Integrated Computational Materials Engineering of Fire-Resistant Steels

We explore a wide compositional space of low-carbon steel, containing 11 alloying elements, assess its feasibility for fire-resistant applications via modeling yield strength at elevated temperatures. We employ the high-throughput CALPHAD-based modeling to calculate the contributions of solid solution, precipitation, and dislocation strengthening for specific compositions. Over 30,000 yield strength predictions are made at two elevated temperatures (600 °C and 700 °C) across about 5,000 unique compositions, each with three different heat treatment conditions. We analyze the big data base and optimize using the machine-learning techniques to understand the significance of different parameters on strength. Experimental validation include thermomechanical treatments, high-temperature tensile tests, and microstructural characterizations. The newly developed alloys demonstrate a yield strength of 520–770 MPa at 600 °C, more than twice the strength of the commercial S355 steel. This approach facilitates the rapid discovery of novel fire-resistant steel compositions and has a high potential for other alloy systems.