**3.2.3 Discussion**

As illustrated in Fig. 19, the absorbed energy of the Al–11mass%Si alloy increases with increasing number of RD-ECAP passes. However, after an abrupt increase in the first few passes, generally 4, the increment of absorbed energy gradually levels off with increased RDECAP passes, indicating that the first four passes have the greatest effect on impact toughness. This result is related to the microstructure of the as-cast Al–11mass%Si alloy. As shown in Fig. 13, the microstructure of the as-cast Al–11mass%Si alloy consists of large aluminum grains, including large dendrites and interdendritic networks of eutectic silicon plates, which are the primary reason for the low impact toughness of this alloy. We therefore conclude that breaking up this microstructure and dispersing the eutectic silicon results in improved impact toughness. It appears that the first four RD-ECAP passes do most of the work of breaking the microstructure of the large aluminum dendrites and interdendritic networks of eutectic silicon in the alloy. In fact, during the first 4 RD-ECAP passes, the grain or grain fragment sizes of this alloy are also significantly refined, as shown in Fig. 17(b).

The signal effect of ECAP, as reported in several recent works, is the modification of the grain boundaries. Misorientation angles of grain boundaries are clearly modified during

Rotary-Die Equal Channel Angular Pressing Method 59

related to grain or grain fragment boundary modification and the increase in the proportion of fine particles (smaller than 1 μm) because, as stated above, the degree of boundary modification and the fine particle content increased with increased numbers of RD-ECAP

Fig. 23. Microstructure of the Al–11mass%Si alloy processed 16 passes at 623 K by RD-ECAP followed solution treatment at 793 K for 2 h. (A. Ma, K. Suzuki, Y. Nishida, N. Saito, I. Shigematsu, M. Takagi, H. Iwata, A. Watazu, T. Imura: Acta Materialia 53 (2005) 211–220.)

Fig. 24. Microstructure of the Al–11mass%Si alloy heat-treated under T6 conditions (793 K for 2 h then at 443 K for 10 h) followed RD-ECAP at 573 K for 16 passes. (A. Ma, K. Suzuki,

In the present work, various other aluminum alloys such as Al-23 mass% Si alloy, Si-whisker/extra super duralumin composite, etc., were studied for processing by

Y. Nishida, N. Saito, I. Shigematsu, M. Takagi, H. Iwata, A. Watazu, T. Imura: Acta

Materialia 53 (2005) 211–220.)

**3.3 Other aluminum alloy** 

passes.

ECAP. In the present study, RD-ECAP had a similar effect on the grain or grain fragment boundaries. Electron backscatter diffraction (EBSD) can be used to analyze the distribution of misorientation angles (h) for the aluminum matrix in the RD-ECAPed samples since the grains or grain fragments are very small; i.e., the second particle phase can be ignored. The results for EBSD show that the fraction of high angle boundaries with h > 15 are 65% and 73% for the samples processed by RD-ECAP at 623 K for 8 and 16 passes, respectively. Therefore, the modified grain boundaries produced during RD-ECAP are likely to be an important factor in enhancing impact toughness.

In fact, the experimental results suggest that modified boundaries affect impact toughness. As shown in Figs. 17(b)–(d) and 3, although the diameter of grains or grain fragments did not clearly decrease and the distribution of the larger particles (over 2 μm in diameter) in the sample did not greatly change with increasing the number of RD-ECAP passes, the absorbed energy steadily increased with increased number of RD-ECAP passes, as shown in Figs. 18(b)–(d) and 8. This means that, except for the grain or grain fragment size and the aspect of the particles, other factors such as modified grain or grain fragment boundaries appear to be having an effect during impact testing for improving toughness. We therefore think that if the modified boundaries were eliminated, the impact toughness of this alloy would greatly decrease. To investigate the effect of the modified grain boundaries and the larger silicon particles on the impact toughness, we made two kinds of samples and measured the absorbed energy using Charpy impact tests:


Impact testing of sample 1 was carried out immediately after the solution treatment. In this case, the recrystallization of the aluminum alloy would take place during the solution treatment. However, since the precipitation of fine particles before impact testing may have been negligible, the effect of particle precipitation at the boundaries could be ignored. As shown in Fig. 12, large particles, including eutectic silicon and intermetallic compounds, other than the small amounts dissolved in the aluminum matrix during solution treatment, were still evenly distributed in the alloy due to the fact that they were evenly distributed during RD-ECAP. The grain or grain fragment size increased to around 8 μm, meaning the modified grain or grain fragment boundaries were eliminated; however, the average silicon particle size also increased with the disappearance of small particles, compared with Fig. 15(b). Impact testing results show that the absorbed energy of sample 1 fell markedly from 7.2 to 4.0 J/cm2 after the solution treatment.

