A high performance organic photovoltaic utilizing PEDOT:PSS and graphene oxide

Abstract

Interface engineering may lead to a high performance organic photovoltaic as well as long lifetime. Incorporating the commonly used conducting polymer PEDOT:PSS with GO as an the anode interfacial layer, poly[[4,8-bis[(2-ethylhexyl)oxy]benzo[1,2-b:4,5-b′]dithiophene-2,6-diyl][3-fluoro-2-[(2-ethylhexyl)carbonyl]thieno[3,4-b]thiophenediyl]] (PTB7):fullerene derivative (PC₇₁BM) bulk heterojunction (BHJ) organic photovoltaics (OPVs) were fabricated. This conventional OPV demonstrated a power conversion efficiency (PCE) of 7.53%, which has outperformed the highest reported PEDOT:PSS combined metal oxide, the highest reported efficiency with a solution processed PEDOT:PSS:GO interfacial layer as well as a PEDOT:PSS-based OPV (reference device). Moreover, the versatility of PEDOT:PSS:GO has also been demonstrated in the medium bandgap BHJ system employing poly[4,8-bis(5-(2-ethylhexyl)thiophen-2-yl)benzo[1,2-b;4,5-b′]dithiophene-2,6-diyl-alt-(4-(2-ethylhexyl)-3-fluorothieno[3,4-b]thiophene-)-2-carboxylate-2-6-diyl] paired with PC₇₁BM yielding an astonishing high PCE of 9.12%. We believe proper interface engineering could lead to high performance devices.

---

1 Introduction

Recently, organic photovoltaics (OPVs) received huge interest as cost-effective, and flexible power sources.^1–6 We have seen a rapid improvement during the last few years in terms of power conversion efficiency (PCE), where Yang's group and Heliatek along with their respective co-workers have surpassed the 10% efficiency barrier in multiple junction solar cells.⁷ In addition, Yusoff and coworkers have reported an 11.83% efficiency in a triple-junction OPV.⁸ This big jump from the previously reported 9.2% PCE from Zhou and co-workers is probably due to the fact that the synthesized photosensitive active materials are able to generate more excitons. Despite these advancements, a major drawback in OPVs is related with instability at the interface between the commonly used transparent electrode, indium tin oxide (ITO) and an anode interfacial layer.^9–11 One of the methods is to make use of metal oxide because a deep valence band which makes this material stable. Another route is to make full use of an inverted architecture, where one can simply avoid the unsolved interface issue; ITO/PEDOT:PSS. Here, we report that combining PEDOT:PSS with other materials such as tungsten oxide (WO_x), molybdenum oxide (MoO_x), carbon nanotube (CNT) or graphene oxide (GO) can improve the device performance as well as their respective lifetime. Interestingly, the fabricated OPV combining PEDOT:PSS with GO results in 7.53% (PTB7:PC₇₁BM) and 9.12% (PBDTTT-EFT:PC₇₁BM) efficiencies compared to 7.27% (PTB7:PC₇₁BM) and 8.63% (PBDTTT-EFT:PC₇₁BM) efficiencies for PEDOT:PSS OPV (reference devices). Our OPVs incorporating the well-characterized and popular electron donating polymer poly[[4,8-bis[(2-ethylhexyl)oxy]benzo[1,2-b:4,5-b0]dithiophene-2,6-diyl][3-fluoro-2-[(2-ethylhexyl)carbonyl]-thieno[3,4-b]thiophenediyl]] (PTB7; Fig. 1a) and the fullerene electron acceptor [6,6]-phenyl-C₇₁-butyric-acid-methyl-ester (PC₇₁BM) and also a medium bandgap polymer poly[4,8-bis(5-(2-ethylhexyl)thiophen-2-yl)benzo[1,2-b;4,5-b′]dithiophene-2,6-diyl-alt-(4-(2-ethylhexyl)-3-fluorothieno[3,4-b]thiophene-)-2-carboxylate-2-6-diyl] (PBDTTT-EFT) paired with PC₇₁BM; Fig. 1a). While the efficiency level is currently state-of-the-art, the performance of PTB7 devices and many other OPV systems is limited by inhomogeneous structural and electrical properties of PEDOT:PSS.^9–11 Thus, many works have been reported to alternative polymeric, cross-linking, self-assembling, and inorganic anode interfacial materials as PEDOT:PSS replacements.^12–15

Fig. 1 Comparative photovoltaic devices with PTB7:PC₇₁BM and PBDTTT-EFT:PC₇₁BM with PEDOT:PSS, PEDOT:PSS:WO_x, PEDOT:PSS:MoO_x, PEDOT:PSS:GO, and PEDOT:PSS:CNT. (a) Chemical structures of the PTB7, PC₇₁BM, PFN-P1, WO_x, MoO_x, GO, and CNT. (b) Schematic structure of conventional OPVs.

