Influence of Partial Er Substitution for Sc on the Microstructure, Mechanical Properties and Corrosion Resistance of Short-Processed Al-4.7Mg-0.6Mn-0.3Zr-0.3Sc Sheets

Abstract

Standard AA5083 (ZSE000), AA5083 modified with 0.3 wt.% Zr and 0.3wt.% Sc (ZSE330) and AA5083 modified with 0.3 wt.% Zr, 0.2wt.% Sc and 0.1wt.%Er(ZSE321) sheets were fabricated through a short process (including a simulated twin-belt continuous casting, subsequent direct rolling, intermediate annealing, cold rolling and stress-relief annealing) to systematically investigate the influence of partially substituting Er for Sc on the microstructure, mechanical properties and corrosion resistance of short-processed Al-4.7Mg-0.6Mn-0.3Zr-0.3Sc sheets. The results show that ZSE321 presents the optimal tensile properties (UTS: 541 MPa; 0.2%PS: 469 MPa and EF:7.7%) among the three experimental sheets. This is attributed to significant grain refinement, the inhibition of the recrystallization and promotion on the precipitation of Al₃(Sc, Zr, Er) nanoparticles. Furthermore, the corrosion properties of the experimental sheets were also explored in this study, and the short-processed ZSE321 sheet presents the optimum corrosion resistance.

1. Introduction

Al-Mg series alloys have been widely used in various industries, including the automotive, marine, packaging and construction industries, due to their excellent weldability, ductility, toughness, formability, and corrosion resistance [1,2,3,4,5]. However, their strength cannot be enhanced through work hardening [6], only by grain refinement [7,8,9]. Nevertheless, many cold-worked or grain-refined Al-Mg series sheets still exhibit limited strength and a poor thermal stability, primarily because of the recrystallization and grain-growth at elevated temperatures [10,11,12]. Sc and Zr elements [13,14] have been conventionally introduced into Al-Mg based alloys as they are beneficial to their mechanical properties. These benefits have been attributed to the refinement of grains and dendrites [15], the inhibition of recrystallization [16] and the formation of nano-sized Al₃Sc_xZr_1−x phases [17]. Notably, a commercial Al-3Mg alloy is fully recrystallized after annealing at 360 °C, whereas the synergy between Zr and Sc resists its recrystallization even when it has been annealed for 8 h for 520 °C [18]. This synergy significantly promotes the annealing to play an important role in improving the strength and ductility of Al-Mg alloys. Y. Miyake et al. [19] have proposed that a mechanically balanced Al-Mg-Mn-Sc-Zr alloy with a high hardness and high formability can be fabricated by not only adding scandium and zirconium but also optimizing the heat treatment conditions. J.J Shen et al. [5] found that the hardness and strength decreased with an increase in the temperature and duration of annealing. Recently, the Er element has emerged as a promising alternative [20] to more expensive Sc, aiming to improve the mechanical properties Al-Mg series alloys.

Amidst the increasingly serious environmental and energy crises, some short processes, such as twin-belt continuous casting (TBCC) and twin-roll casting (TRC), have been gradually employed in aluminum alloys manufacturing processes in recent decades to produce sheets because of its characteristic superiorities which include low investment costs, a low energy consumption and a high efficiency [21,22]. However, compared to traditional direct chill casting (DC) processing route, Al–Mg alloys produced via these short processes present distinct microstructural characteristics and mechanical properties. This difference arises from the higher solidification rate, the homogenization savings and the decrease in plastic deformation. J. Li et al. [23] investigated the recrystallization microstructure and texture of cold-rolled continuous cast AA5083 and 5182 aluminum alloys, and found that a heat treatment prior to cold rolling strongly affected the recrystallized grain structure and recrystallization texture of AA5083 aluminum alloy. Ch. Gras et al. [24,25] systematically observed surface microdefects, center line defects and texture evolution during the twin-roll casting of Al–Mg–Mn alloys. Sanjeev Das et al. [26] explored the effect of the rolling speed on the microstructural and mechanical properties of Al-Mg alloys prepared by twin roll casting, and found that tensile strength and hardness (at the cell interior) decrease with an increase in rolling speed.

