Introduction

Recently, the demand for lightweight aluminum conductors has been growing to meet the energy consumption needs in the electrical industry [1, 2, 3]. Aluminum conductors are divided into two categories: aluminum conductor steel reinforced (ACSR) and all-aluminum alloy conductor (AAAC). In the former (ACSR), commercially pure aluminum alloys such as AA1350 alloy are generally used, and the core consists of the steel to reinforce the conductors in power lines. In recent decades, AAAC has been found as an appropriate candidate in high voltage transmission lines due to it meeting the requirements of having both high mechanical properties and good corrosion resistance. Al–Mg–Si 6xxx alloys, such as AA6101 and AA6201 cables, often serve as conductors in AAAC because their strength can be enhanced via the precipitation hardening, while their electrical conductivity (EC) remains relatively high [4, 5, 6, 7].

Precipitation hardening, work hardening, solute strengthening, and grain boundary strengthening are among the strengthening mechanisms in aluminum alloys. In contrast, the EC is governed by the movement of valence electrons. Any disturbances in the crystal lattice periodicity, which stem from defects in the crystal structure (such as vacancies, dislocations, grain boundaries, and impurity atoms), cause an increase in electrical resistance [8, 9, 10, 11, 12]. Among those defects, dissolved atoms along with Guinier–Preston (GP) zones show the most deleterious impact on EC [9, 10, 13]. It is evident that factors enhancing the strength of aluminum alloys also tend to reduce its EC. As a consequence, the strength and EC are found to be mutually exclusive [8, 9, 10, 14].

Mg and Si are the principal alloying elements in 6xxx alloys, making them age-hardenable [15]. The conventional sequences of precipitation during aging treatment are as follows: clusters of Mg and Si atoms; GP zones; β″; β′; and equilibrium β-Mg2Si [16, 17]. It was found that excessive Si relative to Mg2Si enhanced the precipitation hardening [11, 15, 18, 19], while the additional Mg relative to Mg2Si exhibited a negligible strengthening effect [15, 18]. The excessive Si in 6xxx series alloys leads to an increase in the number density of the β″ strengthening phase [15, 19]. Si columns act as a skeleton for the precipitates, observed in the core of precipitates from GP zones to β″ [20, 21, 22]. Moreover, the higher diffusion of Si atoms leads to the formation of Si/vacancy clusters, which have a higher binding energy than Mg/vacancy clusters, easily evolving into the subsequent strengthening precipitates with high density [15, 23]. Therefore, excessive Si promotes precipitation hardening in the 6xxx series alloys [15, 19].

The effect of excess Si on the electrical and mechanical properties in 6xxx series conductor alloys has recently been taken into consideration [9, 11, 15]. In coarse-grained Al–Mg–Si conductor alloys, it was reported that the alloys with excess Si showed a higher strength [11, 15], of which strengthening mechanisms were not discussed. In addition, in fine-grained Al–Mg–Si conductor alloys, a higher EC was achieved owing to grain boundary precipitates in the alloys with excess Si [11]. Overall, the EC and mechanical properties ought to be simultaneously considered in Al–Mg–Si conductor alloys. The minimum requirements of AA6201 conductor alloys for ultimate tensile strength (UTS), EC, and elongation to failure are 305 MPa, 52.5% IACS, and 3%, respectively [24, 25, 26].

Considering that the Mg/Si ratio of the main strengthening phases is approximately 1 (at.%) in Al–Mg–Si alloys, investigations on the effect of excess Si on the strength and EC are limited. The present work aimed to find the optimum Si level and thermo-mechanical treatments for a balance between mechanical properties and EC. In addition, the underlying mechanisms and microstructures for improving the strength and EC in Al–Mg–Si 6201 alloys were explored.

