Structural and electrochemical behavior of sol-gel ZrO2 ceramic film on chemically pre-treated AZ91D magnesium alloy

Article abstract of DOI:10.1016/j.jallcom.2008.11.123

In the present investigation sol-gel-based ZrO₂ ceramic film was obtained using zirconium acetate as the precursor material. The film was deposited on AZ91D magnesium alloy by a dip-coating technique. An uniform stannate conversion coating as c

Full text of DOI:10.1016/j.jallcom.2008.11.123

Journal of Alloys and Compounds 478 (2009) 544–549
Contents lists available at ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jallcom
Structural and electrochemical behavior of sol–gel ZrO2 ceramic film on
chemically pre-treated AZ91D magnesium alloy
Qing Li∗, Bo Chen, Shuqiang Xu, Hui Gao, Liang Zhang, Chao Liu
School of Chemistry and Chemical Engineering, Southwest University, Chongqing 400715, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 9 July 2008
Received in revised form
15 November 2008
Accepted 21 November 2008
Available online 6 December 2008
In the present investigation sol–gel-based ZrO2 ceramic film was obtained using zirconium acetate as
the precursor material. The film was deposited on AZ91D magnesium alloy by a dip-coating technique.
An uniform stannate conversion coating as chemical pretreatment was employed as an intermediate
layer prior to deposition of the ZrO2 film in order to provide advantage for the formation of sol–gel-
based ZrO2 layer. The corrosion properties, structure, composition and morphology of these coatings on
AZ91D magnesium alloy were studied by potentiodynamic polarization tests, EIS, XRD, SEM, respectively.
According to the electrochemical tests, the corrosion resistance of AZ91D magnesium alloy was found to
be greatly improved by means of this new environment-friendly surface treatment.
Keywords:
Magnesium alloy
Sol–gel processes
Thin films
Crown Copyright © 2008 Published by Elsevier B.V. All rights reserved.
Corrosion
Electrochemical impedance spectroscopy
1. Introduction
In this paper, a new environmental-functionally gradient coat-
ing preparation method was proposed for AZ91D magnesium alloy
Magnesium alloys present a wide range of potential applica-
tions due to their high strength and low density. However, poor
corrosion resistance and high chemical reactivity have hindered
their wide applications [1–3]. One of the most effective ways is
to coat the base materials. At present, the chemical conversion
methods are mainly based on immersion treatment in a solution
which contains hexavalent chromium compounds. Chromate con-
version coatings had shown to provide good corrosion protection
for magnesium and its alloys through their self-healing properties
[4]. However, some environmental laws have been passed by many
governments to impose severe restrictions on chromate use due to
its high toxicity and consequent environmental hazards, and these
severe restrictions have stimulated the research and development
of new environment-friendly surface treatments. The sol–gel pro-
cess and its parameters have been actively studied for the past
several decades and there is a growing body of literatures avail-
able [5]. Sol–gel-based ZrO2 thin films, owing to their high thermal
expansion coefficients and chemical stability, have important appli-
cations as corrosion-resistant coatings on metals [6–8]. Chemical
conversion coatings are easy to apply and cost-effective. The main
factors related to stannate conversion coating on magnesium alloys
had been expounded in the previous investigations [9–12].
to improve its corrosion resistance, which consisted of pre-etching
followed by stannate chemical conversion before applying the
sol–gel-based ZrO2 ceramic film. Etching had been employed in
this method to improve the coating density and corrosion resis-
tance of stannate-coated specimens [9]. The stannate conversion
coating, which acted as chemical pretreatment and a barrier to envi-
ronment, improved the corrosion resistance of the metal to some
extent, and then ZrO2 ceramic film was deposited on the stannate
conversion coating through sol–gel dip-coating technique. A newly
developed ZrO2 sol-based on zirconium acetate prepared by sol–gel
process was also proposed. Electrochemical techniques had been
employed to characterize the corrosion properties of these coatings.
2. Experimental
2.1. Materials
Die-caste AZ91D magnesium alloy with
a chemical composition (wt.%) of
Mg–8.77Al–0.74Zn was used in this study. The dimensions of specimens were
18 mm × 22 mm × 4 mm.
2.2. Specimens preparation
The specimens were mechanically polished by SiC papers up to 2000 grit to
ensure the same surface roughness. They were then degreased with acetone in an
ultrasonic bath and dried under a stream of air. Prior to the immersion in the stan-
nate bath, the plate was pickled in 0.25 wt.% HF + 0.25 wt.% HCl solution at room
temperature. Composition of the stannate bath was 50 g/L Na2SnO3•3H2O, 10 g/L
CH3COONa•3H2O, 50 g/L Na3P2O7 and 10 g/L NaOH. Analytical-grade chemicals were
∗
Corresponding author. Tel.: +86 23 68253884; fax: +86 23 68367675.
E-mail addresses: liqingd@swu.edu.cn, chenbo11-1984@163.com (Q. Li).
0925-8388/$ – see front matter. Crown Copyright © 2008 Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.jallcom.2008.11.123

