On the microstructure, texture and electrochemical properties of severely deformed and artificially aged lightweight AA2050 Al–Cu–Li alloy

Lightweight 3rd generation Al–Cu–Li alloy AA2050 was severely deformed via Multi Axial Forging (MAF) at 170 °C followed by artificial aging at 150 °C. Effect of MAF and post MAF aging on microstructure and precipitation was investigated using transmission electron microscopy (TEM). Formation of deformation bands and large dislocation cells were evident on MAF processed samples. Post MAF peak aging resulted in the distribution of fine T1 precipitates in grain interiors with reduction in grain boundary precipitation. Bulk texture studies reveal the formation of strong Goss and S texture components upon MAF processing. Further, low temperature artificial aging doesn’t exhibit significant changes in texture characteristics, either in terms of texture intensities or texture components. Polarization studies showed that peak aged samples exhibited better corrosion resistance compared to un-aged samples. Overall, 12 pass MAF processed and peak aged samples showed higher corrosion resistance. Further, corrosion surface morphologies examination revealed the change in corrosion mechanisms with thermomechanical treatments. Improved corrosion resistance after MAF and peak aging widens the requirements in aerospace and aircraft applications for such engineering alloys.


Introduction
Development of advanced materials for aerospace and aircraft applications for improving the structural efficiency of aircraft remains never-ending.Aluminium and its alloys with lower density, higher specific strengths, stiffness and better corrosion resistance are extensively used in aerospace and aircraft structural applications [1].Furthermore, Aluminum-Copper-Lithium (Al-Cu-Li) alloys have outperformed conventional 2XXX and 7XXX series Al alloys for aircraft structural applications due to their reduced density, greater strengths, fracture toughness, and fatigue crack growth resistance.With maximum solid solubility of 4.2 wt% Li in Al, for every 1 wt% addition of Li, density reduction by 3% and increase in elastic modulus upto 6% can be achieved [2].Al-Cu-Li alloys evolved in three generations and 3rd generation Al-Cu-Li alloys developed with Li less than 2 wt% have gained more research interest.In recent years, objective of development has been to further strengthen and improve the properties of the alloy, and two approaches were predominantly employed.Primarily by micro alloying to improve the material's characteristics properties.For example, Zhang et al [3] found that adding 0.2 wt% Sc to Al-Cu-Li refined the grain structure and aided in the precipitation of plate-like S/S′ phase during aging, both of which led to materials with outstanding mechanical properties.Similar outcomes of grain refinement and enhanced precipitation kinetics on addition of 0.025 to 0.25 wt% Sc to AA2195 alloy was reported by Suresh et al [4].Secondly, by including suitable thermo-mechanical processing routes further into the process-flow, improved mechanical properties with simultaneously control over inherent anisotropic properties can be achieved.
Most commonly, various researches have been reported on modifying the alloys properties by adopting various thermo-mechanical processing routes.Natural and artificial aging, isothermal and non-isothermal aging, multi-stage or dual-step aging, retrogression and re-aging are the most significant thermal processing techniques adopted in age hardenable Al-Cu-Li alloys.By long-term low temperature aging of low Li containing Al-Cu-Li alloys, a fine combination of δ′, θ′ and T 1 precipitates were achieved by Chen et al [5].Lin et al [6] reported that dual-step aging could promote denser and finer precipitation of δ′ and T 1 phases in Al-Cu-Li alloys.Chen et al [7] reported similar findings of denser, finer and uniform distribution of T 1 precipitates after two-step aging.In addition, thermo-mechanical processing techniques such as pre-straining prior to aging, stress aging (creep aging) [8] and SPD processing prior to aging were also reported.Wang et al [9] reported an enhancement in yield strength due to increases in number density of T 1 precipitates and decreases in their size after pre-strain prior to aging of Al-Cu-Li alloy.Rod like (Al 2 CuLi) T 1 phases/precipitates are the major strengthening phases in 3rd gen Al-Cu-Li alloys with p6/mm symmetry forming on {111} planes of Al matrix [10,11].T-phase plays a crucial role in dispersion strengthening by acting as the barriers to dislocation movements, promotes homogeneous deformation, and significantly reduce the grain size [12].Further, T 1 phases were found to improve fatigue crack resistance in Al alloys [13,14].Previous studies have shown that T 1 precipitates in Al-Cu-Li alloys tend to nucleate at crystallographic defects such as grain boundaries, vacancies, and dislocations.Several researchers investigated on introducing dislocations prior to aging, to promote nucleation of large number of T 1 precipitates.Shanmugasundaram et al reported enhanced aging kinetics with more uniform distribution of strengthening precipitates upon aging of cryorolled 2219 Al alloy [15].ECAP of Al-Zn-Mg-Zr prior to aging carried out by Gubicza et al showed significant effect on final shape and size of the precipitates [16].
