Effect of sealing treatment on the corrosion behavior of anodized AA2099 aluminum-lithium alloy

Pedro Samaniego-Gámez^a, Facundo Almeraya-Calderón^a, Ulises Martin^b, Jacob Ress^b, Citlalli Gaona-Tiburcio^a, Luis Silva-Vidaurri^c, José Cabral-Miramontes^a, José M. Bastidas^d, José G. Chacón-Nava^c, David M. Bastidas^b,*

^aUniversidad Autónoma de Nuevo León. FIME–CIIIA, Ave. Universidad s/n, Ciudad Universitaria, 66455 San Nicolás de los Garza, Nuevo León, México

^bNational Center for Education and Research on Corrosion and Materials Performance, NCERCAMP-UA, Dept. Chemical, Biomolecular, and Corrosion Engineering, The University of Akron, 302 E Buchtel Ave, Akron, OH 44325-3906, United States

^cCentro de Investigación en Materiales Avanzados (CIMAV), Miguel de Cervantes 120, Complejo Industrial Chihuahua, Chihuahua 31136, México

^dNational Center for Metallurgical Research (CENIM, CSIC), Ave. Gregorio del Amo 8, 28040 Madrid, Spain

(*Corresponding author:  dbastidas@uakron.edu)

1. INTRODUCTIONTOP

The anodizing process improves the resistance to corrosion and abrasion of Al alloys (AA) by increasing the thickness of the alumina layer (Al₂O₃) formed. This process involves an electrochemical reaction in acidic electrolytes increasing the thickness of the Al oxide layer. The anodizing process is used mainly for the treatment of Al alloy surfaces for industrial applications. However, in the aeronautic industry, such electrochemical processes are implemented because Al alloys such as 2xxx and 7xxx have a wide range in aerospace applications due to high strength and low density, allowing for an increase in payload and fuel efficiency (Keller et al., 1953; Etienne et al., 2016; Ma et al., 2016; Jaimes-Ramírez et al., 2018; Zhang et al., 2020).

Al-Li AA2099 alloy has been used in the aeronautic industry in recent years due to its low density and high mechanical strength, replacing the conventional 2xxx and 7xxx alloys (Mouritz, 2012; Ma et al., 2016). The chief characteristic of Al-Li alloy is that, by adding 1 wt.% Li, it is possible to decrease the weight by 3% compared to pure Al and the stiffness is increased by 5−6%. Additional advantages of AA2099 include lower in-plane anisotropy of the mechanical properties, greater transverse ductility, excellent resistance to stress corrosion cracking and excellent toughness (Romios et al., 2005; Mouritz, 2012; Deschamps et al., 2013; Ma et al., 2016; Gumbmann et al., 2016a; Gumbmann et al., 2016b; Deng et al., 2017; Deschamps et al., 2017). Due to its low density and high stiffness, Al-Li AA2099 alloy is one of the core materials in aircraft manufacturing applications, for example, in the reinforcement of T-shaped joints for the fuselage panels (Tian et al., 2016). Due to the interaction of Al in contact with the environment, Al alloys form a spontaneous oxide layer on the surface. Such a natural oxide layer does not provide adequate protection in aggressive environments, and Al alloys are susceptible to localized corrosion through microgalvanic coupling between different phases. The high-copper-containing constituent Al–Fe–Mn–Cu particles are responsible for the initiation of localized corrosion in the anodized AA2099 (Ma et al., 2015a). Grains of high stored energy could also contain increased volume fraction of Al₂CuLi intermetallic phase after aging, which might increase corrosion susceptibility (Ma et al., 2015b).

