Modeling of Non-Uniform Composite Microstructures

Abstract

The microstructural effects of non-uniform composite microstructures are modeled. Relationships are observed between the degree of non-uniformity and fiber spa tial information which may be measured via image analysis or derived from Dirichlet cell tessellations. For artifical patterns containing chain-like clustering which simulate com posite microstructures: (a) the nearest-neighbor distances of random and clustered pat terns are smaller than those normally estimated by square or hexagonal arrays, (b) in creased clustering may be associated with increased mean cell volume fraction, cell volume fraction distribution standard deviation, number of cell sides, distribution stan dard deviation, nearest-neighbor distance distribution skewness, and decreased nearest- neighbor distances and (c) accounting for non-uniform fiber diameters is generally un necessary, except possibly at low fiber volume fraction or with patterns exhibiting a high degree of chaining. For Nicalon SiC/zirconia titanate composite samples with micro structures which exhibit clustering, the maximum value of skewness, determined from sub-regions of the sample, correlates with the flexure strength of that sample.