Effects of Manganese and/or Carbon on the Grain Refinement of Mg-3Al Alloy

Abstract

The effects of manganese and/or carbon on the grain refinement of Mg-3Al alloy have been investigated in the present study. Significant grain refinement was obtained for the Mg-3Al alloy modified with either carbon or Al-Mn master alloy. There existed an optimal content of 0:1� 0:2 mass%Mn to obtain refining grain size for the Mg-3Al alloy. The Al-Mn intermetallic particles with molar ratio of 1:1 were considered as potent nuclei for Mg grains. The addition of Mn had no obvious effect on the grain size for the Mg-3Al alloy which was refined by 0.2 mass%C. The Al-C-O-Mn compounds were considered as the potent nuclei for Mg grains. A new hypothesis that the particles of Al-Mn compounds with Al4C3 coating film can act as potent nuclei for Mg grains was proposed. (doi:10.2320/matertrans.MRA2007196) with carbon or not. The ultimate purpose is to provide the important data to develop a suitable and reliable grain refiner to be applied for Mg-Al alloys. 2. Experimental Procedures The raw materials used in the present study included relatively high purity magnesium (99.95 mass%Mg, 0.002 mass%Fe, 0.002 mass%Mn), high purity aluminum (99.99 mass%Al), carbon powder (45 mm in average diameter and purity of higher than 99 mass%), magnesium powder (200� 400 mm in size and purity of higher than 99 mass%) and aluminum powder (70� 150 mm in size and purity of higher than 99.5 mass%). The pellets (about 3 mm in diameter and 5� 10 mm in height) comprised of carbon powder, magnesium powder and aluminum powder at a mass ratio of 1:5:4 were prepared by using a cold isostatic press (CIP) under a pressure of 150 MPa for 1 h. The Al-20 mass%Mn master alloy was prepared in advance as follows. The pure aluminium and electrolytic manganese were melted together at 950 � C, and then the Al-Mn melt was poured into a copper mould at a high cooling rate. The MgO crucible having high purity of more than 99.5 mass% was used in the present study. The pure Mg and Al as well as Al-20 mass%Mn master alloy were melted together in an electric resistance furnace at the temperature of 780 � C. In the present study, the Al content in the prepared samples was controlled to be at constant 3 mass% and the Mn content was controlled to be at 0.05 mass%, 0.1 mass%, 0.2 mass%, 0.5 mass% in turn. After the pellets containing carbon were plunged into the melt, the melt was held for 10 min. Then it was manually stirred for 1 min with a magnesia rod, and