Erbium is a rare-earth metal that, when incorporated in a solid, can emit light at a wavelength of 1.5 μm. It plays a key role in current day telecommunication technology as the principle ingredient of optical fiber amplifiers. In this thesis the control of the Er spontaneous emission in three different types of microphotonic materials is described. Part I of this thesis focuses on the effect of a metallo-dielectric interface on the spontaneous emission of optical emitters in silica glass. It is shown that Er ions near a Ag interface can couple to surface plasmons (SPs) via a near-field interaction. By coupling SPs out into the far field, large changes in the Er photoluminescence emission distribution, spectra, and polarization can be observed. The excitation of SPs also results in an increase of the Er photoluminescence decay rate. The observed decay rates are in good agreement with calculations based on a classical dipole oscillator model. From the change in photoluminescence decay rate of Si nanocrystals near a Ag interface it is shown that Si nanocrystals can efficiently excite SPs and have an internal quantum efficiency of 77 %. Part II focuses on the effect of a microcavity on the spontaneous emission of Er and describes how ion implantation can be used to dope dielectric microresonators with optically active Er ions. The fabrication and characterization of an Er ion-implanted silica microsphere resonator is described that shows lasing at 1.5 μm when pumped above its lasing threshold. Ion implantation is also used to dope toroidal microcavities on a Si chip with Er. The microtoroids are doped by either pre-implantation into the SiO2 base material, or by post-implantation in a fully fabricated microtoroid. The optical activation of Er ions in the microtoroid is investigated and Er lasing at 1.5 μm is observed for both types of microcavities with the lowest threshold (4.5 μW) for the pre-implanted microtoroids. Part III describes the fabrication of an Er-doped Si-inverse opal photonic crystal. These photonic crystals can potentially have a photonic bandgap that can fully inhibit the spontaneous emission of on optical emitter. Fabrication criteria are derived for such a photonic crystal, based on the lattice parameter, filling fraction, and Si refractive index. In the opal photonic crystal composed of both Si and SiO2 we show that Er ions can be selectively excited in both the Si and SiO2 part of the photonic crystal by changing the excitation wavelength and/or the measurement temperature.
Erbium is a rare-earth metal that, when incorporated in a solid, can emit light at a wavelength of 1.5 μm. It plays a key role in current day telecommunication technology as the principle ingredient of optical fiber amplifiers. In this thesis the control of the Er spontaneous emission in three different types of microphotonic materials is described. Part I of this thesis focuses on the effect of a metallo-dielectric interface on the spontaneous emission of optical emitters in silica glass. It is shown that Er ions near a Ag interface can couple to surface plasmons (SPs) via a near-field interaction. By coupling SPs out into the far field, large changes in the Er photoluminescence emission distribution, spectra, and polarization can be observed. The excitation of SPs also results in an increase of the Er photoluminescence decay rate. The observed decay rates are in good agreement with calculations based on a classical dipole oscillator model. From the change in photoluminescence decay rate of Si nanocrystals near a Ag interface it is shown that Si nanocrystals can efficiently excite SPs and have an internal quantum efficiency of 77 %. Part II focuses on the effect of a microcavity on the spontaneous emission of Er and describes how ion implantation can be used to dope dielectric microresonators with optically active Er ions. The fabrication and characterization of an Er ion-implanted silica microsphere resonator is described that shows lasing at 1.5 μm when pumped above its lasing threshold. Ion implantation is also used to dope toroidal microcavities on a Si chip with Er. The microtoroids are doped by either pre-implantation into the SiO2 base material, or by post-implantation in a fully fabricated microtoroid. The optical activation of Er ions in the microtoroid is investigated and Er lasing at 1.5 μm is observed for both types of microcavities with the lowest threshold (4.5 μW) for the pre-implanted microtoroids. Part III describes the fabrication of an Er-doped Si-inverse opal photonic crystal. These photonic crystals can potentially have a photonic bandgap that can fully inhibit the spontaneous emission of on optical emitter. Fabrication criteria are derived for such a photonic crystal, based on the lattice parameter, filling fraction, and Si refractive index. In the opal photonic crystal composed of both Si and SiO2 we show that Er ions can be selectively excited in both the Si and SiO2 part of the photonic crystal by changing the excitation wavelength and/or the measurement temperature.
