Enhancement in Mechanical Properties of AA5052 Sheet for Small Ship by Cold Roll-Bonding Process
Article information
Abstract
Abstract
A cold roll-bonding (CRB) process was applied to enhance the mechanical properties of AA5052 sheet for small ship. Two and four AA5052 sheets of 1 mm thickness, 40 mm width and 300 mm length were stacked, then roll-bonded to a thickness of 0.5 mm and 1 mm respectively by two-pass rolling after such surface treatment as degreasing and wire brushing. The rolling was performed at ambient temperature without lubricant using a 4-high mill at rolling speed of 6.0m/sec. The as roll-bonded AA5052 sheet was then annealed for 1h at various temperatures from 200 to 500°C. The as roll-bonded AA5052 sheets showed a deformation structure in which the grains are elongated to the rolling direction, however above 300°C they exhibited the complete recrystallization structure consisting of equiaxed grains, regardless of the stack number. The grain diameter of specimens annealed at high temperatures above 300°C was smaller in 4-layer stack than in 2-layer stack CRB. Both tensile strength and elongation of the specimens were better in 4-layer stack than in 2-layer stack CRB. The effects of stacking number on microstructure and mechanical properties of AA5052 sheets processed by CRB and subsequent annealing were discussed in detail.
1. Introduction
Recently, in anticipation of energy resource depletion, research on renewable energy sources, including solar and wind power, along with the lightweighting of transportation equipment, has attracted considerable national and global attention. In particular, the lightweighting of transportation equipment is becoming a critical issue not only in automobiles, which are the most commonly used means of transportation, but also in the shipbuilding industry. Steel, which is widely used as a component for transportation equipment, has the advantage of being inexpensive and capable of exhibiting various mechanical properties. However, due to its high density, steel is unfavorable for the lightweighting trend. As a result, lightweight metals such as aluminum1-11) and magnesium12-15), which can replace steel, are receiving significant attention. From a lightweight perspective, magnesium alloys are highly attractive due to having one-fourth the density of iron, but they are more expensive compared to aluminum alloys and are extremely disadvantageous in terms of plasticity13) and corrosion resistance14,15). However, aluminum alloys have long been widely used as structural materials in various fields, including transportation equipment, and their range of applications is also expanding. Compared to steel materials, aluminum alloys have the disadvantages of lower strength and poor formability. However, they have advantages such as being about one-third the weight of steel, high thermal and electrical conductivity, and excellent recyclability. As a result, in developed countries like Europe, which are actively pursuing eco-friendly policies, aluminum alloys have already replaced steel materials in many areas. However, in order for aluminum alloys to expand their application range as materials for transportation equipment, research aimed at improving mechanical properties such as strengthening and formability must be more actively promoted. Especially for small ships, the AA5083 alloy is relatively superior in terms of strength, weldability, and internal corrosion resistance in seawater compared to other aluminum alloys, making it the most commonly used material for fishing boats and leisure vessels. However, although AA5083 exhibits the highest strength among non-heat-treatable aluminum alloys, it has poor formability and is relatively expensive. In contrast, the AA5052 alloy has lower additions of elements such as Mg, Si, and Zn compared to AA5083, making it weaker but with advantages such as excellent formability and lower price. Therefore, this study conducted the cold roll bonding (CRB) process to improve the mechanical properties such as strength and ductility of the AA5052 sheet for its wide application as a material for small ships.
In general, the CRB process must successfully achieve a robust joint between metal sheets while simultaneously undergoing plastic deformation. Hence, optimizing the rolling process conditions, such as the type of metal, rolling reduction, and surface treatment, is crucial. In particular, aluminum reacts with oxygen in the air to form a thin oxide film (Al2O3) on its surface, so a surface treatment process to remove this oxide film must be carried out before the rolling process16). Furthermore, to achieve high joint strength between two aluminum sheets, a high rolling reduction beyond the critical rolling reduction is required to ensure extensive metal bonding between the newly formed surfaces of the aluminum sheets in contact with each other. In previous studies, our research team manufactured various laminated dissimilar aluminum composites consisting of 2 to 4 layers through the roll bonding process of different dissimilar aluminum materials, and we were able to achieve various mechanical properties17-23). Specifically, we found that in the combination of AA1050 and AA6061 alloys, different mechanical properties are exhibited depending on the number of overlaps23). Therefore, this study also focused on how the difference in the number of layers in the CRB of AA5052 alloys affects their mechanical properties.
