Content uploaded by Barbara Kazanski
Author content
All content in this area was uploaded by Barbara Kazanski on Feb 15, 2015
Content may be subject to copyright.
Comparison of electrochemical and chemical corrosion behavior of MRI
230D magnesium alloy with and without Plasma Electrolytic Oxidation
treatment
Barbara Kazanski1,a, Alex Lugovskoy1,b, Ohad Gaon1,c, Michael Zinigrad1,d
1Department of Chemical and Materials Engineering, Ariel University, Ariel, Israel
abarbarak@ariel.ac.il, blugovsa@ariel.ac.il, cohadgaon@gmail.com, dzinigrad@ariel.ac.il
Keywords: Magnesium alloys, plasma electrolytic oxidation, Tafel polarization, Linear Polarization
resistance, Weight loss, Impedance Spectroscopy.
Abstract. Magnesium is one of the lightest metals and magnesium alloys have quite special
properties, interest to which is continuously growing. In particular, their high strength-to-weight
ratio makes magnesium alloys attractive for various applications, such as transportation, aerospace
industry etc. However, magnesium alloys are still not as popular as aluminum alloys, and a major
issue is their corrosion behavior.
The present research investigated the influence of the PEO treatment on the corrosion behavior of
MRI 230M magnesium alloy. Plasma electrolytic oxidation (PEO) of an MRI 230M alloy was
accomplished in a silicate-base electrolyte with KF addition using an AC power source.
The corrosion behavior of both treated and untreated samples was evaluated by open circuit
potential (OCP) measurements, electrochemical impedance spectroscopy (EIS), linear polarization
tests, linear sweep voltammetry (Tafel extrapolation) and chemical methods, such as mass loss and
hydrogen evolution, in neutral 3.0 wt% NaCl solution.
According to the tests results, PEO process can affect the corrosion resistance of MRI 230M
magnesium alloy, though its action is not always unambiguous. An attempt to explain the influence
of the PEO treatment on the corrosion behavior of the alloy is presented.
Introduction
Magnesium and magnesium alloys have become very popular in various applications due to their
excellent physical and mechanical properties [1]. Yet, their relatively poor corrosion resistance [2]
and, therefore, necessity for surface treatments [3-6], lower the practical usage of magnesium and
its alloys. There are number of surface modifications technologies that can be used to improve the
corrosion and wear resistance of magnesium. One of them is Plasma Electrolytic Oxidation (PEO),
which is based on a conventional anodizing process, but capable of producing much thicker oxide
layers. PEO is an environmentally friendly technology, capable of providing a denser and harder
surface structure as compared to those produced by anodizing [7].
In Plasma Electrolytic Oxidation an oxide layer is formed at high voltage in an alkaline-silicate
based electrolyte by micro-plasma discharges. It was established that different electrolyte
compositions [8-10] as well as different applied electrical regimes [11] and oxidation times [12]
result in different oxide thickness and morphology. Another parameter affecting electrochemical
and mechanical properties of PEO films is substrate composition. According to the literature [13-
15], AZ series alloys (containing zinc) demonstrate better electrochemical and mechanical
properties of PEO produced oxide layers than AM series alloys (containing manganese) or WE
series (containing yttrium). However, the detailed effect of the alloy composition on the corrosion
resistance of the coatings produced by PEO is still unknown.
The purpose of this work is to study the corrosion behavior of MRI 230D magnesium alloy with
and without PEO treatment using electrochemical impedance spectroscopy, linear polarization
resistance, Tafel extrapolation, mass loss and hydrogen evolution. The morphology and
composition of the samples are investigated by scanning electron microscopy (SEM) and energy
dispersive spectroscopy (EDS).
Defect and Diffusion Forum Vol. 364 (2015) pp 27-34
© (2015) Trans Tech Publications, Switzerland
doi:10.4028/www.scientific.net/DF.4.27
All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of TTP,
www.ttp.net. (ID: 62.219.54.250, Ariel University Center of Samaria, Ariel, Israel-15/02/15,08:18:50)
Experimental
High-pressure die cast MRI 230D magnesium alloy (6.3–7.2% Al, 1.8–2.6% Ca, 0.15-0.4% Mn,
0.3-1.2% Sn, 0.05-0.4% Sr, 0.08% max Zn, 0.05% max of others) was cut into rectangular flat
plates, polished with #300, #600, #1000 grit diamond abrasive paper, degreased ultrasonically in
MRC D80H bath in tap water with soap for 10 min and dried with acetone.