However, until now, many published works [27,28] have primarily focused on the short process’ parameters and the subsequent process of the aluminum sheets, and few attempts [29] have been carried out to explore the influence of micro-alloying elements and heat treatment on the microstructure, mechanical properties and corrosion behavior of short-processed Al-Mg series sheets. Therefore, as part of a commission from a Luoyang factory to develop Al-Mg alloy sheets for the automobile and marine industries, this paper systematically investigated the influence of 0.1 wt.% Er substituted for Sc on the microstructure, mechanical properties and corrosion resistance of Al-4.7Mg-0.6Mn-0.3Zr-0.3Sc sheets processed through a short process which included a simulated TBCC, subsequent direct rolling, intermediate annealing, cold rolling and stress-relief annealing (abbreviated to TBCCRS).

2. Materials and Methods

The experimental AA5083 alloys with Er and Zr additions were prepared from Al–5Mn (wt.%, hereafter in weight percentage), Al-5Zr, Al-2Sc, Al-2Er and Al-5Cr master alloys, pure commercial Al and pure commercial Mg. A schematic diagram of the short process including a simulated TBCC, subsequent direct rolling, intermediate annealing, cold rolling and stress-relief annealing (TBCCRS) is illustrated in Figure 1. Firstly, the commercial pure Al bulks were melted in a graphite crucible with a capacity of 1.5 L in an electrical resistance furnace. The preheated master alloys and pure commercial Mg were successively added into the furnace at 750 °C, after the pure Al had melted. The melt was degassed with pure Ar with a constant flow rate of 0.5 L/min for 20 min. After standing for about 5 min, the surface impurities were removed and the melt was poured into a copper mold preheated to 220 °C. Subsequently, the cast slabs were quickly removed and maintained at a temperature slightly above 400 °C. The weight of each cast slab was ~1 Kg. When the temperature of the cast slabs was directly decreased into 400 °C, the slabs were immediately rolled in three consecutive passes, through utilizing the heat of the slab itself. A thickness reduction of approximately 30% in each pass was applied. Note that the cooling rate of the melting in the copper mold was about ~100 °C/s, which corresponded to the data actually measured in the factory from Luoyang. On the other hand, before the first pass rolling, the temperature of the cast slabs was selected to being directly decreased into 400 °C from the cast slabs, based on the data measured in the factory. After the direct rolling process with three passes, the intermediate annealing at three temperatures was carried out for these direct-rolled sheets, which promoted the formation of intermetallic made of Sc, Zr and Er elements.

After the intermediate annealing, the sheets were further cold rolled at room temperature to a thickness of ~1 mm, which is a ~90% reduction. The cold rolling totally took place in six passes. Finally, the cold-rolled sheets were subjected to an annealing at 175 °C. The chemical composition of the experimental sheets according to spectral analysis (Oxford Instruments) is presented in

Table 1. Chemical composition of experimental alloys.

Alloys	Mg	Mn	Cr	Ti	Zr	Sc	Er	Al
ZSE000 (AA5083)	4.73	0.60	0.16	0.14	-	-	-	Bal.
ZSE330	4.71	0.61	0.16	0.14	0.29	0.30	-	Bal.
ZSE321	4.71	0.60	0.16	0.14	0.29	0.20	0.10	Bal.

. As can be seen, three sheets named as ZSE000 (a standard AA5083), ZSE330 (AA5083 modified with 0.3 wt.% Zr and 0.3 wt.% Sc, hereafter in weight percentage) and ZSE321 (AA5083 modified with 0.3% Zr, 0.2% Sc and 0.1% Er) were processed by TBCCRS.

The cast slab and rolled sheet specimens were prepared for microstructural observation by the standard mechanical polishing and etching technique. The microstructure of the experimental specimens was examined by an optical microscope (OM) and a Quanta-2000 scanning electron microscope (SEM) equipped with energy dispersive x-ray spectroscopy (EDS). The mean linear intercept method [30] was used within the ImageJ 1.54j software to measure the average grain size of the experimental slabs and sheets. The fine precipitates of the studied samples were identified by a FEI Talos F200X transmission electron microscope (TEM). Thin foils for TEM observation were mechanically ground to ~40 μm in thickness and then polished by a Leica EM RES102 ion milling machine. A FM-800 micro-hardness tester was used to measure the hardness of the experimental sheets at an applied load of 100 g for 15 s. The tensile specimens with a gauge length of 45 mm and weight of 8 mm were machined and polished according to the ASTM E8 standard [31]. Tensile tests were carried out on an Instron 8801 loading frame at a crosshead speed of 0.2 mm/min. Post-loading fractography was also performed by SEM.

