Comparison of PID Shunting in Polycrystalline and Single-Crystal Silicon Modules via Multi-Scale, Multi-Technique Characterization
Steven P Harvey, Dana Sulas, Peter Hacke, Steve Johnston, Mowafak Al-Jassim
National Renewable Energy Laborator, CO, CO, United States

We used the methods we reported last year to investigate potential-induced degradation (PID). We have now applied these methods to single-crystalline silicon modules that have degraded during field deployment, as well as in mini-modules stressed in the laboratory. We will compare these results to the polycrystalline results presented last year. Small cores have been removed from the modules and subjected to analysis. We use a combination of photoluminescence and dark lock-in thermography imaging, laser marking, electron-beam induced current measurements, and subsequent focused ion-beam marking to allow analysis of individual defects via time-of-flight secondary-ion mass spectrometry (TOF-SIMS) to investigate the root-cause mechanism for PID shunting. We see a direct correlation between recombination active shunts and sodium content. The sodium content in shunted areas peaks at the SiN/Si interface and is consistently observed at a concentration of 0.1%–1% in shunted areas. TOF-SIMS data taken on degraded and non-degraded single-crystalline sample areas show a similar trend as in the polycrystalline samples: more sodium is seen in the degraded areas.

Area: Sub-Area 9.2: Device Reliability and Accelerated Testing