Scientists use Avizo to capture the complex dynamics of particle movement in granular materials

Parmesh Gajjar, Chris G. Johnson, James Carr, Kevin Chrispeels, J. M. N. T. Gray & Philip J. Withers - Henry Moseley X-ray Imaging Facility, Department of Materials Science, The University of Manchester, Manchester, UK

Scientists have for the first time captured the complex dynamics of particle movement in granular materials, helping to explain why mixed nuts often see the larger Brazil nuts gather at the top. This phenomenon, known as the ‘Brazil-nut effect’, has huge implications for industries where uneven mixing can critically degrade product quality. The findings could have vital impact on industries struggling with the phenomenon, such as pharmaceuticals and mining.

For the first time, scientists at The University of Manchester have used time-resolved 3D imaging through time-lapse X-ray Computed Tomography to show how the Brazil nuts rise upwards through a pile of nuts as it is repeatedly agitated.

Dr Parmesh Gajjar, lead author of the study, said: “In order to understand the movements, it was necessary to identify each of the individual nuts within the 3D data. This was challenging because of the irregular shape of the nuts, but also because of noisy CT data.

“We have used Avizo for many years at the Henry Moseley X-ray Imaging Facility at The University of Manchester, and the ease of use for 3D data was what made it suitable for this work.

“The 3D capabilities of Avizo allowed a very neat pipeline from raw data, correcting for noise, identifying each nut and analyzing size, shape and orientational statistics for each one. In addition, the TCL programming capabilities made it very simple to extend the pipeline to seamlessly process all 181 datasets.

“Our study highlights the important role of particle shape and orientation in segregation. Further, this ability to track the motion in 3D will pave the way for new experimental studies of segregating mixtures and will open the door to even more realistic simulations and powerful predictive models. This will allow us to better design industrial equipment to minimize size segregation thus leading to more uniform mixtures. This is critical to many industries, for instance ensuring an even distribution of active ingredients in medicinal tablets, but also in food processing, mining and construction.”

Read the news in The University of Manchester Newsroom.

How Amira-Avizo Software is used

Avizo pre-processing operations are used to obtain the best working material to guarantee that the segmentation workflow remains stable through the variable quality of all the datasets. Shading Correction was crucial in this case for grayscale intensities uniformization, as well as NLM for denoising.

The main idea of the workflow is then to perform an object separation using the Watershed Segmentation technique but on the filled nuts rather than on the original nuts (with large empty space in their center). Filling the nuts prior to segmentation helps to avoid undesired separations by keeping only one local extrema for each nut.

The empty space of the nuts are restored afterwards by applying an extrusion of the empty space on the final segmentation result.

Dr Parmesh Gajjar, lead author of the study, said: “A Thermo Fisher Scientific application specialist helped me derive the image processing and segmentation pipeline. He was great to work with, the pipeline he derived was a positive surprise and he was always ready to help answer questions through e-meetings or emails.”