Monte Carlo Simulation of Grain Growth at Heat Affected Zone (HAZ) of Weldment

Abstract

Abstract Average grain size and grain size distribution at HAZ is the dominant microstructural features of weldment. Such features may govern many properties of the weldment, such as strength, ductility, toughness, corrosion resistance and so forth. Accordingly, prediction of the final grain size and grain distribution is crucial to a welding engineer. The nature of grain growth in the weldment is known to be stochastic. Hence, using a deterministic method for predicting the final grain size of weldment should be reconsidered. In this paper Monte Carlo simulation of grain growth at Heat Affected Zone (HAZ) of a 0.5 Mo-Cr-V steel is presented. The technique used for this simulation is discussed. We also use a methodology to obtain a one-to-one correlation between Monte Carlo (MC) and real parameters of grain size and time. The results are compared with the experimental results and a reasonable agreement is observed. To carry out the simulation, we generate a random 200*200 matrix. Each element of this matrix represents a grain site with a unique grain orientation at the beginning of the simulation. We run the simulation for both bulk heating situation and HAZ zone to decide the thermal pinning effect. In the case of HAZ. zone, the center of this matrix represents the fusion zone of the HAZ. Periodic boundary conditions are used so that the grains wrapping around are at the same temperature. To make a comparison, we look at a HAZ location about 120 microns from the fusion zone and compare it with the grain size obtained in an actual weld HAZ at an identical location. The simulation result shows a near match with the experimental data for bulk heating. The grain structure of the simulation result in the weld HAZ also has a reasonable accuracy. It can be concluded that the MC technique could effectively capture the thermal pinning existing in the weld HAZ. The difference between the simulation result and the experimental data suggest the effect of thermal pinning is higher than the effect based on the thermal gradient alone.