2 Results and discussion

In this work, we show that, utilizing novel solution-processed PEDOT:PSS:GO can be deposited as anode interfacial layer onto ITO transparent electrode and at the same time offering better transparency than PEDOT:PSS. The molecular structures of PTB7, PC₇₁BM, WO_x, MoO_x, CNT and GO, the device schematic structure and the energy levels of the used materials in this work are depicted in Fig. 1. The details preparations are documented in the Experimental section. In brief, our conventional OPVs did not undergo any thermal annealing treatment expect for anode interfacial layer. The work function of PEDOT:PSS, PEDOT:PSS:WO_x, PEDOT:PSS:MoO_x, PEDOT:PSS:CNT, and PEDOT:PSS:GO were determined using Kelvin probe measurements. The work function of PEDOT:PSS:GO is approximately 4.6 eV, which is close to the lowest unoccupied molecular orbital (LUMO) of PC₇₁BM (4.5 eV). Hence, hole transport to ITO is desirable.

Fig. 2a and b show the dark current (J–V) and J–V characteristics under AM1.5G illumination at 100 mW cm⁻² for conventional OPVs with different anode interfacial layers (PEDOT:PSS, PEDOT:PSS:WO_x, PEDOT:PSS:MoO_x, PEDOT:PSS:GO, and PEDOT:PSS:CNT), respectively. Table S1† collects the photovoltaic parameters for all devices. The device with PEDOT:PSS anode interfacial layer has a short-circuit current density (J_sc) of 14.5 mA cm⁻², an open-circuit voltage (V_oc) of 0.75 V, a fill factor (FF) of 67% along with PCE of 7.27%. When PEDOT:PSS is combined with WO_x as anode interfacial layer, the V_oc decreases from 0.75 to 0.73 V but both J_sc and FF remain unchanged. However, as we substitute WO_x with MoO_x, both J_sc and FF drop to 11.4 mA cm⁻² and 64.2%, respectively.

Fig. 2 (a) Dark J–V characteristics, (b) J–V characteristics measured under AM1.5G illumination at 100 mW cm⁻², (c) EQE spectra and (d) optical transmittance of the conventional OPVs using PEDOT:PSS, PEDOT:PSS:WO_x, PEDOT:PSS:MoO_x, PEDOT:PSS:GO, and PEDOT:PSS:CNT anode interfacial layers.

Since the V_oc of OPVs is determined by the difference in LUMO of acceptor and HOMO of donor, as well as the work function difference between anode and cathode,^16,17 the decrease in V_oc can be attributed to the enhanced valence band of anode after PEDOT:PSS:MoO_x deposition (4.87 eV for PEDOT:PSS:MoO_x coated ITO vs. 5.04 eV for PEDOT:PSS, see Fig. 3a). This argument is supported by the dark J–V curves in which PEDOT:PSS:MoO_x device shows a low current in the forward direction and high leakage current in reverse direction. The series resistance (R_s) increases to 8.86 Ω cm² and the shunt resistance (R_sh) decreases to 693.1 Ω cm² in PEDOT:PSS:MoO_x device, compared to PEDOT:PSS device which has R_s and R_sh of 7.62 Ω cm² and 714.3 Ω cm², respectively. On the other hand, the combination of PEDOT:PSS and CNT layer enhances J_sc, V_oc and FF concurrently, and thus PCE of 7.48% was achieved.

Fig. 3 (a and b) UPS spectra of HTLs on top of ITO. Energy-level diagrams of (c) PEDOT:PSS, (d) PEDOT:PSS:WO_x, (e) PEDOT:PSS:MoO_x, (f) PEDOT:PSS:GO, and (g) PEDOT:PSS:CNT. (h) Complete energy-levels for each layers of the device.

