Quantum Simulation and Quantum Information Processing with Programmable Rydberg Atom Arrays
Keesling Contreras, Alexander
MetadataShow full item record
CitationKeesling Contreras, Alexander. 2021. Quantum Simulation and Quantum Information Processing with Programmable Rydberg Atom Arrays. Doctoral dissertation, Harvard University Graduate School of Arts and Sciences.
AbstractThe creation and coherent control of systems of many strongly-interacting quantum particles represents one of the most important challenges in quantum science and technology. Devices capable of performing such a task have applications ranging from sensing and simulation, to communications and computation. The past few decades have seen remarkable progress in the development of different platforms to address the challenges of isolating and controlling individual quantum objects and increasing the number of interacting elements.
This thesis presents a novel platform using arrays of individually controlled neutral atoms in optical tweezers which enables high-fidelity control of large systems of strongly interacting particles. Using real-time imaging and feedback, we initialize the positions of hundreds of atoms in arbitrary one- and two-dimensional geometries. By coupling the internal state of the atoms to highly excited Rydberg states, we are able to introduce spatially-dependent interactions to the system. Combining the remarkable coherence properties of isolated neutral atoms with strong Rydberg interactions, we study complex spin Hamiltonians in regimes beyond classical simulability.
In the context of quantum simulation, we use this platform to study the phase diagram of regular arrays, including the critical properties of some of the associated quantum phase transitions. Additionally, we report on the discovery of special slowly-thermalizing states known as quantum many-body scars, which are then used as a starting point to stabilize many-body dynamics across the Hilbert space. Through the addition of specialized local controls, we demonstrate the ability to generate large N-partite entangled states with up to 20 atoms, as well as the ability to implement quantum logic gates. We combine quantum simulation with quantum information processing by preparing states with signatures of quantum spin liquids, which may be developed as topologically protected qubits. Lastly, we explore the potential applications of this platform to classical combinatorial optimization problems by approximating the Maximum Independent Set of different graphs.
Citable link to this pagehttps://nrs.harvard.edu/URN-3:HUL.INSTREPOS:37368497
- FAS Theses and Dissertations