Corrosion Science 50 (2008) 3494–3499

Contents lists available at ScienceDirect

Corrosion Science
journal homepage: www.elsevier.com/locate/corsci

The effects of immersion time on morphology and electrochemical properties

of the Cr(III)-based conversion coatings on zinc coated steel surface
Yu-Tsern Chang a,c,*, Niann-Tsyr Wen b, We-Kun Chen b, Ming-Der Ger b, Guan-Tin Pan c, Thomas C.-K. Yang c
a
Department of Chemical and Materials Engineering, Nanya Institute of Technology, Jhongli 32091, Taiwan
b
Graduate School of Defense Science, Chung Cheng Institute of Technology, National Defense University, Tao-Yuan 335, Taiwan
c
Department of Chemical engineering, Laboratory of EMO Materials and Nanotechnology, National Taipei University of Technology, Taipei 106, Taiwan

a r t i c l e i n f o a b s t r a c t

Article history: The main purpose of this paper is to develop a dynamic and non-destructive method to quantify and cor-
Received 1 June 2007 relate the microstructure changes of the Cr(III) layer by electrochemical techniques. The open circuit
Accepted 5 August 2008 potential (OCP) analysis reveals the nucleation growth mechanisms of the Cr(III) layer and the dissolution
Available online 19 September 2008
phenomena of Zn. In addition, the effects of immersion time to the corrosion behavior of Cr(III)-based
conversion coatings (TCCCs) on electrogalvanized steel were studied using potentiodynamic polarization
Keywords: and electrochemical impedance spectroscopy (EIS) in a 3.5% NaCl solution. Furthermore, surface mor-
A. Zinc
phology of the Cr(III) coatings under different immersion times was examined using both a scanning elec-
B. SEM
B. Polarization
tron microscope and an atomic force microscope.
B. EIS From the potentiodynamic polarization experiment, the corrosion current density (Icorr) of the speci-
C. Cr(III)-based conversion coatings men with immersion time of 60 s was found appreciably small, representing the inheritance of the best
anticorrosion performance. Additionally, the corrosion resistance of the Cr(III)-coating for the specimens
obtained between 30 s and 60 s is two order higher than those of the untreated specimen from the EIS
experiments. Results show that the quality of Cr(III)-based conversion coatings was strongly influenced
by the immersion time of Cr(III) solution. And the optimal immersion time is recommended in the range
of 30–60 s.
Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction Cr(III)-based conversion coatings (TCCCs) have become promis-

ing alternatives to traditional CCCs, because the Cr(III) treatment
The zinc coating are widely used to protect carbon steels from solutions are less toxic than Cr(VI) compounds. Besides, the corro-
corrosion since it behaves as a barrier layer to inhibit the carbon sion resistance of Cr(III) coatings is as good as that of hexavalent
steel from oxidation by outdoor exposure. Zinc coating also pro- chromium coatings [7–10]. The preparation of the Cr(III) coatings
vides galvanic protection, as a sacrificial anode, for exposed areas were implemented in the solution containing nitrate as an oxidant
of the steel substrates [1,2]. The corrosion-resistance qualities of and sodium hypophosphite as a complexant, in order to increase
zinc coatings are improved by means of chemical surface treat- the stability of Cr(III) ions inside the chemical bath. After the appli-
ments and organic coatings. After the application of the zinc coat- cation of bath deposition, the color and the thickness of the conver-
ing on carbon steels, the chromate conversion coatings (CCCs) are sion coatings are determined by the immersion time. During the
subsequently deposited to provide the visual color and perform an chemical bath deposition, the chemical compositions and micro-
effective protection from corrosion. However, certain Cr(VI) com- structure of the conversion coatings change dynamically with the
pounds are classified as carcinogen due to the intrinsic toxicity immersion time and the ion compositions of the solution.
[3–5], and government legislation has limited their usage [5,6] by As a result, the length of the specimen soaking in the solution
the RoHS directives from the European union since 2006. There- determines the corrosion resistance of the anticorrosion treated
fore, the search for an alternative material for the conversion coat- specimen. An insufficient immersion would lead to the formation
ing becomes an urgent task. of a very thin barrier layer so that the corrosion protection is lim-
ited. However, over-soaking would promote the cracking on the Cr
layer and cause the dissolution of zinc from the specimen, which
would deteriorate the properties of the corrosion resistance. There-
* Corresponding author. Address: Department of Chemical and Materials Engi-
neering, Nanya Institute of Technology, Jhongli 32091, Taiwan. Tel.: +886 3
fore, the search of an appropriate immersion time for chemical
4361070x2301; fax: +886 3 4652040. bath deposition appears to be significant and assure the quality
E-mail address: ytchang@nanya.edu.tw (Y.-T. Chang). of the anticorrosion coating [11–17].

