insignificant for the chosen parameters (Fig. 17b). The dependence of Q tot versus
V pre measured for a moderate pulse fluence of 0.7 μJ/cm
2 is shown in Fig. 17c. The
extracted free charge is independent of pre-bias (within 10%, see grey area in
Fig. 17c) for both as-cast and thermally annealed blends, implying field-independent
charge carrier generation. A very weak dependence of free carrier generation on the
electric field was also seen in TDCF experiments performed on chlorobenzene-cast
P3HT:PCBM, with and without thermal annealing [167, 173]. Further evidence for
field-independent free carrier formation in thermally annealed P3HT:PCBM came
from transient photoconductivity experiments [174, 175].
To summarize, field-independent generation is shown to be common to blends of
regioregular P3HT with PCBM. The importance of this finding becomes evident
when considering that the studied devices were processed under a variety of
conditions (different solvents, thermal annealing, solvent annealing, different
layer thicknesses), which resulted in different morphologies and, consequently, in
a wide range of PV performance parameters.
Before discussing possible morphological pictures to explain these findings, we
turn to the involvement of CT states in the exciton-to-polaron conversion in P3HT:
PCBM blends. The observed insensitivity of free carrier generation to the internal
electric field suggests that this process does not involve split-up of bound CT states.
It has been proposed that the dissociation of P3HT excitons at the heterojunction
generates “hot” CT states (Fig. 18a), which possess a sufficient amount of energy to
overcome the Coulomb barrier without the aid of an electric field [169]. Interestingly, “colder” CT states in P3HT:PCBM can be directly excited by using photon
energies of between 1.2 and 1.6 eV, which is below the energy of the vibronically
relaxed S 1 exciton of P3HT. Lee and coworkers measured the external quantum
efficiency and the absorption of annealed P3HT:PCBM over a wide range of photon
energies [163]. The main result of these experiments is displayed in Fig. 18b. The
EQE spectrum was fully reproduced by considering only the wavelength-dependent
absorption of the organic layer in the device while keeping the internal quantum
efficiency constant at around 80%. It was concluded that efficient free charge
generation in P3HT:PCBM does not require the split-up of hot CT states, which
is in contrast to the hot CT model outlined above. In accordance with the interpretation by Lee and coworkers, it was reported that the incident photon energy has no
effect on the charge generation and recombination in either as-prepared or thermally annealed P3HT:PCBM blends [162]. These authors also showed that the
shape of the EQE does not change with bias. Because free carrier generation is
known to be independent of bias upon excitonic excitation, the dissociation of CT
states must also be unaffected by the internal electric field.
An important observation by Lee et al. is that free charge generation is efficient,
even when directly exciting the sub-bandgap CT state. Therefore, a driving force
must exist that counterbalances the mutual Coulomb attraction of these geminate
polaron pairs. One possible cause of this force is the energy landscape in these
blends, arising from its particular three-phase morphology. As pointed out above,
these blends consist of three phases: intermixed regions of P3HT and PCBM,
domains of crystallized P3HT chains, and almost pure agglomerates of PCBM
P3HT-Based Solar Cells: Structural Properties and Photovoltaic Performance
213
V pre measured for a moderate pulse fluence of 0.7 μJ/cm
2 is shown in Fig. 17c. The
extracted free charge is independent of pre-bias (within 10%, see grey area in
Fig. 17c) for both as-cast and thermally annealed blends, implying field-independent
charge carrier generation. A very weak dependence of free carrier generation on the
electric field was also seen in TDCF experiments performed on chlorobenzene-cast
P3HT:PCBM, with and without thermal annealing [167, 173]. Further evidence for
field-independent free carrier formation in thermally annealed P3HT:PCBM came
from transient photoconductivity experiments [174, 175].
To summarize, field-independent generation is shown to be common to blends of
regioregular P3HT with PCBM. The importance of this finding becomes evident
when considering that the studied devices were processed under a variety of
conditions (different solvents, thermal annealing, solvent annealing, different
layer thicknesses), which resulted in different morphologies and, consequently, in
a wide range of PV performance parameters.
Before discussing possible morphological pictures to explain these findings, we
turn to the involvement of CT states in the exciton-to-polaron conversion in P3HT:
PCBM blends. The observed insensitivity of free carrier generation to the internal
electric field suggests that this process does not involve split-up of bound CT states.
It has been proposed that the dissociation of P3HT excitons at the heterojunction
generates “hot” CT states (Fig. 18a), which possess a sufficient amount of energy to
overcome the Coulomb barrier without the aid of an electric field [169]. Interestingly, “colder” CT states in P3HT:PCBM can be directly excited by using photon
energies of between 1.2 and 1.6 eV, which is below the energy of the vibronically
relaxed S 1 exciton of P3HT. Lee and coworkers measured the external quantum
efficiency and the absorption of annealed P3HT:PCBM over a wide range of photon
energies [163]. The main result of these experiments is displayed in Fig. 18b. The
EQE spectrum was fully reproduced by considering only the wavelength-dependent
absorption of the organic layer in the device while keeping the internal quantum
efficiency constant at around 80%. It was concluded that efficient free charge
generation in P3HT:PCBM does not require the split-up of hot CT states, which
is in contrast to the hot CT model outlined above. In accordance with the interpretation by Lee and coworkers, it was reported that the incident photon energy has no
effect on the charge generation and recombination in either as-prepared or thermally annealed P3HT:PCBM blends [162]. These authors also showed that the
shape of the EQE does not change with bias. Because free carrier generation is
known to be independent of bias upon excitonic excitation, the dissociation of CT
states must also be unaffected by the internal electric field.
An important observation by Lee et al. is that free charge generation is efficient,
even when directly exciting the sub-bandgap CT state. Therefore, a driving force
must exist that counterbalances the mutual Coulomb attraction of these geminate
polaron pairs. One possible cause of this force is the energy landscape in these
blends, arising from its particular three-phase morphology. As pointed out above,
these blends consist of three phases: intermixed regions of P3HT and PCBM,
domains of crystallized P3HT chains, and almost pure agglomerates of PCBM
P3HT-Based Solar Cells: Structural Properties and Photovoltaic Performance
213