C 0.69 9.15 45.1 0.90 18.25 234.7 238 0.70 69.1 116.2 36.91

As expected, the electrochemical dissolution of Fe is enhanced by the decrease of solution pH,

as revealed by the low values of Rct and R1 obtained in Sol. B (Table 6). The addition of Br2 further
accelerates the corrosion process, i.e., Rct and R1 values are more than six times lower in Sol. C as
compared to Sol. B.

3.3. Ex-Situ Examinations

3.3.1. XRD Measurements

The XRD spectra of Cu, Zn, Sn, Pb and Fe obtained after 120-h immersion in different bromide-based
electrolytes are shown in Figure 8.
Prior to XRD measurements, the above-mentioned metallic samples were mechanically and
chemically cleaned to remove any traces of corrosion from their surface. XRD diffractograms for all
cleaned samples (pure curves in Figure 8) show that they are pure, without traces of oxides or other
compounds on their surface.
The long immersion of the metals in different bromide-based electrolytes induces structural
modifications for all electrodes, as shown in their XRD diffractograms. In the case of Cu electrode
immersed in Sol. A and Sol. B, the XRD diffractograms shows a combination of CuO and Cu2 O
phases. By exposing the Cu electrode to Sol. C, the corrosion products change their composition and
XRD diffractograms show only peaks related to the Cu2 O phase. In the case of Sn, the additional
diffraction peaks appearing in the diffraction pattern after 120 h were assigned to SnO2 and their
intensity increases in Sol. C. In the case of Pb the influence of the Sol. C is more obvious. The peaks in
the XRD diffractograms were indexed with red arrows (for PbBr2 phase) and black arrows (for Pb3 O4
phase). The predominance of PbBr2 becomes visible.
The XRD diffractograms of Zn and Fe after immersion in Sol. A and Sol. B reveal the presence
of corrosion products, i.e., Fe3 O4 and ZnO, respectively. Instead, the XRD diffractograms for Fe
immersion in Sol. C show only the peaks corresponding to KBr.
From the XRD diffractograms obtained on the metal electrodes, a preferential orientation tendency
of the crystallites of the corrosion products was observed according to certain crystallization planes.
To study the effect of dissolution of metals in different bromide-based systems on the preferential
orientation of crystallites, we assessed the X-ray diffractograms obtained on samples exposed to these
solutions for 120 h. The relative values of the intensities of the diffraction peaks for all samples are
presented in Table 7.
Materials 2020, 13, 3630 18 of 28

Table 7. Preferential orientation tendency of the crystallites of the corrosion products in different
bromide-based electrolytes.

Relative
Sample Sol. A Sol. B Sol. C
Intensities Ratio
Cu I(111) /I(220) 1.01 1.05 1.10
Zn I(002) /I(102) 1.03 1.10 -
Sn I(110) /I(211) 0.97 1.01 1.05
Pb I(110) /I(211) 0.99 1.01 0.98
Fe I(001) /I(201) 1.10 1.12 -

In the case of Cu and Sn electrodes, a preferential orientation of the crystallites of the corrosion
products is observed according to the diffraction planes (111) for Cu2 O and (110) for SnO2 . The increase
in the intensity ratios after these diffraction planes indicates that the corrosion products become more
stable [56], in agreement with potentiodynamic polarization and EIS measurements. In the case of Pb
electrodes, the small variation of the I110 /I211 ratio of the intensities around 1 for Pb3 O4 is probably due
to the fact that the formation of the corrosion product PbBr2 is more accentuated, as can be seen in
Figure 8c.
For the Fe and Zn electrodes, a preferential orientation on the c direction of the crystallites of the
corrosion products is observed, according to the diffraction planes (001) for Fe3 O4 and (002) for ZnO.
The decrease in the solution pH significantly influences the Fe and Zn dissolution process, preventing
the formation of oxidized species on the metallic surface, consistent with EIS measurements.

3.3.2. XPS Measurements

To get deeper insight into the corrosion processes taking place on the metallic surfaces after
60–120 min immersion in Sol. C, X-ray photoelectron spectroscopy was used. XPS might act as a
valuable tool for studying the initial stages of corrosion, when the layer of formed corrosion products
may be too thin for bulk sensitive methods such as for X-ray diffraction.

Cu Samples
The elements identified in the survey scans of Cu samples are Cu, O, Br, K, C and Si. As expected,
sample degradation due to X-ray exposure was observed (Figure 9). The sample degradation was
measured by repeating the Cu 2p narrow scan acquisition for 10 iterations and subsequently fitting the
spectra. Figure 9 shows the changes in the Cu 2p narrow scan data due to sample degradation and the
fit of the initial dataset (Iteration 0). The peak shift observed on the left side of the figure is ascribed to
charging effects. At least two contributions are required to fit the data, which were ascribed to Cu2+
and Cu1+ , respectively.
Materials 2020, 13, x FOR PEER REVIEW 20 of 29

2.1
5.5
0
5.0
2.0
residuals (*1E3 CPS)

scan10
intensity (*1E5 CPS)

4.5 Cu II
scan 9 -10
4.0 1.9
intensity (a.u.)

scan 8 Cu II satellites
3.5
scan 7
3.0 1.8 -20
scan 6 Cu I
2.5
scan 5
2.0 1.7 -30
scan 4
1.5
scan 3 1.6
1.0
scan 2 -40
0.5
scan 1 1.5
0.0
-50
960 950 940 930 920 950 945 940 935 930 925
binding energy (eV) binding energy (eV)
(a) (b)
Figure
Figure 9. Coppersample
9. Copper sampledegradation
degradation (a)
(a)and
andfitfitofof
thethe
initial narrow
initial scanscan
narrow (b). (b).

Table 8. Cu2p3/2 fit parameters (referenced to adventitious carbon C1s peak at 285.1 eV).

Peak Iteration B.E. (eV) FWHM (eV)

Cu 2p3/2 Cu (II) 0 934.2 3.1
Cu 2p3/2 Cu (I) 9 932.6 2.1
Materials 2020, 13, x FOR PEER REVIEW 20 of 29

Materials 2020, 13, 3630 2.1 19 of 28

5.5
0
5.0
2.0

residuals (*1E3 CPS)

scan10

intensity (*1E5 CPS)

4.5 Cu II
Table 8 gives the binding energiesscan
4.0
of the
9 two contributions and their attributed oxidation-10 states.
1.9
intensity (a.u.)
The initial spectrum shows satellite scan 8at the following binding
features Cu II energies:
satellites 940.5, 943.9 and 962.1 eV,
3.5
2+ scan 7 2+
which3.0 are specific to Cu state [57]. The Cu 2p3/2 contribution
1.8 associated with Cu was observed at-20
scan 6 Cu I
934.22.5eV. The Cu 2p3/2 contribution associated with Cu 1+ was observed at 932.6 eV, in agreement with
scan 5
2.0 1.7
formerly reported results [58–60]. Withscan the4 prolonged X-ray exposure, Cu2+ satellite features vanish.
-30
1.5
The intensity
1.0 of the Cu 2p3/2 –Cu2+ peakscan
decreases.
3 Conversely,
1.6 the intensity of the Cu 2p3/2 –Cu1+ peak
-40
increases.
0.5 This indicates the reduction of the Cu component. The degradation of Cu2+ compounds
scan 2 2+
scan 1 1.5
0.0 XPS measurements is a well-known occurrence [44,61].
during -50
960 950 940 930 920 950 945 940 935 930 925
Table 8.binding
Cu2p3/2 energy (eV) (referenced to adventitious binding
fit parameters energy
carbon C1s peak at(eV)
285.1 eV).
(a) (b)
Peak Iteration B.E. (eV) FWHM (eV)
Figure 9. Copper sample degradation (a) and fit of the initial narrow scan (b).
Cu 2p3/2 Cu (II) 0 934.2 3.1
Cu 2p3/2 Cu (I) 9 932.6 2.1
Table 8. Cu2p3/2 fit parameters (referenced to adventitious carbon C1s peak at 285.1 eV).

