Materials witness

Project aims to combine scanning technology and computer modelling to predict how, and where, cracks will develop in critical aircraft parts. Stuart Nathan reports.


Research into modelling the structure of metals could lead to a method for predicting when naval aircraft are likely to suffer from metal fatigue.

Co-ordinated by researchers at Northrop Grumman, the project aims to develop models which will be able to predict how, and where, cracks will develop in critical aircraft parts.

Carrier-borne aircraft have a tough life — landing is closer to a controlled crash than on a tarmac runway — and metal fatigue is a major problem. But according to Anthony Rollett, a materials scientist from CarnegieMellonUniversity in Pittsburgh, nearly all the previous research on metal fatigue has focused on how cracks propagate through metal components. Hardly any have looked at how they start, he said.

Rollet’s research combines examination of metal components with computer modelling, and is based on work carried out by Pittsburgh metals specialist Alcoa on forming 3D digital models of crystalline structures.

‘Companies like Alcoa have invested huge amounts of effort to try to improve the performance of their materials, especially with respect to fatigue. They tend to do it by trial and error, through making many measurements. but making the detailed connection between what you can measure inside the material and how it behaves and performs is far from simple,’ he said.

‘What you would like to be able to is make a very detailed model of the structure of your material, with all the crystalline arrangements and grain boundaries, run a mechanics model, and see how quickly you could generate a crack under certain conditions and how much it grows. That’s what we’re working towards.’

The core of Rollett’s research is measurement — taking aircraft parts and samples of aluminium and subjecting them to a technique called electron back-scattered diffraction (EBSD). This is similar to CAT scanning, he said, but using scanning electron microscopes and electron-beam diffraction to look at ‘slices’ through the material in three orthogonal directions.

‘EBSD maps let us see not the shape of the grains in the metal, and allows us to measure the orientation of the grains and the types of grain boundaries.’

The next step is to turn this into a computer model, using a mathematical technique called a Voronoi tessellation to fill a box with the shapes of the grains. ‘Once you’ve filled the box with this geometry, you have a representation of the grains themselves. Then, you can assign orientations to them until you match the measured texture and grain boundary character of your sample,’ he said.

Any particulate impurities in the aluminium are then added in; this is vital, said Rollett, as fatigue cracks tend to start at some — but not all — of these particles. ‘We’ve got some idea how this happens,’ he said. ‘We believe there are certain crystal orientations that are bad — more likely to give rise to cracks than others.’

Once Rollett and his colleagues have completed the computer model, it is passed on to collaborating researchers at CornellUniversity and Rensellaer Polytechnic, who apply finite element analysis techniques to determine how the structures react to loading. ‘They look for stress concentrations and find the places that cracks might start,’ said Rollett.

The goal of the project is to come up with a model that can predict, from the type of aluminium used in a particular aircraft and its service history, the point at which it is likely to develop fatigue.

‘If you know about the variations of the material and the aircraft’s service —did it fly in bad weather or did it have gentle sea breezes? — you should be able to do a much better job of predicting the end of its useful life,’ said Rollett.