Behind the Paper

Crushing soda cans and the mathematics of corrugation formation

How do liquid-filled cylindrical shell buckle when compressed?

If you have ever fallen down the internet rabbit hole of oddly satisfying videos, chances are you have watched objects being crushed by a hydraulic press. These clips are strangely captivating: a rubber duck flattening into a pancake, a stack of coins folding like paper, or a metal object buckling in dramatic slow motion. One particularly memorable example comes from the popular YouTube channel Hydraulic Press Channel, where the hosts test the limits of everyday objects under enormous pressure. In one video, a drink can is squeezed until it forms a striking series of neat circular ridges. For most viewers, the spectacle is simply entertaining. For us, it sparked a scientific question. Why does a liquid-filled can crumple into such a tidy sequence of rings, while an empty can simply collapses?

This seemingly simple curiosity eventually led us to uncover a surprising piece of mathematics hidden inside a very familiar object. In our recent paper published in Communications Physics, we show that the ripples forming on a crushed drinks can follow a rare mathematical phenomenon known as homoclinic snaking. While it has long fascinated mathematicians studying pattern formation, observing homoclinic snaking clearly in a real physical structure is quite unusual.

And it all started with a soda can.

A simple experiment with a surprising twist

Most people have probably stepped on an empty aluminium can at some point. The result is dramatic but short-lived: the can suddenly buckles flattening into a pancake. But when we crushed a liquid-filled can in the laboratory, the behaviour was completely different. Instead of immediately collapsing, the can developed a single circular wrinkle on its surface. Then another appeared nearby. Then another. Gradually, the entire cylinder wrapped itself in evenly spaced rings, almost like the ridges of a concertina. Even more striking was how orderly the process looked. Each ring appeared sequentially, one after another, rather than all at once. The pattern seemed almost choreographed. In fact, a particularly creative interpretation of this process comes from the performer Smac McCreanor, whose dance piece inspired by hydraulic-press experiments humorously captures the dramatic tension and release of crushing objects. While our research did not involve choreography, watching the rings appear one by one does feel a little like watching a carefully timed performance.

But behind the visual spectacle lies some fascinating physics.

Important differences

The crucial difference between an empty can and a full one is the liquid inside. Liquids are almost incompressible, meaning that their volume changes very little even under high pressures. When a filled can is compressed, the liquid effectively pushes against the metal walls stopping the can from simply folding inward. Instead, the structure must deform while maintaining the volume of the liquid. This constraint forces the metal to buckle in a much more organised way, with the familiar pattern of corrugations appearing gradually.

When mathematics predicts the wrinkles

To understand this behaviour, we combined laboratory experiments with mathematical modelling. We developed a model similar to those used in studying convective patterns and water ripples. What we found was remarkable. Just like in experiments, the pattern appeared gradually, one wrinkle at the time. If you draw this process on a graph, it wiggles back and forth like a snake, which is why the mathematical phenomenon was dubbed homoclinic snaking.
Each additional wrinkle corresponds to another “turn” of the snake.

Mathematicians have predicted before that this type of behaviour could occur in cylindrical shells, though not necessarily in the form of rings. However, clear experimental evidence has been rare. Seeing it so clearly in a simple drinks can was an exciting moment for us. Sometimes, sophisticated mathematics hides in the most everyday objects. 

Tiny imperfections, predictable patterns

Another crucial piece of physics is how aluminium reacts to being stressed. As the can is compressed, this metal alternates between softening and stiffening, naturally producing evenly spaced corrugations. Fortunately, such metal response is not unique to aluminium, so circular ridges are not just a quirky feature of soda cans. Metal deformation also ensures that the pattern does not disappear once the compression is stopped. 

In theory, a perfect cylindrical shell under compression will always develop the first wrinkle in the middle. In reality, however, no two cans are perfectly identical. Small variations in thickness, tiny dents, or slight differences in shape can influence where the first buckle appears. In our experiments, the first wrinkle often formed near the middle of the can, but there were times when it appeared slightly higher or lower. However, once the first wrinkle formed, the next one would always appear adjacent to it, until the whole surface was patterned. 

Why this matters beyond soda cans

At first glance, crushed cans might seem like a curiosity rather than a serious engineering problem. But structures similar to drink cans appear all over modern technology. Many industrial systems rely on liquid-filled cylindrical metal containers. These include storage tanks, pipelines, transportation vessels, energy infrastructure, and even components of rockets. Understanding how these structures buckle under pressure is important for safety. If engineers know how the buckling process unfolds, they may be able to detect early warning signs of failure. The appearance of the first few wrinkles could indicate that a structure is approaching a critical state long before a catastrophic collapse occurs. This knowledge could help improve monitoring systems and lead to safer designs.

From collapse to manufacturing?

Interestingly, the discovery might also suggest new possibilities for manufacturing. Corrugated metal surfaces are widely used because they are stronger and more rigid than smooth sheets. Typically, such shapes are produced using moulds or specialised forming processes. Our results suggest that controlled compression might be able to create similar corrugations after a container is filled. Imagine producing a perfectly smooth can, filling it with liquid, and then applying carefully controlled compression to generate evenly spaced corrugations. In principle, the natural buckling process could do the shaping for you. While this idea is still speculative, it highlights how understanding fundamental physics can sometimes inspire unexpected technological ideas.

Scientific discoveries do not always begin in a laboratory. Sometimes they start with a curious observation in everyday life—or even while watching videos online. The hydraulic press experiments that inspired this work were initially just entertaining demonstrations of extreme force. But by looking a little more closely, we realised that something deeper was happening. The neat sequence of wrinkles on a crushed can was not just visually satisfying. It was the signature of a subtle mathematical process rarely observed in real systems. This reminds us that fascinating science often hides in plain sight—even in the humble soda can sitting on your desk. The next time you see something crushed by a hydraulic press, you might notice more than just the spectacle. You might see mathematics unfold, one wrinkle at a time.