Among various metal forming processes, severe plastic deformation (SPD) has gained research interest in the past two decades for its ability to produce ultra-fine grain structures, thereby improving material properties [17][18][19][20].Furthermore, SPD processing has been found to be more effective in age-hardenable Al alloys [21].By applying significant plastic strains at relatively low temperatures, with little to no dimensional change, SPD induces microstructural modification in a material by introducing a large number of crystalline imperfections [22,23].Among the SPD techniques developed to date, Multi Axial Forging (MAF) has proven to be a simple yet effective SPD technique that can be implemented on existing forging setups [24].Additionally, MAF can be scaled up to industrial levels, allowing for the processing of large bulk samples, making it viable for commercial production.
Compared to other SPD processes, Multi Axial Forging (MAF) is unique in the strain path induced, as the sample is continuous rotated after every pass.However, the evolution of textures during MAF processing is often compared to those of rolling textures, as both processes are based on plain strain condition [25].Research has shown that MAF of AA7075 Al alloy at 200 °C results in formation of Goss and cube texture components from the initial copper texture component [26].Additionally, Moghanaki et al reported the formation of Copper and Brass texture components during MAF processing of AA2024 alloy [27].A classical study by Zhou et al [28] on the stability of textures in FCC materials states that during plastic deformation, the grains rotate and tend to reach a stable end position.This is further supported by various research studies considering microstructural studies showing the variation in grain misorientation upon plastic deformation [29,30].Also, Suresh et al [31] reported that modification of texture components occurs due to destabilization during MAF of AA2195 alloy.
Numerous investigators have demonstrated that Al alloys exhibit tendency towards an increase in intergranular corrosion (IGC) susceptibility during aging heat treatments.Given that IGC is not so desirable in engineering applications of Al alloys, research into IGC resistance has been a primary focus in recent years.Factors that greatly influence the corrosion resistance of age-hardenable Al alloys include the shape, size, and distribution of precipitates within the grain interiors, the continuity of precipitates at grain boundaries and the width of precipitate-free zones (PFZ) [32].Various studies have reported on the influence of heat treatments on IGC behavior in Al-Cu-Li alloys.For example, Ma et al [33] observed localized corrosion in AA 2099-T83 alloy resulting from the selective dissolution of grain/subgrain boundary T 1 phase.Additionally, Zou et al [34] investigated the impact of artificial aging on the corrosion behavior of AA2198 Al-Cu-Li alloy and found a decrease in corrosion resistance from the solution-annealed to peak-aged condition.Re-solution and re-aging treatment of AA2196 Al-Cu-Li alloy resulted in significantly improved IGC resistance, with a change in corrosion mode from pitting to IGC [35].De Sousa et al [36] conducted a comparative study of T3, T8, and T851 thermomechanical heat treatments on the corrosion resistance of AA2198 alloy.T3 heat-treated samples exhibited trenching and shallow pits due to galvanic coupling between cathodic precipitates and the anodic Al matrix.T8 and T851 samples displayed severe localized corrosion resulting from intragranular attack of highly strained grains.
Literature suggests that SPD processing prior to aging of Al-Cu-Li alloys can significantly improve their mechanical properties.Our previous investigation has also examined the effect of MAF on mechanical and microstructural characteristics [37].While a significant amount of literature is available on MAF processing of FCC materials, only few studies have reported on evolution of texture.Several researchers have investigated the influence of precipitates on the corrosion resistance of Al-Cu-Li alloys in various heat-treated conditions.Nevertheless, to the best of current knowledge, no studies have explored the effect of MAF and post MAF aging on texture and corrosion behavior of the 3rd generation AA2050 Al-Cu-Li alloy.In order to address these gaps, an investigation was undertaken to examine the effect of MAF processing prior to aging on microstructure and crystallographic texture evolution using TEM analysis and XRD bulk texture analysis.Additionally, corrosion tests were conducted to investigate corrosion behavior in Cl -containing environment and the results were correlated with the microstructure.

