2023 年 64 巻 2 号 p. 379-384
Vertical-type high-speed twin-roll casting (VT-HSTRC), which is characterized by a high production rate and cooling rate, is a promising method for upgrade recycling of aluminum cast alloy scrap to wrought alloys in the near future. To produce wrought alloy sheets from cast alloy scrap, the strips must be isotropic to achieve good formability. However, in cold-rolled and annealed Al–7% Si alloy and A356 alloy sheets fabricated from the HSTRC strips, average elongation is much greater in the rolling direction than in the transverse direction. This elongation anisotropy results from both the morphology and the alignment of eutectic Si particles. In the present study, the effect of homogenization heat treatment on the microstructure and elongation was investigated. Al–7% Si and Al–11% Si alloy strips were fabricated by HSTRC and were homogenized by heat treatment at 540°C for 10 h and 500°C for 10 h, respectively. The strips were cold rolled at a reduction rate of 50% and annealed. The eutectic Si particles were spheroidized and coarsened by the homogenization heat treatment, and they were uniformly dispersed after cold rolling. There was no significant difference in elongation between the rolling and transverse directions in the Al–7% Si and Al–11% Si alloys. These results show that the homogenization heat treatment of the strips reduced the elongation anisotropy.
TD cross-sectional microstructures and elongation of cold-rolled and annealed Al–7% Si strip. (a) Non-homogenized sheet, (b) homogenized sheet.
Aluminum alloys have been used for automotive components to improve fuel efficiency by reducing the weight.1) As the use of aluminum alloys increases, the amount of scrap is also rising. In addition, aluminum can be recycled using only a few percent of the energy used in the production of virgin ingots, which not only reduces production costs but also reduces carbon dioxide emissions from the production process. For these reasons, there is growing demand for aluminum recycling.2)
However, the general use of the recycled aluminum alloys is still limited to the production of cast products via a process that is called cascade recycling because the quality of the products is lower than that of wrought aluminum. Furthermore, as electric vehicle production continues to increase, the demand for cast aluminum alloys used for engine parts is decreasing. Therefore, there is likely to be an oversupply of aluminum product scrap in the near future. To solve this problem, it is necessary to establish product-to-product recycling (wrought alloy scrap to wrought alloys) and upgrade recycling (cast alloy scrap to wrought alloys).
VT-HSTRC is a method for fabricating aluminum alloy strips directly from the molten metal.3,4) A schematic of the vertical-type high-speed twin-roll caster5) used in this study is shown in Fig. 1. The caster consists of pure copper rolls with a water-cooling mechanism, a series of springs for applying load to the strips, nozzles, and side-dams for making a pool of molten metal. The cooling rate during solidification is about 1000°C/s,6) which is sufficient to refine the microstructure, detoxify impurity elements, and promote supersaturation of solute atoms in the matrix.7–9) Because of these advantages, VT-HSTRC is a candidate method for using cast alloy scrap to produce wrought alloy products.
Schematic of the vertical-type high-speed twin-roll caster.5) 1. Graphite crucible, 2. melt, 3. nozzle, 4. side dam, 5. melt pool height, 6. roll rotation speed, 7. solidification length, 8. initial roll separating force, 9. initial roll gap (pre-casting stage).
Typical cast aluminum alloys include A356 (Al–Si–Mg) and ADC12 (Al–Si–Cu). These alloys contain large amounts of Si to improve castability (A356 contains about 7 mass% Si and ADC12 contains about 11 mass% Si). In these alloys, the amount of eutectic Si particles is large, and their size, morphology, and distribution control the mechanical properties of the material. To produce wrought alloy sheets from cast alloy scrap, the strips produced by VT-HSTRC must be isotropic and highly ductile to achieve good formability. However, the elongation of cold-rolled and annealed A356 sheets produced from the VT-HSTRC strips differs in the rolling direction (RD) and transverse direction (TD) of the sheets.10) Harada et al.11) found that no elongation anisotropy was observed in pure Al, whereas elongation anisotropy was observed in Al–2% Si (RD: about 32%, TD: about 27%) and Al–7% Si (RD: about 30%, TD: about 20%), and it was greater in Al–7% Si. This suggests that the elongation anisotropy was caused by eutectic Si particles. Eutectic Si particles were randomly distributed in the RD cross section of cold-rolled and annealed Al–7% Si binary alloy strips when they were oriented along the RD in the TD cross section. Therefore, during loading in the TD, the voids that formed around the eutectic Si particles were likely to connect at right angles to the loading direction during tensile testing, thereby decreasing the elongation ratio.
Increasing the rolling reduction rate reduces the elongation anisotropy because the eutectic Si particles are uniformly distributed.10,11) However, the reduction of the elongation anisotropy by heat treatment has not been reported. Therefore, we focus on using homogenization heat treatment to obtain a microstructure with uniformly dispersed eutectic Si particles. In this study, we investigated the effect of homogenization heat treatment on the elongation anisotropy in cold-rolled and annealed Al–7% Si and Al–11% Si alloy sheets fabricated from VT-HSTRC strips.
