Correlation of Defect Luminescence and Recombination in Multicrystalline Silicon

Abstract

Defect related luminescence (DRL) of mc-Si wafers, including the four D line emissions D1-D4, is investigated by hyperspectral photoluminescence (PL) imaging. The background subtraction scheme for the hyperspectral imaging setup is improved in order to obtain enhanced possibilities for comparing the DRL of different samples. In combination with PL based techniques for lifetime and iron imaging, the improved hyperspectral imaging technique is used to compare DRL of n-type and p-type mc-Si material, and to study changes of the DRL spectrum along the height of a crystalline ingot. Further, the correlation between DRL and metallic impurities as well as changes of DRL due to solar cell processing steps are investigated. No differences in the D line emissions that with certainty can be attributed to differences in material type (n/p) are found. We suggest that the spectral shape rather is determined by the recombination mechanism through which the charge carrier lifetime of a sample mainly is limited. In regions with high concentration of iron and other contaminations, we observe reduced intensities of the D3 and D4. It is thus likely that precipitates of iron or other impurities partly supress the D4 and D3 emission intensities.

Limiting Defects in n‐Type Multicrystalline Silicon Solar Cells

“…In the following, we summarize the results on limiting defects in n-type mc silicon from Schön et al, [27] Morishige et al, [28] and Schindler et al [35,38,39] Together with material-related efficiency potential analyses with efficiency limiting bulk recombination analysis (ELBA) [40] and multidimensional cell simulation with Quokka3, [41] we show the main limits to overcome the further increased efficiencies of n-type mc-silicon solar cells. More detailed studies of structural defects, such as dislocations, are available for the case of p-type mc silicon (see the previous studies [42][43][44][45][46][47] ) but are not yet available for the case of n-type doping (although some observations are expected to be transferable), such that we restrict our conclusions, in this article, to observations at precipitates and the impact of an empirically gained area fraction of recombination-active structural crystal defects. We finally report on the susceptibility of these cells on degradation.…”