Resistivity Evolution with Temperature of 7050 Aluminium Alloy in Different Casting Methods

Materials Transactions, Vol. 49, No. 10 (08) pp. 2283 to 2287 #08 The Japan Institute of Metals Resistivity Evolution with Temperature of 7050 Aluminium Alloy in Different Casting Methods Chunyan Ban* 1, Fei Fan* 2, Yang Wang* 2, Qixian Ba and Jianzhong Cui The Key Lab of Electromagnetic Processing of Materials, Ministry of Education, Northeastern University, Shenyang, 110004, P.R. China The resistivity-temperature curve of 7050 aluminum alloy prepared in different casting methods ( and ) has been measured with D.C. four-probe method. By analyzing the curve, it was found that a change occurred to the slope at 250 C when the samples were heated, and the resistivity of sample changed with temperature more quickly than that of. Both the solidus and liquidus temperatures of samples were higher than those of samples during the process from room temperature to 900 C. However, the resistivity of samples increased remarkably while the temperature was kept at 900 C, and even before it was decreased to 600 C. With the microstructure observation, the characteristics and phenomenon of the resistivity evolution with temperature were studied. [doi:10.23/matertrans.mra07287] (Received November 21, 07; Accepted June 23, 08; Published September 10, 08) Keywords: aluminum alloy, low frequency electromagnetic casting, resistivity, Lorentz force 1. Introduction Super-high strength Al alloy (Al-Zn-Mg-Cu alloy) is widely used in aeronautical and automotive applications, due to its high strength, low density and good fracture toughness. 1) There are more demands for combining property of high strength and high tough Al alloy in modern industry, thus exploiting and developing super-high strength Al alloy is crucial. Constant efforts are being made to improve the mechanical and corrosion properties of these alloys for a particular application. 2) Low frequency electromagnetic casting () is a new casting technique. It was developed from the CREM (Casting, Refining, Electromagnetic) process by Cui Jianzhong etc. In this process, AC induction coil are arranged around the crystallizer of conventional casting, inletting low frequency current (less than 50 Hz), then Al alloy ingots are prepared by using semi-continuous. Comparing with direct chilling casting (), could not only obtain fine and uniform microstructure and good surface quality which gained by CREM, but also remarkably increase the alloy element content inside grain, enhance the ability of resisting cracking, decrease macrosegregation, and improve the quality of ingots. 3) Resistivity is a sensitive parameter to the structural changes of alloy. By analyzing the influencing factors of resistivity detailedly, we can know the internal structure evolution of alloy. 4,5) Resistivity is applied to the study of liquid metal, because the resistivity has an abrupt variety at liquidus or solidus temperature. Furthermore when the temperature is higher than the liquidus temperature, the structure changes of liquid metal can be reflected sensitively by resistivity variance. 6,7) Therefore, according to the study of resistivity-temperature (-T) curve, we may obtain a lot of important information of structure evolutions with temperature in the liquid and solid metal. * 1 Corresponding author, E-mail: bancy@epm.neu.edu.cn * 2 Graduate Student, Northeastern University The present investigation was aimed at studying -T curves of 7050 Al alloy ingots which were prepared in different casting methods ( and ) in the range from room temperature to 500 C and 900 C, respectively. With the microstructure observation, the reason of the difference of -T curves between the two category ingots was studied. 2. Experimental 7050 Al alloy was made up according to the following component(quality percentage): Zn6.5%, Cu2.5%, Mg2.5%. The samples were cut from the ingot which was cast in the same charging material, but the casting technique is different. The upper part was cast by while the lower part was cast by method. Casting temperature was 725 7 C, casting speed was 55 mm/min, and ingots were cooled at a cooling rate of 90 L/min by water. The coil electrical current was maintained at 160 A with frequency of Hz. The diameter of ingots was 1 mm. A 140 mm long part was cut off from the center section of ingot, the part of edge was removed, from which the square materials with 10 10 mm cross section were prepared, and the cylinder samples (6 90 mm) were machined. One group of the samples were heated from room temperature to 500 C at a heating rate of 2 C/min, held at this temperature for 10 minutes, then cooled to room temperature at the same rate. The resistance of the sample was measured at the temperature interval for every 5 C during the heating and cooling processes. Another group of the samples were heated from room temperature to 900 Cata heating rate of 5 C/min, held at this temperature for 10 minutes, then cooled to room temperature at the same rate. In the same way, the resistance of the sample was measured at the temperature interval for every 5 C. With D. C. four-probe method, the resistances of samples were measured by SB- 22 direct current number resistance meter. Samples with mm in length were cut at the middle part, abraded with abrasive paper, first polished by machine,

