Time-Dependent Corrosion Behavior of Aluminum Using Advanced Electrochemical and Characterization Techniques

Sample preparation

Scanning electron microscopy

Transmission electron microscopy

Secondary ions mass spectroscopy

Cyclic potentiodynamic polarization (CPP), electrochemical impedance spectroscopy (EIS), and electrochemical noise measurements (ECN) tests were performed on pure Al in 0.01 M NaCl (pH ∼ 5.9) using VMP-300 potentiostat (Biologic) with the help of the EC-lab software. The sample preparation for the electrochemical tests involved mounting in epoxy and grounding to 1200 grit using SiC paper finish, followed by ultrasonication in ethanol for 10 min. An adhesive polymeric tape was used to avoid crevice corrosion and maintain the required exposed area. The exposed area for CPP and EIS tests was 0.178 cm², while it was 0.079 cm² for the ECN test. The electrochemical tests were conducted using a classical three-electrode flat cell with a Pt mech counter electrode. In addition, a saturated calomel reference electrode (SCE) was utilized as a reference electrode for all the electrochemical tests. At least three tests were performed on each condition to assure reproducibility.

The open-circuit potential (OCP) of the Al specimen was measured for 30 min and 14 d in 0.01 M NaCl. After that, the polarization was performed with a sweep rate of 0.167 mV.s⁻¹ towards the more noble potential direction starting from 50 mV_SCE below the OCP until the current density reached 200 μA.cm⁻². Then the scanning direction was reversed until the potential reached 100 mV_SCE below the OCP.

The time-dependent corrosion performance of pure Al was investigated using EIS. The sample was immersed in 0.01 M NaCl for 30 min, followed by measurements using an AC sinusoidal voltage signal with 10 mV amplitude with respect to OCP. The test frequency range was varied from 100 kHz to 1 mHz, and ten data points were acquired in each frequency decade. The EIS tests were performed every 12 h until 336 h of immersion.

Electrochemical noise measurements were conducted on a pure Al specimen after 30 min of immersion in 0.01 M NaCl. The current and potential noise were recorded for 1024 s at a 1 Hz frequency. After the initial scan, the ECN data were collected every 12 h for 336 h of immersion. The sample preparation for the ECN involved mounting two samples (not in contact) in an epoxy rod with the same exposed area. Two specimens were electrically interconnected using the working electrode and counter electrode terminals. The working electrode potential was recorded with respect to the SCE reference electrode, and the current pulse between the samples was recorded using the zero-resistance ammeter mode in the potentiostat. The power spectral density (PSD) of potential and current was used to estimate the noise resistance (R_n), and the procedure for the R_n estimation is described in Refs. 33, 37.

SEM and EDXS elemental maps of pure Al after 30 min and 336 h of immersion in 0.01 M NaCl are shown in Figs. 1 and 2, respectively. Pits were formed in both samples (Figs. 1a and 2a). After 30 min of immersion, the average pit diameter and pit area fraction were (2.3 ± 0.5) μm and 0.23%, respectively. However, both the parameters were increased over the immersion, reaching (6.1 ± 3.0) μm average pit size and 1.2% pit area fraction after 336 h of immersion. Figures 1b and 2b revealed the presence of cracks along the edge of the pit. EDXS elemental mapping exhibited Fe-Si-rich and Si-rich particles after 30 min and 336 h of immersion, respectively. Interestingly, the EDXS elemental mapping after 336 h of immersion (Fig. 2) revealed the absence of elemental Fe. This could be attributed to the corrosion products deposition on the Fe-rich particles over the immersion time, which hindered the Fe detection during EDXS analysis. The presence of the above-mentioned Fe-Si-rich and Si-rich particles facilitate the pit initiation due to the galvanic interaction. ³⁸ Apart from the pits around heterogeneous sites, tiny pits were observed throughout the surface (Fig. 2a).

