Prediction of Grain Structure and Texture in Twin-Roll Cast Aluminum Alloys Using Cellular Automaton-Finite Element...

Area of Science:

• Materials Science
• Metallurgy
• Computational Modeling

Background:

• Twin-roll casting (TRC) is efficient for aluminum sheet production but lacks control over initial grain structure.
• Predictive modeling is needed to understand and control microstructure evolution during TRC.

Purpose of the Study:

• To develop and validate a cellular automaton-finite element (CA-FE) model for predicting grain structure and texture in TRC aluminum strips.
• To investigate the solidification behavior and microstructure evolution of pure Al and AA7075 alloys during TRC.

Main Methods:

• A CA-FE model was developed, incorporating a Gaussian nucleation distribution and an equivalent binary approach for multicomponent alloys.
• Pure Al and AA7075 were cast using a pilot-scale horizontal twin-roll caster.
• Experimental characterization of microstructures and texture analysis (IPFs, PFs) were performed.

Main Results:

• The CA-FE model successfully predicted solidification, grain structure, and texture evolution.
• Distinct differences in temperature distribution and solid fraction were observed between pure Al (columnar grains) and AA7075 (equiaxed grains).
• The model accurately captured the <001> texture in pure Al and random texture in AA7075.

Conclusions:

• The developed CA-FE model provides reliable predictions for TRC strip microstructure and texture.
• This model is a valuable tool for optimizing continuous casting processes in aluminum production.