Customized Energy Down-Shift Using Iridium ... - ACS Publications

Oct 14, 2016 - Customized Energy Down-Shift Using Iridium. Complexes for Enhanced Performance of. Polymer Solar Cells. Hyun-Tak Kim,. †. Ji Hoon Seo...
0 downloads 0 Views 4MB Size
Download PDF
Customized Energy Down-Shift Using Iridium Complexes for Enhanced Performance of Polymer Solar Cells Hyun-Tak Kim,† Ji Hoon Seo,‡ Jeong Hyuk Ahn,† Myung-Jin Baek,‡ Han-Don Um,‡ Sojeong Lee,‡ Deok-Ho Roh,† Jun-Ho Yum,§ Tae Joo Shin,# Kwanyong Seo,*,‡ and Tae-Hyuk Kwon*,† †

Department of Chemistry, ‡Department of Energy Engineering, and #UNIST Central Research Facilities & School of Natural Science, Ulsan National Institute of Science and Technology (UNIST), Ulsan 44919, South Korea § Laboratory for Molecular Engineering of Optoelectronic Nanomaterials, École Polytechnique Fédérale de Lausanne (EPFL), Station 6, 1015 Lausanne, Switzerland S Supporting Information *

ABSTRACT: The spectral absorption range of polymer solar cells can be efficiently increased by molecular compatibility and energy level control in the energy transfer system. However, there has been limited research on energy transfer materials for both amorphous and highly crystalline polymer active materials. For the first time, we developed customized iridium (Ir(III)) complexes that are incorporated into the active materials, poly[[4,8-bis[(2-ethylhexyl)oxy]benzo[1,2-b:4,5-b′]dithiophene-2,6-diyl][3fluoro-2-[(2-ethylhexy)carbonyl]thieno[3,4-b]thiophenediyl]] (PTB7, amorphous) or poly(3-hexylthiophene) (P3HT, high crystalline) as energy donor additives. The Ir(III) complex with the 2-phenyl quinolone ligand increased the power conversion efficiency of the corresponding devices by approximately 20%. The enhancements are attributed to the improved molecular compatibility and energy level matching between the Ir(III) complex and the active material, long Förster resonance energy transfer radius, and high energy down-shift efficiency. Overall, we reveal Ir(III) complex additives for amorphous and highly crystalline polymer active materials. These additives would enable efficient energy transfer in polymer solar cells while retaining the desirable active layer morphology, thereby improving the light absorption and conversion. olymer solar cells (PSCs) are flexible, lightweight, and highly customizable, but their efficiency is still lower than that of silicon-based cells. A broad spectral absorption range is critical for high-efficiency PSCs that consist of an energy donor material and a fullerene acceptor.1−4 As single donor materials have limited generation of photocurrent and a narrow absorption region across the solar spectrum, strategies based on multiple donors5,6 and energy transfer systems7−9 are being explored. Most studies on this subject have focused on polymer−polymer-based systems. However, molecular compatibility in such systems remains difficult to control, given that the polymers can exhibit different crystallinities and energy levels. Therefore, it is challenging to ensure efficient charge or energy transfer. Moreover, nonradiative Förster resonance energy transfer (FRET) between polymers is relatively inefficient, owing to their small Stokes shift. In this study, for the first time, we developed customized energy transfer materials for both amorphous and highly crystalline polymer materials and then applied them to a triplet−singlet energy transfer system for enhanced perform-

