- Email: [email protected]
Study of ultrasonic melt treatment on the quality of horizontal continuously cast Al–1%Si alloy
Ultrasonics Sonochemistry 13 (2006) 121–125 www.elsevier.com/locate/ultsonch
Short communication
Study of ultrasonic melt treatment on the quality of horizontal continuously cast Al–1%Si alloy Li Xin-tao *, Li Ting-ju, Li Xi-meng, Jin Jun-ze Research Center of Foundry Engineering, Dalian University of Technology, Dalian 116024, China Received 14 May 2005; accepted 26 August 2005
Abstract The fluctuation of the melt temperature in a tundish was measured during casting and experiments were conducted to investigate the effects of ultrasonic melt treatment on the surface quality and solidification structures of Al–1%Si ingots. The results show that the uniformity of melt temperature was enhanced with the application of ultrasonic melt treatment. When the ultrasonic power is 1000 W, the surface quality was evidently improved and grains of cast ingots were refined. Moreover, EPMA analysis was adopted to study the relationship between the ultrasonic power and boundary segregation of Si element. The result shows that boundary segregation is suppressed with the increase of ultrasonic power and the phenomenon was theoretically interpreted. Ó 2005 Elsevier B.V. All rights reserved. Keywords: Ultrasonic melt treatment; Al–1%Si alloy; Surface quality; Solidification structure; Boundary segregation
1. Introduction A tundish is one of the major components of horizontal continuous casters to deliver melt to the mold(s) at a controlled rate and remove non-metallic inclusions from liquid melt [1]. In horizontal continuous casting, early emphasis was placed on the mold because product quality and productivity are determined mostly by mold practices and parameters. With the development of solidification technology and clusters theory, effects of the melt quality on the final solidification structure are being addressed [2]. Tundish metallurgy is a new concept that was first put forward to improve melt quality by Mclean et al. in 1980s [3,4]. Since then the investigation of the relationship between tundish metallurgy and the product quality has been carried out, which has become a hot topic in recent years [5,6]. Ultrasound is a high frequency mechanical wave. The propagation and interaction of sound waves change the *
Corresponding author. Tel./fax: +86 0411 84708940. E-mail address: [email protected] (X.-t. Li).
1350-4177/$ - see front matter Ó 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.ultsonch.2005.08.005
physical and chemical properties of materials that are subjected to ultrasound. Over the last few years high-intensity ultrasound has been used for producing new materials successfully. For example, ultrasonic treatment is applied in melts during solidification in order to obtain fine-grain solidification microstructures [4–6]. High-intensity ultrasound has been used to fabricate continuous fibre and particulate reinforced metal matrix composites, by improving the wettability or promoting the chemical reaction between the reinforcement and the metal matrix [7–12]. Unfortunately, so far, there have been few reports on the application of high-intensity ultrasonic to the melt treatment in tundish (which is abbreviated to ultrasonic melt treatment in the following) during horizontal continuous casting. Nowadays fine wire bonding is the most popular interconnection method used in packaging electronic devices for electrical connection between the integrated circuit blocks and the external frame and Al–1%Si alloy is the main raw material used to manufacture bonding wire [13,14]. In this paper, the novel processing of ultrasonic melt treatment is proposed to improve the quality of a horizontal continuously cast ingot of Al–1%Si alloy. The
122
X.-t. Li et al. / Ultrasonics Sonochemistry 13 (2006) 121–125
influence of ultrasonic melt treatment on the temperature fluctuation in the tundish, quality of cast ingots such as surface roughness and solidification structure are presented, and discussed. 2. Experimental A schematic view of the experimental apparatus is shown in Fig. 1. The primary mold consists of a copper outer-jacket and a graphite inner-mold. The ultrasonic emitter, made of stainless steel and coated with TiN membrane, was inserted into the melt from the top of the tundish and connected to a generator with a maximum power capacity of 1 kW and frequency of 22.3 kHz. A secondary mold is immediately connected to the primary one, in which cooling water is distributed uniformly over the periphery of the ingot. The experimental conditions are summarized in Table 1.
