Subscriber access provided by Linköping University Library
Communication
Flexible graphene electrode-based organic photovoltaics with record-high efficiency Hyesung Park, Sehoon Chang, Xiang Zhou, Jing Kong, Tomas Palacios, and Silvija Gradecak Nano Lett., Just Accepted Manuscript • DOI: 10.1021/nl501981f • Publication Date (Web): 20 Aug 2014 Downloaded from http://pubs.acs.org on August 22, 2014
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Nano Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Nano Letters
Flexible graphene electrode-based organic photovoltaics with record-high efficiency Hyesung Park1,2†, Sehoon Chang2†, Xiang Zhou2, Jing Kong1*, Tomás Palacios1*, Silvija Gradečak2* 1
Department of Electrical Engineering and Computer Science, Massachusetts Institute of
Technology, Cambridge, Massachusetts 02139 2
Department of Materials Science and Engineering, Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139
* Corresponding author email address: [email protected]; [email protected]; [email protected] †
These authors contributed equally to this work.
ACS Paragon Plus Environment
1
Nano Letters
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 29
Abstract Advancements in the field of flexible high-efficiency solar cells and other optoelectronic devices will strongly depend on the development of electrode materials with good conductivity and flexibility. To address chemical and mechanical instability of currently used indium tin oxide (ITO), graphene has been suggested as a promising flexible transparent electrode, but challenges remain in achieving high efficiency of graphene-based polymer solar cells (PSCs) compared to their ITO-based counterparts. Here we demonstrate graphene anode- and cathodebased flexible PSCs with record-high power conversion efficiencies of 6.1% and 7.1%, respectively. The high efficiencies were achieved via thermal treatment of MoO3 electron blocking layer and direct deposition of ZnO electron transporting layer on graphene. We also demonstrate graphene-based flexible PSCs on polyethylene naphthalate substrates and show the device stability under different bending conditions. Our work paves a way to fully graphene electrode-based flexible solar cells using a simple and reproducible process.
Keywords: flexible solar cell, polymer solar cell, graphene anode, graphene cathode
ACS Paragon Plus Environment
2
Page 3 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Nano Letters
Flexible organic and hybrid organic-inorganic solar cells require both the flexible photoactive media1, 2 and flexible electrodes with good conductivity and transparency. So far, indium tin oxide (ITO) has been an electrode material of choice for studies focusing on optimizing the morphology and chemistry of the photoactive media3 in polymer solar cells (PSCs). For flexible applications, graphene has been proposed as a promising replacement for ITO due to its mechanical and chemical robustness, excellent electrical and optical properties, and potentially low-cost processing4, 5. Recent studies have demonstrated dramatic improvements in the efficiency of PSCs6, 7 that ascertain the bright future towards PCE≥10%, a threshold considered for industrial applications2. In PSCs, and more generally organic solar cells (OSCs), one of the electrodes typically consists of a transparent conductor, among which ITO is most widely used due to its good optical transparency and electrical conductivity. However, due to its non-uniform absorption, chemical and mechanical instability, as well as high cost of indium8, several alternative materials have been proposed, including carbon nanotube9, 10 or metallic nanowires networks11. Furthermore, even at small mechanical stresses, microcracks are produced in ITO resulting in increased film resistance and decreased device performance12, thus limiting its applications for flexible OSCs. Owing to the unique optoelectronic properties of graphene5, several works on graphenebased OSCs have been reported, demonstrating the feasibility of graphene in transparent electrode applications3, 12-26. While the initial demonstrations have been promising, performance of graphene-based devices still falls short (< 3%) of recent advances accomplished in ITO-based devices (8 – 9%)6, 7. Therefore, to demonstrate graphene as an emerging alternative to ITO, it is inevitable to improve the currently low-efficiency of graphene-based solar cells. Moreover, to test feasibility of graphene for flexible applications, high performance and stability of such