P

© XXXX American Chemical Society

ance of PSCs. Compared to the conventional singlet−singlet energy transfer system, triplet systems have several advantages, such as a larger Stokes shift to prevent self-quenching, a longer diffusion length, and reduced charge recombination.10−12 With this aim in mind, we used iridium-based energy donors customized for two representative active polymer materials: the amorphous poly[[4,8-bis[(2-ethylhexyl)oxy]benzo[1,2-b:4,5b′]dithiophene-2,6-diyl][3-fluoro-2-[(2-ethylhexy)carbonyl]thieno[3,4-b]thiophenediyl]] (PTB7) and the high-crystallinity poly(3-hexylthiophene) (P3HT). Iridium (Ir(III)) complexes are used as the energy donors because they exhibit a number of additional advantages, such as high quantum yield (QY) and easy control of the energy levels and molecular compatibility. The long diffusion lengths of these complexes, a result of their long exciton lifetimes, can increase the energy transfer efficiency (ηET). For efficient energy transfer, the energy donor also needs to have (1) an absorption range different Received: October 10, 2016 Accepted: October 14, 2016

991

DOI: 10.1021/acsenergylett.6b00518 ACS Energy Lett. 2016, 1, 991−999

Letter

http://pubs.acs.org/journal/aelccp

Letter

ACS Energy Letters

Figure 1. Optical properties of Ir(III) complexes and schematic of energy transfer in the active layer. (a) Schematic of energy transfer from an Ir(III) complex to PTB7. (b) Chemical structures of Ir-Red, Ir-Orange, Ir-Green, Ir-Blue, PTB7, P3HT, and PC71BM. (c) Absorption (solid symbols; red square with line for Ir-Red, orange circle with line for Ir-Orange, green up-triangle with line for Ir-Green, and blue down-triangle with line for Ir-Blue) and PL (open symbols) spectra of the various Ir(III) complexes.

Table 1. Optical Properties, Electrochemical Properties, QY Values, FRET Radii, and Device Performances of the Ir(III) Complexes active layers PTB7:Ref. PTB7:Ir-Red PTB7:Ir-Green PTB7:Ir-Blue PTB7:Ir-Orange PTB7:Ir-Orange (inserted layer) P3HT/Ref. P3HT/Ir-Orange P3HT/Ir-Blue

UV (nm)

PL (nm)

HOMO (eV)

LUMO (eV)

band gap (eV)

QY

FRET radius (nm)

JSC (mA·cm−2)

Voc (V)

FF (%)

best PCE (%)

462 400 376 455

608 511 471 580

−5.2 −5.4 −5.6 −5.7 −5.5

−3.3 −3.2 −2.9 −3.0 −3.2

1.9 2.2 2.7 2.7 2.3

0.12 0.06 0.21 0.23

7.8 5.7 6.3 8.2

13.3 15.9 15.9 15.9 16.1 13.3

0.75 0.74 0.75 0.74 0.74 0.75

71.7 72.4 70.0 72.3 72.9 69.9

7.37 8.53 8.32 8.52 8.72 7.23

7.23 8.41 8.14 8.41 8.62 7.03

455 376

580 471

−4.8 −5.5 −5.7

−2.7 −3.2 −3.0

2.1 2.3 2.7

0.23 0.21

2.8 2.7

10.3 11.0 11.5

0.57 0.57 0.58

51.7 51.6 55.9

3.02 3.23 3.63

2.96 ± 0.06 3.08 ± 0.15 3.52 ± 0.11

average PCE (%) ± ± ± ± ± ±

0.14 0.12 0.18 0.11 0.10 0.20

dramatically, from 13.3 to 16.1 mA·cm−2, resulting in improvement in the power conversion efficiency (PCE) from 7.37 to 8.72%. Figure 1a shows the simple architecture and a plausible mechanism of the resulting active layer. In the device with P3HT and 0.5 wt % blue-emitting Ir(III) complex, JSC also increased significantly from 10.3 to 11.5 mA·cm−2, corresponding to an increase in PCE from 3.02 to 3.63%. The route employed to synthesize the four Ir(III) complexes used in this study is depicted in Figure S1, and the experimental details are given in Supporting Information (with the spectral characterization of Ir-Red shown in Figures S2−S4). To control the absorption and emission characteristics of the Ir(III) complexes, four different hydrophobic primary ligands (Figure 1b) were used: 1-phenylisoquinoline (1pq), 2-phenylquinoline (2pq), phenylpyridine (ppy), and difluorophenylpyridine (F2ppy). These ligands have been shown to create high-QY