Results

Mechanical properties and EC

Figures 1(a) and 1(b) show the evolution of the microhardness and UTS, respectively, as a function of aging time. The Si4 and Si5 alloys showed peak hardness after 2 h, while the Si6 and Si9 alloys reached peak hardness after 1 h. The higher the level of Si, the quicker the alloy reaches the peak aging condition. The values of peak hardness for the Si4, Si5, Si6, and Si9 alloys are 114, 119, 127, and 140 HV, and the values of UTS after peak aging are 310, 338, 364, and 394 MPa, respectively. Overall, it is found that greater hardness and UTS were achieved in the alloys with higher excessive Si. Figure 1(c) displays the elongation to failure for four alloys according to the aging time, in which the elongation increases rapidly with aging time, then gradually increases over the aging time. Figure 1(d) demonstrates the evolution of the EC as a function of the aging time. The results show that the EC exhibits a rapid increase before the peak aging, then increases moderately with prolonged aging time. The quick increase in EC at the early stage of aging could be attributed to the formation of precipitates at the expense of a large number of Mg and Si atoms. The alloys with higher Si contents show a lower EC at a given aging time, arising from dissolved Si atoms in the aluminum matrix.

Figure 1:
figure 1

Evolution of (a) microhardness, (b) UTS, (c) elongation to failure, and (d) EC as a function of aging time.

Full size image

In Al–Mg–Si 6xxx alloys, it is generally less problematic to meet the required mechanical properties owing to its strong precipitation strengthening. In the peak aging condition [Fig. 1(b)], the strengths of all four alloys are above the minimum UTS requirement (305 MPa) for 6201 conductor alloys. However, the EC values at the peak aging for the Si4, Si5, Si6, and Si9 alloys are 52.3, 52, 49, and 47.5% IACS, respectively [Fig. 1(d)], which are all below the minimum required EC (52.5% IACS). The challenge is that EC and UTS must be concurrently taken into consideration in the conductor alloys to find an optimal tradeoff between EC and UTS. The minimum required EC can only be reached after 3, 4, 5, and 34 h of aging for the Si4, Si5, Si6, and Si9 alloys, respectively, showing that the alloys with higher Si contents needed a longer aging time to enhance EC. Considering the required minimum elongation to failure (3%) [26], it is apparent that all alloys fulfill this requirement under all aged conditions after peak aging [Fig. 1(c)].

Figure 2(a) displays the UTS as a function of EC for the four alloys in the full range of the aging time studied. Figure 2(b) is an enlarged view of the right corner of Fig. 2(a), where it shows the relation between UTS and EC above the minimum UTS and EC requirements (305 MPa and 52.5% IACS, respectively). It reveals that the maximum UTS levels of the alloys above 52.5% IACS for Si4, Si5, Si6, and Si9 alloys are 306, 322, 346, and 318 MPa after 3, 4, 5, 34 h of the aging, respectively. With respect to the UTS above the minimum required EC, the Si6 alloy after 5 h of aging shows the highest strength, followed by the Si5 alloy after 4 h of aging. As observed in Figs. 1(a) and 1(b), higher peak hardness and strength were attained in the Si-excessive Si9 alloy. However, its EC was the lowest among all alloys and increased only slowly with the aging time [Fig. 1(d)]. The Si9 alloy needs a much longer aging time (at least 34 h) to reach the required EC than the other alloys, where its UTS decreased from 394 MPa at the peak aging to 318 MPa. The Mg-excessive Si4 alloy generally exhibited the lowest strength among four alloys, and it reached the required minimum EC after 3 h of aging, but its UTS was at the bottom line (306 MPa).

Figure 2:
figure 2

(a) UTS versus EC of four alloys and (b) the enlarged view of the right corner of (a) above the minimum UTS and EC requirements (305 MPa and 52.5% IACS).

