TWO-PHASE MICROSTRUCTURE GENERATION IN 3D BASED ON 2D SECTIONS OF A NICKEL ALLOY

摘要

A series of new and existing techniques have been developed to generate a three-dimensional microstructure based on geometric and texture characteristics of the primary phase and secondary phase (i.e. carbides) of a nickel alloy, Waspaloy. Each technique is addressed to describe its advantages and proposed modifications to improve the representations of specific features. A Potts-based Monte Carlo grain growth model and Ellipsoid Packing/Cellular Automaton approach were compared to determine which method could construct a grain structure with size and shape distributions that match the experimental distributions. Using the experimental grain orientation distribution (ODF) and misorientation distribution (MDF) as input, an orientation assignment algorithm generated the digital texture distributions. The low volume fraction of (~0.2%) secondary phase particles was inserted onto grain boundaries and grain interiors using an injection technique. The MC grain growth model produced grains with equiaxed morphologies and a lognormal grain size distribution while the orientation assignment algorithm produced orientation and misorientation distributions nearly identical to the nickel alloy. Particle placement on grain boundaries was consistent with the experimental particle-associated misorientation distribution (PMDF); that is, the PMDF suggests a non-random correlation with grain boundary type (disorientation angle) and was reproduced in the simulation microstructure. The agreement between the digital and experimental features indicates that the techniques can be applied to generate a statistically representative two-phase microstructure.