Doping for Controlling Open-Circuit Voltage in Organic Solar Cells

Feb 23, 2018 - Doping, addition of trace amount of p-type and n-type impurities, to form the pn junction is the central technology in inorganic solar ...
0 downloads 0 Views 1004KB Size
Download PDF
Article Cite This: J. Phys. Chem. C 2018, 122, 5248−5253

pubs.acs.org/JPCC

Doping for Controlling Open-Circuit Voltage in Organic Solar Cells Naoto Shintaku, Masahiro Hiramoto, and Seiichiro Izawa* Institute for Molecular Science, 5-1 Higashiyama, Myodaiji, Okazaki, 444-8787 Aichi, Japan SOKENDAI (The Graduate University for Advanced Studies), 5-1 Higashiyama, Myodaiji, Okazaki, Aichi 444-8787, Japan ABSTRACT: Doping, addition of trace amount of p-type and n-type impurities, to form the pn junction is the central technology in inorganic solar cells. However, the doping effect on the energy-level alignment and the performance in organic solar cells (OSCs) are still unclear. Here, we report that the addition of p-type (MoO3) and n-type (Cs2CO3) dopants into a donor layer in phthalocyanine/fullerene planar heterojunction OSCs controls the open-circuit voltage (VOC). The VOC decreased to 0.36 V when a p-type dopant was added to the donor layer, whereas it increased to 0.52 V with an n-type dopant. In contrast to the previous reports where p-type dopants were usually added to the donor layer, the n-type dopant was found to increase the VOC. Energy-level mapping revealed that the origin of the VOC change was the vacuum level shifts occurring near the donor/acceptor (D/A) interface because of the Fermi-level alignment. The results demonstrated that the VOCs in OSCs are largely affected by the energy-level shift near the D/A interface that could be controlled by p-type and n-type doping.

■

INTRODUCTION Open-circuit voltage (VOC) is one of the most important parameters for determining the power conversion efficiency (PCE) of organic solar cells (OSCs).1,2 The VOC in OSCs is primary related to the energy-level difference between the highest occupied molecular orbital (HOMO) of an electron donor material and the lowest unoccupied molecular orbital (LUMO) of an electron acceptor material.3 Previous reports have demonstrated that the eVOC (e: elementary charge) is empirically expressed by the energy-level difference between the HOMO of the donor material and the LUMO of the acceptor material (EDA), as measured by cyclic voltammetry minus 0.3 eV.4 However, the EDA in actual devices is sometimes changed because of the energy-level alignment at the donor/ acceptor (D/A) interface. After different layers are contacted, charge transfer occurs between them to align the Fermi energy level (EF) and the generated electrostatic field near the interface induces a vacuum level (VL) shift.5−9 The generated charges are also trapped at the gap state in the bulk, inducing band bending over a long range along the direction away from the interface.10 The EF alignment occurs at organic/organic interfaces as well as inorganic/organic interfaces.11 The VOC in OSCs can be controlled if the energy-level shift near the D/A interface can be manipulated. Doping, addition of trace amount of p-type and n-type impurities, is the central technology in the inorganic solar cells. It has also been studied in the field of OSCs.12−17 The doping effects in the previous reports of OSCs can be roughly classified into three topics: conductivity enhancement,12,14,15 trap filling,16 and dipole moment formation.17 Liu et al. have reported that the addition of a molecular p-type dopant tetrafluoro-tetracyanoquinodimethane (F4TCNQ) into solution-processed bulk heterojunction (BHJ) OSCs increased the hole conductivity and improved the short-circuit current (JSC) and fill factor (FF).15 Shang et al. have reported that the © 2018 American Chemical Society

