Efficient light emission from bismuth-doped rare earths compounds for Si microphotonics

For almost 50 years, silicon microelectronics has been the engine of the modern information revolution, owing to the advent of the nanotechnology era that has produced always faster, cheaper, smaller and more performing devices inside the same silicon chip. However its fortune might be coming to an end. Indeed, as a consequence of the continuous reduction in size, much longer electrical interconnects are required, thus leading to an increase of signal delays and to electromagnetic interferences that cause power dissipation and limit the total performances of a chip. A possible solution to this problem can be obtained by replacing the electrical interconnections with the optical ones by the realization of a totally integrated photonic circuit on a silicon platform. For this reason, recently strong efforts have been devoted to the development of the main constituents of a photonic circuit, such as waveguides, splitters or multiplexers, detectors, etc. However there is still a lack of integrated infrared light sources that are required to generate logic value 1-0 and to compensate the optical losses in waveguides. In addition recently also the demand for efficient integrated visible light sources for LEDs, displays and lab-on-chip applications is increasing. Different strategies have been proposed to satisfy these requirements and will be reviewed in the following chapter, by evidencing their strengths and their limits. In order to further increase the efficiency of these light sources and to realize integrated photonic circuits, their coupling with passive devices, such as SOI or plasmonic waveguides and photonic crystals, has been recently proposed owing to the existence of several interesting effects. Two completely different approaches will be pursued in this thesis work: (i) the synthesis of silicon compatible REs compounds in which the REs amount can be varied in a continuous way increasing the efficiency of the light source without suffering from detrimental effects and (ii) the contemporary introduction of post-transition metals as either strong emitting elements in the visible range or as sensitizers for the infrared emission, thus suggesting these materials as good candidates for an integrated light source on silicon. In particular in Chapter 2, the needs of new light sources in the visible range will be faced. Bismuth is proposed as an emitter for Si-based transparent materials in place of the most common used REs, as Eu and Tb: this element indeed can improve much more the optical efficiency of the systems thanks to its peculiar electronic configuration and to its high absorption and emission cross sections. Its introduction in two different Si-compatible yttrium based hosts, the yttrium disilicate and the yttrium oxide, will be discussed. These matrices are indeed suitable for the introduction of dopant elements in the Y3+ substitutional position. In particular, the influence of different annealing atmospheres on the structural and optical properties will be presented for both the hosts, thus proposing them as efficient, Si compatible optical materials for applications on Si platforms as down-converters for the solar spectrum and as broad and tunable emitters in the visible range. In Chapter 3, the possibility to exploit also infrared emission by involving erbium-yttrium mixed compounds will be addressed. This approach permits to increase the Er content up to the constituent level without optically inactive clusters formation. The additional introduction of Bi as a sensitizer for Er will be proposed. Therefore by the optimization of the structural and optical properties, the coupling Bi-Er will be demonstrated in order to enhance Er optical emission at 1.54 micron up to 2000 times. This result makes Bi-Er-Y mixed compounds good candidates for light emission and amplification in the telecommunication windows, thus achieving an interesting goal for Si-microphotonics.