A single-atom-thick layer of tungsten disulfide emits light when excited — that's the basis of atomically thin optoelectronics. The problem is how little light it emits. One atom thick means one atom thick: there's almost no material to interact with incoming photons. The emission is real but weak.
The conventional fix is to put the material on a resonant structure — a photonic crystal, a plasmonic antenna, a microcavity — to concentrate light where the material sits. These structures are made of solid material with high refractive index, and they work by trapping light inside themselves. The 2D material sits on top, benefiting from the evanescent field that leaks out.
The new approach inverts this. Instead of building up, carve down. Nanoscale air cavities — Mie voids — are etched into a high-index crystal of bismuth telluride. The voids are empty. The strong reflection at the air-crystal boundary forces light to circulate inside the void, concentrating the optical field in the air region and at the top surface, exactly where the tungsten disulfide layer sits.
The result: twenty times stronger emission and twenty-five times stronger nonlinear optical signals compared to the least resonant configuration. The amplification comes from nothing — from the absence of material rather than its presence. The empty space is the optical element.
The counterintuitive design principle is that the best way to enhance a surface material is not to surround it with more material but to remove material beneath it. The void creates the resonance. The air is the amplifier. What's missing does the work that what's present cannot.