Grain size effects in micro-tensile testing of austenitic stainless steel

Abstract

A micro-tensile test system is under development at the National Institute for Standards and Technology (NIST) and is based on tensile test techniques developed for thin films, but it provides the capability of testing samples from bulk material. Example specimen dimensions are a gage section length of 342 microns, width of 75 microns, and thickness of 80 microns. The micron size scale of the specimens presents the potential to analyze local deformation of a material within a specific region of the microstructure. The micro-tensile system could also be useful in material modeling, because samples can be constructed to match micromechanical modeling simulations of specific microstructures and microstructure size scales. The primary objectives of this research are to fundamentally understand the grain size effect on micro-tensile testing results and to further the goal of using micro-tensile technology as a tool to study mechanical behavior of small volumes of material with minimal size effects. 316L stainless steel was selected as the candidate material because of its austenite stability. Both ASTM E-8 rectangular sheet-tensile specimens and micro-tensile specimens were tested for mean grain sizes of 3.1 microns, 9.5 microns, and 17.5 microns, which were obtained through heat treatments of 40% and 85% cold reduced sheet. The grain sizes were selected such that there was wide range in the number of grains in the micro-tensile specimen cross-section. For all three grain sizes, the yield stresses, flow stresses, and strain hardening rates of the micro-tensile specimens were less than those of the conventional tensile specimens. A specimen size effect scaling model suggests that the lower flow stresses and lower strain hardening rates are attributable to the high grain size to specimen size ratios of the micro-tensile specimens. The scatter in the yield stress and flow stress data was higher for the micro-tensile specimens, but the amount of scatter did not show a strong dependence on grain size. It was expected that the scatter would increase with an increasing grain size to specimen size ratio, but high uncertainty in the calculated stress from the micro-tensile tests could also contribute to the observed scatter. Electron back scatter diffraction (EBSD) was used to calculate the possible standard deviation in the minimum average Taylor factor in the cross-section of the micro-tensile specimens. The increase in the standard deviation of the minimum average Taylor factor with increasing grain size to specimen size ratio could contribute to the observed scatter in the yield stress and possibly the flow stress of the micro-tensile specimens. Recommendations for improving precision and accuracy of measurements associated with the micro-tensile test system are also discussed.