Arabian Journal of Chemistry (2015) 8, 821–827

King Saud University

Arabian Journal of Chemistry

www.ksu.edu.sa
www.sciencedirect.com

ORIGINAL ARTICLE

Investigation of the cut-edge corrosion

of organically-coated galvanized steel after
accelerated atmospheric corrosion test
_
Resßit Yıldız *, Ilyas Dehri

Department of Chemistry, Cukurova University, 01330 Adana, Turkey

Received 7 July 2012; accepted 17 April 2013

Available online 25 April 2013

KEYWORDS Abstract The cut edge corrosion of organically coated (epoxy, polyurethane and polyester) galva-
Organic coatings; nized steel was investigated using electrochemical impedance spectroscopy (EIS). Measurements
Steel; were performed on specimens that had been tested in an accelerated atmospheric corrosion test.
Zinc; The samples were subjected to 10 s fogging and 1 h awaiting cycles in an exposure cabinet (120
EIS; and 180 days) with artificial acid rain solution. According to the investigation, the coatings were
Corrosion damaged from the cut edge into the sheet, this distance was about 0.8 cm. These defects were more
pronounced at after 180 days in proportion to after 120 days.
ª 2013 Production and hosting by Elsevier B.V. on behalf of King Saud University.

1. Introduction The use of organically coated steel products in a wide vari-

Atmospheric corrosion is mostly due to the presence of salt
and gases dissolved in a thin layer on a metal surface. The
atmospheric corrosion rate of metals depends on the thickness
of the electrolyte layer (Tomashow, 1964). Changes in the
thickness of the layer affect the rate of oxygen transport
through the electrolyte layer and solubility of corrosion prod-
ucts and, hence, the metal corrosion rates (Nishikata et al.,
1995).
* Corresponding author. Tel.: +90 (322) 3386084 2465; fax: +90
(322) 3386070.
E-mail address: ryildiz80@gmail.com (R. Yıldız).
Peer review under responsibility of King Saud University.
Production and hosting by Elsevier
1878-5352 ª 2013 Production and hosting by Elsevier B.V. on behalf of King Saud University.
http://dx.doi.org/10.1016/j.arabjc.2013.04.023
ety of industries has grown dramatically, recently, which has
led inevitably to increasing demands on performance. As a re-
sult, it is vitally important, both from a commercial and envi-
ronmental perspective, to optimize products’ lifetime by
minimizing degradation. In organically coated galvanized steel
materials used in the construction industry, one of the most
frequently reported modes of failure is corrosion at the metal-
lic cut edges (Ryan et al., 1994). Coil coatings have long been
used to protect metals and alloys against corrosion. In spite of
the fact that such coatings form a barrier against diffusion of
ions, water and oxygen are considerably more permeable and,
therefore, are able to be present at the metal/coating interface.
This can lead to substrate corrosion and coating delamination
under the same conditions. Profiled galvanized steel sheet is
commonly used for roof and wall cladding. This material is
mostly factory-coated in coil; firstly, with a primer layer
containing corrosion inhibitors such as strontium chromate
and after that with a thicker top coat for additional barrier
822
protection and the desired esthetic appearance. At the cut
edge, however, the bare metal is exposed to the atmosphere,
and corrosion may occur. There is a considerable interest in
limiting this problem as it accounts for a significant proportion
of the failures of coil-coated cladding according to recent
industry surveys (Cox et al., 1985; Jones and Nair, 1985).
One of the failure mechanisms that must be evaluated in
organically coated galvanized materials is a form of localized
corrosion, occurring at cut edges of the organic coated sub-
strate, where sheets have been cut to size (Ryan et al., 1994).
The undermining of the organic coating as a result of anodic
zinc dissolution (1) is a frequent, although not exclusive, cause
of cut edge corrosion. In this case, cathodic oxygen reduction
(2) is localized on the steel. Previous workers have shown that
under these conditions the rate of reactions is controlled by a
mass transport of oxygen to the metal surface (Jones and Nair,
1985; Berke et al., 1985).
ZnðsÞ () Znþ2
ðaqÞ þ 2e

