ICMCTF2000 Session A3-2: Thermal Barrier Coatings

Tuesday, April 11, 2000 1:30 PM in Room Royal Palm Salon 1-3

Tuesday Afternoon

Start	Invited?	Item
1:30 PM	Invited	A3-2-1 Missing - Pettit and Meier Invited Unknown Pettit, Unknown Meier (Consultant)
2:10 PM		A3-2-3 Thermal Conductivity of EB-PVD Thermal Barrier Coatings Evaluated by the Steady-State Laser Heat Flux Technique D. Zhu, R.A.. Miller (NASA Glenn Research Center); B.A. Nagaraj (GE Aircraft Engines) Thermal conductivity of electron beam physical-vapor-deposited (EB-PVD) ZrO2-Y2O3 thermal barrier coatings were determined by a steady-state heat flux laser technique. Thermal conductivity change kinetics of the EB-PVD ceramic coatings were also obtained in real time by measuring the temperature difference across the ceramic coating at high temperatures under the laser generated near-realistic engine temperature and stress gradients, and long term test conditions. The thermal conductivity increase due to micro-pore sintering and the decease due to coating micro-delaminations in the EB-PVD coatings were evaluated and compared for the grooved and non-grooved EB-PVD coating systems under isothermal and thermal cycling conditions. The time and temperature dependence of the coating thermal conductivity was also discussed. The test technique provides a viable means for obtaining coating thermal conductivity data for use in design, development, and life prediction for engine applications.
2:30 PM		A3-2-4 Development of a Micromechanical Life Prediction Model for Plasma Sprayed Thermal Barrier Coatings R. Vaßen, G. Kerkhoff (Forschungszentrum Jülich GmbH, Germany); D. Stöver (Forschungszentrum Jülich GmbH, Germany) A widely used method to produce thermal barrier coating (TBC) systems is the vacuum plasma spraying of a highly dense bondcoat layer with a defined surface roughness and the atmospheric plasma spraying (APS) of a porous (10-15%) Y₂O₃-stabilized zirconia top coat. In thermal cycling operation these systems often fail by crack initiation and propagation close to the bondcoat - top coat interface. This failure is attributed to stresses arising from the formation of a thermally grown oxide (TGO) layer on the rough bondcoat surface. The actual stress situation is rather complex due to TGO formation, creep effects in both bondcoat and top coat and due to the roughness of the bondcoat. All these factors have been take into account in the present work by using a finite element method (FEM) to calculate stress development during thermal loading. These data were then introduced into a crack propagation model to estimate crack development during the thermal cycling operation. The outcome of the calculations are compared to experimental results on the influence of bondcoat roughness on coating life. In these experiments TBC systems with bondcoat layers having three different levels of roughness were cycled in a gas burner rig until failure.
2:50 PM		A3-2-5 Microstructure of TBCs Grown on Rotating Substrates by EB-PVD S.G. Terry, C.G. Levi (University of California, Santa Barbara) Given the important role coating microstructure plays in the thermal and mechanical performance of EB-PVD TBCs, a better understanding of the influence of process parameters on key microstructural features of the TBC is of significant interest. The present research focuses on elucidating the underlying science of these relationships. TBCs were grown under systematically varying conditions of substrate temperature, fixed and varying vapor incidence angle (VIA), deposition rate, and elapsed time. Notable changes in the TBC microstructure (columnar grain diameter, porosity size and distribution, column tip morphology, and crystallographic texture) occur as deposition substrate temperature is increased over the range 0.40 to 0.46 T/T_M. Specifically, the column texture changes from predominantly <111> to a mixed <110> + <113>, although the dominant growth planes remain {111} in all cases. As the vapor incidence departs from normal, the columns become predominantly <110> throughout the temperature range investigated. Concomitant changes in the tip growth morphology have important implications to the shadowing processes that produce the requisite porosity. The discussion will focus on the underlying phenomena over a range of scales going from the atomistic to the mesoscopic.
3:30 PM		A3-2-7 Thermal Fracture of Interfaces in Precracked Thermal Barrier Coatings K. Kokini, A. Banerjee (Purdue University); T.A. Taylor (Praxair Surface Technologies, Inc.) Thermal barrier coatings (TBC's) make it possible to operate gas turbines, aircraft engines and diesel engines at higher temperatures, thus enabling significant improvements in the performance of these systems. The most commonly used material for TBC's is YSZ which is deposited on top of a bond coat (MCrAlY). The material property differences, the applied temperature gradients and large surface temperatures usually lead to cracking, delamination and spalling of the coatings. TBC's are deposited either by plasma spraying or by electron beam vapor deposition. The latter results in a columnar structure, which reduces the stiffness in tension and is considered to be beneficial to the life of the coating. In this study, a similar concept, developed by Praxair was used with plasma sprayed TBC's which contain cracks perpendicular to the surface of the coating. The resistance of these precracked coatings to interfacial fracture was determined both experimentally and analytically. In the experiments performed, beam shaped specimens, with crack densities varying from 0 to 42 cracks/inch, were subjected to a high heat flux generated by a 1.5kW CO₂ laser for a time interval of 4 seconds followed by natural convection cooling. In each case, the surface temperature, which resulted in interface crack initiation, was determined. Similarly, the final interfacial crack length corresponding to the different crack densities was measured as a function of maximum surface temperature. In all cases, it was shown that increasing crack density resulted in decreased interfacial cracking at the end of the thermal shock procedure. A numerical model of the experiment was developed using the finite element method and it was used to study the effect of the laser heating process on an interface crack in a precracked TBC system. The analytical results confirmed the experimentally observed behavior and explained it.