ACS Paragon Plus Environment
3
Nano Letters
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 29
devices must be investigated. In this work, we demonstrate high-efficiency graphene anode- and cathode-based PSCs with PCEs comparable to their ITO-counterparts. We also demonstrate graphene-based flexible PSCs on polyethylene naphthalate (PEN) substrates and show the device stability under different bending conditions. The overall PSC device structure used in this work and the corresponding band diagram are shown in Fig. 1a-b. To fabricate efficient graphene-based OSCs, we used photoactive media composed of a blend of low bandgap semiconducting polymer donor thieno[3,4b]thiophene/benzodithiophene (PTB7) and acceptor [6,6]-phenyl C71-butyric acid methyl ester (PC71BM) prepared using mixed solvents of chlorobenzene:1,8-diiodoctane (CB:DIO, 97:3 vol%)6. The hole injection layer poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) was deposited on the transparent graphene electrode. To ensure uniform coverage over the graphene surface, we used modified PEDOT:PSS with isopropyl alcohol (IPA) at 3:1 (v/v) ratio. Prior to the active device layer deposition, graphene/PEDOT:PSS must be covered by an additional electron blocking layer (MoO3). This process is critical in graphenebased bulk heterojunction PSCs since the additional MoO3 layer prevents charge recombination occurring at the graphene/PEDOT:PSS and the polymer blend interface27. We also note that although ITO does not require additional MoO3 layer, ITO control devices were fabricated with PEDOT:PSS/MoO3 for direct comparison with graphene-based devices. We discovered that one critical aspect of the device fabrication is the thermal annealing of MoO3 layer before spin-coating PTB7:PC71BM. When directly applying PTB7:PC71BM to PEDOT:PSS/MoO3 on ITO or graphene without thermal annealing, considerable degradation in device performance was observed, which did not occur on PEDOT:PSS only device. Fig.1c presents the current density–voltage (J–V) characteristics of PSCs fabricated on ITO substrates,
ACS Paragon Plus Environment
4
Page 5 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Nano Letters
in which addition of MoO3 layer (non-annealed) results in detrimental effects on the shortcircuit current density (JSC, 2.6 mA/cm2), open-circuit voltage (VOC, 0.57 V), and fill factor (FF, 31.3%). However, significant improvements in JSC (16.1 mA/cm2), VOC (0.68 V), FF (60.7%), and the resulting increase in PCE were observed after the MoO3 electron blocking layer thermal annealing (Fig. 1d and Supplementary Table S1). Graphene-based PSCs were fabricated under the same experimental conditions and the resulting J–V characteristics are compared with that of ITO reference device in Fig. 1d. The graphene electrode was made by stacking three monolayers of graphene film prepared by low pressure chemical vapor deposition, with a typical sheet resistance of ~300 Ω/sq and transmittance of ~92% at λ = 550 nm13. With incorporation of appropriate PEDOT:PSS and thermally-treated MoO3, we observed record-high efficiency from graphene (PCE=6.1%) approaching to ITO reference device (PCE=6.7% ) (Supplementary Table S1). Since the efficient graphene PSC with PTB7:PC71BM was accomplished only in the presence of MoO3 electron blocking layer with subsequent thermal treatment, we investigated the effect of CB:DIO solvent on MoO3 via several routes. First, the surface morphology of MoO3 film (20 nm, on glass) was characterized using atomic force microscopy (AFM) after the thermal and solvent treatments (Fig. 2a-d). Figure 2a shows the surface topography of as-evaporated MoO3 film is smooth with root-mean-square (rms) roughness of ~1 nm. The thermal treatment alone does not affect significantly topography of MoO3 (Fig. 2b). However, noticeable change in the surface morphology was observed after spin-coating CB:DIO on non-annealed MoO3 (Fig. 2c), indicative of adverse effects of the solvent. In contrast, after thermal treatment and subsequent CB:DIO spin-coating on MoO3, the surface profile (Fig. 2d) is similar to that of annealed MoO3 (Fig. 2b) indicating that annealed MoO3 film becomes robust upon the solvent treatment.
