Lifetime spectroscopy with high spatial resolution based on temperature- and injection dependent photoluminescence imaging

2019 
Abstract In this paper we present a method for performing temperature- and injection dependent lifetime spectroscopy (TIDLS) with high spatial resolution based on steady state photoluminescence (PL) images taken at a range of excitation intensities and temperatures. The PL lifetime images are calibrated based on temperature dependent photoconductance measurements, thus requiring a minimum of assumptions regarding the temperature dependency of the luminescence signal and detection system. PL image acquisition at varying conditions and subsequent data analysis is automated, allowing for investigations of different samples without the need for excessive operator time. We demonstrate the method by presenting lifetime data and TIDLS analysis from commercial HPMC-Si samples, highlighting the local differences in the recombination properties for high lifetime in-grain areas and lower lifetime areas dominated by dislocation clusters. These regions have been investigated for two etched and passivated neighboring wafers, one in the as-cut state and one where the bulk lifetime had been improved through key solar cell process steps. For the as-cut wafer we find two dominant defects in the low lifetime area, one of which is identified as FeB pairs, with E t  = 0.90 ± 0.01 (above the valence band) and k  = 0.4 ± 0.1. In the in-grain region of this wafer we also observe two dominating defects, one with an energy level of either 0.28 ± 0.01 eV or 0.74 ± 0.01 eV above the valence band and a k value of 25 ± 1. The other defect is shallow, within ~0.27 eV of the band edge, and could not be further identified with the chosen temperature range. For the wafer that had passed through solar cell processing, we find a clear, overall improvement in the lifetime and a change in the type of dominating defects: We observe that a single, deep defect with 0.35  eV E t eV and a capture cross section ratio k  = 10 ± 1 can fully describe the observed lifetime data in the in-grain area after solar cell processing, whereas two defects are needed to describe the behavior of the lower lifetime regions. One of these defects is also a deep defect with 0.34  eV E t eV but with a higher k of 25 ± 2, and is thus similar to one of the defects observed in the in-grain region of the as-cut wafer. The other is a shallow defect with E t within 0.21  eV  of the band edge. Finally, we show how the method can be used to show interesting new properties of the lifetime maps, including maps of the temperature coefficient of the lifetime, maps of the injection dependence of the lifetime and maps at a constant injection level.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    32
    References
    3
    Citations
    NaN
    KQI
    []