**4. Roll-casting**

### **4.1 Single-roll caster equipped with a scraper**

It is known that centerline segregation occurs between the solidification layers in Al-Mg alloy strips cast using a twin-roll caster (TRC). It is difficult to reproduce the occurrence of this type of centerline segregation in strips cast using twin-roll casters. In this study, a single-roll caster equipped with a scraper (SRCS) was used to cast Al-Mg alloys strips without centerline segregation [11].

In the SRCS, the molten metal is solidified on the side of the one roll, and a centerline does not form. The free solidified surface is scribed into a flat surface by the scraper. The scraper load was 0.2 N/mm, and no crack was formed on either surface of the strip because of the small scraper load. A copper roll was used to increase the cooling speed and roll speed. In the conventional TRC, steel rolls are used. The thermal conductivity of copper is much larger than that of steel, and the cooling ability of a copper roll is thus greater than that of a steel roll. The casting speed of the SRCS was 30 m/min, whereas the casting speed of a conventional TRC is usually slower than 2 m/min. The excellent cooling ability of the copper roll enabled high-speed roll casting. Schematic illustrations of the SRCS and the area near the scraper are shown in **Figure 15**. Cross-sections of Al-4.5%Mg strips cast using the high-speed TRC and the SRCS are shown in **Figure 16** [11, 12]. Centerline segregation occurred in the strip cast using the high-speed TRC and not in the strip cast using the SRCS.

### **4.2 Effect of Mg content on strip thickness and surfaces**

Strips of Al-Mg alloys with Mg contents ranging from 4.5–10% could be continuously cast using the SRCS. The strip thickness is plotted against the Mg content in **Figure 17**. The strip became thicker as the Mg content increased. Two potential causes were considered to explain the relationship between the Mg content and the strip thickness. One is that the latent heat of the magnesium is smaller than that of the aluminum; thus, the latent heat of Al-Mg alloy decreases as the Mg content increases, which may then cause the strip thickness to increase with the Mg content. The other is that the amount of scribed and piled aluminum alloy under the scraper becomes greater as the Mg content increases, and the piled

#### **Figure 15.**

*Schematic illustrations of (a) a top-down view of a single-roll caster equipped with a scraper and (b) a view near the scraper.*

*Characteristics of Al-Mg Test Pieces with Fe Impurities Fabricated by Die Casting, Roll Casting… DOI: http://dx.doi.org/10.5772/intechopen.100940*

#### **Figure 16.**

*Cross-sections of Al-4.5%Mg alloys cast using (a) a high-speed twin-roll caster and (b) a single-roll caster equipped with a scraper. The casting speed was 30 m/min.*

**Figure 17.** *Strip thickness plotted against Mg content.*

aluminum alloy becomes a part of the strip [12]. Therefore, the strip becomes thicker as the Mg content increases.

The surfaces of the Al-Mg alloy strips are shown in **Figure 18**. The scribed surface did not have a metallic luster, whereas the roll contact surface did. The Mg content did not influence the surface condition.

#### **4.3 Mechanical properties of roll cast Al-Mg alloy strips**

The mechanical properties of the roll-cast Al-Mg alloy strips were tested by the tensile test. The cast strip was cold-rolled down to 1 mm and annealed at 360°C for 90 min. The dimensions of the test piece for the tensile test are shown in **Figure 19**.

The results of the tensile test are shown in **Figure 20**. The tensile stress increased monotonically at a gradual rate with increasing Mg content. The 0.2% proof stress was almost constant at different Mg contents. The elongation increased with increasing Mg content up to 8% Mg and then substantially decreased at 10% Mg. Comprehensively, judging from the tensile test, the Al-8%Mg showed the best mechanical properties.

A deep drawing test was then conducted to investigate the ability of sheet forming. The cast strip was cold-rolled down to 1 mm and annealed at 360°C for 90 min. The diameter of the punch used for the deep drawing test was 32 mm. The

#### **Figure 18.**

*Surfaces of as-cast Al-Mg alloy strips.*

#### **Figure 19.**

*Size of a test piece for the tensile test.*

**Figure 20.** *Tensile test results for roll-cast Al-Mg alloys with different Mg contents.*

deep drawing test was conducted under two conditions: with the roll-contact side of the strip facing outward and with the scribed surface facing outward. The results of the deep drawing test are shown in **Figure 21**. The limiting drawing ratio (LDR, the maximum ratio of circular blanks to the diameter of the die) of Al-4.5%Mg was 2.0 regardless of which side of the strip was facing outward. The LDR decreased

