ISIJ International, Vol.

56 (2016),
ISIJ International,
No. 12 Vol. 56 (2016), No. 12, pp. 22672275

Effect of Phosphate and Chromate Pigments on Sacrificial

Corrosion Protection by AlZn Coating and Delamination
Mechanism of Pre-painted Galvalume Steel

Yu SUGAWARA,1)* Kohei ARAAKE,1,2) Izumi MUTO,1) Michiyasu TAKAHASHI,3) Masamitsu MATSUMOTO3) and
Nobuyoshi HARA1)

1) Department of Materials Science, Tohoku University, 6-6-02 Aza-Aoba, Aramaki, Aoba-ku, Sendai, 980-8579 Japan.
2) Frontier Research Institute for Interdisciplinary Sciences, Tohoku University, 6-3 Aza-Aoba, Aramaki, Aoba-ku, Sendai, 980-
8578 Japan. 3) Steel Research Laboratories, Nippon Steel and Sumitomo Metal Corporation, 20-1 Shintomi, Futtsu,
Chiba, 293-8511 Japan.
(Received on May 31, 2016; accepted on August 10, 2016; J-STAGE Advance published date: October
27, 2016)

The effect of Mg3(PO4)2 and SrCrO4 pigments in paint layer on the corrosion behavior of 55 mass%
AlZn coating at the cut edges and the delamination mechanism of the organic paint at cross-cut areas of
pre-painted galvalume steels were examined using wet/dry cycle corrosion tests. The Mg3(PO4)2 pigment
did not inhibit white rust formation at the cut edges and the delamination of the organic paint. On the
other hand, the SrCrO4 pigment suppressed white rust formation and improved the delamination resis-
tance of the organic paint. Observations of the cross sections of the cross-cut area indicate that delamina-
tion was the result of the formation of corrosion products in the AlZn layer under the organic paint layer,
thus the SrCrO4 pigment inhibited the delamination of the organic paint layer by reducing the corrosion
rate of the AlZn layer. The Kelvin probe force microscopic measurements revealed that while the addition
of the anti-corrosive pigments increased the corrosion potential of the AlZn layers, it did not affect the
electrochemical properties of the steel substrates. The increases in the corrosion potential can be attrib-
uted to the barrier effect of the corrosion products containing chromate pigments on the AlZn layer
against anodic dissolution. From these results, it is reasonable that chromate pigments effectively increase
the delamination resistance of organic paint layers on pre-painted galvalume steels under atmospheric
corrosion conditions. In addition, phosphate pigments would improve the corrosion resistance of AlZn
layers in atmospheric environments with low chloride concentrations.

KEY WORDS: pre-painted galvalume steel; sacrificial corrosion protection; delamination; phosphate;
chromate; Kelvin probe force microscopy.

natives to chromate based pigments. Zinc phosphate

1. Introduction
Pre-painted hot-dip 55 mass% AlZn alloy coated (galval-
ume) steels are widely used as architectural and automotive
materials due to their high weatherability in atmospheric
environments. The high durability of pre-painted galvalume
steels is derived from the sacrificial protection of steel by
the AlZn coating layer, and anti-corrosive pigments are
added to the primer layers to improve the delamination of
the organic paint. Strontium chromate (SrCrO4) is typical of
such anti-corrosive pigments, and effective against filiform
corrosion on pre-painted steels at cut edges and coating
defects.14) However, due to concerns regarding the toxicity
of chromium (VI)-containing compounds, the use of SrCrO4
has been restricted,5) and the search for an alternative to
chromate base pigments has been extensive.1,614)
Phosphate pigments are traditional and promising alter-
* Corresponding author: E-mail: sugawaray@material.tohoku.ac.jp
DOI: http://dx.doi.org/10.2355/isijinternational.ISIJINT-2016-336
(Zn3(PO4)2) is applied to a conversion treatment for the
improvement of adhesion between organic paint layers
and steels.15,16) Because of the high barrier performance of
phosphate conversion coatings on steel substrates,17) it is
widely believed that phosphate pigment also functions as a
barrier type inhibitor against the dissolution of the AlZn
layer for pre-painted galvalume steels. Some researchers
have focused on the effect of phosphate pigments on the
corrosion resistance of Zn-coated steels.1823) Simes et
al. studied the corrosion inhibition mechanism of sodium
phosphate (Na3PO4) at the cut edges of electrogalvanized
steels using scanning vibration electrode technique (SVET)
and electrochemical impedance spectroscopy (EIS) in phos-
phate-containing 0.1 M NaCl solution. They summarized
that Zn3(PO4)2, which was formed by the reaction between
phosphate and Zn ions, inhibited the anodic dissolution of
Zn;18,19) however, the protective performance of Zn3(PO4)2
was lower than that of zinc chromate.20) Zin et al. reported
that the deposition of compact calcium/zinc phosphate