Al–7% Si and Al–11% Si alloys were obtained by mixing ingots of commercially pure aluminum and Al–25 mass% Si alloy. The chemical compositions of the alloys are listed in Table 1. The strips were fabricated by VT-HSTRC. The diameter and width of the pure copper rolls are 300 and 100 mm, respectively. One roll is fixed rigidly to the pedestal, whereas the other roll is attached to the pedestal by a spring to apply load to the solidifying strip. The spring loading ensures firm contact between the roll surfaces and the solidifying shell, resulting in excellent heat removal and a high cooling rate. The casting parameters are listed in Table 2. The Al–7% Si and Al–11% Si strips were fabricated at speeds of 60 and 40 m/min, respectively. A 20 kN load was applied to one of the rolls by springs before the casting and the initial roll gap was 1 mm. The solidification length, which was the contact length between the nozzle tip and the roll gap along the roll surface, was fixed as 100 mm. The melt pool height was kept at about 100 mm during casting. The alloy placed in the graphite crucible was melted in an electric furnace under an argon gas atmosphere. The melt was degasified by blowing argon gas into the melt at 750°C for 20 min. Thereafter, the crucible containing the molten metal was removed from the furnace and the molten metal was poured into the nozzle at a temperature 20°C higher than the liquidus temperature. About 2.5 kg molten alloy was prepared. Strips approximately 3 m long, 100 mm wide, and 2 mm thick were fabricated. The specimens were cut from the central part in the casting direction of the strips with a constant thickness, and were used for subsequent homogenization heat treatment, microstructural observation, and cold rolling. The homogenization heat treatment was performed at 500 and 540°C for 10 h (Al–7% Si) and at 500°C for 10 h (Al–11% Si), followed by furnace cooling. The RD was the same as the casting direction. The rolling reduction rate to 1 mm thickness was about 50%. The cold-rolled strips were annealed at 540°C for 1 h (Al–7% Si) and at 500°C for 1 h (Al–11% Si), followed by water cooling.
Cross sections of the Al–7% Si alloy and Al–11% Si alloy strips (as-cast, after homogenization heat treatment, and after cold-rolling and annealing) were ground with waterproof abrasive paper sheets up to #4000 grit, and then polished with diamond pastes (6, 3, and 1 µm) and a colloidal silica suspension. The polished surfaces were etched with 5% NaOH aqueous solution. The microstructures were observed with an optical microscope. As in the previous study,11) Voronoi diagram was created on the cold-rolled and annealed Al–7%Si sheet using Image J software to calculate the coefficient of variation (CV) of the Voronoi cell area. This CV value indicates the dispersion state of eutectic Si particles.
2.3 Tensile testsMechanical properties of the cold-rolled and annealed Al–Si alloy sheets were investigated. Small tensile test specimens were taken from the rolled strips at 0° (RD) and 90° (TD) relative to the RD (Fig. 2). The RD and TD samples were taken from different rolled sheets, and 5 samples were respectively used. The specimens were annealed (Al–7% Si: 540°C, 1 h; Al–11% Si: 500°C, 1 h) and water cooled. The tensile test was carried out using an Instron type testing machine at a crosshead speed of 1 mm/min at room temperature. The loading directions were the RD and TD of the rolled strip.
Diagram of tensile test specimen.
The TD cross-sectional microstructures of the as-cast Al–Si strips produced by VT-HSTRC are shown in Fig. 3. Previous study11) has shown that there is no major difference in the α-Al phase and eutectic Si particles in the CD and TD cross sections of the as-cast strip. The bright contrast is the primary α-Al phase and the dark contrast is the eutectic Si particles. The cooling rate near the surface was high; therefore, fine eutectic Si particles that were uniformly distributed were observed near the surface of the strips (Figs. 3(a), (c)). However, the eutectic Si particles agglomerated and formed clusters in the central region of the strip, which was the final solidification region (Figs. 3(b), (d)).
TD cross-sectional microstructures of as-cast strips. (a) Near-surface and (b) central regions in Al–7% Si and (c) near-surface and (d) central regions in Al–11% Si.
Figure 4 shows the cross-sectional microstructures of Al–Si strips after homogenization heat treatment. Compared with the microstructures of the as-cast Al–Si strips shown in Fig. 3, the eutectic Si particles were coarser and no clusters of eutectic Si particles were observed. In the Al–7% Si alloy strips homogenized at 500°C (Figs. 4(a), (b)), there were considerable differences in the microstructures between the near-surface and central regions. The eutectic Si particles were fine and spherical in the near-surface region (Fig. 4(a)), whereas they were coarse and rod-shaped in the central region (Fig. 4(b)). We attempted to spheroidize the eutectic Si particles in the central region by increasing the heat treatment temperature. After heat treatment at 540°C, there was no difference in the microstructures between the near-surface and central regions, and the eutectic Si particles became coarse and spherical (Figs. 4(c), (d)). Thus, the Al–7% Si strip heat-treated at 540°C was cold rolled and annealed, and the mechanical properties were measured. Figures 4(e), (f) show the microstructures of the Al–11% Si alloy strip homogenized at 500°C. There were no major differences in the microstructure between the near-surface and central regions after heat treatment at 500°C, and this strip was also used for cold rolling. These results showed that the distribution and morphology of eutectic Si particles was homogenized by appropriate heat treatment and a uniform microstructure was obtained throughout the material.