addition of a p-type dopant at a low concentration into BHJ OSCs filled the deep traps in the active layer and led to the increase of approximately 0.3% and 0.02 V for the PCE and the VOC, respectively.16 Yu et al. have reported that the addition of F4TCNQ in a polymer layer formed a dipole moment at the D/A interface and increased the VOC.17 However, the effect of doping on the relationship between the EF alignment and the VOC in OSCs is still unclear. In the present study, p-type (MoO3) and n-type (Cs2CO3) dopants were added in the donor layer of phthalocyanine (H2Pc)/fullerene (C60) planar heterojunction (PHJ) devices. The PHJ devices provide an advantage in investigating the effect of doping at the D/A interface because their flat and distinct interface enables us to observe energy-level shifts there.18 Figure 1a shows a schematic of the effects of doping on the energy-level alignment. We have previously found that the EF of an organic layer can be freely controlled by adding an appropriate dopant at an appropriate concentration.19−21 The EF of the films becomes deeper when a p-type dopant is added, whereas it becomes shallower when an n-type dopant is added. After the doped H2Pc and C60 layers are contacted, EF alignment occurs, leading to a VL shift near the D/A interface. The addition of p-type and n-type dopants induces an energylevel shift in the directions of decreasing and increasing EDA, respectively. We hypothesized that the VOC could be controlled by this effect.

■

EXPERIMENTAL SECTION Materials. H2Pc (Dainippon Ink & Chemicals) and C60 (Frontier Carbon, Nano Purple TL) were purified by singleReceived: December 11, 2017 Revised: February 23, 2018 Published: February 23, 2018 5248

DOI: 10.1021/acs.jpcc.7b12203 J. Phys. Chem. C 2018, 122, 5248−5253

Article

The Journal of Physical Chemistry C

indium tin oxide (ITO)-coated glass substrates (ITO thickness: 150 nm; sheet resistance: 10.3 Ω sq−1; Techno Print). Device Fabrication. The MoO3 hole-transporting layer (10 nm, 0.1 nm s−1), H2Pc donor layer (50 nm, 0.1 nm s−1), C60 acceptor layer (50 nm, 0.1 nm s−1), bathocuproine (BCP) electron transport layer (15 nm, 0.1 nm s−1), and Al electrodes (75 nm, 0.4 nm s−1) were deposited via thermal evaporation under high vacuum (∼10−5 Pa) in a vacuum evaporation system (VTS-350M, ULVAC) housed in a glovebox (DSO-1.5S MS3-P, Miwa). MoO3 and Cs2CO3 as the dopants were introduced by codeposition with H2Pc. Measurements. The devices were characterized in a vacuum container for optical measurements (EPITECH) without exposure to air. The J−V characteristics of the devices were measured under simulated solar illumination (AM 1.5, 100 mW cm−2) from a solar simulator based on a 300 W Xe lamp (HAL-320, Asahi Spectra) using a source meter (R6243, Advantest). The light intensity was calibrated with a standard silicon solar cell (CS-20, Asahi Spectra). The active area of the devices was defined using a 0.04 cm2 photomask. The position of the EF was determined using a Kelvin probe (Riken-Keiki, FAC-1) without exposure to air. The Kelvin probe was housed in a glovebox with N2 gas (H2O < 1 ppm, O2 < 0.1 ppm).

■

RESULTS AND DISCUSSION The PHJ devices with a doped H2Pc layer were fabricated by thermal evaporation under high vacuum. The structure of these devices was ITO/MoO3/doped H2Pc/C60/BCP/Al (Figure 1b). MoO3 and Cs2CO3 were used as the p-type and n-type dopants, respectively.19 The dopants were introduced into the H2Pc layer via a codeposition technique, and the dopant concentration relative to the H2Pc volume was controlled by the ratio between the deposition rate of the dopant and that of H2Pc. The energy levels of H2Pc, C60, MoO3, and Cs2CO3 are summarized in Figure 1c. The work function (WF) of the 50 nm H2Pc films on ITO was 4.20 eV in the undoped film, became deeper (4.56 eV) in the film doped with 2000 ppm MoO3, and became shallower (3.52 eV) in the film doped with 2000 ppm Cs2CO3. The J−V curves of the devices are shown in Figure 2a, and the device performances are summarized in Table 1. The PCE and VOC of the undoped device are 1.0% and 0.47 V, respectively; these values are very similar to those reported