ð1Þ

1=2O2ðgÞ þ H2 Oð1Þ þ 2e () 2OH ð2Þ
ðaqÞ
Strontium chromate has been clearly seen to reduce the gal-
vanic corrosion of zinc and steel in a model cell arrangement
that simulates a cut edge (Howard et al., 1996). EIS has been
used in a lot of studies to analyze the general degradation of
organic coatings (Bastos et al., 2011; Bastos and Simoes,
2009; Mansfeld and Tsai, 1991; Zubielewicz and Krolikowska,
2009; Mansfeld et al., 1982; Ranjbar et al., 2004; Walter, 1986;
Touzain, 2010; Deflorian et al., 2005; Penney et al., 2007;
Niknahad et al., 2010; Zin et al., 2005; Gonzalez-Garcia
et al., 2007; Rossi et al., 2005), the impedance response of an
organic coating with artificially-produced defects (Geenen
et al., 1990) and the impedance response of cut-edge corrosion
of organically-coated galvanized steel (Howard et al.,
1999a,b,c; Elvins et al., 2008; Simoes et al., 2009).
In this work, the degradation of the paint film by artificial
acid rain solution at the cut-edge of the organic coated galva-
nized steel was studied with EIS. Measurements were
Figure 1 Experimental test cell set-up.
_ Dehri
R. Yıldız, I.
performed at the end of the 120th and 180th days. The exper-
imental data were fitted to appropriate equivalent electrical cir-
cuits with ZView program.
2. Experimental
Specimens of hot dip galvanized steel (HDG), with a zinc coat-
ing 20 lm thickness on steel were obtained. Sections of galva-
nized steels, with dimensions of 5 · 10 cm, were first coated
with 10 lm of wash primer (polyvinyl butyral), and then 35–
40 lm of epoxy, polyurethane and polyester paint were applied
separately as a top coat. Then organic-coated galvanized steels
were cut 1 cm from edge. A test cell was prepared, with dimen-
sions of 50 · 75 · 40 cm and made from Plexiglass, the samples
were placed at an inclination of 45 relative to the horizon
(Fig. 1), then these samples were exposed to the artificial acid
rain solution according to test conditions. The experiments
were performed in ambient laboratory air and temperature
(25 C). The corrosion tests were performed on these materials.
In order to simulate fogging an experimental set up was con-
structed for the laboratory investigation. Fogging intensity
was regulated by introducing clean compressed air into an ori-
fice of a thin tube with flowing artificial rain, hence, forming
small drops leaving the orifice.
The samples were exposed to a 10 time concentrated artifi-
cial acid rain solution (pH 3.5) to simulate fogging/awaiting
test cells. The composition of a 10 time concentrated artificial
acid rain solution is as follows (Dehri et al., 1999):
Sulfuric acid (1.84s.g.) 31.85 mg/L
Ammonium sulfate 46.2 mg/L
Sodium sulfate 31.94 mg/L
Nitric acid (1.42s.g.) 15.75 mg/L
Sodium nitrate 21.25 mg/L
Sodium chloride 84.85 mg/L
(pH adjusted to 3.5 with sodium carbonate).
Investigation of the cut-edge corrosion of organically-coated galvanized steel after accelerated atmospheric corrosion test 823

Figure 2 Three-electrode system used for the impedance measurements on coated galvanized steel.

Test conditions were 10 s fogging, 1 h awaiting. After 120 -6000

-6e6
and 180 days the samples were removed from the test cell for -100

EIS. PVC tubes, (0.503 cm2) were glued to the substrates and -75
-50
filled with a 10 time concentrated artificial acid rain solution

θ /₀
-25
(Fig. 2). After exposing the substrates into this 10 time concen- 0
-4e6
Z'' / kΩ cm2

trated artificial acid rain solution for 2 h, EIS was performed -4000 25
10-2 10-1 100 101 102 103 104 105
at open circuit potentials in the 100 kHz–10 mHz frequency ƒ/ Hz
range using an amplitude of 10 mV (peak to peak) at room
temperature with CHI 660 C. The counter electrode was a
platinum wire with a 0.7 cm2 surface area and Ag/AgCl (3 M -2e6
-2000
KCI) electrode was used as the reference.

