High-Throughput Computational Screening, Non-Metallic Inclusion Analysis, and Microstructural Characterization of a Novel Medium Manganese Steel

High-throughput computational screening (HTCS) based on CALPHAD (Calculation of Phase Diagram) was employed to investigate potential chemical compositions within the medium manganese steel family for achieving desired austenite stability and stacking fault energy (SFE). The primary objective was to identify optimal alloy compositions that balance the complex effects of various alloying elements on retained austenite fraction and related mechanical properties. Utilising TC-Python Thermo-Calc software coupled with a custom-developed algorithm, two optimised compositions were determined: 0.35C, 9Mn, 1Mo, 3Al, 1Si, 0.05Nb, 0.3V (alloy 353), and 0.35C, 9Mn, 1Mo, 3Al, 1Si, 0.1Nb (alloy 310) in wt.% to be the best fited composition to our selected criteria. The alloys were subsequently produced via open-air induction furnaces, and the microstructure was analysed after the hot forging condition. The initial multiphase as-cast structure, primarily composed of lath martensite, δ-ferrite (34 vol.%), and retained austenite (RA, 5–7 vol.%), experienced notable grain refinement. Forging reduced δ-ferrite grain sizes from 39 µm to 12 µm (alloy 310) and from 46 µm to 9 µm (alloy 353), accompanied by increased RA content (28 vol.% for alloy 310 and 46 vol.% for alloy 353) and reduced RA grain sizes (1.2 µm and 1.9 µm, respectively). Non-metallic inclusions (NMIs) were analysed using field emission scanning electron microscopy coupled with energy dispersive X-ray spectroscopy, classifying inclusions primarily as AlN, MnS, (Mo,Nb)C, or their combinations. No significant differences in inclusion types were observed, but forged samples displayed reduced inclusion sizes. In summary, the results showed that HTSC effectively identified optimal compositions with a high fraction of retained austenite.