Table 1 Experimental conditions Items
Parameters
Casting temperature Water flux Water temperature Casting speed Diameter of ingot
720 °C 40 L/h 20 °C 100 mm/min 10 mm
3. Results and discussion The fluctuation of the melt temperature in the tundish is measured by thermocouples during casting, the positions of which are shown in Fig. 1. Fig. 2 shows that melt temperature declines from position A to position C in the absence of ultrasonic melt treatment and the maximum temperature difference is 11 °C. By contrast, the temperature distributes more uniformly in the tundish after the application of ultrasonic melt treatment, and the maximum temperature difference declines to 2 °C. Fig. 3 indicates the surface of cast ingots with and without the application of ultrasonic melt treatment. It is clearly observed that ripple marks in the longitudinal direction are formed, especially on the bottom side without the application of ultrasonic melt treatment, as seen in Fig. 3(a) and (b). Whereas, when ultrasonic melt treatment was applied with the power of 1000 W, the ripple marks
735
P=0W
Melt temperature,T/ºC
A
Fig. 1. Top: schematic view of experimental apparatus: (1) melt, (2) thermocouples, (3) tundish, (4) ultrasonic emitter, (5) ultrasonic transducer, (6) temperature-gathering equipment, (7) primary mold, (8) secondary mold, (9) ingots. Bottom: top view showing the thermocouple positions for the test.
730
B
P=1000W
725
C
720 715 710 20
40 80 60 Length of the tundish/mm
100
Fig. 2. Distribution of temperature in tundish at h = 40 mm (point A, B, C) corresponding to the thermocouples positions shown in Fig. 1.
Fig. 3. Surface view of cast ingots in the absence and presence of ultrasound. (a) P = 0 W, upper part view, (b) P = 0 W, bottom view, (c) P = 1000 W, upper part view, (d) P = 1000 W, bottom view.
20
P=1000W
1
2 3 4 5 6 7 Length of measured ingot/mm
8
9
0 -10
Surface roughness/µm 0
-30
-30 a
10 b
P=1000W
10
P=0W
Surface roughness/µm -20 -10 0 10
P=0W
123
-20
20
X.-t. Li et al. / Ultrasonics Sonochemistry 13 (2006) 121–125
0
1
7 8 3 4 2 5 6 Length of measured ingot/mm
9
10
Fig. 4. Surface roughness tested by a displacement sensor with and without the application of ultrasonic melt treatment: (a) upper part, (b) bottom side.
disappeared completely and the surface of the ingot is smooth enough to be rolled without the ÔscalpingÕ operation that is required following traditional casting. Fig. 4 presents the surface roughness of the cast ingots measured by a displacement sensor in the two cases. It can be seen the surface roughness of the upper part reduced from 30 lm to 10 lm; on the bottom, the value reduced from 40 lm to 10 lm. As mentioned before, without ultrasonic melt treatment, the temperature difference of the melt is high, which might result in the occurrence of remelting after the initial solidification shell forms. Thus the surface of cast ingots is coarsened. With the application of ultrasonic melt treatment, an alternating pressure is generated in the melt. Therefore, the melt will be stirred and the temperature difference of which decreases. Consequently, remelting of the initial solidification shell is suppressed and the surface quality of the cast ingots is improved. Fig. 5 shows the cross-sectional solidification structure of cast ingots with and without ultrasonic melt treatment. It is found that the solidification structure is mainly composed of coarse grains (Fig. 5(a)) and the average grain size is about 94.1 lm in the absence of ultrasonic melt treatment. After the ultrasonic melt treatment during continuous casting (P = 1000 W), the crystal grains are evidently refined, as shown in Fig. 5(b), and the average grain size reduces to 31.2 lm. It is well known that the real melt always contains vast amount of atom clusters and refractory heterogeneous
phases such as particles of metal oxides. These clusters and heterogeneous phases, as a rule, are not wettable by melt of the alloy and contain gases (mainly hydrogen) absorbed at their surface and in surface defects. However, the situation is changed and the gases start to diffuse from the melt into the cavities with the formation of cavities when ultrasonic melt treatment is applied. Some of the cavities oscillate at a frequency of the imposed ultrasonic field and the gas content inside them remains constant, while other cavities grow intensely under tensile stresses and the inward diffusion of gases dissolved in the melt. Yet another set of the cavities which are not completely filled with gas, start to collapse as the alternating compression and rarefaction phases of the sound wave pass through the melt [15]. The growth of cavities will absorb energy from the melt around and result in local undercooling, which facilitates the formation of a lot of tiny atom clusters, acting as nuclei for a-primary crystals. Moreover, the collapsing cavities will result in instantaneous high temperature, high pressure and violent impact waves in the local zone of the melt, which will also multiply the nuclei by crushing the secondary arms of a-primary crystals. Furthermore, ultrasonic melt treatment is helpful for simultaneous solidification due to the enhancement of the uniformity of melt temperature, as mentioned previously. Thus solidification structures are significantly refined and the uniformity of the grain size is improved by the application of tundish ultrasonic treatment during horizontal continuous casting.
