Radiative Transfer and Spectrophotometric Characterization of Mineral Dust Optics on Photovoltaic Cells
Loading...
Authors
Piedra, Patricio G.
Issue Date
2017
Type
Dissertation
Language
Keywords
dust , photovoltaics , radiative transfer , soiled , solar , two stream
Alternative Title
Abstract
Efficiency of solar cells is degraded by deposition of mineral dust as well as other particles, and experiments reveal that losses can be significant (up to ~85%) depending on various factors. However, little is known about the role of light scattering and absorption in reducing optical transmission to the solar cell semiconductor. This dissertation first develops a fundamental model of optical losses dueto particle-on-substrate scattering for light propagating into the forward direction. We use discrete dipole approximation with surface interaction (DDA-SI), which is a numerical solution of light scattering for an arbitrarily shaped particle-on-substrate. Using DDA-SI, we studied transmission losses due to hemispheric backward scattering (HBS) and absorption. A parameter called the fraction of power lost, defined as the ratio of HBS efficiency plus absorption efficiency to extinction efficiency, was found appropriate to describe optical losses into the forward direction. We found that fine particles lead to higher losses (per optical depth or layer optical thickness) than coarser ones. Losses into the forward direction are maximized when the ratio of skin depth to particles diameter approaches unity.In addition, we conducted a resuspension-deposition experiment with two types of mineral dust, optically absorbing and non-absorbing dust. The dust samples were suspended and deposited onto glass slides, acting as surrogates for solar cells. Dust-deposited glass slides with increasing amounts of mass per area were spectroscopically characterized using a spectrophotometer with an integrating sphere (SIS) detector system. The SIS device allowed us to measure forward-hemisphere scattering, HBS, and direct beam transmission. Transmission into the forward direction was found to decrease as function of optical depth, depending on the absorptivityof the dust. Multiple-scattering radiative transfer theory, specifically the two-stream model as well as Monte Carlo stochastic calculations, were used to describe transmission as function of optical depth for both absorbing and nonabsorbingdust, yielding good agreement with experimental results within ~5%. Two-stream model and Monte Carlo techniques yield a multiple-scattering transmission calculation that depends on the single-scattering parameters of albedo andasymmetry parameter.This study has the potential to help with solar energy forecasting, aiding smart power grids in predicting and adapting to variations in solar cell energy output due to aerosol deposition. In addition, this study can help optimize cleaning procedures and schedules to save water in desert and semi-arid regions by describing transmission losses as function of dust type.