Solar Spectral Splitting for Parallel Multijunction Photovoltaics

Abstract

Solar power presents itself as one of the strongest candidates available for renewable energy power generation. Photovoltaic panels in particular have high efficiencies and low manufacturing costs when compared to its peers; it falls short however, when compared to today's commonly used fossil fuels in both fronts, making the transition to clean, green energy a very rocky one. This thesis addresses some novel methods to improve the efficiency of commercially available photovoltaic panels, with a particular focus on spectral splitting. Spectral Splitting is a term used throughout this study, which refers to splitting the light into its respective spectra via the use of a dispersive optical element (eg prism), or spectral filters to disperse – or ‘split' – the incident light into the wavelengths comprising it on a target receiver, made up of photovoltaic cells. The choice of cells and their arrangement depends on material parameters such as the bandgap (Eg), and external quantum efficiency (EQE) to maximize the power output from the system. Simulations depicting the performance of the chosen solar cells under different spectral conditions were carried out, and their results posted to further augment the validity of the spectral splitting approach. Concentration is also a parameter of interest in our analysis, as it has been established that an increase in the light incident on a cell increases its exciton generation, which in turn improves its efficiency. Two main concentrator setups are considered, the first is a double component concentrator/splitter element based on Masdar City's 100 kWth Beam Down Facility, where the concentration and splitting of the incident light is carried out in two discrete steps. The second is a single component concentrator/splitter element, capable of splitting and concentrating the light propagating through it to producing a continuous band of wavelengths on the receiver. A comparison of the two aforementioned ideas is carried out, with particular focus on the difference in optical losses between both. Finally, a brief section on minor improvements that can be used to enhance the efficiency of currently established concentrated photovoltaic (CPV) technology is discussed with some preliminary results showing the advantages of such novel methodologies.

Date of Award	2012
Original language	American English
Supervisor	Marco Stefancich (Supervisor)