from that of the energy acceptor, (2) efficient spectral overlapping between absorption of the energy acceptor and emission of the energy donor, and (3) a wider energy level than that of the energy acceptor in order to prevent charge transfer.13 Customizable Ir(III) complexes perfectly satisfy these prerequisites of an energy donor.9 Four different iridium-based energy donors were developed for the two aforementioned active materials with different maximum metal-to-ligand charge transfer (MLCT) absorption ranges (350−500 nm). The low absorptivity of PTB7 was compensated for by changing the main ligand. The resulting PTB7:[6,6]-phenyl C71 butyric acid methyl ester (PC71BM) devices incorporating 10 wt % iridium-based energy donors demonstrated high performances, with the best device using the orange-emitting Ir(III) complex. The short-circuit current density (JSC) of the device with PTB7 was increased 992

DOI: 10.1021/acsenergylett.6b00518 ACS Energy Lett. 2016, 1, 991−999

Letter

ACS Energy Letters materials when incorporated with iridium.14−17 They also exhibit emissions of different colors, which range from blue, green, and orange to red as the conjugation length is increased. The hydrophilic 3-hydroxypicolinic acid (pic-OH) was introduced as an ancillary ligand for controlling the molecular compatibility.9 The material (ppy)2Irpic (Ir-Green2) with picolinic acid was also used to study the effect of the hydroxyl group of pic-OH. We observed improvement in the active layer morphology due to the amphiphilic nature of the Ir(III) complexes, using grazing incidence-wide-angle X-ray diffraction (GI-WAXD) analysis and atomic force microscopy (AFM) (see the Supporting Information). This phenomenon is discussed further in the morphological study. Table 1 and Figure 1c show the absorption and emission properties of the developed Ir(III) complexes. The absorption spectra of the Ir(III) complexes in the film state indicate a π−π* ligand-centered transition at wavelengths lower than 350 nm. Ir-Blue, Ir-Green, Ir-Orange, and Ir-Red exhibited MLCT at 376, 400, 455, and 462 nm, and the corresponding maximum emission peaks (λmax) appeared at 471, 511, 580, and 608 nm, respectively. All of the Ir(III) complexes exhibited large Stokes shifts at wavelengths higher than 100 nm, which prevented selfquenching.18 Figure S5 depicts the overlap between the emission spectra of the iridium-based energy donors and the absorption spectrum of PTB7. The degree of spectral overlap, one of the most important factors for efficient energy transfer, follows the order of Ir-Red > Ir-Orange > Ir-Green > Ir-Blue. The QY values of the Ir(III) complexes measured in a CH2Cl2 solution are ordered as follows: Ir-Orange (0.23) > Ir-Blue (0.21) > Ir-Red (0.12) > Ir-Green (0.06) (Table 1). The QY value also affects energy transfer, and the obtained values are discussed below based on a device study. The energy levels of the highest occupied molecular orbitals (HOMO, −5.44 to −5.73 eV) and lowest unoccupied molecular orbitals (LUMO, −3.19 to −2.94 eV) of the Ir(III) complexes included the energy levels of PTB7 (HOMO = −5.15 eV, LUMO = −3.31 eV) and P3HT (HOMO = −5.20 eV, LUMO = −3.21 eV), as shown in Table 1. These results indicate that efficient energy transfer occurred from the energy donor to the acceptor. To confirm the transfer of energy from the iridium-based energy donors to the acceptor, both steady-state and transient photoluminescence (PL) spectra were analyzed using timecorrelated single-photon counting. The PL intensity of PTB7 with 10 wt % Ir-Orange in the film state was twice that of intrinsic PTB7, while the emission of Ir-Orange almost disappeared when excited in the MLCT region (455 nm, Figure 2a). These results could be attributed to the triplet− singlet energy transfer, not the charge transfer. In addition, reduction in the PL intensity of Ir-Orange was directly related to ηET. By integrating the emission spectra of the Ir(III) complexes, their ηET values were calculated to be 98%. The energy transfer efficiency measured from the steady-state PL method is (1 − AD−A/AD) × 100%, where AD−A and AD are the total integral areas of the emission intensity of the Ir complex with and without PTB7, respectively.19 The value of ηET can be determined more accurately by measuring the exciton lifetimes of the Ir(III) complexes, using transient PL spectroscopy with and without PTB7. The ηET value was calculated using the following formula9 ηET = 1 −