2267 2016 ISIJ

 ISIJ International, Vol. 56 (2016), No. 12
species on the Zn coating layer and the steel substrate
highly improved the efficiency of the inhibition effect of
phosphates on cut-edge corrosion.2123) These studies were
carried out in immersion conditions in aqueous solutions.
As such, the corrosion protection mechanisms of phosphate
pigments on pre-painted steels in atmospheric environments
remain unclear.
Because a better understanding of the role of chromates
in the corrosion inhibition supports the development of
alternatives, the effect of chromates on the cut-edge cor-
rosion has been investigated widely. Chromate pigments
are electrochemically reduced in aqueous solutions and are
deposited as Cr(III) oxyhydroxide on the metal surface.2426)
The chromate is believed to affect the galvanic reaction
between the AlZn coating layer and the steel substrate.
Apparently, the delamination resistance of the paint layer is
improved by chromate through the galvanic reaction at cut
edges and cross-cut area.3,4,23) It should be noted, however,
that a detailed understanding of the mechanism has yet to
be presented. The purpose of this study is to elucidate the
corrosion protection mechanism of phosphate and chromate
in the primer layer of pre-painted hot-dip 55 mass% AlZn
alloy coated steel. We focused on the effect of these anti-
corrosive pigments on the electrochemical properties of the
AlZn layer and the steel substrate, and also investigated
the delamination mechanism of the organic paint layer under
atmospheric corrosion conditions.
Kelvin probe force microscopy (KPFM) is a powerful
tool in the analysis of corrosion reactions under thin elec-
trolyte layers in atmospheric corrosion conditions. KPFM,
which is a kind of scanning probe microscopy, can obtain
the Volta potential difference (Kelvin potential) by measur-
ing the electrostatic interaction between a KPFM probe and
a specimen surface. Schmutz et al. demonstrated the cor-
rosion potentials for pure metals in a chloride-containing
solution and their Kelvin potentials in the air after the
immersion followed a linear relationship,27,28) suggesting it
is possible to estimate the corrosion potential distribution
of metals under galvanic coupling from the Kelvin potential
map immediately after corrosion tests. In this study, mag-
nesium phosphate (Mg3(PO4)2) and SrCrO4 were selected
as the anti-corrosive pigments. This is because Mg3(PO4)2
pigment is useful for corrosion inhibition of ferrous met-
als29,30) and SrCrO4 is well known as a typical anti-corrosive
pigment. Wet/dry cycle tests were conducted to clarify the
effect of the anti-corrosive pigments on the cut-edge cor-
rosion behavior and the delamination of the organic paint.
The Kelvin potential distribution at the interface between
the AlZn coating layer and the steel substrate in the cut-
edge specimen under atmospheric corrosion conditions was
measured by KPFM. Polarization curves of the AlZn coat-
ing layers covered with corrosion products were taken to
elucidate the role of the anti-corrosive pigments in cut-edge
corrosion and the delamination of the organic paint.
2. Experimental
2.1. Specimens
Pre-painted hot-dip 55 mass% AlZn alloy coated
(galvalume) steel sheets with and without anti-corrosive
pigments were used as specimens. Either Mg3(PO4)2 or
2016 ISIJ
SrCrO4 was added to the primer layer as the anti-corrosive
pigment. A galvalume steel sheet (0.6 mm in thickness) with
approximately 20 m of 55 mass% AlZn alloy layers on
both sides was used as the starting material. Prior to paint-
ing, no conversion treatment was performed on the AlZn
surface. The front of the galvalume steels was painted with
a polyester primer ca. 5 m in thickness with and without
the anti-corrosive pigments. The concentration of the anti-
corrosive pigments in the primer layer was adjusted to 20
mass%. Polyester topcoat layers of 510 m in thickness
were then formed on both sides of the sheet.
2.2. Wet/dry Cycle Test for Cut Edge Corrosion
The International Organization for Standardization (ISO)
16539 Method A was performed to simulate the initial stage
of atmospheric corrosion at the cut edges of the specimens.
Figure 1 shows the change in temperature and relative
humidity (RH) as a function of time in a wet/dry cycle of
ISO 16539 Method A. This is the simplified pattern derived
from the variations of temperature and RH on a steel sheet
exposed to an outdoor environment.31) The dew point of
water vapor in the air was kept at 301 K to reproduce
the summer season in Japan. It was demonstrated that the
atmospheric corrosion behavior for pre-painted galvanized
and galvalume steels can be reproduced using this corro-
sion test.3,4) An ESPEC SH221 programmable humidified
chamber was used for the corrosion test.
Figure 2(a) shows the model cut-edge specimen used for
this corrosion test for the wet/dry cycle test. The pre-painted