Peak Iteration B.E. (eV) FWHM (eV)

The ratio of the two oxides was estimated by adding the contribution of the Cu2+ satellite peaks
to the contributionCu 2p3/2main
of the Cu (II) 0
Cu2+ peak. Table 9 shows934.2
the quantification3.1
results.
Cu 2p3/2 Cu (I) 9 932.6 2.1
Table 9. Cu oxide ratios.
Table 9. Cu oxide ratios.
Level Cu2+ (at %) Cu1+ /Cu0 (at %)
Level Cu2+ (at %) Cu1+/Cu0 (at %)
Iteration 0 (“as received” sample) 83.7 16.3
Iteration 0 (“as
Iteration 9 received” sample) 25.5 83.7 16.3
74.5
Iteration 9 25.5 74.5

The O1s
The O1s peak
peakpresents
presentsthreethreecomponents
components which
which cancan be ascribed
be ascribed to organic
to organic contaminations
contaminations and
and to metallic oxides, respectively. The component ascribed to metallic oxides
to metallic oxides, respectively. The component ascribed to metallic oxides is observed at 530.1 eV, in is observed at
530.1 eV, in accordance with the values previously reported for Cu 2 O, which
accordance with the values previously reported for Cu2O, which are in the range of 530.2–530.7 eV are in the range of
530.2–530.7 eV [62,63].
[62,63].
The presence
The presence of of KK along
along with
with BrBr could
could indicate
indicate thethe presence
presence of of recrystallized
recrystallized KBr
KBr traces
traces on
on the
the
sample. The measured B.E. of
sample. The measured B.E. of the K2p3/2the K2p and Br3p
3/2 and Br3p3/2
electrons are consistent with the expected
3/2electrons are consistent with the expected values values
for KBr. The K2p 3/2 peak is observed at
for KBr. The K2p3/2 peak is observed at 292.9–293.0 eV292.9–293.0 eV and
and the
the Br3p
Br3p3/2 peakisisobserved
3/2peak observedatat182.1
182.1eV.
eV.
Typical values reported in the literature are in the range of 292.9–293.1 eV for
Typical values reported in the literature are in the range of 292.9–293.1 eV for K in KBr [64,65] and K in KBr [64,65] and
182.1 eV
182.1 eV for
forBrBrininNaBr
NaBr[66].
[66].Depth
Depth profile
profileexperiments
experiments (Figure 10) 10)
(Figure showed
showeda parallel variation
a parallel of peak
variation of
intensity for the K2p 3/2 and the Br3p 3/2 photoelectrons with the etching time.
peak intensity for the K2p3/2 and the Br3p3/2 photoelectrons with the etching time. This further This further indicates the
presence of
indicates theKBr.
presence of KBr.

9000

8000 K2p3
Br3p3
7000
intensity (CPS)

6000

5000

4000

3000

2000

1000

0 50 100 150 200 250 300

etch time (s)
Figure
Figure 10.
10. Variation
Variation of
of K
K (black)
(black) and Br (red) peak intensity with etching time.
Materials 2020,
Materials 13,13,
2020, 3630
x FOR PEER REVIEW 21 20 of 28
of 29

Zn Samples
Zn Samples
The main elements identified in the case of the Zn sample are Zn, K, Br, C and O. The C1s peak
of The main elements
the adventitious identified
species in the case
was observed of the
at 285.0 ZnZn
eV. sample are Zn,
is observed in K,
theBr,
ZnC2+ and O. The
oxidation C1son
state peak
the of
the“as
adventitious species was observed at 285.0 eV. Zn is observed in the Zn 2+ oxidation state on the
received” sample. The Zn 2p3/2 peak appears at 1022.5 eV, in agreement with previously reported
“asvalues
received” sample.
for ZnO The Zn2 [44,67].
or Zn(OH) 2p3/2 peak appears
During depthatprofile
1022.5 experiments,
eV, in agreement withshape
the peak previously
of the reported
Zn 2p
values
and offorthe
ZnOZnor Zn(OH)
LMM 2 [44,67].
electrons During
indicates thedepth profile
emergence ofexperiments,
metallic Zn. No themetallic
peak shapeZn isofobserved
the Zn 2p
and of the
before Zn LMM electrons indicates the emergence of metallic Zn. No metallic Zn is observed
sputtering.
Figure 11 shows the narrow scans for the two types of electrons. The XPS electrons are shown
before sputtering.
after 150 s11
Figure of shows
etching.the
Thenarrow
X-ray induced
scans forAuger electrons
the two typesare presented after
of electrons. 120 selectrons
The XPS of etching.are
In shown
both
examples,
after 150 s of oxide andThe
etching. metallic
X-raycontributions
induced Auger are clearly visible.
electrons This highlights
are presented the spresence
after 120 of the
of etching. In 2+
both
oxide state before etching.
examples, oxide and metallic contributions are clearly visible. This highlights the presence of the 2+
Table
oxide state 10 shows
before the B.E. and the K.E. values corresponding to the two oxidation states.
etching.

4.0 6
10
3.5 0
intensity (*1E5 CPS)

residuals (*1E3 CPS)

intensity (*1E4 CPS)

residuals (*1E3 CPS)

0 ZnLM2 metallic Zn
3.0 5 2+
ZnLM2 Zn
metallic Zn -10 -5
2.5 2+
Zn
-20
2.0 experimental
4
Zn2p3 metallic
Zn2p3 Zn
2+ -10
1.5 background -30
envelope
residuals
1.0 0 mark -40
3 -15

0.5 -50

1026 1025 1024 1023 1022 1021 1020 1019 1018 -20
binding energy (eV) 982 984 986 988 990 992 994
binding energy (eV)

Figure 11. Zn 2p narrow scan (left), Zn LMM narrow scan (right).

Figure
Table 10 shows the B.E. 11.
and Znthe
2p K.E.
narrow scan (left),
values Zn LMM narrow
corresponding to thescan
two(right).
oxidation states.

Table 10. B.E. and K.E. values for Zn peak contributions/oxidation states.
Table 10. B.E. and K.E. values for Zn peak contributions/oxidation states.
Peak Oxidation State Measured B.E. (eV) K.E. (eV) FWHM (eV)
Peak Oxidation State Measured B.E. (eV) K.E. (eV) FWHM (eV)
Zn 2p3/2 metallic 1021.8 - 1.0
Zn 2p3/2
Zn 2p3/2 metallic
Zn2+ 1021.8
1022.5 - - 1.4 1.0
Zn 2p3/2 Zn2+ 1022.5 - 1.4
Zn LMM metallic - 991.9 1.2
Zn LMM metallic - 991.9 1.2
Zn
Zn LMM
LMM Zn
2+
2+ -- 987.5987.5 3.0 3.0
Zn

The measured ratios of Zn:O could be useful in assigning the chemical composition. The Znoxide:
O The measured
atomic ratios of This
ratio is 49.7:50.3. Zn:Oiscould
close be useful
to the in assigning
1:1 expected ratiothe
forchemical
ZnO. composition. The Znoxide :
O atomic ratio is 49.7:50.3. This is close to the 1:1 expected ratio for ZnO.
Figure 12 shows the O1s scan after 270 s of etching. The peak at 531.0 eV was attributed to Zn
Figure
oxide, 12 shows with
in accordance the O1s
otherscan afterdata
reported 270[44,67].
s of etching.
The peaks Thefrompeak atand
532.2 531.0 eVeV
533.1 was attributed
correspond
to to
Znadventitious organic species.
oxide, in accordance withThe peakreported
other intensitydata
in the[44,67].
case of the
Theorganic
peakscomponents
from 532.2isand
decreasing
533.1 eV
rapidly with sputtering, although the peaks remain detectable throughout the entire depth
correspond to adventitious organic species. The peak intensity in the case of the organic components is profile
experiment.
decreasing rapidly with sputtering, although the peaks remain detectable throughout the entire depth
profile experiment.
K and Br are also observed for the Zn sample, but the two elements are removed from the sample
at the same rate during the depth profile experiment. It could therefore be inferred that they belong to
the same compound, namely KBr.
Materials 2020, 13, x FOR PEER REVIEW 22 of 29

