Controlled dynamics in ultracold atomic systems

          @misc{ pirsa_PIRSA:06040004,
            doi = {10.48660/06040004},
            url = {https://pirsa.org/06040004},
            author = {Jaksch, Dieter},
            keywords = {Quantum Information},
            language = {en},
            title = {Controlled dynamics in ultracold atomic systems},
            publisher = {Perimeter Institute},
            year = {2006},
            month = {apr},
            note = {PIRSA:06040004 see, \url{https://pirsa.org}}
          }
          

Abstract

I will investigate the creation and detection of multipartite entangled states in systems of ultracold neutral atoms trapped in an optical lattice. These setups are scalable, highly versatile and controllable at the quantum level. Thus they provide an ideal test bed for studying the properties of multipartite entangled states. I will first present methods exploiting incoherent dynamics for initializing an atomic quantum register. The immersion of an optical lattice in a Bose-Einstein condensate leads to spontaneous emission of phonons. This process can be used for irreversibly loading and cooling atoms within the lowest Bloch band of the lattice. I will describe loading and cooling schemes based on this mechanism and compare them to conventional loading schemes. I will then show how coherent dynamics in a very strongly interacting 1D optical lattice setup can be used for the efficient generation of arbitrary graph states in the atomic quantum register. This system can be mapped onto an XY spin chain which itself is equivalent to a system of non-interacting fermions. By exploiting the anticommutation relations between these fictitious fermions I will discuss how any graph state can be realized in an efficient and robust way. In the final part of my talk I will present a practical method for detecting and characterizing multipartite entangled states in atomic quantum registers. This scheme is based on measuring violations of entropic inequalities using simple quantum networks involving only two copies of the quantum state under consideration. I will investigate the performance of this method under realistic conditions taking into account the most common sources of experimental errors.