0010-938X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.corsci.2008.08.051
 Y.-T. Chang et al. / Corrosion Science 50 (2008) 3494–3499 3495

In this paper, the effects of immersion time on the corrosion

resistance of Cr(III) based coatings on carbon steels are investi-
gated using electrochemical methods and surface analysis tools.

2. Experimental procedure

2.1. Specimen preparation

The carbon steel specimens with 10.0 cm by 6.5 cm were elec-

trogalvanized in a zinc solution (JASCO Hyper Zinc 9000) at room
temperature. The thickness of the zinc coatings was approximately
6.23 lm. The zinc surface was activated with 0.5% HNO3(aq) for 10 s
and then rinsed in a deionized water before the conversion treat-
ment. Trivalent chromium treatment was carried out in chemical
solutions with the primary compositions of 0.3 M sodium nitrate,
0.05 M chromic nitrate, and 0.08 M sodium hydroxy alkyl dicar-
boxylic acids. Galvanized specimens were individually immersed
Fig. 1. Change of OCP as a function of time for Cr(III)-based conversion coating.
in Cr(III)-based chemical solutions, pH = 2, at 35 °C at various soak-
ing periods. The specimens were then rinsed in deionized water,
and dehydrated in an oven at 80 °C for 15 min. of the Cr(III) coatings. The voltage was scanned between 1.4 V
and 0.8 V with the scan rate of 5 mV/s. EIS measurement was
2.2. Morphology used to construct the Nyquist and Bode plots of the Cr(III) coat-
ings. The perturbation amplitude of the AC voltage was 5 mV,
The morphology of the Cr(III) treated and untreated zinc surface and scanning frequencies ranged from 100 MHz to 0.01 Hz. All
was investigated using a scanning electron microscope (SEM; Hit- electrochemical experiments were performed using a potentio-
achi, Model S-3000N, Japan) and atomic force microscope (AFM, stat/galvanostat (Autolab PGSTAT30).
Digital, Model III a).
3. Results and discussion
2.3. Electrochemical experiments
3.1. Open circuit potential measurements
In this work, open circuit potential (OCP), potentiodynamic
polarization, and electrochemical impedance spectroscopy (EIS) Fig. 1 shows the time-dependent OCP of the specimen with
measurements were implemented on each specimen. A three- Cr(III)-based conversion coating s on a zinc in the Cr(III) solution.
electrode cell system was used for all electrochemical experi- Three measurements were made for each sample to determine
ments. A platinum sheet and a Ag/AgCl electrode (197 mV vs. the reproducibility of the measurements. For the untreated speci-
SHE) was served as the counter and reference electrodes. The mens (without Cr(III) coating), the OCP was initially 0.907 V
OCP measurement was carried out in the Cr(III)-based solutions (compared to Ag/AgCl) within the first few seconds and then in-
to record the potential versus reaction time (Eocp vs. time), and creased to a maximum value of 0.885 V (compared to Ag/AgCl)
the working electrode was the electrogalvanized plate. Both at 30 s. After the peak value, the potential then felt rapidly at
potentiodynamic polarization and EIS measurements were con- 60–70 s and then maintained as a stable value at around 300 s.
ducted in a 3.5 wt% sodium chloride solution, in which the work- The initial increase of the OCP was due to the surface activation,
ing electrode was electrogalvanized steel with Cr(III) coatings. The where dissolution of zinc occurred simultaneously with the evolu-
potentiodynamic polarization was employed to investigate the tion of hydrogen. These two parallel processes were proposed for
corrosion current density (Icorr) and polarization resistance (Rp) the Cr(III)-based conversion coatings [18,19].

