Abstract
The adoption of recycled aluminum alloys is crucial for sustainable lightweight automotive design, yet quantitative assessments of their sheet formability remain limited. This study compares the mechanical properties and formability of primary AA6082–P and recycled AA6082–R alloys under three temper conditions: soft annealed (A), solution-treated W-temper (WT), and naturally aged T4 (3 weeks). Compared to AA6082–P, AA6082–R contains higher levels of impurity elements, particularly Fe (0.30 vs 0.22 wt%) and Cu (0.22 vs 0.0025 wt%), while main alloying elements (Mg, Si, Mn, Cr) remain comparable. Uniaxial tensile tests showed that the transverse direction exhibited the highest strength, with T4 and A conditions increasing yield strength (YS) and ultimate tensile strength (UTS) by ∼1.5× compared to WT. For AA6082–P in the ED orientation, YS increased from 102.5 (WT) to 195 MPa (T4), while UTS rose from 155 to 310 MPa; fracture strains increased ∼10% in T4 and A conditions. AA6082–R displayed similar trends with slightly lower strength but comparable ductility. Square cup drawing revealed the greatest localized thinning particularly in AA6082–P (∼2× higher corner thinning than AA6082–R), while T4 specimens achieved higher limiting drawing depths and more ufniform thickness. Soft annealed samples exhibited moderate formability due to incomplete microstructural evolution. Overall, recycled AA6082–R demonstrates tensile and forming performance comparable to primary AA6082–P across all tempers, supporting its application in lightweight sheet-forming processes without significant performance loss.
Keywords
Introduction
Aluminum alloys are among the most widely used structural materials in automotive, transportation, and building applications due to their favorable specific strength, corrosion resistance, recyclability, and manufacturing flexibility. In particular, 6xxx–series alloys (Al–Mg–Si) are frequently selected for lightweight design of body-in-white structures, closure panels, and structural components that demand a balanced combination of strength, ductility, and formability.1,2 As regulatory and market pressures toward decarbonization intensify, the integration of high-fraction recycled aluminum alloys is increasingly viewed as a critical pathway to reduce the environmental footprint associated with vehicle manufacturing and use.1,3
Despite their sustainability benefits, the broader industrial adoption of recycled aluminum alloys is still constrained by concerns about forming performance, especially in deep drawing and high-strain applications. Compared with primary alloys, recycled materials often exhibit subtle but important variations in alloy chemistry, reflected in elevated Fe and impurity levels due to mixed scrap streams and re-melting processes.4–6 These compositional shifts are known to influence strain hardening behavior, plastic anisotropy, and damage evolution, with direct implications for sheet metal formability.
Numerous studies have evaluated tensile behavior of 6xxx alloys under various conditions, including the influence of heat treatment state, natural aging, and processing history.7–9 However, uniaxial tensile tests alone do not fully capture the complex stress–strain histories encountered in industrial forming operations, and therefore may not reliably predict component-level formability.7,10 Forming limit diagrams (FLDs) have long been established as a more representative tool for assessing sheet metal formability by predicting the onset of localized necking and failure under multiaxial strain states.10–12
Recent advances in AA6082 formability research include investigations into thickness-dependent yielding and formability of T6-tempered sheets, demonstrating that increased thickness influences yield locus shape, anisotropy, and FLD characteristics. 13 Other work has explored warm and hot stamping regimes for AA6082, showing enhanced forming limits at elevated temperatures, and the effects of friction stir processing on cold formability.14,15 Computational modeling efforts have also been employed to predict formability by incorporating realistic yield criteria and anisotropy into forming simulations. 16
Formability research on recycled aluminum alloys has lagged behind that on primary materials, with most prior studies focused on simplified stretch forming tests or tensile properties rather than comprehensive deep drawing evaluation.5,17 A few comparative studies have addressed recycled versus primary 6xxx alloys (e.g. AA6181), showing that scrap content can alter alloy composition without drastically degrading ductility, but the direct link between recycling, heat treatment, and industrial-level formability remains underexplored.4,5
The literature also reveals a gap in systematic comparison of recycled and primary AA6082 alloys across multiple temper conditions under industrially relevant forming tests. While characterization of AA6082-O by uniaxial and cupping tests provides useful baseline data, such studies do not directly address deep drawing drawability, particularly with respect to the effects of recycling and temper.17,18
In this study, the mechanical behavior and formability of recycled and primary AA6082 aluminum alloys are investigated through a combination of conventional tensile testing and square cup deep drawing experiments, which concurrently impose plane strain, biaxial stretching, and bending–unbending deformation modes representative of industrial sheet forming operations. Formability metrics include limiting draw depth, punch force–stroke response, and localized thinning/failure patterns. The materials are examined in three distinct temper states: soft annealed (partially solutionized), W-temper (solution heat treated), and T4 (solution heat treated and naturally aged), enabling a comprehensive assessment of how microstructure and thermal history intertwine with recycling effects.