galvalume steels were cut into 10 mm 15 mm squares,
and the back side of the steels was polished to remove the
AlZn layer and paint layers. After being polished, the
specimen was embedded in an epoxy resin as shown in
Fig. 2(a), and the cross section of the pre-painted galvalume
steel was polished with 1 500 grit SiC paper. The surface
was covered with a polyimide tape except for the cut-edge
region (5 mm 15 mm). A thin electrolyte layer of 500 m
thickness was formed on the cut-edge region at the begin-
ning of the corrosion test. The amount of Cl ion deposition
on the specimen was set to 0.1 g m 2 using 5.64 mM NaCl
solution to simulate marine environments. All the solutions
in this study were prepared from deionized water and ana-
Fig. 1. Change in temperature and relative humidity (RH) as a
function of time in the wet/dry cycle test (ISO 16539
Method A).
2268
 ISIJ International, Vol. 56 (2016), No. 12
Fig. 2. (a) Model cut-edge specimen used in the wet/dry cycle test
(ISO 16539 Method A). (b) Cross-cut specimen used in the
wet/dry cycle test (JASO M 609-91). (Online version in
color.)
lytical grade chemicals. Before and after 1 cycle of the wet/
dry cycle test, the surface appearance of the cut edges was
photographed by a digital camera.
2.3. Wet/dry Cycle Test for Paint Delamination
The Japanese Automobile Standards Organization (JASO)
M 609-91, which is a cyclic corrosion test with salt spraying,
was carried out to evaluate the effect of the anti-corrosive
pigments on delamination of the organic paints at the cross-
cut area. The salt water for JASO M 609-91 was 0.856 M
NaCl. Because the chloride concentration of the salt water
for JASO M 609-91 was higher than that for ISO 16539
Method A, JASO M 609-91 was suitable to evaluate the
delamination of the organic paints efficiently. Figure 2(b)
shows the cross-cut specimen used for this wet/dry test of
JASO M 609-91. The cross-cut was made with a mechani-
cal device and a folded back. The cyclic conditions of the
JASO M 609-91 were as follows: 1) salt water spraying for
2 hours; 2) drying at 333 K and 2030% RH for 4 hours;
3) wetting at 323 K and 95% RH for 2 hours. The wet/dry
cycle was conducted for 60 times.
2.4. Kelvin Probe Force Microscopy
Kelvin probe force microscopy (KPFM) was carried out
using a Bruker Multimode 8 to measure the potential distri-
bution of the AlZn coating/steel interfaces. The pre-painted
galvalume steels were cut into 8 mm 15 mm squares, and
the back side of the steel sheet was polished to remove the
AlZn and paint layers. After that, the steel sheets were
embedded in an epoxy resin, and then the cross-section
of the pre-painted galvalume steel was polished with dia-
mond paste down to 1 m. Except for the cut-edge region
(5 mm 5 mm), the specimen surface was covered with
polyimide tape. In order to simulate the initial stage of the
sacrificial protection, the specimens were corroded under a
thin electrolyte layer at 301 K and 90% RH for 1 hour in
the humidified chamber. Because KPFM were not suitable
for the measurements of the specimen with many corrosion
products, 5.64 mM NaCl was used to form an electrolyte
2269
layer 500 m in thickness. After the corrosion test, KPFM
measurements of the specimen were carried out immedi-
ately without any water-rinse treatment. The tip used was a
PtIr coated and electrically conductive Si probe (Bruker,
SCM-PIT), with tip radius, resonant frequency, and spring
constant of 2025 nm, 75 kHz, and 2.8 N m 1, respec-
tively. The topographical images were obtained by atomic
force microscopy (AFM) simultaneously with the potential
distributions by KPFM. The scan rate of the tips during the
KPFM and AFM measurements was 0.2 Hz.
2.5. Polarization Measurements
In order to discuss the Kelvin potential of the AlZn
coated layers in terms of the effect of the anti-corrosive
pigments, both anodic and cathodic polarization curves
were measured. Galvalume steel sheet without a paint
layer, which was the starting material for pre-painting, was
corroded for 3 cycles by ISO 16539 Method A. A thin
electrolyte layer was formed using 5.64 mM NaCl, 5.64
mM NaCl5.64 mM SrCrO4, or 5.64 mM NaCl9.88 M
Mg3(PO4)2 solution. In this case, 9.88 M Mg3(PO4)2 was
the saturated concentration of Mg3(PO4)2 in water at 298 K.
The anti-corrosive pigments were thought to be introduced
into corrosion products, and their electrochemical properties
were probably modified. The surface area of the working
electrodes was 10 mm squares. 0.1 M NaCl at 298 K was
used as the electrolyte in the polarization measurements.
While the chloride concentration of the electrolyte for the
polarization curves was higher than that for the KPFM mea-
surements, 0.1 M NaCl is a commonly used for evaluation
of the corrosion property in chloride-containing environ-
ments. The reference electrode was an Ag/AgCl (3.33 M
