EFFECT OF LATTICE AND NEIGHBOURS IN THE 2D AND 3D MONTE CARLO GRAIN GROWTH SIMULATIONS

Abdelhak AYAD, N. ROUGO

Résumé


Abstract:
Grain growth is largely studied both theoretically and by computer simulations in two
dimensions. but it represents as yet a challenge in three dimensions because the difficulty to
studying the gain growth from physical experiments. Usually, microstructure studies are
performed for cross-sections of experimental specimens, thus providing only 2D
microstructure information?s. Therefore. computer simulation is actually the most effective
tool for studying the 3D grain growth process. In this paper, three and two dimensional Monte
Carlo simulations are conducted to study the isotropic gain growth. This technique involves
representing the microstructure on a discrete set of regular space grid. In this paper we
investigate the effect of lattice that represents the discrete microstructure by using a square
grid for 2D and a simple cubic network for 3D. Grain growth kinetics and their topology are
analyzed with the 1st , 2nd and 3rd neighbour's consideration both on 2D and 3D. 1t can be seen
that a compact grain structure was developed. Various microstructural features of normal
grain growth commonly observed in 2D Monte Carlo simulations were also seen in the
microstructure in the cross-section planes. The first results show that the lattice symmetry and
the neighbour?s number have a fundamental influence on the results of grain growth
simulation. Also the average grain size is a very important parameter because it is closely
related with many properties of simulated microstructures. The mean grain size increases as a
power law of the time both in 2D and 3D simulations but microstructure evolutions on 2D
cross-sections starting from 3D simulation give better results compared to a direct 2D
simulation.

Mots-clés


Monte Carlo; grain growth; topology; lattice; neighbours; 3D; 2D

Texte intégral :

PDF