Electrochemical corrosion tests were conducted at room temperature by using a Gamry Reference 1000 potentiostat/galvanostat. The testing samples were a 0.5 cm² exposed working electrode, a platinum wire coil as the counter electrode and a saturated calomel electrode (SCE) as the reference electrode. The continuous electrochemical impedance spectroscopies (EISs) were measured under constant stirring conditions (250 rpm) in a naturally aerated 3.5 wt.% NaCl solution after the open circuit potential (OCP) was stable, with an amplitude of 10 mV RMS sinusoidal perturbation over a frequency range from 100 kHz to 0.1 Hz. Tafel curves were performed in a naturally aerated 3.5 wt% NaCl solution, at a scan rate of 1.0 mV/s. Nitric acid was substituted by hydrochloric acid for security; thus, the hydrochloric acid mass loss test (HAMLT) was carried out to determine the corrosion susceptibility according to the ASTM-G67 standard [32]. The corrosion rate (mg·cm⁻²·h⁻¹) can be calculated using the published formulas [33,34].

3. Results and Discussion

3.1. Microstructural Evolution

Figure 2 shows the microstructure of the AA5083 casting slabs (Figure 2a–c) micro-alloyed with various Zr, Sc and Er additions and their corresponding sheets (Figure 2d–f) that were processed by three consecutive passes of direct rolling. The unmodified AA5083 (ZSE000) slab presented numerous coarse bulk α-Al dendrites. However, the α-Al grains in AA5083 slab were markedly refined by the 0.3% Zr and 0.3% Sc additions (ZSE330), as shown in Figure 2b. When 0.1% Sc was substituted with 0.1% Er (ZSE321), the grains in the slab were further refined. After undergoing three consecutive passes of direct rolling, the bulk α-Al dendrites were crushed into fine capsule grains. The ZSE330 sheets present finer grains than the ZSE000 sheets, while the ZSE321 sheets show the finest grains among all the experimental sheets, which is the same as the experimental casting slabs. This indicates that the grains can be significantly refined by direct rolling using the residual temperature from the casting.

Figure 3 shows the representative SEM micrographs obtained from the experimental casting slabs (Figure 3a–c) and their subsequent direct-rolled sheets (Figure 3d–f). Some white bulk particles can be found in the AA5083 casting slabs, as depicted in Figure 3a; the EDS analysis indicates that these white bulk particles are in the Al(Mg,Mn,Cr) phase. Upon the addition of 0.3% Zr and 0.3% Sc, the number of white intermetallic particles in the casting slabs increase. According to the EDS analysis, in addition to the Al(Mg,Mn,Cr) intermetallic particles, the presence of the Al₃(Zr,Sc) phase is also confirmed, represented by arrows A and B, respectively, in Figure 3a–c. When 0.1% Er is substituted for Sc, there is a further increase in the number of white intermetallic particles, and the Al₃(Zr,Sc) intermetallic particles are modified into the intermetallic Al₃(Zr,Er,Sc) phase, as evidenced by the EDS analysis. The application of the direct rolling including three consecutive passes resulted in the fragmentation of the bulk particles in the experimental casting slabs, and these fragmented particles were gradually dispersed in a ribbon-like pattern along the rolling direction.

After the direct rolling process, intermediate annealing was performed on the direct-rolled sheets at various temperatures (380 °C, 400 °C and 420 °C). Figure 4 illustrates the microhardness of the direct-rolled sheets at these three annealing temperatures over different durations. The trend in microhardness variation with annealing time indicates a sharp decrease initially, followed by a gradual decline after 2 h, regardless of the specific annealing temperature. After 4 to 6 h of annealing, the hardness variation reaches a stable state. Consequently, an intermediate annealing time of 6 h can be optimally selected.