Q. Li et al. / Journal of Alloys and Compounds 478 (2009) 544–549
used for preparation of all solutions. Zirconium acetate as the precursor was dis-
545
solved in the ethanol, celloidine as the disperser was added dropwise. The volume
ratio between ethanol and celloidin was 4:1, and the concentration of the precur-
sor was 0.08 M. The sol–gel ZrO2 film was obtained for sol–gel dip-coating process.
Firstly, it was dried in the open air, and then sintered at a heating rate of 10 ◦C/min
to 400 ◦C and held there for 2 h. The above procedures were repeated three times for
the preparation of composite coating.
2.3. Tests
The electrochemical behavior of the stannate-coated specimen with or without
ZrO2 ceramic film was investigated by means of potentiodynamic polarization mea-
surements PS-268B (China) and electrochemical impedance spectroscopy (EIS) IM6e
system (Germany). Both of the electrochemical tests were carried out using a clas-
sical three electrodes cell with platinum as counter electrode. The potentiodynamic
polarization curves were obtained in 5 × 10−4 M H2SO4 solution with a scanning rate
of 1 mV/s. Impedance measurements were carried out at open circuit potential in
the 3.5 wt.% NaCl solution at perturbation amplitude of 10 mV and frequency range
from 1 × 105 to 10−2 Hz. The morphology and composition of the treated specimens
was investigated by scanning electron microscopy (SEM) S-4800 (Japan) and X-ray
powder diffraction XD-3 system (China), using Cu K␣ (1.5405 Å) radiation source.
The operating voltage and current are 40 kV and 50 mA, respectively. In order to
exactly analyze the phase structures of ZrO2 ceramic film on the surface of speci-
mens, ZrO2 gel powder was detected considering as the ZrO2 ceramic film coated on
the substrate. The ZrO2 gel powder was sintered at 400 ◦C for 2 h.
3. Results and discussion
3.1. X-ray diffraction
Fig. 1 illustrates the phase components of the coatings formed
on AZ91D alloy substrate after chemical treatment. According to
Fig. 1(a), it is clear that the chemical conversion coating shows sev-
Fig. 2. SEM micrographs for (a) the AZ91D substrate; (b) the stannate conversion
coating; and (c) the ZrO2 ceramic film.
eral peaks corresponding to MgSnO3·H2O. As shown in Fig. 1(b),
the ZrO2 gel powder which had been sintered at 400 ◦C for 2 h con-
sidering as the ZrO2 ceramic film coated on the substrate showed
formations of crystalline ZrO2, giving its characteristic peaks sig-
nifying orientation along (1 1 1), (2 0 0), (2 2 0), (3 1 1) planes. The
average particle size of ZrO2 that calculated along the dominant
(1 1 1) plane using the Scherrer formula was about 10 nm. It can
be concluded that the ZrO2 ceramic film is mainly composed of
tetragonal phase ZrO2. In Fig. 1(a), the characteristic peaks of alloy
substrate could also be obviously observed and the peak intensity
was weaker than that of stannate-coated specimen. This implied
that the stannate chemical conversion coating was not thick. Obvi-
Fig. 1. X-ray diffraction patterns of (a) the AZ91D substrate and the stannate con-
version coating; (b) the ZrO2 gel powder sintered at 400 ◦C for 2 h.

