Imaging of metastable defects in silicon

Photoluminescence Imaging has been proven to be very useful for the characterization of silicon material characterization. By detecting emitted photons from the band-to-band recombination with a CCD camera information about the excess carrier density is gained. Besides carrier lifetime spatially resolved concentration measurements of the most prominent metallic impurities, namely interstitial iron and chromium, can be obtained in boron doped silicon. These measurements are based on a careful preparation of different metastable chemical states of the species under test by appropriate temperature and illumination conditions. Interstitial chromium as well as iron may reversibly bind to boron which changes their electrical defect parameters and thus the injection dependent carrier lifetime. From PL-images in both states the specific quantitative point defect concentration can be extracted. An analysis of the depth dependent carrier profiles gives an estimation on systematic errors, ideal measurement conditions, and a correction procedure for non-ideal measurement conditions. This contribution summarizes the state-of-the-art of impurity imaging methods with PL and introduces new aspects on quantitative PL-analysis of silicon material. A possible impact of lateral carrier diffusion on measurement errors is discussed by simulations of the carrier density distribution around extended defects. We present spatially resolved PL-measurements of the BO-defect concentration which expand the analysis of specific impurities. Being the most prominent factor which limits the carrier lifetime in Czochralski silicon this defect proves to significantly reduce material quality in multicrystalline silicon, too. For the separation of the impact of different metastable species present in the same sample a sequence of conditioning steps and measurements is proposed. Combining all techniques on the detection of metastable defects the quantitative impact of the specific detected impurity concentrations on the electrical material quality is deduced and can be tracked during solar cell processing resulting in a comprehensive analysis of efficiency-limiting defects in silicon.