Grain Days 2024

Granular materials exhibit important structural features at many length scales. At the smallest scales, we can observe the frictional contact of asperities between individual grains. Zooming out we can describe the packing and arrangement of assemblies of particles. At larger scales still we can observe the runout of landslides and the evolution of sand dunes along a beach. The imaging of these systems is characterised by two important phenomena; opacity and heterogeneity. Almost all granular media are optically opaque. Even when the individual grains themselves are transparent, assemblies of particles rapidly become opaque due to refraction of light as it passes between the grains and the interstitial spaces. For this reason, we typically only ever see the outer layers of a granular medium. It is the heterogeneity of the material, and the interstitial spaces, that creates this opacity. The imaging of granular materials, therefore, is an uphill battle at almost all length scales.

In this Lecture we will cover the basics of imaging; from lighting to acquisition and finally data analysis. You will learn how to process both images and videos. We will see examples of reflection and transmission imaging, and put into a basic context more advanced techniques, such as tomography. Finally, we will go into detail on two techniques for measuring displacements of granular media from images. We will introduce both Particle Tracking Velocimetry and Particle Image Velocimetry (also referred to as Digital Image Correlation depending on your background) and provide a brief introduction to their use cases, pitfalls, and freely available software.

In granular materials, force transmission, fluid flow and heat transfer through them are dominated by their microstructure. The development of accurate models to predict these transport processes has been hindered due to the limited ability to capture the microstructural information in traditional physical testing. X-ray computed tomography (XCT) is a powerful emerging technique that can access the microstructure of granular material non-destructively. During an XCT scanning, X-ray lights penetrate a rotating sample and reach a detector that records the attenuation of X-ray energy in a series of 2D X-ray plates. These data are then inversed to generate three-dimensional cross-sectional images of the sample. By applying image processing methods on the (2D and 3D) XCT images, a digital sample can be reconstructed for microstructure characterisation and transport phenomena simulations.

The Lecture will cover the image processing methods to reconstruct a digital sample based on the already inverted 2D XCT image stacks, with the goals to distinguish the different phases and identify individual particles from the sample to calculate microstructural characteristics such as particle size and shape. You will learn the basics of image filters and data analysis on XCT images. The XCT images of Ottawa sand 20-30 will be provided, and we will show the image processing procedures on them using open-source software: ImageJ (Fiji). More advanced examples generated using an XCT image approach under an Artificial Intelligence (AI) framework will be introduced as well.

Co-authors: Negin Amini, Josh Tuohey, John M. Long, Jun Zhang, David A.V. Morton (Deakin University), and Karen E. Daniels, Farnaz Fazelpour (NC State University).

Abstract: Stress visualization in 3D particles can provide insight and understanding of the behaviour of complex particles. Traditional photoelastic methods produce a polarised light signal in response to applied stresses, but can only be used for simple particle shapes such as discs. 3D-printing has created new possibilities for enhancing the scope of stress analysis of physically representative particles with irregular shapes and structures. Using X-ray computed tomography and 3D-printing combined with traditional photoelastic analysis, we are able to visualize strain for particles ranging from simple 2D discs to complex 3D-printed coffee beans, including internal voids. The relative orientation of the print layers and the loading force can affect the optical response of the discs without affecting the mechanical properties. Potential limitations and areas of future interest for stress visualization of 3-dimensional particles are also outlined.

Amini, N., J. Tuohey, J. M. Long, J. Zhang, D. A. V. Morton, K. E. Daniels, F. Fazelpour and K. P. Hapgood (2022). Photoelastic stress response of complex 3D-printed particle shapes. Powder Technology 409: 117852. https://doi.org/10.1016/j.powtec.2022.117852

Bio: Professor Karen Hapgood is an experienced and respected executive leader, who leads the development and implementation of Swinburne's Research strategy and builds impactful reseach partnerships to further Swinburne’s research nationally and globally. Her career has included over 7 years of industry experience in Australian and US, followed by 15 years of leadership roles in the higher education sector, all based around a common thread of manufacturing and STEM. Prior to joining Swinburne, Karen was the Executive Dean of the Faculty of Science Engineering and Built Environment at Deakin University and before that the Dean of Engineering. Karen is a renowned researcher in the fields of chemical engineering and pharmaceutical manufacturing, a Fellow of ATSE, IChemE, RACI and Engineers Australia, and a Graduate of the Australian Institute of Company Directors. She is passionate about diversity and inclusion in research, engineering and STEM.