Journal of Alloys and Compounds 470 (2009) 413–419 Contents lists available at ScienceDirect Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jallcom Effects of stirring parameters on rheocast structure of Al–7.1wt.%Si alloy M. Reisi, B. Niroumand ∗ Department of Materials Engineering, Isfahan University of Technology, Isfahan 84156-83111, Iran article info Article history: Received 12 December 2007 Received in revised form 20 February 2008 Accepted 25 February 2008 Available online 9 April 2008 Keywords: Metals and alloys Crystal growth Microstructure abstract In this study, effects of stirring time and stirring speed on the microstructure of semisolid rheocast (SSR) Al–7.1 wt.%Si were examined. The results demonstrated that the non-dendritic structure could be formed by a short stirring period below liquidus temperature and further stirring had little impact on the final morphology of the primary particles. Stirring was shown, however, to affect the average particles size mainly during the initial stages of solidification. Although the average shape factor of primary particles was relatively insensitive to large variations in the stirring speed, higher stirring speeds made the shape and size of the primary particles more uniform. Higher stirring speeds also rendered smaller and more rounded agglomerates of primary particles. The results of two stability models employed suggest that, in general, the primary particles generated in the initial stages of solidification can attain growth stability before pouring and maintain this stability during the secondary cooling stage. © 2008 Elsevier B.V. All rights reserved. 1. Introduction The desired starting material for semisolid metal (SSM) form- ing is a partially solidified (or partially melted) alloy in which the solid is present as nearly perfect spheroidal particles [1]. In the early days of SSM development, it was thought that in order to pro- duce such a slurry, dendrites had to be formed within the semisolid region and then sheared by strong melt agitation via mechanical or magnetohydrodynamic (MHD) means [1]. However, the researches conducted during the last few years had led to the fundamental dis- covery that the ideal semisolid structure could be produced directly from the melt by a controlled nucleation and growth phenomenon, i.e. by promoting copious nucleation during the initial stages of solidification and controlling the growth of the nuclei formed in the later stages of solidification [2,3]. Based on this finding, various processes have been invented during the last few years for production of semisolid metal slur- ries. In one of these processes named SSR (semisolid rheocasting) process [4], the ideal microstructure for semisolid forming can be produced by a short stirring during the first few percentages of solidification, provided that stirring is combined with rapid heat extraction through the stirrer. In this process, a rotating cold copper or graphite rod is immersed into a melt held just above its liquidus temperature. The immersion of the rotating rod creates a region of high local cooling and drops the bulk melt temperature below the liquidus temperature. At the same time it provides vigorous convec- ∗ Corresponding author. Fax: +98 311 3912752. E-mail address: behzn@cc.iut.ac.ir (B. Niroumand). tion in the melt. These will result in the formation of a high density of nuclei well distributed in the melt and produces non-dendritic primary particles, essentially free of entrapped liquid [5]. In this study, experiments were conducted to study the influence of stirring time and stirring speed on the microstructural features of a SSR-processed Al–7.1wt.%Si alloy. 2. Experimental procedures 2.1. Preparation of SSR slurry In each experiment 520 g of Al–7.1 wt.%Si alloy was melted in a graphite crucible having a bottom pouring arrangement. The melt was then allowed to cool inside the furnace. When the melt temperature reached to approximately 3 ◦ C above its measured liquidus temperature (TL), a cold graphite stirring rod with 27 mm diam- eter was immersed into the melt removing the superheat almost instantly. In each experiment, regardless of the stirring time, the slurry was cooled continuously below the liquidus temperature for about 100s, corresponding to 0.3 solid fraction, and cast into a wedge shaped steel die made of 20mm thick steel plates. The section at which the microstructural studies were performed was about 9mm across. The primary cooling rate of the melt while passing the liquidus temperature, i.e. in the temperature range of TL ± 2 ◦ C, was approximately 0.26 ◦ Cs -1 in all the experiments. In this study, effects of two SSR process parameters, i.e. stirring time and stirring speed, on size and shape of primary features of rheocast Al–7.1wt.%Si were exam- ined. The experiments were also aimed at recognizing the stirring requirements for significant changes to the dendritic microstructure. Details of processing parameters in two sets of experiments conducted are shown in Table 1. In the first experiment, the graphite rod was just immersed into the melt and kept stationary for 100s. In experiments number 2–5, the slurry was stirred for 5, 20, 40 and 70s below the liquidus temperature and then cooling was continued quiescently for another 95, 80, 60 and 30s before pouring, respectively. Similar to the first experiment, the graphite rod remained in the melt for the entire period of cooling until pouring. In experiment number 6, the slurry was stirred below liquidus temperature for 100 s and cast instantly. In experiments number 7–9, the slurry was stirred at 100, 400 and 700 rpm for approximately 100 s and cast into the steel die. 0925-8388/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jallcom.2008.02.104