Accelerating the Optimization of Forging Parameters using Materials Genome Engineering Methodology

 Zaiwang Huang¹*, Jingzhe Wang¹, Siyu Zhang¹, Kechao Zhou¹, Liang Jiang²

¹State key laboratory of powder metallurgy, Central South University, Changsha, P. R. China

²Institute for advanced studies in precision materials, Yantai University, Yantai, P. R. China

ABSTRACT: Even though great progress has been made in materials science, today developing new materials from discovery to deployment often takes 10-20 years, which is time-consuming and labor-intensive. For instance, it costs roughly 18 months to design a new generation of high-pressure turbine disk of areo-engine, but the time from materials discovery to component production costs 15-20 years based on previous experience. In this regard, the innovation of products and the promotion of industrial competitiveness are closely related to the rapid deployment of new materials. For a long time, materials research heavily relies on the empirical “trial-and-error” experimentation, but in past decades, the powerful new tool dubbed by “computational materials by design” emerges to accelerate materials discovery and optimization, the successful application of computational approaches based on physics-based materials models can hugely reduce materials development time in an economic manner, this leads to a new materials research paradigm referring as “theoretical prediction, experimental validation”.

In this research, we investigated hot deformation behavior of metallic materials using double cone specimens to construct processing-microstructure-properties dataset and propose optimized processing parameters for component. In specific, we studied the hot compression behavior of a prototype powder metallurgy nickel base superalloy at a range of temperatures and strain rates. Finite element simulations (FEM) show that the effective strain range from the hub to rim region of double cone specimen is much wider than that of cylinder specimen, meanwhile the experimental observations clearly uncover that the average grain sizes from specimen hub to rim region vary due to the difference of effective strains comparing to cylinder counterpart. Moreover, we found that: (1) under higher temperature and lower strain rate, continuous dynamic recrystallization dominates the deformation process, leading to finer grain sizes. (2) lower temperature and higher strain rate, the deformed grains govern the microstructure. When the temperature is lower than 1066 ℃, cracking occurs in the rim region, which can be explained the competition between tensile strength and hoop stress predicted by the FEM technique. The benefits using double cone specimen to establish the processing-microstructure relationships cover: (1) compare the load-displacement curves predicted by FEM using experimentally measured ones and validate the flow stress data. (2) compare the specimen size change of double cone specimens between experimental and FEM, and validate the flow stress data. (3) measure the grain sizes from specimen hub to rim regions, establish the relationship between grain size and processing parameters, validate the grain size constitutive model based on processing parameters.

In summary, the experimentally validated grain size model can be used to guide the selection and optimization of processing parameters during isothermal forging in powder metallurgy nickel base superalloys, as well as other non-ferrous metals.

Keywords:  Materials genome engineering, powder metallurgy nickel base superalloys, hot deformation, grain size