Theoretical comparison of conventional and multilayer thin silicon solar cells

Abstract

Thin solar cells based on low-quality silicon are assessed for a range of possible material parameter values and device structures. Device thickness is freely optimized for maximum efficiency for a range of doping densities and numbers of junctions, leading to results differing markedly from previous investigations. Modelling of conventional and multilayer structures in this paper indicates little difference in efficiency potential on low-lifetime (<50 ns) crystalline silicon layers. Moderate efficiencies (>15%) are possible given adequate light trapping. Conventional structures (single and double junction cells) are superior if excellent light trapping is assumed. Thicker multilayer structures are advantageous in the case of poor light trapping or surface passivation. In an optimized cell in low-quality silicon, increasing the number of junctions allows a high current to be maintained, but at the cost of a reduced voltage and fill factor caused by increased junction recombination. Formidable practical difficulties are likely to be encountered to realize the theoretical performances discussed.