546
Q. Li et al. / Journal of Alloys and Compounds 478 (2009) 544–549
Table 1
ously, these broadened peaks also implied that fine grains were
formed. Thus, it indicates that a nanocrystalline ZrO2 ceramic film
was successfully obtained in this study.
The relevant electrochemical parameters of potentiodynamic polarization curves
calculated from the Tafel plots.
Sample
Ecorr (mV)
Icorr (A/cm2)
RP (ꢀ cm2)
C.R. (mpy)
Substrate
Chemical conversion
ZrO2 ceramic film
−1758
−1654
−1446
1.24 × 10−3
9.1 × 10−5
3.5 × 10−6
17.57
238.06
6163.55
14.35
1.06
0.04
3.2. Surface morphology
The microstructure and surface morphology of the alloy spec-
imens with and without non-chromate surface treatments are
shown in Fig. 2. As shown in Fig. 2(a), the surface of the uncoated
specimen is uniform and there are no impurity particles on it
before pickling pretreatment, but some fine nicks resulting from
the polishing process are still visible. After pickling pretreatment,
the micrographs were taken when the specimens were immerged in
stannate bath for 1 h, which can be appreciated in Fig. 2(b). As can be
seen from this image, the coating is entirely covered on the surface
of AZ91D alloy substrate, a few spatial gaps between finer hemi-
spherical particles can also be observed. According to the theory
of nucleation [10,12], the conversion coating was formed by initial
nucleation of isolated crystals of MgSnO3·H2O on the ␤ phase on
the alloy surface. Once the nuclei coalesced with each other, their
growth presumably stopped, and the coating reaction occurred on
the areas that either remained free of nuclei or were dotted with
dispersive tiny nuclei. Finally, further nucleation of MgSnO3·H2O
and growth of the pre-existing nuclei of MgSnO3·H2O resulted in
rapid coverage of the surface by MgSnO3·H2O spherical particles up
to the whole surface. The AZ91D magnesium alloy substrate, homo-
geneously covered by composite coating, was shown in Fig. 2(c). It is
well known that cracking is a common phenomena in sol–gel coat-
ing [13], uniformly distributed micro-cracks could also be observed
on this coating. For the composite coating, cracking could be mainly
ascribed to the thickness of the sol–gel coating required to cover the
pores in the stannate conversion coating or the sintering strain of
sol–gel ZrO2 coating during thermal treatment. Although the coat-
ing cracks, the metal surface is still protected, it will be expatiated
in the following text.
obvious and similar reduction in the anodic current density. It is
also obvious that the sol–gel ZrO2 coated specimen shows the best
ant-corrosion performance. After stannate conversion treatment,
corrosion potential (Ecorr) increased to −1.65 V, and the anodic
curve moved to lower current density side, which implied that
the anodic process was restrained by the stannate conversion coat-
ing. When the alloy was treated by means of the composite-coated
approach, the corrosion potential (Ecorr) showed a significant shift
to the positive direction and increased by 310 mV approximately by
contrast with the bare alloy. The corrosion potential (Ecorr), corro-
sion current (Icorr), polarization resistance (RP) and corrosion rate
(C.R.) were measured on the polarization curves by using Tafel
extrapolation methods as is shown in Table 1. In Table 1, it can
be seen that the Icorr value of stannate-coated specimen occurs two
order of magnitude lower than bare magnesium alloy substrate. For
the sol–gel ZrO2 coated specimen, Icorr further reduces one order
magnitude as compared to stannate-coated specimen. Observation
of a significantly less Icorr value for the sol–gel ZrO2 coated specimen
attributes the beneficial effect of stannating in the improvement of
3.3. Corrosion behavior
3.3.1. Polarization curve
Fig. 3 shows the potentiodynamic polarization curves of the
AZ91D magnesium alloy substrate, the stannate-coated specimen
and the sol–gel ZrO2 coated specimen respectively in 5 × 10−4 M
H2SO4 solution at room temperature. As shown in the figure, a con-
siderable ennoblement in the corrosion potential is seen for all the
treated samples and each of the treated specimens gives rise to
Fig. 3. The potentiodynamic polarization curves of the specimens in 5 × 10−4
M
Fig. 4. EIS spectra for the bare substrate for immersion times up to 96 h in 3.5 wt.%
sodium chloride solution.
H2SO4 solution.