Fig. 5. Solidification structure of Al–1%Si cast ingots: (a) P = 0, (b) P = 1000 W.
124
X.-t. Li et al. / Ultrasonics Sonochemistry 13 (2006) 121–125
Fig. 6. The distribution of Si on the cross-section of cast ingot of Al–1%Si alloy by EPMA analysis.
SR ¼
S1 ; S2
where S1 is the average content of Si in the a-matrix; S2 is the average content of Si in the grain boundary. Fig. 7 shows the relationship between the ultrasonic power and SR. It is obvious that SR decreases with the increase of ultrasonic power. When the input power is increased to 1000 W, SR decreases from 83.3 (P = 0 W) to 1.7 (P = 1000 W). This phenomenon may be theoretically interpreted by use of the Arrhenius formula: D ¼ D0 expðQ=RT Þ;
ð1Þ
where D is the diffusion coefficient, D0 is diffusion constant, R is gas constant, Q is activation energy, and T is absolute temperature. Then the value of the diffusion extent of Si, a, can be estimated by the following formula:
90 80 Segregation rate, SR
The reason for the addition of Si is to improve the mechanical properties of Al–1%Si bonding wire. It can be seen from the binary-phase diagram of Al–Si alloy that the maximum solid solubility of Si in the a-matrix is 1.65% and the solid solubility decreases sharply with the decrease of temperature. It is well known that the segregation of Si to grain boundaries will seriously deteriorates the mechanical properties of the bonding wire. EPMA analysis was adopted to investigate the distribution of Si particles in the cross-section of the cast ingot. Fig. 6(a) shows Si segregates at grain boundaries as a network without the application of ultrasonic melt treatment. By contrast, the ultrasonic melt treatment (P = 1000 W) refined the a-primary crystal and distributed Si solute uniformly over the cross-section. Most of the Si-phases in the boundary were disconnected and broken after the ultrasonic melt treatment, as shown in Fig. 6(b). In fact, the course of solid solubility is the diffusion of Si in the a-matrix. To quantitatively investigate the relationship between the ultrasonic power and boundary segregation of Si, the segregation rate (SR) is defined as
70 60 50 40 30 20 10 0
0
400
600 800 Ultrasonic power, P/W
1000
Fig. 7. The relationship between ultrasonic power and segregation rate.
a¼D
s ; l2
ð2Þ
where, s is diffusion time, namely local solidification time, l is diffusion length. When high-intensity ultrasound is imposed, on one hand, the solidification structure is refined, that is to say, the value of l decreases; on the other hand, the value of T is enhanced due to the local high temperature created by the ultrasonic physical effect, thus the value of D is enhanced, which means that the value of a is enhanced. Consequently the grain boundary segregation of Si element is suppressed. 4. Conclusions The main results in this study are as follows: (1) The application of ultrasonic melt treatment can decrease the temperature difference in the tundish. (2) The surface quality is improved by the application of ultrasonic melt treatment and cast ingots free from surface defects can be obtained when the ultrasonic power is 1000 W. (3) Ultrasonic melt treatment can significantly refine the solidification structure and suppress the boundary segregation of the Si element.
