Solidification Characteristics of the Al-8.3Fe-0.8V-0.9Si Alloy K.L. SAHOO, C.S. SIVARAMAKRISHNAN, and A.K. CHAKRABARTI Studies of solidification behavior have been conducted on cast Al-Fe-V-Si alloys. The first phase to precipitate during solidification of an Al-8.3Fe-0.8V-0.9Si alloy is Al 3 Fe(V,Si), which is isostructural with the Al 3 Fe phase. Thereafter, the solidification proceeds through several invariant reactions. The final invariant reaction is associated with a pronounced arrest. The temperature of this arrest is a function of the cooling rate and modification treatment, with magnesium added as an Al-20 pct Mg or Ni-20 pct Mg master alloy. The coarse iron aluminide precipitates in a slow-cooled (1 °C/s) cast structure transform to a ten-armed, star-like morphology upon chill casting the melt (cooling rate 10 °C/s) from 900 °C or upon water quenching from above 800 °C. Treatment with magnesium refines the morphology, size, and distribution of iron aluminide precipitates in slow-cooled alloys. I. INTRODUCTION charge in the crucible coated with alumina paint and heated to about 600 °C. At this stage, a 99.9 pct pure aluminium THE properties of cast Al-Fe-V-Si alloys are determined ingot (ALCAN, Ontario, Canada) was charged into the cruci- by the fineness of their microstructures and the distribution ble. The major impurities present in the aluminium ingot of their phases. Two major factors that affect these are the used in the present study were iron (0.05 pct maximum) and solidification rate and alloy composition. [1] Observable silicon (0.04 pct maximum). Traces of copper (0.01 pct changes in microstructure may occasionally be brought maximum), manganese (0.01 pct maximum), and zinc (0.004 about by chemical treatment of an alloy melt, the notable pct maximum) were also detected. After melting, sufficient example being the modification of Al-Si alloys. [2–5] Thermal time was given for complete homogenization of the melt. analysis is a powerful tool for understanding the transforma- The melt was intermittently agitated with a graphite rod for tion that occurs during solidification and solid-state cooling. complete mixing. The melt was then degassed with dry argon We use this tool, in the present investigation, to attempt to (impurities were 2 ppm oxygen, 3 ppm nitrogen, and 0.2 understand the changes in the microstructures and microanal- ppm hydrocarbons) introduced through lancing tubes, after ysis of phases in the Al-Fe-V-Si alloys that result due to the cover flux was skimmed, and subsequently cast into variations in cooling rates between 1 °C/s and 20 °C/s. These cooling rates approximate the usual sand- and chill-casting different molds to achieve different cooling rates. situations in the foundries. In addition, the feasibility of High-purity magnesium (0.003 pct Al, 0.004 pct Ca, 0.001 modifying the complex intermetallic phases in Al-Fe-V-Si pct Cu, 0.015 pct Fe, 0.005 pct Si, and the remainder Mg) alloys by a magnesium treatment has been examined. Nor- and 80 pct Al-20 pct Mg and 80 pct Ni-20 pct Mg alloys were mally, the Al-Fe-V-Si alloys are processed through a rapid used for modification studies. The approximate maximum solidification–powder compaction route. [6,7] Development concentrations of impurities present in these master alloys of a suitable modification technique may enable the pro- were 0.05 pct iron, 0.04 pct silicon, 0.01 pct manganese, cessing of these alloys through the casting route. 0.01 pct titanium, 0.005 pct chromium, and 0.003 pct cal- cium. The modifiers were introduced into the melt in the temperature range from 880 °C to 900 °C. Pure magnesium II. EXPERIMENTAL PROCEDURE was added in the form of small granules. It was wrapped in The compositions of different alloys investigated in the aluminium foil (0.4 pct Fe, 0.01 pct Cu, 0.003 pct Ca, 0.005 present study are shown in Table I. The experimental alloys pct Si, and the balance Al) and plunged into the melt. The were prepared in an electric resistance heating furnace in a Al-20 pct Mg and Ni-20 pct Mg master alloys were first clay-bonded graphite crucible under the cover of a sodium- preheated to 250 °C and then plunged into the melt in free flux (COVERAL 33FF*), which made up 2 pct of the small pieces. *COVERAL is a trademark of FOSECO (F.S.) Limited, Staffordshire, The degassed melt (both modified and unmodified) was United Kingdom. poured into (1) a 25-mm-diameter sand mold (2) a 30-mm- diameter permanent steel mold and (3) a 30-mm-diameter melt. For alloy preparation, Al-28.6 pct Fe, Al-9.85 pct water-cooled Cu mold. The object was to vary the cooling Fe-5.3 pct V, and Al-19.8 pct Si master alloys were used. rates. The pouring temperature was maintained at 900 °C (Compositions are given in weight percent unless otherwise  5 °C. The fluidity of the melt at this temperature was mentioned.) Weighed quantities of the master alloys were adequate for casting test pieces. Cooling curves were recorded with the help of a strip chart recorder attached attached to a WAHL* temperature indicator/calibrator of K.L. SAHOO, Scientist, and C.S. SIVARAMAKRISHNAN, Senior Dep- uty Director, are with the National Metallurgical Laboratory, Jamshedpur- *WAHL is a trademark of Wahl Instruments Inc., Culver City, CA. 831007, India. A.K. CHAKRABARTI, Professor, is with the Department of Metallurgical and Materials Engineering, Indian Institute of Technology, 0.4 °C accuracy and a chromel-alumel thermocouple of Kharagpur-721302, India. Manuscript submitted January 12, 1999. 0.4 mm in diameter. The thermocouple was placed at the METALLURGICAL AND MATERIALS TRANSACTIONS A VOLUME 31A, JUNE 2000—1599