Data-Driven Design and Rapid Evaluation of Novel Thermal Barrier Coatings

Thermal barrier coatings have been an important component of aero and gas turbine engines since the 1980’s. The fundamental purpose being to enable higher operating temperatures to improve efficiency, durability, and power output of the engine. It is desirable that thermal barrier coatings have low thermal conductivity, high toughness, high thermal cyclability, and high resistance to the many chemistries of sand (i.e., CMAS). Two of the most used materials, YSZ and Gd₂Zr₂O₇, were developed in the 1980’s and early 2000’s; respectively. While many other classes of materials have been evaluated as potential thermal barrier coatings, few have been adopted commercially. With the recent development of the high entropy oxide field, the number of possible compositions has increased exponentially such that this multi-objective problem is intractable for human intuition alone. In this presentation, it will be demonstrated that the conventional wisdom of thermal barrier coating composition design can be enhanced by employing materials informatics. This is accomplished using a combination of thermodynamic modeling, artificial intelligence, and high-throughput experimental validation for refining of the models.