The microstructures of the direct-rolled sheets processed by three intermediate annealing at different temperatures for 6 h has been were assessed with the aim of precisely identifying the optimal intermediate annealing temperature. The polarized optical microscopy (OM) images are presented in Figure 5. Compared with the direct-rolled ZSE000 sheets without intermediate annealing, the ZSE000 sheets subjected to intermediate annealing exhibit an enlargement in the capsule grain size, displaying a gradual increase in grain size as the intermediate annealing temperature is raised. However, there are no significant changes in the grain size of the direct-rolled ZSE330 and ZSE321 sheets and their grain size slightly increases with an increase in the intermediate annealing temperature.

Furthermore, Figure 6 displays SEM images of the direct-rolled sheets processed by three rounds of intermediate annealing at different temperatures for 6 h. During the intermediate annealing, the Al(Mg,Mn,Cr) intermetallic particles in the direct-rolled ZSE000 sheets dissolved into the Al matrix, and the number of dissolved intermetallic particles increased with the intermediate annealing temperature. Figure 7 presents the statistical results for the percentage of the area of the intermediate annealed sheets occupied by intermetallic particles. From Figure 7, it can be seen that the percentage of the area of the ZSE000 sheet annealed at 380 °C occupied by intermetallic particles is 0.27%. The addition of synergetic Zr and Sc significantly increased the percentage of the area occupied by intermetallic particles to 0.71%. When 0.1% Sc is substituted by 0.1% Er, the percentage of the area occupied by intermetallic particles is further increased into 0.89%. On the other hand, for all the intermediate annealed sheets (ZSE000, ZSE330 and ZSE321), the percentage of the area occupied by intermetallic particles slightly decreases as the annealing temperature increases.

Since the variation in the grain sizes and intermetallic particles in the sheets intermediately annealed at three temperatures are insignificant, and the microstructure of the annealed sheet at 400 °C is slightly better, the annealed sheet at 400 °C was selected to carry out the subsequent process and detailed microstructural observations. To further reveal the grain structures in the intermediate annealed sheets along the rolling direction, an EBSD analysis was carried out on the samples intermediately annealed at 400 °C for 6 h. In Figure 8, the EBSD micrographs and corresponding orientation angle distribution maps of the grain boundaries can be seen. In Figure 8a–c, the black lines indicate the high-angle grain boundaries (HAGBs) with s misorientation angle greater than 15°, and the red lines represent the low-angle grain boundaries (LAGBs) with s misorientation of 2°~15°. The histograms in Figure 8d reveal that the grain boundary misorientation angles in the intermediately annealed ZSE000 sheet are both LAGBs (49.8%) and HAGBs (50.2%). In contrast, the histograms in Figure 8e,f exhibit the grain boundary misorientation angles in the ZSE330 and ZSE321 sheets and show that they are significantly dominated LAGBs by rather than HAGBs. The DRX distribution and statistical results for the intermediately annealed sheets further support the observations made from IPF maps. The fraction of recrystallized grains dramatically decreased from 66.1% to 1.0% because of the addition of synergetic Zr and Sc, and further decreased to 0.8% when the 0.1% Sc was substituted with 0.1% Er, as shown in Figure 8g–l. These results indicate the remarkable ability of Sc, Zr and Er to significantly inhibit the recrystallization process in AA5083 (ZSE000) sheets during intermediate annealing.

Following the intermediate annealing, the experimental sheets underwent six passes of cold rolling, with each pass achieving a 20% reduction in thickness. Figure 9a–c shows polarized OM images of the cold-rolled sheets that had been previously processed by intermediate annealing at 400 °C for 6 h. From Figure 9, it can be seen that the grains in all the intermediately annealed sheets were cold rolled into a banded structure. Figure 10a,c show the width of the grains in the cold-rolled sheets with or without relieved-stress annealing. Compared to the cold-rolled ZSE000 sheet, the ZSE330 sheet exhibits a notably narrower banding structure, with a grain width of 7.2 μm. The introduction of 0.1% Er to the ZSE321 sheet by substituting a fraction of Sc leads to an even more pronounced decrease in the grain width, down to 5.3 μm. After stress relief annealing, the grain width of the experimental sheets has no significant change.