2. Experimental Method
2.1 CRB and Annealing
The chemical composition of the materials used in this study is shown in Table 1, and they are commercial AA5052 sheets. The size of the specimens was 1 mm in thickness, 40 mm in width, and 300 mm in length. To remove the residual stress of the commercially available processed sheet, a homogenization treatment was performed at 400°C for 30 minutes. It was then used as the starting material for the experiment. The starting material exhibited a recrystallized structure, with tensile strength, yield strength, and elongation at a break of 207MPa, 75MPa, and 27%, respectively. Fig. 1 shows a schematic diagram of the roll bonding process. First, after performing surface treatment such as degreasing and wire-brushing on the joint areas of the AA5052 sheets, two sheets (2-layer overlap) and four sheets (4-layer overlap) were stacked and subjected to CRB under room temperature and non-lubricated conditions. The rolling was performed using a 4-high mill with a roll diameter of 370 mm at a rotation speed of 5.0 m/s in a two-pass rolling process. In the case of 2-layer overlaps, the initial thickness was reduced from 2 mm to 0.5 mm, and in the case of 4-layer overlaps, it was reduced from 4 mm to a final thickness of 1 mm. Therefore, in both cases, the total rolling reduction was the same at 75%. To calculate this as equivalent strain, assuming the deformation during rolling is plane strain, the equivalent strain ((ε)) can be obtained by the following equation 24).
where r is the rolling reduction. Therefore, substituting r = 0.75 into Equation (1), the equivalent strain becomes 1.6. Annealing was conducted at 200-500°C for 1 hour on the roll-bonded layered AA5052 sheet subjected to the high strain.
2.2 Characterization
The microstructure was observed using an optical microscope (OM) after cutting the roll-bonded specimen in the rolling direction parallel to the TD plane at the center of the sheet, followed by electro-etching in an HClO4: CH3CH2OH = 3:17 solution at a liquid temperature of -5°C and a voltage of 20V. Tensile test specimens were fabricated using a discharge machine, with a size of 1/5 of the KS 5 standard (parallel section width 5 mm, distance between points 10 mm), ensuring that the tensile axis aligned with the rolling direction. A tensile test was conducted at room temperature using an Instron-type testing machine under the initial strain rate of 8.3×10-4s-1. A hardness test was conducted using a Micro-Vickers hardness tester under a load of 0.05 kg and an indentation time of 10 seconds.
3. Results and Discussion
3.1 Microstructure
Fig. 2 shows the changes in the optical microstructure of the upper half-plane of the two-layer overlapping roll-bonded material according to the annealing temperature. Only the upper half-plane is shown here because it represents the vertically symmetric micro- structure. As shown in the figure, the roll-bonded material exhibited a typical deformation structure with grains significantly elongated in the rolling direction throughout all areas, but the grains in the surface area were smaller than those in the central area. This is because, during rolling, additional shear deformation due to friction between the rolling rolls and the material is introduced in the surface area, unlike in the central area where only compressive deformation occurs25). Due to this additional shear deformation, the surface area exhibits a practical equivalent strain value of over 1.6, unlike the central area. The materials annealed at 200°C and 250°C also showed a slight increase in grain size, but they still exhibited a deformed microstructure. It can be observed that the microstructure of the surface area is smaller compared to the central area. However, the material annealed at 300-500°C exhibited complete recrystallization, showing equiaxed recrystallized structures in almost all areas. Furthermore, as the annealing temperature increased, the grain size slightly increased. Even after complete recrystallization, the grain diameter in the surface area was smaller compared to the central area. The grain size distribution was further analyzed in detail using the LAS X Grain Analysis software through the ASTM E112 (Jeffries) method, and the results are shown in Fig. 3. As shown in this figure, the average grain diameter (¯d) of the surface area is 5.5-5.8 μm, which is more than twice smaller than the 11.3-14.5 μm of the central area. The reason the grain diameter in the surface area is smaller than that in the central area is the introduction of redundant shear strain in the surface area due to the friction between the rolling mill and the specimen, along with compressive deformation. This resulted in a larger equivalent strain, thereby increasing the recrystallization nucleation sites. Additionally, as the annealing temperature increased, the grains in both the surface and the central region increased in size due to grain growth.