3. Results and discussion

00
3.1. The evaluation of epoxy-coated galvanized steel cut-edge
00 2000
2e6 4000
4e6 6000
6e6 8000
8e6
impedance diagrams
Z' / kΩ cm2
The Nyquist plot with its Bode diagram as an inset of epoxy-
coated galvanized steel exposed to artificial acid rain solution Figure 3 Nyquist and Bode (as inset) plots of epoxy-coated
for 120 days is given in Fig. 3. It is clear that two capacitive galvanized steel at the cut-edge after exposure to the artificial acid
loops are present in high and low frequency regions. The high rain solution for 120 days (solid lines show fitted results).
frequency capacitive loop is related to the coating resistance
(Rpf) and the low frequency capacitive loop is related to the
pore resistance (Rpo) (Ranjbar et al., 2004; Dehri and Erbil,
2000; Thompson and Campbeel, 1994). The Nyquist plot of
epoxy-coated galvanized steel in artificial acid rain solution
after 120 days does not yield a perfect semicircle, generally
attributed to the frequency dispersion as well as to the inhomo-
geneities of surface and mass transport resistance (Zhang et al.,
2004; Dehri et al., 2003). A slightly depressed semicircular
shape at high frequencies in the Nyquist plot was observed.
It indicates that the corrosion of epoxy-coated galvanized steel Figure 4 Equivalent electrical circuit for defective organic-
in artificial acid rain solution is mainly controlled by a charge coated metals. Rs: solution resistance, Rpf: coating resistance,
transfer process (Zin et al., 2005; Nikravesh et al., 2011). CPE1: coating capacitance, Rpo: pore resistance, CPE2: double
The related electrochemical equivalent circuit used to mod- layer capacitance.
el the epoxy-coated galvanized steel–artificial acid rain solu-
tion interface is shown in Fig. 4, where Rs represents the
solution resistance, Rpf corresponds to the first loop of Nyquist and the accumulated species resistance at the metal/solution
plot, and Rpo represents the pore resistance. Rpo includes interface (Ra). CPE represents a constant phase element to re-
charge transfer resistance (Rct), diffuse layer resistance (Rd), place a double layer capacitance (Cdl) in order to give a more
824 _ Dehri
R. Yıldız, I.

Table 1 Fitting parameters Rs, CPE1, Rpf, CPE2, Rpo, CPE3, and Rdiff for the epoxy coated galvanized steel at the cut-edge after
exposure to the artificial acid rain solution test for 120 and 180 days.
Epoxy coating Rs (kO cm2) Rpf (kO cm2) CPE1 (F cm2) Rpo (kO cm2) CPE2 (F cm2) Rdiff (kO cm2) CPE3 (F cm2)
09 07 –
120 days 0.91 2500 1.4 · 10 10,500 5.14 · 10 –
180 days 0.90 11.5 1.33 · 1008 24.6 2 · 1005 1000 8.2 · 1005

-60000
-60
-40
-30
θ /₀

-50000
- 50 -20
-10
0
-40000
-40 10-2 10-1 100 101 102 103 104 105
ƒ/ Hz
Z'' / kΩ cm2

