On estimating crystal shape for crystal size distribution analysis

Abstract The accurate determination of three-dimensional crystal habit (long–intermediate–short axis) is a requirement to enable correct stereological conversion of 2D crystal size measurements to true 3D crystal size distributions (CSDs). In this contribution, we introduce a database and spreadsheet program which provides an objective estimate of true crystal habit from raw 2D measurements. The database compares the sample's 2D measurements with 2D shape curves for random sections through 703 different habits. The output gives the five best-match curves and corresponding crystal habits based on a least-squares fit between sample and database. The minimum sample size required to give good shape estimates from 2D data is tested using random runs of sections through known shapes with different population densities. At minimum, 75 crystal sections are required to robustly determine crystal habit for CSD measurements if crystals are tabular in shape, with more acicular shapes requiring a minimum of 250 sections, suggesting a sample size of > 250 sections to be used for shape determination in natural examples where the true 3D shape is unknown. The crystal habit of the best-match shape can then be used to provide a more robust CSD analysis, which in turn can be used to investigate magmatic processes with more certainty.