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Representative CPP curves for pure Al after 30 min and 336 h of immersion in 0.01 M NaCl are shown in Fig. 3. The average OCP, average passive current density 30 mV_SCE below pitting potential, and the average pitting potential (E_pit) of each condition are presented in Table I. The average OCP was nobler in 30 min of immersion than that in 336 h of immersion. The average passive current density 30 mV_SCE below pitting potential was ∼(0.9 ± 0.7) μA.cm⁻² after 336 h of immersion, which is 4.5 times higher than that in 30 min. The active OCP and increased passive current density in 336 h of immersion indicate accelerated corrosion over the immersion time. The breakdown potentials were –554 mV_SCE and –542 mV_SCE after 30 min and 336 h of immersion, respectively.

Zoom In Zoom Out Reset image size

Table I. The average electrochemical parameters for pure Al are estimated from Fig. 3.

Immersion time (hours)	OCP (mVSCE)	i (30 mVSCE below Epit )(μA.cm−2)	Epit (mVSCE)
0.5	$-620\pm 8$	$0.2\pm 0.07$	$-554\pm 10$
336	$-890\pm 20$	$0.9\pm 0.7$	$-542\pm 57$

Electrochemical impedance spectroscopic analysis of pure Al for 336 h of immersion in 0.01 M NaCl is shown in Fig. 4. During the EIS test, the current passes through the passive film in two different ways for Al and its alloys depending on the applied frequency of the wave: At high frequencies, the current flows through the passive film, while at lower frequencies, localized corrosion regions such as pits are the current paths, where the passive film becomes an insulator. ^39–42 Therefore, passive film and localized corrosion are characterized using various frequencies in EIS. The EIS results were fitted using Zview software, as shown in Table II. The solution resistance (R_s), constant phase element of the film (CPE_film), and film resistance (R_film) were estimated using a high-frequency region. While the double layer constant phase element (CPE_dl) and the charge transfer resistance (R_ct) were measured at a lower frequency region. A constant phase element was used to represent the ideal capacitor deviation.

Zoom In Zoom Out Reset image size

Figure 4a shows the Nyquist plot for pure Al at different immersion times. The Bode plot (Fig. 4b) revealed an increment of total impedance (|Z|) in pure Al over the immersion time until 336 h. The surface film formed on the Al substrate was characterized by two primary time constants: one was at high frequencies while the other one was at low frequencies. Additionally, a time constant has been observed at around 10⁵ Hz. This high-frequency response could be attributed to artifacts of the cell. After 12 h of immersion, R_film was 23 kΩ.cm² and increased gradually over the immersion time and reached its highest value (101 kΩ.cm²) after 336 h of immersion. This increment could be attributed to the surface film growth over the immersion time. R_ct was 115 kΩ.cm² after 12 h and decreased to 58 kΩ.cm² after 60 h of immersion, possibly due to the accelerated localized dissolution of pure Al. Then, R_ct was increased slowly over the immersion time and reached its highest value (259 kΩ.cm²) after 336 h of immersion. This R_ct enhancement could be attributed to the deposition of corrosion products inside the pit, which could act as a barrier for ion transportation between electrolyte and substrate. ²¹ The highest CPE_film (29 μF.cm⁻²) was noted after 336 h of immersion, starting with a gradual increment from 18 μF.cm⁻² (after 12 h of immersion). This indicates that the surface film's hydration effect due to the various ions penetration is more dominant than the influence of surface film growth over the immersion time. ⁴³ CPE_dl was 252 μF.cm⁻² after 12 h of immersion and fluctuated over the immersion time before reaching its highest value (545 μF.cm⁻²) after 48 h.

