Processing Routes and Deformation Characteristics of High Temperature and High Strain Rate Superplasticity in Aluminum Alloys and Composites

博士 === 國立中山大學 === 材料科學研究所 === 87 === The current study first examined the processing routes for producing the 2024Al/SiC, 6061Al/SiC, pure-Al/SiC, and 7091Al/B4C particulate-reinforced composites exhibiting high strain rate superplasticity (HSRS), as well as the characterization of their HSRS charac...

Full description

Bibliographic Details
Main Authors: Bai-Yao Lou, 樓百堯
Other Authors: J. C. Huang
Format: Others
Language:en_US
Published: 1999
Online Access:http://ndltd.ncl.edu.tw/handle/72554975143332078801
Description
Summary:博士 === 國立中山大學 === 材料科學研究所 === 87 === The current study first examined the processing routes for producing the 2024Al/SiC, 6061Al/SiC, pure-Al/SiC, and 7091Al/B4C particulate-reinforced composites exhibiting high strain rate superplasticity (HSRS), as well as the characterization of their HSRS characteristics over 500-640 oC and 10-4-101 s-1. It was observed that the HSRS characteristics of the composites were process dependent. The critical processing steps included mixing, pressing and working details. The uniform dispersion of reinforcement within the matrix appeared to be most crucial. The best HSRS elongation of 230% occurred at 525 oC and 1x10-1 s-1 for the 2024Al/SiC/15p composites, 280 % occurred at 580 oC and 1x10-1 s-1 for the 6061Al/SiC/15p composites, and 280 % occurred at 640 oC and 1x100 s-1 for pure-Al/SiC/15p composites, suggesting that the optimum test temperature varies apparently for different matrix alloys. The optimum HSRS temperatures were above the incipient melting temperatures, the small amount of liquid phase at interfaces and/or grain boundaries would help to release the stress concentration and help to accommodate the strain induced by grain boundary sliding or interface sliding. The effect of the liquid phase on superplastic deformation was further systematically examined in commercial 1050 and 6061 Al alloys and a commercial 6061/SiC/20w composite at extrahigh test temperatures of 550 to 650 oC and over the strain rates of 2x10-4 to 5x10-1 s-1. The representative strain rate sensitivity m-values was ~0.2 over most strain rates for the 1050 alloy; ~0.5-0.7 at low rates and ~0.2 at higher rates for the 6061 alloy; and ~0.1 at low rate and ~0.3-0.5 at high rates for the 6061/SiCw composites. Power law creep seemed to occur over most strain rates in the 1050 alloy, as well as at higher rate in the 6061 alloy. Liquid-involved deformation has been occurred at low rates in the 6061 alloy. And GBS with the help from a limited amount of liquid was the controlling mechanism in the fine grained 6061/SiCw composites at high rates. At temperatures lower than the incipient melting point, the values of true activation energies for all four materials examined in this study was ~200 kJ/mol. This value was higher than but roughly in agreement with the activation energy for Al lattice diffusion. The activation energy of the 1050 alloy at ~650 oC and the 6061/SiCw composite at 590-610 oC exceeded 1000 kJ/mol. Such an exceptionally high value above the incipient partial melting point does not imply any physical meaning, but appears to represent a transition from plastic flow in a solid state to viscous flow in a liquid phase. Although the activation energy of the 6061 Al alloy at T>590 oC did not increase to over 1000 kJ/mol, as did in the 1050 alloy and the 6061/SiCw composite, this might due to the fact that all stress data over 570-610 oC at low rates were affected by the liquid phase and all bend to lower stress to a similar degree. Anyway, the current results from the 6061 alloy have provided numerous pieces of evidence supporting that the deformation process was indeed affected by a critical amount of the liquid phase and the effect was more pronounced at low strain rates. In addition, the true activation energy for the 6061/SiC/20w composite rationalized in terms of load transfer was not successful since the temperature dependence of load transfer has been overestimated. The predicted local stress caused by sliding was much higher than the flow stress. This implies that the stress concentration caused by sliding was relaxed by the liquid phase.