Low temperature sensitivity of implied voltages from luminescence measured on crystalline silicon solar cells

Performance monitoring of crystalline silicon solar cells often requires terminal voltage measurements, which are strongly influenced by the sample temperature via the large temperature dependence of the intrinsic carrier density. The impact of sample temperature variations can be corrected for by using the temperature coefficient of the terminal voltage, however this relies on having both accurate values for the temperature coefficient and accurate measurements of the sample temperature. This paper demonstrates that in situations where the sample temperature cannot be accurately measured, for example in some high volume production facilities or during module degradation experiments, implied voltages determined from either electroluminescence or photoluminescence provide a more accurate measure of sample performance than the terminal voltage. The results presented here show that implied voltages exhibit a temperature sensitivity that is one order of magnitude lower than that of the terminal voltage. This is largely due to the fact that luminescence intensity is not strongly temperature dependent around room temperature. This is confirmed by experimental temperature dependent measurements on four different crystalline silicon solar cell types. The benefit of using implied voltage measurements over temperature corrected terminal voltage measurements for the monitoring of light and elevated temperature induced degradation in silicon solar modules is demonstrated.