Due to the different conditions that aircraft experience in service, the Al alloys used in the aeronautic industry are anodized before applying primers and topcoats (Ma et al., 2011; Yu et al., 2020; Zhang et al., 2020). By carrying out an anodizing process, it is possible to generate a thicker artificial oxide layer, improving the corrosion resistance, wear resistance and mechanical properties (Runge, 2018). Conversion coatings have proved to impart passivation, such the traditionally applied anodizing electrolyte containing sulfuric acid, glycerol, lactic acid and titanium dilactate ammonium salt (Dale, 1970). The Al₂O₃ layer is characterized by being a compact, fine, porous, and external barrier. The pores act as migration pathways for corrosive agents, therefore a sealing treatment is necessary to stabilize the Al₂O₃ layer, thus reducing the diameter of the pores and increasing the corrosion resistance. The sealing treatment can be carried out through different solutions such as water, sodium dichromate (Na₂Cr₂O₇), nickel acetate, and nickel fluoride. The efficiency of a sealing process depends on the surface reactivity of the porous oxide structure in the different anodized Al alloys (Etienne et al., 2016; Runge, 2018). Sodium dichromate is considered to be one of the most effective sealing methods for corrosion prevention because it not only blocks pores, it also acts as a corrosion inhibitor, a specific property of self-healing corrosion (Zuo et al., 2003; Etienne et al., 2016; Yu et al., 2020).

The aim of this work is to study the corrosion properties of the Al-Li AA2099 alloy in 3.5 wt.% NaCl and 10 vol.% H₂SO₄ solutions. The AA2099 alloy was anodized using a sulfuric acid bath and two current densities, 0.19 and 1.0 A·cm^–2. A posterior sealing process in H₂O or in Na₂Cr₂O₇ was applied. The resulting specimens were exposed to a 3.5 wt.% NaCl or a 10 vol.% H₂SO₄ solution, and their electrochemical corrosion behavior was studied using electrochemical impedance spectroscopy (EIS). Scanning electron microscopy (SEM), and X-ray photoelectron spectroscopy (XPS) analyses were performed to characterize the surface, to determine the morphology, thickness, and chemical composition of the anodized and sealed coatings.

2. MATERIALS AND EXPERIMENTAL PROCEDURETOP

2.1. MaterialsTOP

An Al-Li AA2099 alloy was used as the substrate. The specimens were cut to dimensions of 0.05 m length and 0.005 m in thickness. Specimens were sequentially polished with 400, 600, and 800 SiC grit papers and cleaned with deionized water and acetone.

2.2. Anodizing processTOP

Before anodizing, the AA2099 specimens were degreased and etched in a 50% HCl solution for 5 s at 25 °C and rinsed in distilled water (3 times). Anodizing was carried out in a 16 wt.% H₂SO₄ solution using two current densities, 0.19 and 1.0 A·cm⁻², for 45 min at 25 °C. For the sealing process, the anodized specimens were immersed in H₂O at 95 °C temperature for 25 min, or in 6 wt.% Na₂Cr₂O₇ solution at 95 °C temperature for 25 min. Table 1 shows the nomenclature of the specimens and the variables used in the anodizing and sealing processes.

2.3. Electrochemical studyTOP

A conventional three-electrode cell configuration was used for the electrochemical study, the anodized AA2099 alloy as working electrode, a saturated calomel electrode (SCE) as reference, and a platinum mesh as counter electrode. Electrochemical measurements were carried out using a Gill-AC potentiostat/galvanostat/ZRA from ACM Instruments (UK). Corrosion experiments were performed by immersion of the anodized and sealed Al-Li AA2099 specimens, with an active surface area of 1.0 cm², in 3.5 wt.% NaCl solution and in 10 vol.% H₂SO₄ solution, this latter to simulate an acid rain environment, at 25 °C temperature for 3 h. The EIS measurements were recorded at the corrosion potential (E_corr) over a frequency range from 100 kHz to 1 mHz, obtaining 10 points per decade, and applying an AC potential signal of 10 mV r.m.s. amplitude according to ASTM G106-89 (2015). The results were analyzed using electrical equivalent circuit (EEC) models and the Zview impedance program. All the EIS tests were performed in triplicate.

2.4. Anodized layers characterizationTOP

Analysis of in-plane and cross-section surface morphology was done using SEM, in a JEOL JSM 6510LV model operating at an excitation voltage of 20 kV, and a working distance (WD) of 12 mm. Study of the oxide layer thickness of the cross-section was performed at a magnification of 2000X.

