School Notes: School of Engineering & Applied Science
May/June 2026

A breakthrough via fridge magnets

With a tool known as ARPES, scientists can map out a material’s electron structure. Like an electronic DNA, this structure determines many properties. Under a magnetic field, however, ARPES technology is ineffective because the electrons’ natural path is disrupted. That’s a major drawback, since many materials’ most valuable properties only happen under a magnetic field. Inspired by refrigerator magnets, Professor Yu He and his team replaced a large single magnet with many tiny magnets of alternating polarities, allowing the electron to experience the magnetic field for nanoseconds, and then resume its natural trajectory. The study is featured in the Journal of Physical Chemistry Letters.

The dramatic effect of a pinch of salt

It is commonly assumed that tiny particles go with the flow as they make their way through soil, biological tissue, and other complex materials. But the lab of Professor Amir Pahlavan put this to the test by creating its own porous media out of transparent polymers. 

On a chip, the researchers created a mazelike environment and followed fluorescent particles as they wended through the pathways. Despite the relatively strong force of the flow, a pinch of salt significantly affected the particles’ movement. It’s an insight with potential applications in wastewater treatment facilities, chromatography, agriculture, drug delivery, and other fields. Read the full results in Physical Review Letters.

The shared logic of natural and artificial networks

From our brains and living cells to AI platforms, natural and engineered information-processing networks share a remarkably similar strategy for figuring things out. That’s the finding of Professor Andre Levchenko’s lab, published in Cell Systems. This strategy, Levchenko says, features hierarchical components that process information at different rates and to varying depths. Our brains, for example, respond quickly to certain types of information but delve more deeply and in more detail when warranted. Artificial neural networks also react quickly to input, but similarly demonstrate the capacity for deeper integration. This clearer understanding of how AI networks organize information could illuminate how our biology works and vice versa.