Karakterisering och systemmodellering av Time-Bin-kodad chip-till-chip-kvant Entanglement på tunnfilmslitiumniobat
Emma Rui Yi Koh
Abstract
Quantum Key Distribution (QKD) provides a form of secure communication against quantum computing attacks. Many QKD protocols require a source of entangled photons, which uses the arrival time of the photon to encode qubit information. These entangled photons are ideally produced on-chip to improve scalability. Atpresent, majority of on-chip entangled photons are produced and measured on the same chip, which does not reflect actual QKD systems that send each half of entangled photon pairs to different experimental nodes for measurement. Therefore, this project focus on the characterisation and system modelling of a chip-to-chipsource of entangled photons. The source is time-bin encoded, which uses the arrival time of photons to encode qubit information, and takes place on thin-film lithiumniobate (TFLN) integrated photonics chips. Light pulses (pump pulses) with a time separation of 220 ps per cycle, are coupled into a TFLN chip, and pass through aperiodically poled lithium niobate (PPLN) waveguide. They exit the waveguide as entangled photon pairs. One half of each pair is then sent to a Franson interferometer on the same chip, while the other half is sent to a separate TFLN chip with an identical interferometer. By measuring the outputs of the two interferometers,we can determine if the correlations between the photons are classical or quantum, by testing for the violation of the Clauser–Horne–Shimony–Holt (CHSH) inequality.We tuned the experimental system to balance amplitudes of all states, but slight imbalances still existed. We extracted the amplitude parameters from the experimental system, then finally modelled the experimental system along with an ideal system with perfectly balanced amplitudes. If µ is the mean number of quantumentangled pairs formed by spontaneous parametric down-conversion (SPDC) per in coming pulse into the PPLN, then for the simulated experimental system, we wereable to simulate entanglement for a µ up to 0.09, and for the perfectly balanced ideal system, we simulated entanglement up to a µ of 0.19. Thus we conclude that entanglement was hypothetically proven in the system model, although improvements can still be made.