Design of Oxidation Resistant Alloys using Combinatorial Approaches with Chemically Graded Materials

Abstract This work introduces a new high-throughput method to characterize the oxidation behavior of chemically graded Ni-based alloys in order to feed databases destined to numerical metallurgy approaches. A Ni-wCr-3Al (w ∈ [0, 30]) chemically graded material was obtained from two homogeneous samples by a diffusion couple method at 1300°C for 100h. The composition range was selected in order to observe the three types of oxidation behavior identified in the reference work of Giggins and Pettit [10]. The excellent agreement between simulated and experimental diffusion profiles validated the experimental method used to manufacture the chemically graded material (CGM). The CGM was then oxidized at 1200°C in air. Surface and cross-section characterization were conducted by SEM/EDS and Raman spectroscopy to identify the oxides formed on the CGM. To accelerate the Raman characterization treatment, a method linking Principal Component Analysis (PCA) and K-means unsupervised clustering algorithm was developed [11–12]. It allowed for the identification of the oxide type without peak indexation issues and is well-suited for CGM. These results show that results similar to well-recognized reference experiments [10] can be achieved using only one CGM.