Fig. 2. Photographs for Cr(III)-coating at different dipping times: (a)10 s, (b) 30 s, and (c) 60 s.
3496 Y.-T. Chang et al. / Corrosion Science 50 (2008) 3494–3499

Anodic reaction: Cr3þ þ 3OH ! CrðOHÞ3 ð4Þ

2þ 
Zn ! Zn þ 2e ð1Þ The OCP analysis at first 30 s shows a linear dependence be-
tween potential and immersion time, indicating that the zinc dis-
Cathodic reactions:
solution and conversion coating proceeds through the
þ
2H þ 2e ! H2 ð2Þ instantaneous formation of Cr(III) and Zn(II) hydroxides. After
30 s, the decrease in potential implied that the oxides were dis-
The reduction of hydrogen ions leads to an increase of pH value
solved continuously until the zinc was exposed to the bath solu-
in the vicinity of the Zn surface, which results in the precipitation
tion. The drop of OCP magnitude thus initialed the hydrogen
of chromium hydroxide and zinc hydroxide from Cr(III) and Zn(II)
evolution. Furthermore, the interaction of zinc with bath solution
ions.
only occurred at the condition of the Cr(III) dissolution, which took
Zn2þ þ 2OH ! ZnðOHÞ2 ð3Þ place after 30 s of immersion.

Fig. 3. SEM micrograph of zinc and Zn/Cr(III) deposits on steel surface obtained at different dipping times such as: (a) untreated, (b) 30 s, (c) 60 s, (d) 90 s, (e) 120 s, and (f)
150 s.
 Y.-T. Chang et al. / Corrosion Science 50 (2008) 3494–3499 3497

3.2. Morphology of Cr(III) coatings at different dipping times The AFM topography images of the Zn-coated steel surface be-
fore and after the immersion the Cr(III) solution are shown in Figs.
The Cr(III) coating on an electrogalvanized steel surface was pre- 4a–c. Fig. 4a reveals a smooth surface of the electrodeposited zinc
pared with different coating times. With the increase of dipping coating. After immersion in the Cr(III) solution for 20 s, some oxi-
time, the color of the Cr(III) coating varied from blue, to slightly iri- des or reacted products are deposited on the surface of the Zn
descent yellow and finally to green as shown in Fig. 2 due to the coated steel, as shown in Fig. 4b. When the immersion time is in-
increment in coating thickness. Fig. 3 shows SEM images of Cr(III)- creased to 50 s, the deposition proceeds to form a more compact
based coatings on zinc with different immersion times. Fig. 3a film (Fig.4c).
shows an image of zinc deposited on carbon steel surface. As the
immersion time extends to 30 s, a very thin Cr(III) film was noticed. 3.3. Effects of immersion times on electrochemical properties
For the samples with immersion time longer than 30 s (Fig. 3c), the
appearance of micro cracks become visible. These cracks only oc- 3.3.1. Potentiodynamic polarization
curred superficial by the early stage without penetrating TCCCs. The electrochemical behavior of the Cr(III) coated zinc sub-
Shortly thereafter, micro cracks become intensified at 90 s strates was investigated using potentiodynamic polarization. Figs.
(Fig. 3d). Upon increasing the immersion time up to 120 s (Fig. 3e) 5a–f depicts the polarization curves of the specimen at different
and 150 s (Fig. 3f), the cracks become widened and deepened. Sim- immersion times, including untreated specimen (0 s), 20, 40, 60,
ilar phenomena have been reported by Gigandet et al. [20] that the 80 and 100 s.
growth of CCC on electrogalvanized Zn was accompanied with the As shown in the above figure, the values of cathodic current
dissolution of CCC layers during the immersion of an acidic Cr(VI) tend to decrease with the increase of the soaking time for the
bath. Moreover, Cho et al. [21] reported that an increase of the Cr(III) treated specimen at the initial 30 s. Upon the increase of
immersion time would cause the decrease the thickness of Cr(III)- the soaking time up to 60 s, the cathodic currents decrease appre-
based conversion coating since the Cr(III) deposition rate was ciably. In addition, the currents begin to slightly increase at 100 s.
slower than that of the dissolution rate. In addition, cracks have also After 100 s, the currents of the treated specimen are the same as
been identified on the coatings on tungstate [22] and molybdate that of the untreated sample (Fig. 5a). The difference of cathodic
layers [23,24]. According to earlier reports [18,19,25,26], the pres- currents between untreated and treated substrates was due to
ence of cracks on specimen surface was related to the presence of the inhibition of oxygen by a Cr(III) layer. When the immersion
stress by uneven film thickness or during the dehydration of chro- time was increased to 100 s, the presence of wider and deeper
mate coating at the end of the conversion process. cracks yielded an obvious increase in cathodic current as shown