10

5.0

CPS)(*1E4 CPS)
Materials 2020, 13, 3630 21 of 28

residuals (*1E2
0
Materials 2020, 13, x FOR PEER REVIEW 22 of 29
4.5 metallic oxide
experimental -10
10
O1s oxide
O1s organic
5.0
4.0 O1s organic organic species
background -20

residuals
0

intensity
envelope
residuals
0 mark
4.5
3.5 metallic oxide -30
-10

intensity (*1E4
experimental

CPS)(*1E2 CPS)
O1s oxide
O1s organic
4.0
3.0
O1s organic organic species -40
-20
background
envelope
residuals
0 mark -50
3.5 -30
536 534 532 530 528
binding energy (eV)
3.0 -40
Figure 12. O1s narrow scan, Zn sample.
-50
536 for the Zn
K and Br are also observed 534 sample,532
but the two530
elements528
are removed from the sample
binding energy (eV)
at the same rate during the depth profile experiment. It could therefore be inferred that they belong
to the same compound, namely KBr. Figure12.
Figure 12. O1s
O1s narrow
narrow scan,
scan,Zn
Zn sample.
sample.

Sn Samples
Sn Samples
K and Br are also observed for the Zn sample, but the two elements are removed from the sample
at theInsame
In rateofduring
the case Snthe
the Sn depththe
sample,
sample, profile
the mainexperiment.
main elements It couldin
elementsidentified
identified therefore
inthe
theXPS
XPSbe inferred
survey
survey thatare
spectra
spectrathey belong
Sn,Sn,
are O,O,
K,
to
K, the
Br Br
and same
C. C.
and Thecompound,
SnSn
The namely
3d3d5/2narrow KBr.
narrowscan
scanpresented
presentedininFigure
Figure1313shows
showstwotwowell
well separated
separated peaks. These
peaks. These
5/2
peaks are identified as corresponding to the metallic and Sn4+
peaks states,respectively,
4+ states, respectively, according
according to
to their
their
Sn Samples
B.E values
values (Table
(Table 11).
B.E
In the case of the Sn sample, the main elements identified in the XPS survey spectra are Sn, O,
10
K, Br and C. The Sn 3d5/2 1.6
narrow scan presented in Figure 13 shows two well separated peaks. These
peaks are identified as corresponding to the metallic and Sn 4+ states, respectively,
0 according to their residuals (*1E3
1.4
CPS)(*1E5 CPS)

B.E values (Table 11).

4+
1.2 experimental Sn -10
Sn3d5 metallic
Sn3d5 Sn
4+ 10
1.0
1.6 background -20
envelope
residuals 0
residuals

0.8
1.4 0 mark
-30
intensity

metallic
4+ Sn
0.6
1.2 experimental Sn -10
Sn3d5 metallic
-40
CPS)(*1E3 CPS)
intensity (*1E5

4+
Sn3d5 Sn
0.4
1.0 background -20
envelope
0.2 residuals -50
0.8 0 mark
-30
metallic Sn
0.0
0.6 -60
490 488 486 484 -40
0.4 binding energy (eV)
0.2 Figure 13. -50
Sn 3d5/2 narrowscan,
5/2narrow scan,Sn
Snsample.
sample.
0.0 -60
Table 11. B.E. values for Sn (with
490reference488
the C1s peak due to adventitious C which is observed
Table 11. B.E. values for Sn (with reference the C1s peak486 484
due to adventitious C which is observed at
at 285.0 eV).
285.0 eV).
binding energy (eV)
Oxidation Measured Reported
Peak OxidationState Figure 13. Sn 3dMeasured
Compound 5/2 narrow scan, Sn sample.
B.E. (eV) Reported
B.E. (eV) FWHM
FWHM (eV) References
Peak Compound References
Sn
State B.E. (eV) B.E. (eV)
3d5/2 11. B.E. values for Sn (with reference the C1s peak due to adventitious C which is observed
metallic 0 Sn 485.0 485.0 [68] at
(eV)
0.6
Table
SnSn3d
3d5/2
5/2 oxidized 4+ SnO2 486.8 486.6 1.2 [65]
285.0 eV). 0 Sn 485.0 485.0 0.6 [68]
metallic
SnFigure
3d5/2 Oxidation Measured Reported FWHM from organic
Peak 14 shows4+the O1s peak for2 the Sn sample
Compound
SnO 486.8after 15 s of 486.6
etching. Contributions
1.2 References
[65]
oxidized State B.E. (eV) B.E. (eV)
compounds and the inorganic oxide are initially observed. The O1s peak attributed (eV) to metallic
Sn 3d
oxides was5/2
observed 0at 530.4 eV. During the depth profile, the organic
Sn 485.0 485.0 contributions
0.6 vanish[68]
rapidly.
metallic 4+
The inorganic contribution remains visible as long as the Sn peak remains visible. This behavior is
Sn 3d5/2with the presence of Sn oxide on the surface. Because Sn was found in Sn4+ state, it was
consistent 4+ SnO2 486.8 486.6 1.2 [65]
oxidized
assumed that the sample surface contains SnO . 2
 wasFigure
observed at 530.4
14 shows theeV.
O1sDuring thethe
peak for depth profile,after
Sn sample the 15
organic contributions
s of etching. vanish from
Contributions rapidly. The
organic
inorganic contribution
compounds remains
and the inorganic visible
oxide as longobserved.
are initially as the SnThe4+ peak remains visible. This behavior is
O1s peak attributed to metallic oxides
consistent
was observed withat the presence
530.4 of Snthe
eV. During oxide on the
depth surface.
profile, the Because Sn was found vanish
organic contributions in Sn 4+ rapidly.
state, it was
The
assumed contribution
inorganic that the sample surfacevisible
remains contains
as SnO
long2. as the Sn4+ peak remains visible. This behavior is
consistent
Materials 2020,with the presence of Sn oxide on the surface. Because Sn was found in Sn 4+ state, it
13, 3630 22 was
of 28
10
assumed that the sample surface contains SnO2.

