Dislocation based modelling of deformation and fracture in real engineering alloys

ad52cded-23fe-4833-94cc-5dcac9aaec39

Closed

EPSRC
EPSRC
EPSRC
EPSRC
EPSRC

£3,340,260

Nov. 1, 2015

Jan. 31, 2021

If loaded a small amount, a metal will deform elastically, returning to its original shape when the load is removed. However if the load exceeds some value, then permanent deformation occurs, known as plasticity. Plasticity is far more complex to understand than elasticity as it involves breaking lines of atomic bonds in the metal. These lines of broken atomic bonds are called dislocations. This is analogous to the motion of a caterpillar: which does not attempt to move its whole body forward simultaneously; instead it incrementally moves its body forward in a wave of motion sweeping through the caterpillar's body. Metals contains a huge number of dislocations: these lines sweep through the metal allowing atomic planes to slip over each over, causing the metal to be permanently deformed. When metal is loaded, new dislocations are nucleated and some become trapped at obstacles. However, if the load is applied too quickly or the metal is too cold, then the dislocation lines do not have time to nucleate and move: instead whole planes of atoms are ripped apart, fracturing the metal.

In a nuclear reactor, the fuel rods are cladded in a zirconium alloy: over time, hydrogen from water used to cool the fuel rods, diffuses into the zirconium and is attracted to dislocation lines and to any small cracks or notches in the metal. If the hydrogen concentration becomes too high, hydrogen atoms will clump together to form precipitates which block dislocation motion and can easily fracture.

It is this complex interaction between, dislocations, diffusion, precipitate formation and fracture which I aim to simulate on a computer. This is possible by utilising the power of modern graphics cards (developed to play video games) which allow massively parallel simulations to be performed easily and at little cost. Even then it is only possible to simulate a very small volume of material. Traditional mechanical tests (bending or compressing pieces of metal) were always performed on large specimens, several millimetres in size, meaning it was simply not possible to simulate all the dislocations in the sample explicitly.

In the last decade it has become possible to perform mechanical tests on samples that are only a few microns in size. The samples are so small, that by utilizing the power of modern graphics cards, it will be possible to simulate the experiment including every dislocation in the material explicitly, and watch how they interact with each other and with multiple precipitates. Being able to simulate an entire experiment at this level of detail is unprecedented and it will provide new insights into the details of what exactly goes on when metal deforms plastically and fractures.

The fundamental new insights gained during the project will be used to develop more accurate engineering design rules for industry and involves close collaboration with scientists and engineers at Lawrence Livermore National Laboratory in California, Imperial College London, Culham Centre for Fusion Energy in Oxfordshire, The National Physical Laboratory in Teddington and Rolls-Royce in Derby.

Potential Impact:
Who will benefit from this research?
The project will provide new insight and enable engineering design rules to be developed which will benefit the immediate project partners (Rolls-Royce, the National Physics Laboratory, Imperial College London, Culham Centre for Fusion Energy and Lawrence Livermore National Laboratory) as well as researchers in nuclear materials in Oxford, in both the host Department of Materials and in the Department of Engineering Science. The modelling capability developed during the project will extend the range of problems that can be simulated with dislocation dynamics to include complex alloys and hydrogen diffusion. By publishing the code and training material online, researchers throughout the world will be able to benefit, particularly universities and research groups which currently have little or no dislocation modelling capability. Experimental researchers in micromechanics will be immediate users of these models. Researchers in materials for nuclear reactors will benefit; for example the model will be able to accurately simulate a range of alloys used in current or future reactors including: Zr alloys, Ni alloys, steels and ODS alloys. In the longer term the model could be used to simulate stress corrosion cracking or in virtual prototyping and alloy development. The model will also provide a link between the nano-scale, electronic structure and molecular dynamics simulations (for example research at the Thomas Young Centre in London) and larger scale crystal plasticity finite element modelling (e.g. Fionn Dunne's group at Imperial College, Anish Roy's at Loughborough University) and industrial simulation of whole components (e.g. at Rolls-Royce).

How will they benefit from this research?
The specific benefits to the UK are more accurate modelling capability of engineering alloys. This will enable more accurate engineering design rules to be extracted as well as constitutive laws for use in larger scale engineering models. The project aims to produce a model able to predict delayed hydride cracking and to develop design rules which Rolls-Royce will be able to use. Today dislocation models are not used in industry as they are not able to simulate alloys accurately. It is anticipated that this project will generate not only good science but also have a real and significant industrial impact. A key area of impact that this project will accomplish is the training of the next generation of skilled researchers in 3D discrete dislocation plasticity. This will be achieved through the training of a postdoctoral research assistant and doctoral students who will become knowledgeable in state of the art 3D coupled discrete dislocation and finite element modelling techniques. Interactions with undergraduates will also be developed through the undertaking of final year (Part II) projects to simulate reactor materials. The UK is lacking in these core skills and as such, this project will provide a means by which the UK can once again become a leader in the generation of new materials technology.