Three-dimensional images of the bubbles in a liquid metal foam produced by the new super-fast version of computed tomography. Image: Adv. Mat./PSI/HZB.
Three-dimensional images of the bubbles in a liquid metal foam produced by the new super-fast version of computed tomography. Image: Adv. Mat./PSI/HZB.

Most people are familiar with computed tomography for medicine: a part of the body is X-rayed from all sides to produce a three-dimensional (3D) image, from which sectional images can be created for diagnosis.

This method can also be very useful for material analysis, non-destructive quality testing and the development of new functional materials. But examining such materials at high spatial resolution and in the shortest possible time requires the particularly intense X-ray light of a synchrotron radiation source. Using a synchrotron beam, even rapid changes and processes in material samples can be imaged, making it possible to acquire three-dimensional images in a very short time sequence.

A team of researchers from Helmholtz-Zentrum Berlin (HZB) in Germany, led by Francisco Garcia Moreno, has been working on this kind of fast computed tomography of materials, in conjunction with colleagues from the Swiss Light Source (SLS) at the Paul Scherrer Institute (PSI) in Switzerland. Two years ago, they achieved a record-breaking 200 tomograms per second by developing a new method called fast-imaging tomography.

Now, the researchers have achieved a new world record – 1000 tomograms per second – allowing them to record even faster processes in materials and during manufacturing. As they report in a paper in Advanced Materials, this is achieved without any major compromises in other parameters: the spatial resolution is still very good, at several micrometres; the field of view is several square millimetres; and continuous recording periods for up to several minutes are possible.

To achieve this world record, the researchers placed the sample on a high-speed rotary table developed in-house, whose angular speed can be perfectly synchronized with the camera's acquisition speed. "We used particularly lightweight components for this rotary table so that it can reach 500 Hertz rotation speed stably," says Moreno.

At the TOMCAT beamline at the SLS, which specializes in time-resolved X-ray imaging, PSI physicist Christian Schlepütz used a new high-speed camera and special optics. "This increases the sensitivity very significantly, so that we can take 40 2D projections in one millisecond, from which we create a tomogram," Schlepütz explains. With the planned SLS2.0 upgrade, even faster measurements with higher spatial resolution should be possible from 2025.

The acquisition of 1000 3D data sets per second – over a period of minutes – generates a huge data stream, which was initially stored at the PSI. Paul Kamm at HZB was responsible for the further processing and quantitative evaluation of these data. The reconstruction of the raw data into 3D images was carried out on the high-performance computers at PSI, and the results were then transferred to HZB for further analysis.

The researchers demonstrated the power of this super-fast version of computed tomography on various material samples. They recorded images of the extremely rapid changes during the burning of a sparkler, the formation of dendrites during the solidification of casting alloys, and the growth and coalescence of bubbles in a liquid metal foam. Such metal foams based on aluminium alloys are being investigated as lightweight materials, for use in constructing electric cars, for example. The morphology, size and cross-linking of the bubbles are important to achieve the desired mechanical properties such as strength and stiffness in large components.

"This method opens a door for the non-destructive study of fast processes in materials, which is what many research groups and also industry have been waiting for," says Moreno.

This story is adapted from material from Helmholtz-Zentrum Berlin, with editorial changes made by Materials Today. The views expressed in this article do not necessarily represent those of Elsevier. Link to original source.