Fig. 4 shows the changes in the optical microstructure of the half-plane of a 4-layer roll-bonded AA5052 sheet according to the annealing temperature. In the case of the 4-layer overlapped material, there are three rollbonded interfaces, and the roll-bonded interface on the upper side is indicated by an arrow in Fig. 4(a). As shown in the figure, both the 4-layer roll-bonded material and the 200-250°C annealing material exhibited a typical deformed microstructure with grains significantly elongated in the rolling direction in all areas, similar to the 2-layer overlapped material. Furthermore, the 300-500°C annealing material also exhibited a complete recrystallization structure. As the annealing temperature increases, the grain diameter somewhat increases, and in all recrystallized materials, the grain diameter at the surface is smaller compared to the central region, which is highly similar to the two-layer laminated material. For more detailed analysis, the grain diameter distribution was divided into three parts: the surface area, the upper joint area, and the central area, and the results are shown in Fig. 5. As shown in the figure, the average grain diameter of the surface area increased slightly from 4.7 to 5.1 μm as the annealing temperature increased from 300°C to 500°C. However, the average grain diameters at the upper joint and central areas were 5.3-9.4 μm and 10.3-12.8 μm, respectively, indicating a significant increase in grain diameter with rising annealing temperature. Additionally, in all specimens at 300-500°C, the grain diameter increased in the order of surface area, upper joint area, and central area, clearly showing a difference in grain diameter in the thickness direction of the specimen. Here, it is thought that the smaller grain diameter in the surface area, similar to the two-layer overlapping rolled material, is due to the additional shear deformation introduced in the surface area during the roll bonding process.
3.2 Mechanical Properties
Fig. 6 shows the changes in hardness distribution in the thickness direction (Fig. 6(a)) and average hardness (Fig. 6(b)) of the laminated AA5052 sheets that were roll-bonded in 2-layer and 4-layer overlaps according to increasing annealing temperature. As shown in the figure, the roll-bonded material exhibited higher hardness at the surface compared to the central region in both 2-layer and 4-layer overlaps, resulting in an uneven hardness distribution in the thickness direction. Here, the high hardness in the surface area is due to the additional shear deformation introduced to the surface during rolling, as explained in the microstructure section. In the case of the 200°C annealing material, the 4-layer overlap still shows a high hardness distribution in the surface area, but the 2-layer overlap exhibits a highly uniform hardness distribution. In the case of annealing materials at 250°C or higher, both the 2-layer and 4-layer overlaps exhibited a uniform hardness distribution in the thickness direction. Furthermore, as shown in Fig. 6(b), the average hardness gradually decreased to 250°C regardless of the 2-layer and 4-layer overlaps, significantly decreasing at 300°C or higher. The gradual decrease in hardness up to 250°C is mainly attributed to recovery, while the significant decrease above 300°C is considered to be due to the occurrence of complete recrystallization. The point to note here is that in the case of roll-bonded material, the average hardness of the 4-layer overlapped material is 108 Hv, which is higher than the average hardness of the 2-layer overlapped material at 101 Hv. This is evidence of the fact that even though both the 2-layer and 4-layer overlapped materials were rolled with the same total reduction of 75%, in reality, more equivalent strain was accumulated inside the 4-layer overlapped material compared to the 2-layer overlapped material.
Fig. 7 shows the Nominal stress-Nominal strain curves (s-s curves) (Fig. 7a) and the strength-elongation diagram (Fig. 7b) of the laminated AA5052 sheet manufactured by CRB and annealing through tensile tests. As shown in Fig. 7a, both the 2-layer and 4-layer overlap roll-bonded materials exhibit a typical s-s curve with high strength and extremely low elongation due to their cold-rolled state. However, as the annealing temperature increased, they exhibited a typical s-s softening curve where the strength decreased and the elongation rate increased. In addition, as shown in Fig. 7(b), up to an annealing temperature of 250°C, the difference between tensile strength and yield strength was small, but at 300°C or higher, this difference increased significantly, and the gap remained almost unchanged. This means that in the case of specimens at 300°C or higher, the contribution of yield strength to tensile strength is mostly due to yield strength during work hardening.
Fig. 8 summarizes the mechanical properties of 2-layer and 4-layer overlap roll-bonded materials and annealing materials in terms of tensile strength-strain rate (Fig. 8a) and yield strength-strain rate (Fig. 8b) curves. In this figure, the asterisk indicates the standard strength and elongation values of the AA5052 wrought alloy (O material and H32-H38). As shown in the figure, under all conditions, both the strength and elongation of the CRB material are higher compared to the standard values of the AA5052 wrought alloy. Further- more, it can be observed that the values of the 4-layer overlapped CRB material are higher compared to the 2-layer overlapped CRB material. This clearly demonstrates that the mechanical properties of AA5052 sheet can be improved through CRB and annealing, and that even in CRB, a 4-layer overlap can achieve superior mechanical properties compared to a 2-layer overlap. Additionally, it was observed that this trend was more pronounced in the annealing material compared to the CRB material. In particular, under these experimental conditions, the material annealed at 400°C after a 4-layer overlapped CRB exhibited the best mechanical properties in terms of strength and ductility.
