Experimental Research And Multi-scale Modelling Of High Solid Fraction Semi-solid Die Casting Of 319s Aluminium Alloy

The increasing demand for lightweight of auto industry provides a wide application prospect for aluminum alloy parts, but also raises a new challenge for the technology of manufacturing aluminum components with high performance. The semi-solid die casting process of aluminum alloys has been developed for decades, and has been applied in the auto industry gradually. The semi-solid die castings produced with solid fraction from 0.4 to 0.7 perform as equal mechanical properties as forging parts, while the cost of this process is highly competitive with that for conventional high pressure die casting. In spite of exceptional mechanical properties could be obtained, semi-solid casting process has not yet achieved the wide-spread commercial application that was envisioned during its early days. There are major limitations that are preventing its use on a broader scale, including the high sensitivity of flow pattern to process parameters, the strict requirement for component dimensions to avoid turbulent filling and non-filling defects as well, and an urgent lack of criterions for die design, etc. The scientific issue to overcome these limitations is how to reasonably predict the flow behaviour of semi-solid slurry under different process conditions.This project aims at investigating the rheology and also the flow mechanism of semi-solid slurry. Furthermore, a multi-scale modelling approach (i.e. single phase modelling and multiphase modelling) is developed to accommodate different applications of flow prediction. The alloy used in this study is 319s. The effects of heating rate and soaking time on microstructural evolution during partial melting of semi-solid 319s alloy is investigated. The flow parameters of 319s aluminium alloy are determined by isothermal compression experiments in the semi-solid state. Besides, the flow mechanism is discussed by analysing the characteristics of true stress-true strain curves. By employing the obtained rheology data, a temperature-and time-dependent viscosity model is established and applied to practical process, and then factors affecting flow stability of semi-solid slurry have been discussed in term of a modified "Reynolds Number". At the end, with the understanding of flow mechanism, the multiphase modelling (i.e. a liquid-particle-air system) is also developed to take the different motilities between the liquid and the solid phase into account, indicating the impact of process parameters on particle segregation. This study can be mainly summarized as follows:(1) Heating rate has strong effect on the melting process. Increasing heating rate will accelerate the liquid formation, followed by a wider process window. The primary particles evolution can be divided into four stages:the coarsening and coalescence of dendritic arms due to the dissolution of secondary phases; the coarsening of big grains with the dissolution of small ones; the significant coarsening and spheroidization caused by the melting of polygonal particles' corners; the slight coarsening. The coarsening rate K for 319s during isothermal test is 227?m3/s.(2) The high solid fraction slug is compressed as the particles translation and rotation. The liquid flows from high static pressure zone to lower one. Four stages can be determined with increasing strain:elastic deformation stage; partial breakdown of the deformed skeleton; further breakdown of the skeletal structure; equilibrium between agglomeration and disagglomeration.(3) By employing the obtained viscosities, the time-dependent power law viscosity model is established. The shear strain during the geometric transition is proposed to identify the relevant time-dependent parameters for modelling the die filling process in a given geometry with increasing shear conditions. For the decreasing shear condition during filling, a critical cut-off shear rate is proposed to deal with the instantaneous viscosity.(4) There is a 'sweet spot' in terms of temperature (i.e. fraction liquid), flow velocity and hydraulic diameter (i.e. die design) where the flow front has the maximum stability. This is in contrast with conventional wisdom which would suggest that low fractions liquid would give the most stable flow front. A rationale for this is presented in terms of the particle crowding at the relatively low fraction of liquid.(5) The Eulerian multiphase model is employed to take the properties of the three phases into account separately. The results indicate that the increasing filling velocity will enhance the level of particle segregation, while has no effect on the location of the segregation layer. On the contrary, increasing initial solid fraction or particle size has no distinct effect on the level of particle segregation, but would increasing the thickness of liquid layer near the wall.