-30000
-30
-20000
-20
-10000
-10
0 In the evaluation of Table 1, the Rpf values showed a sharp
10 20000
00 10000 30 40000
20 30000 50 60000
40 50000 60
Z' / kΩ cm2
Figure 5 Nyquist and Bode (as inset) plots of epoxy-coated
galvanized steel at the cut-edge after exposure to the artificial acid
rain solution test for 180 days (solid lines show fitted results).
accurate fit to the experimental results (Zin et al., 2005; Jorcin
et al., 2006). The impedance parameters obtained by fitting the
Nyquist plot to the equivalent circuit are listed in Table 1. In
Fig. 3, high and low-frequency loops were attributed to coat-
ing capacitance and double layer capacitance, respectively.
The total capacitance of the capacitors is equal to the sum of
the double layer capacitance and coating capacitance.
The Nyquist diagram obtained at the sample exposed to
artificial acid rain solution for 180 days is shown in Fig. 5.
By increasing days (from 120 to 180 days), the significant influ-
ence of exposure time on the size of the impedance spectra was
observed. As it is seen from the inset of Fig. 5 there are 3 time
constants, two capacitive loops in high and middle frequency
regions and followed a straight line in low frequencies. The
high and middle frequency capacitive loops are related to Rpf
and Rpo, respectively. The low frequency straight line implies
that the corrosion of epoxy-coated galvanized steel is diffusion
controlled. The representative equivalent circuit model is given
in Fig. 6. It is known that, the anodic dissolutions of galva-
nized steel and the cathodic oxygen reduction reactions take
place simultaneously on the metal surface. This was repre-
sented as Rdiff (Rdiff = diffuse layer resistance, CPE3 = dif-
fuse layer capacitance) (Zhang et al., 2004; Özcan et al.,
2002; Bonora et al., 1996). The diffusion process may be due
to either the transportation of corrosive ions and soluble cor-
rosion products at the metal/solutions interface or the diffu-
sion of dissolved oxygen to the galvanized steel surface
Figure 6 Equivalent electrical circuit for defective organic-
coated metals. Rs: solution resistance, Rpf: coating resistance,
CPE1: coating capacitance, CPE2: double layer capacitance, Rpo:
pore resistance, CPE3: diffuse layer capacitance, Rdiff: diffuse layer
resistance.
(Worsley et al., 2001). However, the second one is more
probable.
decrease from 2500 kO cm2 for 120 days to 11.5 kO cm2 after
180 days of exposure, indicating a more corrosive process. It
is apparent from Table 1 that the CPE values tend to decrease,
as the Rpf values increased. This decrease in the CPE can be
attributed to the decrease in local dielectric constant or an in-
crease in the thickness of the electrical double layer (Lavaert
et al., 2002).
3.2. The evaluation of polyurethane-coated galvanized steel cut-
edge impedance diagrams
The Nyquist diagrams of the electrochemical impedance data
for the polyurethane-coated galvanized steel exposed to artifi-
cial acid rain for 120 and 180 days are shown in Figs. 7 and 8,
respectively. The related Bode diagrams are added to Figs. 7
and 8 as the inset. In Figs. 7 and 8 it is clear that all the imped-
ance spectra obtained in the artificial acid rain solution consist
of two capacitive loops. The related electrochemical equivalent
circuit used to model the polyurethane-coated galvanized steel/
artificial acid rain solution interface is shown in Fig. 4. The fit-
ted parameters are given in Table 2. As it can be seen from Ta-
ble 2, the coating resistance and pore resistance decreased with
an increase in exposure time. The decrease in the Rpf and Rpo
values was attributed to the same corrosion process occurred
on the metal surface and pores (Dehri et al., 1999; Ranjbar
et al., 2004; Dehri and Erbil, 2000). CPE values obtained from
EIS measurements are in good agreement with each other.
3.3. The evaluation of polyester-coated galvanized steel cut-edge
impedance diagrams
The representative Nyquist and Bode (as inset) plots of polyes-
ter-coated galvanized steel obtained in artificial acid rain solu-
tion are given in Figs. 9 and 10, respectively. As it is seen from
Investigation of the cut-edge corrosion of organically-coated galvanized steel after accelerated atmospheric corrosion test 825

-300000
-300 -75000
-75
-100 -30
-25
-75 -20

θ /₀
θ /₀
-50 -15
-200000
-200 -10
Z'' / kΩ cm2

-25 -50000
-50 -5
10-2 10-1 100 101 102 103 104 105

Z'' / kΩ cm2
0
10-2 10-1 10 0
10 1
10 2
10 3
10 4
10 5 ƒ/ Hz
ƒ/ Hz
-100000
-100
-25000
-25

00
00 100000
100 200000
200 300000
300
00
Z' / kΩ cm 2 00 25
25000 50
50000 75
75000

Z' / kΩ cm2
Figure 7 Nyquist and Bode (as inset) plots of polyurethane-
coated galvanized steel at the cut-edge after exposure to the
artificial acid rain solution test for 120 days (solid lines show fitted Figure 9 Nyquist and Bode (as inset) plots of polyester-coated
results). galvanized steel at the cut-edge after exposure to the artificial acid
rain solution for 120 days (solid lines show fitted results).

-200000
-200
-75
-20000
-20
-30
-50
θ /₀

-25 -20
-150000
-150 θ/₀
0 -15000
-15
-10
10-2 10-1 100 101 102 103 104 105
Z'' / kΩ cm2

ƒ/ Hz 0
Z'' / kΩ cm2

-100000
-100 10-2 10-1 100 101 102 103 104 105
-10000
-10
ƒ/ Hz

-50000
-50
-5000
-5

00
00
00 50
50000 100000
100 150
150000 200
200000 00 5000
5 10000
10 15000
15 20000
20
Z' / kΩ cm2 Z' / kΩ cm 2

Figure 8 Nyquist and Bode (as inset) plots of polyurethane- Figure 10 Nyquist and Bode (as inset) plots of polyester-coated
coated galvanized steel at the cut-edge after exposure to the galvanized steel at the cut-edge after exposure to the artificial acid
artificial acid rain solution for 180 days (solid lines show fitted rain solution for 180 days (solid lines show fitted results).
results).

