Quantifying Processing Map Uncertainties by Modeling the Hot-Compression Behavior of a Zr-2.5Nb Alloy

Abstract

Compression dilatometer tests were used to study the hot deformation response of a Zr-2.5Nb alloy over the temperature range 650°C – 850°C and strain rates of 10 #$.& s #) – 10 +) s #) . A high number of test conditions was used (72, with every test duplicated), in order to assess how differences in data processing influence the resulting relationships between flow stress, temperature and strain-rate. Particular attention was paid to ‘processing maps’, showing strain-rate sensitivity over the processing domain, commonly cited in the field and widely used as a basis to determine optimum processing conditions. Significant variations in these maps were found to depend on the number of data-points included and the fitting procedure used to smooth the data. A finite element (FE) model of the test demonstrates the order of the corrections that can be required to the flow stress and the consequent processing maps due to friction at the platen-workpiece interface, and non-uniform temperature and deformation in the test piece. Changes in crystallographic texture, measured using electron-backscatter diffraction (EBSD), illustrate the effect of temperature, strain and strain-rate on the deformation, phase transformation and recrystallization mechanisms. A significant spread in response arises as a result of variation in micro-texture between samples and the tendency for flow to localise, giving rise to scatter in the measurements and generating artefacts in the processing map. Although the processing map methodology is strongly affected by experimental uncertainty, a detailed analysis of the final microstructures in the test samples show similar features to those produced during industrial-scale processing, providing insight into the deformation mechanisms in dual-phase Zr-Nb alloys.