A granular material, such as sand, coffee beans, or balls in ball pit, is a collection of particles that interact with each other and dissipate energy. These materials can act like solids, flow like liquids, or suddenly transition between the two phases – for example, in a landslide, the soil stops holding its shape and flows. The Granular and Particulate Networks Workshop, PARNET19, brought together the physicists, engineers, and mathematicians who study these materials in a series of lectures and discussions.
PARNET19 took place at the Max Planck Institute in Dresden, Germany on July 8-10, 2019. I attended to represent Softbites at the science communication panel and to present my research on fly larvae as an active granular material.
The focus of the workshop was exploring the networks formed by the forces in granular materials. When granular materials are stretched or squeezed, they form networks of high forces known as force chains. These networks can be visualized with photoelastic disks, as described by this previous Softbites post. In a series of 30 minute to 1 hour long scientific talks at PARNET19, the experimentalists who study granular materials and mathematicians who study topological networks discussed how network math can be applied to the force chains found in granular materials. Unusually, talks were followed by 30-minute discussion sessions in which the previous speakers answered questions and posed some of their own.
Modeling granular materials is difficult because they are made up of many individual particles. Simulating the interactions of all of the particles takes a very long time, even with a powerful computer: imagine trying to predict the motion of each sand grain on a beach! The other traditional way to model a granular material is with a continuum model — considering the material as a smooth (continuous) mass, instead of keeping track of individual particles. This works for materials like fluids or solids because the individual molecules that make them up are so small that their individual interactions don’t need to be understood. However, the relevant particles in a granular material are much bigger, relative to the size of the material as a whole, than molecules, which makes the interactions between the particles important. In a granular material, the critical interactions between the particles can result in sudden transitions such as landslides.
The approach taken by the PARNET workshop was to model the part of granular materials that will cause the entire material to change if it moves — the force chains through the grains. The goal of the workshop was to apply existing mathematical theories used to model networks, such as the connection of roads on a map, to understanding the connections of force chains in granular materials. For example, understanding when a force chain in the rocks making up a cliff is likely to fail can inform workers near the cliff about impending danger and allow them to evacuate before a landslide occurs.
The scientific communication panel I was part of discussed a variety of topics, such as publishing journal articles in high or low impact factor journals, making scientific journals open access, and writing for a broad audience. A result of the discussion, we made the Softbites style guide publicly available – everyone wanted to read how we write and edit our posts!
For me, the main takeaway of the workshop was that the network view of granular materials is a promising one to predict catastrophic events. Understanding what causes a force chain to break can explain why some arrangements of granular materials are stable for a long time while others come crashing down with no obvious warning. However, connecting the complex and chaotic real-life granular materials in 3D to the purely theoretical math behind topological networks will prove challenging. Mathematical models of networks can be very abstract, and these theories need to be connected to physics in the real world. As with any theory, it is important to verify predictions with real-life experiments, but the force chains inside granular materials are difficult to measure.
Overall, this was one of the best conferences I’ve attended as a graduate student. The format of longer discussion sessions was very effective, as it allowed more time for elaborating on each speaker’s points than the traditional 5 minute long Q&A sessions. The PARNET workshop was a useful introduction to a new (to me) way of thinking about granular materials, one which I am implementing in my own research. If complex systems, such as granular materials, can be modeled by a simple set of topological equations, they will be much easier to understand and predict in future studies.