residuals
0

CPS) CPS)
9 experimental
10 O1s metal oxide
O1s metal oxide
-20

residuals
8 O1s organic
0

(*1E4 (*1E4
9 O1s organic
experimental
background O1s metal oxide -40

(*1E2 CPS)
O1s metal oxide
7 envelope -20
8 O1s organic
residuals

intensity
O1s
zeroorganic
mark -60
6 background -40

(*1E2 CPS)
7 envelope
O1s organic species
residuals
-80

intensity
5 zero mark -60
6 -100
4 O1s organic species -80
5 -120
536 534 532 530 528-100
4 binding energy (eV)
-120
536 Figure 14. O1s narrow
534 532 scan, 530
Sn sample. 528
binding energy (eV)
Pb Samples
Figure 14. O1s narrow scan, Sn sample.
The main elements observed in the XPS survey spectra of the Pb sample were only Pb, Br, C and
Pb
Pb Samples
O.Samples
The C1s peak is observed at 285.0 eV. C and O are ascribed only to the adventitious organic species.
TheTheintensity
main of these peaks
elements is decreasing
observed in the
the XPSrapidly
XPS surveyand at the of
spectra same
the rate, suggesting
Pb sample
sample werethat
onlyboth elements
Pb, Br,
Br, C
C and
and
The main elements observed in survey spectra of the Pb were only Pb,
belong
O. The
The C1sto the
C1s peak same
peak is compounds.
is observed
observed at at285.0
285.0eV.eV. C
C and
and O Oare
areascribed
ascribedonly onlytotothe
theadventitious
adventitiousorganic
organicspecies.
species.
O.
The Figure of
intensity 15 these
shows the is
peaks narrow scan rapidly
decreasing on the and
Pb 4fat7/2the
peak.
same The
rate,two observedthat
suggesting components
both are
elements
The intensity of these peaks is decreasing rapidly and at the same rate, suggesting that both elements
attributed
belong to to theto metallic
the same
same compounds.Pb
compounds.and Pb 2+, respectively. The B.E. of Pb 4f 7/2 peak observed at 139.0 eV is
belong
consistent with the presence of PbBr2 on the surface of the Pb sample, in agreement with previously
Figure 15
Figure 15shows
showstheThethenarrow
narrowscanscanon the
onPb 4f7/2
the Pbpeak.
4fat The two observed components are attributed
7/2 peak. The two observed components are
reported data [69,70]. Br 3d5/2 peak
2+ , respectively. was observed 68.8 eV. When Br is forming PbBr2, a similar
to metallic
attributed Pb and
to metallic Pb Pb and Pb eV
2+ The B.E.
, respectively. of Pb 4f
The B.E. peak
7/2 of Pb observed
4f7/2 peak at 139.0 eVatis139.0
observed consistent
eV is
Br 3d5/2 peak was reported at 68.7 [71].
with the presence
consistent with the ofpresence
PbBr2 on of the
PbBr surface of the Pb sample, in agreement with previously reported
2 on the surface of the Pb sample, in agreement with previously
Table 12 shows the experimentally determined B.E. values for the two Pb components and the
data [69,70].
reported The Br 3dThe5/2 peak was observed at 68.8 eV. When Br is forming PbBr2 , a similar Br 3d5/2
assigneddata [69,70].
oxidation states.Br 3d5/2 peak was observed at 68.8 eV. When Br is forming PbBr 2, a similar
peak
Br 3d5/2 peak was reportedeV
was reported at 68.7 at [71].
68.7 eV [71].
Table 12 shows the experimentally determined B.E. values for the two Pb components and the
4.5
assigned oxidation states. 10 9
4.0 0
residualsresiduals

0 8
intensity (*1E5 CPS)

experimental
(*1E4 CPS)

3.5 -10 Br3d3 Br3d5 -5

residualsresiduals
2+ Pb4f7 metallic
Pb 7
4.5 Pb4f7 Pb
2+
3.0 -20
10 9 experimental -10
background 6
4.0 Br3d5 0
2.5 envelope 0-30 8
intensity (*1E5 CPS)

Br3d3 -15
(*1E3 CPS)

residuals
experimental 5 background
(*1E4 CPS)

3.5 2+ 0 mark
Pb4f7 metallic -40
-10 Br3d3 Br3d5 -5
2.0 Pb 7 envelope
-20
(*1E3 CPS)

2+
Pb4f7 Pb -50 4 residuals
3.0 -20
intensityintensity

1.5 background experimental

0 mark -10
-60 6 Br3d5 -25
2.5 envelope
metallic Pb -30 3 Br3d3 -15
(*1E3 CPS)

1.0 residuals 5
-70
-40 background -30
2.0 0 mark 2 envelope
0.5 -80 -20
(*1E3 CPS)

-50 4 residuals -35

1.5 1 0 mark
0.0 -90
-60 -25
3 -40
1.0 142 141 140 139 138metallic
137 Pb136 135
72 70 68 66 -30
-70
binding energy (eV) 2 binding energy (eV)
0.5 -80 -35
1
0.0 -90 (left); and Br 3d narrow scan (right).
Figure 15. Pb 4f7/2 narrow scan
142 141 140 139 138 137 136 135 -40
72 70 68 66
binding energy
Figure 15. Pb(eV)
4f7/2 narrow scan (left);B.E.
and values
Br 3d narrow scan
binding (right).
energy (eV)
Table 12 shows the experimentally determined for the two Pb components and the
assigned oxidation states.

Figure 15. Pb 4f7/2 narrow scanB.E.

Table 12. (left); and for
values Br 3d
Pb.narrow scan (right).

Peak Oxidation State B.E. (eV) FWHM (eV)

Pb 4f7/2 metallic 0 137.0 1.0
Pb 4f7/2 oxidized 2+ 139.0 1.1
 Materials 2020, 13, x FOR PEER REVIEW 24 of 29

Table 12. B.E. values for Pb.

Peak Oxidation State B.E. (eV) FWHM (eV)

Materials 2020, 13, 3630 Pb 4f7/2 metallic 0 137.0 1.0 23 of 28
Pb 4f7/2 oxidized 2+ 139.0 1.1

The
ThePb:Br
Pb:Brratio
ratio determined
determined fromfrom the
the XPS
XPS data is in
data is in the
the range
range of 35.6:64.4 ≈≈1:1.81,
of 35.6:64.4 1:1.81,which
whichisisclose
close
to the expected 1:2 ratio. It should be noted that the occurrence of PbBr on
to the expected 1:2 ratio. It should be noted that the occurrence of PbBr 22on lead surface was alsolead surface was also
confirmed
confirmedby byXRD
XRDmeasurements.
measurements.
The featureless residueobtained
obtainedwith
withonly
only two components, 2+ peak
The featureless residue two components, thethe relatively
relatively narrow
narrow Pb2+ Pb
peak and
and
the the rapid
rapid decrease
decrease of the
of the O1sO1s peak
peak intensity,suggests
intensity, suggeststhe theabsence
absenceofofany
anyother Pb2+2+containing
other Pb containing
compounds.
compounds.This Thisisisinin
contrast with
contrast XRD
with data,
XRD which
data, indicated
which the occurrence
indicated of Pb3of
the occurrence O4 Pb
on3Othe sample,
4 on the
along with PbBr
sample, along with 2 . The absence of Pb
PbBr2. The absence O in the XPS spectra could be the result of the different
3 4of Pb3O4 in the XPS spectra could be the result of the different scanning
depth
scanning and/or
depthdifferent
and/orscanning
different area typical
scanning fortypical
area the twofortechniques. The metallic
the two techniques. component
The metallic of the Pb
component
4fof7/2the
peak is probably the effect of sample reduction due to the Ar sputtering.
Pb 4f7/2 peak is probably the effect of sample reduction due to the Ar sputtering.

Fe
FeSamples
Samples
The
Theelements
elements identified
identified in
in the
the XPS
XPS spectra
spectra for the Fe
for the Fe sample
sample are
are Fe,
Fe, O,
O, K,
K, Br
Brand
andC.
C.
Figure 16 shows the Fe 2p 3/2 narrow scan spectrum, collected on the sample,
Figure 16 shows the Fe 2p3/2 narrow scan spectrum, collected on the sample, before before (left) andand
(left) after
15 s of etching (right).
after 15 s of etching (right).