Figure 2. Energy transfer analysis with PL spectroscopy. (a) PL spectra of Ir-Orange with (open black circle with line) and without (solid orange square with line) PTB7 and only PTB7 (solid black circle) the in film state. The magnified spectrum at 700−850 nm is shown in the inset. The intensity for the Ir-Orange film decreases when PTB7 is added, while that for PTB7 increases when IrOrange is added. (b) Transient PL spectra of Ir-Orange with (orange line) and without (black line) PTB7 in the film state. Exciton lifetimes of Ir-Orange with and without PTB7 are calculated from the transient PL spectra. The exciton lifetime of the Ir-Orange layer (1236 ns) is much longer than that of the IrOrange and PTB7 blended layer (14 ns).

where τA,D and τD are the exciton lifetimes of the energy donor Ir(III) complexes with and without the energy acceptor (PTB7), respectively (Figure S7). The obtained results are shown in Table S1. The pristine lifetimes of the Ir(III) complexes were as follows: 859 ns for Ir-Red, 1236 ns for IrOrange, 1466 ns for Ir-Green, and 1929 ns for Ir-Blue. The longer exciton lifetimes compared to those of the polymers helped increase the FRET radius, resulting in an increased ηET.20 When the Ir(III) complexes were used with PTB7, their exciton lifetimes excited in the MLCT region decreased dramatically to 9 ns for Ir-Red, 14 ns for Ir-Orange, 77 ns for Ir-Green, and 90 ns for Ir-Blue (Figures 2b and S6). These correspond to the ηET values of 99, 99, 95, and 95%, respectively. A sharp decrease in the lifetime of the energy donor is characteristic of energy transfer,19 and all of the iridium-based energy donors studied here exhibited high ηET values (>95%). This was in keeping with the values obtained using steady-state PL spectroscopy. Inverted PSCs were fabricated with the structure of glass/ indium tin oxide (ITO)/zinc oxide (ZnO)/active layer/ molybdenum trioxide (MoO3)/Ag, as shown in Figure 3a,b. Table 1 shows the photovoltaic characteristics of these devices fabricated with and without the Ir(III) complexes, measured under AM1.5G illumination. Incorporating the four iridiumbased energy donors at the optimized ratio of 10 wt % (see

τA,D τD

(1) 993

DOI: 10.1021/acsenergylett.6b00518 ACS Energy Lett. 2016, 1, 991−999

Letter

ACS Energy Letters

Figure 3. Device performance using Ir(III) complexes. (a) PSC with the glass/ITO/ZnO/active layer/MoO3/Ag structure, with the Ir(III) complexes added into the active layer. (b) Energy diagram of PSC with Ir(III) complexes. The energy band gap with the Ir(III) complex is wider than that of the polymers. Thus, the absorbed light energy can be transferred to the polymers. (c) J−V curves of the PTB7-based PSC device with (orange circle with line) and without (black square with line) 10 wt % Ir-Orange under 1000 W·m−2 irradiation. (d) The EQE spectra and (e) IQE of the PSC devices with (orange circle with line) and without (black square with line) 10 wt % Ir-Orange. The blue area is the region where the energy transfer effect dominates the IQE enhancement, and the green area is the region where the morphological effect dominates.

