Solar Simulator for Multijunction Solar Cells

For the measurement and calibration of multijunction solar cells, Optosolar offers solar simulators with multiple light sources to adjust the spectrum of the light. In a spectral metric, they can be used to analyse cell properties at multiple spectra (see: Publication List).

Depending on the required spectral range and illumination area, these light sources can be combinations of (multiple) Xenon lamps and Tungsten lamps.

For tandem solar cells, triple solar cells and multijunction solar cells, it is a common misconception that the solar simulator spectrum must be a very good representation of AM 1.5 or AM 0.

For series connected multijunction cells, at a given spectrum (e.g. AM 1.5 or AM 0 for space solar cells), one of the cells will limit the current in the series connection of cells. All solar simulators deviate from the true AM 1.5 spectrum, and they are only classified in wide spectral bands (100nm width). So even for class A or A+ solar simulators, it can easily happen that the bandgaps of the multijunction cells are such distributed that the 'wrong' cell will limit the current, if a solar simulator with fixed spectrum is used. In the 70s and 80s, when these measurements were first developed, intercomparisons showed deviations of cell efficiency in fixed spectrum solar simulators of more than 40%.

So, the right way for multijunction measurements is to have one light source per constituent solar cell, and then to adjust all light sources such that the carrier generation in each cell is identical to the carrier generation under the reference spectrum (be it AM 1.5 or AM 0). This may not be the best representation of the spectrum with respect to solar simulator classification - one rather needs to find the best spectrum approximating the sub solar cell carrier generation. This can be found in an iterative settings process, after measurement of the spectral response of the multijunction solar cell (for the measurement process: see Publication List).