ACS Paragon Plus Environment
5
Nano Letters
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 29
We further characterized the effect of solvent treatment on MoO3 via scanning transmission electron microscopy (STEM), as shown in Fig. 2e-h, which revealed similar structural changes. STEM image of as-deposited MoO3 is largely featureless (Fig. 2e) with minimum contrast differences indicating a uniform structure. Significant difference is observed from the asdeposited MoO3 film after CB:DIO treatment (Fig. 2g). Numerous contrast features with sizes on the order of 50 – 100 nm were observed, which indicates that solvent treatment of the pristine MoO3 electron blocking layer significantly degrades its structural uniformity. In contrast, solvent treatment did not significantly alter the morphology of annealed MoO3, as shown in Fig. 2h, indicating that the annealing prior to CB:CIO treatment makes the MoO3 film more robust and resistant to the solvent erosion. Both the AFM and STEM studies reveal that annealing step stabilizes the MoO3 structure and is thus crucial for the improved device performance. The MoO3 morphology and device performance annealed at different temperatures are shown in Supplementary Fig. S1. Based on these results, annealing temperature of 150°C was chosen by considering the solvent resistance of MoO3 film to CB:DIO and the glass transition temperature of PEN (155°C) substrate for flexible devices. Next, ultraviolet photoelectron spectroscopy (UPS) was performed to investigate the electronic structure of MoO3 film after the thermal and solvent treatment. Figure 2i-j shows the He I UPS spectra in its full scan, including the photoemission onset. For the pristine MoO3 film (Fig. 2i), the photoemission onset occurs at 15.97 eV corresponding to a work function of 5.25 eV, and the solvent treatment shifts the photoemission onset to a higher binding energy (16.43 eV) leading to downward-shift of the vacuum level and reduction in the work function (4.79 eV). Annealing the films does not cause significant changes in the UPS spectra with the photoemission onset of 15.94 eV (work function=5.28 eV) and 16.46 eV (work function=4.76
ACS Paragon Plus Environment
6
Page 7 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Nano Letters
eV) before and after the solvent treatment, respectively (Fig. 2j). For both as-evaporated and annealed cases, MoO3 is slightly n-doped after the CB:DIO spin-coating as a result of oxygen vacancies in the bulk of the material due to the adsorption of various species from the solvent onto MoO328. Despite the slight decrease in work function of both types of the film, the work function values still remain relatively high suitable for the hole transporting material. Therefore, we suggest that the degraded device performance (Fig. 1c, blue) primarily originates from the drastic morphology changes of MoO3 resulting upon the interaction with CB:DIO (Fig. 2a-h), which can be remedied via simple anneal treatment introduced in this work. In addition to the graphene anode-based devices discussed above, we also investigated inverted cathode-based PSC configuration that can provide improved device stability by avoiding easily oxidized low work function metal electrodes such as Al or Ca8, 29. In this structure, n-type semiconducting metal oxides such as ZnO or TiOx can be utilized as an effective electron transporting path from the photoactive polymer layer to the cathode. After achieving uniform coverage of the electron transporting ZnO layer directly on graphene surface by a simple spin-coating method (see Supplementary Fig. S2), graphene cathode-based inverted PSCs and their ITO-based counterparts were demonstrated with the following device structure: graphene (or ITO)/ZnO/PTB7:PC71BM/MoO3/Ag. The device structure, corresponding band diagram, and measured J–V characteristics are shown in Fig. 3. As illustrated in Fig. 3c, both graphene- and ITO-based inverted solar cells show great similarities in device performance with PCEs of 6.9% and 7.6%, respectively. We note that the inverted cathode-based PSCs fabricated on graphene electrodes without the ZnO layer exhibit negligible photoresponse (Supplementary Fig. S3), indicating the importance of electron transporting layer on the graphene cathode.
ACS Paragon Plus Environment
7
Nano Letters
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 29
With growing interests in flexible electronic devices30, as the final step we have explored the potential of our device structures to realize flexible graphene-based PSCs. We have realized both anode- and cathode-based device architectures on PEN substrates and tested their performance under mechanical tensile bending conditions. For that purpose, the devices were subjected to consecutive flexing cycles at 5 mm radius, corresponding to the strain of ~4.3%. The resulting graphene PSCs on PEN substrates show excellent device performance for both anode (PCE=6.1%, Fig. 4a and Supplementary Table S1) and cathode (PCE=7.1%, Fig. 4b and Supplementary Table S2) configurations. Our graphene-based flexible PSCs are robust under mechanical deformations, which is highly desirable for low-cost productions such as roll-to roll processing and applications that require flexibility. As shown in Fig. 4c, the device (graphene anode) did not display any significant performance changes up to 100 tensile flexing cycles. Key photovoltaic parameters at intermediate flexing cycles are shown in Fig.4e-f and Supplementary Table S3. We demonstrated highly efficient flexible graphene-based PSCs with record power conversion efficiency of 7.1%. The performances of our graphene-based devices are the highest reported in the literature: Figure 5 summarizes the device performance of graphene-based OSCs with various polymer acceptors reported in the literature along with our results for both normal (graphene anode) and inverted (graphene cathode) cells with low bandgap PTB7. The successful demonstration of both types of graphene electrode-based flexible devices in this work is an important milestone toward fully graphene electrode integrated flexible solar cells. This work is accomplished via thermal treatment of MoO3 hole transporting layer on graphene anode, which resolves the compatibility issue of pristine MoO3 with CB:DIO solvent for polymer deposition. Furthermore, we developed a simple spin-coating method for direct deposition of ZnO electron
ACS Paragon Plus Environment
8
Page 9 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Nano Letters
transporting layer on graphene for the cathode application. The demonstration of high-efficiency graphene-based flexible PSCs with remarkable mechanical robustness opens up a bright future for variety of graphene-based flexible optoelectronic devices.