*Characteristics of Al-Mg Test Pieces with Fe Impurities Fabricated by Die Casting, Roll Casting… DOI: http://dx.doi.org/10.5772/intechopen.100940*

**Figure 21.** *Limiting drawing ratio at different Mg contents.*

with increasing Mg content. When the Mg content was 4.5% or 6%, the LDR was not affected by which side faced outward; in contrast, when the Mg content was 8% or 10%, the LDR was better when the roll-contact side faced outward. The difference between the LDRs in these two cases is not suitable for sheet forming. The Al-4.5%Mg was most suitable for sheet forming. The optimal Mg content for deep drawing was different from that obtained from the elongation in the tensile test. These results demonstrate that 514.0 aluminum alloy is suitable for sheet forming, and 518.0 aluminum alloy is suitable for the easy shape plate, which needs strength and elongation. This shows that the choice of Mg content depends on the purpose. The forming ability is the most important property for sheets used in automobile manufacture, and thus the Al-4.5%Mg is suitable for this purpose. The Al-4.5%Mg was used to make the model alloy of recycled Al-Mg alloys.

### **4.4 Mechanical properties of Al-4.5%Mg with Fe**

Impurities of 0.2%, 0.4%, 0.6%, and 0.8% Fe were added to the Al-4.5%Mg to model the recycled Al-Mg alloy. The Al-4.5%Mg with Fe could be cast into a strip, as the addition of the Fe did not affect the ability of the roll casting; however, the addition of the Fe makes the strip hard and brittle. Edge cracks with lengths of 3 mm or less occurred in the Al-4.5%Mg with 0.8%Fe, and the cold rolling could be conducted on the strip down to 1 mm without breaking. When the added Fe content was less than 0.6%, edge cracking did not occur. The surfaces of the as-cast and the cold-rolled strips of the Al-4.5%Mg and the Al-4.5%Mg with 0.8%Fe are shown in **Figure 22**. There was no difference between the scribed and roll-contact surfaces of the cold-rolled virgin Al-4.5%Mg and Al-4.5%Mg with 0.8%Fe strips. It is thought that the increase in the Fe content does not affect the surface properties of the Al-4.5%Mg sheet cast by the SRCS after cold rolling.

Cross-sections of the virgin Al-4.5%Mg and Al-4.5%Mg with 0.8%Fe strips are shown in **Figure 23**. The grain of the as-cast Al-4.5%Mg strip was almost uniform in the thickness direction. In the as-cast strip of Al-4.5%Mg with 0.8%Fe, the grain of the roll-contact side of the strip was finer than that of the scribed side. The effect of cooling speed on the grain size of the Al-4.5%Mg with 0.8%Fe was more apparent than in the Al-4.5%Mg. This is the influence of the added Fe. The Fe formed a crystal nucleus, and many crystals were made. As a result, the grain number increased and the grain size became small near the roll-contact side. The structure became a fine deformation structure after cold rolling and annealing.

**Figure 22.**

*Surfaces of as-cast and cold rolled strips of Al-4.5%Mg and Al-4.5%Mg with 8%Fe.*

**Figure 23.**

*Cross-sections of as-cast strip and cold rolled and annealed strip of Al-4.5%Mg and Al-4.5%Mg with 0.8%Fe. Annealing: 360°C for 90 min.*

The results of the tensile test of the Al-4.5%Mg with Fe are shown in **Figure 24**. The tensile stress was almost uniform for the added Fe content. The 0.2% proof stress gradually increased with increasing Mg content, and the elongation very gradually decreased. The elongations of the Al-4.5%Mg and Al-4.5%Mg with 0.8%Fe were 30.3% and 28.6%, respectively. The reduction of the elongation with the addition of Fe *Characteristics of Al-Mg Test Pieces with Fe Impurities Fabricated by Die Casting, Roll Casting… DOI: http://dx.doi.org/10.5772/intechopen.100940*

**Figure 24.** *Result of tensile test of Al-4.5%Mg with Fe.*

**Figure 25.**

*Limiting drawing ratio of Al-4.5%Mg with different Fe Contents.*

was very small. The intermetallic compound including Fe may be very fine because of the rapid solidification of the rolling caster, and it did not make the elongation worse.

The LDR of Al-4.5%Mg with Fe is shown in **Figure 25**. The LDR did not decrease from 2.0 until the addition of 0.4% Fe. When the Fe content was 0.6%, the LDR was 1.9. The LDR when the scribed surface faced outward was the same as that when the roll contact surface faced outward until the Fe content was 0.6%. Therefore, the ultimate addition of Fe to the Al-4.5%Mg was 0.4%.
