In this thesis, we focus our attention on the excitation of Surface Plasmon Polaritons (SPPs) and their propagation along metal stripes. Plasmons are characterized by losses into the metal, therefore an important step is to investigate the effect of these losses on their quantum properties. This is a field not yet fully investigated and the work presented here will give us the possibility to understand the effect of losses on the plasmons quantum properties. This will allow us to prove that plasmons can be used in the quantum information technology field, since they keep the quantum information regardless of their lossy character. Another key property yet to be fully investigated is the bosonic character of single surface plasmon polaritons (SPPs). The quasi-particle nature of SPPs, consisting of a photon (boson) coupled to a charge density wave of electrons (fermions), makes them an unusual type of quantum excitation. It is, as of yet, unclear whether SPPs are bosons, fermions, or a hybrid mixture. Here, we will prove the bosonic character of plasmons, making use of interference experiments. This study will open opportunities for controlling quantum states of light in ultra-compact nanophotonic plasmonic circuitry. First of all, the mean excitation rates, intensity correlations and Fock state populations are studied by using heralded single photons generated via spontaneous parametric down conversion as sources of light. One downconverted beam is used as a trigger, the other one is the signal we send to the metal stripes to excite the plasmons. After an introduction on the meaning of coherence functions, we explain how we couple photons into a gold waveguide with gratings on both sides, where the coupled plasmon is confined at the interface between gold and air. By measuring the second-order quantum coherence function g2(t), we demonstrated the ability to excite single SPPs. Moreover, the effect of losses incurred during propagation of the single SPPs is consistent with the classical exponential behaviour and does not change the value of g2(t), providing evidence that a linear uncorrelated Markovian loss model is valid for SPP damping at the single quanta level. Then, we move onto more complicated devices, such as X-shaped stripes that act as a plasmonic beamsplitter, in order to observe nonclassical effects in the interference of two single plasmons. This is an important step along the way to understanding better the behaviour of single surface plasmons at the quantum level and how one can build more complicated quantum interference networks, such as plasmonic-based quantum logic gates. In order to fully verify the bosonic nature of single excitations in the quantum regime it is vital to observe quantum interference. A natural thing to probe in the most basic type of scatterer (a 50/50 beamsplitter) operating in the quantum regime, is how it acts on two separate single surface plasmons. Here the launching method is the same as for the previous experiment, except that the waveguide structure is in the form of a 50/50 beamsplitter (X-shape) and both photons from the parametric down-conversion type-I generation are sent onto the two inputs of the plasmonic beamsplitter. By this way, both the beams, generated by the nonlinear crystal, act as signal beams. If the SPPs are truly bosonic and indistinguishable then they tend to bunch together when they interact at the beamsplitter -this is the well known Hong-Ou and Mandel quantum interference effect. In this work we report the first direct observation of quantum interference in the HOM effect for single SPPs, demonstrating by this way the bosonic nature of plasmons.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:656553 |
| Date | January 2014 |
| Creators | Di Martino, Giuliana |
| Contributors | Maier, Stefan |
| Publisher | Imperial College London |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://hdl.handle.net/10044/1/24543 |
Page generated in 0.0017 seconds