Fig. 4a. The AFM topography images of the Zn coated steel surface.

Fig. 4b. The AFM topography images of Zn coated steel surface after immersion in the Cr(III) solution for 20 s.
3498 Y.-T. Chang et al. / Corrosion Science 50 (2008) 3494–3499

Fig. 4c. AFM topography images of Zn coated steel surface after immersion in the Cr(III) solution for 50 s.

ler than that of cathodic currents. Similarly, when the conversion

-0.8
time increased to 100 s, a sudden increase of anodic current was
noted due to the increment of Zn dissolution through the cracks
-0.9 formed in this condition. This phenomenon is also observed by
E(V) vs. Ag/AgCl

c
d the SEM micrograph of Fig. 2f. The above results indicated that a
-1.0 e a Cr(III) functioned as a barrier layer, which hindering the transport
b
of oxygen to the zinc substrate and delaying the dissolution of zinc.
-1.1 a. untreated Hence, the rate of corrosion on zinc surface is retarded. However,
b. 20s once the barrier layer was destroyed, the Cr(VI) coating lost its
c. 40s self-healing ability [27–31].
-1.2
d. 60s
e. 80s 3.3.2. Impedance spectroscopy
-1.3 f. 100s The electrochemical behavior of Cr(III) coated specimen was
investigated using the electrochemical impedance spectroscopy.
-1.4 Fig. 6 shows the impedance results for the samples obtained at dif-
1E-9 1E-7 1E-5 1E-3 0.1
2
ferent dipping times in a Cr(III) solution.
Current density(A/cm ) Fig. 6a and b show the impedance spectra of samples obtained
at dipping times of 10, 20, 30, 60, and 100 s and Zn. The imped-
Fig. 5. Potentiodynamic polarization curves obtained on the Cr(III) coated galva-
nized steel obtained at different dipping times: (a) untreated, (b) 20 s, (c) 40 s, (d) ance at range of the low frequency (102 Hz, Fig. 6a) increases
60 s, (e) 80 s, and (f) 100 s obtained from Cr(III)-based chemical solutions at 35 °C, gradually with the immersion time until 60 s. In addition, the
pH value 2. intensity of total impedance of specimen obtained between 30
and 60 s is two orders of magnitude higher than that of the un-
in Fig. 2f. Table 1 shows the response of Icorr and Rp with an in- treated specimen. However, with increase of dipping time to
crease in dipping time. Results show that the polarization resis- 100 s, the total resistance at the same frequency dropped appre-
tance increased with dipping time until it reached to a maximum ciably due to the presence of wider and deeper cracks, which
value of 1.85  104 X cm at 60 s. Similarly, the corrosion current accelerating the dissolution of Zn.
increased from 1.12  106 A cm2 to 5.29  107 A cm2 at the Fig. 6b represents the Bode plot of various specimens prepared
same period of time. The results show that the increase of dipping at different dipping time. There are at least two time-constants in
time causes the increased polarization resistance until 60 s. Base the phase angle/frequency plot of the TCCCs, indicating that the
on these results, the appropriate immersion time for optimum equivalent circuit of the TCCCs contains more than two capaci-
anticorrosion performance was found to be in the range of 30–60 s. tances. Similar experiment was reported by Zheludkevich et al.
The anodic curves formed a similar trend as the performance of who investigated the conversion coating on an Al alloy pre-treated
cathodic currents. However, the drop of anodic currents was smal- with cerium nitrate. The coating structures after corrosion were
interpreted by the phase angle versus frequency plot [32]. Accord-
ing to their research, the capacitance of the coating and corrosion
Table 1
Potentiodynamic polarization data at difference immersion time layer could be distinguished from different frequency range in
the Bode plot of the conversion coatings. In the Bode plot of fre-
Dipping time (s) ba (V/dec) bc (V/dec) Ecorr (V) Icorr (A cm2) Rp (X cm)
quency response, the high-frequency signal is contributed to the
0 0.018 0.022 1.276 1.12E6 1.60E + 2 capacitance of oxide compounds, but the electrolyte capacitance
20 0.096 0.058 1.099 1.25E6 1.27E + 4
presents at the lower frequency. At the early stage of immersion,
40 0.059 0.051 1.075 1.18E6 1.01E + 4
60 0.051 0.041 1.073 5.29E7 1.85E + 4
such as 10 s and 20 s, the thickness of the oxide layer is very thin
80 0.084 0.058 1.089 1.96E6 7.61E + 3 and loose; therefore, the peak in the Bode plot would initially ap-
100 0.049 0.034 1.110 1.91E6 4.55E + 3 pear at a lower frequency. However, the curve at 20 s has three
Icorr = B/Rp, B = 2.3(ba + bc)/(babc), Rp = dg/dIg = 0, Rp is the polarization resistance, B is
time-constants; one capacitance peak appears at low frequency
the Stern-Geary constant, ba, bc are Tafel slops for the anodic and cathodic reactions, of 100.16 Hz and two capacitance signals show at high frequencies
and g is the overpotential for the reactions. of 102.70 Hz and 104.0 Hz. Increasing the dipping time to 60 s, two
 Y.-T. Chang et al. / Corrosion Science 50 (2008) 3494–3499 3499