KCl) electrode, and the counter electrode was a Pt sheet.
The anodic and cathodic polarization measurements were
performed in deaerated and air-saturated solutions, respec-
tively. The scan rate of the electrode potential was 3.8
10 4 V s 1 (23 mV min 1).
2.6. Surface and Cross Section Observations
The surface appearances of the specimens before and
after the wet/dry corrosion tests were taken by a Nikon D1
digital camera. Cross sections of the pre-painted galvalume
steels were observed by a JEOL JSM-7100F scanning elec-
tron microscope (SEM) equipped with a JEOL JED-2300
energy-dispersive X-ray spectrometer (EDS). Prior to the
SEM observation, the cross-section of the specimens were
prepared by a JEOL IB-09010CP cross section polisher
(Ar + ion milling), and a Au thin film was deposited on the
observed areas to avoid the charge up of the specimens.
3. Results and Discussion
3.1. Distributions of Anti-corrosive Pigments in Primer
Layer
To confirm the particle size and distribution morphology
of the anti-corrosive pigments in the primer layer, SEM/
EDS analyses were performed. Figures 3(a) and 3(b) show
the cross-sectional SEM images of the specimens with the
Mg3(PO4)2 and SrCrO4 pigments. The large white (pale
gray) particles observed in the primer layer were thought
to be the anti-corrosive pigments. Figure 3(c) indicates the
2016 ISIJ
 ISIJ International, Vol. 56 (2016), No. 12
comparison of the EDS spectra obtained from inside (Point
A1) and outside (Point A2) of the particle in the primer layer
with Mg3(PO4)2 pigment. The intensities of Mg, P, and O
at Point A1 were higher than those at Point A2, suggesting
Fig. 3. (a, b) Cross-sectional SEM images of the specimens with
Mg3(PO4)2 or SrCrO4 pigment in the primer layers. (c, d)
EDS spectra obtained from inside (Points A1 and B1) and
outside (Points A2 and B2) the particles in the primer lay-
ers shown in (a) and (b).
2016 ISIJ
that the large particles in the primer layers corresponded
to the Mg3(PO4)2 pigment. Also in the specimen with the
SrCrO4 pigment (Fig. 3(d)), the intensities of Sr, Cr, and O
inside the particle (Point B1) were higher than those outside
the particle (Point B2), implying that the large particles in
the primer layers corresponded to the SrCrO4 pigments. As
seen in Figs. 3(a) and 3(b), because pigments are normally
produced by milling processes,32) the pigment particles were
coarsely distributed over the range from 0.5 to 5 m.
In addition to the above characteristics of the distribu-
tion of the anti-corrosive pigments, the microstructure of
55% AlZn layer was confirmed to be dual-phase. It is well
known that the primary solidification phases are Al-rich with
a granular-shape, and that Zn-rich phases exist at the inter-
granular region. In Figs. 3(a), 3(b), the white (pale gray) areas
along grain boundaries were likely to be the Zn-rich phases.
3.2. White Rust Formation at Cut Edges
The wet/dry cyclic corrosion test (ISO 16539 Method A)
was performed to investigate the effect of Mg3(PO4)2 and
SrCrO4 pigments on the corrosion of the AlZn layer at
the cut edges. Figure 4 shows the surface appearances of
the model cut-edge specimens with and without the anti-
corrosive pigments after 1 cycle. The AlZn coating and
the organic paint layers are located in the lower side of the
images shown in Fig. 4. It was clearly observed that less
white rust formed on the SrCrO4-containg specimen than on
the Mg3(PO4)2-containing and the pigment-free specimens.
The addition of SrCrO4 was confirmed to prevent the Zn
dissolution at cut edges, but Mg3(PO4)2 was thought to be
relatively ineffective. The results obtained in this study are
consistent with those in the literature.3,4,21)
Fig. 4. Surface appearances of the model cut-edge specimens
with and without the anti-corrosive pigments in the primer
layer after 1 cycle of the wet/dry corrosion test (ISO 16539
Method A).
2270
 ISIJ International, Vol. 56 (2016), No. 12
3.3. Delamination of Paint Layer
The cyclic wet/dry test (JASO M 609-91) was carried out
to ascertain the effect of Mg3(PO4)2 and SrCrO4 pigments on
the delamination resistance of organic paint layers. Figure 5
shows the maximum delamination length of the paint layers
from the cross-cut on the pre-painted galvalume steels with
and without the anti-corrosive pigments after 60 cycles in
the wet/dry cycle test. The delamination resistance of the
organic paint layers at the cross-cut on the specimen was
improved by the addition of SrCrO4 pigment. On the other
hand, the delamination length of the paint layer with the
Mg3(PO4)2 pigment was comparable to that without the
anti-corrosive pigments. From the result of the cyclic wet/
dry corrosion tests (ISO 16539 Method A and JASO M 609-
91), it appears that the Mg3(PO4)2 pigment had little effect
on the corrosion resistance of the AlZn layer at cut edges
and the delamination resistance of the organic paint layers.
