Towards efficient quantum repeater nodes based on solid-state quantum memories

A la portada: QPSA - Quantum Photonics with Solids and Atoms Group, ICFO - the Institute of Photonic Sciences (English) Quantum repeaters are the foundation of future long-distance quantum networks. In most architectures, their functional core is constituted by quantum memories, which are devices that can store and re-emit photonic quantum information on-demand. The goal of this thesis is to progress towards efficient quantum repeater nodes enabling quantum correlations between telecom photons and matter qubits. To these ends, we performed three main experiments. In our first work, we built a solid-state entanglement photon source with embedded storage capabilities. This emissive quantum memory was implemented in a Pr3+:Y2SiO5 crystal, by means of the atomic frequency comb (AFC) protocol. Thanks to the AFC, we were able to adapt the Duan-Lukin-Cirac-Zoller (DLCZ) protocol, initially conceived for cold atoms, to a solid-state ensemble. This experiment proved that we can produce light-matter entanglement between a heralding photon, at 606 nm, and a spin-wave excitation delocalized inside the ensemble. The matter excitation could be read on-demand at a later time with a read pulse, and mapped as a second photon, at 606 nm as well, emitted by the memory. Quantum correlations between the two photons were measured, enabling the violation of a Bell inequality, thus demonstrating the presence of entanglement. The read-out efficiency of this experiment was low, 1.6%, but solutions were identified to increase this value. In the second experiment, we laid the groundwork for the quantum frequency conversion (QFC) of these photons to the telecom band. The long duration of these photons, up to 1 µs, makes their conversion with high signal-to-noise ratio (SNR) challenging. The conversion from the visible 606 nm wavelength to the telecom regime (1552 nm) was achieved by difference-frequency generation (DFG) in a PPLN waveguide using a strong pump field at 994 nm. A proof of principle with weak coherent pulses showed that we can convert µs-long photons with the low heralding efficiency of the previous experiment with a SNR around 2.6. This sets the stage for interfacing an AFC-DLCZ memory, working at 606 nm, with the telecom network and with material systems working at a different wave length. Finally, in the last experiment, we implemented an AFC impedance-matched cavity (IMC) storage experiment. It has been demonstrated theoretically and experimentally that the IMC enhan