Full size image

Differential scanning calorimetry analysis

Figure 3(a) shows the differential scanning calorimetry (DSC) results of precipitation for the as-drawn alloys during heating with a heating rate of 10 °C/min, in which the exothermic peak I is ascribed to the precipitation of β′, and peak II is attributed to the precipitation of the equilibrium β phase. The precipitation sequence was disrupted due to the presence of a high number of dislocations from cold wire drawing, acting as heterogeneous nucleation sites for β′ phases [27, 28, 29]. Therefore, the precipitation of β″ was suppressed in all as-drawn alloys. Given that precipitates formed at the peak I are the main strengthening phase in the studied alloys, a close examination of the first exothermic peaks is given in Fig. 3(b). It was determined that the onset temperature of the Si4, Si5, Si6, and Si9 alloys are 213, 213, 212, and 208 °C, and the corresponding peak temperatures are 235, 235, 233, and 230 °C, respectively. The onset and peak temperatures for precipitates shifted to lower temperatures with increasing Si levels. It reveals that the higher the Si content, the higher the precipitation kinetics. In addition, the precipitate amount is proportional to the size (height and width) of the peak. Precipitation peak I became higher and broader with increasing Si levels. Therefore, it is evident that excess Si promotes the higher amount of precipitates, resulting in alloys with higher Si content exhibiting higher peak hardness and UTS, as shown in Figs. 1(a) and 1(b).

Figure 3:
figure 3

(a) DSC curves of the different alloys in the as-drawn condition and (b) the enlarged view of the precipitation peak I.

Full size image

Strain hardening of cold wire drawing

For the strength of final conductor wires, the strain hardening during cold wire drawing is also an important contribution. The strain hardening effect from wire drawing was determined by measuring the UTS and hardness difference before and after cold wire drawing without any natural aging. Yin et al. [30] observed that applying pre-straining following the solution treatment only led to the formation of a large number of dislocations without any apparent precipitates. An increase in UTS from wire drawing for Si4, Si6, and Si9 alloys is 79, 83, and 106 MPa, and an increase in microhardness is 25.5, 28.6, and 33.1 HV, respectively. It is worth mentioning that the aging at 180 °C could not affect the strain hardening contribution, which was proved in our preliminary work by the tensile testing on commercially pure drawn aluminum at different aging times up to 34 h at 180 °C. It can be deduced that the strain hardening effect increases with the Si level. Compared with the other two alloys, the Si-excessive Si9 alloy shows the largest strain hardening contribution.

Microstructures

For the microstructural examination, the Si4 alloy after 3 h, the Si6 alloy after 5 h, and the Si9 alloy after 34 h of aging were selected, representing a compromise between UTS and EC, designated as Si4-3 h, Si6-5 h, and Si9-34 h alloys. Figure 4 shows the grain structures for these three alloys in the drawing direction. The grains are elongated along the drawing direction. The average grain size was determined by measuring over 300 grains, being 150, 130, and 138 μm for Si4-3 h, Si6-5 h, and Si9-34 h alloys, respectively, as shown in Fig. 4(d). Given the fact that the experimental process is the same for all alloys and the aging at 180 °C would barely lead to the recrystallization, it is expected that the grain sizes of three alloys are relatively similar. The large grain size in all the samples could be attributed to the casting without grain refiner.

Figure 4:
figure 4

Grain structures of (a) Si4-3 h, (b) Si6-5 h, and (c) Si9-34 h alloys parallel in the drawing direction; (d) the average grain sizes of three alloys.

Full size image

Figure 5 shows the bright-field transmission electron microscope (TEM) micrographs of Si4-3 h, Si6-5 h, and Si9-34 h alloys, revealing the main precipitates in the aluminum matrix. Quantitative analysis was performed on the precipitates to identify the strengthening contribution of precipitates [Figs. 5(d) and 5(e)]. It was observed that the precipitate structures formed in the as-deformed samples are only hexagonal, being the structure of β′ precipitates [27]. Accordingly, the conventional precipitation sequence could be bypassed by the formation of β′ precipitates instead of β″ phases with the introduction of a large number of dislocations by cold deformation [27, 28]. The average lengths of the β′ precipitates were 58.2, 82.8, and 171.3 nm for Si4-3 h, Si6-5 h, and Si9-34 h alloys, while the number densities of the precipitates were 4674, 5177, and 3171 μm−3, respectively. The β′ precipitates were slightly coarsened in the Si6-5 h alloy compared with the Si4-3 h alloy, while the β′ precipitates were coarsened to a great extent in the Si9-34 h alloy with a long aging time in the overaging condition, as shown in Fig. 5. It is observed that the Si-excessive Si9-34 h alloy has the largest average length and lowest number density of the precipitates, suggesting that the Si9-34 h alloy possesses low contribution to strengthening by precipitation. According to DSC results, it could be deduced that the highest amount of the precipitates was achieved in Si9 alloy in the peak aged condition among all the alloys. However, it was required to significantly prolong the aging time to enhance the EC at the expense of the strength. Therefore, the Si9 alloy was subjected to a long time in the overaging stage to reach the required minimum EC, causing the coarsening of the precipitates. Owing to the highest number of precipitates, the Si6-5 h alloy is expected to have the highest contribution of precipitate strengthening. Based on the DSC results, it might be inferred that the Si6 alloy has a higher precipitate amount relative to the Si4 alloy in the peak aged condition. Even after the slightly long aging time, the Si6-5 h alloy still has a higher number density of the precipitates compared with the Si4-3 h alloys when they both reached the minimum required EC. The grain boundary precipitates barely occur in the coarse-grained Al–Mg–Si alloys, and precipitates mainly form in the grain interiors in all three alloys [11].

