Applied Surface Science 257 (2011) 1859–1863

Contents lists available at ScienceDirect

Applied Surface Science

journal homepage: www.elsevier.com/locate/apsusc

Effect of alkaline cleaning and activation on aluminum alloy 7075-T6

Simon Joshi ∗ , William G. Fahrenholtz, Matthew J. O’Keefe
Department of Materials Science and Engineering, Missouri University of Science and Technology, Rolla, MO 65409, USA

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

Article history: The effect of alkaline cleaning and activation on the composition and thickness of the oxide layer on
Received 12 February 2010 aluminum alloy 7075-T6 was studied. E–pH diagrams were developed to predict the effect of alkaline
Received in revised form 20 July 2010 cleaning and activation solutions on the stability of the oxide surface layers. The thickness of the native
Accepted 30 August 2010
oxide layer was determined to be ∼30 nm by Auger electron spectroscopy depth profiling analysis. The
Available online 21 September 2010
outer ∼20 nm was rich in magnesium while the remaining ∼10 nm was rich in aluminum. Cleaning in a
9.1 pH alkaline solution was found to remove the magnesium-rich layer and leave behind an aluminum-
Keywords:
rich oxide layer ∼10 nm thick. Activation in alkaline solutions of NaOH (pH > 12.9) or Na2 CO3 (pH > 11.5)
Aluminum
Magnesium
produced an oxide that was ∼20 to 60 nm thick and rich in magnesium. Alkaline cleaning and activation
Oxide altered the oxide composition and thickness making it possible for deposition of thicker cerium-based
NaOH conversion coatings (∼100 to 250 nm) compared to only alkaline cleaning (∼30 nm), with application of
Na2 CO3 one spray cycle of deposition solution.
Degreasing © 2010 Elsevier B.V. All rights reserved.

1. Introduction when magnesium-containing (2.64 wt%) Al 5052 was heat treated,

Aluminum alloys have high strength-to-weight ratio that makes
them attractive for applications in the aerospace industry [1–13].
The metals are alloyed to increase the strength by forming inter-
metallic compounds (IMCs); however, IMCs that are more cathodic
than aluminum can also increase the susceptibility of the alloys to
corrosion. The most frequently used classes of aluminum alloys in
the aerospace industry are the 2000 and 7000 series, which have
copper and zinc, respectively, as the predominant alloying ele-
ments [1,2,7,8]. Two common alloys are 2024 with nominal alloying
additions (all in wt%) of Cu-4.4%, Mg-1.5%, and Mn-0.6% and 7075
with nominal alloying additions of Zn-5.6%, Mg-2.5%, Cu-1.6%, and
Cr-0.23% [1,12,14]. Both alloys contain magnesium as a secondary
alloying addition.
Aluminum is highly reactive and forms a self-healing, passive
oxide layer that can protect the underlying alloy from atmospheric
corrosion [1,3,5,15–17]. For magnesium-containing alloys such as
7075, magnesium diffuses to the surface during heat treatment,
thereby producing a magnesium-rich surface oxide [1,13,18,19].
Holub and Matienzo suggested that magnesium diffuses to the
surface through the Al2 O3 grain boundaries to form MgO [18,19].
Further, magnesium can even reduce the native alumina to metal-
lic aluminum [18] so that the aluminum-rich oxide layer may also
contain some metallic aluminum. The Holub study was conducted
using aluminum alloys 5052, 3003, and 3004. It was found that
∗ Corresponding author. Tel.: +1 57 334 16461; fax: +1 57 334 12071.
E-mail address: sjwt5@mst.edu (S. Joshi).
the surface consisted almost entirely of MgO, and not MgAl2 O4
as might be expected in the presence of Al2 O3 . The aluminum
detected by Auger electron spectroscopy (AES) was consistent with
the presence of metallic aluminum and not of Al2 O3 . Thus, Al 7075-
T6, which is also rich in magnesium (2.5 wt%), is likely to have a
magnesium-rich oxide layer on the surface.
To deposit coatings for decoration or corrosion resistance, alloy
surfaces are typically cleaned to remove organic contaminants,
native surface oxides, and/or corrosion products to ensure proper
bonding of the coatings to the substrates [14,17,20]. Surface clean-
ing can be done using a variety of different solutions that may
be acidic or alkaline in nature. The effect of these solutions on
the stability of the oxide surface layers can be predicted using
the appropriate E–pH diagrams. Depending on the elements and
compounds that are present, the corrosive, passive, and immune
regions change as a function of pH. For example, alumina is stable in
aqueous solutions in the pH range of 4.0–9.0. At high pH, aluminum
or aluminum-rich oxides will react with the solution to form AlO2 −
(aluminate ions) while at low pH aluminum or aluminum-rich
oxides dissolve to form Al3+ ions [3,4]. Likewise, Mg(OH)2 is sta-
ble in aqueous solutions of pH greater than about 11; below this
pH, it dissolves as Mg2+ [21,22].
Previous research has described the role of different processing
parameters on the thickness and corrosion resistance of cerium-
based conversion coatings (CeCCs) [8,14,16,23–30]. Recent studies
have shown that alkaline cleaning (also referred to as degreas-
ing) followed by acid activation using sulfuric acid increased the
deposition rate and corrosion resistance of CeCCs on Al 2024-T3
[23,24]. For example, degreasing alone resulted in CeCCs thickness