Oxidation experiments were carried out with a home-made 40 kVA PEO station with a water-
cooled bath made of stainless steel, which served as the counter electrode; in AC mode by 50 Hz
sow edged voltage (±100 V) at the initial RMS current density of approximately 10.0 ± 2.0 A/dm2,
for 30 minutes. After the oxidation samples were rinsed in distilled water to prevent additional
deposition of electrolyte components during the drying. The electrolyte contained 0.05M Na2SiO3
(pentahydrate, Spectrum, practical grade), 0.05M KOH (Finkelman Chemicals, technical grade) and
0.05M KF (Merck, 99%) in tap water. Conductivities and рН of the electrolytes were measured by
an YK-2005WA pH/CD meter, the thickness of oxide layers was measured by MiniTest 730 with
SIDSP Thickness gauge.
The surface morphology and composition of both oxidized and non-oxidized samples were
inspected on SEM JEOL JSM6510LV equipped with an NSS7 EDS analyzer (Correction Method
Proza – Phi-Pho-Z was used for the quantitative analysis).
The electrochemical behavior of all the samples was studied on an Ivium Potentiostat in the three-
electrode cell arrangement, during the exposure to 3% NaCl solution. A coated sample, with a
tested area of 1.13 cm2, was used as a working electrode, the counter electrode was a platinum wire
and the reference electrode used was a standard Ag|AgCl electrode. EIS data were obtained at the
open-circuit potential over a frequency range of 10 kHz to 0.1 Hz using 5 mV amplitude of
sinusoidal voltage. The corrosion rates in 3% NaCl solution of both oxidized and non-oxidized
samples were determined using linear polarization resistance (LPR) tests and Tafel plots. Potential
sweep rate of 0.05 and 1.0 mV/sec, correspondently, was applied after the constant open-circuit
potential (OCP) was established (~80 h), starting from the OCP to the cathodic and, independently,
to the anodic direction. All the electrochemical experiments were repeated three times to verify the
reproducibility of the results.
Mass loss measurements were also performed on both oxidized and non-oxidized samples.
Specimens were immersed in 3.0 wt.% NaCl solution at room temperature (pH 5.4, 20±30C) for
several days. Prior to the tests, specimens were measured and weighed. Once the test was finished,
the samples were extracted, rinsed in distilled water and then etched with the standard reagent (10
gr CrO3, 0.5 gr AgNO3 in 50 ml H2O) in order to remove the corrosion products. After that, the
samples were weighed again in order to calculate the mass loss per unit surface area. In order to
obtain further information, hydrogen evolution during the corrosion process was measured using a
simple procedure as shown in Fig. 1.
Fig. 1. Hydrogen evolution and mass loss procedure to measure the corrosion rate.
28 Recent Developments of Diffusion Processes and their Applications: Fluid,
Heat and Mass
Results and Discussion
The OCP measurements for samples with and without Plasma Electrolytic Oxidation after their
immersion for approximately 3 days in 3% NaCl solution are shown in Fig. 2.
1.45
1.5
0 10 20 30 40 50 60
OCP with PEO (-V)
OCP without PEO (-V)
OCP (-V)
Time (hr)
Fig. 2. Stabilization of the corrosion potential for MRI 230D alloy in 3% NaCl.
After the achievement of a stable potential electrochemical impedance spectroscopy, linear
polarization resistance and Tafel extrapolation were used for the elucidation of the corrosion
parameters.
The corrosion rate, current and potential, as well as Tafel slopes values was determined from the
extrapolation of the Tafel slopes, build to the polarization curves measured in the potential range of
Ess ± 0.25V, where Ess is the stable free corrosion potential. Typical examples of Tafel plots for
both treated and untreated samples are presented in Fig. 3. The comparison shows that the corrosion
potential for an oxidized sample is higher than for the untreated one, while the corrosion current
remains practically the same. That means that oxidized samples are nobler and less affected by the
corrosion, but as the corrosion has started it progresses at the same rate for both samples. The
calculated values of corrosion parameters are presented in Table 1.