Table S2) in the PTB7:PC71BM film led to significant enhancements in the JSC and thereby the PCE. In the device with 10 wt % Ir-Orange with respect to PTB7, JSC = 16.1 mA· cm−2, the open-circuit voltage was (VOC) = 0.74 V, the fill factor was (FF) = 72.9%, and PCE = 8.72%. The JSC and PCE values from the current density−voltage (J−V) curves were 21 and 18% higher than those of the reference device, respectively (Figure 3c and Table 1). The efficient energy transfer increased JSC, as confirmed by the external quantum efficiency (EQE) spectra (Figure 3d). Interestingly, within the broad range from 380 to 720 nm, the EQE of the PSC with the Ir-Orange complex was notably improved, even at the long wavelength where the absorption of Ir-Orange is extremely weak. To clarify the effect of Ir-Orange on energy transfer, Figure 3e compares the internal quantum efficiency (IQE) spectra of the devices with and without the Ir-Orange complex. Specifically, IQE was notably improved between 380 and 510 nm, which is the absorption range of the Ir-Orange complex. This directly proves that efficient energy transfer occurred between the iridium-

based energy donor and the energy acceptor PTB7. The IQE spectrum was also enhanced by Ir-Orange for wavelengths greater than 510 nm, even though the reflectance spectra of devices with and without the Ir-Orange complex have similar profiles in the long-wavelength region (Figure S8). This might be related to the improved active layer morphology. The enhanced JSC values can be estimated from the IQE spectra by using the following equation JSC =

q hc

∫

SsQ x(1 − R )λ dλ

(2)

where q, h, c, Ss, Qx, R, and λ represent the electron charge, Planck’s constant, speed of light, AM1.5G solar spectrum, IQE spectrum, reflectance, and light wavelength, respectively. A significant JSC gain was obtained by employing the Ir-Orange complex. The value of JSC was 1.22 mA·cm−2 at short wavelength (380−510 nm) as energy transfer of the Ir-Orange complex would dominate the enhancement of JSC due its significant light absorption. In addition, a JSC gain of 1.50 mA· 994

DOI: 10.1021/acsenergylett.6b00518 ACS Energy Lett. 2016, 1, 991−999

Letter

ACS Energy Letters

Figure 4. GI-WAXD patterns and line-cut profiles of the blended films. 2D GI-WAXD patterns of (a) PTB7, (b) PTB7:Ir-Red (10:1), (c) PTB7:Ir-Orange (10:1), (d) PTB7:Ir-Green (10:1), and (e) PTB7:Ir-Blue (10:1). (f) Chemical structures of the incorporated fluorine substituents of Ir-Blue and the thiophene spacers of PTB7. (g) GI-WAXD line-cut profiles of the PTB7:Ir(III) complex system, in-plane. Black square with line: PTB7. Red circle, orange up-triangle, green down-triangle, and blue diamond with lines are for Ir-Red, Ir-Orange, Ir-Green, and Ir-Blue with PTB7, respectively. (h) GI-WAXD line-cut profiles of the PTB7:Ir(III) complex system, out-of-plane.

cm−2 was also achieved at long wavelength (510−740 nm) because the morphological improvement would be more dominant compared to energy transfer of the Ir complex. The sum of the two JSC gains (2.72 mA·cm−2) was well matched with the JSC improvement measured by the solar simulator (2.80 mA·cm−2). The morphological effect will be discussed later. When other iridium-based energy donors were incorporated in the PTB7:PC71BM film, the PCEs of the resulting PSCs were 8.53% (Ir-Red), 8.52% (Ir-Blue), and 8.32% (Ir-Green), all of which were still higher than that of the control device (Figure S9a). The corresponding EQEs matched well with the increase in JSC and were higher than that of the control device (Figure S9b). The increase in the PCE values could be arranged in the following order: Ir-Orange > Ir-Red > Ir-Blue > Ir-Green. The fact that devices based on Ir-Orange exhibited the highest PCE can be explained in terms of the ηET value and the FRET radius. First, the ηET value corresponding to energy transfer was the highest in the case of Ir-Orange (99%

as calculated earlier). Second, Ir-Orange as an energy donor exhibited the largest FRET radius R0 (8.2 nm, Table 1), which was calculated using the following equation13 R0 =

