| dc.description.abstract | The distinct properties of nanoparticles compared to their bulk counterparts makes them a suitable candidate for the development of many emerging technologies, and this presents a demand for scalable production routes. Gas-phase nanoparticle synthesis is the most scalable method to achieve this goal, but developing these techniques requires diagnostics that can characterize nanoparticle morphology in situ. On the other hand, the incidental release of nanoparticles to the environment contributes to climate change and severely influences human health. Therefore, effective and robust diagnostics for assessing the morphology of aerosolized nanoparticles are essential for understanding and mitigating its impacts, assessing and developing environmental regulations, understanding how nanoparticles are formed in combustion and gas-phase synthesis systems, and providing feedback for closed-loop control systems for nanoparticle production.
In situ optical methods such as multi-angle light scattering (MALS) and time-resolved laser-induced incandescence (TiRe-LII) have a robust temporal and spatial resolution, which are less expensive and time-consuming compared to transmission electron microscopy, and permit online control. Both of these methods depend on precise spectroscopic models to connect the observed signals to the aerosol properties. In the case of MALS, the morphological parameters of polydisperse aerosolized soot can be found by regressing modeled angularly-resolved elastic light scattering to experimental measurements, but this is an ill-posed problem in the presence of measurement noise or model error. Rayleigh-Debye-Gans Fractal Aggregate (RDG-FA) theory provides a closed-form solution for the light scattering kernel in the measurement model but can be subject to as much as 30% model error compared to the exact solution, which is amplified by the ill-posedness of the inference problem into significant errors in the recovered morphological parameters. More precise approaches, e.g. the multi-sphere T-matrix method (MSTM), are too expensive for inference problems, which require repeated evaluation of the forward model. The efficiency of RDG-FA and the accuracy of MSTM can be combined by modeling the approximation error. The error function is derived from a principal component analysis on error matrices generated using randomly-sampled aggregates. The error model is then used to correct the RDG-FA kernel in the forward model for a particular set of fractal parameters. The model is then used to estimate probability densities of the size distribution and aggregate fractal parameters via Bayesian inference. Alternatively, an artificial feed-forward multi-layered neural network (ANN) can be trained using MSTM scattering simulations on randomly-generated soot aggregates. The ANN is then used to approximate the light scattering kernel in the measurement model, which is incorporated into the Bayesian inference procedure. The Bayesian/ANN approach is shown to be more accurate compared to the Bayesian approximation error technique. The Bayesian/ANN is then applied to in-flame measurements of soot and results are compared with transmission electron microscopy results from the literature.
While MALS is mainly used to infer the size distribution of aggregates and usually a deterministic primary particle size is assumed in the model, TiRe-LII is increasingly applied to characterize the size distribution of soot primary particles and non-carbonaceous nanoparticles as well as the thermophysical properties of the bulk material. However, there exist several measurement phenomena, particularly from signals generated from metal nanoaerosols, that cannot be explained using traditional models. This thesis shows that some of these phenomena may be due to errors caused by using the Rayleigh approximation of Mie theory, which is a standard approach for modeling the spectral absorption of carbonaceous nanoparticles but is generally invalid for metal nanoparticles.
There has also been speculation that several commonly-observed measurement phenomena in TiRe-LII measurement data may be caused by bremsstrahlung emission from a laser-induced plasma, a phenomenon known to occur at higher fluences typical of laser-induced breakdown spectroscopy. This thesis presents the theoretical framework to investigate a laser-induced plasma formation under LII measurement conditions and explores how this plasma may affect time-resolved spectral intensity measurements. At fluences larger than 8 mJ/mm2, the absorption cross-section of the laser-energized nanoparticle is enhanced due to inverse bremsstrahlung absorption, and bremsstrahlung emission results in an overestimation of the nanoparticle temperature due to the corruption of the incandescence signal.
In the case that aerosolized nanoparticles with a low absorption cross-section at the laser wavelength, the neutral bremsstrahlung emission can be detected during the experiments due to the absence of nanoparticle incandescence emission contingent on electron emission to the gas phase from the nanoparticle. Measurements carried on silver (Ag) and gold (Au) nanoparticles within the size range of 30 nm to 60 nm excited with a 1064 nm nanosecond Nd:YAG laser pulse. Assuming that the detected signals are due to incandescence from the nanoparticle a pyrometric temperature is defined which varied with buffer gas molecule type and showed a linear relation with laser fluence that suggests that the signal is not, in fact, incandescence. A new model is proposed based on plasmonically-enhanced photoemission of electrons from the nanoparticles. The interaction of the electrons with buffer gas neutral species leads to inverse neutral bremsstrahlung absorption of the laser pulse as well as neutral bremsstrahlung emission.
In summary, this thesis not only improves the spectroscopic models of two aerosol metrology techniques but also proposes new methods that could lead to a faster inference of aerosol properties and presents a set of new approaches to explain the TiRe-LII model deficiencies. Also, newly proposed models for TiRe-LII could connect the field to other research areas such as laser-induced plasma diagnostic, LIBS on aerosols, and nanoantenna study. | en |
