Date of Graduation

5-2013

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Engineering (PhD)

Degree Level

Graduate

Department

Mechanical Engineering

Advisor/Mentor

Saxena, Ashok

Committee Member

Couvillion, Rick J.

Second Committee Member

Holdsworth, Stuart R.

Third Committee Member

Spearot, Douglas E.

Fourth Committee Member

Zou, Min

Keywords

Pure sciences; Applied sciences; Chromium; Constitutive modeling; Creep; Fatigue; Ferritic steel; Molybdenum; P91 steel

Abstract

Grade P91 steel, from the class of advanced high-chrome ferritic steels, is one of the preferred materials for many elevated temperature structural components. Creep-fatigue (C-F) interactions, along with oxidation, can accelerate the kinetics of damage accumulation and consequently reduce such components' life. Hence, reliable C-F test data is required for meticulous consideration of C-F interactions and oxidation, which in turn is vital for sound design practices. It is also imperative to develop analytical constitutive models that can simulate and predict material response under various long-term in-service conditions using experimental data from short-term laboratory experiments. Consequently, the major objectives of the proposed research are to characterize the creep, fatigue and C-F behavior of grade P91 steels at 625 C and develop robust constitutive models for simulating/predicting their microstructural response under different loading conditions.

This work will utilize experimental data from 16 laboratories worldwide that conducted tests (creep, fatigue and C-F) on grade P91 steel at 625°C in a round-robin (RR) program. Along with 7 creep deformation and rupture tests, 32 pure fatigue and 46 C-F tests from the RR are considered in this work. A phenomenological constitutive model formulated in this work needs just five fitting parameters to simulate/predict the monotonic, pure fatigue and C-F behavior of grade P91 at 625 C. A modified version of an existing constitutive model is also presented for particularly simulating its isothermal creep deformation and rupture behavior.

Experimental results indicate that specimen C-F lives, as measured by the 2% load drop criterion, seem to decrease with increasing strain ranges and increasing hold times at 625°C. Metallographic assessment of the tested specimens shows that the damage mode in both pure fatigue and 600 seconds hold time cyclic tests is predominantly transgranular fatigue with some presence of oxidation spikes. The damage mode in 1800 second hold time cyclic tests is an interaction of transgranular fatigue with dominant oxide spikes and creep cavitation. Other experimental results including the statistical analysis and inter- and intra-laboratory variability in the C-F lifetimes are provided in the text. Scatter factor for any of creep, monotonic, pure fatigue and C-F simulations is shown to be at a maximum of ~ 1.3, in comparison to > 5 expected for a RR. Moreover, the microstructural variability between nominally homogeneous specimens can be inherently accounted by the formulated constitutive model.

Share

COinS