Accordingly, this study addresses two central questions:
Experiments
Materials
Two Al–Mg–Si alloy (AA6082) materials were investigated: Extruded plate from primary alloy production, hereafter denoted AA6082–P and recycled alloy, denoted AA6082–R, produced through the similar extrusion process. The thickness of the extruded plates is 2 mm. Chemical compositions are summarized in Table 1. Compared to the primary alloy AA6082–P, the recycled alloy AA6082–R contains higher concentrations of impurity elements, particularly Fe (0.30 vs 0.22 wt%) and Cu (0.22 vs 0.0025 wt%), reflecting the unavoidable accumulation of residual elements during recycling, while the main alloying elements (Mg, Si, Mn, and Cr) remain essentially similar. Both alloys fall within the industry specification standard of AA6082 and, thus, support a direct comparison of formability performance of industrial alloys made from recycled (post-consumer scrap) and primary metal.
Chemical composition of the conventional and recycled Al–Mg–Si alloys (wt%).
The extruded plates were received in a soft-annealed condition (partially solutionized, hereafter referred to as condition A) and were subsequently subjected to controlled heat treatments, including solution heat treatment (hereafter W) and natural aging following solution heat treatment (hereafter T4). The applied heat-treatment sequence is illustrated in Figure 1(a). To obtain the W temper, the plates were heated to 565°C, with heating rate approximately 10°C/min using Nabertherm 2943 oven. After holding for 1 h, the samples were water quenched. T4-temper samples were produced by naturally aging the W-tempered material at room temperature for 3 weeks, as shown in Figure 1(a). For the W-temper condition, formability tests were conducted within 15 min after quenching to preserve the supersaturated solid-solution state characteristic of freshly quenched material, consistent with industrial in-line solutionizing practices for aluminum alloys. 19
(a) Heat-treatment procedure and (b) schematic diagram of square cup drawing.
The initial microstructures of AA6082–P and AA6082–R were examined using optical microscopy (OM) and scanning electron microscopy (SEM). Representative OM images of the extruded plates are shown in Figure 2. Both alloys exhibit elongated grains aligned along the extrusion direction, which is characteristic of extruded Al–Mg–Si alloys. The overall grain morphology of the recycled alloy (AA6082–R) is comparable to that of the primary alloy (AA6082–P), indicating that the recycling route and subsequent extrusion process produced a similar deformation texture and grain structure. No visible porosity or casting-related defects were observed in either material, confirming the metallurgical integrity of the recycled alloy.
Microstructural comparison of as received (A) AA6082: (a)–(b) optical microscopy images along the extrusion direction for AA6082–P and AA6082–R (scale bar: 1 mm) and (c)–(d) SEM backscattered electron images AA6082–P and AA6082–R (scale bar: 10 µm).
