Search

In the past few decades, Laminated composites have become the most favored material system for a wide range of industrial applications. Factors such as the potential for structural tailoring, superior strength, and stiffness for a given weight along with higher toughness and mechanical damping etc., have justified their increased use. Due to this high usage, the accurate behavior of laminated composites under various loading conditions remains a concern and hence this dissertation focuses on building simple and accurate uni-directional micro-mechanical material models.Firstly, an elastic micromechanical model based on the physically viable sub-cell boundary conditions is developed and implemented for use with uni-directional composite laminates in the explicit finite element software LS-DYNA®. Stress-strain relations have been presented in a three-dimensional context and hence can be used with solid elements. The effect of these boundary conditions in accurately estimating the elastic properties has been studied. Transverse and shear modulus have been analyzed in detail alongside popular analytical methods and verified against available experimental results for various volume fractions. Good agreements have been observed for the presented model in comparison with the experimental results.Secondly, a non-linear micro-mechanical composite material model is developed to simulate the behavior of uni-directional composites under impact loading conditions. The strain-rate and pressure dependency in the composite material model is accounted by the resin, which uses previously developed state-variable viscoplastic equations. These equations were modified to account for the significant contributions of hydrostatic stresses typically observed in polymers. The material model also uses a continuum damage mechanics (CDM) based failure model to incorporate the progressive post-failure behavior. A set of Weibull distribution functions are used to quantify this behavior and an iterative methodology of characterizing the softening parameters used in these functions is presented. The impact response of composite laminate plates has been simulated and compared to the experiments. It has been observed that the predicted results compare favorably to the experiments. In addition, the effectiveness of studying inter-laminar delamination in laminated composites with the help of thickness-stretch shell elements which utilize a 3-D material model sub-routine as compared to the traditional plane-stress shell elements has also been investigated. Ply-level progressive failure criteria have been used to simulate the initiation and propagation of delamination. The numerical delamination growth has been qualitatively analyzed against the experimental C-scan images for multiple impact events on different composite plates.Lastly, since it is well known that strain-softening behavior in composite material models leads to a pathological mesh sensitivity in finite element simulations. Non-local continuum theory has been used and three non-local damage models with isotropic weight function have been implemented to work in conjunction with the non-linear composite micro-mechanical material to address this problem. The effect of weighing function on each of these formulations has also been studied in detail. All three non-local formulations have been observed to produce a nice smeared effect of damage, unlike the local damage models.

The work involved studies of two separate aspects of natural fibre composites. The first part of the work examined the use of high aspect ratio short fibres in thermoplastic matrix composites. This required that the technical fibres were first broken down into elementary fibres. Kenaf fibres were used in this part of the study with high-density polyethylene (HDPE) as the matrix. Several different treatments were examined for isolating the elementary fibres from the technical fibres, the most successful being 60% nitric acid treatment and 20% hydrogen peroxide/glacial acetic acid treatment. The hydrogen peroxide/acetic acid treatment produced full length elementaries with an average length of 2.3 mm and an average aspect ratio of 180. However, the nitric acid treatment caused fragmentation of the elementary fibres resulting in an average length of only 0.2 mm and an average aspect ratio of 15. It also caused an increase in the defect density of the fibres. Both treatments increased the cellulose crystallinity but caused some oxidation of the fibres. The elementary fibres were used to produce HDPE composites with a 40 wt% fibre fraction. The behaviour of these composites was then compared with that of composites of the same fibre fraction, prepared using chopped technical fibres with an average length of 0.7 mm and an average aspect ratio of 8. Breakup of the elementaries occurred during extrusion reducing the fibre length by as much as a factor of 10. However, the chopped technical fibres were unaffected. No improvement in modulus or strength over that of the chopped fibre composites was obtained for the nitric acid treated fibre composites, due to the very low fibre aspect ratio. However, an improvement of 20% in both modulus and strength was obtained for the hydrogen peroxide/acetic acid treated fibre composites. The second part of the work examined the suitability of using the results obtained from single fibre testing and from flat fibre bundle testing to predict the tensile properties of unidirectional composites made from the fibres. Flax fibres were used in this part of the study. Unidirectional composites were prepared with a fibre volume fraction of 25% using vinyl ester as the matrix resin. The modulus and strength obtained by backing out the fibre properties from tensile data obtained from the composites were both within 7% of those obtained from single fibre testing, once proper account was taken of the true fibre cross sectional shape and the effect of fibre length on fibre strength. The flat fibre bundle tests gave values which were only 43% of the single fibre data. However, the scatter in results was much lower indicating that this test may be useful for assessing batch to batch variation.