Investigation into the effect of nucleation parameters on grain formation during solidification using a cellular automat
- PDF / 1,335,729 Bytes
- 14 Pages / 612 x 792 pts (letter) Page_size
- 49 Downloads / 202 Views
C.J. Davidson Commonwealth Scientific and Industrial Research Organization (CSIRO)—Manufacturing Science and Technology, Kenmore, 4069 QLD, Australia
D.H. StJohn CRC Centre for Metals Manufacturing (CAST) Cooperative Research Centre, School of Engineering, University of Queensland, Brisbane, 4072 QLD, Australia (Received 14 December 2007; accepted 28 February 2008)
A cellular automation (CA) model has successfully been used to model the development of microstructure of an aluminum alloy during solidification to produce detailed structure maps for the solidified alloys. More recently, the application of CA models to practical castings/solidification conditions has attracted increasing research interest. However, the determination of the calculation parameters of any model associated with nucleation is difficult. Accordingly, this work investigates the detailed effect of the six parameters of nucleation on microstructure formation and morphology as well as the grain size by cellular automaton-finite control volume method (CAFVM). The nucleation parameters can be determined or estimated by comparing the calculated and experimental results, which enables a more practical prediction of the microstructure (morphology and grain size).
I. INTRODUCTION
The control of solidification microstructures is essential to the modification and improvement of the quality and properties of the final cast products. Prediction of the formation and evolution of the cast microstructure is an essential element in the control of these microstructures during solidification and therefore the modeling of these microstructures is of considerable interest. As a result, several kinds of models have been developed in the last couple of decades.1–10 These include deterministic models1–4 and other models,5–14 which include both deterministic and probabilistic aspects. Several excellent reviews in this area have been published recently (RafiiTabar and Chirazi15 and Stefanescu16). Each kind of model has its advantages and disadvantages. For example, deterministic models, which are based on solute diffusion and its coupling with mass and momentum conservation, use a strong physical background to describe the nucleation and crystal growth during solidification.15 However, the deterministic models are limited because
a)
Address all correspondence to this author. e-mail: [email protected] DOI: 10.1557/JMR.2008.0303 2312
http://journals.cambridge.org
J. Mater. Res., Vol. 23, No. 9, Sep 2008 Downloaded: 10 Feb 2015
the nucleation typically requires a stochastic approach. Thus these models cannot account for the grain distribution,17 which occurs close to the mold surface and gives rise to the columnar region.18,19 These limitations led to the development of probabilistic models, based on several well-known techniques such as the Monte Carlo (MC) procedure,5–7 the phase field approach,20–22 and the cellular automation approach (CA).17 These models can reproduce more accurately the distribution of grains in the columnar zone and the columnar-to-equiaxed
Data Loading...