-4 -3 -2 -1 0 1 2
without PEO
with PEO
-1.7
-1.6
-1.5
-1.4
-1.3
-1.2
-1.1
Potential (V)
log i (mA)
Fig. 3. Tafel polarization plot for MRI 230D alloy in 3% NaCl.
Defect and Diffusion Forum Vol. 364 29
Table 1. Corrosion parameters measured by Tafel slope extrapolation technique for MRI 230D
samples with and without Plasma electrolytic oxidation treatment.
Ecorr (V) jcorr
(mA/cm2)
Rp (Ω) βa(V/dec) βc(V/dec) Corrosion rate
(mm/year)
with
PEO -1.207 0.0236 1834 0.184 0.293 0.516
without
PEO -1.367 0.0244 1480 0.124 0.414 0.551
The values of Ecorr and jcorr were also obtained from linear polarization curves measured in the
potential range of Ess ±10mV, both before and after PEO, which are presented in Fig. 4 and Table 2.
The values of the cathodic (βc) and anodic (βa) Tafel slops were used for the calculation of the
Stern-Geary constant B (Eq. 1)
= ·
.() (1)
and the calculation of polarization resistance Rp (Eq. 2).
=
(2)
It is seen from Table 2 that Plasma Electrolytic Oxidation treatment results in slower corrosion
rates and the corrosion potential into the positive direction, which corresponds with results obtained
from Tafel extrapolation. However, the polarization resistance values show some discrepancy: the
polarization resistance of an untreated sample is higher than of an oxidized one.
-0.04
-0.02
0
0.02
0.04
0.06
0.08
0.1
-1.51 -1.5 -1.49 -1.48 -1.47 -1.46 -1.45
with PEO
without PEO
i (mA)
E (V)
Fig. 4. Linear polarization resistance plot for MRI 230D alloy in 3% NaCl.
It can be assumed that either the preciseness of the corrosion current measurements for the clearly
localized corrosion (Fig. 5) is simply not sufficient for the discrimination in the range of µA/cm2, or
that the LPR measurements hint that the corrosion inside the micrometer-range pores in the PEO
coating may proceed very actively though be confined in very small pits.
30 Recent Developments of Diffusion Processes and their Applications: Fluid,
Heat and Mass
Fig. 5. Photo images of MRI 230D alloy samples after electrochemical experiments with (a) and
without (b) Plasma Electrolytic Oxidation treatment: a round window only is open to the corrosion
process.
Table 2. Corrosion parameters measured by linear polarization resistance technique for MRI 230D
samples with and without Plasma Electrolytic Oxidation treatment.
E
corr
(V) j
corr
(mA/cm
2
) R
p
(Ω) B(V/dec)
Corrosion rate (mm/year)
with PEO -1.467 0.0539 911.5 0.049 0.176
without PEO -1.486 0.0266 1562 0.042 0.599
The EIS measurements were carried out in order to confirm the data obtained by polarization
resistance and Tafel extrapolation, and to learn about the possible corrosion mechanism. The
Nyquist plots for MRI 230D alloy samples before (a) and after (b) PEO treatment are shown in Fig.
6.
Fig. 6. EIS Nyquist plots for MRI 230D alloy (a) with PEO and (b) without PEO in 3% NaCl.
The general profile of the Nyquist plot of non-oxidized sample indicates a single process, while
the profile of the treated sample shows at least two different processes probably caused by a thick
oxide layer on the sample surface. The presence of an inductive loop in both impedance spectra can
be explained by the surface sorption of metal ions from the corrosive medium. Because of the
complex structure of the EIS spectra and the fact that the corrosion mechanisms with and without
PEO are visually different, we did no calculation of the corrosion parameters for the impedance
measurement.
The weight loss corrosion rate, P
W
(mm/year), was measured by Eq. 3, according to the literature
[16, 17]:
=
∙∙
(3)
where W
i
and W
f
(gr) are a sample’s initial and final weight, accordingly, A (cm
2
) is the surface
area of the sample, ρ (gr/cm
3
) is the sample’s density and time is the overall immersion time (years).