Materials and experimentation
3rd generation Al-Cu-Li alloy AA2050, supplied in the form of 50 mm thick hot rolled slabs were used in this study.The elemental composition of the alloy, as determined by inductively coupled plasma optical emission spectroscopy (ICP-OES), is presented in table 1 and falls within the range specified by the aluminum association (AA).Several rectangular blocks measuring 24 × 30 × 30 mm were machined for MAF.Blocks were solution heat treated (SHT) at 520 °C for 1.5 h and immediately water quenched prior to MAF.Specially designed die setup made of hardened die steel equipped with heating control setup was used to carry out MAF at 170 °C.A schematic representation of MAF process, illustrating the sample and die interaction during each pass, is shown in figure 1.Several samples underwent 6 and 12 MAF passes, resulting in a cumulative strain of ∼1.6 and ∼3.2 respectively.After MAF, samples for various characterizations were extracted from the near mid plane perpendicular to previous forging axis.Additionally, several sets of samples from the SHT, 6 and 12 pass conditions were artificially peak aged for 40 h at 150 °C.
Standard sample preparation technique was followed for TEM sample preparation.Initially, mechanical polishing was performed using silicon carbide emery papers, until a thickness of 300 microns was achieved.Subsequently, 3 mm discs were punched out from the mechanically polished sheets.These discs were further polished using a disc grinder to reduce the thickness to approximately 80 microns, followed by dimple grinding, resulting in further thinning.To achieve electron transparency necessary for TEM examination, the specimens were thinned using twin-jet electro polishing in an electrolyte composed of 15% nitric acid and 85% methanol.Samples were observed under a Zeiss EM912 TEM apparatus operating at 200 kV.For Corrosion surface morphologies and cross-sections examination, Jeol SEM apparatus is used.Cross-sections of the corroded samples were carefully extracted by machined using the metallurgical sample cutting saw for cross-section observations.For XRD macro texture analysis, 1 cm 3 samples were machined using the metallurgical sample cutting saw, followed by electropolishing to obtain the mirror like strain free surface.To measure the XRD macro texture, the commonly used Schulz Method of texture measurement was adopted [38].This method involves stereographic projection of pole densities in the form of intensities obtained from x-ray diffractions at a particular hkl plane.This stereographic projection also known as pole figures depicts the position of pole densities of that crystallographic plane with respect to the sample reference axis.Stereographic projections, also known as pole figures, depicts the position of pole densities of that crystallographic plane with respect to the sample reference axis.Diffraction data was obtained using Cobalt K α x-ray source for three different hkl planes using PAN Analytical EMPYREAN diffractometer.The cobalt source was selected to eliminate the possibility of fluorescence resulting from the sample.The sample was tilted up to 75°with standard step sizes of 5°and for every tilt, the sample was rotated from 0°to 360°.To compensate for the effect of defocusing error, the maximum tilt was limited to 75°.Furthermore, bulk texture measurements were analyzed on ATEX software.
All potentiodynamic polarization studies were performed on mirror-polished 1 cm 2 area samples extracted from the mid-sections of MAF processed and peak aged blocks.Open-circuit potential (OCP) and potentiodynamic polarization measurements were carried out as a function of immersion time in a 3.5 wt% NaCl aqueous solution using a Solartron ECI 1286 potentiostat workstation.A standard three-electrode electrochemical cell setup was employed, with the Al alloy sample as the working electrode (WE), Ag/AgCl as the reference electrode (RE), and a platinum wire as the counter electrode (CE).Samples were immersed in the solution for 5 min prior to OCP measurements to ensure stability in potentials.OCP measurements were conducted for 1000 s between RE and WE with no current transfer to CE. Immediately following the OCP measurements, polarization measurements were taken within potential drift of ±250 mV with respect to the OCP.For IGC tests, the samples were exposed to electrolyte for 10 h, and longitudinal to longitudinal transverse (L-LT) surfaces were extracted and mechanically polished, followed by diamond paste polishing.Further, polished surfaces were examined using SEM.