Electrochemical noise measurements of pure Al for different immersion times in 0.01 M NaCl are presented in Fig. 5a. The potential noise fluctuated significantly after 30 min of immersion and reached better stabilization over the immersion. The magnitude of the current transients and the average current density increased over the immersion time and reached the highest current transition after 168 h of immersion. This indicates the increment of localized corrosion activities, such as the initiation of metastable pits and the propagation of stable pits over the immersion time. The continuous corrosion inside of a stable pit compared to that of a metastable pit, which repassivates results in larger radii pits. ²⁸ In 336 h, magnitude, and the average current density dropped and were closer to those values in 30 min of immersion. The lifetimes of an arbitrarily selected current noise after 30 min and 336 h of immersion are presented in Figs. 5c and 5d, respectively. A formed current noise quickly disappeared after 30 min of immersion than that in 336 h, revealing the reduction of the repassivation tendency of the localized activities over the immersion. ⁴⁴ Figure 5b shows the R_n fluctuation over the immersion time. The highest magnitude and average R_n were 932 kΩ.cm² and 60 kΩ.cm², respectively, observed after 30 min of immersion. However, a gradual reduction of the average R_n (Table III) was observed until 60 h of immersion, reaching the lowest R_n (36 kΩ.cm²) followed by a slight increase for the rest of the immersion time.

Zoom In Zoom Out Reset image size

Transmission electron microscopic analysis was performed on the cross-section of pure Al after 30 min of immersion in 0.01 M NaCl, as shown in Fig. 6. Low magnification bright field image (Fig. 6a) of the cross-section revealed the presence of several grains. Figure 6b shows the passive film at the distinct grains and grain boundaries. A zoomed-in region of a grain boundary was captured (Fig. 6b) and exhibited the formation of a non-uniform passive film on top of the surface. The formed passive film on both grains was amorphous in nature (Fig. 6c). The average passive film thickness formed on grain 1 was (2.3 ± 0.7) nm, while it was (5.0 ± 0.8) nm for grain 2, indicating the influence of the grain orientation on passivation. Figures 6d–6f exhibit the Al metal/film interface. Figure 6f exhibits a distinguished interface of the film/metal, revealing the crystalline Al substrate and the amorphous oxide film.

Zoom In Zoom Out Reset image size

Figure 7 shows the STEM images of the cross-section of pure Al after 168 h of immersion in 0.01 M NaCl. The deposition of a non-uniform corrosion products layer throughout the surface was observed in Fig. 7a. This could be attributed to the continuous dissolution of the surface film, which increased the number of new pitting sites over the immersion time. The low-magnification image (Fig. 7a) shows several pit formations within grains and grain boundaries. A cross-section of a pit shown in red square in Fig. 7a is presented in Fig. 7b. The EDXS elemental map of Fig. 7b is shown in Figs. 7c–7f, which reveals the distribution of Al, O, C, Cl, and Pt. The O elemental map exhibited significant corrosion product deposition inside the pit. This deposition could cause the prevention of ion transportation between the pit and the bulk electrolyte. ²¹

Zoom In Zoom Out Reset image size

The depth profile of pure Al was analyzed using secondary ion mass spectroscopy after 30 min, 168 h, and 336 h of immersion in 0.01 M NaCl and is presented in Fig. 8. Al₂‾, AlO₂‾, and ¹⁸O₂‾ in the curves represent metallic Al, oxidized Al, and oxide, respectively. Two regions were defined as surface film and substrate, where the metallic Al intensity became steady, and oxide ions of AlO₂‾ and ¹⁸O₂‾ signals dropped. ^45,46 However, Al₂‾ ion count was used to define the surface film/substrate interface as oxide ions of AlO₂ ⁻ and ¹⁸O⁻ signals do not drop instantly with the sputtering time in 168 h and 336 h of immersion. The intensity of Al₂‾ was steady after ∼82 s, 490 s, and 1024 s sputtering time after 30 min, 168 h, and 336 h of immersion, respectively. This reveals the surface film growth over the immersion time and could be attributed to the surface film resistance improvement, as indicated in the EIS analysis. The outermost surface film (gray shaded area) shows the presence of OH‾ and Cl‾ ions, indicating the presence of hydroxide formation. The thickness of this hydroxide layer increased over the immersion time from ∼22 s (30 min.) to ∼325 s (336 h). Moving towards the metal surface, the AlO₂‾ ion indicates the formation of the passive oxide layer. After 30 min of immersion, the OH‾ and Cl‾ ions showed a sudden drop as the AlO₂‾ ion spiked up, indicating an effective barrier due to the passive film and preventing chloride ions penetration. With the increase in immersion time (i.e., 168 and 336 h), the progression of the Cl‾ ion signal within the oxide region reveals the successful penetration of chloride through the passive film, which gradually dropped with the sputtering time. Meanwhile, the progression of OH‾ ion shows the formation of corrosion product over the passive film. The gradual penetration of Cl‾ ion through the passive film with the immersion time could have reduced passive film stability.