Figure 5:
figure 5

Bright-field TEM micrographs for (a) Si4-3 h, (b) Si6-5 h, and (c) Si9-34 h alloys; (d) the mean precipitate length and (e) the average number density of precipitates.

Full size image

Discussion

Correlation between strength and EC

In the peak aged condition, it was observed that the alloys with higher Si content showed greater strength, but EC is lower compared with the alloys with lower silicon content (Fig. 2), presumably resulting from the dissolved Si in the matrix [21]. When considering the strength and EC simultaneously in each alloy with different aging times, it was found that the best tradeoff between UTS and EC for Si4, Si5, Si6, and Si9 alloys were achieved after 3, 4, 5, and 34 h of aging, respectively. Figure 6(a) shows the highest UTS at the minimum required EC as a function of the Mg/Si ratio for four alloys. The UTS increases first with increasing Mg/Si ratio and reaches the maximum at Mg/Si ratio of ~1, and then dropped with a further increase of Mg/Si ratio. The order of the highest achievable strengths of the alloys with the minimum required EC is Si6-5 h, Si5-4 h, Si9-34 h, and finally Si4-3 h.

Figure 6:
figure 6

(a) UTS as a function of Mg/Si ratio and (b) hardness as a function of EC above the minimum required EC (52.5% IACS) for four alloys.

Full size image

Figure 6(b) displays the microhardness values above the minimum required EC as a function of EC, which provides an overview of the range of achievable properties in the Al–Mg–Si alloys studied. In general, the strength (hardness) decreases with increasing EC. If the strength is prioritized, the Si6 alloy with Mg/Si ratio ~1 is considered as the best candidate because it shows the highest strength up to 54% IACS among all the alloys. However, there is a restriction on the usage of the Si6 alloy at the higher EC (>54% IACS), where the strength tends to decrease sharply. If the EC is the primary concern in the conductor alloys, the Mg-excessive Si5 alloy with an Mg/Si ratio of ~1.5 provides a better compromise between strength and EC. On the high end of EC (54–56% IACS), the Si5 alloy can still maintain a reasonably high level of the strength compared with other alloys. For the Si-excessive Si9 alloy, its EC could increase no further, even after the prolonged aging time [Fig. 1(d)], and therefore, the achievable property window above 52.5% IACS is very narrow.

Strengthening mechanisms

In this section, the constitutive equations of the classical strengthening mechanisms are applied to quantitatively describe the contributions of the microstructural features at multiple levels. Knowing the importance of the individual contribution, the analysis can also provide a theoretical basis for developing new kinds of Al–Mg–Si conductor alloys with enhanced properties. The strength (hardness) is mainly determined by the forest of dislocations, precipitates, grain boundaries, and solute atoms. Correspondingly, the hardness (HV) can be expressed in the following equation, considering that the hardness is approximately equal to one-third of the yield stress (ΔHV ≈ Δσ /3) [9, 11, 31].

$${\rm H}{\rm V}_{{\rm total}} = {\rm H}{\rm V}^{{\rm Al}} + \Delta {\rm H}{\rm V}^{{\rm dislo}} + \Delta {\rm H}{\rm V}^{{\rm prec}} + \Delta {\rm H}{\rm V}^{{\rm gb}} + \Sigma \Delta {\rm H}{\rm V}_i^{{\rm sol}},$$
((1))

where HVtotal is the total hardness, HVAl is the hardness of the aluminum matrix, ΔHVdislo is the contribution of forest dislocations via strain hardening in the wire drawing, ΔHVprec is the contribution of the precipitation hardening, ΔHVgb is the contribution of grain boundary strengthening, and ΣΔHVisol is the hardness contribution via solute strengthening [9, 11].

