Description
Rare-earth quantum memories could fulfil several roles in future quantum networks, including long-lived memories and quantum light sources for repeaters, high-capacity cache memories for quantum computers, and interconnects for heterogeneous quantum processors. These applications demand memory efficiencies above 90%, fidelities exceeding 99.9%, storage capacities of more than 1000 modes, and data rates beyond 1 MHz. While recent demonstrations have achieved fidelities above 99% and steadily improving efficiencies, further progress is increasingly limited by the simplicity of the models used to guide experiments.
Semiclassical Maxwell–Bloch models accurately describe light–matter interactions in rare-earth crystals, but those commonly used were adapted from atomic gas systems and neglect important solid-state effects. In particular, they ignore crystalline birefringence and, in erbium systems, the mixed electric–magnetic nature of the optical transition dipole moment. These omissions are known to limit quantitative accuracy, but we show they also lead to qualitatively incorrect predictions [1]. Most importantly, conventional models completely fail to describe the interaction of polarised light with the crystal, forcing existing quantum-memory protocols to ignore the polarisation degree of freedom.
Here we use an extended Maxwell–Bloch framework that incorporates birefringence and the mixed moment to investigate the storage and retrieval of polarised light in rare-earth quantum memories. We show how the mixed-parity dipole moment can be exploited, explain the poor performance observed in some previous memory demonstrations, and identify protocol modifications that enable storage of polarisation qubits. We further demonstrate how polarisation can be used to enhance memory efficiency and functionality.
| I am the presenting author | Yes |
|---|