Since the advent of vertical cavity surface-emitting lasers, a renewed interest has also emerged for the spontaneous emission from Fabry-Pérot microcavities. Changes in spontaneous emission characteristics in a 1D microcavity were predicted long ago,1 but have been demonstrated only recently in the optical domain for solid-state, e.g. semiconductor, structures where changes in spectral purity, lifetime, and emission intensity were observed.2 However, the interpretation of these effects is somewhat obscure, because these structures are highly absorbing and often involve band-to-band, rather than atomic, transitions. In the present study, we use optically transparent Si/SiO2 cavities, consisting of two coplanar Si/SiO2 distributed Bragg reflectors (DBR) around a (λ/2) SiO2 active region, which was doped with ~0.2 at.% Er3+ ions through MeV ion implantation. The Er3+ ions in SiO2 exhibit a sharp, atomic-like, intra-4f transition around λem=1.54 μm, which is effectively shielded by the outerlying filled 5s2 and 5p6 shells. Recently, we reported on the considerable enhancement of the spontaneous emission intensity out of such an Er-doped microcavity.3 In this paper, we present a more detailed study of the modified spontaneous emission, including fluorescence decay measurements to study the effect of the cavity on the spontaneous emission rate.
Photonic crystals, i.e. materials that have a periodic variation in refractive index, form an interesting new class of materials that can be used to modify spontaneous emission and manipulate optical modes in ways that were impossible so far. This thesis is divided in three parts. \nPart I discusses the design and fabrication of two-dimensional photonic crystals in silicon using deep anisotropic etching with a SF6/O2 plasma. The etching process was optimized for the fabrication of two-dimensional photonic crystals by tuning the main parameters of the etching process, i.e. temperature, bias voltage and O2 flow. Vertical confinement in these structures is provided by integrating the structures in a dielectric waveguide. For this purpose, amorphous silicon, silicon-on-insulator and SiGe structures were considered. Fabrication of structures in both amorphous silicon and silicon-on-insulator was successfully demonstrated. The incorporation of luminescent species, such as laser dyes, was demonstrated using a new wet chemical coating technique that forms thin silica layers on a substrate.\nPart II discusses the modification of spontaneous emission in one dimensional systems by studying the decay rate of luminescing Cr ions close to a dielectric interface. The decay rate of the Cr ions can be changed by bringing the samples into contact with a range of liquids with different refractive indices. The change in radiative decay rate can be calculated by calculating the local density of states. To explain the experimental results additional non-radiative decay channels have to be introduced and yields a quantum efficiency of ~50% for the Cr R-line luminescence. This concept was further extended to a thin silica layer on silicon implanted with erbium ions and resulted in the radiative rate of erbium in pure silica: 54 s-1. This number was used to analyze the decay rate of erbium ions in silica colloidal spheres that can be used as building block for three-dimensional photonic crystals by self-assembly.\nFinally, Part III discusses the optical properties and modified spontaneous emission from a three-dimensional silicon photonic crystal of finite (5-layers) thickness. The crystals are made in a layer-by-layer approach using lithographic tools and show near 100% reflection in the 1.4-1.7 mu m wavelength range indicative of a photonic stopgap. A direct comparison with the calculated reflectivity reveals that some features in the reflectivity can be ascribed to the finite thickness of the crystal, while other features can be explained in terms of a superstructure that leads to zone folding of the photonic bandstructure. The collected spontaneous emission from erbium implanted crystals is strongly reduced for wavelengths in the stopgap from 1.4-1.7 mu m. The changes in collected luminescence intensity are explained in terms of a rate equation model that takes into account the effect of Bragg scattering, the local density of states and the quantum efficiency of the emitters inside the crystal. Using this model a spectral attenuation of 5 dB per unit cell at 1.539 mu m wavelength is obtained from the experimental data, which is in perfect agreement with existing theory and transmission data.
