Ultracold bosons in optical lattices for quantum measurement and simulation

Abstract

Ultracold atoms provide a platform that allows for pristine control of a physical system, and have found uses in both the fields of quantum measurement and quantum simulation. Optical lattices, created by the AC Stark shift of a coherent laser beam, are a versatile tool to control ultracold atoms and implement novel Hamiltonians. In this thesis, I report on three experiments using the bosonic species Rubidium-87 trapped in optical lattices. I first discuss our work in simulating the Harper-Hofstadter Hamiltonian, which describes charged particles in high magnetic fields, and has connections to topological physics. To simulate the charged particles, we use laser-assisted tunneling to add a complex phase to tunneling in the optical lattice. For the first time, we have condensed bosons into the ground state of the Harper-Hofstadter Hamiltonian.

In addition, we have demonstrated that we can add strong on-site interactions to the effective Hamiltonian, opening the door to studies of interesting states near the Mott insulator transition. Next, I present a novel technique to preserve phase coherence between separated quantum systems, called superfluid shielding. Phase coherence is important for both quantum measurement and simulation, and is fundamentally limited by projection noise. When an interacting quantum system is split, frozen-in number fluctuations lead to fluctuations of the relative phase between separated subsystems. We cancel the effect of these fluctuations by immersing the separated subsystems in a common superfluid bath, and demonstrate that we can increase coherence lifetime beyond the projection noise limit. Finally, I discuss our efforts in simulating magnetic ordering in the spin-1 Heisen- berg Hamiltonian.

It is hard to adiabatically ramp into magnetically ordered ground states, because they often have gapless excitations. Instead, we use a spin-dependent lattice to modify interspin interactions, allowing us to ramp into the spin Mott insulator, which has a gap and can therefore act as a cold starting point for exploration of the rest of the phase diagram. We have achieved a cold spin temperature in the spin Mott insulator, and I discuss plans to also achieve a cold charge temperature and then ramp to the the xy-ferromagnet, which has spin-charge separation.