The cross-cut areas of the specimens after the cyclic wet/
dry test (JASO M 609-91) were observed by SEM to clarify
the delamination mechanisms of the organic paint layer. To
avoid chemical species dissolving in water by wet-polishing,
the cross section polisher (Ar + ion milling) was employed
in the preparation of the cross-sections. Figures 6(a)6(c)
show the cross-sectional SEM images and EDS maps in the
vicinity of the cross-cuts on the pre-painted galvalume steels
with and without the anti-corrosive pigments. The delami-
nated areas were marked by the horizontal thick arrows in
Figs. 6(a)6(c). In the case of the specimen without the
anti-corrosive pigments (Fig. 6(a)), signals of Cl and O were
found to co-exist at the region between the organic paint and
AlZn layers (marked by C1). In this co-existence region,
signals of Al and Zn were detected. It was likely that the
corrosion products of Al and Zn were generated under the
paint layer by the migration of Cl ions in the wet/dry cycle
test, which caused the delamination of the organic paint on
the AlZn layer. In addition to this, a characteristic feature
of corrosion morphology was observed in the AlZn layer.
The co-existence region of Cl and O signals was extended
into the AlZn layer (marked by C2). The intergranular Zn-
Fig. 5. Maximum delamination length of the organic paint from
the cross-cut on the pre-painted galvalume steels with and
without anti-corrosive pigments after 60 cycles of the wet/
dry (JASO M 609-91).
2271
rich phases were thought to corrode preferentially under the
wet/dry conditions, and the corrosion products of Zn, such
as ZnCl2, Zn(OH)2, Zn5(CO3)2(OH)6, Zn5(OH)8Cl2H2O etc.
were formed in the grain boundaries in the AlZn alloy
layer.33) These results show that volume expansion due to
the formation of corrosion products of Zn and Al between
the organic paint and the coating layers was likely to be the
main factor of the delamination of the organic paint layer.
Figure 6(b) shows the EDS maps of the specimen with the
Mg3(PO4)2-containing organic paint layer after the wet/dry
cycle test. Also in this case, signals of Cl and O co-existed,
and that of Mg was not detected in the delaminated region
(marked by D1). Moreover, even in the un-delaminated
(intact) region, Cl and O signals were detected (marked by
D2), implying that the formation of the corrosion products of
the AlZn layer caused the delamination of the organic paint
in the same way as the pigment-free specimen. According
to this mechanism, the corrosion rate of the AlZn layer is
the predominant factor affecting the delamination resistance
of the paint layers. As shown in Fig. 4, almost the same
amount of white rust formed at the cut edge on the specimen
with the Mg3(PO4)2 pigment as on the specimen without the
anti-corrosive pigments, which explains the similarity in the
delamination length of the specimen with the Mg3(PO4)2
pigment and that of the specimen without the anti-corrosive
pigments, as shown in Fig. 5.
Figure 6(c) exhibits the EDS maps of the specimen with
the SrCrO4-containing organic paint layer after the wet/dry
cycle test. In this case, the signals of O and Cl co-existed
in the delaminated region, and Sr and Cr did not exist
at the same region. This suggests that the specimen with
the SrCrO4-containing organic paint followed the same
delamination mechanism as that mentioned above. How-
ever, extremely low amounts of Cl and O were found for
the specimen with SrCrO4-containing pigment compared
to those with the Mg3(PO4)2 pigment and without the anti-
corrosive pigments. The results shown in Fig. 4 indicate
that the addition of SrCrO4 pigment was effective for the
suppression of the corrosion of the AlZn layer. Suppress-
ing the dissolution of the AlZn layer by anti-corrosive
pigments is concluded to be one of the key points in the
improvement of the delamination resistance of organic
paints on galvalume steels.
3.4. Potential Distribution at AlZn Layer/steel Inter-
faces
As mentioned above, the dissolution of the AlZn layer
results in the delamination of the organic paints, and hence
the effect of anti-corrosive pigments on the AlZn layers
provides important information regarding the corrosion pro-
tection mechanisms of the anti-corrosive pigments. Kelvin
potential distributions at the AlZn layer/steel substrate
interface were measured by the KPFM. Since the KPFM
measurements were conducted immediately after the cor-
rosion test at 301 K and 90% RH for 1 hour, the measured
Kelvin potentials are closely related to the corrosion poten-
tial on the specimen surfaces.27,28) The Kelvin potential (Vk)
measured by KPFM is referred to the PtIr tip in this study;
Vk is proportional to the corrosion potential.
Figure 7 shows the AFM height and KPFM potential
images at the AlZn layer/steel interface of the pigment-free
2016 ISIJ
 ISIJ International, Vol. 56 (2016), No. 12