The yield stress of the aluminum matrix was reported to be 34 MPa for a commercially pure 1100-O aluminum alloy [32]. Therefore, HVAl could be considered as approximately 11 HV, which stems from friction stress.

The dislocations in the present alloys were mainly generated from the strain hardening by cold wire drawing with an area reduction of 50%. As mentioned earlier, the hardness increases from strain hardening were experimentally determined, and they are 25.5, 28.6, and 33.1 HV for Si4, Si6, and Si9 alloys, respectively. The Bailey–Hirsch relationship for strain hardening could be used to estimate the number density of dislocations for each alloy [13].

$${\Delta {\rm \sigma }^{{\rm dislo}} = {\alpha MGb}{\rm \rho }^{1/2}},$$
((2))

where α is a dimensionless constant with a value of 0.3 for FCC Al [13], M is the Taylor factor (constant), which is 2 for polycrystalline FCC Al [32, 33], G is the shear modulus with a value of 26.9 GPa in FCC Al, b is the burgers dislocation vector with a value of 0.29 nm in FCC Al(110) [13], and ρ is the number density of dislocations. By knowing the strain hardening contribution and (ΔHV ≈ Δσ /3) in Eq. (2), the number density of dislocations for Si4, Si6, and Si9 alloys could be estimated to be 2.7 × 1014, 3.4 × 1014, and 4.5 × 1014 m−2, respectively, which is in a good agreement with the data reported in Al–Mg–Si alloys by Sauvage et al. [13] and Zhang et al. [34].

By assuming the precipitates are evenly distributed (Fig. 5), the contribution of precipitation hardening could be calculated using the Orowan equation for non-shearable precipitates [13].

$$\Delta {\rm \sigma }^{{\rm prec}}=\displaystyle{{2M{\rm \beta }Gb} \over L},$$
((3))

where β is 0.28 (constant) and M, G, and b were defined above. L is the average inter-precipitate spacing, which can be estimated from the precipitate density (N).

$$L = \displaystyle{1 \over {{( N ) }^{1/3}}}.$$
((4))

Based on the experimental measurements on precipitate parameters (Fig. 5), the contribution of precipitates hardening in the hardness of Si4-3 h, Si6-5 h, and Si9-34 h alloys were calculated to be 48.5, 50.5, and 42.5 HV, respectively.

The contribution of grain boundary strengthening can be calculated by using the well-known Hall–Petch equation [13].

$${\rm \Delta H}{\rm V}^{\rm gb} = k_{{\rm HP}}{d}^{{-}1/2},$$
((5))

Where HVgb is the contribution of the grain boundary hardness, kHP is a scaling constant typically considered to be 35 (HV μm1/2) for 6xxx series aluminum alloys [13] and d is the average grain size. Owing to the similar grain sizes in the present alloys (Fig. 4), the contribution of the grain boundary strengthening was calculated to be approximately 3 HV for all alloys, which is low compared to other contributors.

For the solute strengthening contribution, the following equation could be used [13].

$$\Delta {\rm \sigma }_i = k_iC_i^{2/3},$$
((6))