X-ray photoelectron spectroscopy (XPS) was used to determine the chemical composition of the Al oxide layer, a Thermo Scientific Escalab 250 Xi equipment was utilized. The pressure in the analysis chamber was maintained below 10⁻⁹ torr. The excitation of the analyzed photoelectrons was carried out with a monochromatic Al K_α anode X-ray source (1486 eV). The analyzed regions of interest were Al 2p and O 1s. The peaks were fitted using a Gasussian-Lorentzian mixed function, after a Shirley background subtraction (Fajardo et al., 2019). Spectra were obtained at a take-off angle of 45° (Cumpson, 1995).

3. RESULTS AND DISCUSSIONTOP

3.1. SEM, surface and cross-sectional morphologiesTOP

The study of the surface and the cross-sectional morphologies of the anodic oxide layer of Al-Li AA2099 alloy is of great importance to understand the corrosion resistance. In Fig. 1, the surface morphology of the anodized AA2099 samples can be observed with varying current density conditions (0.19 and 1.0 A·cm⁻²) and sealing solutions (H₂O and 6 wt.% Na₂Cr₂O₇) within the anodizing process. Micrographs of the AA2099 show the conventional honeycomb morphology. One of the disadvantages of the Al oxide layer is its porous structure, favoring the accumulation of corrosion products. In order to reduce the Al₂O₃ layer porosity, two sealing solutions (H₂O and 6 wt.% Na₂Cr₂O₇) were used to reduce the diameter of the pores. The efficiency of the sealing process depends mainly of the surface reactivity and structure of the passive layer (Etienne et al., 2016; Yang et al., 2019; Yu et al., 2020).

Figure 1(a, c) shows a less homogeneous surface compared to the micrographs of Fig. 1(b, d). This may be due to the amount of current density employed during the anodizing process, resulting in a uniform surface area for the 1.0 A·cm⁻² current density sample. Some imperfections can be observed on the micrographs (see Fig. 1(b, d)). However, the imperfections tend to be in greater abundance on the anodized specimens using current density of 0.19 A·cm⁻² (see Fig. 1(a, c)).

Figure 2 shows the cross-section morphology obtained by SEM of the anodized specimens. For samples anodized using 0.19 A·cm⁻² current density (see Fig. 2(a, c)), a more irregular and cracked oxide layer can be observed, compared to specimens anodized using 1.0 A·cm⁻² current density (see Fig. 2(b, d)) which have a more stable Al₂O₃ layer. The current density influences the structure of the Al₂O₃ layer, but not the thickness, as the four specimens have a thickness between 6−9 μm. The specimens anodized using 1.0 A·cm⁻² current density provides greater stability of the Al₂O₃ layer, as revealed in the micrographs of Fig. 2(a, c) in which a structure with more imperfections in comparison with the micrographs of Fig. 2(b, d) can be observed. The average thickness obtained in the cross-section of AA2099 specimens sealed in H₂O was 7.44 μm, while for the specimens sealed in 6 wt.% Na₂Cr₂O₇ was 9.51 μm.

3.2. EIS dataTOP

Figure 3 shows Nyquist plots, along with Bode plots (top-inset) for anodized and sealed AA2099 alloy immersed in 3.5 wt.% NaCl solution for 3 h. For all specimens, a capacitive behavior is observed, as well as a high impedance value (>450 kΩ cm²). The best defined semicircle is shown by the M4 specimen (1.0 A·cm⁻² current density and 6 wt.% Na₂Cr₂O₇ sealing solution), while the lower impedance was found for the M1 sample, see Fig. 3 bottom inset. The impedance results were fitted using the electrical equivalent circuit (EEC) model in Fig. 4, which includes two-time constants (Yang et al., 2019). The upper inset of Fig. 3 shows the Bode plots for all impedance data, two peaks can be observed in the phase angle (θ) vs. frequency plot, indicating that AA2099/3.5 wt.% NaCl interface presents two-time constants. This cannot be observed on the Nyquist plots of Fig. 3 because the two-time constants are overlapped. An excellent correlation between the experimental and the fitted data can be observed, see Fig. 3, the fitting procedure matches well with experimental results shown as individual points. Table 2 summarizes the fitting parameters of the EIS data, the χ² values are in the range of 1.54−7.92×10⁻³.