X.-t. Li et al. / Ultrasonics Sonochemistry 13 (2006) 121–125
Acknowledgements The authors gratefully acknowledge the support of the National Natural Science Foundation of China (No. 50454055, No. 50274017) and the Joint Fund of Iron and Steel Research of Baosteel Group Corporation. References [1] Mustafa R. Ozgu, Can. Metall. Quart. 3 (1996) 199. [2] Yaohe Zhou, Zhuanglin Hu, Wanqi Jie, Solidification Technology, Mechanical Industry Press, Beijing, 1998, p. 98. [3] L.J. Heaslip, A. Mclean, I.D. Sommerville, Chemical and Physical Interactions during Transfer Operations—Continuous Casting, vol. 1, ISSAIME, Warrendale, PA, 1983. [4] A. Mclean, in: Steelmaking Conf. Proc., 1988, vol. 71, pp. 3–23. [5] M.M. Wolf, in: Steelmaking Conf. Proc., 1996, vol. 79, pp. 367–381.
125
[6] Jianju Wang, Yanping Bao, Ying Qu, Tundish Metallurgy, Metallurgical Industry Press, Beijing, 2001, 1–10. [7] H. Nakanishi, Y. Tsunekawa, M. Okumiya, N. Mohri, Mater. Trans. Jpn. Inst. Metal. 34 (1993) 62. [8] H. Nakanishi, Y. Tsunekawa, M. Okumiya, N. Mohri, J. Mater. Sci. Lett. 12 (1993) 1212. [9] Y. Tsunekawa, H. Nakanishi, N. Mohri, AFS Trans. 102 (1994) 953. [10] J. Pan, D.M. Yang, X.F. Yin, H. Fukunaga, J. Mater. Res. 10 (1995) 596. [11] H. Wan, J. Pan, D.M. Yang, in: A. Pursartip, K. Street (Eds.), Proceedings of the 10th International Conference on Composite Materials, 2, Woodhead, Cambridge, 1995, p. 161. [12] Y. Genma, Y. Tsunekawa, M. Okumiya, N. Mohri, Mater. Trans. Jpn. Inst. Metal. 38 (1997) 232. [13] Andrew J.G. Strandjord, Scott Popelar, Christine Jauerning, Microelectron. Reliab. 42 (2002) 265–283. [14] S. Ramminger, N. Seliger, G. Wachutka, Microelectron. Reliab. 40 (2000) 1521–1525. [15] G.I. Eskin, Ultrason. Sonochem. 8 (2001) 319–325.
Short communication
Study of ultrasonic melt treatment on the quality of horizontal continuously cast Al–1%Si alloy Li Xin-tao *, Li Ting-ju, Li Xi-meng, Jin Jun-ze Research Center of Foundry Engineering, Dalian University of Technology, Dalian 116024, China Received 14 May 2005; accepted 26 August 2005
Abstract The fluctuation of the melt temperature in a tundish was measured during casting and experiments were conducted to investigate the effects of ultrasonic melt treatment on the surface quality and solidification structures of Al–1%Si ingots. The results show that the uniformity of melt temperature was enhanced with the application of ultrasonic melt treatment. When the ultrasonic power is 1000 W, the surface quality was evidently improved and grains of cast ingots were refined. Moreover, EPMA analysis was adopted to study the relationship between the ultrasonic power and boundary segregation of Si element. The result shows that boundary segregation is suppressed with the increase of ultrasonic power and the phenomenon was theoretically interpreted. Ó 2005 Elsevier B.V. All rights reserved. Keywords: Ultrasonic melt treatment; Al–1%Si alloy; Surface quality; Solidification structure; Boundary segregation
1. Introduction A tundish is one of the major components of horizontal continuous casters to deliver melt to the mold(s) at a controlled rate and remove non-metallic inclusions from liquid melt [1]. In horizontal continuous casting, early emphasis was placed on the mold because product quality and productivity are determined mostly by mold practices and parameters. With the development of solidification technology and clusters theory, effects of the melt quality on the final solidification structure are being addressed [2]. Tundish metallurgy is a new concept that was first put forward to improve melt quality by Mclean et al. in 1980s [3,4]. Since then the investigation of the relationship between tundish metallurgy and the product quality has been carried out, which has become a hot topic in recent years [5,6]. Ultrasound is a high frequency mechanical wave. The propagation and interaction of sound waves change the *
Corresponding author. Tel./fax: +86 0411 84708940. E-mail address: [email protected] (X.-t. Li).