1.2 10
2.5 0
0 experimental

residuals (*1E2 CPS)

Fe2p3 metallic
residuals (*1E2 CPS)

3+
-10
intensity (*1E4 CPS)
3+ Fe 2+ 3+
intensity (*1E4 CPS)

1.0 Fe Fe Fe Fe
2+ Fe2p3 multiplet
-10 2.0 satellite satellite Fe2p3 Fe
2+
2+ -20
3+
Fe satellite Fe 2+
Fe2p3 Fe satellite
-30 Fe2p3 Fe
3+
Fe2p3
0.8 metallic Fe -20 3+
Fe2p3 Fe satellite
1.5 -40
multiplet Fe2p3 background
metallic -50 envelope
-30 residuals
0.6 Fe2p3 -60
1.0 Fe2p3 0 mark
metallic Fe
2
Fe satellite metallic -70
-40 multiplet
Fe 2p1
metallic Fe Fe2p1 -80
0.4 0.5
metallic
-50 -90
720 715 710 705 700 720 715 710 705 700
binding energy (eV) binding energy (eV)
(a) (b)
Figure 16.
Figure 16. Fe 2p3 narrow scan, initial spectrum
spectrum (a)
(a) and
and spectrum
spectrumafter
after15
15ssof
ofetching
etching(b).
(b).

The
Thenarrow
narrowscans
scansshowed
showed thatthat Fe
Fe was
was found in three oxidation states, namelynamely metallic, Fe2+2+and
metallic, Fe and
Fe 3+ Before etching,
etching, the
the spectra
spectra is is dominated
dominated by by the
the oxide
oxidecontributions.
contributions.After
Afteretching,
etching,the
themetallic
metallic
Fe . Before
3+

component
componentbecomes
becomes dominant.
dominant. The metallic component
component shows
shows multiplet
multipletsplitting.
splitting.The
Theoxide
oxidephases
phases
show characteristic satellite features.
show characteristic satellite features. The chemical states are identified according to the B.E. at
chemical states are identified according to the B.E. at which which
they
theyareareobserved
observed and
and by
by their
their satellite
satellite fingerprint. The
The observed
observedresults
resultsagree
agreewith
withthe
thepreviously
previously
reportedvalues
reported values[44,72,73].
[44,72,73].
Table13
Table 13shows
showsthetheB.E.
B.E.of
of Fe
Fe components
components and their respective oxidation
oxidation states.
states.
The O1s peak shows contributions from organic species and inorganic oxides. The two types of
contributions can be separated based Table B.E. values
on13.their for Fe.
response to the Ar ion etching. The organic
contributions vanish more quickly than the contributions associated to the Fe oxides. As expected,
Oxidation Measured B.E. FWHM Reported
the presencePeak
of KBr on the metallic surface(eV)
was also evidenced. References
State (eV) B.E. (eV)
Fe 2p3/2 metallic 0 706.8 1.0 706.8 [44,72–74]
0.9
Fe 2p3/2 multiplet 0 707.7 1.0 [44,72,73]
(from main peak)
Fe 2p3/2 (II) oxide 2+ 709.8 3.5 709.6–709.9 [44,72,73,75]
Fe 2p3/2 (II) satellite 2+ 716.1 3.5 - -
Fe 2p3/2 (III) oxide 3+ 712.7 3.5 710.8–711.4 [44,72,73,76]
Fe 2p3/2 (III) satellite 3+ 719.3 3.5 - -

The O1s peak shows contributions from organic species and inorganic oxides. The two types of
contributions can be separated based on their response to the Ar ion etching. The organic contributions
 Fe 2p3/2 multiplet 0 707.7 1.0 (from main
1.0 (from main [44,72,73] peak)
peak) Fe 2p3/2 (II) oxide 2+ 709.8 3.5 709.6–709.9
3.5 709.6–709.9 [44,72,73,75] Fe 2p3/2 (II)
2+ 716.1 3.5 -
3.5 - 2020, 13, 3630 - satellite
Materials 24 of 28
Fe 2p3/2 (III)
3+ 712.7 3.5 710.8–711.4
3.5 710.8–711.4 [44,72,73,76] oxide
Fe 2p3/2 (III)
vanish more quickly than the contributions associated to the Fe oxides. As expected, the presence of
3+ 719.3 3.5 -
3.5 KBr on - the metallic surface
- was alsosatellite
evidenced.