Fig. 6. Cross-sectional SEM images and EDS maps in the vicinity of the cross-cuts on the pre-painted galvalume steels
after the cyclic wet/dry test of JASO M 609-91: (a) specimen without the anti-corrosive pigment, (b) with
Mg3(PO4)2 pigment, and (c) with SrCrO4 pigment in the primer layer.

Fig. 7. AFM height image and corresponding Kelvin potential map at the AlZn layer/steel interface of the pigment-
free specimen after the corrosion test for 1 h.

2016 ISIJ 2272

 ISIJ International, Vol. 56 (2016), No. 12

specimen. From the AFM height image, it was confirmed
that the dissolution of the AlZn layer in the vicinity of the
steel was relatively high, and the corrosion products were
deposited on the steel substrate and the AlZn layer. More
corrosion products were deposited on the steel substrate than
on the AlZn layer, suggesting that corrosion products were
mainly deposited along with the oxygen reduction reaction
on the steel substrate. It was confirmed that the galvanic
coupling between the AlZn layer and the steel substrate
takes place under a thin electrolyte layer in the atmospheric
corrosion test, which was performed prior to KPFM mea-
surements. The Kelvin potential of the steel substrate was
lower than that of the AlZn layer, indicating that the cor-
rosion potential of the steel substrate was higher than that
of the AlZn layer.
Figure 8 exhibits the AFM height and KPFM potential
images at the AlZn layer/steel interface of the specimen
with the Mg3(PO4)2 pigment. As was the case for the pig-
ment-free specimen, the AlZn layer dissolved to a depth of
ca. 2 m, and the finer corrosion products were deposited on
the steel substrate (marked by E1). The large particles were
thought to be NaCl from the electrolyte. A smaller Kelvin
potential difference was noted between the AlZn layer and
the steel of the specimen with the Mg3(PO4)2 pigment than
that of the pigment-free specimen. In particular, it appeared
that the Kelvin potential of the AlZn layer for the speci-
men with the Mg3(PO4)2 pigment was lower than that of the
specimen without the anti-corrosive pigments. On the other
hand, the Kelvin potential of the steel substrate was almost
the same as that of the specimen without the anti-corrosive
pigments.
Figure 9 shows the AFM height and KPFM potential
images at the AlZn layer/steel interface of the specimen
with the SrCrO4 pigment. From the AFM height image,
neither the dissolution of the AlZn layer nor the deposition
of corrosion products was severe in this case. This result
corresponds to the inhibition of the white rust formation at
the cut edge shown in Fig. 4. The Kelvin potential differ-
ence between the AlZn layer and the steel of the specimen
with the SrCrO4 pigment was also smaller than that of the
pigment-free specimen.
The line profiles of the Kelvin potentials along the black
arrows shown in Figs. 79 are summarized in Fig. 10. The
horizontal axis is the distance apart from the left sides of
the images of Figs. 79. From Fig. 10, it is seen that the
Mg3(PO4)2 and SrCrO4 pigments increased the corrosion
potentials of the AlZn layers. On the other hand, the corro-
sion potentials of the steel substrates were almost the same;
that is, the anti-corrosive pigments did not affect the poten-
tials of the steel substrates. In all the known earlier studies,
the protection mechanism of anti-corrosive pigments against
the corrosion of galvanized and galvalume steels has been
studied in aqueous solutions. In phosphate or chromate con-
taining solutions, a deposition layer composed of phosphates
or Cr(III) oxyhydroxides was probably formed on steel
substrate, inhibiting the oxygen reduction reaction.4,18,21,26)

Fig. 8. AFM height image and corresponding Kelvin potential map at the AlZn layer/steel interface of the specimen
with Mg3(PO4)2 pigment in the primer layer after the corrosion test for 1 h.

Fig. 9. AFM height image and corresponding Kelvin potential map at the AlZn layer/steel interface of the specimen
with SrCrO4 pigment in the primer layer after the corrosion test for 1 h.