Figure 1. (a) Schematic of the EF alignment at the D/A interface when the H2Pc layer is doped with p-type and n-type dopants. (b) Schematic of the OSC device with a doped H2Pc layer. (c) Chemical structures and energy levels of H2Pc and C60. The dashed lines indicate the WF of undoped (black), 200 ppm MoO3 doped (green), 2000 ppm MoO3 doped (blue), 200 ppm Cs2CO3 doped (orange), and 2000 ppm Cs2CO3 doped (red) H2Pc layers (50 nm) on ITO substrates.

crystal sublimation. MoO3 (Alfa Aesar, 99.9995%) and Cs2CO3 (Sigma-Aldrich, 99.995%) were used as the p-type and n-type dopants, respectively. The OSC devices were fabricated on

Figure 2. (a) J−V curves of undoped H2Pc/C60 devices (black) and doped H2Pc/C60 devices with MoO3 concentrations of 200 ppm (green), 2000 ppm (blue), and 20 000 ppm (purple) and Cs2CO3 concentrations of 200 ppm (yellow) and 2000 ppm (red) under AM 1.5 irradiation (100 mW cm−2, solid lines) or in the dark (dashed lines). (b) VOC of doped H2Pc/C60 devices as a function of dopant concentration. 5249

DOI: 10.1021/acs.jpcc.7b12203 J. Phys. Chem. C 2018, 122, 5248−5253

Article

The Journal of Physical Chemistry C

Table 1. Summary of the Performances of the Doped H2Pc/C60 Devices with Different Concentrations of MoO3 and Cs2CO3a

a

dopant

MoO3

concentration/ppm JSC/mA cm−2 VOC/V FF/% PCE/%

20 000 1.91 (0.04) 0.355 (0.0054) 58.6 (0.34) 0.40 (0.0080)

2000 2.36 (0.42) 0.396 (0.0087) 57.7 (1.0) 0.54 (0.11)

200 3.30 (0.15) 0.432 (0.0038) 59.6 (0.21) 0.85 (0.042)

undoped

Cs2CO3

0 3.68 (0.18) 0.466 (0.0044) 58.8 (0.34) 1.01 (0.057)

200 3.37 (0.0022) 0.509 (0.0030) 53.8 (0.15) 0.92 (0.0073)

2000 0.53 (0.054) 0.519 (0.000084) 25.5 (0.32) 0.07 (0.0062)

Averaged values are reported. Values inside parentheses are standard deviations.

Figure 3. (a) WF of C60 films and H2Pc films without dopant (black) and with 2000 ppm of MoO3 (green) and Cs2CO3 (red), as measured by the Kelvin probe method and plotted as a function of the thickness. The C60 and H2Pc films were stacked on BCP/Al/ITO substrates. The value of 0 nm corresponds to the WF of the surface of the BCP/Al/ITO substrate. The region corresponding to the H2Pc layer thickness less than 10 nm is highlighted. (b) Energy levels relative to the EF of C60 films and H2Pc films without dopant (black) and with 2000 ppm of MoO3 (green) and Cs2CO3 (red); these results are based on the results in Figure 3a. We assumed that the positions of the HOMO and the LUMO energy levels relative to the VL are constant irrespective of the presence of the dopant.

previously.22 In the case of PHJ devices with MoO3-doped H2Pc, the VOC gradually decreased from 0.43 to 0.40 and 0.36 V for devices with dopant concentrations of 200, 2000, and 20 000 ppm. These devices show similar FF values (∼59%), indicating that charge recombination processes were not affected by doping. Therefore, the decrease in the VOC with increasing MoO3 dopant concentration is attributed to the change in the WF of the H2Pc layer. In the case of PHJ devices with Cs2CO3-doped H2Pc, the VOC increased from 0.47 V in the undoped film to 0.51 and 0.52 V in films with dopant concentrations of 200 and 2000 ppm, respectively. In contrast to the MoO3 case, the JSC and the FF steeply decreased with increasing Cs2CO3 concentration. The forward current under the dark condition decreased with increasing Cs 2 CO 3 concentration, indicating that Cs2CO3 doping decreased the hole conductivity in the H2Pc layer. Cs2CO3 doping caused a decrease of either the hole concentration or the hole mobility in the H2Pc layer. Nevertheless, the VOC change upon Cs2CO3 doping resulted in the opposite trend as the MoO3 case; therefore, we concluded that the WF decrease in the H2Pc film by Cs2CO3 doping was responsible for the VOC increase. The WF difference in the H2Pc film induced by p-type and n-type doping resulted in VOC values ranging from 0.36 V at a MoO3 concentration of 20 000 ppm to 0.52 V at a Cs 2 CO 3 concentration of 2000 ppm (Figure 2b). The large VOC change of 0.16 V in total suggested that energy levels near the D/A interface were greatly shifted by the addition of MoO3 and Cs2CO3 in the H2Pc layer. Noteworthy point is that most of the previous reports of the doping in OSCs added a p-type dopant in donor layers to enhance the hole conductivity.12−16 However, we found that n-type doping in the donor layer increased the VOC.