Cross-sectional microstructures of strips after homogenization heat treatment. (a) Near-surface and (b) central regions in Al–7% Si homogenized at 500°C, (c) near-surface and (d) central regions in Al–7% Si homogenized at 540°C, and (e) near-surface and (f) central regions in Al–11% Si homogenized at 500°C.
Figure 5 shows the cross-sectional microstructures of cold-rolled and annealed Al–Si sheets. In the Al–7% Si alloy (Figs. 5(a)–(d)) and Al–11% Si alloy (Figs. 5(e)–(h)), the eutectic Si particles were uniformly distributed in the TD cross sections of both the near-surface and central regions, as in the RD cross sections. Without homogenization heat treatment before cold-rolling, in the cold-rolled and annealed Al–Si alloy sheets, the eutectic Si particles in the central region were uniformly distributed in the RD cross section, whereas these particles were oriented in the RD in the TD cross section.11) However, the eutectic Si particles were uniformly distributed in the TD cross section of the homogenization heat-treated sheets as well as in the RD cross section (Figs. 5(b), (d) and (f), (h)). As a result of measuring the CV value of Voronoi cell area for Al–7%Si, it was 0.61 and 0.59 for the RD cross section and the TD cross section, respectively, with no difference. This indicates that the degree of dispersion of eutectic Si particles in the RD and TD cross sections is comparable.
Cross-sectional microstructures of cold-rolled and annealed sheets. (a), (b) RD cross section and (c), (d) TD cross section in Al–7% Si and (e), (f) RD cross section and (g), (h) TD cross section in Al–11% Si.
The tensile properties of the cold-rolled and annealed strips are shown in Fig. 6. The 0.2% proof stress and the ultimate tensile strength were similar in the RD and TD in the Al–7% Si alloy and Al–11% Si alloy, that is, there was no anisotropy. The elongation of Al–7% Si was 24.8% and 28.3% in the RD and TD, respectively. Compared with the values in the previous study11) as shown in Fig. 7, the elongation in the TD was greatly improved and the anisotropy was improved. In contrast, the elongation of Al–11% Si was 28.5% and 24.5% in the RD and TD, respectively. The difference in elongation was larger than that of the Al–7% Si alloy sheet. In the previous study,11) the difference between elongation in the RD and TD also increased with increasing Si content. The results (28.5% in RD and 24.5% in TD) for the Al–11% Si alloy suggested that the elongation anisotropy was suppressed at high Si contents. Homogenization heat treatment of the as-cast strips eliminated clusters of eutectic Si particles, and the uniform distribution of the eutectic Si particles in both the RD and TD cross sections in the cold-rolled and annealed sheets reduced the elongation anisotropy. In the central region of the as-cast strip, coarse plate-like eutectic Si particles were observed (Fig. 3). Goda and Kumai reported that these plate-like particles were crushed and elongated along the RD and cracks formed. Upon increasing the rolling reduction, these particles were homogeneously dispersed.10) In this study, the eutectic Si particles in the Al–7% Si and Al–11% Si alloy sheets had already been spheroidized and were uniformly dispersed by homogenization heat treatment before cold rolling. In addition, no clusters of the fine eutectic Si particles were observed. If the clusters were present, these would have been oriented in the RD after cold-rolling due to the narrow spacing of the eutectic Si particles. The homogenization heat treatment activated the diffusion of Si atoms, fine Si particles in the clusters grew, and the distance among the particles increased, resulting in a uniform distribution of eutectic Si particles after cold rolling. Furthermore, no cracking of the eutectic Si particles was observed in the microstructure of the cold-rolled and annealed sheets, which was attributed to the spheroidization of the eutectic Si particles before rolling by homogenization heat treatment and the suppression of particle fracture. The elongation anisotropy in the Al–7% Si alloy and Al–11% Si alloy sheets was reduced by homogenization heat treatment without cold rolling at the high reduction rate of the VT-HSTRC strips. Therefore, homogenization heat treatment is a candidate method for making cast alloys or recycled alloys into wrought alloy products by VT-HSTRC.
Mechanical properties of cold-rolled and annealed sheets. (a) Al–7% Si and (b) Al–11% Si.
Elongation of cold-rolled and annealed sheets in Al–7%Si. (a) without homogenization and (b) with homogenization before cold-rolling.
The effect of homogenization heat treatment on the elongation anisotropy in cold-rolled and annealed Al–7% Si and Al–11% Si alloy sheets fabricated from VT-HSTRC strips was investigated.