While the grain morphology appears broadly similar in OM observations, differences become more evident at higher magnification. SEM backscattered electron (BSE) imaging reveals the presence of intermetallic particles distributed throughout the aluminum matrix. In AA6082–P, the particle population is relatively sparse and mainly consists of fine Al–Mg–Si precipitate-related phases together with a limited number of Fe-containing intermetallics. In contrast, AA6082–R exhibits a higher density of coarse, bright particles, which are attributed primarily to Fe- and Cu-containing intermetallic phases such as Al–Fe–Mn–Cu compounds. Quantitative image analysis supports these observations: the area fraction of Al–Fe–Mn–Cu particles increase from approximately 1.25% in AA6082–P to 2.10% in AA6082–R, while the fraction of Al–Mg–Si particles decrease from 1.36% to 0.87%, respectively. The increased population of Fe-rich intermetallic particles in the recycled alloy is consistent with its higher impurity element content reported in Table 1. These particles are expected to influence local deformation behavior during forming by acting as potential strain localization sites, which may contribute to differences in thinning behavior observed in subsequent formability tests.
Tensile test
Tensile tests were conducted with samples cut from the 2 mm thick extruded plates with dimensions following ASTM E-8. Young’s modulus (E), yield stress (
Mechanical properties characterized by tensile tests.
Square cup drawing test
The square cup drawing test was employed to evaluate the formability of the two alloys under various heat treatment conditions. Figure 3 illustrates the testing procedure, including sample fabrication, heat treatment and drawing with square punch. Schematic diagram of experimental setup, which consists of the square punch, die, and blank holder is described in Figure 1(b). Corresponding geometric dimensions are provided alongside the figure.
Overview of experimental process involving sample fabrication, heat treatment, and square cup drawing tests.
Samples were cut to approximately 100 mm × 100 mm. The cutting ring diameter and drawing allowance (die opening) were 80 mm; thus, a blank size of about 100 mm provided a sufficient region for gripping. To ensure consistency between the forming experiments and the quasi-static tensile tests (strain rate of 0.001 s−1), the punch speed was determined using an equivalent strain-rate criterion for square cup forming. Accordingly, the punch speed was set to 0.5 mm/min, corresponding to an overall material deformation rate of approximately 0.001 s−1, which is in line with previous study. 20 A blank-holding pressure of 11 MPa was applied, as established through preliminary trials, to provide adequate clamping and prevent slippage in 2 mm thick AA6xxx samples, thereby promoting a mid-range drawing deformation mode (Appendix 1). Friction conditions were controlled by inserting Teflon film coated with MOLYKOTE between the samples and the punch. 21
Results and discussion
Mechanical properties of extrusion sheets
Figure 4(a) and (b) present the strain-hardening behavior obtained from uniaxial tensile tests of AA6082–P and AA6082–R. The calibrated mechanical properties obtained from the tensile tests in different directions and material conditions are summarized in Table 2. In addition to the strength and ductility parameters, the strain hardening exponent (n-value) and anisotropy coefficient (R-value) were determined to better characterize the material deformation behavior. The n-value describes the strain hardening capability of the material, while the R-value reflects the plastic anisotropy of the sheet. To facilitate comparison of the calibrated mechanical properties presented in Table 2, a graphical representation is provided in Figure 4(c). The figure summarizes the variation of the key parameters across the different material conditions and loading directions. Tensile stress–strain responses were evaluated along three material orientations—the extrusion direction (ED), diagonal direction (DD), and transverse direction (TD)—under the investigated heat-treatment conditions. For both alloys, the TD consistently exhibited higher initial yield strength and ultimate tensile strength than the ED and DD, while fracture elongation showed only limited difference between the three orientations. Compared with the solutionized (W) condition, both the soft annealed (A) condition and the naturally aged for 3 weeks (T4) displayed a significant increase in strength. Across all orientations, the yield and tensile strengths in the T4 and NA states were approximately 1.5 times higher than those measured in the W condition.
Measured tensile test properties of (a) recycled alloy AA6082–R, (b) primary alloy AA6082–P, and (c) comparison view of characterized mechanical parameters in three different orientations, following distinct heat treatments.