Q. Li et al. / Journal of Alloys and Compounds 478 (2009) 544–549
547
corrosion resistances of ZrO2 ceramic film. This implied that com-
pared with the bare specimen, the chemical conversion treatment
improved the corrosion resistance of the AZ91D alloy to some extent
and the composite coating could provide the best protection to
the substrate. This may be that the sol–gel ZrO2 coated specimen
had a nanocrystalline structure which could serve as a considerably
effective barrier against corrosion electrolyte.
3.3.2. EIS test
The electrochemical impedance spectroscopy (EIS) is a non-
destructive powerful electrochemical method of testing for the
evaluation of corrosion resistance properties of sol–gel coated
substrate [14,15]. EIS represents the way in which charges are trans-
ferred or impeded between various phases, which may be studied
by the status and behavior of interfaces among conducting phases
[16]. In order to check the reproducibility of results, at least five
samples were measured for each coating. EIS measurements were
also carried out on the bare substrate as a comparison.
Fig. 4 presents the evolution of the EIS spectra of the bare sub-
strate after different times of immersion in 3.5 wt.% sodium chloride
solution, revealing a well defined time constant around 500 Hz at
the beginning of immersion tests. It is clear that the impedance
magnitude (|Z|) increases with immersion time at the beginning,
from 1 to 48 h, due to the formation of a passive film of corrosion
products on the substrate surface. However, longer immersion leads
to degradation of the passive film because of the starting of localized
corrosion attack causing gradually decrease of impedance magni-
Fig. 6. EIS spectra for the composite coating during 192 h of immersion in 3.5 wt.%
sodium chloride solution.
tude (|Z|). These results are similar to that of literature [17] which
employed AZ31B magnesium alloy as the substrate and 0.005 M
sodium chloride solution as electrolyte.
The EIS spectra obtained for the stannate-coated specimen
(Fig. 5) show different shapes during immersion. After 24 h of
immersion, the impedance magnitude (|Z|) of the coating shows
a remarkable reduction, nearly close to the impedance magnitude
(|Z|) of the bare substrate, which may be attributed to pitting corro-
sion resulting from the electrolyte partially penetrating the film
through the cracks existing on the stannate conversion coating.
This suggests that the stannate conversion coating, in some way,
does not act as an effective barrier to prevent the penetration of
the electrolyte. According to the figure, the impedance magnitude
(|Z|) increases with 148 h immersion time, may be the formation of
a passive film of corrosion products on the substrate surface.
The impedance spectra of the specimen treated with compos-
ite coating are shown in Fig. 6. As shown in the figure, it can be
seen that after 1 h of immersion, the spectrum reveals the broadest
phase angle and highest impedance value, which can be ascribed
to good protective property of sol–gel ZrO2 ceramic film rather
than the stannate conversion coating. Over half an order of mag-
nitude of impedance reduces during the immersion period from
1 to 48 h, may attribute to that electrolyte penetrates through the
ZrO2 ceramic film. After prolonged immersion time, the electrolyte
containing Cl− continues to attack the composite coating, and the
composite coating reveals a decrease of resistance which main-
tains a impedance magnitude (|Z|) that protect the substrate from
corrosion. After 96 h immersion, messy dots appear in the low fre-
Fig. 5. EIS spectra for the stannate-coated specimen for immersion times up to 144 h
in 3.5 wt.% sodium chloride solution.

548
Q. Li et al. / Journal of Alloys and Compounds 478 (2009) 544–549
quency. This may attribute to that as prolonged immersion time,
the electrolyte penetrated the composite coating and interact with
the substrate at last. After 192 h immersion, the impedance mag-
nitude (|Z|) of the coating remained higher than that of the bare
substrate, which indicates that the severe corrosion have taken
place in the local area of the composite coating. Generally, the
impedance magnitude of the composite coating decreases slightly,
which illuminates that the composite coating is stable and the cor-
rosion resistance can be maintained for long immersion time in
3.5 wt.% sodium chloride solution.
In order to rationalize the corrosion resistance mechanism
attributed by the composite coating, EIS equivalent circuits are con-
structed in Fig. 7 (after 48 h) and Fig. 8 (after 192 h). In Fig. 7, we
consider that Rs represents solution resistance, Rc refers to the inter-
face resistance between work electrode and reference electrode in
parallel with Cc, RZiror and RStan represent resistance of the sol–gel
ZrO2 ceramic film and the stannate conversion coating in paral-
lel with QZiror and QStan, respectively. After 48 h of immersion in
3.5 wt.% sodium chloride solution results in electrolyte diffusing
through the pores or defects of the ZrO2 ceramic film and the elec-
trolyte reaches the intermediate layer. At this time, the impedance
magnitude (|Z|) is nearly equivalent to that of the stannate con-
version coating, it can be seen in Figs. 5 and 6. Therefore, the
intermediate layer behaves like a capacitor because there is no
uptake of the electrolyte in the intermediate layer; hence, there
is no corrosion of substrate.
After 192 h of immersion, the electrolyte has interacted with
the substrate and the corresponding equivalent circuit is shown
Fig. 8. Equivalent circuit (a) used for numerical simulation of the EIS spectra; (b)
fitted spectra for composite coating after 192 h of immersion in 3.5 wt.% sodium
chloride solution.
in Fig. 8. The fit obtained with the equivalent circuit is very good
both for the impedance module and for the phase angle (Fig. 8b).
Rcoat refers to the resistance of the composite coating in parallel
with Ccoat, Rpore and Rct are respectively resistance of the pore exist-
ing in the composite coating and the charge transfer resistance, L
is the inductance. It can be known that in Fig. 8 that after 192 h
of immersion in 3.5 wt.% sodium chloride solution, the electrolyte
penetrates the composite coating and interact with the substrate.
Eventually, the coating develops further conductive pathways and
becomes saturated.
Compared with the result of the potentiodynamic polarization
experiment, the trend shown by the results of EIS is similar. Thus,
applying the sol–gel ZrO2 ceramic film on chemically pre-treated
AZ91D magnesium alloy is feasible.
4. Conclusions
In summary, we have successfully developed an environmental-
friendly novel approach to improve the anti-corrosion properties of
the AZ91D magnesium alloy. It is determined by the analysis of XRD
that nanocrystalline ZrO2 ceramic film had been obtained on pre-
stannate coated specimen using the sol–gel dip-coating technique,
and the ZrO2 ceramic film is mainly composed of tetragonal phase
ZrO2. In H2SO4 corrosion electrolyte, the ZrO2 ceramic film exhib-
ited a slow corrosion rate of 0.04 mpy and the corrosion potential
(Ecorr) increased by 310 mV compared with the bare magnesium
Fig. 7. Equivalent circuit (a) used for numerical simulation of the EIS spectra; (b) fit-
ted spectra for composite coating after 48 h of immersion in 3.5 wt.% sodium chloride
solution.

