Molecular fluorophores and semiconductor quantum dots (QDs) have been used as cellular imaging agents for biomedical research, but each class has challenges associated with their use, including poor photostability or toxicity. Silicon is a semiconductor material that is inexpensive and relatively environmental benign in comparison to heavy metal-containing quantum dots. Thus, red-emitting silicon nanoparticles (Si NPs) are desirable to prepare for cellular imaging application to be used in place of more toxic QDs. However, Si NPs currently suffer poorly understood photoinstability, and furthermore, the origin of the PL remains under debate. This dissertation first describes the use of diatomaceous earth as a new precursor for the synthesis of photoluminescent Si NPs. Second, the stabilization of red PL from Si NPs in aqueous solution via micellar encapsulation is reported. Thirdly, red to blue PL conversion of decane-terminated Si NPs in alcohol dispersions is described and the origins (i.e., color centers) of the emission events were studied with a comprehensive characterization suite including FT-IR, UV-vis, photoluminescence excitation, and time-resolved photoluminescence spectroscopies in order to determine size or chemical changes underlying the PL color change. In this study, the red and blue PL was determined to result from intrinsic and surface states, respectively. Lastly, we determined that the blue emission band assigned to a surface state can be introduced by base addition in originally red-emitting silicon nanoparticles, and that red PL can be restored by subsequent acid addition. This experimentally demonstrates blue PL is surface state related and can overcome the intrinsic state related excitonic recombination pathway in red PL event. Based on all the data collected and analyzed, we present a simple energy level diagram detailing the multiple origins of Si NP PL, which are related to both size and surface chemistry.

While silicon has long been utilized for its electronic properties, its use as an optical material has largely been limited due to the poor efficiency of interband transitions. However, discovery of visible photoluminescence (PL) from nanocrystalline silicon in 1990 triggered many ensuing research efforts to optimize PL from nanocrystalline silicon for optical applications. Currently, use of photoluminescent silicon nanoparticles (Si NPs) is commercially limited by: 1) the instability of the energy and intensity of the PL, and 2) the low quantum yield of interband PL from Si NPs. Herein, red-emitting, hydrogen-passivated silicon nanoparticles (H-Si NPs) were synthesized by thermally-induced disproportionation of a HSiCl3-derived (HSiO1.5)n polymer. The H-Si NPs produced by this method were then subjected to various chemical and physical environments to assess the long-term stability of the optical properties as a function of changing surface composition. This dissertation is intended to elucidate correlations between the reported PL instability and the observed changes in the Si NP surface chemistry over time and as a function of environment. First, the stability of the H-Si NP surface at slightly elevated temperatures towards reactivity with a simple alkane was probed. The H-Si NPs were observed by FT-IR spectroscopy to undergo partial hydrosilylation upon heating in refluxing hexane, in addition to varying degrees of surface oxidation. The unexpected reactivity of the Si surface in n-hexane supports the unstable nature of the H-Si NP surface, and furthermore implicates the presence of highly-reactive Si radicals on the surfaces of the Si NPs. We propose that reaction of alkene impurities with the Si surface radicals is largely responsible for the observed surface alkylation. However, we also present an alternate mechanism by which Si surface radicals could react with alkanes to result in alkylation of the surface. Next, the energy and intensity stability of the interband PL from H-Si NPs in the presence of a radical trap was probed. Upon addition of (2,2,6,6,-tetramethyl-piperidin-1-yl)oxyl (TEMPO), the energy and intensity of the interband transition was observed to change over time, dependent on the reaction conditions. First, when the reaction occurred at 4ºC with minimal light exposure, the interband transition exhibited a gradual hypsochromic shift to between 595 nm and 655 nm, versus the λmax of the original low energy emission peak at 700 nm, depending on the amount of TEMPO in the sample. Second, when the reaction proceeded at room temperature with frequent exposure to 360 nm irradiation, the original interband transition at 660 nm was quenched while a new peak at 575 nm developed. Based on all the data collected and analyzed, we assign the 595 -- 655 nm transition as due to interband exciton recombination from Si NPs with reduced diameters relative to the original Si NPs. We furthermore assign the 575 nm transition as due to an oxide-related defect state resulting from rapid oxidation of photo-excited Si NPs.