trochemical properties were studied using polarization and elec-

trochemical impedance spectroscopy in 3.5% NaCl solution. With
the increase of immersion time the color of the Cr(III) coating var-
ied from blue to slightly iridescent yellow, and finally to green. The
above results show that a compact Cr(III) was obtained at immer-
sion time range of 30–60 s in Cr(III)) solution. The chromium Cr(III)
compound functioned as a barrier layer, hindering the transport of
oxygen to the zinc substrate.

Acknowledgment

The authors gratefully acknowledge the financial support from

the Ministry of Education (Project No. E9415).

References

[1] T.E. Graedel, J. Electrochem. Soc. 136 (1989) 193c–203c.

[2] D.C.H. Nevison, in: ASM Handbook, ASM International, Materials Park, OH,
1987. pp. 755–769.
[3] G.D. Wilcox, J.A. Wharton, Trans. IMF 75 (1997) B140–B142.
[4] C.S. Jeffcoate, H.S. Isaacs, A.J. Aldykiewicz Jr., M.P. Ryan, J. Electrochem. Soc. 147
(2000) 540–547.
[5] P.L. Hagans, in: ASM Handbook, ASM International, Materials Park, OH, 1994.
pp. 405–411.
[6] G.D. Wilcow, Trans. Inst. Met. Fin. 81 (2003) B13–B15.
[7] P.C. Wynn, C.V. Bishop, Trans. Inst. Met. Fin. 79 (2001) B27–B30.
[8] A. Gardner, J. Scharf Soc. Automot. Eng. SP-1614 (2001) 91–95.
[9] T. Lionel, Galvanotechnik 91 (12) (2000) 3373–3377.
[10] C. Barnes, J.J.B. Ward, T.S. Sehmbhi, V.E. Carter, Trans. Inst. Met. Fin. 60 (1982)
45–48.
[11] H. Krawiec, B. Stypuła, J. Stoch, M. Mikołajczyk, Corros. Sci. 48 (2006) 595–607.
[12] A.C. Bastos, M.G. Ferreira, A.M. Simões, Corros. Sci. 48 (2006) 1500–1512.
[13] Chao-Min Wang, Han-Chih Liau, Wen-Ta Tsai, Mater. Chem. Phys. 102 (2007)
207–213.
[14] S.M.A. Shibli, B. Jabeera, R.I. Anupama, Surf. Coat. Technol. 200 (2006) 3903–
3906.
[15] F. Contu, B. Elsener, H. Böhni, Corros. Sci. 47 (2005) 1863–1875.
Fig. 6. (a) Impedance spectra and (b) shape of the phase angle verses frequency [16] B.I. Podlovchenko, E.A. Kolyadko, J. Electroanal. Chem. 506 (2001) 11–21.
curve for the samples obtained at different at different dipping times, including [17] S. Frangini, S. Loreti, J. Power Sources 160 (2006) 800–804.
10 s, 20 s, 30 s, 60 s, 100 s, and untreated specimen obtained from Cr(III)-based [18] A.A.O. Magalhaes, B. Tribollet, O.R. Mattos, I.C.P. Margarit, O.E. Barcia, J.
