Computational Design of Novel Ni Superalloys with Low Crack Susceptibility for Additive Manufacturing

Additive manufacturing (AM) has enabled the fabrication of innovative, geometrically complex components of metallic materials with minimal material waste. However, the application of AM in processing Ni superalloys remains challenging because of the high crack susceptibility. To address this problem, a computationally design method was applied to design novel Ni superalloys with low crack proneness as well as decent service performance. Different types of crack formation mechanisms during AM processing, i.e., the hot tear cracks and strain-age cracks, have been considered, while a series of corresponding criteria to quantitively describe the crack susceptibility were parallelly analyzed. To select the most proper criteria as optimization index in alloy design regime, a series of prototype alloys with wide range of alloying elements were printed, with their crack behaviors being quantitively captured. The freezing range criterion and strain-age crack criterion were then determined as the optimization index in the design routine. A genetic algorithm was applied to probe the compositional search domain, while a Pareto front linking the hot tear crack criterion and strain-age crack criterion was constructed by combining results of two separate optimization routes. Novel Ni compositions with optimized crack resistance as well as decent microstructural features were thereby identified. The new solutions are predicted to possess fairly better crack formation resistance compared to existing Ni superalloys. Moreover, the newly designed alloys manage to computationally outperform the existing printable alloys in all aspects of temperature capability, creep properties and oxidation resistance.