Exciton and Charge Dynamics in Polymer Solar Cells Studied by Transient Absorption Spectroscopy - Organic Solar Cells: Materials and Device Physics - page 130

still two orders of magnitude higher than the time-independent trap-free bimo-
lecular recombination rate c & 10 -12 cm 3 s -1 . As reported previously, the
reduced bimolecular recombination rate is partly ascribed to phase-separated bi-
continuous networks of RR-P3HT and PCBM domains, which are beneficial for
reducing bimolecular recombination loss [ 67 , 74 ]. It is also possible that the
recombination is not diffusion limited but depends on the electron transfer rate at
the interface. More importantly, the lifetime of trap-free carriers is estimated to be
s = (c n 0 ) -1 & 10 ls under the 1 sun condition, which is longer than a charge
collection time (*2 ls) to extract *50 % charges under the 1 sun open-circuit
conditions [ 25 ]. This finding suggests that the majority of trap-free charge carriers
could reach the electrode before the bimolecular recombination even under near
open-circuit condition. Under the short-circuit condition, the recombination loss
has been reported to be negligible, because the short-circuit current increases
linearly with the illumination intensity. We therefore conclude that trap-free
polarons play a major role in the charge transport, resulting in the recombination-
lossless performance in RR-P3HT:PCBM solar cells under not only the short
circuit but also near-open-circuit condition. This is consistent with the relatively
high fill factors (0.6-0.7) and EQEs ([80 %) reported for this device in com-
parison with other combination devices [ 5 - 9 ].
5.6.4 Relevance to Device Performance
As summarized in Table 5.1 , we can evaluate all the efficiency of photovoltaic
events in polymer solar cells. As described above, we can evaluate the efficiency of
g ED , g CT , and g CD from the transient absorption study. The remaining charge
collection efficiency g CC can be estimated from the internal quantum efficiency
(IQE) reported in previous studies [ 57 , 78 - 80 ]. Note that the efficiency is not
absolute one but should depend on the film morphology. Indeed, the difference in
the IQE is due to the different film morphology depending on the preparation
conditions. Nonetheless, the efficiency listed in the table demonstrates which loss
process is dominant in the device performance qualitatively. For RR-P3HT:PCBM
solar cells, all the efficiency is more than 90 % after the thermal annealing. For
RRa-P3HT:PCBM solar cells, on the other hand, the low efficiency in g CD and g CC
is a major cause of the poor device performance. More specifically, the exciton
diffusion g ED is *100 % for RRa-P3HT:PCBM blend films, 93 % for
RR-P3HT:PCBM blend films before the thermal annealing, and 89 % after the
thermal annealing. In terms of the exciton collection to the interface, therefore,
homogeneously mixed blend structures of RRa-P3HT:PCBM films are more
desirable than phase-separated blend structures of RR-P3HT:PCBM films. This is
consistent with the PL quenching results, indicating that there is still room to
further improve the exciton diffusion efficiency in RR-P3HT:PCBM [ 81 ]. Indeed,
such unquenched P3HT excitons can be effectively collected to the interface
through the long-range energy transfer by loading appropriate dye molecules into