Grain boundary networks and shape preferred orientation: A fresh angle on pattern quantification with GBPaQ

Abstract

A quantitative understanding of grain shape preferred orientation (SPO) and grain boundary networks as fundamental characteristics of rocks and other crystalline solids is of major interest in geology and material science. Grain boundary networks contain useful information on the deformation history of polycrystalline aggregates, and their diagenetic and metamorphic histories. SPO can have a major impact on material characteristics such as permeability, acoustic velocity and mechanical strength, and on reaction surfaces. The objective of this study is to present a semi-automated toolbox of MATLAB™ scripts, named Grain Boundary Pattern Quantification (GBPaQ), that incorporate different methods for grain boundary pattern quantification for their application to, for example, seismic wave attenuation estimation. GBPaQ uses grain boundary statistics and calculates radial scan line intercepts. In this paper, GBPaQ is tested on two example grain boundary patterns, a granular texture and a foam texture with equant grains, which have been digitally stretched (deformed) to analyse their SPO evolution. The results show that a combination of grain ellipse, grain boundary segment orientation, and grain boundary segment intercept density rose diagrams provide a complete, detailed quantification of grain boundary pattern anisotropy. Grain boundary segment intercept (GBSI) analysis using GBPaQ yields a new grain boundary network parameter – the minimum intensity of grain boundary intercepts (Imin) – which follows a power law relationship with the average axial ratio of grain-fitted ellipses (r) during SPO development. We propose that Imin can be used for the quantitative analysis of SPO strength as a useful tool to assess the deformation history of polycrystalline aggregates. Further studies involving a broader range of different patterns and strain histories are necessary to fully investigate the potential of Imin versus r diagrams.