To estimate the energy-level shift near the D/A interface, we carried out energy-level mapping using the Kelvin probe method.23,24 The WFs of C60 layers with different thicknesses were measured on ITO/Al/BCP substrates. Subsequently, the WFs of the H2Pc layer with and without the dopants were measured on C60 layers. The results of the energy-level mapping are shown in Figure 3a. The WF of the C60 films on ITO/Al/BCP gradually increased from 3.12 eV at a film thickness of 0 nm to 4.18 eV at a film thickness of 50 nm. The WF of undoped H2Pc films on C60 layers steeply increased from 4.18 eV at the surface of the C60 film to 3.88 eV at a H2Pc film thickness of 10 nm and became constant as the thickness was increased further. The negative shift originated from the charge transfer induced by the EF difference between the H2Pc and C60 layers; this EF difference formed a large electrostatic field near the D/A interface, which in turn led to the VL shift. The WF of the film doped with 2000 ppm MoO3 shifted in the positive direction, showing a WF of 4.18 eV at the surface of the C60 film to 4.55 eV at a doped H2Pc film thickness of 50 nm because the EF of the p-doped H2Pc became deeper than that of C60. By contrast, the WF of the H2Pc film doped with 2000 ppm Cs2CO3 showed a large negative WF change from 4.18 eV at the surface of the C60 film to 3.54 eV at 50 nm. The large EF difference between the Cs2CO3-doped H2Pc layer and the C60 layer formed a large electrostatic field. The energy diagrams based on the results of band mapping are shown in Figure 3b. We assume that the positions of the HOMO and LUMO energy levels relative to the VL are constant irrespective of the presence of the dopant.25,26 The VL shifted steeply within 10 nm near the D/A interface because of the charge transfer, and the direction of the shift varied depending on the dopant. MoO3 doping induced an energy-level shift that decreased the 5250

DOI: 10.1021/acs.jpcc.7b12203 J. Phys. Chem. C 2018, 122, 5248−5253

Article

The Journal of Physical Chemistry C

Figure 4. Schematic of the structure of trilayer devices and the J−V curves of the devices with (a) 0 nm (black), 5 (green), 10 (blue), and 50 (purple) nm of 2000 ppm MoO3-doped H2Pc layer and (b) 0 nm (black), 5 (yellow), 10 (pink), and 50 (red) nm of 2000 ppm Cs2CO3-doped H2Pc layer under AM 1.5 irradiation (100 mW cm−2, solid lines) or in the dark (dashed lines). (c) VOC of trilayer devices as a function of the thickness of the doped layers. The region corresponding to a thickness less than 10 nm is highlighted.

Table 2. Summary of the Performances of the Trilayer Devices with Different Thicknesses of MoO3- and Cs2CO3-Doped H2Pc Layersa

a

dopant

MoO3

thickness (nm) JSC (mA cm−2) VOC (V) FF (%) PCE (%)

5 3.64 (0.039) 0.444 (0.0017) 59.1 (0.61) 0.96 (0.0091)

Cs2CO3 10 2.53 (0.12) 0.411 (0.0079) 58.8 (3.9) 0.61 (0.022)

50 2.36 (0.42) 0.396 (0.0086) 57.7 (1.0) 0.54 (0.11)

5 2.99 (0.066) 0.504 (0.0083) 42.9 (1.7) 0.65 (0.042)

10 1.86 (0.17) 0.521 (0.0027) 31.3 (0.41) 0.30 (0.030)

50 1.36 (0.64) 0.518 (0.012) 25.2 (3.8) 0.19 (0.12)

The table shows averaged values. Values inside parentheses are standard deviations.

