Compensation engineering for silicon solar cells

This paper discusses the role of compensation engineering as a means to allow higher concentrations of dopants in silicon than would otherwise be acceptable for solar cell fabrication. Special attention is given to tri-doping, a technique consisting on the addition of gallium to boron and phosphorus doped UMG-Si (upgraded metallurgical grade silicon) feedstock to better control the net dopant density. Firstly, we review the current understanding of compensated silicon, focusing on the fundamental electronic properties of charge carriers: their density, mobility and lifetime. Based on those parameters, we then model solar cell efficiency in order to identify the advantages and limitations of compensation engineering. Given the current uncertainty of the majority and minority carrier mobilities, we study the possible impact of different levels of mobility reduction on solar cell efficiency. This modelling indicates that it is possible to achieve reasonable solar cell efficiencies, around 18%, even in cases of strong dopant compensation and mobility reduction. Lastly, the alternative of using n-type compensated silicon is briefly discussed, taking into account recent evidence that such material can degrade significantly upon illumination.