Analysis of the performance of multi-junction solar cells under realistic operating conditions

With this work important contributions on the area of the characterization and calibration of multi-junction solar cells under standard testing conditions and under concentration, as well as for a better understanding and for the modeling of concentrator systems, were made.In the frame of this work a flash based measurement set-up was built up. In this way, the calibration opportunities of the calibration laboratory at Fraunhofer ISE could be extended, by developing a routine for the reduction of measurement uncertainty for the calibration of multi-junction solar cells under concentration. The ISE CalLab is thus, besides NREL in the USA, the only institution being able to perform calibrated measurements of multi-junction cells under concentration. However, a prerequisite for the calibration under concentration is the calibration at standard testing conditions and the verification of the cell s linearity, for which a method based on the spectral response measurement, was proposed. With the help of this method, effects, indicting a non-linearity, could firstly be observed, however not finally be explained. In this context the work performed here corresponds to pioneer work and further works have to concentrate on this topic. Concerning the calibration under concentration, the effect of the use of a poorly matched spectrum, was quantified. The effect of reduction in fill factor under concentration with excess current of the top cell could be reproduced. For the first time in this work a possible explanation for this effect could be given, basing on a carrier transfer from the top to the middle subcell by radiative recombination.Another main topic of the work is the temperature dependence of the IV parameters of III-V solar cells. Starting from a theoretical consideration and based on the up to now most detailed measurement campaign worldwide, a set of temperature coefficients was generated, which is essential for the modeling of concentrator systems. The unexpected effect of a decrease in EQE of the top cell with increasing temperature could be explained by the increasing absorption in the window layer. All in all a contribution towards a better understanding of the temperature dependence of III-V solar cells could be made. Especially the data sets of the temperature coefficients of dual- and triple-junction solar cells under concentration are unique up to now.Besides the indoor calibration of concentrator solar cells also outdoor measurements of concentrator modules were one focus of this work. Thereto a set-up for the continuous measurement of concentrator modules has been designed. As a start, the irradiation resource for concentrator modules has been discussed in detail. It turned out, that only for very sunny locations a concentrator module will have a higher irradiation resource available in comparison to a fixed mounted flat plate module. Nevertheless, analysis of long term measurement data showed, that even for Freiburg (which is not a preferable location for concentrator PV) a FLATCON module of the first generation delivers a yearly sum of energy, which is comparable to standard flat plate modules. For sunny locations and application of FLATCON modules with higher efficiency that are available now, this ratio will be shifted in favor of the concentrator modules.Starting from the analysis of the measurements of a FLATCON module on one certain day, the difficulty of the dependence of the module parameters on the incident spectrum was illustrated - a simple correlation of module performance with meteorological data is not possible. With the aid of measured direct sun spectra it could be shown that if such spectral data is available, a correlation of the module s short circuit current with the solar spectrum can be established. Topic of future works will be amongst other things the definition of a simple measurement method in order to assess the parameter "spectrum".Finally a method has been presented, with which the heating of the solar cells in a concentrator module can be determined, based on indoor measurements of the solar cells. It could be shown that although only passive cooling is used, the temperature of the solar cells does not necessarily have to be drastically above the temperature of flat plate modules. Using the data on temperature dependence of multi-junction cells, it could be shown, that the temperature related losses in a FLATCON module is smaller by a factor of two compared to standard flat plate modules