Combinatorial optimization of carbide-free bainitic nanostructures

Thermodynamic calculations in combination with a neural network model are employed to predict the conditions under which nanostructured carbide-free bainite can be formed. The method recovers well the conditions under which the alloys reported in the literature display such features. Aluminium and silicon are shown to be equally effective in suppressing cementite. Manganese reduction appears to be the most effective means to accelerate bainite formation at low temperatures. A new low-manganese high-chromium steel grade capable of transforming into a nanostructured carbide-free structure is proposed, in which thermokinetic calculation and experiment show that low-temperature bainite forms faster and displays greater hardness than the alloys previously reported in the literature.