Furthermore, Figure 11a–c displays the SEM images of the cold-rolled sheets that had previously been processed by intermediate annealing at 400 °C for 6 h. Compared with the intermediately annealed sheets, the dissolved intermetallic particles in the cold-rolled sheets are finer and distributed in bands. Like the situation in the cast slabs and the intermediately annealed sheets, the addition of synergetic Zr and Sc increases the number of white intermetallic particles in the cold-rolled AA5083 sheet, and the fractional Er substitution for Sc causes a further increase in the number of intermetallic particles. Figure 10b presents the statistical results on the size of the intermetallic in the cold-rolled sheets. It was found that the diameter (4.0 μm) of the intermetallic particles in the cold-rolled ZSE321 sheet is larger than that of the particles in the other cold-rolled sheets.

As is well known, cold rolling enhances the strength of aluminum sheets but often comes at the cost of reduced plasticity. Thus, stress-relief annealing must be carried out on the cold-rolled sheets, to ensure they have a better formability during the parts manufacturing process. In this study, the cold-rolled experimental sheets were subjected to annealing at 175 °C for 8 h. Figure 9d–f show the polarized OM images of the cold-rolled sheets under stress-relief annealing at 175 °C for 8 h, and Figure 10d–f shows their SEM images. In Figure 9 and Figure 10, the detailed variations in the banded grains and intermetallic particles in the cold-rolled sheets after the stress-relief annealing could be not observed; thus, statistical measurements were carried out, as shown in Figure 10c,d. It is can clearly be seen that the ZSE321 sheet after stress-relief annealing exhibits the lowest width of grains and the highest percentage of the area occupied by intermetallic particles among the experimental sheet, which is consistent with the results obtained for the experimental cast slabs.

Figure 12 shows TEM images and the EDS analysis of the experimental sheets subjected to the stress-relief annealing at 175 °C for 8 h. From Figure 12a, it can be seen that some spheroidal nanoparticles with a size of ~100 nm were found in the ZSE000 sheets after the stress-relieving annealing. These particles were identified as being in the Al₆Mn phase according to the EDS analysis and previous studies [35]. In addition to the Al₆Mn nanoparticles, the synergistic addition of Zr and Sc promotes the precipitation of other finer nanoparticles (~10 nm) in the sheets. The EDS analysis suggests that these finer nanoparticles are likely to be in the Al₃(Sc,Zr) phase. Furthermore, following the fractional substitution of Er for Sc, the Al₃(Sc,Zr) nano-particles were modified into finer Al₃(Sc,Zr,Er) nano-particles [36].

3.2. Microhardness

Figure 13 describes the correlation between the microhardness and the intermediate annealing temperature of the experimental sheets treated with stress-relief annealing at 175 °C for 8 h. As illustrated in Figure 13, the peak microhardness value for each distinct micro-alloyed sheet is attained at an intermediate annealing temperature of 400 °C. It is further demonstrated that the optimal intermediate annealing temperature for the experimental sheets is 400 °C, which complements the above microstructural observation. For sheets intermediately annealed at 400 °C, the combined addition of Zr and Sc increases the microhardness of AA5083 sheets from the 142.9 HV to 155.2 HV, following stress-relief annealing. When fractional Sc is substituted by Er (ZSE321), the microhardness value further increases to 159.4 HV.

3.3. Tensile Properties

The typical engineering stress–strain curves and tensile properties of the cold-rolled sheets with/without the stress-relief annealing at 175 °C for 8 h are shown in Figure 14. From Figure 14a, it can be seen that, prior to stress-relief annealing, the ZSE000 cold-rolled sheet exhibited an ultimate tensile strength (UTS) of 526 MPa, a 0.2% proof stress (or 2.0% offset yield strength) of 473 MPa and an elongation to failure (EF) of 5.1%. When 0.3% Zr and 0.3% Sc were added, the strength of the sheet increased with a slight reduction in the EF. The partial substitution of Sc with Er leads to a further enhancement of the ultimate tensile strength (UTS) and 2.0% proof stress (PS), reaching 577 MPa and 539 MPa, respectively, without compromising the ductility. Additionally, stress-relief annealing at 175 °C for 8 h results in a reduction in the strength of all the cold-rolled sheets while significantly improving their ductility, as is evident from Figure 14b. Compared with the ZSE000 sheet, the ZSE330 sheet shows an enhanced strength alongside a slight reduction in ductility. The strength is further increased without reducing the ductility because of the fractional substitution of Sc with Er. Some quality indexes [37,38,39] and modified indexes [40,41,42] are usually used to evaluate the synergy of the strength and ductility of aluminum alloys to balance the common inverse relationship between the strength and ductility of metal materials. In this work, the following modified quality index was selected, based on the above published quality indexes: $Q=UTS+0.6UTS\cdot \log (EF)$. Hence, the Q value of the ZSE321 sheet following the stress-relief annealing can be calculated to be ~828.8 MPa, which is 9.5% higher than that of the ZSE000 (AA5083) sheet. The improvement in the tensile properties is attributed to grain refinement and the precipitation of the Al₃(Sc,Zr,Er) nano-sized particles. At the end of the stretching process, some curves showed stretching serrations, which is due to the PLC effect. In addition, due to the dense stress sawtooth oscillation and its location below the envelope, it is speculated that the PLC type is a C-type sawtooth.

