QUBE: QUasi-Brittle fracture: a 3D Experimentally-validated approach

Ductile materials, like metals and alloys, are widely used in engineering structures either by themselves or as reinforcement. They usually can sustain a lot of plastic damage before failing. Engineers understand quite well the ways that metals fail and how tolerant they are to damage, so efficient and less massive structures may be designed with well-defined margins of safety or reserve strength to cope with extreme events. By comparison, elastic brittle materials such as glasses and ceramics can fail without prior warning, so much larger safety margins are needed.
Quasi-brittle materials are an important class of structural materials. They are brittle materials with some tolerance to damage and include concrete, polygranular graphite, ceramic-matrix composites, geological structures like rocks and bio-medical materials such as bone and bone replacements. Although their damage tolerance is much less than many metals and alloys, it can be quite significant compared to brittle materials such as ceramics and glasses. But this is not accounted for very well when engineers design with, or assess, quasi-brittle materials, as there is not an adequate understanding of the role on their damage tolerance of factors such as the microstructure of the material or the state of stress. Quasi-brittle materials are usually treated as fully brittle, taking little or no account of their damage tolerance, so assessments incorporate very significant safety margins, leading to designs that may be inefficient and unnecessarily bulky. Even when some assessment of damage tolerance is included, the microstructure can change as the material ages over time, and we need ways to measure the effects of this and to predict what it will do to the safety of the structure. This project aims to develop a method to predict the performance and evaluate the integrity of structures and components made from quasi-brittle materials. This will extend opportunities for their use in engineering applications, enabling more efficient design with greater confidence in safety.
Quasi-brittleness is a property that emerges from the material's microstructure. A quasi-brittle material can be made from a connected network of very brittle parts (for instance, a porous ceramic). It exhibits a characteristic "graceful" failure as parts break locally when loaded sufficiently, which gives it damage tolerance. The "gracefulness" of the failure is affected by the random variations of strength and stiffness of the network and the form of the connections. Such networks represent a key part of the microstructure of the material, and to understand quasi-brittle fracture we need to construct models that properly describe the microstructure. There is a need to understand and define the mechanisms that control the fracture at the small and the large scale within these quasi-brittle materials. This will allow us to capture sensitivity to microstructure differences and degradation, and to produce general models that are suitable for the wide range of quasi-brittle materials and applications.
Three-dimensional models that are faithful to the microstructure can be created using modern 3D microscopy methods, such as X-ray computed tomography. But these models are far too complex to simply scale up to structures very large relative to the microstructure. There is no computer than can do this, yet. We will develop modelling methods that sufficiently represent the complexity of quasi-brittle microstructures over a wide range of length scales, such as cellular automata finite elements. We will use advanced tomography and strain mapping techniques to observe how damage develops and to test and refine our models. We will then use this and the understanding that we gain to design new material tests and characterisation methods so that our methods may be used in a wide range of materials, from concretes to advanced nuclear composites, bone replacement biomaterials and geological materials.

By improving basic understanding of damage mechanisms, we will create a framework to predict fracture behaviour and strength of quasi-brittle materials, validated by unique experimental observations of damage development at the microstructural scale. The challenge of different length-scales will be overcome by protocols for large-scale cellular automata finite element simulations of damage from high fidelity, three-dimensional descriptions of microstructure and damage mechanisms.
In a specific case, the low carbon nuclear energy sector will benefit in the UK and internationally, with impact on environmental sustainability and protection. New data and understanding in AGR nuclear graphite are relevant to the UK regulator and EdF Energy. Obtained for graphite without irradiation or oxidation, it supports confidence in the conservatism of integrity assessments for stress concentration features such as moderator keyways, whose predicted failure may limit reactor life. Our recommendations on small specimen property measurement and microstructure characterization to calibrate models for highly oxidized graphite will be sufficiently timely, with development, to support AGR safety cases for life-extension in the next decade, making best use of core inspection samples and accelerated aging tests. Graphite is also critical for future nuclear plants: USA NGNP (Next Generation Nuclear Plant); European GFR (Gas-cooled Fast Reactor: Allegro); and Pebble Bed Modular Reactor (the South African PBMR is postponed, but construction continues in China). Assured life without graphite core replacement is critical to their economic development. Improved design methods, underpinned by this project's outputs, will increase confidence in structural integrity and improve the design of specimens to monitor material ageing. For the Magnox stations improved mechanistic understanding will provide further confidence in operation to end of life.
Our methodology applies to other quasi-brittle materials; SiC-SiC composites (e.g. fuel sleeves), concrete containment structures (including for nuclear waste), mining technologies such as rock drilling and tunnel structural integrity as well as oil and gas recovery via methods such as hydraulic fracturing. Designers of synthetic biomaterials, e.g. bone replacements, require large-scale simulations of the integrity of components with complex geometry and porous microstructures to optimize performance within constraints of size, shape and weight. Benefiting from our methods of predicting material strength under complex stress states by models derived from high-resolution characterization of microstructure, this will have significant potential impact on the health and life quality of the general public.
The postdoctoral researchers will gain enhanced skills, benefiting from our cross-disciplinary approach with real links between groups expert in modelling and experiment. Collaboration will enhance all our skills and capabilities, sharing unique facilities, innovative modelling tools, data and giving exposure to wider networks of academic and industrial partners.
We will publish in international journals and conferences relevant to engineering structural integrity and computational modelling of materials. We will disseminate outputs to industry and its stakeholders through our interactions with regulatory advisory groups and industrial networks, also through a workshop on quasi-brittle material fracture; a successor to the recently held UK Forum for Engineering Structural Integrity (FESI) workshop, 'Modelling fracture in quasi-brittle materials: achieving consistency between different length scales'. We will start the process of developing design guidelines through the international networks that create engineering standards. Publications, open-source software codes, guidelines and example applications will be public on the web, with explanations of our methods and motivations suitable for the general public.