Physical metallurgy of rolling and finishing : precipitation behaviour in the ferrite range of soft microalloyed ULC steel grades

Thin sheet material for the automotive industry, produced with an excellent balance between strength level and formability and an additional bake hardening behaviour, is still gaining importance. One possible approach to meet this goal - particularly for the production via a hot-dip galvanising line - is the use of partially stabilised ULC grades. Due to the very low alloy content of this material the precipitation of carbides, which controls the carbon content in solid solution and thus the BH level, takes place in the ferritic range. In comparison to the austenitic range, few investigations have been carried out in the past for the ferritic range. During this project such partially stabilised thin sheet material was produced on an industrial scale as well as via pilot lines and annealing simulators. Additionally, the hot-rolling step was simulated with compression tests. The microstructure and mechanical properties were investigated in detail on the samples. Particular attention was given to the characterisation of the precipitates with TEM and the measurement of the carbon content in solid solution. Simultaneously thermodynamic and kinetic modelling of the precipitates was performed and the results were compared with the experimental results from the TEM investigations and the results from the internal friction measurements. A thermodynamic and kinetic model was developed in the course of the project and the agreement between the results from experimental investigations - mainly from TEM and internal friction measurement - and modelling were acceptable in general. The experimental results show that the C content in solid solution and hence the BH behaviour are mainly determined by the chemistry; the other processing parameters such as coiling temperature, cooling rate on the run-out table and the annealing parameters have a minimal influence. However, the mechanical properties such as strength level and particularly the r-values are significantly influenced by the processing parameters.