Design rules for failure-resistant materials

Ziel

Material failures caused by long-term static or cyclic loading are among the most serious challenges in materials engineering. They account for up to 90 % of structural failures and result in billions of dollars in annual economic losses worldwide. The gradual accumulation of microscopic damage often goes undetected until catastrophic failure occurs, and despite decades of study, these processes are still described mainly by empirical laws that reproduce behavior but offer no mechanistic path to prevention. The fundamental challenge is to move from describing when materials fail to understanding how deformation evolves and how it can be controlled through design.

This transition has been blocked by a key barrier: deformation involves many interacting microscopic events that are experimentally observable but whose causal relationships (how events trigger others) have been inaccessible, preventing the development of predictive, physics-grounded design rules. Only now, through the convergence of high-resolution imaging, physics models of collective deformation, and modern causal-inference algorithms, can these mechanisms be reconstructed directly from experiments.

I will overcome this barrier by establishing a physics-based Descriptors-Mechanisms-Design pipeline that transforms deformation data into actionable design principles. Advanced imaging and statistical-physics analysis identifies robust deformation descriptors, physics-informed causal inference reconstructs how local deformation events interact to drive or stabilize failure, and these insights yield microstructural design rules validated through proof-of-concept materials.

Together this will establish a preventive, physics-informed approach to material design, transforming failure prediction into failure resistance. The outcome will be a general method for creating durable, sustainable materials and a conceptual breakthrough linking deformation physics, causal reasoning, and materials engineering.

Wissenschaftliches Gebiet (EuroSciVoc)

Dieses Projekt wurde noch nicht bei EuroSciVoc klassifiziert.
Schlagen Sie die Wissenschaftsbereiche vor, die Ihrer Einschätzung nach besonders relevant sind, und helfen Sie uns, unseren Klassifizierungsdienst zu verbessern.