The EEC model of Fig. 4 contains the electrolyte resistance (R_s), attributed to the ohmic resistance, between the working electrode and the reference electrode. R_s is in series with a parallel combination (R₁//CPE₁) for the high- and intermediate-frequency range response containing the oxide film resistance (R₁) and a constant phase element (CPE₁), both attributed to the oxide films generated on the AA2099 alloy surface in the anodizing and sealing processes. Additionally, R₁ is in series with a second R₂//CPE₂ sub-circuit in parallel, for the low-frequency response measurements, which may be attributed to the corrosion process, where the CPE₂ represents the double-layer capacitance, and the R₂ is the charge transfer resistance, which is inversely proportional to the corrosion rate. A CPE is defined as an empirical function depending on the frequency and its impedance is defined as: Z_CPE=(Y)⁻¹(jω)⁻ⁿ, where Y is the admittance (a real frequency-independent constant), j²=(−1) is the imaginary number, ω is the angular frequency, and n is a dimensionless fraction exponent (−1<n<+1). When n=+1 the CPE is an ideal capacitor, and when n=−1 the CPE is an inductor (Scully et al., 1993; Perez, 2004; Bastidas, 2007; Hirschorn et al., 2010; Jinlong et al., 2016; Evertsson et al., 2017; Ayagou et al., 2018; Halvorsen et al., 2019).

Table 2 shows the four specimens (M1−M4) have a similar R_s value in the range of 29.38−33.65 Ω cm². The value of the dielectric properties of the oxide films generated on the anodized and sealed AA2099 alloy surface, which correspond to the admittance (Y₁) in the CPE₁ (Bastidas, 2007), were found to be of the same order of magnitude for each specimen (1.07−2.06 μS·cm⁻² sⁿ¹), indicating protective properties. The R₁ parameter for each specimen was found to be of the same order of magnitude, except for the specimen M3 (0.19 A·cm⁻² current density, and 6 wt.% Na₂Cr₂O₇ sealing solution) having a R₁ value one order of magnitude lower (750.5 Ω cm²). This behavior is observed because the 6 wt.% Na₂Cr₂O₇ solution does not reduce the diameter of the pores, therefore it behaves as a corrosion inhibitor by a precipitation mechanism inside the pores, which hinders the corrosion process and causes cracking, as was observed in Fig. 2c (Zuo et al., 2003; Yang et al., 2019). The n₁ parameter presents high values inducing non-depressed semicircles, which indicates a homogeneous oxide film surface (Djellab et al., 2019). Figure 3 shows that the smallest defined semicircle was found for the sample M1 (0.19 A·cm⁻² current density, and deionized water sealing solution) having a R₂ value one order of magnitude lower (178.32 Ω cm²) than M2, M3 and M4 samples (see Table 2). The effect of water as a sealing solution is to produce a reduction of the porous diameter of the honeycomb structure, probably the 0.19 A·cm⁻² current density presents some limitations generating a low corrosion resistant oxide film. The capacitance of the electrochemical double-layer (Y₂) was in the range of 1.14−4.33 μS·cm⁻² sⁿ², and the n₂ parameter presented high values (0.80−0.97), very close to an ideal-capacitor behavior (n=1), see Table 2.

