Simulating high flux solar radiation and assessing its influence on a sooty flame

Abstract

Integrating concentrated solar thermal energy into fossil-fuels for the production of power/clean fuels is receiving growing attention as the combination of the two energy sources can provide lower emissions of carbon and other pollutants, lower cost, and continuous supply. Various types of hybrid concepts have been proposed. However, all of these concepts employ stand-alone solar receivers and standalone combustors. The University of Adelaide has developed an alternative approach with which to fully integrate a combustor into a solar cavity receiver. This offers the potential for significant savings from reduced infrastructure investment and reduced start-up and shut-down losses. In addition, this hybrid also results in the direct interaction between concentrated solar radiation and a flame, which is theoretically known to be coupled. However, the influence of concentrated solar radiation (CSR) on the flame has not been experimentally investigated. Hence this thesis aims at filling this gap. High flux solar simulators, comprising an array of high-intensity-discharge lamps coupled with elliptical reflectors, have been widely employed to study concentrated solar thermal energy systems. The use of electrical solar simulators holds the advantage over natural solar radiation in providing repeatable performance without the variability of the solar resource. Reliable models which predict the heat flux generated by a solar simulator are desirable because they enable efficient and systematic optimization of the system to meet the required trade-off between cost and performance. To this end, a concentric multilayer model of the light source is developed in this study to accurately predict the spatial distribution of the heat flux at the focus using a commercial Monte Carlo ray-tracing code. These simulations were validated with measurements of both the radiant intensity of the light source and the distribution of the concentrated heat flux. Further to that, on the experimentally validated ray tracing model, the geometry and surface reflectance of the additional concentrators were also assessed of two high flux solar simulators: one employs a single lamp, the other uses a seven-lamp array. In addition, the time-resolved spectra of solar simulators employing a metal halide and a xenon arc lamp are also measured, which provides the first experimental results of this kind that acquired from the same spectrometer to allow for direct comparison. This thesis also reports the first set of measurements of the influence of concentrated solar radiation on the soot volume fraction and temperature in a laminar sooty flame. Detailed laser diagnostics was performed on a laminar sooty flame with and without the irradiance of CSR, because laser diagnostics are demonstrated to hold the advantages of being non-intrusive, lower interferences and of being applicable to environments with high flux radiation. The current measurement using laser induced incandescence shows that the soot volume within the laminar flame was increased by 55% by CSR. In addition, the measurement of temperature using two-line atomic fluorescence shows that the flame temperature was increased by around 8% under CSR. In addition to the detailed laser diagnostics, an assessment of the influence of soot volume fraction on the global performance of the flames was also performed through a systematic study of flames using fuels of different soot propensities, which is achieved by blending hydrogen into hydrocarbon fuels, with hydrogen volume fraction ranging from 0 to 100%. Results show that flames with higher soot volume fraction have higher radiant fraction and lower NOx emissions. The principle contribution of the thesis is that the first measurement of the influence of concentrated solar radiation on the soot volume fraction and temperature of a flame was performed, which pushed forward the existing understanding of the interaction between broadband solar radiation and combustion. Its second major contribution is establishing an experimentally validated ray-tracing model that accurately predicts the concentrated heat flux from the solar simulator, and on this model, new design and optimization of solar simulators were performed. While this ray-tracing model is developed for metal halide lamps, the methodology is applicable more generally to solar simulators employing other types of discharge arc lamps.