Location and Properties of Carrier Traps in mc‐Si Solar Cells Subjected to Degradation at Elevated Temperatures

Abstract

Multi‐crystalline Si (mc‐Si) solar cells subjected to carrier‐induced efficiency degradation at elevated temperatures are studied using deep‐level transient spectroscopy (DLTS), capacitance‐voltage, and electroluminescence (EL) measurements. Commercially available passivated emitter and rear cell (PERC) mc‐Si solar cells are investigated after short annealing in the dark (at T = 200 °C for 20 min), after degradation by a constant forward current at 70 °C, and after regeneration annealing at 200 °C for 20 min. The degradation is detected in situ by EL imaging of the surface of the cell. Several n+p mesa‐diodes, selectively fabricated on the front surface of the solar cell, allow the characterization of locations with various levels of degradation and density of extended defects. The degree of the degradation correlates with the local active boron concentration. At all stages, traps associated with extended defects (i.e., dislocations, grain boundaries, precipitates, etc.) are detected by DLTS, and these traps are not affected by changes in the degree of degradation. Two minority‐carrier traps with energy level positions in the bandgap at Ec −0.19 eV and Ec −0.34 eV are detected only in the degraded solar cells in concentrations that correlate with the local degree of cell degradation.

Progress in the understanding of light‐ and elevated temperature‐induced degradation in silicon solar cells: A review

“…on mesa diodes fabricated from commercial PERC mc-Si solar cells, an almost linear correlation between doping concentration and relative degradation from electroluminescence (EL) imaging was found, as shown in Figure 5. 64 To our knowledge, this is the first time that such a clear correlation between doping and relative degradation has been reported in literature. Despite the fact this correlation is not conclusive on the constituents of the LeTID defect, it suggests dopants are at least interacting to alter the LeTID kinetics and deserves further analysis.…”

“…found that degradation rate variations may also arise from lateral doping differences in mc-Si wafers, which increase local currents as suggested by Mchedlidze et al 64 65 (see Figure 4). Even though no conclusive evidence was provided on Niewelt's study, it was argued that the relative changes in PL intensity with degradation (up to 10%) could not be solely accounted by the doping variation in FZ wafers (<5%).…”

“…Due to the metastable nature of Fe i in silicon and overlaps in activation energy extracted from the DLTS spectra, the authors could not, however, exclude the contributions of other transition metals. Similarly, also on p-type mc-Si PERC solar cells, Mchedlidze et al 64 obtained two minority carrier traps with energy levels located at E C -0.19 eV and E C -0.34 eV.…”

“…suggest significantly faster degradation rates on cells made from the bottom of the brick, which agrees with the observations in, 39 where the degradation extent was also evaluated. Many more studies on the dependence of LeTID kinetics on illumination, 14 dopant type 43,45,51 and doping concentration, 64 among others, have been published.…”

“…Mchedlidze et al 64 using DLTS have provided some insight into what the defect may be. Chunlan et al 24 reported two traps in degraded p-type mc-Si PERC cells with deep energy levels of E V + 0.485eV associated with an interstitial iron (Fe i ) complex and E V + 0.295 eV associated with iron-hydrogen (Fe-H) pairs.…”