Sharif University of Technology, Material Science and Engineering Department Tehran, Iran.

Transcription

1 Microstructural Investigation of Semisolid Aluminum A356 Alloy Prepared by the Combination of Electromagnetic Stirring and Gas Induction Nora Nafari 1,a, Farnoush Yekani 1,b, Hossein Aashuri 1,c* 1 Sharif University of Technology, Material Science and Engineering Department Tehran, Iran. a nora.nafari@gmail.com, b farnoush.yekani@yahoo.com, c ashuri@sharif.edu Keywords: Semi-Solid Process, Electromagnetic Stirring, Gas Induced Semi-Solid (GISS), Globular Microstructure, A356 Alloy. Abstract. A three phase electromagnetic stirrer was used to agitate aluminum A356 slurry and a dry and oxygen free argon gas was introduced in to the slurry by a porous graphite core at a same time. The prepared semi-solid slurry was then transferred into a metallic mold and was compacted by a drop weight. Results demonstrated a favorable increase in shape factor, decrease in aspect ratio and average diameter size at different intensities of stirring. The intensity of stirring was changed by altering the current passed through the magnetic coil and also bubbling intensity via the porous graphite diffuser. Different time intervals for electromagnetic stirring and gas induction were applied. Agitating the slurry for 90 Sec. separately by electromagnetic stirrer and GISS method, gave better results in terms of shape factor, decrease in average diameter of the globules and aspect ratio. Introduction Semi-solid metal (SSM) casting is being used as a method of producing parts with non-dendritic structure since its invention in MIT by Spencer et al. [1]more than 40 years ago. It is known that Aluminum alloys, especially A356, have been the subject of SSM experimental studies because of its wide solid-liquid range and superior fluidity [2]. As Martinez and Flemings [3] stated, Several methods of SSM were invented in order to obtain a globular structure through agitating the molten metal, causing dendrite fragmentation, spreading them throughout the melt and subsequently forming new grains. According to Fan [4], of all the methods suggested for SSM, electromagnetic stirring (EMS), which was developed by Vives in 1981, has become wildly accepted on a commercial scale. Moffatt [5] demonstrated that when an alternating magnetic field is applied to a conductor, it will induce Lorentz force distribution that is generally rotational. If the conductor is fluid, it is set in motion. However as Tzimas and Zavaliangos mentioned [6], EMS method suffers nonhomogeneous microstructure in radial direction and presence of non-globular primary phases. In an effort to reduce the shortcomings of EMS, Zhu et al. [7] developed a new process called annular electromagnetic stirring (A-EMS). It was shown that by eliminating the central part of the EMS chamber, a more uniform and grain refined structure is formed compared to normal EMS. In a computational modeling study of electromagnetic stirring by Xing-run et al. [8], it was demonstrated that in the central part of the electromagnetic stirring chamber, the electromagnetic force was at its lowest. Therefore by eliminating the central part, the magnetic and temperature fields were more uniformly distributed in this narrow gap and thus a more uniform structure is obtained. A two-dimensional computational model of EMS has also been developed by Vanluu et al. [9] to investigate the effects of process parameters on temperature and flow fields. They concluded that increasing stirring frequency and current will increase electromagnetic field and flow field density, but will not cause a uniform distribution of grains.

2 A method of SSM was suggested in 2006 by Wannasin et al. [10]. Gas induced semi solid (GISS) is a process in which molten metal is agitated by introducing gas bubbles. Different configurations for such purpose, mainly a graphite diffuser, a nozzle and a copper chill were studied. An optimal range of gas flow was determined. The localized chill effect was used to heighten the cooling rate of the melt and to produce mother dendrites. According to Wannasin et al. [11] vigorous convection causes these dendrites to breakup and form new grains in the melt. Subsequently a highly grain refined globular structure is obtained. GISS process is reportedly developed for industrial scale applications [12]. Microstructure of A380 aluminum alloy was studied by Shabestari et al. [13], using GISS process. Non-dendritic structure containing fine globular structure and well distributed α-al phase was reported. In the present paper, a combination of A-EMS and GISS methods were used in order to obtain a grain refined, globular structure. A bubbling device was located in the central part of the stirring chamber in order to introduce gas bubble into the melt. Two types of bubbling device were used in this investigation. The first one was consisted of a hollow body cylinder made of 316 stainless steel contained a pure iron bar as a core. Small holes were made in the bottom part of the cylinder in order to introduce gas bubbles into the melt [14]. The second one consisted of a graphite diffuser which is explained in the following section. Experimental Aluminum A356 alloy was prepared using standard ingots. The chemical composition of the alloy is shown in table 1. Table 1. Chemical composition of aluminum A356 alloy Element Si Fe Cu Mn Mg Zn Ti Al Wt. Percent ± remaining In order to investigate the effects of electromagnetic field on the melt, an electromagnetic system with a three-phase AC electromagnetic stirrer and a stainless steel container was used. Also a graphite diffuser was made in order to introduce gas bubbles into the melt. A schematic of the diffuser is shown in figure 1. Fig. 1. Graphite diffuser.

