Micro-forming and grain refinement: effects of microstructural and geometric scale on metal formability
thesisposted on 31.01.2017 by Gu, Chengfan
In order to distinguish essays and pre-prints from academic theses, we have a separate category. These are often much longer text based documents than a paper.
The effects of grain size and geometric scale on metal processing were investigated for several processing techniques. The motivation for conducting this research was to improve our understanding of the microstructural evolution and thermo-mechanical behavior of UFG copper as a ‘model material’ used in micro-forming processes. Coarse grained (CG) copper was also investigated for comparison. The starting UFG material was obtained by equal channel angular pressing (ECAP). Micro deep-drawing and micro-extrusion were the two micro-forming processes investigated; geometry, crystallographic orientation effects and thermal stability were all examined. Crystallographic orientation was determined using electron back scattered diffraction (EBSD) and X-ray diffraction; grain size was determined by EBSD; and thermal stability was measured by differential scanning calorimetry and nano-indentation. Geometry effects were inferred from mechanical tests coupled with material characterization data. These experiments were supported by application of a grain fragmentation model coupled to a visco-plastic self consistent (VPSC) simulation in ECAP and micro-extrusion, and a VPSC simulation in micro-deep drawing. In order to characterize the starting material a comparative study was conducted for different strain paths: ECAP and cold rolling. The refined grain size was significantly smaller in ECAP than in rolling. Grain fragmentation modeling demonstrated that these differences are attributable to the different lattice rotation rate in ECAP and cold rolling, and this provides important insight into the grain refinement process to obtain UFG structures. EBSD measurements of ECAP-ed copper demonstrate that the misorientation distribution depends on the measurement plane. A new procedure is used which separates the misorientation distributions into two parts: a distribution that describes misorientations between subgrains within the original parent grain interior and a distribution that is measured across the original parent grain boundaries. The procedure permits us to track microstructure as it evolves from the original so-called ‘parent grains’. It is based on measurements on two planes of misorientation, grain shape evolution; and subgrain size variations. The results obtained showed large misorientations in the boundary regions of the parent grains and mostly low misorientations for the new grains emerging in the interior of the parent grain. In order to examine the effect of texture evolution on material behavior, micro deep drawing of ECAP-ed copper was investigated. Two blank thicknesses were investigated, and it was found experimentally that when the grain size was about one tenth of the thickness, the CG copper was less drawable. Through-thickness variations in the texture of the UFG copper were found by experiment and this was explained by simulations. It was shown that rigid body rotation during the drawing process play an important role in the evolution of the texture. Such variation is important in cup-forming if the grain size is very small. Thermal stability during further processing is one of the main concerns in the application of UFG materials for micro-forming. Micro-extruded copper with a range of grain sizes, spanning from CG to UFG copper, have been studied at different extrusion speeds. It was clear that the grain size after micro-extrusion depends on the processing speed and was always larger than the grain size expected from purely geometrical considerations of the process. Differential scanning calorimetry, dislocation density measurements and hardness all support the occurrence of dynamic recrystallization (DRX) of the UFG copper – but importantly not the CG copper – during micro-extrusion at the highest extrusion speed of 25mm/min. The effect of recovery was incorporated in the grain refinement model by changing the curvature induced dislocation density represented by an adjustable parameter in model for CG copper at the extrusion speed of 1 mm/min. Good agreement with the experimental observation was obtained. Importantly, although the CG copper was micro-extrudable, the thermal stability of UFG copper was superior after high speed micro-extrusion because of the dynamic recrystallization and this justifies the use of UFG copper in this case.