The columnar to equiaxed transition in tin-lead alloys
- PDF / 1,267,305 Bytes
- 8 Pages / 583.28 x 777.28 pts Page_size
- 37 Downloads / 236 Views
I.
INTRODUCTION
A number of mechanisms have been proposed to account for the columnar to equiaxed transition (CET) in chill castings.~'z One mechanism associates the CET with the generation of nuclei, particularly the remelting of secondary dendrite branches which produce nuclei ahead of the advancing columnar dendrites. This mechanism assumes that the controlling factor for the CET is nucleation. Dendrite remelting cannot be used to predict when the CET will occur, since it requires specific data on the interdendritic liquid temperature and flow velocity, and the melting temperature of the dendrite branch roots which is generally unknown. An alternative mechanism for the CET has been proposed by Hunt and his coworkers, 3 based on melt supercooling. In careful measurements they have shown that the advancing columnar dendrite tip is supercooled, primarily due to solute concentration ahead of the tip, the amount of supercooling depending on the growth velocity, temperature gradient, and alloy composition. 4'5 The mechanism for the CET assumes heterogeneous nuclei are present at a specified level and grow at a specific supercooling. A columnar or equiaxed structure is then predicted, under given growth conditions, based on heat and mass flow considerations. The transition from columnar to equiaxed can then occur progressively for a given alloy as either the temperature gradient or growth rate is changed in the transition region. Using the model, predictions are made for the CET in A1Cu alloys. 3 In a directionally solidified chill casting, the growth rate and temperature gradient are coupled, both depending primarily on the rate of heat removal at the chill, and the thermal conductivity of the alloy. The present investigation was undertaken to examine experimentally the CET in SnPb alloys solidified directionally under varying growth conditions, and to determine to what extent the proposed mechanisms for the CET can account for the results. II.
EXPERIMENTAL P R O C E D U R E
Alloys of Sn containing 5, 10, and 15 wt pct Pb were prepared from 99.9 pet Sn and Pb and solidified directionally in the apparatus Shown in Figure 1. A cylindrical R.B. MAHAPATRA, Graduate Student, and F. WEINBERG, Professor, are with the Department of Metallurgical Engineering, University of British Columbia, Vancouver, BC, Canada, V6T 1W5. Manuscript submitted September 10, 1986.
METALLURGICALTRANSACTIONS B
copper mold was used having an internal diameter of 35 ram, height 110 mm, and wall thickness of 3.2 mm. The inner and outer vertical surfaces of the mold were covered with a layer of fiberfax, 2 mm thick, to suppress radial heat losses. Heat was extracted from the bottom of the mold through a stainless steel sheet into a water cooled copper block. The heat transfer from the mold to the copper block was changed by changing the thickness of the steel sheet. In a normal test, the melt was heated to a uniform specified temperature, and cold water then passed through the copper chill. Solidification occurred dendritically from the chill fa
Data Loading...