The evolved hydrogen was collected during the mass loss test and the average corrosion rate, P
H
(mm/year), was calculated by Eq. 4, using the ideal gas law:
=
∙
(4)
where V is the total volume of the evolved hydrogen, A (cm
2
) is the surface area of the sample and
time is the overall immersion time (years).
a
b
a
b
Defect and Diffusion Forum Vol. 364 31
Fig. 7 shows the specific hydrogen volume as a function of time for both treated and untreated
samples and the calculated results are presented in Table 3.
Table 3. Corrosion rate values measured by mass loss (PW) and hydrogen evolution (PH) techniques
for MRI 230D samples with and without Plasma Electrolytic Oxidation treatment.
PW (mm/year) PH (mm/year)
with PEO 0.526 0.363
without PEO 0.707 0.846
The mass loss and gas evolution data obtained from oxidized sample are in good agreement with
the data obtained from Tafel and linear polarization. However, corrosion rates of the non-oxidized
sample obtained by non-electrochemical methods are slightly higher than the corrosion rates
obtained by electrochemical techniques. Even though, all the data are on the same order of
magnitude, while the minor differences can be explained by the difficulty in the determining of the
exact sample area after the immersion for 20 days.
0
1
2
3
4
5
6
7
0 100 200 30 0 400 500 600
with PEO
without PEO
V/A (ml/cm2)
time (hr)
Fig. 7. Hydrogen evolution plot for MRI 230D alloy in 3% NaCl.
In order to differentiate between the surface morphology of the treated and untreated samples
after an immersion in 3% NaCl solution, SEM investigations were carried out. Fig. 8 shows the
SEM images of (a) the surface including a border between the area exposed to the corrosive media
and the area that was covered with a polymer protective layer, (b) the morphology of the corrosion
products deposit on a sample without the PEO treatment.
Fig. 8. SEI SEM images (a) x30, (b) x1000 of the non-oxidized sample
a
b
32 Recent Developments of Diffusion Processes and their Applications: Fluid,
Heat and Mass
The SEM images of an oxidized sample are presented in three different magnifications and show
different areas. Fig. 9 shows (a) a surface exposed to the corrosive media with a corrosion pit in a
low magnification, (b) the surface outside the corrosion pit and (c) the surface inside the corrosion
pit.
Fig. 9. SEI SEM images (a) x30, (b) x400, (c) x1000 of the oxidized sample
As follows from the low magnitude SEM images (Fig. 8a and Fig. 9a), as well as from Fig. 5, the
non-oxidized samples have a more continuous corrosion behavior, while the corrosion of samples
after the PEO process is clearly localized (pitting).
The weight percentage of the elements of the corresponding EDS profile of the selected areas on
the SEM images shown in Fig. 8b were 46.01% O, 47.78% Mg, 0.18% Al, and 0.17% Sn. This
suggests that the compound formed on the untreated Mg alloy is mainly magnesium oxide. On the
other hand, the weight percentages of the elements found in the pitting area of SEM image shown in
Fig. 9c were 59.05% O, 31.47% Mg, and 3.92% Si. This can indicate the formation of a magnesium
silicate film instead of magnesium oxide.
Conclusions
1. Magnesium alloy MRI 230D shown an excellent corrosion resistance, with a good correlation to
the literature [18].
2. Corrosion rates measured by different electrochemical and chemical methods are consistent to at
list an order of magnitude.
3. The Plasma Electrolytic Oxidation treatment seems to only slightly improve the corrosion
protection of the oxidized samples by shifting the corrosion potential to the cathodic direction.
However, the samples with PEO show a lower polarization resistance than untreated ones due to
the pitting character of corrosion.
Acknowledgements
The authors extend their appreciation to the Dead Sea Magnesium Ltd. and personally to Dr. B.
Bronfin for the provided alloy samples and technical information, and to Eng. Natalya Litvak for
SEM-EDS imaging and analysis.