Zoom In Zoom Out Reset image size

SIMS analysis revealed the chloride penetration through the surface film once the sample is exposed to a chloride-containing environment. This chloride penetration could be ascribed to the oxygen vacancies in the passive film, as widely accepted in the literature. ^5,7,19,47,48 As a result, the surface film stability reduces, allowing the transportation of water molecules through the surface film. ⁵ High OH‾ intensity at the top surface showed that the surface film dissolution occurs at the film/solution interface. The hydrating passive film involves the reaction between Al-O-Al bonds and water molecules to generate Al-OH bonds. ⁴⁹ Each water molecule consumption results in the formation of two Al-OH bonds. Additionally, hydrolysis could lead to lattice modification and changing bond lengths. ⁴⁹ Such a structural modification in the passive film increases the passive film's corrosion susceptibility, leading to the surface film's continuous dissolution over the immersion time. The chloride penetration increased significantly over the immersion time. This could have resulted in the continuous dissolution of the surface film over the exposure time, as evidenced by the increased intensity of OH‾ ions over the immersion in SIMS analysis. The continuous dissolution would lead to the enhanced conductivity of the surface film, which resulted in the surface film capacitance increments over the immersion time in the EIS analysis.

It has been widely accepted that the pitting initiation takes place near the metal film interface in Al, resulting in the formation of nanoscopic voids and/or blisters within occulled cell. ^5,22,50 The localized dissolution of formed Al³⁺ within the pitting site causes localized acidification. Therefore, chloride ion transportation takes place toward the pit to maintain charge neutrality inside the pit. ⁵¹ The formation of a critical pit environment causes pit growth, resulting in localized cracking of the passive film and direct exposure between the pit and the electrolyte. Once the pit has been exposed to the electrolyte, the pitting kinetics has been accelerated. To support that, a considerably decreased charge transfer and noise resistance within 12–60 h of immersion has been observed (Tables II and III), which indicates increased localized corrosion kinetics over the immersion time until 60 h. In addition, a significant increase in the double-layer CPE was detected between 12–48 h of immersion. The faster dissolution of Al from the matrix could lead to excess Al³⁺ ions at the outermost part of the surface film. This phenomenon possibly caused strong charge separation between the outmost surface at localized dissolution points and the electrolyte, resulting in a higher double-layer CPE. The stability of these formed pits is mainly determined by the sustainability of the pit critical pit solution, which prevents the repassivation inside the pit. ⁵¹ It has been reported that the chloride ion concentration also plays a critical role in pit current increment because it helps to maintain the acidic environment inside the pits. ⁵ However, ion transportation between the electrolyte and the pit solution is the key to enhanced corrosion kinetics over time. Interestingly, the deposition of corrosion products within the pits was observed in the cross-sectional analysis of the pit over the immersion time (Figs. 7b and 7d). This deposition could act as a barrier to ion transportation between the electrolyte and the substrate. ²¹ Thus, it can be proposed that the deposition of corrosion products inside the stable pits resulted in an increase in the charge transfer resistance, which controlled localized corrosion kinetics after 60 h of immersion. Although the continuous dissolution of the surface film resulted in the increment of overall corrosion kinetics of pure Al with the increase in the immersion time, as indicated by the passive current density increment in Table I and ECN resistance reduction (Table III).