EDA, whereas Cs2CO3 doping increased the EDA. The direction of the EDA change corresponded to the VOC difference. However, the eVOC difference of 0.16 eV is much smaller than the HOMO energy-level difference of 0.87 eV between MoO3- and Cs2CO3-doped H2Pc films 10 nm from the D/A interface. A recent report suggested that some portion of the generated charges formed by the EF alignment were transferred from the interface to the substrate; as a result, a weak electrostatic field was applied over the entire range of the film instead of near the interface.27 The shift obtained by energylevel mapping might be overestimated. Another noteworthy point is that the energy-level shift was almost completed less than 10 nm from the D/A interface. Therefore, the doping effect near the interface might determine the device performance.

To investigate the effect of doping near the D/A interface, we fabricated trilayer devices with a structure of ITO/MoO3/ H2Pc/doped H2Pc/C60/BCP/Al. The H2Pc/doped H2Pc layers were stacked, with thicknesses of 45 nm/5 nm and 40 nm/10 nm. The thin doped H2Pc layer constituted the D/A interface of the trilayer devices, and the undoped H2Pc layer existed apart from the D/A interface. The J−V curves of the trilayer devices with a 2000 ppm MoO3-doped layer or a 2000 ppm Cs2CO3-doped layer are shown in Figure 4a,b, respectively, and the device performances are summarized in Table 2. The device with a 5 nm MoO3-doped layer showed a VOC of 0.44 V, whereas the device with a 10 nm MoO3-doped layer showed a decrease in the VOC to 0.41 V, which was approximately the same value as the bilayer device with a 50 nm MoO3-doped layer (0.40 V). The FF and JSC of the trilayer devices with 5 and 5251