The reduced work function of combined PEDOT:PSS and CNT (3.8 eV from 4.5 eV, Fig. 3b) increases the work function difference between anode and cathode, increasing the V_oc to 0.74 V. The R_s reduces to 6.26 Ω cm² and R_sh decreases to 533.3 Ω cm². The FF and J_sc increase from 67 to 68.1% and from 14.5 to 14.8 mA cm⁻², respectively. Finally, blending PEDOT:PSS and GO improves all photovoltaic parameters including J_sc, V_oc, and FF along with PCE of about 7.53%. Further reduction in work function of combined PEDOT:PSS and GO (3.8 eV from 4.5 eV) increases the work function difference between anode and cathode, increasing the V_oc to 0.75 V. The R_s reduces to 7.5 Ω cm² and R_sh decreases to 702.9 Ω cm². Both FF and J_sc enhance from 67 to 68.3% and from 14.5 to 14.9 mA cm⁻², respectively. In both cases (PEDOT:PSS:CNT and PEDOT:PSS:GO) the outstanding diode quality factor (1.3595 and 1.3574, respectively) along very low negative bias leakage current density and high rectification ratio are obviously shown in Fig. 2a. While the role of GO in the enhancement of J_sc and FF of device is not clear at this point, we speculate that the increased built-in potential and internal electric field, due to reduced work function of cathode, might facilitate the charge carriers to escape shallow traps and reduce trap-assisted recombination.

We performed the UPS analysis to understand the influence of the proposed HTLs on the device performances. Fig. 3b–f show the energy-level diagrams for various HTLs on the ITO layer. The Δ_VB for ITO/PEDOT:PSS layer is 0.50 eV. When the WO_x, and MoO_x were blended into the PEDOT:PSS, the Δ_VB values were decreased from 0.50 eV to 0.40 eV, 0.35 eV, respectively while the Δ_VB increased from 0.50 eV to 0.55 eV. The anticipated hole injection barrier can be estimated by the difference between the Fermi level and the valance band of the HTL. The results show that a decrease or increase of hole injection barrier, and thereby demonstrate that hole injection barrier was influenced by the introduction of additional interlayers and also show that interface engineering is particularly important in designing a high performance organic photovoltaic device. Thus, the decrease or increase of hole injection barrier caused by shift of ΔE_vac lead to a larger and smaller hole injection barrier. On the other hand, when we introduced GO into PEDOT:PSS, the hole would transport more efficiently from the active layer to the anode.

To support our speculation, Fig. 4 demonstrates a significant decrease in hole injection barrier in the PEDOT:PSS:GO case via the downshift of the ΔE_vac that accounts for the enhanced J_sc and PCE.

Fig. 4 Energy level diagrams of the ITO with PEDOT:PSS and PEDOT:PSS blended with WO_x, MoO_x, GO and CNT.

To explore the origins of the differences in device performance, the surface topography of each anode interfacial layer was studied by atomic force microscopy (AFM). Fig. 5 presents AFM images of the PEDOT:PSS, PEDOT:PSS:WO_x, PEDOT:PSS:MoO_x, PEDOT:PSS:CNT, and PEDOT:PSS:GO films on ITO substrates. Note the significant vertical scale differences in the two images, which illustrate the very different surface roughness of these five anode interfacial layers. In particular, the PEDOT:PSS film (Fig. 5a) exhibits an overall root-mean-square (RMS) roughness of 2.1 nm, whereas Fig. 4b reveals a slightly higher RMS roughness of 2.4 nm for the PEDOT:PSS:WO_x layer. Fig. 5c shows a distinctive feature, where RMS roughness of 4.2 nm was obtained for the PEDOT:PSS:MoO_x layer. Whereas, Fig. 5d and e depict a 2.1 nm RMS roughness for both PEDOT:PSS:CNT and PEDOT:PSS:GO layers, respectively. Fig. S1† shows the SEM images of these interlayers.

Fig. 5 AFM images displaying the different film surface morphologies of (a) PEDOT:PSS, (b) PEDOT:PSS:WO_x, (c) PEDOT:PSS:MoO_x, (d) PEDOT:PSS:GO, and (e) PEDOT:PSS:CNT anode interfacial layers with different RMS roughness of 2.08, 2.49, 4.20, 2.11, and 2.06 nm, respectively. The scan size is 5 μm × 5 μm at a scan speed rate of 1 Hz.

