Sympathetic cooling of a membrane oscillator in a hybrid mechanical-atomic system

The quantum behaviour of macroscopic mechanical oscillators is currently being investigated using a variety of mechanical systems and techniques such as optomechanical cooling and cold damping. As mechanical systems are also very versatile transducers between different physical systems, it is possible to build hybrid systems that combine the advantages of their constituents. This opens up new possibilities for fundamental studies of quantum physics, precision sensing and quantum information processing. Ultra-cold atoms represent one of the best-controlled systems available, thus making a well-developed toolbox for quantum manipulation available to mechanical oscillators in a hybrid system.
In this thesis, I report on the realization of a hybrid mechanical-atomic system consisting of a Si3N4 membrane inside an optical cavity coupled to an ensemble of atoms. The coupling is mediated by a light field that couples the atomic motion to the membrane motion over a large distance. By laser cooling the atomic motion, the membrane is sympathetically cooled via its interaction with the atoms to a temperature of 0.7 K starting from room temperature, despite the enormous mass ratio of 10^10 between the membrane and the atomic ensemble. Up to now, sympathetic cooling had only been used to cool microscopic particles with much lower masses. The system reported in this thesis is the first hybrid system where the back-action of the atoms onto the mechanical oscillator is sufficiently large for practical applications. It represents a significant improvement over a previous experiment in our laboratory, where the atom’s influence onto the mechanical oscillator was barely detectable. An atom-membrane cooperativity C > 1 is achieved, thus enabling the study of effects such as a mechanical analog of electromagnetically induced transparency in the system, which will be investigated in the future. The quantitative analysis of the coupling mechanism also allows to predict experimental requirements for future ground state cooling of the mechanical oscillator, which are within reach. Interestingly, hybrid systems such as ours can provide ground-state cooling of low-frequency mechanical oscillators in a regime, where neither cavity optomechanical cooling nor cold damping can reach the ground state.