where Δσi is the increased yield stress due to the solute i, ki is a scaling factor for the solute i, and Ci is the concentration of the solute i (in wt%). In the present Al–Mg–Si alloys, only Mg and Si solutes in the aluminum matrix were considered, and kMg and kSi were 29 and 66.3 MPa (wt%)−2/3 [13]. It is apparent that the strengthening effect of Si solutes is much stronger than that of Mg solutes per wt%. Li et al. [21] reported that over half of Si amount remained in the aluminum matrix after 8 h aging at 175 °C in an Al–Mg–Si alloy with an Mg/Si ratio of 1, while the majority of Mg was consumed by the precipitates. Based on this result, it could be assumed that half of Si remains in the matrix in the high Si-containing Si6 and Si9 alloys for solute strengthening, while half of Mg is left in the matrix of the Mg-excessive Si4-3 h alloy. As an approximate estimate using Eq. (6), the contribution of solute strengthening could reach 5, 11, and 13 HV in the Si4-3 h, Si6-5 h, and Si9-34 h alloys, respectively. It is worth mentioning that the Si solute strengthening effect seems to be beyond what it is calculated with the constitutive equation since it is found that the increased hardness by rising Si from 0.4 to 0.89 wt% is approximately 90 MPa in the as-quenched condition before wire drawing, while the constitutive equation shows only 25 MPa. Therefore, the actual Si solute strengthening effect and the Si solute level remained in the matrix ought to be accounted for when identifying the solute strengthening effect.

Figure 7 shows the calculated hardness values for the Si4-3 h, Si6-5 h, and Si9-34 h alloys with different strengthening contributions, and the measured total hardness values are also listed for comparison. Although the calculated values are slightly lower than the measured ones, the general trend is in good agreement with the experimental data. It can be clearly seen that the major strength contributions come from the precipitates and strain hardening in the present alloys. Based on the calculated data and Fig. 7, the higher hardness of the Si6-5 h alloy relative to the Si4-3 h alloy might be ascribed to higher contributions of the precipitation, strain, and solute hardening. Furthermore, the lower hardness of the Si9-34 h alloy compared with the Si6-5 h alloy could mainly be attributed to a lower precipitate hardening. It is also noteworthy to mention that although the Si9-34 h alloy has lower precipitation strengthening compared with the Si4-3 h alloy, the Si9-34 h alloy exhibited a higher total hardness, owing to high contributions from strain hardening and particularly Si solute strengthening.

Figure 7:
figure 7

Comparison between calculated and experimentally measured hardness for the Si4-3 h, Si6-5 h, and Si9-34 h alloys.

Full size image

To better understand the relation between the microstructural features and EC, Matthiessen’s rule is adopted to estimate the electrical resistivity [13].

$${\rm \rho }_{{\rm total}} = {\rm \rho }^{{\rm Alpure}} + L^{{\rm dislo}}\Delta {\rm \rho }^{{\rm dislo}} + S^{{\rm GB}}\Delta {\rm \rho }^{{\rm GB}} + \displaystyle{1 \over {{( L^{{\rm prec}}) }^{1/2}}}\Delta {\rm \rho }^{{\rm Prec}} + \sum {C}_i^{{\rm sol}} \Delta {\rm \rho }_i^{{\rm sol}}, $$
((7))

where ρAlpure is the resistivity of the pure aluminum at room temperature (2.655 × 10−6 Ω cm) [13], and Δρdislo, ΔρGB, ΔρPrec, and Δρisol are the resistivity constants accounting for the contribution of dislocations, grain boundaries, precipitates, and solute elements in the aluminum matrix, respectively. Δρdislo and ΔρGB were reported to be 2.7 × 10−25 Ω m3 and 2.6 × 10−16 Ω m2 [13], and ΔρPrec to be 12 Ω(nm)3/2 [35], respectively. Ldislo is the dislocation density, and SGB is defined as the fraction of grain boundaries, which is equal to 6/d (d is the mean grain size), assuming that grains are spherical. Based on the experimental data, the contributions of electrical resistivity from dislocations for Si4-3 h, Si6-5 h, and Si9-34 h alloys are approximately 7 × 10−9, 9 × 10−9, 12 × 10−9 Ω cm, respectively, and the electrical resistivity from grain boundaries is 1 × 10−9 Ω cm for all alloys, i.e., several orders of magnitude lower than ρAlpure. Considering the electrical resistivity, ρtotal, of all alloys with the minimum requested EC (~53% IACS) being 3.25 × 10−6 Ω cm, the electrical resistivity contributions from dislocations and grain boundaries are not the controlling factors and may be neglected. LPrec is defined as the precipitate spacing (in nm) in Eq. (7) [35]. The electrical resistivity induced by precipitates for Si4-3 h, Si6-5 h, and Si9-34 h alloys with the precipitate spacing of 59.8, 57.8, and 68 nm is estimated to be 0.155 × 10−6, 0.157 × 10−6, and 0.145 × 10−6 Ω cm, respectively, which are less than 5% of the total electrical resistivity. Due to similar precipitate spacings, the difference in the electrical resistivity from precipitates in the three alloys is almost negligible.

