SEARCH

SEARCH BY CITATION

Keywords:

  • impurities;
  • feedstock;
  • polysilicon;
  • solar-grade silicon;
  • silicon;
  • solar cell;
  • efficiency;
  • degradation

ABSTRACT

For the first time, the sensitivity to impurities of the solar cell conversion efficiency is reported for a state-of-the-art (i.e., 18%) and advanced device architecture (i.e., 23%). The data are based on the experimental results obtained in the CrystalClear project for the state-of-the-art cell process and extrapolated to a device with excellent front and rear surface passivation. Both device structures are not assumed to work in low injection level as several studies assumed before, but real operating conditions are considered. This is a fundamental difference with the past and required for modeling future high efficiency devices. The impurity with highest impact is Ti, followed by Cu, Cr, Ni and Fe, which form together a group two order of magnitude less sensitive than the former. In high efficiency devices, a large reduction of the impurity impact is visible for impurities with large capture cross-section ratio like Fe, which reduces its relative difference in comparison with, for example, Cr, which has a small capture cross-section ratio. In general, advanced devices will be more sensitive to the impurity content than the state-of-the-art cell design. This effect is partly compensated by a reduction of the substrate thickness. The impurity sensitivity as function of the substrate thickness is reported. Copyright © 2012 John Wiley & Sons, Ltd.