Effect of artificial aging heat treatment on microstructures
The effect of peak aging on microstructure and precipitation behaviour of AA2050 Al-Cu-Li alloy was investigated for specimens subjected to SHT, 6 pass, and 12 pass MAF processing.Aging of Al-Cu-Li alloys results in evolution of various precipitates depending on the initial processed conditions and aging parameters.Based upon the previous studies, to identify the complex precipitates evolved, the schematic representation of the typical diffraction spots and streaks of selected area electron diffraction (SAED) patterns is provided in figure 4.This schematic depicts the major precipitates (T 1 , θ′, δ′/β′) observed in 3rd generation Al-Cu-Li alloys in their main zone axes.The bright field transmission electron micrograph of the SHT and peak-aged samples reveal strain-free grains with continuous intergranular phases at the grain boundaries (GBs) as depicted in figure 3(a).Typically, in polycrystalline materials, GBs serve as heterogeneous nucleation sites for secondary phases/precipitates due to the high density of dislocations and vacancies.In Al-Cu-Li alloys, Cu atoms diffuse into GBs, leading to the precipitation of continuous Cu-rich phases.Moreover, the grain interiors exhibit substantial precipitation of rod-like T 1 phases as shown in figure 3(b).SAED patterns of SHT + peak aged sample confirms the formation of T 1 precipitates upon aging.The excessive precipitation of Cu-rich phases at GBs and grain interiors causes Cu depletion regions, resulting in the formation of precipitatefree zones (PFZ) adjacent to GBs, as evident in figure 3(a).
Bright field TEM images of 6 pass and 12 pass peak-aged samples that exhibit dislocation cells resulting from large strains induced during MAF processing is shown in figures 3(d) and (g).As depicted in figures 4(e), (h), peak-aged 6 pass and 12 pass samples exhibit a high density of fine, rod-shaped T 1 precipitates within the grain interiors.In Al-Cu-Li alloys, crystallographic defects such as dislocations, vacancies, and GBs act as preferential nucleation sites for T 1 precipitates.The large dislocations induced during the MAF serve as the diffusion paths for solute atoms to nucleate T 1 phases.Thus, denser T 1 phases nucleate and grow within grain interiors in MAF processed and peak-aged samples compared to conventionally aged samples.The substantial precipitation of T 1 phases inside the grain interiors after peak aging leads to Cu depletion in the matrix.Additionally, the large dislocation pileup caused by MAF impedes the diffusion of Cu atoms towards GBs, thus hindering GB precipitation.

XRD macro texture analysis
Schematic representation of typical orientation and ideal texture components of a FCC material in {111} and {200} pole figures are as shown in figure 5.In Schulz method of XRD measurements [41], background intensities vary with increase in tilt angle, thus background error creeps in.To overcome the effect of background intensities on actual peak intensities, background correction is made by measuring the background intensities separately along the 2θ angle and integrating it over the j.Thus, the corrected pole figure intensity is obtained from the equation (1) [42].
Where, I corrected is the corrected pole figure intensity, I measured (Ø,j) is the experimentally measured intensity and I background (Ø) is the background intensity measured separately.200) and (220) hkl planes respectively.It is apparent that there is a significant difference in texture characteristics between the SHT and MAF processed samples.Samples that underwent 6 pass MAF processing exhibit a strong S and Goss texture.Furthermore, MAF processing up to 12 pass results in strengthening of Goss texture component.This change in texture characteristics is primarily due to the change in the reference plane of deformation between each individual pass during MAF.Similar texture modifications were reported by Gurao et al in samples that underwent crossed rolling, which involves a rotation of the deformation direction along the normal direction.200) and (220) hkl planes, respectively.The texture characteristics of SHT + peak-aged samples display a slight variation compared to un-aged sample conditions, as observed in figure 7(a).Furthermore, figures 7(b) and (c) which depict the pole figures of peak-aged 6 pass and 12 pass MAF samples, do not exhibit any significant changes in texture characteristics, either in terms of texture intensities or texture components.This is due to the low aging temperature, which does not promote recrystallization of the alloy, resulting in minimal microstructural modification.

