Constitutive modelling of the strain hardening behaviour of metastable AISI 301LN austenitic stainless steel as a function of strain and temperature

Abstract

The automotive industry currently demands materials with improved formability and crash performance. Austenitic stainless steels have been singled out for potential development of high strength steels to achieve exceptional combinations of strength and ductility due to their high strain hardening abilities. Austenitic 301LN stainless steel is the least alloyed and most metastable among the 300-series austenitic stainless steels. Plastic deformation at a temperature below Md transforms the metastable austenite phase to a thermodynamically more stable martensite phase accompanied with enhanced strain hardening. There are two different deformation mechanisms of austenitic stainless steels which are TRIP and TWIP effects. In this work, both mechanisms were observed at different deformation temperatures with both phase transformations and twin formations contributing towards the strain hardening. This research work concentrated largely on the derivation of constitutive equations of both volume fraction of martensitic transformation and mechanical response of a metastable austenitic stainless steel alloy as a function of applied strain and temperature. The alloy investigated was a lean version of AISI 301LN. A calibration evaluation of the Ferritescope values was performed with the use of magnetizing tests (VSM), X-ray and neutron diffraction analyses to arrive at a reliable methodology for the determination of the martensite content during the tests. A calibration factor of 1.70 was obtained when tensile deformed samples were used (with analyses done using Vibrating Sample Magnetometer measurements, X-ray and Neutron diffraction techniques) and a calibration factor of 1.62 was obtained when cold rolled samples were used with analysis done using the neutron diffraction technique only. A series of interrupted uniaxial tensile tests at temperatures ranging between -60 and 180 °C at a constant strain rate of 6.67 x 10-4 s-1 were performed. A low strain rate and a small interruption interval were chosen to minimize the heating effect due to adiabatic heating. The strain hardening behavior of AISI 301LN metastable austenitic stainless steels was observed to be a complex process which is related not only to the generation of a dislocation structure but transformation and twinning hardening as well depending on the deformation temperature. Strain hardening curves were derived at different temperatures and were found to be following the same basic mathematical equation for the formation of the strain-induced martensitic transformation product as a function of true strain. Prior cold rolling was also done to different gauges ranging from 5% and 70% at ambient temperature, with small reduction passes applied to minimize adiabatic heating. A series of interrupted uniaxial tensile tests were done on the prior cold rolled samples at a low strain rate of 6.67 x 10-4 s-1. All the derived strain hardening curves were extended up to the true strain levels of 1.0, to arrive at estimates of the strength coefficient, K. The strength coefficient, K was found to be in the range of 1500 MPa ~ 1780 MPa, as calculated from the convergence of sigmoidal hardening curves at a log stress of ~ 3.25 (at log true strain of 0).This was found to be in accordance with the tensile strength of 1715 MPa after a cold rolling of 63.2% (which is equivalent to the compressive true strain of 1.0). The calculated values of strain hardening and martensite formed, using the developed constitutive equations, agree well with the experimental results for a wide range of deformation temperatures and prior cold rolling percentages. A linear variation of stress as a function of strain-induced martensite, observed at moderate martensite fraction levels, was explained as being due to the dispersion hardening effect. An abrupt change from a linear variation that occurred on exceeding a threshold value of martensite formed, was believed to be due to “percolation effect of martensite,” where clusters of martensite forms a continuous network linking up in 3D, adding more blockage to dislocation movement in the austenite phase. A percolation threshold of martensite was found to be in the range of 30 ~ 45%. This was found to be the percentage of martensite present when the rate of martensitic transformation reaches a maximum. At the percolation threshold of martensite, there is an interchange of roles of martensite and austenite, where martensite behaves as a matrix phase and austenite as dispersions embedded in the martensite phase. This results in higher strength as more stress is required to move the dislocations past the percolated martensite barriers which also reduces the plasticity of the austenite.