Methods Graphene synthesis. Graphene films were synthesized via low pressure chemical vapor deposition on Cu foil (25 µm in thickness). The growth chamber was evacuated to a base pressure of ~50 mTorr and the Cu was annealed at 1000°C for 30 min under hydrogen gas (9 sccm, ~300 mTorr). Subsequently, methane gas (17 sccm, ~800 mTorr) was added and the graphene growth was performed for 30 min. Graphene sheets were transferred to quartz substrates using PMMA13 and the PMMA layer was consequently removed by acetone. Repeated transfers were performed to obtain three-layer graphene stacks. PEDOT:PSS modification for graphene film. PEDOT:PSS (Clevios™ P VP AI 4083) solution was modified by isopropyl alcohol via thorough mixing: after filtering the PEDOT::PSS (0.45 µm), it was mixed with IPA (3:1, v/v ratio) followed by 24 h of rigorous stirring at room temperature. This mixed solution was spin-coated in air at 4000 rpm for 60 s, and annealed at 170°C for 5 min in air. Solar cell fabrication. Pre-patterned ITO substrates (Thin Film Devices, 150 nm thick, 20 Ω/sq, 85%T) were cleaned by sonication in soap water, DI water, acetone and isopropanol, followed by oxygen plasma cleaning. Patterned graphene substrates were rinsed by acetone and IPA. ZnO layer was deposited by spin-coating 300 mM of zinc acetate dehydrate methanol solution followed by annealing at 175ºC for 10 min. Polymer blend solution was prepared by dissolving PTB7 (12 mg/ml, 1d-material) and PC71BM (40mg/ml, Sigma-Aldrich) in CB:DIO (97%:3% by
ACS Paragon Plus Environment
9
Nano Letters
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 29
volume) and mixing them at 2:1 volume ratio. The mixed solution was spin-coated at 1000 rpm for 2 min in nitrogen-filled glovebox. MoO3, Ag, Ca, and Al were thermally evaporated through the shadow mask at 1.0 Å/s. MoO3 was annealed at 150 ºC for 10 min. The device area (1.21 mm2) was defined by the overlap between the top and bottom electrodes. The substrate size was 12.5×12.5 mm for both ITO and graphene. Measurement and characterization. Current-voltage characteristics of the solar cells were recorded using a Keithley 6487 picoammeter source-meter in a nitrogen-filled glovebox. 100 mW/cm2 illumination was provided by a 150 W xenon arc-lamp (Newport 96000) equipped with an AM 1.5G filter. For the bending test of flexible graphene anode devices, J–V measurements were recorded after 20, 50, and 100 compressive flexing cycles at ~5 mm radius. The surface morphologies of MoO3 were characterized from Digital Instruments Veeco Dimension 3100 AFM operated in tapping mode. STEM images were obtained using a JEOL 2010F with an accelerating voltage of 200 kV. MoO3 (20 nm) films were deposited on rigid TEM grids with 30 nm SiNx windows, and were subjected to identical solvent and thermal treatment as the AFM analysis. UPS measurement. After thermally evaporating the MoO3 (20 nm) on Au coated glass substrate, the samples were transferred to UPS analysis chamber (
Communication
Flexible graphene electrode-based organic photovoltaics with record-high efficiency Hyesung Park, Sehoon Chang, Xiang Zhou, Jing Kong, Tomas Palacios, and Silvija Gradecak Nano Lett., Just Accepted Manuscript • DOI: 10.1021/nl501981f • Publication Date (Web): 20 Aug 2014 Downloaded from http://pubs.acs.org on August 22, 2014
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Nano Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Nano Letters
Flexible graphene electrode-based organic photovoltaics with record-high efficiency Hyesung Park1,2†, Sehoon Chang2†, Xiang Zhou2, Jing Kong1*, Tomás Palacios1*, Silvija Gradečak2* 1
Department of Electrical Engineering and Computer Science, Massachusetts Institute of
Technology, Cambridge, Massachusetts 02139 2
Department of Materials Science and Engineering, Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139
* Corresponding author email address: [email protected]; [email protected]; [email protected] †
These authors contributed equally to this work.