4. Conclusions 4. Conclusions
Based on the obtained results, some
Basedconclusions can be
on the obtained drawn:some conclusions can be drawn:
results,
an be drawn:
• Electrochemical measurements  showed that themeasurements
Electrochemical addition of bromineshowedtothat thethe
system favors
addition of to a
bromine to the syste
addition of bromine togreatthe system favors to a
extent the dissolution process
greatof extent
all studied
the metals as compared
dissolution processto ofbromine-free
all studied electrolytes.
metals as compared to
udied metals as compared to bromine-free
Better understanding of the reaction mechanisms
electrolytes. Betterand understanding
evaluating the electrochemical
of the reaction parameters
mechanisms and eva
eaction mechanisms ofand evaluating the
the systems are necessary before any technological
electrochemical approach.
parameters of theThe results
systems areofnecessary
the study before
concern any technological a
essary before any technological approach.
the dissolution The but atresults
processes, the same of time they open
the study a realthe
concern possibility
dissolution to recover the metals
processes, but atof the same time they
cesses, but at the samepractical
time they open a real
importance from bromine-based
possibility systems.
to recover the metals of practical importance from bromine-based syste
rtance from bromine-based
• In thesystems. 
investigated experimental In the investigated
conditions, experimental
the highest conditions,
dissolution rates of the
thehighest
metalsdissolution
were rates of the
highest dissolution rates of theinmetals
obtained were
acidic bromide obtained
solution in acidic
containing bromide
0.01 M Br2 solution
and theycontaining
vary in the0.01 M Br 2 and
following order:they vary in the foll
01 M Br 2 and they vary in the following order:
Zn >> Sn > Pb > Fe> Cu. Zn >> Sn > Pb > Fe> Cu.
• XPS chemical assessment allowed XPSthe chemical assessment
identification of theallowed the identification
dissolution products formed of theondissolution
the products fo
on of the dissolution products formed on the metallic surfaces after exposure to bromine-containing solution in initial stages of
metallic surfaces after exposure to bromine-containing solution in initial stages of the corrosion
ining solution in initial stages of the corrosion
process. They consist mainly of process. They consist
metallic oxides mainly
in the case of metallic
of Cu, Zn, Sn and oxides in the case
Fe samples, whileof the
Cu, Zn, Sn and Fe sa
n the case of Cu, Zn, Sn and Fe samples, while
presence of a layer of PbBr2 wasthe presence
noticed of aPb
on the layer of PbBr2 was noticed on the Pb surface.
surface.
e Pb surface.  XRD the measurements showed that the immersion of the metals in different br
• XRD measurements showed that immersion of the metals in different bromide-based
n of the metals in different bromide-based
electrolytes induces structural electrolytes
modifications induces
for allstructural
electrodes. modifications
From the XRD for diffractograms
all electrodes. From the XRD d
all electrodes. From the XRD diffractograms obtained on all studied metals, a preferential orientation tendency of the cryst
obtained on all studied metals, a preferential orientation tendency of the crystallites of the corrosion
rientation tendency of the crystallites of the corrosion products was observed according to certain crystallization planes.
products was observed according to certain crystallization planes.
rtain crystallization planes.  areThe bromide-based systems are promising alternatives for lixiviants
• The bromide-based systems promising alternatives for lixiviants in hydrometallurgical route in hydrometal
tives for lixiviants in hydrometallurgical route
of metals recovery from WPCBs. Since the major economic driver for WPCBs recycling relies on for WPCBs recyc
of metals recovery from WPCBs. Since the major economic driver
economic driver for WPCBs recycling relies on
the efficient metal recovery, in athe
firstefficient metal recovery,
step, an effective dissolution in aof first
the rawstep, an effective
material dissolution of the raw
is necessary.
effective dissolution of the raw material is necessary. The results of our work show that all studied metals could be successful
The results of our work show that all studied metals could be successfully leached by using
udied metals could be successfully leached by
bromine-based systems and indicate good premises for selective and efficient metal recovery selective and ef
using bromine-based systems and indicate good premises for
d premises for selective and efficient metal recovery through a multi-step hydrometallurgical processing route.
through a multi-step hydrometallurgical processing route.
l processing route.
Supplementary Materials: The following Supplementary Materials:
are available online The following are available online at www.mdpi.com/xxx/s1, Figure S
at http://www.mdpi.com/1996-1944/13/16/3630/s1,
e at www.mdpi.com/xxx/s1,
Figure S1: Figure S1: Comparison
Comparison of the experimental
of the experimental impedance impedance data corresponding
data corresponding to the corrosion
to the corrosion of differentof metals
different metals in sol. C a
rrosion of different metals
in sol. inCsol.
andC the
andtransfer
the transfer
function calculated
function usingusing
calculated K–K K–Krelations. Metals:
relations. (a)(a)
Metals: Cu;
Cu;(b)
(b)Zn;
Zn;(c)
(c) Sn; (d)
(d) Pb;
Pb;(e) Fe. Experimenta
Zn; (c) Sn; (d) Pb; (e)(e)
Fe.Fe.
Experimental
Experimental data: (●) (•)
data: Real Realcomponent
component(Z’);(Z’); (( ) Imaginary
Imaginary component
component(Z”);
(Z”);Calculated
Calculatedpoints:
points:()(■)Real
Real component (Z’);
d points: (■) Real component
component (Z’);
(Z’); (♦) Imaginary component
( ) Imaginary component (Z”),
(Z”),Figure
FigureS2:
S2:SEM
SEMmicrographs
micrographsof ofthe
the metallic
metallic specimens
specimens after
after different immersio
different
c specimens after different immersionstimes
immersions timesininsol.
sol. C: 60 minutes in the case of Cu (a), Sn (b), Zn (c) and 120 minutes for Pb (d) and
C: 60 minutes in the case of Cu (a), Sn (b), Zn (c) and 120 minutes for Pb (d) and Fe (e). Inserts: S
nutes for Pb (d) and Fe (e). Inserts: SEM pictures inenlarged
Fe (e). Inserts: SEM pictures in scales.
enlarged scales.
Author Contributions: Conceptualization, S.V., S.-A.D., L.M.M., D.M. and P.I.; Data curation, A.O., R.B. and D.M.;
Formal analysis, S.V., A.O. and D.M.; Author Contributions:
Funding acquisition,Conceptualization, S.V., S.-A.D.,S.V.,
S.-A.D. and P.I.; Investigation, L.M.M.,
A.O.,D.M.
R.B.,and
M.P.P.I.; Data curation,
M.M., D.M. and P.I.;and Data curation,
D.M.; A.O., R.B.
Methodology, and
S.V., S.-A.D., A.O.,
D.M.; L.M.M.
Formal and D.M.;
analysis, S.V.,Project
A.O. andadministration,
D.M.; Funding S.-A.D. and P.I.;S.-A.D.
acquisition, Resources,
and P.I.; Investigation, S
tion, S.-A.D. and P.I.; Investigation,
S.-A.D. and P.I.; S.V., A.O., R.B.,
Supervision, S.V.; Validation, S.V., R.B.
M.P. and D.M.; and M.P.; Visualization,
Methodology, S.V., S.-A.D., S.V.,
A.O.,S.-A.D.,
L.M.M.L.M.M.,
and D.M.; M.P.,Project
D.M. administration, S.
and P.I.; Writing—original
and D.M.; Project administration, S.-A.D. anddraft,
P.I.; S.V., A.O., L.M.M. and D.M.; and Writing—review and editing, S.V., S.-A.D., A.O.,
Resources, S.-A.D. and P.I.; Supervision, S.V.; Validation, S.V., R.B. and M.P.; Visualization
S.V., R.B. and M.P.;L.M.M., D.M. and S.V.,
Visualization, P.I. All authors have read and agreed to the published version of the manuscript.
S.-A.D.,
Funding: This work was supported by a grant of the Romanian Ministry of Research and Innovation,
CCCDI-UEFISCDI, project number PN-III-P1-1.2-PCCDI-2017-0652/84PCCDI⁄ 2018, within PNCDI III.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Kaya, M. Recovery of metals and nonmetals from electronic waste by physical and chemical recycling
processes. Waste Manag. 2016, 57, 64–90. [CrossRef] [PubMed]
2. Montero, R.; Guevara, A.; De la Torre, E. Recovery of gold, silver, copper and niobium from printed circuit
boards using leaching column. J. Earth Sci. Eng. 2012, 2, 590–595.
Materials 2020, 13, 3630 25 of 28

3. Ha, V.H.; Lee, J.; Jeong, J.; Haia, H.T.; Jha, M.K. Thiosulfate leaching of gold from waste mobile phones.
J. Hazard. Mater. 2010, 178, 1115–1119. [CrossRef] [PubMed]
4. Abbruzzese, C.; Fornari, P.; Massidda, R.; Vegliò, F.; Ubaldini, S. Thiosulphate leaching for gold
hydrometallurgy. Hydrometallurgy 1995, 39, 265–276. [CrossRef]
5. Behnamfard, A.; Salarirad, M.M.; Veglio, F. Process development for recovery of copper and precious
metals from waste printed circuit boards with emphasize on palladium and gold leaching and precipitation.
Waste Manag. 2013, 33, 2354–2363. [CrossRef]
6. Jing-ying, L.; Xiu-li, X.; Wen-quan, L. Thiourea leaching gold and silver from the printed circuit boards of
waste mobile phones. Waste Manag. 2012, 32, 1209–1212. [CrossRef]
7. Birloaga, I.; De Michelis, I.; Ferella, F.; Bazatu, M.; Veglio, F. Study on the influence of various factors in the
hydrometallurgical processing of waste printed circuit boards for copper and gold recovery. Waste Manag.
2013, 33, 935–941. [CrossRef]
8. Kumar, M.; Lee, J.; Kim, M.; Jeong, J.; Yoo, K. Leaching of metals from waste printed circuit boards (WPCBs)
using sulfuric acid and nitric acids. Environ. Eng. Manag. J. 2014, 13, 2601–2607. [CrossRef]
9. Syed, S. A green technology for recovery of gold from non-metallic secondary sources. Hydrometallurgy 2006,
82, 48–53. [CrossRef]
10. Xu, Q.; Chen, D.; Chen, L.; Huang, M.H. Gold leaching from waste printed circuit board by iodine process.
Nonferrous. Met. 2010, 62, 88–90.
11. Altansukh, B.; Haga, K.; Ariunbolor, N.; Kawamura, S.; Shibayama, A. Leaching and adsorption of gold
from waste printed circuit boards using iodine-iodide solution and activated carbon. Eng. J. 2016, 20, 29–40.
[CrossRef]
12. Serpe, A.; Rigoldi, A.; Marras, C.; Artizzu, F.; Mercuri, M.L.; Deplano, P. Chameleon behaviour of iodine
in recovering noble-metals from WEEE: Towards sustainability and zero waste. Green Chem. 2015, 17,
2208–2216. [CrossRef]
13. Sahin, M.; Akcil, A.; Erust, C.; Altynbek, S.; Gahan, C.S.; Tuncuk, A. A potential alternative for precious
metal recovery from e-waste: Iodine leaching. Sep. Sci. Technol. 2015, 50, 2587–2595. [CrossRef]
14. Quinet, P.; Proost, J.; Van Lierde, A. Recovery of precious metals from electronic scrap by hydrometallurgical
processing routes. Min. Metall. Process 2005, 22, 17–22. [CrossRef]
15. Sousa, R.; Futuro, A.; Fiúza, A.; Vila, M.C.; Dinis, M.L. Bromine leaching as an alternative method for gold
dissolution. Min. Eng. 2018, 118, 16–23. [CrossRef]
16. Melashvili, M.; Fleming, C.; Dymov, I.; Manimaran, M.; O’Day, J. Study of gold leaching with bromine
and bromide and the influence of sulphide minerals on this reaction. In Proceedings of the Conference of
Metallurgists (COM 2014), Vancouver, BC, Canada, 28 September–1 October 2014; Canadian Institute of
Mining, Metallurgy and Petroleum: Westmount, QC, Canada, 2014. ISBN 978-1-926872-24-7.
17. Hojo, M.; Iwasaki, S.; Okamura, K. Pure gold dissolution with hydrogen peroxide as the oxidizer in HBr or
HI solution. J. Mol. Liq. 2017, 246, 372–378. [CrossRef]
18. Dorneanu, S.A. Electrochemical recycling of waste printed circuit boards in bromide media. Part I: Preliminary
leaching and dismantling tests. Studia Univ. Babeş-Bolyai Chem. 2017, 62, 177–186. [CrossRef]
19. Stern, M.; Geary, A.L. Electrochemical polarization: I. A theoretical analysis of the shape of polarization
curves. J. Electrochem. Soc. 1957, 104, 56–63. [CrossRef]
20. Marusic, K.; Otmacic Curkovic, H.; Takenouti, H. Inhibiting effect of 4-methyl-1-p-tolylimidazole to the
corrosion of bronze patinated in sulphate medium. Electrochim. Acta 2011, 56, 7491–7502. [CrossRef]
21. Ives, M.; Lu, Y.; Luo, J. Cathodic reactions involved in metallic corrosion in chlorinated saline environments.
Corros. Sci. 1991, 32, 91–102. [CrossRef]
22. Aben, T. Anodic polarization behavior of copper in aqueous bromide and bromide/benzotriazole solutions.
J. Electrochem. Soc. 1995, 142, 398. [CrossRef]
23. Brolo, A.G.; Temperini, M.L.A.; Agostinho, S.M.L. Copper dissolution in bromide medium in the absence
and presence of hexamethylenetetramine (HMTA). Electrochim. Acta 1998, 44, 559–571. [CrossRef]
24. Fernández-Domene, R.M.; Blasco-Tamarit, E.; García-García, D.M.; García-Antón, J. Thermogalvanic effects
on the corrosion of copper in heavy brine LiBr solutions. Corros. Sci. 2012, 63, 304–315. [CrossRef]
25. Baugh, L.M. Corrosion and polarization characteristics of zinc in neutral-acid media-I. Pure zinc in solutions
of various sodium salts. Electrochim. Acta 1979, 24, 657–667. [CrossRef]
Materials 2020, 13, 3630 26 of 28