1350-4177/$ - see front matter Ó 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.ultsonch.2005.08.005
physical and chemical properties of materials that are subjected to ultrasound. Over the last few years high-intensity ultrasound has been used for producing new materials successfully. For example, ultrasonic treatment is applied in melts during solidification in order to obtain fine-grain solidification microstructures [4–6]. High-intensity ultrasound has been used to fabricate continuous fibre and particulate reinforced metal matrix composites, by improving the wettability or promoting the chemical reaction between the reinforcement and the metal matrix [7–12]. Unfortunately, so far, there have been few reports on the application of high-intensity ultrasonic to the melt treatment in tundish (which is abbreviated to ultrasonic melt treatment in the following) during horizontal continuous casting. Nowadays fine wire bonding is the most popular interconnection method used in packaging electronic devices for electrical connection between the integrated circuit blocks and the external frame and Al–1%Si alloy is the main raw material used to manufacture bonding wire [13,14]. In this paper, the novel processing of ultrasonic melt treatment is proposed to improve the quality of a horizontal continuously cast ingot of Al–1%Si alloy. The
122
X.-t. Li et al. / Ultrasonics Sonochemistry 13 (2006) 121–125
influence of ultrasonic melt treatment on the temperature fluctuation in the tundish, quality of cast ingots such as surface roughness and solidification structure are presented, and discussed. 2. Experimental A schematic view of the experimental apparatus is shown in Fig. 1. The primary mold consists of a copper outer-jacket and a graphite inner-mold. The ultrasonic emitter, made of stainless steel and coated with TiN membrane, was inserted into the melt from the top of the tundish and connected to a generator with a maximum power capacity of 1 kW and frequency of 22.3 kHz. A secondary mold is immediately connected to the primary one, in which cooling water is distributed uniformly over the periphery of the ingot. The experimental conditions are summarized in Table 1.
Table 1 Experimental conditions Items
Parameters
Casting temperature Water flux Water temperature Casting speed Diameter of ingot
720 °C 40 L/h 20 °C 100 mm/min 10 mm
3. Results and discussion The fluctuation of the melt temperature in the tundish is measured by thermocouples during casting, the positions of which are shown in Fig. 1. Fig. 2 shows that melt temperature declines from position A to position C in the absence of ultrasonic melt treatment and the maximum temperature difference is 11 °C. By contrast, the temperature distributes more uniformly in the tundish after the application of ultrasonic melt treatment, and the maximum temperature difference declines to 2 °C. Fig. 3 indicates the surface of cast ingots with and without the application of ultrasonic melt treatment. It is clearly observed that ripple marks in the longitudinal direction are formed, especially on the bottom side without the application of ultrasonic melt treatment, as seen in Fig. 3(a) and (b). Whereas, when ultrasonic melt treatment was applied with the power of 1000 W, the ripple marks
735
P=0W
Melt temperature,T/ºC
A
Fig. 1. Top: schematic view of experimental apparatus: (1) melt, (2) thermocouples, (3) tundish, (4) ultrasonic emitter, (5) ultrasonic transducer, (6) temperature-gathering equipment, (7) primary mold, (8) secondary mold, (9) ingots. Bottom: top view showing the thermocouple positions for the test.
730
B
P=1000W
725
C
720 715 710 20
40 80 60 Length of the tundish/mm
100
Fig. 2. Distribution of temperature in tundish at h = 40 mm (point A, B, C) corresponding to the thermocouples positions shown in Fig. 1.
Fig. 3. Surface view of cast ingots in the absence and presence of ultrasound. (a) P = 0 W, upper part view, (b) P = 0 W, bottom view, (c) P = 1000 W, upper part view, (d) P = 1000 W, bottom view.
