We investigate the spontaneous emission modifications when ensembles of quantum dots (QDs) with differing emission frequencies and finite Lorentzian linewidths are coupled to a microcavity. Using contour integrals we develop a general expression for the rate enhancement when neither the emitter nor the cavity resonance can be treated as a delta function. We show that the ensemble cavity-coupled luminescence lifetimes are generally suppressed in the case of spherical cavities and that the spontaneous emission dynamics of the cavity coupled component becomes increasingly stretched as the coupling factor increases. The Q-factor measured from the luminescence spectrum can be much lower than the intrinsic cavity Q-factor, and is in many practical situations limited by the QD spectral width. The mode spectrum observed in the photoluminescence (PL) spectrum can be largely determined by the QD emission linewidth, permitting this parameter to be extracted without requiring single-particle spectroscopy. In the case of Si-QDs, the linewidth cannot be significantly greater than 10 meV in order to observe spherical cavity resonances in the PL spectrum.

The luminescent properties of SiOx ranging in composition between x=0 and x=2 are presented. Luminescence in the SiOx system is found to be tunable across the full visible spectrum and into the near infrared. The data are used to generate an emission color map for the complete SiOx system. At the lower annealing temperatures, several lines of evidence suggest that the luminescence is due to the presence of amorphous silicon nanoclusters, whereas for higher annealing temperatures the emission is dominated by silicon nanocrystals.

Oxide-embedded silicon nanocrystals (Si-NCs) are a promising material for microphotonics, particularly when coupled to high quality factor (Q-factor) optical cavity modes. By glazing smooth nanocrystalline films from a solution-based precursor onto the surface of optical fibers, the Si-NC luminescence can be strongly coupled into the optical modes of the fiber. Well-developed whispering gallery modes occur in the luminescence of the Si-NCs measured perpendicular to the fiber axis, showing detection-limited Q-factors in the emission spectrum. In addition to providing high Q-factor fluorescence without the need for lithographic procedures, the physical versatility of a nanocrystal-coated fiber suggests possible refractometric applications.

Wavelength tunability is a key aspect for light-emitting materials, and is one of the critical properties driving research on semiconductor quantum dots [1] and organic microcavity devices.[2] Narrow and well-defined emission spectra are important with respect to the development of optical displays, optical amplifiers, light-emitting diodes, and ªtunableº quantum dot lasers. [3] Silicon-based light emitters, in particular, represent a key technology that may challenge the bottlenecks currently facing the microelectronics industry; [4] however, widely tunable and spectrally narrow ªquantum-dot-likeº emission has remained problematic. Here, we demonstrate that siliconrich oxide can be processed at temperatures compatible with silicon microelectronics, while at the same time, the emission wavelength can be tuned throughout the visible spectrum and extended to the fiber-optic transparency windows in the near infrared. This approach covers most of the 0.4 to 1.6 lm wavelength range needed for silicon-based photonics, [4] produces the widest spectral tunability achieved for any single material, and presents an attractive silicon-based alternative to quantum dot or microcavity organic light-emitting materials for photonic applications. Silicon nanocrystal composites are currently the focus of the intense international research efforts that may foreshadow the ªoptical age of siliconº.[5] These efforts are driven by potential photonic applications of silicon-based materials in light-emitting diodes (see, for example, the work of Valenta and co-workers [6] ), nanocrystal lasers, [7,8] and erbium-doped waveguide amplifiers (see, for example, the work of Franzò and co-workers [9] ). Silicon nanocrystal composites represent one of the best hopes for photon creation and transmission in next-generation optical interconnects.[4] Silicon nanocrystals in SiO 2 have a broad luminescence band centered in the near infrared, with a photoluminescence quantum efficiency several orders of magnitude higher than that of bulk silicon.[10]When placed within a distributed Bragg reflector (DBR) structure, the emission is intensified and spectrally narrowed, [11,12] although it is still within the 700±900 nm range characteristic of the natural emission. Nevertheless, this type of fine wavelength control is of recognized importance with respect to luminescent devices. [8] A simpler microcavity consisting of a bottom dielectric stack and a top silver mirror was recently used for a silicon nanocomposite, [13] but wavelength tunability was not demonstrated and the optical modes were broad. One of the main drawbacks with respect to device integration is the high processing temperature (generally at least 1000 C) required to form luminescent silicon nanoparticles via phase separation of a material with excess atomic silicon. Such high temperatures are incompatible with CMOS (complementary metal oxide semiconductor) technology [14] and, from the point of view of DBRs, make the fabrication of devices based on these materials more difficult, ...

The luminescent infrared transitions in Nd3+ can be activated via a transfer mechanism from amorphous silicon nanoclusters. The Nd photoluminescence (PL) has some unusual characteristics, including a weak temperature dependence of the PL intensity. The data are explained using a simple rate equation model which enables an effective nanocluster-to-neodymium transfer time of ∼0.15μs to be extracted. This is short enough to dominate the intrinsic nanocluster decay rates at low temperatures but long enough to imply that the coupling between the nanoclusters and the Nd ions is, in fact, weaker than for Nd-doped bulk silicon or other semiconductors.