In addition to strength enhancement, fracture elongation increased by approximately 10% in the A and T4 conditions, with the improvement being more pronounced in the DD and TD orientations than in the ED orientation. These trends are consistent with previous studies reporting that solution treatment followed by natural aging promotes an increase in the ductility in Al-Mg-Si alloy.22–24
A similar overall tendency to that of the primary alloy was observed for the recycled alloy AA6082–R, as shown in Figure 4(b). The recycled material exhibited strain-hardening behavior and orientation-dependent responses that followed the same qualitative trends as those of AA6082–P across the investigated heat-treatment conditions, indicating a comparable mechanical response to the applied thermal processing. Compared with AA6082–P, AA6082–R showed lower yield and ultimate tensile strengths, most notably along the ED orientation. However, the relative differences were consistent across orientations and heat-treatment states. The fracture elongation of AA6082–R in the T4 and NA conditions was within the same range as that of the primary alloy. Within the scope of the uniaxial tensile tests, these results indicate that the incorporation of recycled feedstock in AA6082–R does not lead to a pronounced degradation in tensile ductility or alter the overall trends in strength evolution for different heat treatment. Consequently, AA6082–R exhibits a tensile mechanical response that is broadly comparable to that of AA6082–P under the conditions investigated in this study.
Square cup drawing results
Square cup deep drawing tests were conducted to evaluate formability as a function of heat-treatment condition for both primary (AA6082–P) and recycled (AA6082–R) alloys. Representative drawn cups are shown in Figure 5, where the loading curve and drawing depth are presented together. The thickness distribution of failed samples, as a key forming metric, is provided in Figure 6.
Square cup drawing results – loading curve and view of samples after failure of (a) as received, (b) WT, (c) NA condition, (d) comparison in loading curve with heat treatments of AA6082 – P, and (e) AA6082 – R.
Thickness distribution and thickness reduction of square cup–drawn AA6082–P and AA6082–R under different heat-treatment conditions: (a) cross-sectional measurement line, (b and c) optical cross sections showing localized thinning at the cup corner (blue circles), and (d–g) measured thickness and calculated thickness reduction profiles, for AA6082-P and AA6082-R each.
As described in Section 2, AA6082–R contains slightly higher levels of residual elements, particularly Fe and Cu, while the main alloying elements remain comparable to AA6082–P.
The higher forming forces observed for the recycled AA6082–R alloy can be explained by the increased population of coarse Fe- and Cu-rich intermetallic particles. These particles are known to impede dislocation motion, which can locally constrain plastic flow, and lead to higher matrix flow stress during forming.25,26 In contrast, the primary AA6082–P alloy, with fewer and finer intermetallics, exhibits lower forming forces and more uniform deformation. These findings, together with microstructure investigation (section 2.1), indicate that microstructural modifications induced by recycling primarily can affect intermetallic particle density and morphology, rather than overall grain structure. Consequently, while grain alignment remains similar between the two alloys, differences in intermetallic populations directly influence forming behavior, including punch force and thinning distribution.
In addition, the observed thinning and strain localization behavior can be directly related to the underlying microstructural features. An increased fraction of coarse intermetallic particles, particularly Fe- and Cu-rich phases, promotes strain localization by acting as stress concentrators and restricting homogeneous plastic deformation.27,28 A higher particle density can therefore accelerate the onset of localized thinning, reducing effective formability. Conversely, a more homogeneous distribution of finer particles supports more uniform strain accommodation and delays necking. 29 This relationship explains the slightly higher forming forces and localized thinning tendencies observed in the recycled alloy, linking microstructural heterogeneity to macroscopic forming response.
Additionally, square cup drawing results reveal that formability is strongly influenced by the material’s thermal history and storage time. Notably, the soft annealed A condition, which is partially solutionized, exhibited reduced formability compared to T4 material. In contrast, fully solutionized and rapidly quenched T4 samples showed improved formability under the same testing conditions, confirming that the degradation observed in the A temper is primarily associated with its incomplete microstructure evolution from partial solutionization.19,30,31
For the solution-treated W-temper condition (W), the microstructure consists predominantly of a supersaturated solid solution with limited obstacles to dislocation motion, resulting in a relatively low work-hardening capacity. Previous studies have linked this microstructural state to early strain localization under multiaxial deformation conditions typical of forming processes.22,23 In the present study, the reduced limiting drawing depth observed for W-temper samples, compared with freshly prepared T4 or short-term naturally aged material, is therefore attributed to the absence of effective strengthening mechanisms that would otherwise delay localized thinning and stabilize plastic deformation.