DOI: 10.1021/acs.jpcc.7b12203 J. Phys. Chem. C 2018, 122, 5248−5253

Article

The Journal of Physical Chemistry C

Heterojunction Solar Cells - Towards 10 % Energy-Conversion Efficiency. Adv. Mater. 2006, 18, 789−794. (5) Tang, J. X.; Lee, C. S.; Lee, S. T. Electronic Structures of Organic/organic Heterojunctions: From Vacuum Level Alignment to Fermi Level Pinning. J. Appl. Phys. 2007, 101, 064504. (6) Braun, S.; Salaneck, W. R.; Fahlman, M. Energy-Level Alignment at Organic/Metal and Organic/Organic Interfaces. Adv. Mater. 2009, 21, 1450−1472. (7) Oehzelt, M.; Akaike, K.; Koch, N.; Heimel, G. Energy-Level Alignment at Organic Heterointerfaces. Sci. Adv. 2015, 1, No. e1501127. (8) Zhong, S.; Zhong, J. Q.; Wang, X. Z.; Huang, M. Y.; Qi, D. C.; Chen, Z. K.; Chen, W. Investigation of Interface Properties for ClAlPc/C60 Heterojunction-Based Inverted Organic Solar Cell. J. Phys. Chem. C 2012, 116, 2521−2526. (9) Yost, S. R.; Van Voorhis, T. Electrostatic Effects at Organic Semiconductor Interfaces: A Mechanism for “Cold” Exciton Breakup. J. Phys. Chem. C 2013, 117, 5617−5625. (10) Mao, H. Y.; Bussolotti, F.; Qi, D.-C.; Wang, R.; Kera, S.; Ueno, N.; Wee, A. T. S.; Chen, W. Mechanism of the Fermi Level Pinning at Organic Donor−Acceptor Heterojunction Interfaces. Org. Electron. 2011, 12, 534−540. (11) Lee, J. H.; Shin, J.-H.; Song, J. Y.; Wang, W.; Schlaf, R.; Kim, K. J.; Yi, Y. Interface Formation Between ZnO Nanorod Arrays and Polymers (PCBM and P3HT) for Organic Solar Cells. J. Phys. Chem. C 2012, 116, 26342−26348. (12) Zhang, Y.; Zhou, H.; Seifter, J.; Ying, L.; Mikhailovsky, A.; Heeger, A. J.; Bazan, G. C.; Nguyen, T.-Q. Molecular Doping Enhances Photoconductivity in Polymer Bulk Heterojunction Solar Cells. Adv. Mater. 2013, 25, 7038−7044. (13) Ohashi, C.; Shinmura, Y.; Kubo, M.; Hiramoto, M. Effects of Doping at the ppm Level in Simple n+p-Homojunction Organic Photovoltaic Cells. Org. Electron. 2015, 27, 151−154. (14) Herrmann, F.; Muhsin, B.; Singh, C. R.; Shokhovets, S.; Gobsch, G.; Hoppe, H.; Presselt, M. Influence of Interface Doping on ChargeCarrier Mobilities and Sub-Bandgap Absorption in Organic Solar Cells. J. Phys. Chem. C 2015, 119, 9036−9040. (15) Liu, C.; Li, Z.; Zhang, Z.; Zhang, X.; Shen, L.; Guo, W.; Zhang, L.; Long, Y.; Ruan, S. Improving the Charge Carrier Transport of Organic Solar Cells by Incorporating a Deep Energy Level Molecule. Phys. Chem. Chem. Phys. 2017, 19, 245−250. (16) Shang, Z.; Heumueller, T.; Prasanna, R.; Burkhard, G. F.; Naab, B. D.; Bao, Z.; McGehee, M. D.; Salleo, A. Trade-Off between Trap Filling, Trap Creation, and Charge Recombination Results in Performance Increase at Ultralow Doping Levels in Bulk Heterojunction Solar Cells. Adv. Energy Mater. 2016, 6, 1601149. (17) Yu, S.; Frisch, J.; Opitz, A.; Cohen, E.; Bendikov, M.; Koch, N.; Salzmann, I. Effect of Molecular Electrical Doping on Polyfuran Based Photovoltaic Cells. Appl. Phys. Lett. 2015, 106, 203301. (18) Nakano, K.; Tajima, K. Organic Planar Heterojunctions: From Models for Interfaces in Bulk Heterojunctions to High-Performance Solar Cells. Adv. Mater. 2017, 29, 1603269. (19) Shinmura, Y.; Kubo, M.; Ishiyama, N.; Kaji, T.; Hiramoto, M. pn-Control and pn-Homojunction Formation of Metal-Free Phthalocyanine by Doping. AIP Adv. 2012, 2, 032145. (20) Shinmura, Y.; Yamashina, Y.; Kaji, T.; Hiramoto, M. Ionization Sensitization of Doping in Co-Deposited Organic Semiconductor Films. Appl. Phys. Lett. 2014, 105, 183306. (21) Hiramoto, M.; Kubo, M.; Shinmura, Y.; Ishiyama, N.; Kaji, T.; Sakai, K.; Ohno, T.; Izaki, M. Bandgap Science for Organic Solar Cells. Electronics 2014, 3, 351−380. (22) Kubo, M.; Shinmura, Y.; Ishiyama, N.; Kaji, T.; Hiramoto, M. Invertible Organic Photovoltaic Cells with Heavily Doped Organic/ Metal Ohmic Contacts. Appl. Phys. Express 2012, 5, 092302. (23) Ishii, H.; Hayashi, N.; Ito, E.; Washizu, Y.; Sugi, K.; Kimura, Y.; Niwano, M.; Ouchi, Y.; Seki, K. Kelvin Probe Study of Band Bending at Organic Semiconductor/Metal Interfaces: Examination of Fermi Level Alignment. Phys. Status Solidi A 2004, 201, 1075−1094.

