Pouring the Remaining Melt as a Method to Reduce the Red-Zone in the Top Region of mc-Silicon Ingots

38th European Photovoltaic Solar Energy Conference and Exhibition; 334-338
Completing our investigations on different melt removal techniques presented at previous PVSEC we installed an appliance in our SCU-G1 furnace for pouring the impurity-enriched silicon melt at the top of the ingot at the end of crystallization. With this method any amount of silicon melt between 0 and 25% (compared to the original mass of the G1 ingot) can be removed completely within a few seconds. Thus, the top surface of the growing crystal can be freed from the impurity enriched melt immediately and completely before the end of the regular crystallization process. This can be detected by optical investigation of the ingot’s top surface. However, in all of our experiments a top red-zone of at least 1.7 cm thickness in the mc-Si ingot remained. Compared to reference ingots, where no melt was removed, no significant thickness reduction was achieved. A deeper analysis of the minority charge carrier lifetime by a diffusion model [9] revealed that, typically, the Fei content in the ingot’s top region was reduced by a factor of about 3 compared to reference ingots. Yet, this seems to be not enough to affect the thickness of the top red zone. We suppose additional contamination of the ingot via the gas phase to be a reason for this behavior.