4:10 PM		A3-2-9 Influence of Impurity Content and Porosity of Plasma Sprayed Yttria Stabilised Zirconia Layers on the Sintering Behaviour N. Czech (Siemens AG, Power Generation Group (KWU), Germany); W. Malléner (Forschungszentrum Jülich GmbH, Germany); W. Stamm (Siemens AG, Power Generation Group (KWU), Germany); D. Stöver (Forschungszentrum Jülich GmbH, Germany); R. Vaßen (Forschungszentrum Jülich GmbH, Germany) Yttria stabilised zirconia (YSZ) powders from different manufactures have been used to prepare atmospheric plasma sprayed (APS) ceramic coatings with different porosity levels. While the particle morphology of the different powders was similar the amount of impurities, especially silica, was different varying between 100ppm and 1500 ppm. APS-coatings were removed from the substrates and the porosity distribution was measured by mercury porosimetry. Typically porosity levels between 10 and 15 % have been used. Free standing coatings were investigated in a dilatometer during long term (>50 h) annealing at 1200°C. Additionally, the coefficient of thermal expansion was determined from expansion during heating. It turned out that the higher porosity levels as well as the high impurity levels led to a significant increase of the sintering rate of the coating during high temperature annealing. Based on the porosity and impurity level a detailed discussion of the sintering behaviour at the beginning and at the end of the annealing process will be presented.
4:30 PM		A3-2-10 Mitigation of Detrimental Sulfur Effects by Platinum Additions to CVD Aluminide Bond Coats J.A. Haynes (Oak Ridge National Laboratory); Y. Zhang (The University of Tennessee); W.Y. Lee (Stevens Institute of Technology); B.A. Pint, I.G. Wright, K.M. Cooley (Oak Ridge National Laboratory) The adhesion of alumina scales to a metallic bond coating is often the life-limiting factor for advanced thermal barrier coating systems. It has been demonstrated that scale adhesion to both superalloys and aluminide bond coatings can be substantially improved by reducing sulfur impurities. It is also well known that scale adhesion to aluminide coatings is enhanced by Pt additions. The objective of this research was to further investigate the influence of Pt on scale adhesion to aluminide bond coatings on superalloy substrates with controlled sulfur contents. Single-phase NiAl and NiPtAl bond coats were fabricated on two types of single-crystal René N5 (low sulfur and high sulfur) by a low aluminum activity chemical vapor deposition (CVD) process. Scale adhesion was investigated by isothermal and cyclic oxidation testing at 1150°C. CVD NiAl on high-sulfur (HS) substrates exhibited very poor scale adhesion, whereas scale adhesion to CVD NiPtAl on HS substrates was excellent (i.e., comparable to CVD NiPtAl on low-sulfur substrates). Upon cooling from isothermal oxidation, scales spalled from the grain boundaries of CVD NiAl on HS substrates due to the formation of numerous large voids at the surface of the coatings. However, during an identical isothermal oxidation test, scales did not spall and voids did not form in CVD NiPtAl on HS substrates. Apparently Pt mitigates the detrimental effects of sulfur, and drastically reduces or eliminates void growth at the oxide-metal interface.
4:50 PM		A3-2-11 The Reduction of Plasma Sprayed Zirconia J. Thornton (Airframes and Engines Division, DSTO, Australia); A.J. Becker, G. McAdam (DSTO, Australia) Plasma sprayed zirconia is used extensively for thermal barrier coatings in gas turbine engines. The zirconia is usually alloyed with other oxides, such as yttria or ceria, which stabilise the zirconia in a tetragonal form. This prevents the undesirable tetragonal-to-monoclinic phase transformation. The transformation is unwanted because the accompanying strain (1.5 %) may cause the coating to spall or disintegrate. However, strains of up to 0.3% in magnitude can also be generated by heating as-sprayed zirconia alloys in air or by then heating in a vacuum or a reducing atmosphere footnote1. We report on the effects that heating in air, in vacuum, and in a reducing atmosphere had on plasma sprayed zirconia alloys. We examined the changes in the crystalline structure, local atomic structure, microstructure, and stiffness of plasma sprayed thermal barrier coatings by using X-ray and neutron diffraction, X-ray absorption fine structure, secondary electron microscopy, and resonant frequency measurements. The results indicate that the zirconia can be reduced during heating in a reducing atmosphere and also during the coating deposition process. The reduction changes the size of the unit cell and will therefore generate strains in the coating. Extreme reduction can produce severe disorder in the crystal. The effects of reduction are most pronounced in ceria stabilised zirconia, where the ceria is observed to change valency (4+ to 3+) and the coating is observed to change its stiffness. The likely consequences of these changes on the durability of thermal barrier coatings are discussed. ¹ J. Thornton, D. Cookson, and E. Pescott, The Measurement of Strains Within the Bulk of Aged and As-Sprayed Thermal Barrier Coatings Using Synchrotron Radiation, presented at ICMTF99, to be published in Surface and Coatings Technology 1999.