10 nm MoO3-doped layers were similar to those of the bilayer device with a 50 nm MoO3-doped layer. Similar VOC trends were observed for the devices with Cs2CO3-doped films. The device with a 5 nm Cs2CO3-doped layer showed a VOC of 0.50 V, and the device with a 10 nm Cs2CO3-doped layer showed an increased VOC of 0.52 V; these values are approximately the same as those of the bilayer device with a 50 nm Cs2CO3-doped layer (0.52 V). These results indicate that 10 nm thick doped layers determined the magnitude of the increase or decrease of the VOC, consistent with the band-mapping results (Figures 3 and 4c). Notably, the device with a thinner Cs2CO3-doped layer showed larger JSC and FF values. These results indicate that the origin of the decrease of the JSC and FF in the devices with a Cs2CO3-doped layer was deterioration of the hole transport in the bulk and that this effect can be avoided by replacing the doped bulk layer with a undoped bulk layer. These results also indicate that the addition of dopants only near the D/A interface could result in an increase in the VOC while maintaining the JSC and FF values.

■

CONCLUSIONS In summary, we demonstrated that p-type and n-type doping strongly affects the VOC in H2Pc/C60 PHJ devices. The addition of a p-type dopant (MoO3) in the H2Pc layer decreased the VOC, whereas the addition of an n-type dopant (Cs2CO3) increased it. The wide range of VOC change was explained by the large energy-level shift within 10 nm of the D/A interface as a result of EF alignment between the donor and acceptor layers. The results demonstrated that the VOC in OSCs is determined not only by the choice of the donor and acceptor materials but also by the energy-level shift near the D/A interface. The results also showed that the shift and the VOC could be controlled through the addition of the dopants. The next challenge is to control the energy-level shift near the D/A interface and the VOC in BHJ devices via doping. Such control could be realized through selectively introducing the dopant into the donor or acceptor phases in the blend films.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Seiichiro Izawa: 0000-0002-1761-2682 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was supported in part by JSPS KAKENHI (Grant-in-Aid for Research Activity Start-up, no. 16H07421). The authors are grateful to Dr. Kouki Akaike at Tokyo University of Science for helpful scientific discussions.

■

REFERENCES

(1) Chen, H.-Y.; Hou, J.; Zhang, S.; Liang, Y.; Yang, G.; Yang, Y.; Yu, L.; Wu, Y.; Li, G. Polymer Solar Cells with Enhanced Open-Circuit Voltage and Efficiency. Nat. Photonics 2009, 3, 649−653. (2) Lu, L.; Zheng, T.; Wu, Q.; Schneider, A. M.; Zhao, D.; Yu, L. Recent Advances in Bulk Heterojunction Polymer Solar Cells. Chem. Rev. 2015, 115, 12666−12731. (3) Elumalai, N. K.; Uddin, A. Open Circuit Voltage of Organic Solar Cells: An in-Depth Review. Energy Environ. Sci. 2016, 9, 391−410. (4) Scharber, M. C.; Mühlbacher, D.; Koppe, M.; Denk, P.; Waldauf, C.; Heeger, A. J.; Brabec, C. J. Design Rules for Donors in Bulk5252

DOI: 10.1021/acs.jpcc.7b12203 J. Phys. Chem. C 2018, 122, 5248−5253

Article

The Journal of Physical Chemistry C (24) Yamashina, Y.; Shinmura, Y.; Ishiyama, N.; Kaji, T.; Hiramoto, M. Mapping of Band-Bending in Organic pn-Homojunctions. J. Appl. Phys. 2015, 117, 125501. (25) Ha, S. D.; Meyer, J.; Kahn, A. Molecular-Scale Properties of MoO3-Doped Pentacene. Phys. Rev. B: Condens. Matter Mater. Phys. 2010, 82, 155434. (26) Kröger, M.; Hamwi, S.; Meyer, J.; Riedl, T.; Kowalsky, W.; Kahn, A. p-Type Doping of Organic Wide Band Gap Materials by Transition Metal Oxides: A Case-Study on Molybdenum Trioxide. Org. Electron. 2009, 10, 932−938. (27) Akaike, K.; Koch, N.; Oehzelt, M. Fermi Level Pinning Induced Electrostatic Fields and Band Bending at Organic Heterojunctions. Appl. Phys. Lett. 2014, 105, 223303.

5253

DOI: 10.1021/acs.jpcc.7b12203 J. Phys. Chem. C 2018, 122, 5248−5253