Date of Completion

4-20-2018

Embargo Period

4-20-2018

Keywords

Thermal Barrier Coating, TBC, CMAS, Thermal Gradient Testing

Major Advisor

Eric Jordan

Associate Advisor

Maurice Gell

Associate Advisor

Mark Aindow

Field of Study

Materials Science and Engineering

Degree

Doctor of Philosophy

Open Access

Open Access

Abstract

Thermal barrier coatings (TBCs) are critical in modern gas turbine engines, increasing operating temperature, efficiency, and component life. Engines have reached temperatures at which ingested debris (CMAS) forms silicate melts that chemically and mechanically attack TBCs, leading to premature failure. New, CMAS-resistant coatings must be validated under conditions that recreate real-world TBC-CMAS interactions. No standardized testing to perform these analyses currently exists.

A cyclic thermal gradient rig with incremental CMAS deposition was developed based on modified literature designs. Tests performed using literature-based parameters showed TBC-CMAS interactions and failure morphology deemed not real-world representative by an engine manufacturer. The results were rationalized with the understanding that, in an engine, “cooling air” is relatively hot (~400 °C). Minimizing transient thermal gradients across the TBC coupon resulted in more representative test outcomes.

A second generation thermal gradient rig was developed with higher heat flux and sample throughput. This rig incrementally deposits CMAS powder, a feature not found on existing rigs. Heterogeneous CMAS was deposited onto an EBPVD YSZ TBC coupon. The CMAS layer had intriguing non-uniformities in accumulation and chemical heterogeneity. This has implications for CMAS materials used for testing. Engine manufacturers need to model TBC life reduction from CMAS attack for different engine parameters and CMAS environments.

Preliminary experiments were performed that provided insight for such models. First, EBPVD YSZ coupons were cycled with varying CMAS dose rates. Over the range investigated, the lightest CMAS dose rate used 80% less CMAS to cause failure compared to the heaviest rate, disproving the concept of a “critical CMAS dose” for failure. Rather, failure is a mix of cycling and CMAS damage. As hot time and the number of thermal cycles increase, TGO growth and accumulated damage effectively reduce the toughness of the coating, making it more susceptible to spallation with less CMAS penetration. Second, differences between testing with homogeneous and heterogeneous CMAS were investigated on APS YSZ TBCs. While failures were similar, partial-life microstructures revealed differences in melting kinetics, which may have implications on how reactive TBC compositions interact with CMAS.

COinS