Cisol is the concentration of solute in the matrix, and ΔρMg sol and ΔρSi sol are 0.445 × 10−6 and 0.496 × 10−6 Ω cm (at.%)−1, respectively [13]. It is apparent that the main controlling factor of electrical resistivity originates from the solute concentrations in the Al–Mg–Si alloys. As mentioned above, a significant portion of Si remains in the matrix of the Si6-5 h and Si9-34 h alloys but not in that of Si4-3 h alloy, and thus, Si solute atoms play an important role in limiting the EC in the present alloys. DSC analysis by Ding et al. [36] also confirmed that the excessive Si remained in the aluminum matrix. This is also the reason why the EC of the high Si-containing Si6 and Si9 alloys could not largely be improved relative to the low Si-containing Si4 and Si5 alloys [Fig. 6(b)]. In Si4-3 h, the majority of electrical resistivity stems from the Mg solute, which has less effect on solute strengthening, while the main resistivity in Si6-5 h and Si9-34 h alloys comes from Si solute, which has a considerable solute strengthening, as aforementioned. Furthermore, the Si solute contracts the lattice of the aluminum alloys owing to the smaller atomic size of Si compared with Al, toward the pure Al lattice parameter [15].

It is well known that precipitation from a solid solution will increase the EC of aluminum alloys because the detrimental effect of precipitates on EC is much smaller than that of solutes in the solid solution. Before reaching the peak aging, the EC sharply increases with aging time [Fig. 1(d)]. However, the EC increased only slowly with increasing aging time during overaging, but the strength rapidly decreases with aging time [Fig. 1(a)]. In other words, after peak aging, a small increase in the EC is accompanied by a substantial reduction in strength, which can provide little benefit in optimizing the combination of strength and EC.

It is worthwhile to point out that strain hardening (dislocation contribution) is one of the key factors in increasing the strength in the present alloys (Fig. 7) while having a negligible effect on EC. Moreover, Liu et al. [14] found that dislocations aid in the extraction of alloying elements from the matrix, leading to an improvement in the EC. Consequently, promoting strain hardening can be considered as a new avenue for the design of the aluminum conductor alloys with an enhanced combination of strength and EC.

Comparison of achievable strength and EC in Al–Mg–Si conductor alloys

For two common Al–Mg–Si 6xxx conductor alloys that are drawn and aged, the standard hardness and EC are ~100 HV and ~53% IACS for AA6201, and ~90 HV and ~54% IACS for AA6101 [13]. Figure 8 is shown to compare the achievable properties of two promising candidates (Si5 and Si6 alloys) in this work with the available data of Al–Mg–Si conductor alloys in the literature, although the Mg/Si ratio, process, and methodology in each study would be different. The Si6 alloy showed superior hardness in the low end of EC (52–54% IACS) relative to the standard AA6201 alloy. Similarly, the hardness values of the Si5 alloy in the high end of EC (54–56% IACS) are higher compared with the standard AA6101 alloy. As shown in Fig. 8, the achievable combination of hardness and EC of the Si5 and Si6 alloys are also superior to the data provided in the recent publications. The present work demonstrated that the appropriate chemical composition, Mg/Si ratio, and processing could further optimize the combination of strength and EC in Al–Mg–Si conductor alloys.

Figure 8:
figure 8

Comparison of hardness and EC of various Al–Mg–Si conductor alloys.

Full size image

Conclusions

  1. 1.

    The excess Si (low Mg/Si ratio) of Al–Mg–Si 6201 conductor alloys produced an increase in peak strength due to high precipitation hardening, while the corresponding EC dropped in Si-excessive alloys due to having more Si dissolved in the matrix. To fulfill the minimum required EC (52.5% IACS), the alloys with low Mg/Si ratios required a longer aging time after peak aging to increase the EC.