26. Hassan, H.H.; Fahmy, K. Pitting corrosion of tin by acetate anion in acidic media. Int. J. Electrochem. Sci.
2008, 3, 29–43.
27. El Boraei, N.F.; Abd El Rehim, S.S. Electrochemical behavior of tin anode in ascorbic acid solutions.
Mater. Chem. Phys. 2018, 215, 332–338. [CrossRef]
28. Abd El Rehim, S.S.; Hassan, H.H. Anodic behaviour of tin in maleic acid solution and the effect of some
inorganic inhibitors. Corros. Sci. 2004, 46, 1071–1082. [CrossRef]
29. Johnson, J.W.; Liu, E.C. The anodic dissolution of tin in acidic chloride solutions. J. Less-Common Metals 1974,
34, 113–120. [CrossRef]
30. Barradas, R.G.; Belinko, K.; Ambrose, J. Electrochemical behavior of the lead electrode in HCl and NaCl
aqueous electrolytes. Can. J. Chem. 1975, 53, 389–406. [CrossRef]
31. Helal, N.H.; El-Rabiee, M.M.; Abd El-Hafez, G.M.; Badawy, W.A. Environmentally safe corrosion inhibition
of Pb in aqueous solutions. J. Alloy. Compd. 2008, 456, 372–378. [CrossRef]
32. Tanno, K.; Itoh, M.; Takahashi, T.; Yashiro, H.; Kumagai, N. The corrosion of carbon steel in lithium bromide
solution at moderate temperatures. Corros. Sci. 1993, 34, 1441–1451. [CrossRef]
33. Popova, A.; Christov, M.; Vasilev, A. Mono- and dicationic benzothiazolic quaternary ammonium bromides
as mild steel corrosion inhibitors. Part II: Electrochemical impedance and polarisation resistance results.
Corros. Sci. 2011, 53, 1770–1777. [CrossRef]
34. Raistrick, I.D.; MacDonald, J.R.; Franceschetti, D.R. The Electrical Analogs of Physical and Chemical Processes.
In Impedance Spectroscopy Emphasizing Solid Materials and Systems; MacDonald, J.R., Ed.; John Wiley & Sons:
New York, NY, USA, 1987; pp. 27–84.
35. Sherif, E.M.; Park, S.M. Inhibition of copper corrosion in acidic pickling solutions by N-phenyl-1,4-
phenylenediamine. Electrochim. Acta 2006, 51, 4665–4673. [CrossRef]
36. Abelev, E.; Starosvetsky, D.; Ein-Eli, Y. Enhanced copper surface protection in aqueous solutions containing
short-chain alkanoic acid potassium salts. Langmuir 2007, 23, 11281–11288. [CrossRef]
37. Assaf, F.H.; Abd El-Rehiem, S.S.; Zaky, A.M. Pitting corrosion of zinc in neutral halide solutions. Mater. Chem.
Phys. 1999, 58, 58–63. [CrossRef]
38. Chiu, S.L.; Selman, J.R. Determination of electrode kinetics by corrosion potential measurements: Zinc
corrosion by bromine. J. App. Electrochem. 1992, 22, 28–37. [CrossRef]
39. Cachet, C.; Wiart, R. Reaction mechanism for zinc dissolution in chloride electrolytes. J. Electroanal. Chem.
1981, 129, 103–114. [CrossRef]
40. Ahlberg, E.; Anderson, H. Simulating the impedance spectroscopy from a mechanistic point of view: Theory
and Simulations. Acta Chem. Scand. 1992, 46, 1–14. [CrossRef]
41. Cachet, C.; Ganne, F.; Maurin, G.; Petitjean, J.; Vivier, V.; Wiart, R. EIS investigation of zinc dissolution in
aerated sulfate medium, Part I: Bulk zinc. Electrochim. Acta 2001, 47, 509–518. [CrossRef]
42. Deslouis, C.; Duprat, M.; Tournillon, C. The kinetics of zinc dissolution in aerated sodium sulphate solutions.
a measurement of the corrosion rate by impedance techniques. Corros. Sci. 1989, 29, 13–30. [CrossRef]
43. Pilbáth, A.; Nyikos, L.; Bertóti, I.; Kálmán, E. Zinc corrosion protection with 1,5-diphosphono-pentane.
Corros. Sci. 2008, 50, 3314–3321. [CrossRef]
44. Biesinger, M.C.; Lau, L.W.M.; Gerson, A.R.; Smart, R.S.C. Resolving surface chemical states in XPS analysis of
first row transition metals, oxides and hydroxides: Sc, Ti, V, Cu and Zn. Appl. Surf. Sci. 2010, 257, 2717–2730.
[CrossRef]
45. Magaino, S.; Soga, M.; Sobue, K.; Kawaguchi, A.; Ishida, N.; Imai, H. Zinc corrosion in simulated acid rain.
Electrochim. Acta 1999, 44, 4307–4312. [CrossRef]
46. Cachet, C.; Wiart, R. The kinetics of zinc dissolution in chloride electrolytes: Impedance measurements and
electrode morphology. J. Electroanal. Chem. 1980, 111, 235–246. [CrossRef]
47. El Ibrahimi, B.; Jmiai, A.; El Mouaden, K.; Baddouh, A.; El Issami, S.; Bazzi, L.; Hilali, M. Effect of solution’s
pH and molecular structure of three linear α-amino acids on the corrosion of tin in salt solution: A combined
experimental and theoretical approach. J. Mol. Struct. 2019, 1196, 105–118. [CrossRef]
48. Huang, B.X.; Tornatore, P.; Li, Y.-S. IR and Raman spectroelectrochemical studies of corrosion films on tin.
Electrochim. Acta 2001, 46, 671–679. [CrossRef]
49. Jafarian, M.; Gobal, F.; Danaee, I.; Biabani, R.; Mahjani, M.G. Electrochemical studies of the pitting corrosion
of tin in citric acid solution containing Cl− . Electrochim. Acta 2008, 53, 4528–4536. [CrossRef]
Materials 2020, 13, 3630 27 of 28

