New image-based computational method can show the performance of an aircraft wing

Carbon fibre composites are a type of carbon fibre reinforced plastic that is widely used in the aerospace and wind power industries, among others, because it is a very light and strong material. By determining how the carbon fibres are oriented and distributed in a composite, it is possible to calculate the mechanical properties of the material, such as its strength and stiffness, which is important for the design and quality testing of load-bearing structures.

To determine the orientation and distribution of the fibres in a carbon fibre composite, you can use computed tomography, an X-ray technique that produces 3D images. As the fibres are very thin, about one twentieth of a human hair, very high resolution is required to see them. The high resolution leads to a large amount of data and the volume that can be analysed is very small, typically a few cubic millimetres.

New method allows analysis of real components

Researchers at Chalmers, ETH, DTU and the Paul Scherrer Institute have now developed a new image-based computational method that makes it possible to analyse the stiffness of carbon fibre composites even at low resolution, i.e. when the individual fibres cannot actually be seen. As a result, material volumes of a few cubic decimetres can be analysed, a million times larger than previously possible.

“Because the high resolution limits the size of the volume that can be scanned, it has not been possible to analyse real components. Instead, you have had to cut out very small sample pieces,” says Robert Auenhammer, Doctor in Material and Computational Mechanics at the Department of Industrial and Materials Science.

“Apart from the fact that you have to cut up what you want to analyse, the disadvantage is that you cannot be sure that the small sample is representative of the whole large structure.”

With the new computational method, however, you are no longer limited to small samples, but can determine the stiffness of real components, which could have major benefits for the aerospace and wind power industries, among others.

“The ability to determine the properties of real components with high precision can help industry improve their products and reduce both material consumption and costs. For example, you can scan whole structural details of an aircraft wing and find out if there are any hidden defects from the manufacturing process,” says Robert Auenhammer.

Determining the stiffness of composites is important for designing lighter load-bearing structures. Carbon fibre composites do not have the same stiffness in all directions – it is higher along the fibres and lower across them. This makes the material lighter than if the stiffness had been high in all directions.

“One advantage of carbon fibre composites is that you can choose in which direction the stiffness is needed most. This allows you to make a lighter structure than with materials that are equally stiff in all directions, such as steel. For example, wind turbine blades and aircraft wings require the highest stiffness along the length of the blade or wing, and are therefore designed with the fibres oriented in that direction,” says Robert Auenhammer.

Potential to calculate more properties

The research behind the new computational method also opens the door to further advances in tomography-based materials research.

“Our research shows the enormous potential of tomography-based computational methods. Our contribution is a first step in this field, but more research is needed to take it further. Similar to the way we have developed our method, methods could be developed to calculate other properties of composites, such as strength,” says Robert Auenhammer.

The collaboration between Chalmers, ETH, DTU and the Paul Scherrer Institute was established through the EU-funded research project Mummering, which aimed to develop a research tool for managing and analysing 3D image data. The project was organised by DTU together with leading European researchers in the field of computed tomography.

“The Mummering project was very successful and particularly beneficial for us at the Department of Material and Computational Mechanics, as we have a lot of expertise in material modelling, which is needed to take advantage of advances in tomography technology,” says Leif Asp, Professor of Lightweight Composite Materials and Structures at the Department of Industrial and Materials Science.

More about computed tomography

The type of computed tomography used in materials research is based on the same technology used in hospital CT scans to detect tumours in patients, for example. The X-ray object is imaged in layers by taking 2D images from different directions and constructing a 3D image.

However, materials research typically uses higher resolution and higher radiation doses than hospital X-rays, and the X-ray process can take up to several hours or days. Another difference between the techniques is that in a CT scan, the radiation equipment rotates around the patient, whereas in materials science, the sample itself rotates instead of the equipment.

The imaging technique on which the new computational method is based is called X-ray scattering tensor tomography and was developed at the Paul Scherrer Institute. The imaging method uses special optical elements and adapted mathematical computational methods to obtain information about the fibre orientation of composite materials at low resolution.

In addition to smaller machines that can fit into a laboratory, large scale facilities, known as synchrotrons, are used in materials research. The MAX IV laboratory in Lund in Sweden and the Swiss Light Source at the Paul Scherrer Institute in Switzerland are examples of such facilities. A synchrotron consists of a large circular tube in which charged particles are accelerated to speeds close to the speed of light. As the particles accelerate, they emit radiation, which is then used by dedicated instruments (beamlines) to create detailed images of materials.