Now, let us consider the reasons why the mechanical properties of the 4-layer overlapped CRB are superior compared to the 2-layer overlap. Generally, the yield strength of metallic materials is determined by solid solution strengthening, grain refinement strengthening, precipitation strengthening, dislocation strengthening, etc26). In this study, most of the conditions such as the materials used, rolling reduction, and annealing temperature were the same for both 2-layer and 4-layer overlap. However, the differences lie in the number of layers being 2 and 4-layer overlap, and the initial thicknesses (2 mm and 4 mm) and final thicknesses (0.5 mm and 1 mm). Due to the differences in these rolling conditions, the shear strain introduced during the 4-layer overlap CRB was greater compared to the 2-layer overlap, which is believed to have caused the differences in hardness and strength due to dislocation. In other words, the shear strain introduced during rolling is influenced by the ratio of the average thickness of the sheet before and after rolling to the length of contact between the roll and the workpiece. Since this value would differ between the 2-layer and 4-layer overlaps, it is estimated that the magnitude of the introduced shear strain would also differ. Furthermore, it is believed that the superior mechanical properties of the 4-layer overlap compared to the 2-layer overlap in low-temperature annealing materials at 200 and 250°C are also due to the same reason. However, complete recrystallization occurs in the case of annealing materials at 300°C or higher. Hence, grain refinement strengthening would have operated as a more important strengthening mechanism than dislocation strengthening. Grain refinement strengthening is well explained by the following Equation(2)27,28)
where σY is the yield strength, σ0 is the friction stress that hinders the motion of dislocations, k is a constant, and d is the grain diameter. The average grain diameters of the 2-layer overlapped specimens after annealing at 300-500°C were 8.4, 9.6, and 10.2 μm, respectively, while for the 4-layer specimens, they were 6.8, 7.5, and 9.1 μm. Therefore, it is believed that under the same annealing conditions, the smaller grain diameter (d) of the 4-layer overlapped material compared to the 2-layer overlapped material contributed more to the increase in yield strength.
Another major difference between the 2-layer and 4- layer overlaps is the number of roll-bonded interfaces. In other words, the 2-layer overlap has one roll-bonded interface, but the 4-layer overlap has three roll-bonded interfaces. The presence of more roll-bonded interfaces means that a greater number of wire-brushing operations are performed to ensure a strong bond between the AA5052 sheets. The wire-brushing process is a type of friction process that removes the robust Al2O3 film on the surface of the aluminum sheets. In this case, the aluminum surface layer experiences a temperature increase due to frictional heat, resulting in recrystallization and the formation of areas composed of ultrafine grains. According to Professor Tsuji’s research team at Kyoto University in Japan, wire-brushing forms ultra-fine grains of approximately 200 nm up to 17 μm below the surface29). The formation of such ultrafine grains is also believed to have contributed to the increase in the yield strength of the 4-layer overlapped material with a larger number of roll-bonded interfaces.
4. Conclusion
This study successfully fabricated AA5052 sheets through the CRB process with varying overlap counts for the 2nd and 4th layers, leading to the following conclusions.
1) Both the 2-layer and 4-layer overlapped CRB AA5052 materials exhibited a typical deformed microstructure with grains significantly elongated in the rolling direction. The 200 and 250°C annealed materials also showed a slight increase in grain thickness due to recovery, but still exhibited a deformed microstructure. In annealing materials at 300°C or higher, complete recrystallization occurred, resulting in an equiaxed grain structure.
2) Both the 2-layer and 4-layer overlapped materials exhibited coarse and uneven microstructures in the central region compared to the surface region. Under the same annealing conditions, the grain diameter of the 4-layer overlapped material was smaller than that of the 2-layer overlapped material.
3) The CRB material in both the 2-layer and 4-layer overlap showed higher hardness at the surface compared to the central region, resulting in an uneven hardness distribution in the thickness direction. However, the annealing material at 250°C or higher showed a highly uniform hardness distribution. Furthermore, the average hardness gradually decreased up to 250°C, and then significantly decreased at 300°C or higher where complete recrystallization occurred.
4) The 2-layer and 4-layer overlapped materials displayed a characteristic softening curve, indicating a reduction in tensile strength and an increase in elongation with rising annealing temperatures. The 4-layer overlapped material exhibited enhanced strength and elongation properties relative to the 2-layer overlapped material.
5) From the above results, it can be concluded that it is possible to manufacture laminated AA5052 sheets with various strength-ductility combinations using 2- layer and 4-layer overlapped CRB. Furthermore, differences in the number of layers and annealing temperature significantly affected the mechanical properties of the laminated AA5052 sheets.
Acknowledgement
This work was supported by Korea Institute for Advancement of Technology (KIAT) grant funded by the Korean Government (MOTIE)(P0017006, HRD Program for Industrial Innovation, 2024).