Analysis of impact of non-uniformities on thin-film solar cells and modules with two-dimesional simulations.

Clean and environmentally friendly photovoltaic (PV) technologies are now generally recognized as an alternative solution to many global-scale problems such as energy demand, pollution, and environment safety. The cost ($/kWh) is the primary challenge for all PV technologies. In that respect, thin-film polycrystalline PV technology (CdTe, Cu(In,Ga)Se2, etc), due to its fast production line, large area panels and low material usage, is one of the most promising low-cost technologies. Due to their granular structure, thin-film solar cells are inherently non-uniform. Also, inevitable fluctuations during the multistep deposition process of large area thin-film solar panels and specific manufacturing procedures such as scribing result in non-uniformities. Furthermore, non-uniformities can occur, become more severe, or increase in size during the solar-panel's life cycle due to various environmental conditions (i.e. temperature variation, shading, hail impact, etc). Non-uniformities generally reduce the overall efficiency of solar cells and modules, and their effects therefore need to be well understood. This thesis focuses on the analysis of the effect of non-uniformities on small size solar cells and modules with the help of numerical simulations. Even though the 2-D model developed here can analyze the effect of non-uniformities of any nature, only two specific types of microscopic non-uniformities were addressed here: shunts and weak-diodes. One type of macroscopic non-uniformity, partial shading, was also addressed. The circuit model developed here is a network of diodes, current-sources, and transparent-conductive-oxide (TCO) resistors. An analytic relation between the TCO-resistor, which is the primary model parameter, and TCO sheet resistance rhoS, which is the corresponding physical parameter, was derived. Based on the model several useful general results regarding a uniform cell were deduced. In particular, a global parameter delta, which determines the performance of a uniform solar cell depending on sheet resistance rhoS, cell length L, and other basic parameters, was found. The expression for the lumped series resistance in terms of physical parameters was also derived. Primary power loss mechanisms in the uniform case and their dependence on rhoS, L, and light generated current JL were determined. Similarly, power losses in a small-area solar cell with either a shunt or a weak-microdiode were identified and their dependence on rho S, JL, and location of the non-uniformity with respect to the current collecting contact was studied. The impact of multiple identical non-uniformities (shunts or weak-diodes) on the performance of a module was analyzed and estimates of efficiency loss were presented. It was found that the efficiency of the module strongly depends not only on the severity and fractional area of non-uniformities but also on their distribution pattern. A numerical parameter characterizing distribution pattern of non-uniformities was introduced. The most and least favorable distribution patterns of shunts and weak-diodes over the module area were determined. Experimentally, non-uniformities may be detected with the help of spatially resolved measurements such as electroluminescence (EL). The 2-D circuit model was also used to develop the general framework to extract useful information from experimental EL data. In particular, a protocol that can help distinguish a shunt from a weak-diode and estimate the severity of the non-uniformity based on the EL data was developed. Parts of these simulation results were verified with experimental EL data obtained by other authors. The thesis also discusses the effect of...