The T4 aged for 3 weeks at room temperature following solution treatment, is reported to exhibit a moderate increase in flow stress due to the formation of solute clusters and early GP zones, 31 while retaining sufficient strain-hardening capacity to delay localization. This combination of increased strength and preserved strain distribution resulted in higher drawing forces but maintained or slightly improved forming depth relative to the W-temper state, indicating an improved balance between strength and work-hardening capacity that governs formability under complex strain paths. Overall, differences between primary and recycled AA6082 were minor under all conditions investigated, with both alloy variants exhibiting similar forming trends when accounting for storage effects.18,19
Thinning behavior
To investigate the thinning behavior of square cup–drawn AA6082–P and AA6082–R under different heat-treatment conditions, the thickness distribution and corresponding percentage reduction were quantified from cross-sectional specimens after forming. All specimens were extracted from the same extruded plate, and the initial sheet thickness was measured prior to forming using a micrometer, confirming a consistent starting thickness of approximately 2 mm across all samples and conditions. This ensured a uniform initial geometry for all thickness measurements and cross-sectional comparisons. Thickness measurements were performed using optical microscopy in combination with a micrometer along a representative cross section, as illustrated in Figure 6(a). Measurements were taken at 2 mm intervals over an approximately 40 mm long section spanning the cup bottom, corner, and sidewall regions. Figure 6(d) and (f) present the measured thickness distributions for AA6082–P and AA6082–R, respectively, while the corresponding percentage thickness reductions relative to the initial plate thickness are shown in Figure 6(e) and (g).
The results indicate that, for both alloys, the solution-treated W-temper (WT) specimens exhibit the most pronounced thickness reduction, particularly at the cup corners, as highlighted by the blue circles in Figure 6(b) and (c). In contrast, the soft annealed (A) and naturally aged (T4) conditions display more uniform thickness profiles with substantially reduced localized thinning. AA6082–P shows slightly greater variation in thickness reduction across heat-treatment conditions than AA6082–R, indicating a higher sensitivity to temper history. Notably, for the WT condition, AA6082–P exhibits severe localized thinning at the punch corner—approximately twice that observed in AA6082–R—while showing less thinning in the cup center. This behavior suggests early strain localization followed by rapid strain accumulation in the corner region, leading to premature failure. Importantly, this localized thinning occurs at comparable limiting drawing depths for both alloys in the WT condition, as shown in Figure 5(b).
For both materials, the limiting drawing depths achieved in the WT condition are lower than those reached by the T4 conditions, underscoring the reduced formability of WT material under multiaxial loading despite its relatively high uniaxial tensile ductility. This discrepancy between tensile ductility and forming performance is consistent with previous studies,20–22 which have demonstrated that resistance to necking under uniaxial tension does not necessarily translate to improved formability under multiaxial stress states. Prior work attributes this behavior to anisotropic strain evolution during forming, which governs localized thinning and failure.23,24 In contrast, the T4 condition, produced through controlled natural aging, exhibits higher limiting drawing depths and more moderate thinning behavior, indicating sufficient strain-hardening capacity to promote more uniform strain distribution and delay localized thinning compared with the WT condition. Similarly, the soft annealed A tempered materials show less severe localized thinning, although the partially solutionized state degrades its overall ductility relative to the well-controlled T4 condition.
Although a full forming limit diagram (FLD) was not constructed in the present study, the formability can be evaluated by combining the thickness distribution, force–displacement response, and strain hardening behavior. The onset of localized thinning, particularly in the cup corner regions, indicates that deformation approaches a localized necking condition under multiaxial stress states. In this framework, the measured thickness reduction provides an alternative indication of strain localization and the material’s capacity to accommodate plastic deformation in regions prone to necking.32–34
Differences in formability between heat-treatment conditions can also be interpreted in relation to the strain hardening exponent (n-value). Higher n-values correspond to a greater ability to distribute strain and delay localization. From a microstructural perspective, this behavior is strongly influenced by the distribution and fraction of intermetallic particles and precipitates in Al–Mg–Si alloys, as discussed in section 2.1 and 3.1. A higher density of coarse intermetallic particles can promote stress concentration and accelerate localized thinning, while a more uniform precipitate distribution enhances strain hardening capacity and postpones necking.27–29,34 Overall, the present analysis provides an assessment of formability, showing that differences in limiting drawing depth and localized thinning are governed by possible interplay between strain localization and strain hardening capacity, which in turn are linked to microstructural features.