6

9000κ 2ηD ln 10 5 4

128π n NA

∫0

∞

fd (λ)ϵA (λ)λ 4 dλ

(3)

where κ2 is the dipole orientation factor, ηD is the QY value, n is the refractive index of the medium, NA is Avogadro’s number, f D(λ) is the normalized energy donor emission spectrum, εA(λ) is the molar extinction coefficient of the energy acceptor, and the integral represents the spectral overlap. Equation 3 suggests that the most important factors determining the FRET radius are QY and the degree of spectral overlap. From Table 1, the FRET radii of the iridium-based energy donors are 8.2 nm for Ir-Orange, 7.8 nm for Ir-Red, 6.3 nm for Ir-Blue, and 5.7 nm for Ir-Green. These values correlate very well with the observed increases in PCE (Figure S10). Although the FRET radius of Ir995

DOI: 10.1021/acsenergylett.6b00518 ACS Energy Lett. 2016, 1, 991−999

Letter

ACS Energy Letters Table 2. 2D GI-WAXD Parameters of the PTB7:Ir(III) Complex/PC71BM Blended Systems PTB7:Ir(III) complex:PC71BM ratio 1:0:0 1:Ir-Red 10 wt %:0 1:Ir-Orange 10 wt %:0 1:Ir-Green 10 wt %:0 1:Ir-Blue 10 wt %:0 1:0:1.5 1:Ir-Red 10 wt %:1.5 1:Ir-Orange 10 wt %:1.5 1:Ir-Green 10 wt %:1.5 1:Ir-Blue 10 wt %:1.5

qxy (Å−1) 0.346 0.342 0.343 0.344 0.345 0.356 0.353 0.351 0.355 0.352

± ± ± ± ± ± ± ± ± ±

0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001

Δqxy(Å−1) 0.128 0.119 0.110 0.122 0.114 0.095 0.089 0.089 0.097 0.088

± ± ± ± ± ± ± ± ± ±

0.001 0.001 0.002 0.001 0.001 0.001 0.001 0.001 0.002 0.001

LC(Å) 46.1 49.6 53.6 48.4 51.8 62.2 66.4 66.4 60.9 67.1

± ± ± ± ± ± ± ± ± ±

0.4 0.4 0.7 0.4 0.3 0.4 0.4 0.4 0.7 0.3

qz(Å−1)

1.37 1.37 1.36 1.35 1.38

± ± ± ± ±

0.02 0.01 0.01 0.01 0.02

Δqz(Å−1)

0.353 0.345 0.322 0.303 0.314

± ± ± ± ±

0.003 0.001 0.004 0.004 0.001

LC(Å)

