Cryogenic thermal noise behavior in release-free optomechanical crystals

Doktorsavhandling, 2026

Nothing with a temperature is ever completely still. The concept that we call temperature originates from the erratic thermal motion of individual atoms, which serve as the building blocks of physical matter. To living beings such as humans, an elevated temperature is a sign of life; but to a sensitive quantum system, the opposite is generally true. Schrödinger's cat in a box is the typical example of a quantum system, and can be thought of as simultaneously dead and alive as long as no information about the cat's well-being can escape to the outside world. The natural way to probe the cat's state is to open the box and take a look, but even the thermal motion of the box's atoms can transmit similar information by bouncing off the cat's paws. In theory, one could therefore infer the cat's state by studying the box itself, without ever having to open the lid.

The effect of atoms' random motion on quantum systems is generally referred to as thermal noise; and for the optomechanical crystal -- the main object of study in this thesis -- reducing this noise is of primary concern. The optomechanical crystal is a nanofabricated device much smaller than the width of a human hair. Made from silicon, it allows laser light to drive high-frequency acoustic vibrations -- and vice versa. Applications of this effect are many, but to enable the most sensitive ones, where quantum states are transferred between light and acoustic vibrations, thermal noise must be kept to a minimum. To achieve this, devices are generally cooled to cryogenic temperatures only ten-thousandths of a degree above absolute zero. But even so, the devices absorb some of the laser light used to operate them, leading to detrimental local heating.

In this thesis, the challenge of laser-induced noise in optomechanical crystals is tackled through the invention of a new class of device. Conventional fabrication typically calls for released -- or suspended -- silicon nanobeams to accommodate pure acoustic vibrations. In contrast, our design is release-free, making devices thermally robust through proper anchoring to the chip surface -- without detrimentally sacrificing acoustic purity. The thesis contributes to our understanding of thermal noise in these devices by studying and comparing its behavior in both release-free and suspended designs. Finally, we begin evaluation of the release-free device's potential as part of a flagship application for optomechanical crystals: a full microwave-to-optical transducer.