Corrosion studies 3.3.1. Open circuit potential (OCP) of AA2050
Figure 8 illustrates the variation of open circuit potential (OCP) in 3.5% NaCl solution and corresponding microhardness values with respect to aging time for various processing conditions of AA2050 Al alloy.OCP represents the corrosion potential of the material in environmental conditions without the influence of external potential or current.The overall OCP values (E ocp ) varied within the range of − 0.75 ± 0.06 mV and are in good agreement with the E ocp values reported for various Al-Cu-Li alloys [34] and are relatively positive in comparison to other Al alloys, indicating good corrosion resistance.As shown in figure 8, 6th and 12th pass MAF processed samples exhibit higher hardness compared to the SHT sample.Additionally, the OCP of 6 and 12 pass samples is more noble than the SHT sample, indicating stable oxide film formation during the study period.However, during aging, with an increase in aging time, the hardness further increases and reaches a peak value.The peak hardness conditions were found to be 162 and 171 VHN for 6 and 12 pass samples at 20 h and 80 VHN for SHT at 40 h of aging time, respectively.It's worth noting that, for all processed conditions, upon artificial aging, the potentials shift towards more negative values.

Potentiodynamic polarization studies
Corrosion susceptibility of AA2050 Al-Cu-Li alloy under different processing conditions were evaluated using polarization studies.As depicted in figure 9(a), the initial anodic curves of the 6 pass and 12 pass MAF processed samples exhibit minimal changes in current density with increasing applied potential, indicating the initial passivation of the alloy.However, after the applied potential surpasses ∼700 mV versus SCE, the MAF processed samples display an increase in current density with increasing applied potential, indicating de-passivation due to dissolution of the alloy [43].Potential at which de-passivation occurs is referred to as the breakaway potential (E b ) and was found to be approximately the same for both 6 pass and 12 pass samples.In contrast, the anodic   curve of the SHT sample exhibits early de-passivation, indicating that the oxide film on the SHT sample is less stable compared to the MAF processed samples.Figure 9(b) illustrates the anodic polarization curves at peak aged conditions of SHT, 6 pass and 12 pass MAF processed samples.In contrast to non-aged samples, peak aged samples exhibit a greater change in current density with minimal increase in applied potential and no signs of anodic passivation.However, the corrosion potentials shift to more positive values after peak aging in comparison to non-aged samples of all processed conditions.Jiang et al also reported similar trends of corrosion potential becoming more noble with increasing aging time in Al-Cu-Li alloys [44].Tafel analysis was performed to determine the current densities (I corr ), corrosion potential (E corr ), and is reported in table 2. SHT and MAF processed samples displayed lower resistance to corrosion due to the presence of large crystalline defects.However, peak aged samples of all processed conditions exhibited improved corrosion resistance compared to un-aged sample conditions, with the 12 pass and peak aged sample showing the highest resistance to corrosion.This is likely due to the formation of large grain/subgrain boundaries, fine distribution of intragranular precipitates, and the prevention of GB precipitates as a result of induced strains during MAF processing.

