Prediction of Fiber Orientation During Processing of Short Fiber Composites

231 p.Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1987.The fiber orientation pattern created by the flow of a suspension of short fibers controls the physical and mechanical properties of a short fiber composite material. The orientation of the fibers at any point in the composite plays the role of a structural state variable. Two sets of even-order tensors are used to describe the distribution of orientation of fibers in suspensions and composites containing short rigid fibers. These orientation tensors offer a description which is concise, unambiguous and independent of coordinate system. They provide very efficient calculations for planar orientations and make the prediction of three-dimensional orientation states feasible. Equations of change for the tensors are derived to predict the orientation of fibers caused by flow during processing. A closure approximation is required in the equations of change. The accuracy of various closure approximations is explored through calculations of mechanical properties of solid composites and the dynamics of orientation in flowing suspensions. A hybrid closure approximation, which combines previous linear and quadratic forms, is introduced. It performs best and permits the use of the second order tensor alone. A Galerkin finite element method is formulated for solution of the equations of change in arbitrary two-dimensional geometries. A computer simulation is developed, which can perform fiber orientation predictions for the compression mold filling process. The simulation is based on the finite element method and is capable of predicting planar orientation in flat parts of otherwise complicated shape. Comparison with experiments show that the simulation accurately predicts the orientation behavior of fibers during compression molding of sheet molding compound and in a model suspension of nylon monofilaments in silicone oil.U of I OnlyRestricted to the U of I community idenfinitely during batch ingest of legacy ETD