To gain further insight into the relative OPV performance of the PEDOT:PSS, PEDOT:PSS:WO_x, PEDOT:PSS:MoO_x, PEDOT:PSS:GO, or PEDOT:PSS:CNT anode interfacial layers, optical transmission (Fig. 2c) and external quantum efficiency (EQE) (Fig. 2d) measurements were carried out as a function of wavelength. The transmittance plots show that the transparencies of PEDOT:PSS:WO_x, PEDOT:PSS:MoO_x, PEDOT:PSS:GO, and PEDOT:PSS:CNT layers are better compared to the PEDOT:PSS layer across the entire visible spectrum. Moreover, device fabricated with PEDOT:PSS:GO anode interfacial layer outperformed the one with PEDOT:PSS anode interfacial layer in the 400–700 nm wavelength range.

The EQE value for PEDOT:PSS:GO device approaches 70.4% at around 460 nm. The integral J_sc calculated from the EQE spectrum from 300 nm to 800 nm is 14.71 mA cm⁻² is indeed in a good agreement (1.27% error) with J_sc of 14.9 mA cm⁻² obtained from J–V characteristics (Fig. 2b). Other integrated J_sc's values from the EQE spectra are also in good agreement with 1–3% error with J_sc obtained from J–V plots (Fig. 2b). In addition, we also tested solvent resistance measurements for all these anode interfacial layers including chlorobenzene, dichlorobenzene, and trichlorobenzene. All interfacial layers survived both dichlorobenzene and trichlorobenzene solvents, however, all interfacial layers were damaged after being treated with chlorobenzene solvent. This can be beneficial especially in using these interfacial layers as interconnecting layer in tandem solar cell (Fig. S2†).

In order to understand the capability of PEDOT:PSS:GO as the promising anode interfacial layer, we designed and characterized another series of OPVs with a newly developed medium bandgap polymer, poly[4,8-bis(5-(2-ethylhexyl)thiophen-2-yl)benzo[1,2-b;4,5-b′]dithiophene-2,6-diyl-alt-(4-(2-ethylhexyl)-3-fluorothieno[3,4-b]thiophene-)-2-carboxylate-2-6-diyl] paired with PC₇₁BM. Fig. 6a shows the J–V curves of PBDTTT-EFT:PC₇₁BM measured under AM1.5G illumination (25 °C, 100 mW cm⁻²). The complete conventional device comprises of ITO/HTL/PBDTTT-EFT:PC₇₁BM/PFN/Al. It is worth noting that no annealing treatment has been carried out toward the active layer. Details fabrication procedure can be obtained in the Experimental section. The J_sc, V_oc, FF and PCE along with the R_s and R_sh values for OPVs are listed in Table S2.† As can be seen from Table S2,† the V_oc is similar for all fabricated devices, which is reasonable because V_oc is governed by the energetic relationship between the donor and acceptor in the BHJ-based OPVs. As shown, the cells with PEDOT:PSS:MoO_x of (35 nm) show relatively poor performance due to a significantly low FF compared to that of other devices, mainly due to the high R_s and low R_sh values. The poor performance of PEDOT:PSS:MoO_x-based device can be supported by the fact that it demonstrated lower currents and higher leakage currents in the forward and reverse directions, respectively (Fig. 6b). All photovoltaic parameters are enhanced by the integration of PEDOT:PSS:GO (34 nm) as the anode interfacial layer. The PCE of the cells with PEDOT:PSS:GO improved by 4.23% from 8.75 to 9.12%. This is largely due to the increased in the J_sc value to 16.9 mA cm⁻² compared to that of PEDOT:PSS-based device with only 16.1 mA cm⁻². The improvement in PCE can be evidenced by the downshift of the ΔE_vac (Fig. S1†) leading to efficient hole transport process from the PBDTTT-EFT:PC₇₁BM to the anode.

Fig. 6 (a) J–V characteristics measured under AM1.5G illumination at 100 mW cm⁻², (b) dark J–V characteristics, and (c) EQE spectra the conventional OPVs using PEDOT:PSS, PEDOT:PSS:WO_x, PEDOT:PSS:MoO_x, PEDOT:PSS:GO, and PEDOT:PSS:CNT anode interfacial layers.