Conclusion
Square cup deep drawing tests were performed to evaluate the formability of primary (AA6082–P) and recycled (AA6082–R) aluminum alloys under different temper conditions. Uniaxial tensile tests showed comparable strength and ductility for both materials; however, multiaxial forming revealed clear differences in localized deformation. Specimens in this condition exhibited early localized thinning, limiting the achievable drawing depth despite relatively high tensile ductility. Soft annealed (A temper) samples demonstrated moderate formability, while fully solutionized and naturally aged T4 samples achieved the highest drawing depths and forming forces.
Microstructural analysis revealed that both alloys have elongated grains aligned along the extrusion direction. The primary AA6082–P contains fine and sparsely distributed Al–Mg–Si precipitates with limited Fe-containing intermetallics, whereas the recycled AA6082–R shows a higher density of coarse Fe- and Cu-rich intermetallic particles. These coarse particles act as rigid obstacles to dislocation motion, locally constraining plastic flow and increasing heterogeneous strain accumulation. This explains the slightly higher forming forces observed in AA6082–R during deep drawing and the differences in localized thinning between the two materials.
From these observations, the key findings are:
These findings provide a mechanistic understanding of the influence of recycling on AA6082 formability, supporting the use of recycled aluminum alloys in lightweight sheet-forming applications.
Footnotes
Appendix 1. Preliminary assessment of blank holding force
A preliminary study was performed to evaluate the influence of blank holding pressure (BHP) on the drawing response and to determine a suitable blank holding force for the main experimental campaign. Blank holding pressures of 2, 11, 15, and 19 MPa were investigated, corresponding to a force range of approximately 0.8–14 kN. The assessment focused on identifying potential slippage at the clamped region and quantifying the sensitivity of the drawing response to variations in BHP.
Slippage was evaluated by drawing fine reference lines at the clamped region prior to forming and inspecting their relative displacement after testing (Appendix Figure 7(b)). No measurable slippage was observed within the investigated BHP range of 2–19 MPa, indicating that sufficient frictional restraint was achieved even at the lowest pressure. However, supplementary trials conducted at higher BHP levels revealed that when the pressure exceeded approximately 23 MPa, fracture initiated near the clamped line, suggesting excessive restriction of material flow and stress concentration at the blank–holder interface.
Appendix Figure 7(b) presents the force–displacement (loading) curves obtained under the four investigated BHP conditions, with displacement plotted along the x-axis and drawing force along the y-axis. The curves are overlaid to facilitate direct comparison. An increase in blank holding pressure resulted in higher peak drawing forces and noticeable changes in the displacement at fracture, reflecting the strong dependence of drawing deformation on the boundary conditions imposed by the blank holder.
The influence of BHP on the forming limits is further summarized in Appendix Figure 7(c), which shows a scatter plot of the maximum displacement and corresponding fracture force as a function of blank holding force. Both quantities exhibit a systematic dependence on BHP, confirming that blank holding force affects the global forming response even in the absence of observable slippage. Based on these findings, a blank holding force of 6 kN, corresponding to a BHP of approximately 11 MPa, was selected for all subsequent experiments. This condition ensured stable gripping without clamp-induced fracture and provided consistent boundary conditions for comparative evaluation of material state and heat-treatment effects.
Acknowledgements
The authors gratefully acknowledge the support from the Norwegian University of Science and Technology (NTNU) and NTNU Aluminum Product Innovation Center (NAPIC). The authors sincerely acknowledge the assistance of the HEXAL project members in providing microstructure information and material support throughout the course of this research.
Handling Editor: Aarthy Esakkiappan
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The work was supported by the TaFF project (Grant No.: 317777) and HEXAL (Grant No.:346433) by the Research Council of Norway.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.