Below we highlight two main achievements from ATOMAG, one focused on efficient atom-light interfaces for quantum information processing and the other focused on generating exotic many-body states.
Project 1: Selective radiance in super-wavelength atomic arrays
A new generation of efficient atom-light interfaces has recently been proposed based on the notion of selective radiance, where destructive wave interference is utilized as resource to suppress emission into unwanted optical modes. It is generally assumed that this strategy requires dense atomic arrays with sub-wavelength lattice constants. While a 2D super-wavelength array is a very poor atom-light interface, due to photons being scattered into multiple diffraction orders, we demonstrate that one can restore the selective radiance by stacking multiple 2D layers. Using an idealized model, we identify a range of super-wavelength mirror configurations that selectively radiate into the target specular mode at an enhanced rate, while scattering into all diffraction orders is eliminated through interlayer destructive interference. Guided by this intuition, we show that realistic super-wavelength arrays can almost perfectly reflect a weak classical beam on resonance, and also be functionalized into efficient quantum memories for single photons. In particular, by locally optimizing the atomic positions we show that one can in principle achieve errors on the order of ~1% with only around ~100 atoms.
Project 2: Emergence of quantum spin liquids from global atom-cavity interactions
Conventional cavity QED phenomena, such as Dicke superradiance, are typically dominated by the semi-classical behaviour of the ensemble's collective spin (S ~ N/2). To move towards complexity, we have demonstrated that global cavity-induced fluctuations can melt classical Ising magnets into quantum spin liquids (QSLs) that exhibit fractionalized excitations and emergent gauge fields. Our key idea is to utilize a strong cavity to project the system into the global singlet sector (S = 0), thereby evading collective-spin physics. By introducing short-range Ising perturbations, such as those arising from Rydberg interactions, the low-energy states map exactly onto the singlet sector of the corresponding short-range Heisenberg model, which can host a variety of QSLs. Focusing on the J1-J2 square lattice model as a paradigmatic example, we show that the cavity initially squeezes the classical antiferromagnetic states by generating EPR-like entanglement between sublattices, and then magnon-magnon interactions melt the classical order into the candidate QSL ground state of the Heisenberg model.