ACS Paragon Plus Environment
1
Nano Letters
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 29
Abstract Advancements in the field of flexible high-efficiency solar cells and other optoelectronic devices will strongly depend on the development of electrode materials with good conductivity and flexibility. To address chemical and mechanical instability of currently used indium tin oxide (ITO), graphene has been suggested as a promising flexible transparent electrode, but challenges remain in achieving high efficiency of graphene-based polymer solar cells (PSCs) compared to their ITO-based counterparts. Here we demonstrate graphene anode- and cathodebased flexible PSCs with record-high power conversion efficiencies of 6.1% and 7.1%, respectively. The high efficiencies were achieved via thermal treatment of MoO3 electron blocking layer and direct deposition of ZnO electron transporting layer on graphene. We also demonstrate graphene-based flexible PSCs on polyethylene naphthalate substrates and show the device stability under different bending conditions. Our work paves a way to fully graphene electrode-based flexible solar cells using a simple and reproducible process.
Keywords: flexible solar cell, polymer solar cell, graphene anode, graphene cathode
ACS Paragon Plus Environment
2
Page 3 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Nano Letters
Flexible organic and hybrid organic-inorganic solar cells require both the flexible photoactive media1, 2 and flexible electrodes with good conductivity and transparency. So far, indium tin oxide (ITO) has been an electrode material of choice for studies focusing on optimizing the morphology and chemistry of the photoactive media3 in polymer solar cells (PSCs). For flexible applications, graphene has been proposed as a promising replacement for ITO due to its mechanical and chemical robustness, excellent electrical and optical properties, and potentially low-cost processing4, 5. Recent studies have demonstrated dramatic improvements in the efficiency of PSCs6, 7 that ascertain the bright future towards PCE≥10%, a threshold considered for industrial applications2. In PSCs, and more generally organic solar cells (OSCs), one of the electrodes typically consists of a transparent conductor, among which ITO is most widely used due to its good optical transparency and electrical conductivity. However, due to its non-uniform absorption, chemical and mechanical instability, as well as high cost of indium8, several alternative materials have been proposed, including carbon nanotube9, 10 or metallic nanowires networks11. Furthermore, even at small mechanical stresses, microcracks are produced in ITO resulting in increased film resistance and decreased device performance12, thus limiting its applications for flexible OSCs. Owing to the unique optoelectronic properties of graphene5, several works on graphenebased OSCs have been reported, demonstrating the feasibility of graphene in transparent electrode applications3, 12-26. While the initial demonstrations have been promising, performance of graphene-based devices still falls short (< 3%) of recent advances accomplished in ITO-based devices (8 – 9%)6, 7. Therefore, to demonstrate graphene as an emerging alternative to ITO, it is inevitable to improve the currently low-efficiency of graphene-based solar cells. Moreover, to test feasibility of graphene for flexible applications, high performance and stability of such
ACS Paragon Plus Environment
3
Nano Letters
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 29
devices must be investigated. In this work, we demonstrate high-efficiency graphene anode- and cathode-based PSCs with PCEs comparable to their ITO-counterparts. We also demonstrate graphene-based flexible PSCs on polyethylene naphthalate (PEN) substrates and show the device stability under different bending conditions. The overall PSC device structure used in this work and the corresponding band diagram are shown in Fig. 1a-b. To fabricate efficient graphene-based OSCs, we used photoactive media composed of a blend of low bandgap semiconducting polymer donor thieno[3,4b]thiophene/benzodithiophene (PTB7) and acceptor [6,6]-phenyl C71-butyric acid methyl ester (PC71BM) prepared using mixed solvents of chlorobenzene:1,8-diiodoctane (CB:DIO, 97:3 vol%)6. The hole injection layer poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) was deposited on the transparent graphene electrode. To ensure uniform coverage over the graphene surface, we used modified PEDOT:PSS with isopropyl alcohol (IPA) at 3:1 (v/v) ratio. Prior to the active device layer deposition, graphene/PEDOT:PSS must be covered by an additional electron blocking layer (MoO3). This process is critical in graphenebased bulk heterojunction PSCs since the additional MoO3 layer prevents charge recombination occurring at the graphene/PEDOT:PSS and the polymer blend interface27. We also note that although ITO does not require additional MoO3 layer, ITO control devices were fabricated with PEDOT:PSS/MoO3 for direct comparison with graphene-based devices. We discovered that one critical aspect of the device fabrication is the thermal annealing of MoO3 layer before spin-coating PTB7:PC71BM. When directly applying PTB7:PC71BM to PEDOT:PSS/MoO3 on ITO or graphene without thermal annealing, considerable degradation in device performance was observed, which did not occur on PEDOT:PSS only device. Fig.1c presents the current density–voltage (J–V) characteristics of PSCs fabricated on ITO substrates,