0169-4332/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.apsusc.2010.08.126
1860 S. Joshi et al. / Applied Surface Science 257 (2011) 1859–1863
Fig. 1. E–pH diagram for Al–Mg–H2 O system at 298 K.
of ∼275 nm after 40 spray cycles whereas surface activation with
sulfuric acid resulted in a coating thickness of ∼200 nm after only
one spray cycle [24]. However, no similar study has been performed
for other alloys.
The purpose of this paper was to investigate the role of degreas-
ing and activation solutions on the composition and thickness of
surface oxides on Al 7075-T6.
2. Experimental
Aluminum alloy 7075-T6 (Alcoa lot # 760003) sheets that were
0.08 cm thick were cut to 2.5 cm by 7.6 cm panels. First, the pan-
els were rinsed with acetone followed by degreasing in an aqueous
solution of a commercial alkaline cleaner (5 wt% Turco 4215 NC-LT
in deionized water) for 5 min at 55 ◦ C. After degreasing, the pan-
els were activated by immersion in one of three different alkaline
solutions at room temperature: (1) 20 s immersion in a solution
containing 1 wt% NaOH (Fisher Chemical; 99.8%) in deionized water
(pH of the solution 12.9), which is referred to as 1NaOH; (2) 20 s
immersion in a solution containing 2 wt% NaOH in deionized water
(pH of the solution 13.1), which is referred to as 2NaOH; and (3) 30 s
immersion in an aqueous solution containing 5 wt% Na2 CO3 ·H2 O
(Alfa Aesar; 99.5%) in deionized water (pH of the solution 11.5),
which is referred to as 5Na2 CO3 . Following cleaning and activation,
the panels were rinsed with deionized water and characterized or
placed on a tray at an angle of 60◦ to horizontal for coating depo-
sition. Between each step of the surface preparation process and
prior to coating, the panels were rinsed with deionized water.
The CeCC deposition solution was prepared from a stock solu-
tion consisting of 40 g CeCl3 •xH2 O (Alfa Aesar, 99.9%), 780 g of
deionized water, pH adjusted to 2.07 with HCl. For the deposi-
tion solution, 205 g of the stock solution was mixed with 0.8 g of a
water soluble gelatin (DSF, Rousselot) dissolved in 25 g of deionized
water, and 20 mL of H2 O2 (Fisher Chemical, 30 wt%). Coatings were
deposited using a detail spray gun (Husky, Model # 515-547), held
at a distance of about 25 cm from the panels. Activated panels were
sprayed for ∼5 s and then allowed to drain for ∼35 s. Multiple spray-
drain cycles were used to achieve the desired coating thickness
and uniformity. Coated panels were post-treated by immersion
for 5 minute in a water solution containing 2.5 wt% Na3 PO4 (pH
adjusted to 4.5 with phosphoric acid) that was heated to 85 ◦ C. The
corrosion resistances of the coated panels were evaluated using salt
spray testing (Q-Fog, Q-Panel Lab Products) per ASTM standards
B117.
The composition of surface oxide layers was characterized using
AES depth profiling analysis (PerkinElmer, Model 545). A sputtering
rate of 9 nm/min was assumed based on a standard that consisted
of a 100 nm Ta2 O5 coating on tantalum. Each AES depth profile
was conducted two times on different panels and average values
reported. FactSage software version 5 was used to construct E–pH
diagrams.
3. Results and discussion
Stable species for aluminum and magnesium were determined
at various pH values by calculating E–pH diagrams. Diagrams were
constructed at room temperature with and without the presence
of carbonate ions (Figs. 1 and 2). For dilute concentrations of mag-
nesium and aluminum (10−6 M) at zero solution potential, Fig. 1
(without carbonate ions) shows that aluminum is stable as AlOOH
(designated Al2 O3 ·H2 O) in the pH range 4–11.5. Below pH 4, alu-
minum dissolves as Al3+ ions, while above pH of 11.5, aluminum
dissolves in the form of aluminate (AlO2 − ) ions. When the temper-
ature was increased to 55 ◦ C, the stability regions did not change
significantly. Therefore, Fig. 1 can be used over the range of temper-
atures of interest in this study, which was room temperature to the
temperature of the degreasing solution (55 ◦ C). Considering mag-
nesium species, Mg(OH)2 is stable at pH values greater than ∼11.5,
but dissolves as Mg2+ at lower pH values. Based on the diagram, alu-
minum should form a stable, protective oxide when the pH value
is between 4 and 11.5, but should dissolve or corrode at higher or
lower values. In contrast, magnesium dissolves at pH values below
∼11.5, but is protected by Mg(OH)2 at higher pH values.
Fig. 2 shows the effect of the presence of carbonate ions (CO3 −
0.6 M) on stability of magnesium and aluminum species. Carbonate
ions do not affect the stability of aluminum species. In contrast, the
presence of CO3 − ions favors the formation of MgCO3 at pH values
from ∼9.5 up to 12.2 at potentials above approximately −0.5 V. The
 S. Joshi et al. / Applied Surface Science 257 (2011) 1859–1863 1861

