In this section we introduce a simple characterisation technique based on the absorption and emission spectra of fluorescent collectors used for a spectral-based analysis of the performance of the collectors using the two – flux model outlined in Section 9.2.3. The analysis is based on careful measurements of the edge fluorescence of the collector and comparing that with the ‘first’ generation fluorescence spectra obtained from samples with no re-absorption at low dye concentrations. The re-absorption loss that occurs from a partial overlap of the absorption and emission bands is evaluated by scaling the measured edge fluorescence to the first generation fluorescence.
The photon flux emitted from the edge of the collector is usually observed with a fibre optic or an integrating sphere. These measurements, however, imply photon collection from a limited angular range (restricted to the edge escape cone) and observation of a different photon flux than the one received by the solar cell with perfect optical coupling to the edge, as described by the Weber and Lambe theory16 (see Section 9.2.2).
The experimental spectral re-absorption probability of different dye concentration collector samples obtained in Section 9.2 are compared with Weber and Lambe’s theory and the modified Weber and Lambe’s theory in Figure 9.13(a), where the first generation fluorescence spectrum was normalised to the edge fluorescence of BASF Red 305 collector samples in the
wavelength range from 650 to 850 nm, where the absorbance of all the collector samples was negligible. The probability of re-absorption r(l) for each collector was estimated from the ratio of the two scaled spectra. Figure 9.13(b) shows that the experimental re-absorption probability fits well with the modified Weber and Lambe’s theory.
Through this analysis, it is possible to estimate the performance of any fluorescent collector based on knowledge of its absorption and emission spectra. The analysis is not restricted to single dye-doped collectors. This simple model estimates the re-absorption probability within the collector and can set an upper limit to the achievable collection efficiency QC assuming no other losses are present except re-absorption and escape cone losses. The effective absorption coefficient, introduced in ref. 28 and defined by:
where la and ls are the thicknesses of the absorbing layer and substrate glass, respectively, and a is the absorption coefficient of the absorbing layer, provides a convenient vehicle for comparison with uniform block collectors of the same absorbance.
