Creep behaviour investigation of additively manufactured IN738LC superalloy based on Materials Genome approach

Abstract

The additively manufactured Ni-based superalloy IN738LC holds significant potential for applications in aerospace high-temperature components due to its exceptional creep properties. However, a limited understanding of the high-temperature creep behaviour impedes its engineering applications. This research delves into comprehending the creep behaviour of additively manufactured Ni-based superalloy IN738LC by integrating the Materials Genome Initiative concept with high-throughput creep experiments and machine learning. The samples of this typical high cracking tendency alloy are prepared using the laser powder bed fusion process, and then the printed microcracks are entirely eliminated through the liquid-induced healing strategy. Advanced high-throughput compression creep tests are conducted under 24 creep conditions, revealing superior creep performance compared to existing Ni-based, Co-based, and Ni-Co-based superalloys. Based on the P-parameter method and deep learning techniques, predictive models exhibit excellent alignment with experimental data, thereby enabling creep behaviour prediction under any temperature and stress conditions. Microstructural examination has shed light on the complex interactions of dislocations with twin, crystal defects and precipitates, which collectively underpin the enhanced creep resistance. This research has provided valuable insights into the creep behaviour of additively manufactured IN738LC superalloy. Moreover, we have established a pathway integrating high-throughput creep testing with machine learning within the framework of the Materials Genome Initiative for materials investigation. This approach offers an efficient method for constructing models to predict creep behaviour and potentially can be applied to other materials.