Fig. 2. E–pH diagram of the Al–Mg–C–H2 O system at 298 K.

stable magnesium species was Mg(OH)2 at higher pH values and

potentials above approximately −0.6 V. At the lower pH values,
magnesium should dissolve as MgHCO3 + ions at higher solution
potentials. At lower potentials Mg(C2 H3 O2 )+ or Mg(C2 O4 )2 2− were
the stable solution species. Hence, the addition of carbonate ions
increases the pH range for stability of magnesium by forming a
stable carbonate phase.
The pH of the degreasing solution used in this study was 9.1
and the solution temperature was maintained at 55 ◦ C. Fig. 1
suggests the stable species for aluminum and magnesium are
AlOOH and Mg2+ . During degreasing, magnesium on the sur-
face should be converted to Mg2+ , but alumina should not be
affected. Therefore, immersion in the degreasing solution should
dissolve a magnesium-rich oxide, but not affect an aluminum-rich
oxide. Fig. 3. AES depth profile of acetone-rinsed Al 7075-T6.
The pH values of the 1NaOH and 2NaOH solutions were both
above 12.9. At these higher pH values, Fig. 1 suggests the stable
Fig. 4 shows the AES depth profile after degreasing. The alu-
species should be Mg(OH)2 and aluminate ions. During surface
minum line crossed the oxygen line at ∼10 nm, suggesting that
activation in NaOH solutions, magnesium should be converted to
the oxide layer was ∼10 nm thick. Further, the oxide layer was
stable Mg(OH)2 while aluminum should be dissolve as aluminate
aluminum-rich compared to the native oxide. The magnesium con-
ions. The pH of 5Na2 CO3 is 11.5. In the presence of carbonate
tent at the surface decreased from the original value of ∼20 at%
ions at this pH, magnesium on the surface should be converted
in the acetone rinsed substrate (Fig. 3) to less than 10 at% (Fig. 4)
to stable MgCO3 , while aluminum should be dissolved as alu-
minate ions (Fig. 2). While more complex than the effect of the
degreasing solution, exposure of degreased surfaces to the alkaline
activation solutions should dissolve aluminum species, but stabi-
lize magnesium-species as Mg(OH)2 or MgCO3 .
Fig. 3 shows the AES depth profiles for acetone rinsed Al 7075-
T6. Acetone should remove surface contamination such as dust or
oil without altering the surface oxides. After acetone rinsing, the
concentrations measured at the surface were 4 at% Al, 22 at% Mg,
and 22 at% O, suggesting a magnesium-rich oxide was present. The
total oxide thickness was ∼30 nm, based on the point where the
aluminum and oxygen lines crossed. The ∼20 nm of the upper sur-
face oxide was magnesium-rich [13,18]. The aluminum content
increased near the oxide–metal interface, but the oxide still had
a significant magnesium content (∼20 at%). Therefore, magnesium
appeared to be surface active in the aluminum alloy because it was
present in higher quantities (>20 at%) compared to its content in
the alloy (2.5 wt% or 2.9 at%). Fig. 4. AES depth profiling of Al7075-T6 after degreasing.
1862 S. Joshi et al. / Applied Surface Science 257 (2011) 1859–1863
Fig. 5. AES depth profiling of Al7075-T6 after degreasing followed by activation in
1NaOH.
after degreasing. Based on the E–pH diagrams, magnesium-species
should dissolve as Mg2+ ions at the pH of the degreasing solution,
but aluminum species are stable as AlOOH. Comparing the AES
results in Fig. 3 with those in Fig. 4, it appears that outer portion of
the magnesium-rich native oxide layer was removed by degreasing,
which left an aluminum-rich oxide that was ∼10 nm thick.
AES depth profile analyses were performed on Al 7075-T6 after
surface activation in three different solutions: (1) 1NaOH (Fig. 5),
(2) 2NaOH (Fig. 6), and (3) 5Na2 CO3 (Fig. 7). Comparison of AES data
for degreasing (Fig. 4) with those of surface activations (Figs. 5–7)
suggest surface activation caused an increase in surface oxide thick-
