MICROSCALE TESTING AND CHARACTERIZATION TECHNIQUES FOR BENCHMARKING CRYSTAL PLASTICITY MODELS
Johns Hopkins University
The desire to improve the performance of engineering alloys and introduce new materials into service has led to the development of advanced, multi-scale material property models that can accurately predict the deformation response of polycrystalline microstructures. These microstructure-dependent, multi-scale models have the ability to provide insight into the connections between material processing, microstructure and properties in a way that has not been available before. However, these advanced modeling techniques require microstructural characterization and experimentally obtained benchmarks at salient length scales. Accordingly, microtensile tests of the polycrystalline Ni-base superalloy René 88DT have been carried out in order to guide and benchmark parallel crystal plasticity finite element method (CPFEM) modeling of this material at appropriate length scales. Microscale machining processes, including wire electrical discharge machining (EDM), focused ion beam (FIB) and femtosecond laser machining, have been developed and optimized for machining microtensile samples across multiple sizes. Loading in uniaxial tension provides the full stress-strain behavior from which quantitative mechanical benchmarks such as yield strength, strain hardening, and modulus can be extracted. The effect of sample size was studied to observe the underlying effects of microstructural variations. It was found that average sample strength decreased, and stochasticity of strength increased, as sample size decreased, owing to a finite sampling of grain orientations with a biased distribution towards higher Schmid factor values for grains in a randomly textured FCC material. In addition, local strain accumulation on the surface of tested oligocrystalline samples, with a computationally tractable number of grains, has been measured through the use of 2D digital image correlation (DIC). It was observed that strain concentrations formed in regions of the microstructure where there was a significant mismatch in Schmid factor and elastic modulus across grain and twin boundaries, a microstructural feature that leads to local stress concentrations. These observations help to guide model development in highlighting deformation mechanisms in the material, and the developed strain maps provide both quantitative and qualitative benchmarks that can be directly compared with modeling results. The scale of these experiments allows for 3D characterization, via serial sectioning and electron backscatter diffraction (EBSD), of tested samples through collection of critical microstructural data, including size, shape and orientation of grains and twins within the tested volume. Experimentally capturing explicit microstructures, at a scale that is also computationally tractable in crystal plasticity modeling, and their attendant mechanical behavior highlights stochastic nature of plasticity in small volumes and provides quantitative metrics for model development.