2284 C. Ban, F. Fan, Y. Wang, Q. Ba and J. Cui 2mm 2mm (c) (d) Fig. 1 Casting macrostructure and microstructure of 7050 aluminum alloy (c) (d). Resistivity, ρ/ µ Ω cm 10 8 6 10 8 6 4 0 100 0 0 400 500 Temperature,T/ C 4 0 100 0 0 400 500 Temperature,T/ C Fig. 2 -T curve of 7050 aluminum alloy from room temperature to 500 C. heating curve cooling curve. then polished with shammy carefully, at last etched with mixed acid. The microstructures were observed by an optical microscope, Leica DMR. The solute distribution from the center to edge region of the ingot was detected by chemical analysis. The instrment used for chemical analysis is Inductively Coupled Plasma Optical Emission Spectrometer (ICP-OES), Optima 40DV. 3. Result Figure 1 and 1 show cast macrostructure of 7050 Al alloy in different casting methods. It indicates that the macrostructure of sample is finer than that of, and there are massive columnar crystals in sample, while the grains of sample are globular. Figure 1(c) and 1(d) show their microstructure respectively. The grain boundaries of sample are thinner than those of,the different thickness of grain boundaries is made by different casting methods. The -T curves of samples heated from room temperature to 500 C are given in Fig. 2. Figure 2 is -T curve of the heating process. The curve is approximately made up of two linear regions. The slope changed at about 250 C. Figure 2 shows the -T curve of cooling process which almost decrease linearly with single slope coefficient. Both in the heating curve and cooling curve, the -T curves of samples are higher than those of, and change more rapidly along with the temperature. Comparing the resistivity of room temperature in Fig. 2 and 2, it was discovered that the resistivitiy of samples which experienced heating and cooling processes are less than those of the original. Microstructure of the measured samples is showed in Fig. 3. The coarse grain boundaries turned into pin stripe and became discontinuous, and there are large numbers of needlelike precipitates inside the grains. The original cast samples were heated from room temperature to 400 C at a rate of 2 C/min, held at this temperature for minutes, then water quenching. The microstructure is illustrated in Fig. 4 and 4. Compared with Fig. 1(c) and 1(d), the grain boundaries became thin, because the solute components on the grain boundary dissolved into -Al matrix. Contrasted Fig. 3 and 4, the grain boundaries of the former are thinner, because more solute on the grain boundary dissolved into -Al matrix with the temperature ascending to 500 C, however the latter has no precipitates from the intracrystalline, thus the precipitates formed while the temperature fell to lower temperature. Figure 5 illustrates the heating and cooling curves from room temperature to 900 C. In the course of heating to 900 C, the resisitivity of samples is higher than that

Resistivity Evolution with Temperature of 7050 Aluminium Alloy in Different Casting Methods 2285 Fig. 3 Microstructure of 7050 aluminum alloy after measuring resistivity from room temperature to 500 C.. Fig. 4 Microstructure of 7050 aluminum alloy quench at 400 C.. 25 15 10 5 50 40 0 0 400 600 800 1000 0 0 400 600 800 1000 Temperature,T/ C Temperature,T/ C 10 Fig. 5 -T curves of 7050 Al alloy from room temperature to 900 C. heating curve cooling curve. Table 1 Solidus and liquidus temperature of and samples. casting method solidus temperature/ C liquidus temperature/ C 565 650 575 660 of. The -T curve is linear from room temperature to 250 C, the -T curve is still linear above 250 C, but the slope is somewhat large. Above around 570 C, the -T curve goes up rapidly, and the curve goes smooth at about 650 C. The former is the solidus temperature of Al alloy, the latter is the liquidus temperature. The solidus and liquidus temperatures of and are listed in Table 1. The solidus and liquidus temperatures of are higher than those of. There should be an error about the solidus and liquidus temperatures obtained by this method, in this paper, we only want to compared the solidus and liquidus temperatures of and samples relatively. The -T curve goes up rapidly again at about 880 C, this change is likely be concerned with the solution of the solute clusters, so the resistivity abrupt. 7) Figure 5 is the cooling curve, the resistivity is upward from 900 C to 600 C, and the Fig. 6 70 60 50 40 0 40 60 80 100 1 time, t /min resistivity of samples is upward more obviously. The resistivity decreased monotonically, when the temperature dropped to 600 C. Unlike the heating curve, the -T curve of samples is higher than that of from 900 Cto room temperature. To know the evolution of resistivity in the course of thermal preservation and cooling, the sample was heated to 900 C, held at this temperature for an hour, then cooled to room temperature. The change of and T with time t are shown in Fig. 6. The resistivity increased all along while the 900 800 700 600 500 400 0 ðtþ-t curve from room temperature to 900 C of sample. /C T