Figs. 9 and 10 the Nyquist plots obtained in artificial acid rain ance parameters are given in Table 3. As it is shown in
media contain two capacitive loops at high and low frequency Table 3, the Rpf and Rpo values decreased after 120 days in
regions. EIS data were fitted according to the electrical equiv- the artificial acid rain media which may be due to the forma-
alent circuit diagrams given in Fig. 4. The calculated imped- tion of some defects on the coating leading to the access of

Table 2 Fitting parameters Rs, CPE1, Rpf, CPE2 and Rpo for the polyurethane coated galvanized steel at the cut-edge after exposure
to the artificial acid rain solution test for 120 and 180 days.
Polyurethane coating Rs (kO cm2) Rpf (kO cm2) CPE1 (F cm2) Rpo (kO cm2) CPE2 (F cm2)
120 days 0.89 130 2.3 · 1009 250 5.96 · 1006
180 days 0.95 67 9.6 · 1009 121.7 7.95 · 1006
826 _ Dehri
R. Yıldız, I.

Table 3 Fitting parameters Rs, CPE1, Rpf, CPE2 and Rpo for the polyester coated galvanized steel at the cut-edge after exposure to
the artificial acid rain solution test for 120 and 180 days.
Polyester coating Rs (kO cm2) Rpf (kO cm2) CPE1 (F cm2) Rpo (kO cm2) CPE2 (F cm2)
7
120 days 0.90 11.6 1.53 · 10 146.1 2.84 · 105
180 days 0.92 7.8 1.75 · 107 61.5 1.25 · 104

Figure 11 0, 120 and 180 days during the appearance of test specimens exposed to artificial acid rain solution. Epoxy 0 day (A), Epoxy
120 days (A0 ), Epoxy 180 days (A00 ), Polyurethane 0 day (B), Polyurethane 120 days (B0 ), Polyurethane 180 days (B00 ), Polyester 0 day (C),
Polyester 120 days (C0 ), Polyester 180 days (C00 ).

