Beat the diffraction limit with geometric potentials

Berry's phase characterizes the geometric properties of the parameter space of a Hamiltonian, where after adiabatically following a closed path in the parameter space, the wavefunction of a system acquires a net phase. In this talk, I will explain how to use this geometric phase to overcome the diffraction limit in the optical manipulation of cold atoms. I will report our creation of a conservative optical potential with a feature size that is below λ/50. This is achieved by engineering the geometric phase associated with the dark state of a three-level atom in the EIT configuration. The ability to manipulate atoms on a subwavelength scale opens many exciting opportunities, including tunnel junctions for atomtronics applications, nearly perfect box-like atom traps, and synthesizing (arbitrary) subwavelength optical potentials.