20
P=1000W
1
2 3 4 5 6 7 Length of measured ingot/mm
8
9
0 -10
Surface roughness/µm 0
-30
-30 a
10 b
P=1000W
10
P=0W
Surface roughness/µm -20 -10 0 10
P=0W
123
-20
20
X.-t. Li et al. / Ultrasonics Sonochemistry 13 (2006) 121–125
0
1
7 8 3 4 2 5 6 Length of measured ingot/mm
9
10
Fig. 4. Surface roughness tested by a displacement sensor with and without the application of ultrasonic melt treatment: (a) upper part, (b) bottom side.
disappeared completely and the surface of the ingot is smooth enough to be rolled without the ÔscalpingÕ operation that is required following traditional casting. Fig. 4 presents the surface roughness of the cast ingots measured by a displacement sensor in the two cases. It can be seen the surface roughness of the upper part reduced from 30 lm to 10 lm; on the bottom, the value reduced from 40 lm to 10 lm. As mentioned before, without ultrasonic melt treatment, the temperature difference of the melt is high, which might result in the occurrence of remelting after the initial solidification shell forms. Thus the surface of cast ingots is coarsened. With the application of ultrasonic melt treatment, an alternating pressure is generated in the melt. Therefore, the melt will be stirred and the temperature difference of which decreases. Consequently, remelting of the initial solidification shell is suppressed and the surface quality of the cast ingots is improved. Fig. 5 shows the cross-sectional solidification structure of cast ingots with and without ultrasonic melt treatment. It is found that the solidification structure is mainly composed of coarse grains (Fig. 5(a)) and the average grain size is about 94.1 lm in the absence of ultrasonic melt treatment. After the ultrasonic melt treatment during continuous casting (P = 1000 W), the crystal grains are evidently refined, as shown in Fig. 5(b), and the average grain size reduces to 31.2 lm. It is well known that the real melt always contains vast amount of atom clusters and refractory heterogeneous
phases such as particles of metal oxides. These clusters and heterogeneous phases, as a rule, are not wettable by melt of the alloy and contain gases (mainly hydrogen) absorbed at their surface and in surface defects. However, the situation is changed and the gases start to diffuse from the melt into the cavities with the formation of cavities when ultrasonic melt treatment is applied. Some of the cavities oscillate at a frequency of the imposed ultrasonic field and the gas content inside them remains constant, while other cavities grow intensely under tensile stresses and the inward diffusion of gases dissolved in the melt. Yet another set of the cavities which are not completely filled with gas, start to collapse as the alternating compression and rarefaction phases of the sound wave pass through the melt [15]. The growth of cavities will absorb energy from the melt around and result in local undercooling, which facilitates the formation of a lot of tiny atom clusters, acting as nuclei for a-primary crystals. Moreover, the collapsing cavities will result in instantaneous high temperature, high pressure and violent impact waves in the local zone of the melt, which will also multiply the nuclei by crushing the secondary arms of a-primary crystals. Furthermore, ultrasonic melt treatment is helpful for simultaneous solidification due to the enhancement of the uniformity of melt temperature, as mentioned previously. Thus solidification structures are significantly refined and the uniformity of the grain size is improved by the application of tundish ultrasonic treatment during horizontal continuous casting.
Fig. 5. Solidification structure of Al–1%Si cast ingots: (a) P = 0, (b) P = 1000 W.
124
X.-t. Li et al. / Ultrasonics Sonochemistry 13 (2006) 121–125
Fig. 6. The distribution of Si on the cross-section of cast ingot of Al–1%Si alloy by EPMA analysis.
SR ¼
S1 ; S2
where S1 is the average content of Si in the a-matrix; S2 is the average content of Si in the grain boundary. Fig. 7 shows the relationship between the ultrasonic power and SR. It is obvious that SR decreases with the increase of ultrasonic power. When the input power is increased to 1000 W, SR decreases from 83.3 (P = 0 W) to 1.7 (P = 1000 W). This phenomenon may be theoretically interpreted by use of the Arrhenius formula: D ¼ D0 expðQ=RT Þ;
ð1Þ
where D is the diffusion coefficient, D0 is diffusion constant, R is gas constant, Q is activation energy, and T is absolute temperature. Then the value of the diffusion extent of Si, a, can be estimated by the following formula:
90 80 Segregation rate, SR
The reason for the addition of Si is to improve the mechanical properties of Al–1%Si bonding wire. It can be seen from the binary-phase diagram of Al–Si alloy that the maximum solid solubility of Si in the a-matrix is 1.65% and the solid solubility decreases sharply with the decrease of temperature. It is well known that the segregation of Si to grain boundaries will seriously deteriorates the mechanical properties of the bonding wire. EPMA analysis was adopted to investigate the distribution of Si particles in the cross-section of the cast ingot. Fig. 6(a) shows Si segregates at grain boundaries as a network without the application of ultrasonic melt treatment. By contrast, the ultrasonic melt treatment (P = 1000 W) refined the a-primary crystal and distributed Si solute uniformly over the cross-section. Most of the Si-phases in the boundary were disconnected and broken after the ultrasonic melt treatment, as shown in Fig. 6(b). In fact, the course of solid solubility is the diffusion of Si in the a-matrix. To quantitatively investigate the relationship between the ultrasonic power and boundary segregation of Si, the segregation rate (SR) is defined as
70 60 50 40 30 20 10 0
0
400
600 800 Ultrasonic power, P/W
1000
Fig. 7. The relationship between ultrasonic power and segregation rate.