aggressive ions to the metal/organic substance interface. Rpf
and Rpo values of polyester-coated galvanized steel in artificial
acid rain media decreased from 11.6 and 146.1 kO cm2 to 7.8
and 61.5 kO cm2, respectively, with increasing exposure time.
The appearance of the test specimens exposed to acid rain
solution was given for 0, 120 and 180 days in Fig. 11. The
cut edge corrosion was observed moving toward the inner
parts of the metal at the 120th day of epoxy (A0 ), polyurethane
(B0 ) and polyester (C0 ). More degradation was observed at the
inner parts of the metal at the end of the 180th day (A00 ) for
epoxy. Blisters were observed at the inner parts of the metal
at the end of the 180th day (B00 ) more intensely for polyure-
thane. The corrosion on the surface of the metal was observed
thoroughly at the end of the 180th day (C00 ) for polyester. As it
is seen in Fig. 11 the polyester coating was more damaged than
the polyurethane and epoxy coatings.
4. Conclusions
The cut edge corrosion behavior of three organically coated
(epoxy, polyurethane and polyester) galvanized steels was
studied using EIS techniques in artificial acid rain solution
for long exposure days. From the results obtained, the follow-
ing points can be emphasized;
1. By increasing days in atmospheric conditions, artificial acid
rain solutions accelerated the damage and corrosion rate of
the organic-coated galvanized steel.
Investigation of the cut-edge corrosion of organically-coated galvanized steel after accelerated atmospheric corrosion test
2. The experimental data were fitted using the ZView program
and appropriate values for each equivalent circuit element
have been identified for organic-coated galvanized steel
after exposure to artificial acid rain solutions in atmo-
spheric corrosion tests.
3. After exposure to artificial acid rain solutions for 120 and
180 days in a test cell, the organic-coated specimens experi-
enced corrosion, and the corrosion process is confirmed
through localized impedance measurements.
4. According to the Nyquist and Bode plots, after 120 days
the Rpf decreased significantly indicating the defects on
the organic coatings. The distance of this defect from cut
edge to that into the sheet was about 0.8 cm.
5. The surface photographs showed that the degradation of
coatings (that is defective coating areas near to the cut
edge) increased with increasing the exposure.
Acknowledgement
This study was supported by Cukurova University Research
Found (FEF2010D10).
References
Bastos, A.C., Simoes, A.M.P., 2009. Prog. Org. Coat 65, 295.
Bastos, A.C., Ferreira, M.G.S., Simoes, A.M.P., 2011. Electrochim.
Acta 56, 7825.
Berke, N.S., Friel, J.J., Baboian, G.S., Haynes, R. (Eds.), 1985.
Laboratory Corrosion Tests and Standards. ASTM STP 866,
ASTM Philadelphia.
Bonora, P.L., Deflorain, F., Fedrizzi, L., 1996. Electrochim. Acta 41,
1073.
Cox, R.N., Kempster, R.N., Bassi, R.A., 1985. Survey of Performence
of Organic-coated Metal Roof Sheeting. Building Research Estab-
lishment Report, Watford, p. 1193.
Deflorian, F., Rossi, S., Fedizzi, L., Bonora, P.L., 2005. Prog. Org.
Coat. 52, 271.
_ Erbil, M., 2000. Corros. Sci. 42, 969.
Dehri, I.,
_ Howard, R.L., Lyon, S.B., 1999. Corros. Sci. 41, 141.
Dehri, I.,
_ Sözüsağlam, H., Erbil, M., 2003. Prog. Org. Coat. 48, 118.
Dehri, I.,
827
Elvins, J., Spittle, J.A., Sullivan, J.H., Worsley, D.A., 2008. Corros.
Sci. 50, 1650.
Geenen, F.M., de Wit, J.H.W., Van Westing, E.P.M., 1990. Prog. Org.
Coat. 18, 299.
Gonzalez-Garcia, Y., Gonzalez, S., Souto, R.M., 2007. Corros. Sci. 49,
3517.
Howard, R.L., Lyon, S.B., Scantlebury, J.D., 1996. Proc. 13th ICC.
Melbourne, Australia, 25 November paper.
Howard, R.L., Zin, I.M., Scantlebury, J.D., Lyon, S.B., 1999a. Prog.
Org. Coat. 37, 83.
Howard, R.L., Lyon, S.B., Scantlebury, J.D., 1999b. Prog. Org. Coat.
37, 99.
Howard, R.L., Lyon, S.B., Scantlebury, J.D., 1999c. Prog. Org. Coat.
37, 91.
Jones, D.A., Nair, N.R., 1985. Corrosion 41, 357.
Jorcin, J.B., Orazem, M.E., Pebere, N., Tribollet, B., 2006. Electro-
chim. Acta 51, 1473.
Lavaert, V., Moors, M., Wettınck, E., 2002. J. Appl. Electrochem. 32,
853–857.
Mansfeld, F., Tsai, C.H., 1991. Corrosion 47, 958.
Mansfeld, F., Kendig, M.W., Tsai, S., 1982. Corrosion 38, 478.
Niknahad, M., Moradian, S., Mirabedini, S.M., 2010. Corros. Sci. 52,
1948.
Nikravesh, B., Ramezanzadeh, B., Sarabi, A.A., Kasiriha, S.M., 2011.
Corros Sci. 53, 1592.
Nishikata, A., Ichihara, Y., Tsuru, T., 1995. Corros. Sci. 37, 897.
_ Erbil, M., 2002. Prog. Org. Coat. 44, 279.
Özcan, M., Dehri, I.,
Penney, D.J., Sullivan, J.H., Worsley, D.A., 2007. Corros. Sci. 49,
1321.
Ranjbar, Z., Moradian, S., Attar, M.R.M.Z., 2004. Prog. Org. Coat.
51, 87.
Rossi, S., Deflorian, F., Fontanari, L., Cambruzzi, A., Bonora, P.L.,
2005. Prog. Org. Coat. 52, 288.
Ryan, P.A., Wolstenholme, R.P., Howell, D.M., 1994. Durability of
Cladding. Bath Press, Bath.
Simoes, A.M., Torres, J., Picciochi, R., Fernandes, J.C.S., 2009.
Electrochim. Acta 54, 3857.
Thompson, I., Campbeel, D., 1994. Corros. Sci. 36, 187.
Tomashow, N.D., 1964. Corrosion 20, 76.
Touzain, S., 2010. Electrochim. Acta 55, 6190.
Walter, G.W., 1986. Corros. Sci. 26, 681.
Worsley, D.A., Williams, D., Ling, J.S.G., 2001. Corros. Sci. 43, 2335.
Zhang, J., Hu, J., Zhang, J.Q., Cau, C., 2004. Prog. Org. Coat. 51, 145.
Zin, I.M., Lyon, S.B., Hussain, A., 2005. Prog. Org. Coat. 52, 126.
Zubielewicz, M., Krolikowska, A., 2009. Prog. Org. Coat. 66, 129.