2286 C. Ban, F. Fan, Y. Wang, Q. Ba and J. Cui temperature increased and held at 900 C, and even before it decreased to 600 C. The resistivity began to decrease when the temperature droped below 600 C. 4. Discussions As we know, under the effects of the alternating current, the inductor generates a variable magnetic field in the melt, which in turn gives rise to an induced current. Thus, the melt is subject to electromagnetic body forces caused by the interaction of the eddy currents J and the magnetic field B. Lorentz force can be solved by Maxwell s electromagnetic equations as follows: f ¼ J B ¼ 1=ðB rþb 1=ð2ÞrB 2 ð1þ where is magnetic permeability. Considering Al is a weak magnetic substance, ¼ 0 ¼ 4 10 7 H/m. The first term on the right side of eq. (1) is a rotational component which results in a forced convection and a flow in the melt, it brings intracrystalline solute of alloy to increase solidified under AC magnetic field. 3,8) The second term on the right side of eq. (1) is an irrotationality force f ¼ 1=ð2 0 ÞrB 2 ð2þ because there is a relation between B and H, eq. (2) is written as follows: f ¼ 0 =2ð1 þ ÞrH 2 ð3þ where is magnetic susceptibility. Because the non-uniform distribution of magnetic line along longtitudinal direction and the skin effect of metal, there are the longitudinal component r == H 2 and the transverse component r? H 2 about rh 2. Correspondingly the irrotationality force is broken up the longitudinal component f == and the transverse component f?. Considering the magnetic line distribution in coil along the axis of the ingot roughly, the direction of axis is z axis, a column coordinate system ðr;;zþ is built. The transverse component is f? ¼ 0 =2ð1 þ Þ @ @r H2 Its direction is the radial of ingot. The difference between Lorentz force acting on the -Al matrix f?b and the impurity phase f?p is written as 9) f ¼ f?p f?b ¼ 0 =2ð b p Þ @ @r H2 where b and p are magnetic susceptibility of -Al matrix and the impurity phase, respectively. It can be deduced that if p > b, the solute would be subject to a magnetic force which points to axis, so the solute would move towards the center of the ingot. If p < b, the solute would move towards the edge of ingot. There are mainly Zn, Cu and Mg in 7050 Al alloy, their magnetic susceptibility is less than that of Al, so they would move towards the edge of ingot. We took samples from the center to edge of and ingots, the chemical analysis is shown in Fig. 7. In both methods, there are more solutes in the edge than in the center. The ingots have less Zn, Cu and Mg in the center while more Zn, Cu and Mg in the edge than the ð4þ ð5þ Solute content(%) Fig. 7 6.3 6.2 6.1 6.0 2.4 2.3 2.2 2.1 0 10 40 50 60 Distance from the center of ingot, D/mm The solute distribution in and casting ingots. ingots. The chief reason of the difference of resistivity between the and samples is not due to the difference of component, but due to the difference of solute content inside the grains. It is necessary to indicate that the solute component which was determined by chemical analysis is the total solute content of intracrystalline and grain boundaries. The solute of intracrystalline includes the solute which dissolves in the solid solution and which is out of the solid solution. When the temperature is invariable, the consistency of the solute which dissolves in the solid solution determines the resistance value of the solid solution, so it determines the resistance value of the alloy which has low solute content. In Fig. 2 and Fig. 5, the electrical resistivity of sample is almost as same as that of sample at room temperature, which indicates their saturation solid solubility are approximative at room temperature. The electromagnetic stirring can refine grain, so it leads to the difference of grain size in the Fig. 1 and 1. Moreover the electromagnetic stirring can increase the intracrystalline solute content. As shown in Fig. 1(c) and 1(d), the grain boundary of sample is much thinner than that of sample because the intracrystalline solute content of sample is increased, thus the precipitates on the grain boundary is decreased. The resistance of alloy which has low solute content depends on the resistance of solid solution. 10) The resistivity of solid solution is ¼ i þ m ðtþ ð6þ where i is relate to the content of solute, m ðtþ is the resistance of dissolvant, depending on the crystal lattice scattering, it is relate to the temperature T. When the temperature is higher than room temperature, there is a linear relation between T and m m ðtþ ¼CT where C is a constant. The impurity scattering would generate when the solute dissolved into solid solution, it is another source of resistance of metal. When the solute consistency n i is smaller, there is an approximate relation between i and n i i ¼ Kn i ð8þ where K is the proportion constant. Zn Mg Cu ð7þ

