Photoluminescence Due To Oxygen Precipitates Distinguished from the D Lines in Annealed Si

About the D1 and D2 Dislocation Luminescence and Its Correlation with Oxygen Segregation

“…Our arguments about the oxygen involvement in the D1 and D2 luminescence are partially supported by a recent work of Tajima et al [2], who submitted a dislocated Cz wafer to an oxygen nucleation and precipitation stage (450 C, 64 h + 1000 C, 16 h) and showed that the very broad D1 band detected at 0.81 eV in the temperature range 11±125 K shifts to 0.77 eV, with a 100-fold decrease in intensity, at 280 K. They, therefore, suggest that this band (called D b band) is the fingerprint of oxygen precipitates at dislocations and it is not traceable to D1/D2 luminescence. Some evidence of oxygen related luminescence has been also shown by Clays et al [1] who observed that after an oxygen precipitation annealing, the D1 and D2 lines are clearly evident in the PL spectrum, whose intensity depends on the initial oxygen content.…”

“…Also the dislocation density (and the degree of deformation) strongly influences the intensity and the halfwidth of the different D bands in a very specific manner, with the D1 and D2 appearing only at high dislocation densities and with a broadening of D emission from sharp spectral lines to broad bands depending on the temperature and other unknown factors [1,2]. Limiting our concern to D1 and D2 emission, none of the models reported in literature fit entirely with their very complex behaviour, considering also that at least the D1 band is the convolution of several sub-bands, as it has been shown by Ossipyan and coworkers [3] and that a band at 0.846 eV (D5 band), intermediate between D1 and D2, can be observed in certain circumstances, as in Er-doped dislocated samples [4].…”

Defect Analysis in Solar Cell Silicon by Photoluminescence Spectroscopy and Topography

“…This finding clearly shows that the two emissions have the different origins. The author et al observed a similar broad band at 0.77 eV (D b band in their notation) in annealed Czochralski-grown Si crystals and concluded that the band is due to the oxygen precipitation (15). The D b band showed the same temperature dependence and the same intensity distribution around dislocations as the present 0.78 eV band.…”

Luminescence emission from forward- and reverse-biased multicrystalline silicon solar cells

“…38 Moreover, it was undoubtedly shown that the origin of the D1/D2 lines found at low temperatures and the defect band at around 0.8 eV detectable at room temperature are different. 38 Oxygen-related room temperature luminescence at around 1600 nm was also reported to appear after prolonged annealing at 450-470°C of Czochralski-grown silicon [39][40][41] and after high-temperature processes at around 1200°C followed by a slow cooling. 42 While the first temperature treatment is known to result in the formation of thermal donors, the latter one typically results in larger oxygen precipitates.…”