Corrosion surface morphology
Corroded surface morphologies of AA2050 alloy in various processed conditions are examined.Corroded surface of SHT specimen (figures 10(a) and (b)) is similar to that of a nearly flat surface without the formation of pits.The exfoliation of scales formed with corrosion particles embedded on the surface is apparent.In contrast, as seen in figures 10(c) and (d), the corroded surface morphology of the 6th pass MAF processed sample exhibits a reduction in corrosion products and a heavier corrosion film.It is apparent from figure 10(d) that the heavier film has been peeled off, revealing the flat internal matrix surface.These observations align with the Tafel plots (figure 9(a)), which indicate that de-passivation occurred much earlier in the SHT sample compared to the MAF processed samples.In 12th pass sample, in addition to the exfoliation of the corrosion film, deepest corrosion ditches are also apparent as seen in figures 10(e) and (f).Compared to each other, the SHT condition sample displayed flatter corrosion surfaces, indicating uniform corrosion.Previous studies by Zou et al [34] and Wen et al [45] found that Al-Cu-Li alloy shows improved corrosion resistance when solution heat-treated, due to absence of intermetallic phases/precipitates.With an increase in strain due to MAF processing, the corrosion resistance is reduced, and it becomes the least in the 12th pass samples, forming shallow and wider corrosion ditches.
Figure 11 illustrates SEM morphologies of the corroded surfaces of SHT, 6th pass, and 12th pass samples in peak aged condition.A change in corrosion susceptibility and morphology after aging heat treatment was noticeable.The SHT + peak aged samples exhibit pits on the corroded surface, indicative of severe localized corrosion (SLC).The size of the pits was observed to be larger in the 6th pass + peak aged samples.An additional increase in the width of the pits was observed in the 12th pass + peak aged samples.
Figure 12(a) illustrates the corrosion morphology of SHT specimen along the L-LT planes following a 10-h exposure to a 3.5% NaCl solution.Corrosion anomalies observed along the L-LT planes of the SHT specimen were predominantly located within the grain interiors with a homogeneous distribution resembling uniform, less detrimental intragranular corrosion.6 pass and 12 pass MAF processed samples also displayed similar susceptibility to intragranular corrosion as depicted in figures 12(b) and (c).However, in MAF treated specimens, the corrosion areas were discontinuous, with wider and deeper corrosion depths indicative of pitting corrosion.
As depicted in figure 12(d), in peak aged sample after SHT, the corrosion mode was primarily IGC and the observed results were in concordance with TEM results.The corrosion potential of T 1 precipitates were found to be more negative in comparison to the matrix, with a value of −1.076 V/SCE for T 1 precipitates and −0.855 V/ SCE for the matrix in a 4% NaCl solution [46].Furthermore, as evidenced in figure 3(a), SHT + peak aged specimens exhibited large T 1 precipitates at grain boundaries with PFZ formation.This resulted in a shift in GB potential to a more negative value, leading to galvanic coupling between the precipitates and the adjacent matrix, resulting in intergranular corrosion.Further propagation of corrosion damage was observed to be severe.Similar findings of a change in corrosion susceptibility from intragranular to IGC in Al-Cu-Li alloys were previously reported by Jiang et al [44].However, peak aged samples after MAF processing displayed a different mode of corrosion compared to conventionally aged specimens (SHT + peak aged).The 6 pass + peak aged specimen as depicted in figure 12(e), exhibited a corrosion characteristics primarily composed of pitting with only slight IGC.Additionally, it was also evident that the corrosion was not uniform throughout the L-LT surface.Furthermore, the 12 pass + peak aged specimen solely exhibited susceptibility to intragranular corrosion with U-shaped pits as depicted in figure 12(f).As discussed in TEM characterization in section 3.1.2,MAF prior to aging promoted intragranular fine precipitation by consumption of Cu solute atoms in the matrix, resulting in suppression of subgrain/grain boundary precipitates.The presence of a high density of fine T 1 precipitates within the grains and the absence of GB precipitates may result in a shift in the intragranular matrix's corrosion potential towards a more negative value compared to the subgrain/grain boundary.This makes the grain interiors more susceptible to corrosion resulting in intragranular corrosion.

Conclusion
In this study, we present and compare the microstructure, macro texture, and corrosion behavior of MAF processed and artificially aged AA2050 Al-Cu-Li alloy with that of the conventionally heat-treated alloy (Solution heat treatment and artificially peak aging).Notable findings from the study include: 1. MAF processed samples exhibited a high density of dislocations and deformation bands due to the substantial strain induced during MAF processing.SHT + peak aged samples displayed large continuous GB precipitates and T 1 precipitates at the grain interiors with PFZ formation at the grain boundaries.In contrast, Artificial aging of MAF processed samples resulted in a distribution of fine T 1 precipitates in the grain interiors.This fine distribution of precipitates can further significantly influence the mechanical and tribological behaviour of the ally paving way for further studies.
2. MAF processing led to the development of strong S and Goss texture components 6 pass sample, with intensities increasing in 12 pass sample due to the continuous change in compression axis between each MAF pass.Artificially peak aging of MAF processed and SHT samples did not result in any significant textural modifications when compared to their un-aged counterparts.Further investigations can reveal the effect of this textural modifications on the in-plane anisotropy properties that are commonly observed in Al-Cu-Li alloys.
3. Potentiodynamic polarization studies revealed the formation of a corrosion film at initial corrosion potentials and breakdown at higher potentials in SHT and MAF processed samples.However, peak aged samples did not show corrosion film formation.The findings further validate that all sample conditions display improved corrosion resistance after peak aging, establishing them as suitable materials for aerospace structural applications.
4. Corrosion mechanism related to the peak aged samples is attributed to galvanic coupling between the cathodic constituent particles and the anodic matrix.SHT peak aged samples are characterized by intergranular corrosion due to cathodic GB precipitates, while MAF peak aged samples exhibit intragranular corrosion due to the presence of fine, denser T 1 precipitates in the grain interiors.