Energy transfer in erbium doped optical waveguides based on silicon This thesis describes the energy transfer processes\noccurring in materials that can be used for the fabrication of silicon compatible optical integrated circuits, operating at 1.54 mm.The thesis consists of three parts: Part I discusses energy transfer in optical waveguide amplifiers on silicon. The waveguide amplifiers use an intra-4f transition of the rare earth ion erbium occurring at 1.54 mm. To achieve significant amplification, high\n Er concentrations (> 10 20 cm -3 ) are required. At these concentrations Er ions can interact, resulting in co-operative\n upconversion, in which one excited Er ion is excited into a higher energy state, while a neighboring Er ion returns to the ground\n state. In Part I this and other gain limiting effects are investigated. In particular, it is shown that the atomic-scale Er distribution\n is of crucial importance for the performance of Er doped waveguide amplifiers. Part II discusses the energy transfer processes\n occurring in Er doped silicon. It is shown that Er in silicon can be excited by electron-hole pair recombination in an\n impurity-Auger effect. This effect may be used to fabricate an electrically pumped Er doped waveguide amplifier. The reverse\n process can also occur: an excited Er ion can de-excite and generate an electron-hole pair in the silicon, due to a non-radiative\n backtransfer process. This process has an efficiency of 70% at room temperature. This observation enables the fabrication of an\n Er doped photo-detector in a Si waveguide operating at 1.54 mm. The performance of such a detector is discussed. Part III\n discusses the interaction between Er and silicon nanocrystals embedded in SiO2. Si nanocrystals efficiently absorb visible light,\n resulting in the generation of quantum confined electron-hole pairs. The electron-hole pairs efficiently transfer energy to Er.\n Due to the energy transfer, the effective Er absorption cross section is five orders of magnitude larger than for direct optical\n excitation. Silicon nanocrystals may thus be used to increase the pumping efficiency of an Er doped waveguide amplifier. It is\n shown that a single nanocrystal can excite ~1 Er ion at a time and that an optical gain of 0.6 dB/cm may be achieved in a typical\n nanocrystal sensitized waveguide amplifier. Due to the broad absorption band of Si nanocrystals, a broad band light source can\n be used as pump source in such a nanocrystal sensitized Er doped waveguide amplifier.
Spherical SiO2 colloids with two different diameters (175 nm, 340 nm) were doped with erbium at different concentrations. The spheres show sharply peaked photoluminescence centered at 1.535 μm, due to intra-4f transitions in Er3+. From measurements of the Er decay rate for different Er concentrations the decay rate of isolated Er ions (i.e., in absence of concentration quenching) was determined for the two colloid diameters. The data were compared to spontaneous emission rates derived from calculations of the local optical density of states in the colloids. The calculation predicts a large difference in the spontaneous emission rate for both colloid sizes (61 vs 40 s−1), in perfect agreement with the measured data.
Summary form only given: Amplification in rare-earth doped solid state planar optical structures has attracted a lot of interest in recent years due to the fact that the rare earth ions show sharp absorption and emission bands and relatively long luminescence lifetimes. In particular the erbium (Er) ion is very attractive because of its intra 4f-transition from the first excited state to the ground state around 1.53 mum (4I13/2-4I15/2). The main problem of the erbium doped waveguide amplifiers (EDWAs) is that high values of the Er3+concentration are required to reach high gain figures in a compact length, but in this way the concentration quenching effects become very considerable and detrimental to the optical gain, since they reduce the population in the first excited state. A good solution to overcome this dramatic drawback is to consider erbium doped photonic crystal amplifier (EDPCA). In fact the small group velocity of electromagnetic eigenmodes brings about the enhancement of stimulated emission if the photonic crystals are doped by active media as the rare earth ions E3+r or Yt3+. This phenomenon originates from the long interaction time between radiation field and the matter caused by the small group velocity near the edge of the photonic band gap. The efficient confinement of light is achievable not only by tuning the input signal at the band edge of the gap of the photonic crystal but it is also obtainable by introducing local defects inside photonic crystals. The density of electromagnetic states inside a cavity is significantly modified and the spontaneous emission of atoms in a cavity can be either enhanced or inhibited. Erbium doped silica-titania one-dimensional photonic crystal microcavities have been analyzed by solving the erbium rate equations and the Maxwell equations by means of an auxiliary differential equation finite difference time domain technique. Since the performance of optical amplifiers depends on the populations of the energy levels participating in radiative transitions, determined principally by the distribution and concentration of the rare-earth ions and by the pumping conditions, the relations between these parameters and the transmittance of the EDPCMA have been studied
The need for higher bandwidth and people's desire to be "always connected" have spurred a new era of silicon photonics. The traditional integrated electrical transmission lines have been an obstacle preventing ultra high speed communication. Using monolithic chips of integrated optoelectronic circuits from silicon provides an economic way to realize Tetra Byte/Second bandwidth in a variety of areas such as "fiber to the home" and the buses linking chips inside computer. The heart of such optoelectronics-silicon laser-is still in pursuit. One of the most promising approaches is the erbium doped silicon nanocrystals embedded in silica system. External photon or hot electrons injection excites the silicon nanocrystals, which then transfer their energies to nearby erbium ions to emit light at 1.55 μm wavelength range. In this thesis, we investigate the effects of cavity and electric field on energy transfer from Si nanocrystals (Si-nc's) to Er ions, and simulate material gain in such systems. Our results show that microcavity can enhance the Forster energy transfer and material gain, reducing the requirements for Si-nc pumping. The electric field will hinder the radiation decay of Si-nc, but we have to further explore the tunneling mechanism before concluding the overall effect of electric field. Some future work needs to be done, which will shine some light on the design of the silicon laser.