50. Foad El-Sherbini, E.E.; Abd-El-Wahab, S.M.; Amin, M.A.; Deyab, M.A. Electrochemical behavior of tin in
sodium borate solutions and the effect of halide ions and some inorganic inhibitors. Corros. Sci. 2006, 48,
1885–1898. [CrossRef]
51. Yang, C.J.; Ko, Y.; Park, S.M. Fourier transform electrochemical impedance spectroscopic studies on anodic
reaction of lead. Electrochim. Acta 2012, 7, 615–622. [CrossRef]
52. Chen, Y.; Jepson, W.P. EIS measurement for corrosion monitoring under multiphase flow conditions.
Electrochim. Acta 1999, 44, 4453–4464. [CrossRef]
53. Keddam, M.; Mattos, O.R.; Takenouti, H. Reaction model for iron dissolution studied by electrode impedance
II. determination of the reaction model. J. Electrochem. Soc. 1981, 128, 257–266. [CrossRef]
54. Amin, M.A.; Khaled, K.F.; Mohsen, Q.; Arida, H.A. A study of the inhibition of iron corrosion in HCl solutions
by some amino acids. Corros. Sci. 2010, 52, 1684–1695. [CrossRef]
55. MacFarlane, D.R.; Smedley, S.I. The Dissolution mechanism of iron in chloride solutions. J. Electrochem. Soc.
1986, 133, 2240–2244. [CrossRef]
56. Montañés, M.T.; Pérez-Herranz, V.; García-Antón, J.; Guiñón, J.L. Evolution with exposure time of
copper corrosion in a concentrated lithium bromide solution. Characterization of corrosion products
by energy-dispersive X-ray analysis and X-ray Diffraction. Corrosion 2006, 62, 64–73. [CrossRef]
57. Biesinger, M.C. Advanced analysis of copper X-ray photoelectron spectra. Surf. Interface Anal. 2017, 49,
1325–1334. [CrossRef]
58. Capece, F.M.; Dicastro, V.; Furlani, C.; Mattogno, G.; Fragale, C.; Gargano, M.; Rossi, M. “Copper chromite”
catalysts: XPS structure elucidation and correlation with catalytic activity. J. Electron Spectrosc. Relat. Phenom.
1982, 27, 119–128. [CrossRef]
59. Nakamura, T.; Tomizuka, H.; Takahashi, M.; Hoshi, T.J. Methods of powder sample mounting and their
evaluations in XPS analysis. Surf. Sci. Soc. Jpn. 1995, 16, 515–520. [CrossRef]
60. Jolley, J.G.; Geesey, G.G.; Haukins, M.R.; Write, R.B.; Wichlacz, P.L. Auger electron and X-ray photoelectron
spectroscopic study of the biocorrosion of copper by alginic acid polysaccharide. Appl. Surf. Sci. 1989, 37,
469–480. [CrossRef]
61. Koberstein, J.T. Quantitative analysis of copper oxide nanoparticle composition and structure by X-ray
photoelectron spectroscopy. Chem. Mater. 2006, 18, 6054–6058. [CrossRef]
62. Dube, C.E.; Workie, B.; Kounaves, S.P.; Robbat, A., Jr.; Aksu, M.L.; Davies, G. Electrodeposition of metal alloy
and mixed oxide films using a single-precursor tetranuclear copper-nickel complex. J. Electrochem. Soc. 1995,
142, 3357–3365. [CrossRef]
63. Otamiri, J.C.; Andersson, S.L.T.; Andersson, A. Ammoxidation of toluene by YBa2 Cu3 O6+x and copper
oxides: Activity and XPS studies. Appl. Catal. 1990, 65, 159–174. [CrossRef]
64. Morgan, W.E.; Van Wazer, J.R.; Stec, W.J. Inner-orbital photoelectron spectroscopy of the alkali metal halides,
perchlorates, phosphates, and pyrophosphates. J. Am. Chem. Soc. 1973, 95, 751–755. [CrossRef]
65. Wagner, C.D.; Riggs, W.M.; Davis, L.E.; Moulder, J.F.; Muilenberg, G.E. Handbook of X-ray Photoelectron
Spectroscopy; Perkin-Elmer Corporation, Physical Electronics Division: Eden Prairie, MN, USA, 1979.
[CrossRef]
66. Sharma, J.; Iqbal, Z. X-ray photoelectron spectroscopy of brominated (SN)x and S4 N4 . Chem. Phys. Lett. 1978,
56, 373–376. [CrossRef]
67. Hosking, N.C.; Ström, M.A.; Shipway, P.H.; Rudd, C.D. Corrosion resistance of zinc–magnesium coated steel.
Corros. Sci. 2007, 49, 3669–3695. [CrossRef]
68. Sen, P.; Hegde, M.S.; Rao, C.N.R. Surface oxidation of cadmium, indium, tin and antimony by photoelectron
and Auger spectroscopy. Appl. Surf. Sci. 1982, 10, 63–74. [CrossRef]
69. Pederson, L.R. Two-dimensional chemical-state plot for lead using XPS. J. Electron. Spectrosc. Relat. Phenom.
1982, 28, 203–209. [CrossRef]
70. Powell, C.J. Recommended auger parameters for 42 elemental solids. J. Electron. Spectrosc. Relat. Phenom.
2012, 185, 1–3. [CrossRef]
71. Nefedov, V.I. A comparison of results of an ESCA study of nonconducting solids using spectrometers of
different constructions. J. Electron. Spectrosc. Relat. Phenom. 1982, 25, 29–47. [CrossRef]
72. Uhlig, I.; Szargan, R.; Nesbitt, H.W.; Laajalehto, K. Surface states and reactivity of pyrite and marcasite.
Appl. Surf. Sci. 2001, 179, 222–229. [CrossRef]
Materials 2020, 13, 3630 28 of 28

73. Grosvenor, A.P.; Kobe, B.A.; Biesinger, M.C.; McIntyre, N.S. Investigation of multiplet splitting of Fe 2p XPS
spectra and bonding in iron compounds. Surf. Interface Anal. 2004, 36, 1564–1574. [CrossRef]
74. Marcus, P.; Grimal, J.M. The anodic dissolution and passivation of Ni-Cr-Fe alloys studied by ESCA.
Corros. Sci. 1992, 33, 805–814. [CrossRef]
75. Hawn, D.D.; DeKoven, B.M. Deconvolution as a correction for photoelectron inelastic energy losses in the
core level XPS spectra of iron oxides. Surf. Interface Anal. 1987, 10, 63–74. [CrossRef]
76. Langevoort, J.C.; Sutherland, I.f.; Hanekamp, L.J.; Gellings, P.J. On the oxide formation on stainless steels
AISI 304 and incoloy 800H investigated with XPS. Appl. Surf. Sci. 1987, 28, 167–179. [CrossRef]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).