Physicists have manufactured a small can of light and shown it could be used to make optical devices such as tiny sensors and light-based computer chips.

The team from ANU, ITMO University and Korea University, trapped light waves for a record length of time inside a cylinder about one hundredth the diameter of a human hair, made not from tin, but a solid piece of the semiconductor material gallium arsenide.

Trapping light is essential to signal processing, said Mr Kirill Koshelev, from ANU Research School of Physics, who was the first author of the publication reporting the work in Science .

“Because we know the structure, we can engineer the optical properties cleverly to get a resonator which confines light for a very long time. This allows light signals to be processed efficiently, for example to convert frequency or create a laser,” Koshelev said.

The ANU team, from the Nonlinear Physics Centre in the ANU Research School of Physics, was led by Professor Yuri Kivshar and included Ms Elizaveta Melik-Gaykazyan and Dr Sergey Kruk.

Laser light has already revolutionised data transfer speeds in the guise of optic fibres, but is not yet used for signal processing at the nanoscale such as amplification or splitting signals. This is because light-based componentry has proven harder to miniaturise than conventional electronics, which can easily trap and store electrons.

“Earlier attempts to trap light for long periods had only been successful with much larger resonators,” Koshelev said.

However, in 2017 Koshelev and his ANU colleagues, along with Dr Andrey Bogdanov at ITMO University in St Petersburg, came up with a theory for a new mechanism to overcome these limitations, based on an idea first proposed in 1929 by von Neumann and Wigner.

Von Neumann and Wigner had realised that in a structure smaller than the wavelength of the light, a particular configuration of interference between waves could lock a specific frequency within a structure, a phenomenon known as a bound state in a continuum.

In 1929, appropriate materials and the technology to manufacture such small devices was not readily available. However in 2019, Koshelev and his colleagues were able to enlist experts in nanofabrication at Korea University, Dr Jae-Hyuck Choi and group leader Professor Hong-Gyu Park, to make the tiny device. The team selected gallium arsenide, a semiconductor with especially strong non-linear interactions with light, that enable light processing such as frequency shifting.

The first trials of the device, however, were disappointing and sent the team back to the drawing board. Reviewing their theory, the team found no error in the calculations, so they persisted with the same sized cylinder: 700 nanometers high, with a diameter of around 900 nanometers.

“We decided to improve the sample by making fabrication even more precise and also using a three-layer substrate which partially reflects the light,” Koshelev said.

“This helped and we were able to get the record value immediately,” Koshelev said.

Where other experiments with such small structures had only managed to hold the light for five to ten oscillations of the light waves, the new cylinder trapped the light for 200 oscillations.

The team put their can of light’s signal processing capabilities to the test by attempting frequency doubling, in which their infrared signal beam was converted to visible red light. This time success was immediate, showing an efficiency 100 times greater than previous experiments on frequency doubling for structures of a similar size.

To use these tiny structures for everyday-sized devices, Koshelev envisages many of them being arrayed alongside each other, forming what is known as a metasurface – recently listed in the top ten emerging technologies by Scientific American.

“If we combine our particles into an array, the efficiency of frequency doubling will be even more effective by a few orders of magnitude,” Koshelev said.