ness from ∼10 nm after degreasing to 20–30 nm after activation
in 1NaOH or 5Na2 CO3 and ∼60 nm after activation in 2NaOH. The
increase in thickness of the oxide layer on the panel activated with
2NaOH is an indication of the higher activity of this solution com-
pared to the other two solutions. The magnesium content of the
oxide layer after activation was higher than it was after degreasing.
At the high pH of the activation solutions, aluminum species should
dissolve as aluminate ions, but magnesium is stable as Mg(OH)2 or
MgCO3 , depending on the activation solution used. AES data for
each surface activation showed the concentrations at the panel
surfaces were 25–30 at% Al, 15–20 at% Mg, and ∼50 at% O. These
compositions were similar through the oxide to within ∼10 nm
of the aluminum alloy substrates surface. The 10 nm or so of the
oxide layer closest to the interface with the substrate appeared to
Fig. 6. AES depth profiling of Al7075-T6 after degreasing followed by activation in
2NaOH.
Fig. 7. AES depth profiling of Al7075-T6 after degreasing followed by activation in
5Na2 CO3 .
be similar to the native oxide present on the surface after degreas-
ing. Based on the AES results, during activation it appears that a
magnesium-rich oxide formed on top of the aluminum-rich layer
that was present after degreasing.
Degreasing and alkaline activation are part of the standard pro-
cess for deposition of CeCCs. The purpose of these steps is to alter
the surface oxide composition and thickness to promote deposi-
tion of uniform CeCCs with sufficient thickness to provide corrosion
protection. After only degreasing, AES depth profiling showed that
the CeCC thickness after one spray cycle was ∼30 nm. For com-
parison, CeCCs on alkaline activated panels had thicknesses in the
range of 100–250 nm after one spray cycle (Table 1). Visually, no
color change was observed for the surface of degreased panels
after one spray cycle, whereas the color of alkaline activated pan-
els changed to yellow after one spray cycle, indicating deposition
of cerium-based coatings. Increasing the number of spray cycles to
five on a degreased panel showed only a faint color change indi-
cating insignificant deposition (thickness was not measured). In
contrast, five spray cycles produced a more intense color change
on alkaline activated panels indicating thicker deposition, which
was determined to be in the range of 250–500 nm for the differ-
ent activation processes (Table 1). While degreasing removed the
magnesium-rich oxide layer, it did not alter the dense native alu-
mina layer on the panels. Apparently, this layer protected the alloy
surface during deposition and did not allow the coating solution to
penetrate the oxide and react with the substrate, which is required
for spontaneous deposition of coatings. In the case of alkaline acti-
vated panels, the surface oxide was altered such that the deposition
solution could react with the substrate to produce coatings. Pan-
els cleaned in 5Na2 CO3 produced thinner coatings than 1NaOH
and 2NaOH, which indicated that it was less aggressive than the
NaOH solutions. Based on observed color changes and measured
coating thicknesses, degreasing followed by activation altered the
surface oxide composition and thickness in such a way to promote
deposition of uniform CeCCs.
Table 1
Coating thickness by AES depth profiling of spray deposited CeCCs after degreasing
and alkaline activation. – denotes not measured.
Degreasing 5Na2 CO3 1NaOH 2NaOH
1 Spray (nm) 30 100 175 250
5 Sprays (nm) – 230 380 500
 S. Joshi et al. / Applied Surface Science 257 (2011) 1859–1863
The corrosion resistance of CeCCs on degreased and alkaline
activated panels was evaluated in salt spray testing conducted in
accordance with ASTM B117. With only one spray cycle, degreased
