Jacob Startt1 Tanvi Dave2 Eric Hoar2 Hamid Garmestani2 Chaitanya Deo1

1, Georgia Institute of Technology, Atlanta, Georgia, United States
2, Georgia Institute of Technology, Atlanta, Georgia, United States

In this work, we attempt to show that microstructural texture analysis coupled with optical microscopy and morphological analysis can be used to directly elicit and quantify the process-path history of a Zirconium-Niobium alloy. We do this by first determining the processing-path relationships belonging to a specific process by performing the experimental processing steps and measuring the texture evolution for a series of samples. We then use those relationships as input to a reverse process-path model that allows us to predict the initial microstructure and processing history of the sample. Lastly, we show that programs such as the Visco-Plastic Self Consistent (VPSC) model can be used to accurately simulate the texture evolution during processing, thereby reducing the need for experimentation and opening this method up to material systems in which the necessary processing-path relationships may not be readily available.

We do this by incrementally warm-rolling monotectoid Zr-18.8w%Nb, and measuring the texture evolution through both X-ray Diffraction (XRD) and electron back-scatter diffraction (EBSD) methods. As these methods provide slightly different views of the texture, we attempt to quantify this difference in relation to the reverse process-path model. We also perform several VPSC calculations on the BCC Zr-Nb system in which we attempt to match the experimental texture growth by varying the active slip systems and hardening laws. In general, we find that slight variations in the processing paths might significantly affect the final texture but this does not prevent us from acquiring an adequately accurate process-path from our model. For instance, in the case of warm-rolling, the process of repeatedly heating the sample before each roll introduces a recrystallization effect which appears in the formation of a strong γ-fiber in the orientation distribution functions (ODFs), particularly at the {111}<112> and {111}<110> components. However, the typical BCC rolling textures generally remain along the α- and θ-fibers and at the rotated cube components. Similar results were also found from VPSC simulations as the variations in processing made it difficult, but not impossible to reproduce the texture evolution. It was found that if the commonly active BCC slip systems are accounted for the texture can generally be fairly well predicted as long as the hardening law components are chosen carefully to best represent the process and material system.