Experimental characterisation and computational constitutive modelling of high temperature degradation in 9Cr steels including microstructural effects
Barrett, Richard A.
MetadataShow full item record
This item's downloads: 566 (view details)
The transition from base-load to intermittent power plant operation has led to the requirement for increased knowledge and more accurate prediction of the effects of flexible loading on plant components. This thesis is concerned with the experimental characterisation and novel computational modelling, including dislocation-mechanics based modelling, of high temperature cyclic plasticity and fatigue for current and next generation power plant materials. The modelling is designed to capture the constitutive behaviour of 9Cr steels across a range of operating conditions, calibrated and validated via measured test data. Phenomenological and microstructurally-based models are developed at the continuum level and applied to isothermal fatigue, thermo-mechanical fatigue and multi-axial loading conditions relevant to next generation power plant. High temperature low cycle fatigue test programmes are conducted on P91 steel and MarBN, a new high temperature alloy currently in development. The results illustrate cyclic softening due to dynamic recovery and dependent on the initial microstructure. A program of thermo-mechanical fatigue experiments is presented for P91 steel across a range of strain-rates, with a significant reduction in fatigue life observed for out-of-phase loading. This highlights the effect of thermal transients on the material performance. The observed strain-rate effect is incorporated within a phenomenological material model via a hyperbolic sine constitutive equation. The hyperbolic sine framework enables variable strain-rate sensitivity in the model and the successful prediction of the strain-rate effect at higher and intermediate strain-rates. A phenomenological kinematic hardening model enables prediction of the effect of the various strengthening mechanisms, with isotropic softening included to predict the cyclic softening behaviour. The material model is implemented in a UMAT user material subroutine, including consistent tangent stiffness, for use with the commercial finite element code Abaqus and calibrated using a step-by-step parameter identification methodology. The material model is successfully validated against isothermal fatigue, thermo-mechanical fatigue and stress relaxation behaviour via comparison with experimental data, and demonstrated for multi-axial applications to notched specimen geometries and thin-walled pipes. A novel dislocation-mechanics based constitutive model is developed, including mean values of microstructural parameters related to the key strengthening mechanisms in 9-12Cr steels. This model includes kinematic hardening based on back-stresses developed due to the presence of precipitates, as well as the pile-up of dislocations at high-angle grain boundaries. A dislocation-mechanics approach is implemented to predict recovery in the material. The novel material model is applied to high temperature fatigue of P91 steel and successfully compared to experimental test data and measured evolutions of key microstructural parameters, viz. dislocation density and martensitic lath width. These results illustrate the ability of the microstructure-sensitive framework to capture both the constitutive behaviour and mean microstructural evolutions of 9Cr steels. This model enables successful advancement of the microstructure under flexible operation within a wider creep-fatigue modelling methodology at the macro-scale and component levels. The microstructure-sensitive material model developed in this work will allow for more accurate simulations of the constitutive behaviour of complex 9Cr welded connections in power plant components. This will, in turn, enable power plant designers to account for microstructural degradation during cyclic viscoplastic deformation.
This item is available under the Attribution-NonCommercial-NoDerivs 3.0 Ireland. No item may be reproduced for commercial purposes. Please refer to the publisher's URL where this is made available, or to notes contained in the item itself. Other terms may apply.
The following license files are associated with this item: