Illumination-induced errors associated with suns-V(OC) measurements of silicon solar cells

Rev Sci Instrum. 2009 Mar;80(3):033106. doi: 10.1063/1.3095441.

Abstract

I(SC)-V(OC) curves measured by the suns-V(OC) method are widely used for solar cell characterization due to its being unaffected by series resistance effects. A common setup for this measurement system uses a xenon photoflash for illumination purposes, resulting in a fast acquisition of the suns-V(OC) measurement data during the decaying edge of one flash. However, the use of a xenon photoflash accompanies also several disadvantages. Measurement errors are expected from the imperfect illumination homogeneity on the measurement stage. Also the discrepancy of the flash spectrum compared to the standard AM 1.5G spectrum leads to spectral mismatch between the sample and monitor cells when their spectral response differs. In addition, the divergence of the flash light leads to different illumination densities on the sample and the monitor cell if the height of these two cells differs. In this article these photoflash-caused measurement errors are investigated in detail, analyzing the resulting deviation in illumination density. The error due to an inhomogeneous illumination is negligible under most circumstances, while the error due to a spectral mismatch has to be considered but can be reduced drastically if an additional short-pass filter is used. The measurement error due to different cell highs should be taken into account but can be accounted for using an analytical correction.