Afrl-osr-va-tr-2015-0051 Topological Quantum Information in a 3d Neutral Atom Array

Experimental and theoretical work was performed with the goal of developing the system of neutral atoms in a 3D optical lattice into a flexible platform for quantum information. The effort was specifically directed toward implementing the Kitaev toric code Hamiltonian model, but accomplishing that goal requires building a universal quantum computer with at least ~25 qubits. Toward that end we have demonstrated the execution of single qubit gates on any arbitrary sequence of individual lattice sites in a 5×5×5 array. This entailed improving laser cooling in a 3D large spacing lattice, developing flexible state manipulation techniques, and demonstrating long atomic coherence times (exceeding 5 seconds). We designed, built and installed two MEMS mirror-controlled addressing beams that allow us to rapidly shift target atoms into resonance with microwave fields for the execution of gates. We demonstrated that we can perform single qubit gates in ~500 μs on target atoms without affecting quantum information in non-target atoms. On the theoretical side, we developed a paradigm for implementing digital quantum simulations in finite systems, including a dissipative mechanism that allows thermalization to arbitrary temperatures, and applied this to realization of the toric code Hamiltonian within our trapped neutral atoms architecture. We have made significant progress on all the remaining stages along the path of turning the system into a universal quantum computer, and thus implementing the Kitaev model. The next step is perfectly filling the lattice, which requires the demonstrated single qubit technology, the rapid and perfect determination of site occupancy, which we have demonstrated, and implementation of state-dependent site translations, which are integral to our cooling procedure. The penultimate step will be entanglement of nearby atoms, which requires the overlapping of two each of two color MEMS controlled beams, which technology is nearly complete. It also requires narrow-linewidth Rydberg excitation lasers, the development of which is in progress. Theoretical optimization of entangling operations in the presence of experimental noise is critical to the success of this venture, and this has now been done using methods of coherent control. The final step will be executing all these advances at the same time in the experiment.