3 Shape Factor The Argon gas was dehydrated by passing through a moisture trap and oxygen was removed from the gas before entering the molten metal. Schematic diagram of the gas preparation system is shown in figure 2. The consisting of silica gel was used for the removal of trace moisture in the gas. The oxygen-removal furnace was filled with the copper turnings and operated at 450 C. Fig. 2. Schematic diagram of gas preparation system [15]. The alloy was melted in a resistance furnace at 750 C and poured into the EMS container. The flow of the induced gas were 2.5 Lit/min and kept constant for all experiments. Agitating the molten metal was done using two different procedures. In the first one, annular electromagnetic stirring (A- EMS) and GISS methods were applied simultaneously for 90 seconds using four different currents: 45, 60, 75 and 90A. The 90Sec. was the maximum time to keep the slurry fluid. In the second procedure, GISS and subsequently A-EMS were applied, 45 seconds each for the four different currents as formerly mentioned. In order to eliminate the middle part of the EMS chamber, the apparatus for gas introduction was not removed after the predetermined time for GISS was over. The prepared semi-solid slurry was then transferred into a die and was squeezed by a 50Kg drop device released from 50cm height. Optical microscopy was carried out on the unetched samples. Shape factor, aspect ratio and average diameter were determined by using an image analyzer (Clemex). Results and Discussion As it can be seen in figure 3, shape factor results obtained from separate stirring mode are always higher than that of the simultaneous stirring. The reason for this can be found in interference of electromagnetic stirring and gas induction for circulating the particles in the slurry. It also can be deducted from figure 4 that while for currents equal to 75 and 90A the aspect ratio values are lower for separated stirring mode compared to simultaneous stirring mode. According to figure 5 average diameter in separated stirring is lower at 45A in comparison with simultaneous stirring mode, whereas obtained values for the other tests are very close to each other Current(A) 45GISS+45EMS 90 Simul Fig. 3. Shape factor vs. current for two stirring modes.

4 Average Diameter(µm) Aspect Ratio When the processes were applied to the slurry simultaneously, the agitation resulting from electromagnetic stirring may cause the gas bubbles to spread throughout the melt and therefore not colliding with diffuser wall and defragmenting the mother dendrites. But when the gas induction was applied first, the dendrites which are normally formed on the diffuser wall are detached, therefore new nucleus are created and a finer structure is formed Current(A) 45GISS+45EMS 90 Simul Fig. 4. Aspect ratio vs. current for two stirring modes Current(A) 45GISS+45EMS 90 Simul Fig. 5. Average Diameter vs. current for two stirring modes. By observing metallographic pictures of figure 6, homogeneity in microstructure can be discussed. As it can be seen, while stirring is applied separately, the microstructure exhibits more homogeneity. More globules can be seen in figure 6-b compared with figure 6-a. Figures 6-c, 6-d. show that globularity is increased when the current is increased from 45 to 90A while the stirring time was constant. A combination of dendrites and globules can be seen in the figure 6-a, whereas figure 6-d shows mostly globular structure. The reason for increasing diameter of globules with increasing gas induction time can be investigated in the fact that gas induction causes less agitation in comparison with radial stirring caused by electromagnetic field.

