Reassessing Polymer Lifetime Prediction Methods with Improved Characterization and Diagnostics

Abstract

A combination of Magnetic Resonance Imaging (MRI), Multiple Quantum Nuclear Magnetic Resonance (MQ-NMR), Molecular Dynamics modeling, and traditional mechanical testing approaches have been used to provide a more scientific prediction of the aging behaviors of two silica-filled siloxane polymers. These materials are especially prone to part-to-part and service condition variablilities, and thus a combination of non-destructive magnetic resonance techniques and atomistic modeling has been used to determine physical and chemical inhomogeneities which are, respectively, already present in the material and potentially occurring as a result of cavitation upon applied stress. To elucidate the overall degradation behavior of the polymers studied and thus add scientific evidence to their lifetime predictions several different damage mechanisms (thermal, radiation, and mechanical) have been studied individually and in combination to elucidate the overall aging behavior for a variety of service conditions. Running concurrent to these experimental and modeling efforts, an analysis has been done to more precisely define the operational capabilities of the polymers relative to their requirements, leading to an even more accurate lifetime prediction capability due to the ability to define performance margins with their associated uncertainties.