Characterization and classification of irradiation
Light accounts for the main part of the energy input in photoelectrochemical measurements. It is common for photoelectrochemical experiments to use the global standard spectrum AM1.5G (IEC 60904-3) and to calibrate its intensity to 100 mW cm–2 with a reference solar cell. The quality of a solar simulator is defined in IEC 60904-9 by three parameters which are spectral match, spatial non-uniformity and temporal instability. Since the test setup contains a mirror which possibly affects the irradiation quality, the parameters were determined and compared with the characteristics of the solar simulator.
The spectral distribution was measured with a spectroradiometer on four points in the plane of the photoelectrode. The classification is based on the cumulative intensity in six spectral ranges between 300 and 1200 nm. The measurements found that the deviation is maximal in the predefined spectral range of 772–919 nm, where it is 14.8 %. This resulting classification of »A« is lower than the classification of the solar simulator (»A+«), which is probably due to a spectral dependence of reflection of the mirror.
The spatial non-uniformity of the light is especially important for the investigation of large photoelectrodes. It was be measured with an array of intensity-calibrated cSi solar cells. The classification is based on the ratio of the largest and smallest irradiance according to (I maxI min)/(I max+I min). Depending on the investigated area, classifications of »A«, »B« and »C« are obtained in the investigated setup (Table 1). The degrading uniformity on larger areas is a result of the limited mirror size.
Temporal instability of irradiance is defined by short term and the long instability, of which only the latter is relevant for PEC measurements. It can be determined from the light intensities before and after a PEC experiment by (I maxI min)/(I max+I min). A value of below 0.02 was found in the investigated setup, which allows a classification of »A«.
It has to be noted that the use of Si solar cells for the measurement light intensity, spatial non-uniformity and temporal instability is legitimate only for absorbers with a similar band gap.[11] Since many semiconductors for water splitting experiments have higher band gaps (e.g ., WO3, 2.8 eV) an uncertainty is induced which will be especially prominent when solar simulators with poor spectral fit are used.