16.7 17.1 18.3 19.5 18.8

± ± ± ± ±

0.2 0.1 0.2 0.2 0.1

values of the peaks of the active layers with the Ir(III) complexes indicate that incorporation of the Ir(III) complexes into the PTB7:PC71BM blended films increased the size of the formed nanocrystallites, compared to those in the pristine film (Table 2).6,24 In the case of Ir-Orange, LC(100) increased from 46.1 (PTB7 film) to 51.8 Å (PTB7:Ir-Orange film). It is intriguing that when PTB7 and the Ir(III) complexes were blended together the amphiphilic nature of the complexes enhanced the crystallinities of PTB7 and PC71BM, as evidenced by the changes in the GI-WAXD peaks of the periodic layers of PTB7 and Ir(III) complexes and their corresponding FWHMs (Figure S13 and Table 2). For example, in the case of Ir-Blue, the LC(100) was enhanced from 62.2 to 67.1 Å, while LC(010), the π−π stacking, was enhanced from 16.7 to 18.8 Å. Ir-Green exhibited the biggest increase in LC(010) from 16.7 to 19.5 Å, corresponding to about five layers of π−π stacking. The enhanced molecular compatibility was related to the 3hydroxypicolinic acid unit of the Ir(III) complexes. By comparing the GI-WAXD peaks of Ir-Green with those of (ppy)2Irpic (Ir-Green2, no hydroxyl unit) in PTB7:PC71BM blended films, Ir-Green2 clearly had an adverse effect on the morphology of PTB7:PC71BM, resulting in weaker GI-WAXD peaks (Figure S13). This effect was also confirmed by the AFM experiments, which showed a higher root-mean-square value for Ir-Green2 than that for Ir-Green (Figure S14). It was also found that the Ir(III) complexes enhance the dipole moments of Ir(III) complexes:PTB7 incorporated molecules (Figure S15). The enhanced dipole moment contributed to the alignment of PTB7 molecules as well as their crystallinity.25 Ir-Blue showed optimal dipole momenta and nanocrystallite sizes of PTB7 and PC71BM. Therefore, Ir-Blue enhances the PCE as much as Ir-Red does, although Ir-Blue has a lower spectral overlap with PTB7 (Table 1). Interestingly, when 10 wt % Ir-Blue was added to the PTB7 film (PTB7:Ir-Blue 10 wt %), a new scattering peak occurred at qxy = 0.712 ± 0.003 Å−1 between those of PTB7 layers, rather than individual peaks (Figure 4g), indicating the incorporation of thiophene spacers of PTB7 and fluorine substituents of Ir-Blue at the molecular level (Figure 4f).26 We also fabricated devices with the Ir(III) complexes as the energy donor, the high-crystallinity polymer P3HT as the energy acceptor and charge donor, and PC61BM as the charge acceptor. The complexes Ir-Blue and Ir-Orange, which have relatively higher FRET radii (Figure S16), were selected to determine a suitable energy donor for P3HT. The value of JSC of the reference P3HT device increased from 10.3 to 11.5 and 11.0 mA·cm−2 after adding Ir-Blue and Ir-Orange, respectively. They correspond to PCE values of 3.63% for P3HT with IrBlue and 3.23% for P3HT with Ir-Orange (Figure S17 and

Blue was much smaller than that of Ir-Red, the PCE values of the corresponding devices were very similar (8.52% for Ir-Blue vs 8.53% for Ir-Red). This was because Ir-Blue exhibited better molecular compatibility with PTB7, resulting in an increase in the crystallinity of the active film. This is further discussed in morphological study. To better clarify the energy transfer effect of Ir(III) complexes, we fabricated devices with different fullerene derivative charge acceptor, PC60BM, which has no absorption in range of 350−500 nm. Figure S11 shows the enhancement of EQE and Jsc in the PTB7:PC60BM:Ir-Orange device (13.4 mA· cm−2) compared to that for the PTB7:PC60BM device (12.1 mA·cm−2). This means that Ir(III) complexes can contribute to an increase in the device performance regardless of the absorption ability of fullerene derivative materials. We also fabricated devices with the configuration glass/ITO/ZnO/IrOrange/PTB7:PC71BM/MoO3/Ag in order to elucidate the energy transfer effect in the active layer. Figure S12 shows the J−V curve with a separate Ir-Orange complex layer inserted between the ZnO and active layers, and Table 1 shows the corresponding device results. The device with the inserted IrOrange layer did not exhibit any increase in JSC (13.3 mA·cm−2 for both). Moreover, the VOC values were almost identical (0.75 V), and the FF value was slightly lower than that of the reference device (69.9 vs 71.7%). These results demonstrate that there was no energy transfer in the case of the inserted IrOrange layer because the FRET radius (