Nanoindentation with a Knoop indenter tip can reveal the plastic response of brittle materials, which is contained in the measured Knoop hardness anisotropy and force–displacement curves. Due to short length scales, details of the local dislocation distribution and dislocation stresses cannot be ignored in the analysis of experimental measurements. Inclusion of such effects leads to a non-local theory of plasticity, where gradients of strain fields are present in the expressions for stress. In this work, a non-local plasticity model is developed and implemented in the Abaqus finite element software. The geometrically necessary dislocations (GND) are quantified via the Nye tensor, and the backstress tensor is found from gradients of the Nye tensor by summation of dislocation stresses. Micro- and nanoindentation experiments on pentaerythritol tetranitrate (PETN) are simulated with the developed model. The average errors between simulated and measured micro- and nanohardness are around 13% and 25%, respectively, while the maximum errors are around 20% and 30%, respectively. The set of active slip systems for PETN that can match the experimental hardness anisotropy trends is {110}⟨11¯1⟩, {100}⟨011⟩, and {101}⟨101¯⟩. The GND hardening, the backstress, and the indenter shape are investigated in relation to the indentation size effects. The model predicts a weak effect of backstress on hardness, while the coupled effects of indenter shape and GND hardening are predominantly responsible for predicted size effects.
Milovan Zečević, Morgan C. Chamberlain, Alexandra C. Burch et al.· Modelling and Simulation in...· 0 citations
Predicting thermal stability during handling and storage is essential for the design of safe and reliable energetic materials. However, experimental measurements vary significantly across laboratories due to differences in protocols and analysis methods, making it difficult to train reliable predictive models. We address this challenge through differential learning. Rather than predicting absolute decomposition temperatures, we instead train message passing neural networks to predict relative differences between pairs of molecules. This approach reduces sensitivity to systematic experimental errors and achieves>85% accuracy in ranking compounds by thermal stability, outperforming conventional regression methods on the same heterogeneous dataset. To understand what drives these predictions, we compare neural network models with interpretable alternatives built from descriptors derived from ab initio calculations and cheminformatics software. This analysis identifies bond dissociation enthalpy as a key determinant of thermal stability rankings, providing further insight into the complex chemistry of thermal decomposition. The differential learning framework generalizes across model architectures, from graph neural networks to classical descriptor-based approaches. Our results demonstrate that learning relative properties rather than absolute values offers a practical solution for modeling noisy experimental data, with direct applications in materials design where thermal stability predictions inform safety protocols.
M. Davis, R. Ullberg, J. Schroeder et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