3. 1 .
Transmission electron microscopy 3.1.1.Effect of MAF on microstructures Bright-field TEM micrograph and selected area electron diffraction (SAED) pattern of SHT sample is shown in the figures 2(a) and (d).Microstructure appears to be free from strain with no indication of dislocation clusters and the corresponding SAED patterns display periodically arranged clear spots indexed to various Al lattice planes.Furthermore, SAED patterns do not show any major spots or streaks between Al matrix spots corresponding to secondary phases.This indicates the complete dissolution of secondary phases into Al matrix leading to the formation of supersaturated solid solution during solution heat treatment[39].The microstructure of 6 and 12 pass samples is characterized by a high density of dislocation forests, revealing the formation of highly strained grains, as seen in figures 2(b) and (c).Additionally, the 12th pass sample exhibits more fine grains that are predominantly surrounded by large dislocation clusters.From figures 2(b) and (c), it is also evident that a significant number of deformation bands were formed upon MAF.No traces of precipitates were observed along the grain boundaries or grain interiors.The SAED patterns display narrow ring-like patterns for the 6 and 12 pass samples (figures 2(e) and (f)), indicating the formation of fine and deformed grains with no spots or streaks corresponding to precipitates.Murayama et al reported similar dissolution of precipitates during equal channel angular pressing of Al-Cu alloy[40].

Figure 1 .
Figure 1.Schematic representation of front view of sample-die interface during MAF processing.

Figure 4 .
Figure 4. Schematic SAED pattern representation of spots and streaks corresponding to various precipitates observed in Al-Cu-Li alloy.

Figure 6
Figure 6 illustrates the complete pole figures of SHT, 6 pass and 12 pass MAF processed sample condition for (111), (200) and (220) hkl planes respectively.As evident in figure 6(a), SHT samples showed strong cube and Goss texture.Figures 6(b) and (c) depicts the pole figures of 6 pass and 12 pass MAF processed samples after SHT for (111), (200) and (220) hkl planes respectively.It is apparent that there is a significant difference in texture characteristics between the SHT and MAF processed samples.Samples that underwent 6 pass MAF processing exhibit a strong S and Goss texture.Furthermore, MAF processing up to 12 pass results in strengthening of Goss texture component.This change in texture characteristics is primarily due to the change in the reference plane of deformation between each individual pass during MAF.Similar texture modifications were reported by Gurao et al in samples that underwent crossed rolling, which involves a rotation of the deformation direction along the normal direction.

Figure 5 .
Figure 5. Schematic representation of typical orientation and ideal texture components of a FCC material in {111} and {200} pole figures.

Figure 7
Figure 7 depict the pole figures at peak-aged conditions of SHT, 6 pass, and 12 pass MAF processed samples for (111), (200) and (220) hkl planes, respectively.The texture characteristics of SHT + peak-aged samples display a slight variation compared to un-aged sample conditions, as observed in figure 7(a).Furthermore, figures 7(b) and (c) which depict the pole figures of peak-aged 6 pass and 12 pass MAF samples, do not exhibit any significant changes in texture characteristics, either in terms of texture intensities or texture components.This is due to the low aging temperature, which does not promote recrystallization of the alloy, resulting in minimal microstructural modification.

Figure 9 .
Figure 9. Polarization curves of AA2050 alloy in different processed conditions (a) before artificial aging, and (b) after peak aging in a 3.5 wt% NaCl solution.

Figure 8 .
Figure 8. Change in hardness and open circuit potential over aging time at 150 °C for various processing conditions of AA 2050 Al-Cu-Li alloy.

Figure 10 .
Figure 10.SEM surface morphologies of corrosion surfaces after immersion in 3.5 wt% NaCl aqueous solution of AA2050 Al-Cu-Li alloy in different processed conditions: (a,b) solution heat treated, (c,d) 6th pass MAF processed, and (e,f) 12th pass MAF processed.

Table 2 .
Electrochemical parameters of AA2050 alloy under different processed conditions derived from polarization curves.