(313c) Experimental Determination Of The Relationship Between Fiber Orientation Distribution And Stress Growth In Start-Up Of Flow For Non-Newtonian Fluids Containing Short Glass Fibers
AIChE Annual Meeting
Tuesday, November 6, 2007 - 4:15pm to 4:30pm
In this paper we investigate the nonlinear viscoelastic behavior in simple shear flow of polymeric fluids containing short glass fibers. These composite melts are notorious for exhibiting a relatively large overshoot in both the shear stress and first normal stress growth functions when compared to the rheology of the neat suspending medium. Interestingly this behavior is not reversible. The overshoot does not reappear after the sample has reached steady state and the flow is removed or interrupted. Even after a long rest time between flows. Mechanisms have been proposed that account for such behavior but no thorough analysis to confirm the relationship between fiber orientation and the transient stresses that occur in start-up of flow has been reported. Understanding this relationship is imperative to the development of a constitutive equation that can not only predict the rheology of glass fiber filled non-Newtonian fluids but the corresponding fiber orientation. Our interest lies in determining the relationship between the overshoot in the stress growth functions and the evolution of the fibers' orientation distribution within the sample. For this study we use a short glass fiber (30 wt%, L/D = 16) filled polybutylene terephethalate. Stress growth experiments in start-up of flow are performed on a Rheometrics Mechanical Spectrometer (RMS-800) using cone and plate geometry. Samples at rest and in equilibrium are deformed at a constant rate for a specific time (i.e. strain) that correlates to various points on a stress growth vs. strain plot and then the flow is stopped. The sample temperature is then lowered below the melt temperature of the suspension to ?freeze? the fibers' orientation. The fiber orientation distribution within the sample is then determined using X-ray tomography and/or by using the optical analysis termed the ?Leeds method? at Oak Ridge National Labs. The fiber orientation distribution within samples subject to various flow histories is then compared.