Why the big nuts end up on top, and why physicists still argue about it
The idea
Open a can of mixed nuts that has travelled in a lorry and the large Brazil nuts are usually at the top. This is not a quirk of packing. When a mixture of grains of different sizes is shaken, the grains tend to sort themselves by size, and the big ones often rise. Physicists call it the Brazil nut effect, a name popularised by a 1987 paper in Physical Review Letters titled “Why the Brazil nuts are on top”.
That paper was a computer study. The authors used a Monte Carlo simulationMonte Carlo simulation A computer method that studies a system by running many repeated random trials and looking at what usually happens., a method that moves simulated particles around by repeated random trials, and found that a single large ball in a shaken container rises to the top even when it is denser than the small particles around it. Their explanation was local and geometric. Each shake lifts the large particle slightly, small grains slip into the gap that opens beneath it, and the large one cannot fall back. Repeat this many times and it ratchets upward. The paper separated this from sifting, where small grains simply fall through spaces between large ones.
The interest for a non-specialist is that a granular materialgranular material A collection of many separate solid grains, such as sand, rice or powder, that can flow but is neither a true liquid nor a true solid., such as sand, cereal or a pharmaceutical powder, behaves like neither a solid nor a liquid. A heavy stone sinks in water. A heavy nut can rise through rice.
Competing explanations
In 1993 James Knight, Heinrich Jaeger and Sidney Nagel at the University of Chicago shook columns of grains in the laboratory and reached a different conclusion. They argued that the separation in their setup came from convectionconvection A slow circulating flow in which material rises in one place and sinks in another, carrying things along with it., a slow circulation of the whole bed of grains. In their account, the circulating flow carries a large particle upward, and once it reaches the top the flow cannot take it back down, so it stays trapped there. In their words, separation “arises from convective processes rather than from local rearrangements”.
Then the air got involved. In a 2001 Nature paper, Matthias Möbius, Benjamin Lauderdale, Nagel and Jaeger reported that how fast a large intruder moves depends on its density in a non-monotonic way, and that this dependence is sensitive to the background air pressure. The gas trapped between grains, called interstitial gasinterstitial gas The air or other gas filling the small spaces between grains in a pile or container of particles., matters. A Jaeger lab summary of the work notes that with the container evacuated, the intruder simply moves with the surrounding particles. A follow-up from the same group, posted in 2004 and using magnetic resonanceresonance A large build-up of vibration that happens when something is pushed repeatedly at the same rhythm it naturally swings at. imaging and high-speed video, found that light intruders can either rise or sink depending on where they start, the container walls and the gas pressure, and modelled this with pressure differences across the bed during each shake.
So three mechanisms have strong support from serious work: geometric ratcheting, convection, and air pressure. They are not mutually exclusive, and the evidence suggests each can dominate under different conditions.
Why it might change a decision
The effect predicts something you can test in a kitchen. Fill a clear jar with uncooked rice and bury a table-tennis ball or a walnut near the bottom, then shake it up and down for a minute. The large object should work its way up. The more interesting prediction comes from the Chicago work: the outcome should depend on details people usually ignore, such as how the jar is shaken, how rough its walls are and whether it is sealed.
The same logic applies where mixing matters commercially. A blend of particles of different sizes that is well mixed at the factory can separate on the way to the user simply by being vibrated. Whether a manufacturer should fix this by changing particle sizes, container shape or packing pressure depends on which mechanism is at work in that particular setup.
How solid is this?
The basic observation is well established: large particles moving relative to small ones under shaking appears in simulation and in independent laboratory experiments using different methods. The sources here are physics studies of controlled systems, not surveys of real food packaging, so the jump from laboratory columns to a can of nuts is plausible but not measured in these papers. The verified abstracts used for this post do not report effect sizes such as typical rise times, so none are given here.
Where the evidence is thin is in the popular version of the story. Explanations online often present the geometric ratchet as the answer. The work since 1993 suggests that is at best one answer: convection was found to drive separation in one experimental setup, and air pressure changed the density dependence in another. The strongest competing view to “one mechanism” is that granular segregation is a family of effects that share a symptom.
Sources
- Rosato, A., Prinz, F. and colleagues (1987). “Why the Brazil nuts are on top: Size segregation of particulate matter by shaking”. Physical Review Letters 58(10), 1038-1040. https://doi.org/10.1103/PhysRevLett.58.1038
- Knight, J. B., Jaeger, H. M. and Nagel, S. R. (1993). “Vibration-induced size separation in granular media: The convection connection”. Physical Review Letters 70(24), 3728-3731. https://doi.org/10.1103/PhysRevLett.70.3728
- Möbius, M. E., Lauderdale, B. E., Nagel, S. R. and Jaeger, H. M. (2001). “Brazil-nut effect: Size separation of granular particles”. Nature 414, 270. https://doi.org/10.1038/35104697
- Möbius, M. E., Cheng, X., Karczmar, G. S., Nagel, S. R. and Jaeger, H. M. (2004). “Intruders in the Dust: Air-Driven Granular Size Separation”. arXiv preprintpreprint A research paper shared publicly before, or without, formal peer review by a journal. cond-mat/0405328. https://arxiv.org/abs/cond-mat/0405328