Evidence for bandgap narrowing effects in silicon solar cell performance

Abstract

The effects of bandgap narrowing on silicon solar cell performance are demonstrated by showing that experimental values for open-circuit voltage (V oc ) and spectral quantum efficiency (QE) at 0.4 µm can both be calculated with accuracy only by including high-doping effects. Experimental performance data were collected from high-quality single crystal silicon solar cells of both p+-n-n+and n+-p-p+types. The cells exhibited maximum power conversion efficiencies of ∼18% (at 25 suns AMl). Reflection as a function of wavelength was measured on the cells so that an intrinsic value for QE was determined. In all cases the measured values of QE at 0.4 µm were above 0.75. Calculations consisted of solution to the current transport equations including effects of Auger plus deep trap recombination, Fermi statistics, bandgap narrowing, and the concentration and field dependence for the mobility. All parameters used in the caluclation were taken from the literature; consequently, the calculation is a first-principles type. The calculated results were within 5% of the measured parameters when the Lanyon-Tuft bandgap narrowing model was used. Experimental agreement for V oc can be achieved without the effects of bandgap narrowing by changing the minority carrier lifetime in the heavily doped regions. However, all attempts to use the same model for V oc and QE failed unless high-doping effects were included.