3.4. Fractography

Figure 15 exhibits the tensile fractographs of the cold-rolled sheets with/without stress-relief annealing at 175 °C for 8 h. The fracture surfaces of the tensile samples from all the cold-rolled sheets without stress-relief annealing show a distinct brittle fracture nature, and the quasi-cleavage feature can be observed, as shown in Figure 15a–c, leading to a lower EF value. Conversely, following stress-relief annealing, the cold-rolled sheets display a mixed morphology comprising both quasi-cleavage and dimple structures, with dimple sizes of approximately 10 μm (Figure 15d–f). This indicates that the stress-relief annealing significantly increases the EF value of the cold-rolled sheets. Compared with the ZSE000 sheet, although the dimple morphology on the fracture surface of the ZSE330 sheet remains relatively unchanged, a slight reduction in the number of dimples accompanied by a minor increase in the quasi-cleavage is observed in Figure 15e. As can be seen in Figure 15f, the fractional substitution of Er for Sc in the ZSE330 sheet introduces few changes in the fracture surface microstructure of ZSE330 sheet, suggesting that the EF value of the ZSE321 sheet remains equivalent to that of the ZSE330 sheet.

3.5. Electrochemical Corrosion Behavior

Figure 16 depicts the polarization curves of the cold-rolled sheets with/without stress-relief annealing at 175 °C for 8 h. The electrochemical corrosion parameters presented in

Table 2. Corrosion parameters obtained from potentiodynamic polarization curves.

Samples	Ecorr (V)	Icorr (A·cm−2)
ZSE000-175 °C × 8 h	−0.874	3.12 × 10−6
ZSE330-175 °C × 8 h	−0.852	2.39 × 10−6
ZSE321-175 °C × 8 h	−0.833	1.51 × 10−6
ZSE000-400 °C × 6 h–175 °C × 8 h	−0.867	2.94 × 10−6
ZSE330-400 °C × 6 h–175 °C × 8 h	−0.838	1.85 × 10−6
ZSE321-400 °C × 6 h–175 °C × 8 h	−0.816	1.07 × 10−6

, were determined through Tafel extrapolation. The corrosion potential (E_corr) usually signifies the tendency of the sheets to corrode, and the corrosion current density (I_corr) characterizes the corrosion rate [43]. As can be seen, the Tafel plot’s shape reveals that the corrosion reaction of the experimental sheets remains consistent with or without the stress-relief annealing. From Table 2, it can be seen that the E_corr values of the cold-rolled sheets have shifted positively after the stress-relief annealing. The positive-shifted E_corr implies that stress-relief annealing reduces the kinetics of the cathodic area and the self-corrosion hydrogen evolution reaction. After the stress-relief annealing, the synergetic addition Zr and Sc induced a positive shift on E_corr value of the AA5083 sheet from −0.867 V(SCE) to −0.838 V(SCE), and the fractional Er substitution for Sc further increases the positive shift in the E_corr. Similarly, stress-relief annealing decreases the I_corr of the cold-rolled sheets, which indicates that their corrosion rates are reduced because of the stress-relief annealing. After the stress-relief annealing, the lowest I_corr value (1.07 × 10⁻⁶ A·cm⁻²) can be found in ZSE321 sheet, which implies the lowest corrosion rates were related to ZSE321 sheet with the stress-relief annealing.