5 a b c d Fig. 6. Samples of A356 alloy poured at 750 C prepared with a combination of electromagnetic stirring and gas induction methods with different currents in a 90 second time interval: a)75a-giss and EMS applied simultaneously, b)75a-giss and EMS applied separately 45 seconds each, c) 45A-GISS and EMS applied separately 45 seconds each, d) 90A-GISS and EMS applied separately 45 seconds each. Conclusion Shape factor results demonstrated better globularity for separate stirring mode in comparison with simultaneous stirring mode. Also aspect ratio results showed an improvement while stirring separately, particularly for 75 and 90 amperes. Average diameter had lower values in separate stirring mode. 45 seconds of gas induction applied before electromagnetic stirring resulted in improved microstructure mainly for aspect ratio in 75 and 90A. Increasing current caused an increase in average diameter due to agglomeration of particles while the stirring mode is separated, whereas gas bubbles prevented agglomeration in simultaneous stirring of the slurry. Also increasing gas induction time caused the microstructure to be more homogeneous. Acknowledgement The authors would like to thank financial support of the Sharif University Deputy Research Council and the valuable support of the technical staff at the university. Special thanks are also due to Mr. Alavi, Mr. Mansoori and Mr Ghaderi for their technical assistant to this work.

6 References [1] D. B. Spencer, R. Mehrabian, M. C. Flemings, Theological Behavior of Sn-15%Pct Pb in the Crystallization Range. Met. Trans. 3, (1972), [2] L. Zheng, M. Wei-Min, Z. Zheng-Duo, Effect of Pouring Temperature on Semi-Solid Slurry of A356 Alloy Prepared by Weak Electromagnetic Stirring. Trans. Nonferrous Met. Soc. China. 16,(2006), [3] R. A. Martinez, M. C. Flemings, Evolution of Particle Morphology in Semisolid Processing. Metall. Mat. Trans. A. 36A, (2005), [4] Z. Fan, Semisolid Metal Processing. Int. Mater.Rev. 47-2, (2002), [5] H. K. Moffatt, Electromagnetic Stirring. Phys. Fluids A3(5), (1991), [6] E. Tzimas, A. Zavaliangos, Evolution of Near-Equiaxed Microstructure in the Semisolid State. Mat. Sci. Eng. A289, (2000), [7] G. Zhu, J. Xu, Z. Zhang, Y. Bai, L. Shi, Annular Electromagnetic Stirring-A New Method for the Production of Semi-solid A357 Aluminum Alloy Slurry.Acta Metall. Sing. (Engl. Lett.). 22-6, (2009), [8] C. Xing-run, Z. Zhi-feng, X. Jun, Effects of Annular Electromagnetic Stirring Processing Parameters on Semi-Solid Slurry Production. J. Trans. Nonferrous Met. Soc. China. 20, (2010), s873-s877. [9] D. Vanluu, Z. Shengdun, L. Wenjie, Numerical and Experimental Study on Effect of Process Parameters on Preparation of A356 Aluminum Alloy Semi-Solid Slurry by Electromagnetic Stirring. International Journal of Applied Electromagnetics and Mechanics 41, (2013), [10] J. Wannasin, R. A. Martinez, M. C. Flemings, A Novel Technique to Produce Metal Slurries for Semi-Solid Metal Processing. Solid State Phenom , (2006), [11] J. Wannasin, R. A. Martinez, M. C. Flemings, Grain Refinement of an Aluminum Alloy by Introducing Gas Bubbles During Solidification.Scripta. Mater. 55, (2006), [12] J. Wannasin, Applications of Semi-solid Slurry Casting Using the Gas Induced Semi-Solid Technique. Solid State Phenom , (2013), [13] S. G. Shabestari, M. Honarmand, H. Saghafian, MicrostructuralEvolution of A380 Aluminum Alloy Produced by Gas-Induced Semi-Solid Technique (GISS). Advances in Materials and Processing Technologies, 1:1-2, (2015), [14] N. Nafari, H. Aashuri, Microstructure of Semi-Solid A356 Aluminum Alloy Prepared by Electromagnetic Stirring and Gas Induction.3 rd International Conference- Engineering Materials and Metallurgy, Nov. (2014), Tehran, Iran. [15] M. J. Alvani, H. Aashuri, A. Kokabi,, R. BeygiSemisolid Joining of Aluminum A356 Alloy by Partial Remelting and Mechanical Stirring. J. Trans. Nonferrous. Met. Soc. China. 20, (2010),