Q. Li et al. / Journal of Alloys and Compounds 478 (2009) 544–549
549
alloy, while the resistance value of the composite coating enhanced
one and a half orders of magnitude than bare magnesium alloy
substrate in 3.5 wt.% sodium chloride solution.
[5] M. Guglielmi, J. Sol–Gel Sci. Technol. 8 (1997) 443–449.
[6] G. Gusmano, G. Montesperelli, M. Rapone, G. Padeletti, A. Cusma, S. Kaciulis, A.
Mezzi, R.D. Maggio, Surf. Coat. Technol. 201 (2007) 5822–5828.
[7] F. Andreatta, P. Aldighieri, L. Paussa, R. Di Maggio, S. Rossi, L. Fedrizzi, Elec-
trochim. Acta 52 (2007) 7545–7555.
[8] L. Fedrizzi, F.J. Rodriguez, S. Rossi, F. Deflorian, R. Di Maggio, Electrochim. Acta
46 (2001) 3715–3724.
Acknowledgements
[9] H.H. Elsentriecy, K. Azumi, H. Konno, Surf. Coat. Technol. 202 (2007) 532–
537.
The authors thank the supports of the Natural Science Foun-
dation of Chongqing, China (CSTC. 2005BB4055) and High-Tech
Cultivation Program of Southwest Normal University (No. XSGX06).
[10] C.S. Lin, H.C. Lin, K.M. Lin, W.C. Lai, Corros. Sci. 48 (2006) 93–109.
[11] H.H. Elsentriecy, K. Azumi, H. Konno, Electrochim. Acta 53 (2007) 1006–1012.
[12] H. Huo, Y. Li, F. Wang, Corros. Sci. 46 (2004) 1467–1477.
[13] P. Shi, W.F. Ng, M.H. Wong, F.T. Cheng, J. Alloy Compd., in press.
[14] J. Gallardo, A. Durán, J.J. de Damborenea, Corros. Sci. 46 (2004) 795–806.
[15] A. Conde, A. Durán, J. de Damborenea, Prog. Org. Coat. 46 (2003) 288–296.
[16] B.E. Conway, Electrochemical Supercapacitors Scientific Fundamentals and
Technological Applications, Kluwer Academic/Plenum Publishers, New York,
1999, pp. 210–211.
References
[1] J.E. Gray, B. Luan, J. Alloy Compd. 336 (2002) 88–113.
[2] T.S. Shih, J.B. Liu, P.S. Wei, Mater. Chem. Phys. 104 (2007) 497–504.
[3] H.F. Guo, M.Z. An, Appl. Surf. Sci. 246 (2005) 229–238.
[4] O. Lunder, J.C. Walmsley, P. Mack, K. Nisancioglu, Corros. Sci. 47 (2005)
1604–1624.
[17] S.V. Lamaka, M.F. Montemor, A.F. Galio, M.L. Zheludkevich, C. Trindade, L.F. Dick,
M.G.S. Ferreira, Electrochim. Acta 53 (2008) 4773–4783.