Designing spectrally-selective mirrors for use in luminescent solar concentrators

Spectrally-selective mirrors improve the performance of luminescent solar concentrators (LSCs) by trapping emitted light within the waveguide. However, this beneficial property comes with a spectral restriction on incident sunlight that enters the concentrator. Especially for luminophores with overlap between the absorption and emission bands, design of the spectrally-selective mirrors requires a tradeoff between transmission of incident sunlight and trapping of luminescent photons. In this paper, we explore how the design of a spectrally-selective top mirror changes for LSCs containing luminophores of varying loading fractions, quantum yield, and overlap between the absorption and emission spectra, as well as LSCs with different back reflectors and lateral sizes. Using CdSe/CdS core/shell nanocrystals as the luminophore, we find that specific conditions favor different mirror designs. Mirrors designed to trap luminescent light have higher predicted performance than mirrors designed for sunlight transmission when the luminophore quantum yield is greater than 0.85, the luminophore optical density is less than 1.4 at 450 nm, the lateral size of the concentrator is greater than 10 cm, or there is low overlap between the luminophore absorption and emission. Mirrors optimized for either transmission or luminescence trapping have comparable performance for quantum yields less than 0.85, and the other conditions favor mirrors optimized for light transmission. For a LSC with unity quantum yield, a lateral size of 1 m x 1 m, and a mirror designed to trap luminescent light, a concentration factor of 37x is possible, as compared to 10x for a LSC with an open top. This research indicates the importance of tailoring the design of the spectrally-selective top mirror to the properties of the luminophore and LSC.