and 5Na2 CO3 activated panels showed significant corrosion in the
form of pits and salt tails within 24 h. The coatings on these pan-
els were too thin to provide significant protection. Increasing the
spray cycle to five for degreased panel did not improve the corro-
sion resistance and the panels showed the formation of pits and salt
tails within 24 h. In contrast, coatings produced by one spray-drain
cycle on panels activated in1NaOH or 2NaOH performed better and
inhibited the formation of pits and salt tails for at least 24 h. When
the number of deposition spray cycles was increased to five, the
coatings on the substrate prepared using any of the three alka-
line activation solutions provided better corrosion protection as
compared to the coatings deposited with one spray cycle. In this
analysis, seven out of 10 CeCCs prepared using five spray-drain
cycles on alkaline activated panels inhibited the formation of visual
pits and salt tails for up to 2 weeks in salt spray testing. Thus, alka-
line surface activation on Al 7075-T6 not only helped promoted
deposition of uniform CeCCs with increased thickness, but the sub-
sequent CeCCs also demonstrated enhanced corrosion resistance.
4. Conclusion
The effect of degreasing and alkaline activation on the compo-
sition and thickness of the surface oxide layer on aluminum alloy
7075-T6 was studied. AES showed the initial native oxide on Al
7075-T6 was about 30 nm thick and was comprised of an outer
layer about ∼20 nm thick that was rich in MgO and an inner layer
∼10 nm thick that was rich in Al2 O3 . Degreasing in a solution of
a commercial alkaline cleaner with pH ∼9.1 selectively removed
the magnesium-rich outer oxide layer, leaving behind the native
aluminum-rich oxide that was ∼10 nm thick. Subsequent alkaline
activation in NaOH or Na2 CO3 solutions with pH > 11.5 increased
the thickness of the oxide to ∼20 to 60 nm. At these high pH val-
ues, E–pH diagrams indicated that aluminum would dissolve as
aluminate ions and magnesium would be stable as Mg(OH)2 or
MgCO3 , depending on the activation solution used. The majority
of the surface oxide thickness was presumably a reaction product
formed during activation due to the interaction of the activation
solution with the ∼10 nm thick native aluminum-rich oxide layer.
After surface activation, cerium-based conversion coatings could
be deposited, with 2 wt% NaOH surface activation resulting in the
thickest coating (∼250 nm) after one spray cycles. After five spray
cycles, the thicknesses ranged from 250 to 500 nm and the CeCCs
demonstrated good corrosion protection. Not only did alkaline acti-
vation result in the deposition of thicker CeCCs, but the resulting
coatings had improved corrosion resistance.
Acknowledgements
The authors would like to acknowledge Bruce Sartwell of
the Strategic Environmental Research and Development Program
(SERDP) for his support and guidance. This project was financially
supported by SERDP under Project WP-1618, Contract W912HQ-
08-C-0008. The authors would also like to acknowledge Jeff Wight
at Missouri S&T for conducting the AES analysis.
1863
References
[1] J.E. Hatch, Aluminum: Properties and Physical Metallurgy, ASM, 1984.
[2] B.Y. Johnson, J. Edington, M.J. O’Keefe, Effect of coating parameters on the
microstructure of cerium oxide conversion coatings, J. Mater. Sci. Eng. A361
(2003) 225–231.
[3] J.R. Davis, Corrosion of Aluminum and Aluminum Alloys, ASM International,
1999.
[4] T.J. O’Keefe, P. Yu, Encyclopedia of Materials: Science and Technology (2001)
7774–7781. ISBN: 0-08-0431526.
[5] D.A. Jones, Principles and Prevention of Corrosion, second ed., Prentice Hall,
1996.
[6] N. Birbilis, R.G. Buchheit, Electrochemical characteristics of intermetallic
phases in aluminum alloys, J. Electrochem. Soc. 152 (2005) B140–B151.
