We discuss the preparation of many-body states of cold fermionic atoms in an optical lattice via controlled dissipative processes induced by coupling the system to a reservoir. Based on a mechanism combining Pauli blocking and phase locking between adjacent sites, we construct complete sets of jump operators describing coupling to a reservoir that leads to dissipative preparation of pairing states for fermions with various symmetries in the absence of direct inter-particle interactions. We discuss the uniqueness of these states, and demonstrate it with small-scale numerical simulations. In the late-time dissipative dynamics, we identify a 'dissipative gap' that persists in the thermodynamic limit. This gap implies exponential convergence of all many-body observables to their steady-state values. We then investigate how these pairing states can be used as a starting point for the preparation of the ground state of the Fermi–Hubbard Hamiltonian via an adiabatic state preparation process also involving the parent Hamiltonian of the pairing state. We also provide a proof-of-principle example for implementing these dissipative processes and the parent Hamiltonians of the pairing states, based on 171 Yb atoms in optical lattice potentials.
We discuss the preparation of many-body states of cold fermionic atoms in an optical lattice via controlled dissipative processes induced by coupling the system to a reservoir. Based on a mechanism combining Pauli blocking and phase locking between adjacent sites, we construct complete sets of jump operators describing coupling to a reservoir that leads to dissipative preparation of pairing states for fermions with various symmetries in the absence of direct inter-particle interactions. We discuss the uniqueness of these states, and demonstrate it with small-scale numerical simulations. In the late-time dissipative dynamics, we identify a 'dissipative gap' that persists in the thermodynamic limit. This gap implies exponential convergence of all many-body observables to their steady-state values. We then investigate how these pairing states can be used as a starting point for the preparation of the ground state of the Fermi–Hubbard Hamiltonian via an adiabatic state preparation process also involving the parent Hamiltonian of the pairing state. We also provide a proof-of-principle example for implementing these dissipative processes and the parent Hamiltonians of the pairing states, based on 171 Yb atoms in optical lattice potentials.
We show how dissipative dynamics can give rise to pairing for two-component fermions on a lattice. In particular, we construct a parent Liouvillian operator so that a BCS-type state of a given symmetry, e.g., a d-wave state, is reached for arbitrary initial states in the absence of conservative forces. The system-bath couplings describe single-particle, number-conserving and quasilocal processes. The pairing mechanism crucially relies on Fermi statistics. We show how such Liouvillians can be realized via reservoir engineering with cold atoms representing a driven dissipative dynamics.
We propose a fault-tolerant loading scheme to produce an array of fermions in an optical lattice of the high fidelity required for applications in quantum-information processing and the modeling of strongly correlated systems. A cold reservoir of fermions plays a dual role as a source of atoms to be loaded into the lattice via a Raman process and as a heat bath for sympathetic cooling of lattice atoms. Atoms are initially transferred into an excited motional state in each lattice site and then decay to the motional ground state, creating particle-hole pairs in the reservoir. Atoms transferred into the ground motional level are no longer coupled back to the reservoir, and doubly occupied sites in the motional ground state are prevented by Pauli blocking. This scheme has strong conceptual connections with optical pumping and can be extended to load high-fidelity patterns of atoms.
The controlled preparation of quantum many-body states is a central challenge for quantum technologies. While preparing target states coherently is possible in principle, it is often fragile and experimentally demanding. Dissipative state preparation, where engineered environments drive the system toward desired steady states, provides a complementary and powerful paradigm. This thesis explores dissipative methods for preparing complex many-body states along three distinct directions. First, we investigate the dissipative stabilization of injective matrix product states (MPS). Injective MPS are particularly well suited for dissipative preparation because they arise as ground states of frustration-free parent Hamiltonians, allowing for tailored jump operators to steer the system directly into the target state. We showcase the surprising result that such state preparation works even in the dilute limit where steering only acts on a single link. Moreover, we introduce a formalism for engineering additional coherent interactions, extending the set of accessible states. Second, and complementary to steering, we develop and analyze a universal cooling scheme capable of preparing ground states even in frustrated quantum systems. In contrast to steering methods, universal cooling can overcome the glass floor that forbids steering methods to prepare such frustrated ground states. A key insight here is that the presented cooling scheme is completely system-agnostic making it extremely robust. Finally, we pursue dressing of the system eigenstates by means of either classical drives or coherent interactions as a means to influence the action of dissipation on the system. Focusing first on time-dependent systems, we propose a generalized numerical implementation of non-adiabatic master equations suitable for the use of optimal control tools. Additionally, we investigate how the collective coupling between a spin ensemble and a cavity can steer the system into a spin-squeezed steady state. Together, these three avenues illustrate how tailored dissipative dynamics can extend the frontier of many-body state preparation, from stabilizing injective MPS to cooling complex frustrated systems and generating non-classical steady states in the dressed framework.