Figure 5 shows Nyquist plots for anodized and sealed AA2099 alloy immersed in 10 vol.% H₂SO₄ solution for 3 h. The four Nyquist plots show the same trend, a depressed capacitive loop behavior from high- to intermediate-frequency, and a depressed inductive loop at low-frequencies in the fourth quadrant. The inset of Fig. 5 shows the Bode plots. The impedance results were fitted using the EEC of Fig. 4, having two-time constants as indicated above, where R_s, CPE₁, and R₁ have the same meaning as Fig. 3, and the R₂//CPE₂ sub-circuit is associated with the inductive loop, which may be explained by the possible adsorption on the AA2099 alloy surface of some species such as H⁺ and/or SO₄²⁻ generated in the aggressive 10 vol.% H₂SO₄ medium (Pivac and Barbir, 2016; Djellab et al., 2019; Klotz, 2019). This process was simulated with CPE₂ and R₂ parameters having negative values. Negative capacitance (Y₂) and resistance (R₂) arise from surface coverage and adsorption-desorption processes, respectively (Bastidas et al., 2001; Jinlong et al., 2016; Jirón-Lazos et al., 2018). The highest R₁ value was obtained for the specimen M3 (32285.5 Ω cm²) (see Table 2), showing allows the precipitation mechanism of dichromate acting as a corrosion inhibitor on the porous Al₂O₃ layer. The corrosion inhibition observed for treated AA2099 alloys can be primarily attributed to the reduction of the active surface area by the anodized layer (Ardelean et al., 2009). The n₁ parameter value is high (0.87−0.95), meaning a homogeneous surface of the anodic oxide layer. It can be observed on Fig. 5 that R₂ value for M1 and M2 specimens (−40.80 and −20.39 kΩ cm²) is the highest, which may be attributed to a reduction in the diameter of the pores of the anodic oxide layer (Zuo et al., 2003; Yang et al., 2019). The n₂ parameter presents low values (see Table 2), producing depressed semicircles, which may be attributed to the heterogeneity of the surface or to the interfacial phenomena (Djellab et al., 2019).

3.3. XPS surface analysisTOP

XPS experiments were performed to determine the chemical composition of the oxide layer generated by anodizing using 0.19 or 1.0 A·cm⁻² current density, and sealing using H₂O or 6 wt.% Na₂Cr₂O₇ solution, on the AA2099 alloy. Figure 6 (a-h) shows high resolution XPS spectra for Al 2p and O 1s. Spectra for Al 2p shows five chemical species assigned to Al₂O₃, Al₂S₃, Al₂(SO₄)₃, AlO(OH) and Al₄C₃, (74.5 eV binding energy), and metallic Al, see Fig. 6 (a, c, e, g). The XPS spectra of O 1s reveal the typical binding energies for O 1s, 531.0−532.5 eV (Romios et al., 2005; Pérez et al., 2010; Abrahami et al., 2015), see Fig. 6 (b, d, f, h). In the current study Cr 2p_3/2 signal was not detected, according to Jaimes-Ramirez et al. (2018), this may be due to a migration process of the chromate ions into the interior of the Al₂O₃ layer (Jaimes-Ramirez et al., 2018). AA2099 specimens sealed in H₂O presented a surface film rich in Al₂(SO₄)₃, this is attributed to the initial H₂SO₄ anodizing process, while for the Na₂Cr₂O₇ sealing the AlO(OH) was the main oxyhydroxide found on the surface. Among the results, it is possible to observe compounds where the presence of significant amounts of C is found on the surface of the oxide layer, which may be attributed to the accumulation of pollutants during exposure to air (Skeldon et al., 1997; Ryu et al., 1999; Romios et al., 2005; Liou et al., 2013; Abrahami et al., 2015; Khan et al., 2019).

4. CONCLUSIONSTOP

ACKNOWLEDGEMENTSTOP

The authors would like to thank the Mexican National Council for Science and Technology (CONACYT) for the support provided for the development of the projects CB 253272 and A1-S-8882, the UANL-CA-316 working group and Universidad Autónoma de Nuevo León (UANL) for the facilities given to develop this investigation. U. Martin, J. Ress and D.M. Bastidas acknowledge funding from Firestone Research Grant 639430 and The University of Akron.

Conflict of InterestTOP

The authors declare that they have no conflict of interest.

REFERENCESTOP