[7] W.G. Fahrenholtz, M.J. O’Keefe, H. Zhou, J.T. Grant, Characterization of cerium-
based conversion coatings for corrosion protection of aluminum alloys, J. Surf.
Coat. Technol. 155 (2002) 208–213.
[8] L.E.M. Palomino, I.V. Aoki, H.G. de Melo, Microstructural and electrochemical
characterization of Ce conversion layers formed on Al alloy 2024-T3 covered
with Cu-rich smut, J. Electrochim. Acta 51 (2006) 5943–5953.
[9] M. Dixit, R.S. Mishra, K.K. Sankaran, Structure–property correlations in Al 7050
and Al 7055 high-strength aluminum alloys, J. Mater. Sci. Eng. A 478 (2008)
163–172.
[10] M.J. Starink, S.C. Wang, A model for the yield strength of overaged
Al–Zn–Mg–Cu alloys, J. Acta Mater. 51 (2003) 5131–5150.
[11] J.F. Nie, B.C. Muddle, Microstructure and design of high-strength aluminum
alloys, J. Phase Equilib. 19 (1998) 543–551.
[12] D. Ortiz, M. Abdelshehid, R. Dalton, et al., Effect of cold work on the tensile
properties of 6061, 2024, and 7075 Al alloys, J. Mater. Eng. Perform. 16 (2007)
515–520.
[13] D.G. Altenpohl, Aluminum: Technology, Applications, and Environment, TMS,
1998.
[14] A. de Frutos, M.A. Arenas, Y. Liu, et al., Influence in cerium conversion treat-
ment of AA2024-T3 and 7075-T6 alloys, J. Surf. Coat. Technol. 202 (2008) 3797–
3807.
[15] M.S. Hunter, P. Fowle, Natural and thermally formed oxide films on aluminum,
J. Electrochem. Soc. 103 (1956) 482–485.
[16] M. Dabala, L. Armelao, A. Buchberger, I. Calliari, Cerium-based conversion layers
on aluminum alloys, J. Appl. Surf. Sci. 172 (2001) 312–322.
[17] K.J.H. Nelson, A.E. Hughes, et al., Characterisation of aluminum alloys after
HNO3 /HF–NaOH–HNO3 /HF pretreatment, J. Mater. Sci. Technol. 17 (2001)
1211–1221.
[18] K.J. Holub, L.J. Matienzo, Magnesium diffusion in several aluminum alloys, J.
Appl. Surf. Sci. 9 (1981) P22–38.
[19] T. Fujita, Z. Horita, T.G. Langdon, Characteristics of diffusion in Al–Mg alloys
with ultrafine grain size, Philos. Mag. 82 (2002) 2249–2262.
[20] C. Dunbar, Surface Preparation for Continuously Applied Coatings, ASM Hand-
book, 2002.
[21] R.W. Revie, Uhlig’s Corrosion Handbook, Second ed., John Wiley & Sons, Inc.,
2000.
[22] V. Fournier, P. Marcus, I. Olefjord, Oxidation of magnesium, J. Surf. Interface
Anal. 34 (2002) 494–497.
[23] W. Pinc, P. Yu, M. O’Keefe, W. Fahrenholtz, Effect of gelatin additions on the
corrosion resistance of cerium based conversion coatings spray deposited on
Al 2024-T3, J. Surf. Coat. Technol. 203 (2009) 3533–3540.
[24] W. Pinc, S. Geng, M. O’Keefe, W. Fahrenholtz, T. O’Keefe, Effects of acid and alka-
line based surface preparations on spray deposited cerium based conversion
coatings on Al 2024-T3, J. Appl. Surf. Sci. 255 (2009) 4061–4065.
[25] C. Wang, F. Jiang, F. Wang, Cerium chemical conversion coating for aluminum
alloy 2024-T3 and its corrosion resistance, J. Corros. Sci. 60 (2004) 237–243.
[26] A.E. Hughes, J.D. Gorman, et al., Development of cerium-based conversion coat-
ings on 2024-T3 Al alloy after rare-earth desmutting, J. Surf. Interface Anal. 36
(2004) 290–303.
[27] A. Decroly, J.-P. Petitjean, Study of the deposition of cerium oxide by conversion
on aluminum alloys, J. Surf. Coat. Technol. 194 (2005) 1–9.
[28] M. Bethencourt, F.J. Botana, M.J. Cano, M. Marcos, High protective environ-
mentally friendly and short-time developed conversion coatings for aluminum
alloys, J. Appl. Surf. Sci. 189 (2002) 162–173.
[29] P. Campestrini, H. Terryn, A. Hovestad, J.H.W. de Wit, Formation of cerium-
based conversion coatings on AA2024: relationship with the microstructure, J.
Surf. Coat. Technol. 176 (2004) 365–384.
[30] F.H. Scholes, C. Soste, A.E. Hughes, S.G. Hardin, P.R. Curtis, The role of hydrogen
peroxide in the deposition of cerium-based conversion coatings, J. Appl. Surf.
Sci. 253 (2006) 1770–1780.