As shown in Fig. 13, the large particle size in sample 2 is as large as that in sample 1 but clearly larger than that in the sample shown in Fig. 4(a). However, sample 2 exhibited a higher impact toughness (absorbed energy is 7.1 J/cm2) than the sample shown in Fig. 4(a). This result indicates that increased silicon particle size is not the reason for the impact toughness reduction of sample 1. Therefore, three factors are likely to be the chief reasons for the loss of impact toughness after the solution treatment: (a) elimination of the modified boundaries, (b) increased grain or grain fragment size, and (c) disappearance of small particles. On the other hand, although the grain or grain fragment size did not clearly decrease with the increased number of RD-ECAP passes over 4 passes, the impact toughness still markedly increased on increasing the number of RD-ECAP passes from 4 to 32, as shown in Fig. 8. This result means that the incremental value of impact toughness may be

ECAP. In the present study, RD-ECAP had a similar effect on the grain or grain fragment boundaries. Electron backscatter diffraction (EBSD) can be used to analyze the distribution of misorientation angles (h) for the aluminum matrix in the RD-ECAPed samples since the grains or grain fragments are very small; i.e., the second particle phase can be ignored. The results for EBSD show that the fraction of high angle boundaries with h > 15 are 65% and 73% for the samples processed by RD-ECAP at 623 K for 8 and 16 passes, respectively. Therefore, the modified grain boundaries produced during RD-ECAP are likely to be an

In fact, the experimental results suggest that modified boundaries affect impact toughness. As shown in Figs. 17(b)–(d) and 3, although the diameter of grains or grain fragments did not clearly decrease and the distribution of the larger particles (over 2 μm in diameter) in the sample did not greatly change with increasing the number of RD-ECAP passes, the absorbed energy steadily increased with increased number of RD-ECAP passes, as shown in Figs. 18(b)–(d) and 8. This means that, except for the grain or grain fragment size and the aspect of the particles, other factors such as modified grain or grain fragment boundaries appear to be having an effect during impact testing for improving toughness. We therefore think that if the modified boundaries were eliminated, the impact toughness of this alloy would greatly decrease. To investigate the effect of the modified grain boundaries and the larger silicon particles on the impact toughness, we made two kinds of samples and

• Sample 1 was processed with 16 passes at 623 K by RD-ECAP, then heat-treated at 793

• Sample 2 was heat-treated under T6 conditions (after solution treatment, aged at 443 K

Impact testing of sample 1 was carried out immediately after the solution treatment. In this case, the recrystallization of the aluminum alloy would take place during the solution treatment. However, since the precipitation of fine particles before impact testing may have been negligible, the effect of particle precipitation at the boundaries could be ignored. As shown in Fig. 12, large particles, including eutectic silicon and intermetallic compounds, other than the small amounts dissolved in the aluminum matrix during solution treatment, were still evenly distributed in the alloy due to the fact that they were evenly distributed during RD-ECAP. The grain or grain fragment size increased to around 8 μm, meaning the modified grain or grain fragment boundaries were eliminated; however, the average silicon particle size also increased with the disappearance of small particles, compared with Fig. 15(b). Impact testing results show that the absorbed energy of sample 1 fell markedly from

As shown in Fig. 13, the large particle size in sample 2 is as large as that in sample 1 but clearly larger than that in the sample shown in Fig. 4(a). However, sample 2 exhibited a higher impact toughness (absorbed energy is 7.1 J/cm2) than the sample shown in Fig. 4(a). This result indicates that increased silicon particle size is not the reason for the impact toughness reduction of sample 1. Therefore, three factors are likely to be the chief reasons for the loss of impact toughness after the solution treatment: (a) elimination of the modified boundaries, (b) increased grain or grain fragment size, and (c) disappearance of small particles. On the other hand, although the grain or grain fragment size did not clearly decrease with the increased number of RD-ECAP passes over 4 passes, the impact toughness still markedly increased on increasing the number of RD-ECAP passes from 4 to 32, as shown in Fig. 8. This result means that the incremental value of impact toughness may be

important factor in enhancing impact toughness.

measured the absorbed energy using Charpy impact tests:

7.2 to 4.0 J/cm2 after the solution treatment.

K for 2 h followed by water quenching (solution treatment).

for 10 h) followed by RD-ECAP for 16 passes at 573 K.

related to grain or grain fragment boundary modification and the increase in the proportion of fine particles (smaller than 1 μm) because, as stated above, the degree of boundary modification and the fine particle content increased with increased numbers of RD-ECAP passes.

Fig. 23. Microstructure of the Al–11mass%Si alloy processed 16 passes at 623 K by RD-ECAP followed solution treatment at 793 K for 2 h. (A. Ma, K. Suzuki, Y. Nishida, N. Saito, I. Shigematsu, M. Takagi, H. Iwata, A. Watazu, T. Imura: Acta Materialia 53 (2005) 211–220.)

Fig. 24. Microstructure of the Al–11mass%Si alloy heat-treated under T6 conditions (793 K for 2 h then at 443 K for 10 h) followed RD-ECAP at 573 K for 16 passes. (A. Ma, K. Suzuki, Y. Nishida, N. Saito, I. Shigematsu, M. Takagi, H. Iwata, A. Watazu, T. Imura: Acta Materialia 53 (2005) 211–220.)