Resistivity Evolution with Temperature of 7050 Aluminium Alloy in Different Casting Methods 2287 Because of the electromagnetism stirring, the intracrystalline solute content of sample is increased, and the solute content of -Al solid solution is increased, too. 8) So the -T curve of sample is higher than that of sample when the samples were heated from room temperature to 500 C and 900 C, respectively. The solute consistency of - Al solid solution is almost invariable when the temperature is lower. With the temperature ascending, the vibration of crystal lattice enhanced, the resistivity is increased linearly. When the temperature is above 250 C, perhaps the solubility of the solute in -Al is increased, here not only the crystal lattice scattering but also the impurity scattering increase along with the temperature increase, so there is a turning point at 250 C in the -T curve. It is noticeable that the solute on intracrystalline is easier to dissolve into -Al than that on grain boundary. The intracrystalline solute content of sample is higher, so the increases rapidly with the temperature increase. The measuring process from room temperature to 500 Cis actually a process that heating and cooling the samples slowly. The solubility which the solute dissolves into -Al is increased at a high temperature, which leads to the resistance increase. The solute would precipitate from -Al with the temperature falling. Because the cooling rate of measuring process is much lower than that of the casting, there is more chances for the solute to precipitate from -Al, so the intracrystalline solute content of samples after measuring is lower than that of the original ingots, the resistivity decreases after measuring. It is approved that the eq. (6) to eq. (8) are still fit for liquid metal. 11) At about 8 C, the resistivity of liquid alloy increased obviously, it is estimated that, at this temperature the solute clusters began to dissolve, more solute dissolved into liquid Al. Because the solute content of sample is higher than those of sample, the total solute content is higher in the liquid Al, which results in the of sample exceeds that of sample above 880 C, and the resistivity is upward when the temperature held at 900 C, and even before cooled to 600 C. Although the solute precipitated continuously from -Al from 900 C to room temperature, but the solubility of sample is higher than that of sample all along, so the -T curve of sample is higher than that of sample from 900 Cto room temperature. 5. Conclusions (1) 7050 Al alloy ingots were prepared in two different methods ( and ), there were two linear regions in the -T curve, and the resistivity of samples changed rapidly. (2) The -T curves abrupted at the solidus and liquidus temperature when the samples were heated from the room temperature to 900 C, and at about 880 C the curves abrupted again. The heating curve of sample was higher than that of sample; while the cooling curve of sample was higher than that of sample. (3) The reason of the above result is that makes the Zn, Cu and Mg move towards the edge of ingot and makes solid solubility of the sample increase. Acknowledgements This research was supported by the National Natural Science Foundation of China (No. 506740, 50404007). REFERENCES 1) K. S. Alrubaie, M. A. Delgrande, D. N. Travessa and K. R. Cardoso: Mater. Sci. Eng. A 464 (07) 141 150. 2) M. Dixit, R. S. Mishra and K. K. Sankaran: Mater. Sci. Eng. A 464 (07) 1 2. 3) B. J. Zhang, J. Z. Cui and G. M. Lu: Journal of Mater. Sci. Technol. 18 (02) 1 3. 4) H. X. Chen: The physics capability and measurement method of metal, (The metallurgy industry publishing company, Beijing, 1987). 5) M. R. Corfield, I. R. Harris and A. J. Williams: Journal of Magnetism and Magnetic Materials 316 (07) 59 66. 6) X. F. Li, F. Q. Zu, H. F. Ding, J. Yu, L. J. Liu and Y. Xi: Phy. Letters A354 (06) 325 329. 7) Q. Wang, J. Jia, P. J. Li and P. L. Wu: Casting 3 (1998) 7 9. 8) C. Y. Ban, Q. X. Ba, J. Z. Cui and G. Y. Zeng: Journal of Northeastern University (Natural Science) 23 (02) 799 782. 9) Y. Han, C. Y. Ban, H. T. Zhang, H. Nagaumi, Q. X. Ba and J. Z. Cui: Mater. Trans. 47 (06) 92 98. 10) J. X. Fang and D. Lu: Physics of solid state, (ShangHai, Science and Technology Press in ShangHai, 1980) pp. 4 7. 11) M. Shimoji: Liquid metals an introduction to the physics and chemistry of metals in the liquid state, (London, Academic Press, 1977) pp. 251 281.