References
[1] B.L. Mordike, T. Ebert, “Magnesium: Properties,applications, potential”, Mater. Sci. Eng.,
A302, 37–45, (2001)
[2] B. A. Shaw, “Corrosion Resistanceof Magnesium Alloys”, ASM Handbook, 13A, 692-695,
(2003)
[3] A. L. Rudd, C. B. Breslin, F. Mansfeld, “The corrosion protection afforded by rareearth
conversion coatings applied tomagnesium”, Corros. Sci., 42, 275-288, (2000)
[4] S.Shadanbaz, G. J. Dias, “Calcium phosphate coatings on magnesium alloys for biomedical
applications: A review”, Acta Biomaterialia, 8, 20–30, (2012)
[5] C.E. Barchiche, E. Rocca, C. Juers, J. Hazan, J. Steinmetz, “Corrosion resistance of plasma-
anodized AZ91D magnesiumalloy by electrochemical methods”, Electrochim.Acta, 53, 417–425,
(2007)
a
b
c
Defect and Diffusion Forum Vol. 364 33
[6] R.G. Hu, S. Zhang, J.F. Bu, Ch.J. Lin, G.L. Song, “Recent progress in corrosion protection of
magnesium alloys by organic coatings”, Prog. Org. Coat., 73, 129– 141, (2012)
[7] K. Murakami, M. Hino, K.Nakai, S. Kobayashi,A.Saijo, T.Kanadani, “Mechanism of Corrosion
Protection of Anodized Magnesium Alloys”, Mater.Transact.,49 (5),1057–1064, (2008)
[8] B. Kazanski, A. Kossenko, M. Zinigrad, A. Lugovskoy, “Fluoride ions as modifiers of the oxide
layer produced by plasma electrolytic oxidation on AZ91D magnesium alloy”, App. Surf.
Sci.,287,461– 466, (2013)
[9] J. Li, Y. Tian, Z. Cui, Z. Huang, “Effects of rare earths on the microarc oxidation of a
magnesium alloy”, Rare Metals, 27 (I), 50– 54, (2008)
[10] H.Duan, C. Yan, F. Wang, “Effect of electrolyte additives on performance of plasma
electrolyticoxidation films formed on magnesium alloy AZ91D”, Electrochim. Acta, 52, 3785–
3793, (2007)
[11] R.F. Zhang, D.Y. Shan, R.S. Chen, E.H. Han, “Effects of electric parameters on properties of
anodic coatings formed on magnesium alloys”, Mater. Chem. and Phys., 107, 356–363, (2008)
[12] C. Blawert, V. Heitmann, W. Dietzel, H.M. Nykyforchyn, M.D. Klapkiv, “Influence of process
parameters on the corrosion properties of electrolyticconversion plasma coated magnesium alloys”,
Surf.Coat. Tech., 200, 68– 72, (2005)
[13] E.Cakmak, K. C. Tekin, U.Malayoglu, S. Shrestha, “The effect of substrate composition on the
electrochemical and mechanicalproperties of PEO coatings on Mg alloys”, Surf. Coat. Tech., 204,
1305–1313, (2010)
[14] R. Arrabal, E. Matykina, F. Viejo, P. Skeldon, G.E. Thompson, “Corrosion resistance of WE43
and AZ91D magnesium alloys with phosphate PEO coatings”, Corros. Sci., 50, 1744–1752, (2008)
[15] X. Zeng, Y. Wang, W. Ding, A. A. Luo,A. K. Sachdev, “Effect of Strontium on the
Microstructure, MechanicalProperties, and Fracture Behavior of AZ31 Magnesium Alloy”,
Metal.Mater. Trans., 37A, 1333–1341, (2006)
[16] Z. Shi, M. Liu, A. Atrens, “Measurement of the corrosion rate of magnesium alloys using Tafel
extrapolation”, Corros. Sci., 52, 579–588, (2010)
[17] F. Cao, Z. Shi, J. Hofstetter, P. J. Uggowitzer, G. Song, M. Liu, A. Atrens, “Corrosion of ultra-
high-purity Mg in 3.5% NaCl solution saturated with Mg(OH)2”, Corros. Sci., 75, 78–99, (2013)
[18] E. Aghion, B. Bronfin, F. von Buch, S. Schumann, H. Friedrich, “The art of developing new
magnesium alloys for high temperature applications”, Mater. Sci. Forum, 419, 407-418, (2003)
34 Recent Developments of Diffusion Processes and their Applications: Fluid,
Heat and Mass