Strongly correlated quantum many-body systems feature a competition of various energy scales as well as a large number of degrees of freedom. The interplay between them leads to a complex variety of quantum phases, while at the same time yielding the possibility to tune them through external perturbations. A way to gain control over the behavior of these quantum systems, is the engineering of couplings to an external bath. Controlling the dissipative coupling of the system to the environment while introducing a coherent drive allows to tailor quantum many-body phases of matter and light with some desired properties, an approach that has been applied to both ultracold atoms and solid state systems. The system that is central to this work is a system of ultracold interacting fermionic atoms on an optical lattice that are globally coupled to a dissipative optical cavity mode. It features an intricate interplay of energy scales, among which are the atomic short-range interactions, a cavity mediated global interaction and the kinetic motion of the atoms. This results in complex quantum phases and self-ordering processes occurring in the system and makes it a perfect platform to study them. This thesis approaches the many-body atom-cavity coupled system and the possible manifestation of non-trivial quantum phases in two main ways. On one hand, we investigate the steady state phase diagram using our method of many-body adiabatic elimination that goes beyond standard mean-field methods. By including fluctuations in the atoms-cavity coupling in our analytical model we are able to show their importance to determine the mixed state character of the self-ordering phase transition. We identify many-body processes within the self-ordered phase that strongly influence the physical nature of the steady states. Our method captures cavity-cooling and many-body cooling resonances, as well as a novel mechanism, the fluctuation-induced bistability, that emerges from the excited eigenstate spectrum of the effective Hamiltonian. We provide arguments, that these phenomena can be generalized to a large class of atoms-cavity models with comparable atomic and photonic energy scales. On the other hand, we investigate the dynamics across the self-ordering transition using newly developed numerically exact tensor-network based algorithms.
Floquet engineering is an important tool for realizing topologically nontrivial band structures for charge-neutral atoms in optical lattices. However, the preparation of a topological-band-insulator-type state of fermions, with one nontrivial quasi-energy band filled completely and the others empty, is challenging as a result of both driving induced heating as well as imperfect adiabatic state preparation (with the latter induced by the unavoidable gap closing when passing the topological transition). An alternative procedure that has been proposed is to prepare such states dissipatively, i.e. as a steady state that emerges when coupling the system to reservoirs. Here we discuss a concrete scheme that couples the system to a weakly interacting Bose condensate given by second atomic species acting as a heat bath. Our strategy relies on the engineering of the potential for the bath particles, so that they occupy weakly coupled tubes perpendicular to the two-dimensional system. Using Floquet-Born-Markov theory, we show that the resulting nonequilibrium steady state of the driven-dissipative system approximates a topological insulator. We even find indications for the approximate stabilization of an anomalous Floquet topological insulator, a state that is impossible to realize in equilibrium.
We discuss how an η condensate, corresponding to an exact excited eigenstate of the Fermi-Hubbard model, can be produced with cold atoms in an optical lattice. Using time-dependent density matrix renormalization group methods, we analyze a state preparation scheme beginning from a band insulator state in an optical superlattice. This state can act as an important test case, both for adiabatic preparation methods and the implementation of the many-body Hamiltonian, and measurements on the final state can be used to help detect associated errors.