a¼D
s ; l2
ð2Þ
where, s is diffusion time, namely local solidification time, l is diffusion length. When high-intensity ultrasound is imposed, on one hand, the solidification structure is refined, that is to say, the value of l decreases; on the other hand, the value of T is enhanced due to the local high temperature created by the ultrasonic physical effect, thus the value of D is enhanced, which means that the value of a is enhanced. Consequently the grain boundary segregation of Si element is suppressed. 4. Conclusions The main results in this study are as follows: (1) The application of ultrasonic melt treatment can decrease the temperature difference in the tundish. (2) The surface quality is improved by the application of ultrasonic melt treatment and cast ingots free from surface defects can be obtained when the ultrasonic power is 1000 W. (3) Ultrasonic melt treatment can significantly refine the solidification structure and suppress the boundary segregation of the Si element.
X.-t. Li et al. / Ultrasonics Sonochemistry 13 (2006) 121–125
Acknowledgements The authors gratefully acknowledge the support of the National Natural Science Foundation of China (No. 50454055, No. 50274017) and the Joint Fund of Iron and Steel Research of Baosteel Group Corporation. References [1] Mustafa R. Ozgu, Can. Metall. Quart. 3 (1996) 199. [2] Yaohe Zhou, Zhuanglin Hu, Wanqi Jie, Solidification Technology, Mechanical Industry Press, Beijing, 1998, p. 98. [3] L.J. Heaslip, A. Mclean, I.D. Sommerville, Chemical and Physical Interactions during Transfer Operations—Continuous Casting, vol. 1, ISSAIME, Warrendale, PA, 1983. [4] A. Mclean, in: Steelmaking Conf. Proc., 1988, vol. 71, pp. 3–23. [5] M.M. Wolf, in: Steelmaking Conf. Proc., 1996, vol. 79, pp. 367–381.
125
[6] Jianju Wang, Yanping Bao, Ying Qu, Tundish Metallurgy, Metallurgical Industry Press, Beijing, 2001, 1–10. [7] H. Nakanishi, Y. Tsunekawa, M. Okumiya, N. Mohri, Mater. Trans. Jpn. Inst. Metal. 34 (1993) 62. [8] H. Nakanishi, Y. Tsunekawa, M. Okumiya, N. Mohri, J. Mater. Sci. Lett. 12 (1993) 1212. [9] Y. Tsunekawa, H. Nakanishi, N. Mohri, AFS Trans. 102 (1994) 953. [10] J. Pan, D.M. Yang, X.F. Yin, H. Fukunaga, J. Mater. Res. 10 (1995) 596. [11] H. Wan, J. Pan, D.M. Yang, in: A. Pursartip, K. Street (Eds.), Proceedings of the 10th International Conference on Composite Materials, 2, Woodhead, Cambridge, 1995, p. 161. [12] Y. Genma, Y. Tsunekawa, M. Okumiya, N. Mohri, Mater. Trans. Jpn. Inst. Metal. 38 (1997) 232. [13] Andrew J.G. Strandjord, Scott Popelar, Christine Jauerning, Microelectron. Reliab. 42 (2002) 265–283. [14] S. Ramminger, N. Seliger, G. Wachutka, Microelectron. Reliab. 40 (2000) 1521–1525. [15] G.I. Eskin, Ultrason. Sonochem. 8 (2001) 319–325.