2273 2016 ISIJ

 ISIJ International, Vol. 56 (2016), No. 12
Fig. 10. Effect of anti-corrosive pigments on the line profiles of
Kelvin potentials along the arrows shown in Figs. 79.
However, the KPFM results of this study indicate that the
corrosion potential of the steel substrates was not signifi-
cantly influenced by the anti-corrosive pigments; as such,
the pigments themselves appear to act as anodic inhibitors
that suppress the dissolution rate of the AlZn layer.
3.5. Effect of Anti-corrosive Pigments on Electrochemi-
cal Property of AlZn Layer
It is likely that the anti-corrosive pigments were intro-
duced into the corrosion products of the AlZn layer, and
that their electrochemical properties were modified. In
order to confirm that the anti-corrosive pigments act as the
anodic inhibitor of the AlZn layer, the anodic and cathodic
polarization curves of the 55 mass% AlZn layer (galval-
ume) covered by a white rust layer were measured in 0.1
M NaCl solution. The galvalume steel steels were corroded
under the thin electrolyte layers with and without the anti-
corrosive pigments, and then the corroded surface of the
AlZn layer was used as a working electrode. Figure 11
shows the experimental results of the polarization behavior.
As seen in this figure, there was little or no effect of the
anti-corrosive pigments on cathodic polarization behavior.
On the other hand, the anodic polarization curves of the
AlZn alloy layer covered with the white rust formed in
the SrCrO4-containing electrolyte layer differed from those
obtained from the specimens corroded in the electrolytes
with the Mg3(PO4)2 and without the pigments. It is clear
that the white rust generated in the SrCrO4-containing elec-
trolyte suppressed the anodic dissolution rate of the AlZn
layer. The addition of SrCrO4 pigment to the primer layer
is thought to increase the corrosion potential of the AlZn
layer; however, the corrosion potential of the AlZn layer
remained lower than that of the steel substrate. As a result,
the AlZn layer can offer a sacrificial corrosion protection
effect to the steel substrate. Furthermore, because the dis-
solution rate of AlZn layer is suppressed, the sacrificial
corrosion protection effect of the AlZn layer continues
for a long period of time even under thin electrolyte layers.
This explains how the SrCrO4 pigment in the primer layer
inhibited the white rust formation at cut-edge specimen and
the subsequent delamination of the organic paint layer, as
shown in Figs. 4 and 5.
As shown in Fig. 11, the anodic polarization behavior
of the rust layer formed in the Mg3(PO4)2-containing solu-
2016 ISIJ
Fig. 11. Anodic and cathodic polarization curves for the galval-
ume steel surfaces covered with white rust layer formed
in wet/dry cycle test (ISO 16539 Method A) for 3 cycles.
Corrosion tests were performed under a thin electrolyte
layer with and without anti-corrosive pigments
(Mg3(PO4)2 and SrCrO4). The anodic and cathodic polar-
ization measurements were carried out in 0.1 M NaCl at
298 K under deaerated and air-saturated conditions,
respectively.
tion was similar to that in the pigment-free solution. The
Mg3(PO4)2 pigment does not improve the protective ability
of the corrosion products of the AlZn layer in 0.1 M NaCl.
Simes et al. reported that a Zn3(PO4)2 film formed on the
Zn layer was locally destroyed in a chloride-containing solu-
tion, and the anodic dissolution of Zn occurred even when
the Zn surface was covered by the corrosion products.18) It
was concluded that the anodic dissolution of the AlZn was
not inhibited due to the local breakdown of the layer of the
corrosion products on the AlZn surface in the polariza-
tion measurements. The polarization measurements were
performed in 0.1 M NaCl solution, and the cyclic wet/dry
test (JASO M 609-91) was carried out using 0.856 M NaCl
solution as the salt water. On the other hand, because the
KPFM measurements were conducted after the corrosion
test under the electrolyte layer of 5.64 mM NaCl solution,
the Mg3(PO4)2 pigment was thought to increase the cor-
rosion potential of the AlZn layer. It was expected that
the Mg3(PO4)2 pigment addition in the primer layer would
improve the atmospheric corrosion resistance of the pre-
painted galvalume steels in a specific environment with low
chloride concentrations.
4. Conclusions
(1) Less white rust formed on the pre-painted 55 mass%
AlZn alloy coated (galvalume) steel with the SrCrO4 pig-
ment at the cut edge than on those with the Mg3(PO4)2
and without anti-corrosive pigments in the wet/dry cyclic
corrosion test (ISO 16539 Method A). In addition, less
delamination was noted in the organic paint layer of the
SrCrO4-containing cross-cut specimen than in that of the
