| dc.description.abstract | Hydrolytic degradation at elevated temperatures is a key reason for failure in offshore flexible risers. Polyamide 11 (PA11) is one of the most common materials used for the internal pressure sheaths in flexible risers due to good fatigue and creep resistance as well as decent barrier properties towards oil and natural gas. In the experimental part of this thesis the ageing of polyamide 11 in deoxygenated water at 90°C and 120°C was studied. Tensile and DMTA tests were performed to measure changes in mechanical properties. Viscometry, gravimetric measurements, DSC and TGA were used to link these properties with morphological changes.
Accelerated aging tests, followed by the Arrhenius based extrapolation, are the conventional way to evaluate long-term degradation of polymers, in particular for offshore flexible risers. In the theoretical part of this thesis a multiscale model has been developed combining diffusion, chemical kinetic reactions, structure-property relationships and composite models to provide faster and less labor extensive property predictions. A general methodology was presented and applied to predict morphology evolution and mechanical properties during the hydrolytic degradation of PA11. Results for density, degree of crystallinity, elastic modulus, tensile strength and embrittlement threshold have been compared with experimental aging in deoxygenated water at 120°C.
General trends observed in the experimental study were increased stiffness, tensile strength and glass transition temperature as well as decreased glassy state damping efficiency with increased ageing times. Changes can be initially ascribed to plasticizer depletion and then to interplay between molecular weight decrease and crystallinity increase. Viscosity at hydrolysis equilibrium indicated that brittle failure typically involves oxidation or UV exposure. For both density and degree of crystallinity the modeled trend was close to the experimental test results. Accurate prediction of the morphological parameters during degradation allowed extension of the multiscale model for the prediction of mechanical properties. Similarly, in the case of mechanical properties the model correctly predicted the experimental test results confirming hydrolysis induced chain scission and chemicrystalisation as the two main mechanisms of property change. This suggests that the multiscale modeling methodology can provide a valuable alternative to accelerated aging tests. The model also indicated that the crystalline phase does play a role in the plastic deformation as opposed to the view that this is solely a matter of the amorphous phase. Moreover, the mechanical equilibrium between effects of macromolecule degradation and an increased degree of crystallinity has been described.
This work is structured as an article-based thesis. The papers that constitute the main part of this thesis are given in Part II. Part I places the included articles into a wider context and provides their summary. | nb_NO |