  2. 2.

    Above the minimum required EC, the Si6 alloy with the Mg/Si ratio of ~1 was considered to be the best candidate because it showed the highest strength in the EC range of 52–54% IACS among all alloys studied. On the high end of EC (54–56% IACS), the Si5 alloy with an Mg/Si ratio of ~1.5 provided a better compromise between strength and EC.

  3. 3.

    Although the Si-excessive Si9 alloy with an Mg/Si ratio of 0.86 exhibited the highest peak strength, it needed a much longer aging time (>34 h) to reach the minimum required EC, resulting in a significant drop in the strength and a narrow window for achievable properties.

  4. 4.

    The constitutive equations of the classical strengthening mechanisms were applied to provide a reasonable estimate of different strengthening contributions from dislocations via strain hardening, precipitates, grain boundaries, and solid solution. The calculations were in good agreement with the experimental data. The results of the quantitative analysis indicated that precipitation strengthening and strain hardening were the two major factors contributing to strength in the present alloys. In combination with EC analysis, it was found that the strain hardening can provide a significant strength contribution with a negligible impact on EC.

Experimental

Experiments were carried out on four Al–Mg–Si 6201 alloys with an Mg/Si ratio of 2, 1.5, 1, and 0.86 (designated as Si4, Si5, Si6, and Si9, respectively, in accordance with their Si levels). In the alloys, the Mg level was kept constant (0.65 wt%). The Si6 alloy with the Mg/Si ratio of ~1 was considered to be the reference alloy. The Si4 and Si5 alloys with Mg/Si ratios higher than 1 are regarded as Mg-excessive alloys, while the Si9 alloy with an Mg/Si ratio less than 1 is considered a Si-excessive alloy. The chemical compositions of the four alloys, as analyzed by optical emission spectroscopy, are presented in Table 1. To prepare the alloys, commercially pure Al, Al-50 wt% Si master alloy, and pure Mg were melted using an electrical resistance furnace. The melt was poured into a permanent mold preheated at 250 °C to obtain cast ingots with a dimension of 30 mm × 40 mm × 80 mm. After casting, the cast ingots were scalped (1.75 mm thick from each surface) and homogenized at 560 °C for 6 h, followed by hot rolling at 350–480 °C (74% area reduction). The hot-rolled sheets were cut and machined into square bars. Afterward, a solution treatment at 540 °C for 2 h was conducted, followed by cold wire drawing (50% area reduction) to obtain the final wire diameter of 4.7 mm. Artificial aging treatment at 180 °C with different aging times was applied to the wire samples. The schematics of the wire fabrication process is shown in Fig. 9.

Figure 9:
figure 9

Scheme of the wire fabrication process.

Full size image
Table 1:
figure Tab1

The chemical composition of experimental alloys with various Mg/Si ratios.

Full size image

To investigate the precipitation of the different phases, DSC was conducted on the as-drawn samples with a heating rate of 10 °C/min in an argon atmosphere. Microhardness and tensile tests were performed to evaluate the mechanical properties. Eight measurements were conducted for the microhardness test with a force and dwell time of 25 g and 20 s, respectively. The tensile tests were performed at least three times for each condition, according to ASTM B557 and ASTM E8, with a strain rate of 8 × 10−4 s−1. In addition, EC was measured via a Sigmascope unit based on ASTM E1004 with a frequency of 480 kHz, and the average of 15 measurements was reported.

An optical microscope was used to investigate the grain structures along the drawing direction, and 15 images were taken from different areas of each sample. A TEM (JEM−2100) operated at 200 kV was used to characterize the size and the number density of precipitates. The TEM samples were taken from the section perpendicular to the drawing direction and prepared using electropolishing with an electrolyte of 30% HNO3 and 70% methanol at a temperature between −20 and −30 °C. The TEM images were obtained in the <001> zone axis direction of the Al matrix. Ten TEM bright-field images were used for quantitative analysis according to the methodology provided in Ref. [37].