chemical solutions at 35 °C, pH value 2. Electrochem. Soc. 150 (1) (2003) B16.
[19] N.M. Martyak, Surf. Coat. Technol. 88 (1996) 139.
[20] M.P. Gigandet, J. Faucheu, M. Tachez, Surf. Coat. Technol. 89 (1997) 285.
high-frequency peaks shown in the 20 s become overlapped. Obvi- [21] K.W. Cho, V.S. Rao, H.S. Kwon, Electrochim. Acta 52 (2007) 4449–4456.
ously, when Cr(III)-based conversion coating forms completely, the [22] C.G. da Silva, A.N. Correia, P.de. Lima-Neto, I.C.P. Margarit, O.R. Mattos, Corros.
TCCCs have a three-layer structure (outer, intermediate, and inner Sci. 47 (2005) 709–722.
[23] E. Almeida, T.C. Diamantino, M.O. Figueiredo, C. Sa, J. Electrochem. Soc. 106
layers). Therefore, three time-constants appear in the Bode plot. (1998) 8–17.
The related positions of the outer, intermediate, and inner capaci- [24] A.A.O. Magalhaes, I.C.P. Margarit, O.R. Mattos, in: Proceeding of the 7th
tances of the TCCCs in the Bode plot curves (Fig. 6b) were marked International Symposium on Electrochemical Methods in Corrosion Research,
Budapest, Hungary, 2000 (Paper No. 100).
as Qout, Qinter, and Qin, respectively. Furthermore, the phase angle [25] X. Zhang, W.G. Sloof, A. Hovestad, E.P.M. VanWesting, H. Terryn, J.H.W. De Wit,
apparently fell at the frequency of 102.70 Hz (Qinter), as shown by Surf. Coat. Technol. 197 (2005) 168.
comparing that of 60 s and 100 s curves. This phenomenon indi- [26] Z.L. Long, Y.C. Zhou, L. Xiao, Appl. Surf. Sci. 218 (2003) 123.
[27] M.W. Kendig, A.J. Davenport, H.S. Isaacs, Corros. Sci. 33 (1992) 281.
cates that the intermediate layer of the Cr(III) coating may become
[28] F.W. Lytle, R.B. Greegor, G.L. Bibbins, K.Y. Blohawiak, R.E. Smith, G.D. Tss,
more porous when the immersion time increases to 100 s. Corros. Sci. 37 (1995) 349.
[29] J. Zhao, G. Frankel, R.L. McCreery, J. Electrochem. Soc. 145 (1998) 2258.
[30] J.D. Ramsey, R.L. McCreery, J. Electrochem. Soc. 146 (1999) 4076.
4. Conclusions [31] C.S. Jeffcoate, H.S. Isaacs, A.J. Aldykiewicz Jr., M.P. Ryan, J. Electrochem. Soc.
147 (2000) 540.
The Cr(III)-based conversion coatings on electrogalvanized steel [32] M.L. Zheludkevich, R. Serra, M.F. Montemor, K.A. Yasakau, I.M. Miranda
Salvado, M.G.S. Ferreira, Electrochim. Acta 51 (2005) 208–217.
were developed under different experimental conditions, and elec-