Mg3(PO4)2-containing and the pigment-free specimens in
the wet/dry cyclic corrosion test (JASO M 609-91).
(2) The delamination of the organic paint layers for
the pre-painted galvalume steels can be attributed to the
formation of the corrosion products of Al and Zn between
2274
 ISIJ International, Vol. 56 (2016), No. 12
the organic paint and the AlZn coating layer. The SrCrO4
pigment inhibited the delamination of the organic paint layer
by reducing the corrosion rate of the AlZn layer.
(3) The Mg3(PO4)2 and SrCrO4 pigments increased the
corrosion potentials of the AlZn coated layers, and did not
affect the potentials of the steel substrates. This indicates
the Mg3(PO4)2 and SrCrO4 pigments act as anodic inhibitors.
(4) Chromate pigments reduce the corrosion rate of Al
Zn layers, and increase their corrosion potential. However,
because the corrosion potential of AlZn layers remains
lower than that of steel substrates, the duration of the sac-
rificial protection by AlZn coated layer for anti-corrosive
pigments-containing specimens is longer than that for
pigments-free specimens. It is expected that chromate pig-
ments effectively improve the delamination resistance of
organic paint layers for pre-painted galvalume steels under
atmospheric corrosion conditions.
(5) The anodic polarization curves in 0.1 M NaCl
showed that the corrosion products generated in the
Mg3(PO4)2-containing electrolyte did not suppress the
anodic dissolution of the AlZn layer. However, because the
Mg3(PO4)2 pigment increased the corrosion potential of the
AlZn layer, it is expected that phosphate pigments would
enhance the corrosion resistance of pre-painted galvalume
steels in atmospheric environments with low chloride con-
centrations.
REFERENCES
1) J. Sinko: Prog. Org. Coat., 42 (2001), 267.
2) A. Nishikata, T. Tsuda and T. Tsuru: Tetsu-to-Hagan, 95 (2009),
565.
3) I. Muto, K. Sato and N. Hara: ECS Trans., 25 (2010), No. 29, 59.
4) A. Chiba, I. Muto and N. Hara: J. Electrochem. Soc., 158 (2011),
C42.
5) S. A. Katz and H. Salem: J. Appl. Toxicol., 13 (1993), 217.
6) A. Seth, W. J. van Ooij, P. Puomi, Z. Yin, A. Ashirgade, S. Bafna
and C. Shivane: Prog. Org. Coat., 58 (2007), 136.
7) S. P. V. Mahajanam and R. G. Buchheit: Corrosion, 64 (2008), 230.
2275
8) D. Snihirova, S. V. Lamaka, M. Taryba, A. N. Salak, S. Kallip, M.
L. Zheludkevich, M. G. S. Ferreira and M. F. Montemor: ACS Appl.
Mater. Interfaces, 2 (2010), 3011.
9) J. Carneiro, J. Tedim, S. C. M. Fernandes, C. S. R. Freire, A. J. D.
Silvestre, A. Gandini, M. G. S. Ferreira and M. L. Zheludkevich:
Prog. Org. Coat., 75 (2012), 8.
10) M. L. Zheludkevich, S. K. Poznyak, L. M. Rodrigues, D. Raps, T.
Hack, L. F. Dick, T. Nunes and M. G. S. Ferreira: Corros. Sci., 52
(2010), 602.
11) S. Bohm, H. N. McMurray, S. M. Powell and D. A. Worsley: Mater.
Corros., 52 (2001), 896.
12) G. Williams, H. N. McMurray and D. A. Worsley: J. Electrochem.
Soc., 149 (2002), B154.
13) G. Williams, H. N. McMurray and M. J. Loveridge: Electrochim.
Acta, 55 (2010), 1740.
14) G. Williams, H. N. McMurray and A. Bennett: Mater. Corros., 65
(2014), 401.
15) D. B. Freeman: Phosphating and Metal Pre-treatment, Woodhead-
Faulkner, Cambridge, (1986), 52.
16) T. S. N. Sankara Narayanan: Rev. Adv. Mater. Sci., 9 (2005), 130.
17) D. Weng, P. Jokiel, A. Uebleis and H. Boehni: Surf. Coat. Technol.,
88 (1996), 147.
18) A. M. Simes, J. Torres, R. Picciochi and J. C. S. Fernandes:
Electrochim. Acta, 54 (2009), 3857.
19) A. M. Simes and J. C. S. Fernandes: Prog. Org. Coat., 69 (2010),
219.
20) A. C. Bastos, M. G. S. Ferreira and A. M. Simes: Prog. Org. Coat.,
52 (2005), 339.
21) I. M. Zin, V. I. Pokhmurskii, J. D. Scantlebury and S. B. Lyon: J.
Electrochem. Soc., 148 (2001), B293.
22) I. M. Zin, S. B. Lyon and V. I. Pokhmurskii: Corros. Sci., 45 (2003),
777.
23) I. M. Zin, S. B. Lyon, V. I. Pokhmurskii and M. C. Simmonds:
Corros. Eng. Sci. Technol., 39 (2004), 167.
24) A. F. Diaz and D. Schermer: J. Electrochem. Soc., 132 (1985), 2571.
25) G. Lindbergh and D. Simonsson: Electrochim. Acta, 36 (1991), 1985.
26) W. J. Clark and R. L. McCreery: J. Electrochem. Soc., 149 (2002),
B379.
27) P. Schmutz and G. S. Frankel: J. Electrochem. Soc., 145 (1998),
2285.
28) P. Schmutz and G. S. Frankel: J. Electrochem. Soc., 145 (1998),
2295.
29) L. Chromy and E. Kamiska: Prog. Org. Coat., 18 (1990), 319.
30) Coating Compositions with Magnesium Phosphate Pigment, US
Patent 3, 960, 611, (1976).
31) I. Muto and K. Sugimoto: Zairyo-to-Kankyo, 47 (1998), 519.
32) F. J. Maile, G. Pfaff and P. Reynders: Prog. Org. Coat., 54 (2005),
150.
33) C. Leygraf and T. E. Graedel: Atmospheric Corrosion, John Wiley &
Sons, New York, (2000), 131.
2016 ISIJ