Failure of thermal barrier coatings under thermal and mechanical fatigue loading: microstructural observations and modelling aspects
2004 (English)Doctoral thesis, comprehensive summary (Other academic)
Industrial and air-borne gas turbine hot components suffer from creep, oxidation, corrosion and microstructural degradation if not shielded from the hot and aggressive combustion gases. Two major strategies commercially available are adopted; film cooling by pressurised air and application of protective coatings. Protective coatings form a slow-growing oxide that protects from oxidation and corrosion. By application of a thermal insulator, a thermal barrier coating, the material will be protected from high temperature through good insulation properties of the coating system.
If thermal barrier coatings are to be used in situations where capabilities and possibilities for inspections are limited, better knowledge of the fatigue properties of the coatings is also needed. Therefore development of a reliable fatigue life model is needed. The present work aims at serving as a basis from which a general physically founded thermal barrier coating life model can be formulated. The effects of exposure to high temperatures and mechanical loads on thermal barrier coatings under service like conditions have been investigated in the present thesis. Emphasis is put on the coupling between materials science and solid mechanics approaches in order to establish a better knowledge concerning degradation mechanisms and fatigue life issues than what is common if only one discipline is explored.
Investigations of material exposed to isothermal oxidation and thermal cyclic fatigue were performed on plasma-sprayed systems with NiCoCrAIY or NiCrAIY bond coats and yttria partially stabilised zirconia top coats. It has been shown that the thermally grown oxide that will form upon high temperature exposure influences the failure behaviour. If the oxide is composed mainly of alumina, the fatigue properties are good since the adhesion between the ceramic top coat and the metallic bond coat is good. This is also shown in a comparison between different plasma sprayed thermal barrier coating systems. If the oxide formed is based on alurnina and spinel is avoided the fatigue properties benefit from a relatively flat interface where out-of plane stresses are low in comparison to a rough interface between top- and bond coat. These findings indicate that the bonding in air-plasma sprayed systems is dependent on so called chemical bonding if the thermally grown oxide is not voluminous with high growth stresses.
It is possible to establish a fatigue life model for thermal barrier coatings. This has been shown with a model based on a modified Paris law formulation. The formulation needs to be modified with regards to mode rnixity of growth. Results achieved in the present project show that it is possible to extract crack growth data for interfacial crack growth. However a combination of mechanical testing and finite element modelling is needed since the load situation in critical areas cannot be measured. Crack growth results are presented and crack growth data are compared to predictions with good agreement.
Place, publisher, year, edition, pages
Linköping: Linköpings universitet , 2004. , 73 p.
Linköping Studies in Science and Technology. Dissertations, ISSN 0345-7524 ; 898
thermal barrier coating, TBC, delamination, crack initiation, crack propagation, crack growth, oxidation, alumina, spinel, MCrAIY, diffusion, fatigue, modelling, modeling, degradation
Engineering and Technology
IdentifiersURN: urn:nbn:se:liu:diva-22793Local ID: 2126ISBN: 91-852-9540-XOAI: oai:DiVA.org:liu-22793DiVA: diva2:243106
2004-10-29, Sal C3, Hus C, Linköpings Universitet, Linköping, 10:15 (Swedish)
Karlsson, Anette M., Professor
List of papers