Laser Assisted Forming and Phase Reversal Heat Treatment of Metastable Austenitic Stainless Steels

Harm Kooiker is a PhD student in the department of Mechanics of Solids, Surfaces & Systems (MS3). His supervisor is A.H. van den Boogaard from the Faculty of Engineering Technology (ET).

Laser assisted forming of sheet metal is an innovative processing technology capable of increasing the formability of the material and increasing the accuracy of the assisted process. A typical process sequence consists of local heating of the forming zone, followed by forming whilst the material is still hot.

Cold rolled grades of metastable austenitic stainless steel have a dual-phase composition of austenite and martensite owing to the martensitic phase transformation upon cold rolling. Hot forming of this material induces several complex processes that affect both the microstructural and macroscopic response, of which the most important are phase reversal transformation and discrete dynamic recrystallization.

Designers tasked with developing a robust, first-time-right, laser assisted forming process require both fundamental insights into the aforementioned transformation and recrystallization and constitutive equations describing the material behaviour during the process to enable finite element-based process design. The first step in developing constitutive equations is to perform suitable experiments to characterize the material behaviour. In the case of cold rolled austenitic stainless steel, the characterization has to be performed at high heating rate and high strain rate. Therefore, a custom laser-integrated tensile test setup was designed and constructed, capable of meeting these requirements. A novel system was installed to measure the local strain field during hot tensile deformation by means of digital image correlation, resulting in the capability to measure the strain development over the entire surface of the specimen with a high spatial resolution. With the new setup, a series of experiments were performed to characterize both the phase reversal transformation and subsequent hot forming behaviour.

During phase reversal transformation, mechanically induced martensite is transformed back to ultrafine-grained austenite of (sub)micron scale. It is widely accepted that there are two mechanisms of phase reversal transformation: shear-type reversion and isothermal transformation. Shear-type reversion is an athermal transformation and depends solely on temperature, whereas isothermal reversion depends on both time and temperature. The results of heat treatment experiments performed on the new laser heated setup reveal that, contrary to some interpretations, reverse transformation of AISI 301 is not dictated solely by an athermal mechanism. Instead, the transformation depends on the heating rate and can be better described by a combination of isothermal and athermal transformation. A new rate-dependent kinetic model for reverse transformation was developed and it was shown that it is capable of describing the transformation experiments accurately.

Using the newly developed setup, hot tensile tests were performed for a range of temperatures and strain rates. Results prove that hot forming enables excellent formability, even at high strain rates. Microstructural analysis reveals that the behaviour during hot forming is governed by dynamic recrystallization. Importantly, it is revealed that the ultrafine-grained austenite, resulting from reverse transformation, is subject to fast grain growth during the heating stage of a hot forming process, whereas the evolution of the grain size is dominated by dynamic recrystallization during hot forming.

A new rate-dependent time-based continuum model for dynamic recrystallization was developed. It was designed to describe the effect of dynamic recrystallization on the yield stress for a wide range of temperatures and strain rates. A new explanation is offered for fast recrystallization observed during high strain-rate hot forming. It is proposed that the driving pressure for recrystallization is not only dependent on the elastic energy caused by dislocations, but also depends on the elastic energy caused by dynamic stress. The model was calibrated to experimental data of two steels, showing good agreement for a wide-range of temperatures and strain rates including high strain rate. The dynamic recrystallization model was further extended to be able to account for the effect of varying initial grain size. This was achieved by making both nucleation and growth rate of the model, dependent on the availability of grain edges. It is shown that this extended model is capable of accurately describing the effect of initial grain size on the kinetics of dynamic recrystallization for a wide range of initial grain sizes without requiring any additional parameters.