Figure 16b illustrates the electrochemical impedance spectra curves of the cold-rolled sheets with/without the stress-relief annealing. From the Nyquist plots, it can be seen that the sheets processed using the stress-relief annealing have bigger capacitive loops than the cold-rolled sheets. For the sheets that underwent the stress-relief annealing, the ZSE330 sheets show a bigger capacitive loop compared to the ZSE000 sheet. The fractional Er substitution for Sc causes an expansion on the capacitive loop in ZSE330 sheet. The results indicate that the ZSE321 sheet treated with the stress-relief annealing presents the best corrosion resistance, corroborating the findings observed in the polarization curves. Generally, the grain boundary energy of HAGBs is higher than that of LAGBs [44]. This is due to the HAGBs, in which recrystallization occurs, having a high amount of energy and easily forming a continuous distribution of nano-sized precipitated particles, while the LAGBs, in which recrystallization does not occur, have a lower amount of energy and the number of nano-sized particles precipitated at the grain boundary is less, which can effectively prevent the continued expansion of grain boundary corrosion [45,46]. In addition, the Al₃(Sc,Zr,Er) phase exhibits superior electrochemical compatibility with the aluminum matrix, contributing to the enhanced corrosion resistance of the ZSE321 sheet. Consequently, the ZSE321 sheet outperforms the ZSE000 sheet in terms of corrosion resistance, making it the optimal choice.

The surface corrosion morphologies of the sheets with the stress-relief annealing immersed in the 5.0 wt.% HCl solution were observed using SEM, and the results are given in Figure 17. It can be seen from Figure 17 that the surface corrosion morphologies of all the experimental sheets show severe exfoliation along the rolling direction. The corrosion is constantly aggravated with the increase in the duration immersion time. For all the experimental sheets that underwent stress-relief annealing, fewer and shallower corrosion pits can be found on the corrosion morphologies of the ZSE321 sheet, which implies the lowest degree of corrosion is found for the ZSE321 sheet.

The results of the hydrochloric acid mass loss test (HAMLT) for the sheets that underwent the stress-relief annealing are provided in Figure 18. In Figure 18, it is clearly demonstrated that the corrosion rate consistently escalates with an increase in the immersion duration. The ZSE321 sheet exhibits the lowest corrosion rate among all the experimental sheets tested, indicating that it possesses superior corrosion resistance compared to the others.

As shown in Figure 19, the radar chart evaluates the tensile properties and corrosion resistance performance of the AA5083 modified with Zr, Sc and Er, based on five performance metrics: UTS (MPa), 0.2% PS (MPa), EF (%), I_corr (A·cm⁻²), E_corr (V-SCE). A shift of five data points toward the outer boundary of the radar chart reveals an improvement in the performance metrics. The ZSE321 sheet exhibits more enhanced strength, the best ductility, the best corrosion resistance and the lowest corrosion rate.

4. Conclusions

In summary, the influence of substituting Sc with Er on the microstructure, mechanical properties and corrosion resistance of Al-Mg-Mn-Sc-Zr sheets processed using a short process (including a simulated twin-belt continuous casting, subsequent direct rolling, intermediate annealing, cold rolling and stress-relief annealing) was systematically investigated in this work. Compared with the standard AA5083 (ZSE000) sheet, the addition of 0.3 wt.% Sc and 0.3wt.% Zr (ZSE330) caused a significant amount of grain refinement in the short-processed AA5083 sheet. When 0.1 wt.% Sc was substituted with Er (ZSE321), the grains in the short-processed sheet were further refined. The intermediate annealing induced between the direct rolling and the cold rolling inhibited the recrystallization and promoted the precipitation of Al₃(Sc,Zr,Er) nanoparticles. Hence, optimization of the tensile properties (UTS~541 MPa, 0.2% PS~469 MPa and EF~7.7%) of the short-processed ZSE321 sheet were achieved. Furthermore, the short-processed ZSE321 sheet exhibits the best corrosion resistance (E_corr~−0.816 V(SCE), I_corr~1.0710⁻⁶ A·cm⁻²) among the experimental sheets; this is mainly because it contains many low-angle grain boundaries and there is better electrochemical compatibility between the Al₃(Sc,Zr,Er) phase and the aluminum matrix.