Image courtesy of Mt Kalmont.
Avalanches do not solely take the form of snow falling down a mountain. Earthquakes, landslides, power grids, and even crashing stock markets are all avalanches in their own ways. Physicists have often used the 1D random-field Ising model (1D-RFIM) to approximate avalanche responses, but no experimental evidence has ever shown that the model verifies classical theoretical model predictions. A team of researchers from the Schiffer Lab at Yale and collaborators have provided the first-ever experimental realization of the 1D model.
Previously, the Schiffer Lab developed artificial spin ice (ASI), a collection of nanomagnets where each one behaves like a single magnetic spin, possessing one north pole and one south pole. These magnets can be arranged into any desired geometrical setup to give rise to various behaviors, including that of avalanches. But what exactly is avalanche behavior?
Consider a row of nanomagnets with their north poles pointing down in a magnetic system. If a magnetic field is applied, then a single magnetic pole may flip, triggering neighboring magnetic poles to flip continuously down the chain due to the repulsive forces when like poles interact. This cascading event is also characteristic of avalanches.
For the experiment, the researchers used a square arrangement of the ASI—four islands (a single nanomagnet) pointing towards a vertex. This is the simplest possible case, one with very well-defined initial conditions. Different magnitudes of the magnetic field were applied from this starting point, causing some islands to flip their poles. Using a magnetic force microscope, the researchers imaged the ASI sample and counted the number of islands that changed course.
“This process takes a long time. On each [nanomagnetic lattice], we fabricated three different array sizes and numbers of islands, and four copies of each sample were made for every magnetic field to get better statistics. We’re talking about hundreds of magnetic force microscope images, which took about six months. A single imaging takes 40 minutes, and not every image turned out usable,” Nicholas Bingham, Associate Research Scientist of the Schiffer Lab, said.
The results indicated that avalanche size caused by the different island sizes and magnetic field strengths aligned well with the 1D model approximations. The beauty of this experimental realization is that the methods used in this study can be used to measure avalanche behavior in various other systems. For snowy avalanches, the framework could be instrumental for hazard prevention.
“Now that we showed the model works for the system, we can also work backward: design materials that behave in any way we want them to behave,” Bingham said. Using the model, physicists know which parameters give rise to particular types of behavior, allowing for the specific design of materials with pre-determined behaviors. The Schiffer Lab now aims to understand avalanche behavior using more complicated materials.