To understand the origin of this remarkable improvement, Fig. 6c demonstrates the EQE spectra for PBDTTT-EFT:PC₇₁BM-based OPVs. As one can see from these spectra, the conventional single-junction OPVs realizing PEDOT:PSS:GO exhibited the maximum EQE of 75.4%, which corresponds to efficient photo-to-electron conversion. In addition, the lowest EQE comes from the cells with PEDOT:PSS:MoO_x, where the maximum EQE is 73.7%. Moreover, for all PBDTTT-EFT:PC₇₁BM-based devices, the integral current density deduced by EQE spectra are in good agreement with our experimental recorded J_sc values. The different between measured and the calculated current density values are within 5%, indicating that the photovoltaic measurement is reliable.

3 Conclusion

We have successfully demonstrated a conventional OPV with 7.53% and 9.12% with PTB7:PC₇₁BM and PBDTTT-EFT:PC₇₁BM, respectively using PEDOT:PSS:GO anode interfacial layer. Blending PEDOT:PSS with GO increases the efficiency of the conventional OPV as a result of reducing the work function of the anode. The PEDOT:PSS:GO HTL plays a vital role in enhancing J_sc and PCE in OPV and thus demonstrates the highest PCE of 7.53% and 9.12% and the EQE of 70% and 75.4% for PTB7:PC₇₁BM and PBDTTT-EFT:PC₇₁BM, respectively. Thus, it is believe that proper interface engineering could lead to highly efficient OPV.

Acknowledgements

This work was supported by MOTIE Ministry of Trade, Industry & Energy (10052044) and KDRC (Korea Display Research Corporation) support program for the development of future device technology for display industry.

References

1. C. J. Brabec, N. S. Sariciftci and J. C. Hummelen, Adv. Funct. Mater., 2008, 11, 15–26.
2. S. E. Shaheen, D. S. Ginley and G. E. Jabbour, MRS Bull., 2005, 30, 10.
3. J. Peet, J. Y. Kim, N. E. Coates, W. L. Ma, D. Moses and A. J. Heeger, Nat. Mater., 2007, 6, 497–500.
4. W. Ma, C. Yang, X. Gong, K. Lee and A. J. Heeger, Adv. Funct. Mater., 2005, 15, 1617–1622.
5. P. Wang, A. Abrusci, H. M. P. Wong, M. Svensson, M. R. Andersson and N. C. Greenham, Nano Lett., 2006, 6, 1789–1793.
6. A. R. B. M. Yusoff, S. J. Lee, H. P. Kim, F. K. Shneider, W. J. da Silva and J. Jang, Adv. Funct. Mater., 2014, 24, 2240.
7. J. You, L. Dou, K. Yoshimura, T. Kato, K. Ohya, T. Moriarty, K. Emery, C. C. Chen, J. Gao, G. Li and Y. Yang, Nat. Commun., 2013, 4, 1446.
8. A. R. B. M. Yusoff, D. Kim, H. P. Kim, F. K. Schneider, W. J. da Silva and J. Jang, Energy Environ. Sci., 2015, 8, 303.
9. W. J. da Silva, H. P. Kim, A. R. B. M. Yusoff and J. Jang, Nanoscale, 2013, 5, 9324.
10. T. M. Barness, J. D. Bergeson, R. C. Tenent, B. A. Larsen, G. Teeter, K. M. Jones, J. L. Blackburn and J. van de Lagemaat, Appl. Phys. Lett., 2010, 96, 243309.
11. F. C. Krebs, M. Hösel, M. Corazza, B. Roth, M. V. Madsen, S. A. Gevorgyan, R. R. Søndergaard, D. Karg and M. Jørgensen, Energy Technol., 2013, 1, 378.
12. Y. Zhou, C. F. Hernandez, J. W. Shim, T. M. Khan and B. Kipplen, Energy Environ. Sci., 2012, 5, 9827.
13. G. Brotas, J. Farinhas, Q. Ferreira, R. Rodrigues, I. L. Martins, J. Morgado and A. Charas, J. Polym. Sci., Part A: Polym. Chem., 2014, 52, 652.
14. S. K. Nisha and S. K. Asha, J. Mater. Chem. C, 2014, 2, 2015.
15. A. R. B. M. Yusoff, H. P. Kim and J. Jang, Nanoscale, 2014, 6, 1537.
16. L. M. Chen, Z. Xu, Z. Hong and Y. Yang, J. Mater. Chem., 2010, 20, 2575.
17. B. Yang, F. Guo, Y. Yuan, Z. Xiao, Y. Lu, Q. Dong and J. Huang, Adv. Mater., 2013, 25, 572.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra04376c

---