3 Percolation Toward Lateral Junctions
69
condition (70 V·cm
−1 ) and even smaller electric field near the open-circuit condition (<<70 V·cm
−1 ) (Fig. 3.19a). The photocurrent density was proportional to the
simulated solar light intensity between 25 and 500 mW·cm
−2 (Fig. 3.19b). The
photocurrent density per cross-sectional area of the multilayered film reached 0.4
A·cm
−2 ((Fig. 3.19b, right vertical axis). Thus, the photovoltaic operation of lateral
alternating multilayered junctions with a macroscopic lateral electrode distance of
0.14 mm was demonstrated.
A corresponding vertical cell with an ITO/MoO 3 /C8-BTBT(50 nm)/PTCDIC8(50 nm)/BCP/Ag structure was usually operated with a built-in electric field under
short-circuit conditions that are 10
4 times larger (10
5 V·cm
−1 ) when a cell thickness
of 100 nm and V OC of 1 V are considered.
By increasing the number of layers from 2 (black curve) to 4 (blue curve) and 10
(red curve), i.e., by decreasing the D/A interface distance from 50 to 25 and 10 nm, a
dramatic increase in the photocurrent was observed. Photogenerated excitons within
a diffusion length of the exciton from the D/A interface can reach the D/A interface
(active layer) (Fig. 3.19e, red region). However, excitons photogenerated at distances
greater than the exciton diffusion length cannot reach the D/A interface (dead layer)
(Fig. 3.19e, pink region). Gradual elimination of the dead layer by decreasing the
D/A interface distance (Fig. 3.19e, from left to right) increases the exciton-collection
efficiency (η EC ), thus dramatically increasing the photocurrent.
The action spectra of the internal quantum efficiency (IQE) of the short-circuit
photocurrent (J SC ) for these cells are shown in Fig. 3.19c. By increasing the number
of layers from 2 to 4 and 10, i.e., by decreasing the D/A interface distance, dramatic
increases in the IQE were observed. The exciton-collection efficiency (η EC ) could
be quantitatively evaluated by using the IQE. The photocurrent owing to irradiation
with visible wavelengths (400–600 nm) was generated by PTCDI-C8 because C8BTBT is transparent. The semilogarithmic plots of the IQE at various wavelengths
versus the PTCDI-C8 thickness are linear (Fig. 3.19d). The slope of the plot gives
an exciton diffusion length of 12 nm, which is the distance at which 50% of the
excitons can reach the D/A interface. By assuming that η EC is 100% at 0 nm, η EC for
five monolayers of PTCDI-C8 with a thickness of 10 nm was 75% (Fig. 3.19e, right
enlarged illustration). Thus, we concluded that the lateral alternating multilayered
junction could collect most of the photogenerated excitons at the D/A interfaces.
This analysis is identical to that for the vertical junctions mentioned in Sect. 3.3.
The clear saturation behavior of the photocurrent that is independent of the reverse
bias (Fig. 3.19a) strongly suggests that most of the photogenerated electrons and holes
are extracted laterally. Moreover, the proportional relationship between photocurrent
and light intensity (Fig. 3.19b) suggests that the carrier recombination during lateral
carrier transport was negligible. Thus, we concluded that the carrier-collection efficiency (η CC ) in the region of photocurrent saturation is unity. Therefore, the η CC of
93%, which is determined by the ratio of the J SC value to the photocurrent in the
saturated region [34], was obtained (Fig. 3.19a, red curve).
Figure 3.20 shows the photocurrent–voltage (J-V) characteristics of the lateral
alternating multilayered junction cell (red curve), vertical multilayered cell (blue
curve), and bulk heterojunction (blended) cell (black curve). The total thickness of
69
condition (70 V·cm
−1 ) and even smaller electric field near the open-circuit condition (<<70 V·cm
−1 ) (Fig. 3.19a). The photocurrent density was proportional to the
simulated solar light intensity between 25 and 500 mW·cm
−2 (Fig. 3.19b). The
photocurrent density per cross-sectional area of the multilayered film reached 0.4
A·cm
−2 ((Fig. 3.19b, right vertical axis). Thus, the photovoltaic operation of lateral
alternating multilayered junctions with a macroscopic lateral electrode distance of
0.14 mm was demonstrated.
A corresponding vertical cell with an ITO/MoO 3 /C8-BTBT(50 nm)/PTCDIC8(50 nm)/BCP/Ag structure was usually operated with a built-in electric field under
short-circuit conditions that are 10
4 times larger (10
5 V·cm
−1 ) when a cell thickness
of 100 nm and V OC of 1 V are considered.
By increasing the number of layers from 2 (black curve) to 4 (blue curve) and 10
(red curve), i.e., by decreasing the D/A interface distance from 50 to 25 and 10 nm, a
dramatic increase in the photocurrent was observed. Photogenerated excitons within
a diffusion length of the exciton from the D/A interface can reach the D/A interface
(active layer) (Fig. 3.19e, red region). However, excitons photogenerated at distances
greater than the exciton diffusion length cannot reach the D/A interface (dead layer)
(Fig. 3.19e, pink region). Gradual elimination of the dead layer by decreasing the
D/A interface distance (Fig. 3.19e, from left to right) increases the exciton-collection
efficiency (η EC ), thus dramatically increasing the photocurrent.
The action spectra of the internal quantum efficiency (IQE) of the short-circuit
photocurrent (J SC ) for these cells are shown in Fig. 3.19c. By increasing the number
of layers from 2 to 4 and 10, i.e., by decreasing the D/A interface distance, dramatic
increases in the IQE were observed. The exciton-collection efficiency (η EC ) could
be quantitatively evaluated by using the IQE. The photocurrent owing to irradiation
with visible wavelengths (400–600 nm) was generated by PTCDI-C8 because C8BTBT is transparent. The semilogarithmic plots of the IQE at various wavelengths
versus the PTCDI-C8 thickness are linear (Fig. 3.19d). The slope of the plot gives
an exciton diffusion length of 12 nm, which is the distance at which 50% of the
excitons can reach the D/A interface. By assuming that η EC is 100% at 0 nm, η EC for
five monolayers of PTCDI-C8 with a thickness of 10 nm was 75% (Fig. 3.19e, right
enlarged illustration). Thus, we concluded that the lateral alternating multilayered
junction could collect most of the photogenerated excitons at the D/A interfaces.
This analysis is identical to that for the vertical junctions mentioned in Sect. 3.3.
The clear saturation behavior of the photocurrent that is independent of the reverse
bias (Fig. 3.19a) strongly suggests that most of the photogenerated electrons and holes
are extracted laterally. Moreover, the proportional relationship between photocurrent
and light intensity (Fig. 3.19b) suggests that the carrier recombination during lateral
carrier transport was negligible. Thus, we concluded that the carrier-collection efficiency (η CC ) in the region of photocurrent saturation is unity. Therefore, the η CC of
93%, which is determined by the ratio of the J SC value to the photocurrent in the
saturated region [34], was obtained (Fig. 3.19a, red curve).
Figure 3.20 shows the photocurrent–voltage (J-V) characteristics of the lateral
alternating multilayered junction cell (red curve), vertical multilayered cell (blue
curve), and bulk heterojunction (blended) cell (black curve). The total thickness of