Meta surfaces steer RF, light or sound arbitrarily

“Existing solutions for controlling reflection of waves have low efficiency or difficult implementation,” said Aalto researcher Dr Ana Díaz-Rubio. “We solved both of those problems. Not only did we figure out a way to design high efficient meta-surfaces, we can also adapt the design for different functionalities. These meta-surfaces are a versatile platform for arbitrary control of reflection.”

The surfaces have been dubbed meta-mirrors, and the key to creating them is applying new physics that describe reflection by sub-wavelength surface features.

“It was only recently that the physics of this wave transformation by meta-surfaces was properly understood. It was shown theoretically that the non-local properties, required for high-efficiency reflections into arbitrary directions, can be, in principle, realised by carefully engineering the surface reactance profile,” according to the paper  ‘Power flow-conformal metamirrors for engineering wave reflections‘ (available free in full) which describes the work in Science Advances.

According to the team, there is no parasitic scattering, nor any need for evanescent fields close to the meta-surface – fields in front of the meta-mirror are combinations of the desired propagating plane waves away from the surface and in its close vicinity.

Given a plane waves incident on the surface, almost all the energy can directed away at a chosen angle, or at angles, determined by fixed small structures on the surface, shaped depending on frequency, the intended entry angle and required exit angles.

The same maths and theory (‘power-flow conformal meta-surfaces’) applies, whether the waves are sound, light or radio waves.

The paper describes two meta-mirrors, both designed to reflect light or sound (when made with a suitable material) arriving at right angles to the mirror surface.

While a standard mirror would reflect it straight back, one of the example mirrors reflects energy away at 70° off to one side, and the other reflects some at 70° to one side, and the rest at 70° to the other side – the latter creating a beam-splitter.


In a proof-of-concept, the 70°-to-one-side situation was designed for 3kHz sound wave use, then both modelled and built (see photo, detailed geometry is described in the paper).

The 3D printed repeated unit in for the 3kHz mirror is 120mm long and 63mm deep (see photo) and consists of a collection of variable depth 8mm rectangular ‘tubes’ – the 8mm width of each tube is smaller than 0.1λ at 3kHz. The physical experiment used 12 repeated units. Experimental results agreed well with the models.

“This is really an exciting result,” said Díaz-Rubio. “We have figured out a way to design such a device and we test it for controlling sound waves. Moreover, this idea can be applied to electromagnetic fields.”