ACS Paragon Plus Environment
4
Page 5 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Nano Letters
in which addition of MoO3 layer (non-annealed) results in detrimental effects on the shortcircuit current density (JSC, 2.6 mA/cm2), open-circuit voltage (VOC, 0.57 V), and fill factor (FF, 31.3%). However, significant improvements in JSC (16.1 mA/cm2), VOC (0.68 V), FF (60.7%), and the resulting increase in PCE were observed after the MoO3 electron blocking layer thermal annealing (Fig. 1d and Supplementary Table S1). Graphene-based PSCs were fabricated under the same experimental conditions and the resulting J–V characteristics are compared with that of ITO reference device in Fig. 1d. The graphene electrode was made by stacking three monolayers of graphene film prepared by low pressure chemical vapor deposition, with a typical sheet resistance of ~300 Ω/sq and transmittance of ~92% at λ = 550 nm13. With incorporation of appropriate PEDOT:PSS and thermally-treated MoO3, we observed record-high efficiency from graphene (PCE=6.1%) approaching to ITO reference device (PCE=6.7% ) (Supplementary Table S1). Since the efficient graphene PSC with PTB7:PC71BM was accomplished only in the presence of MoO3 electron blocking layer with subsequent thermal treatment, we investigated the effect of CB:DIO solvent on MoO3 via several routes. First, the surface morphology of MoO3 film (20 nm, on glass) was characterized using atomic force microscopy (AFM) after the thermal and solvent treatments (Fig. 2a-d). Figure 2a shows the surface topography of as-evaporated MoO3 film is smooth with root-mean-square (rms) roughness of ~1 nm. The thermal treatment alone does not affect significantly topography of MoO3 (Fig. 2b). However, noticeable change in the surface morphology was observed after spin-coating CB:DIO on non-annealed MoO3 (Fig. 2c), indicative of adverse effects of the solvent. In contrast, after thermal treatment and subsequent CB:DIO spin-coating on MoO3, the surface profile (Fig. 2d) is similar to that of annealed MoO3 (Fig. 2b) indicating that annealed MoO3 film becomes robust upon the solvent treatment.
ACS Paragon Plus Environment
5
Nano Letters
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 29
We further characterized the effect of solvent treatment on MoO3 via scanning transmission electron microscopy (STEM), as shown in Fig. 2e-h, which revealed similar structural changes. STEM image of as-deposited MoO3 is largely featureless (Fig. 2e) with minimum contrast differences indicating a uniform structure. Significant difference is observed from the asdeposited MoO3 film after CB:DIO treatment (Fig. 2g). Numerous contrast features with sizes on the order of 50 – 100 nm were observed, which indicates that solvent treatment of the pristine MoO3 electron blocking layer significantly degrades its structural uniformity. In contrast, solvent treatment did not significantly alter the morphology of annealed MoO3, as shown in Fig. 2h, indicating that the annealing prior to CB:CIO treatment makes the MoO3 film more robust and resistant to the solvent erosion. Both the AFM and STEM studies reveal that annealing step stabilizes the MoO3 structure and is thus crucial for the improved device performance. The MoO3 morphology and device performance annealed at different temperatures are shown in Supplementary Fig. S1. Based on these results, annealing temperature of 150°C was chosen by considering the solvent resistance of MoO3 film to CB:DIO and the glass transition temperature of PEN (155°C) substrate for flexible devices. Next, ultraviolet photoelectron spectroscopy (UPS) was performed to investigate the electronic structure of MoO3 film after the thermal and solvent treatment. Figure 2i-j shows the He I UPS spectra in its full scan, including the photoemission onset. For the pristine MoO3 film (Fig. 2i), the photoemission onset occurs at 15.97 eV corresponding to a work function of 5.25 eV, and the solvent treatment shifts the photoemission onset to a higher binding energy (16.43 eV) leading to downward-shift of the vacuum level and reduction in the work function (4.79 eV). Annealing the films does not cause significant changes in the UPS spectra with the photoemission onset of 15.94 eV (work function=5.28 eV) and 16.46 eV (work function=4.76
ACS Paragon Plus Environment
6
Page 7 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Nano Letters
eV) before and after the solvent treatment, respectively (Fig. 2j). For both as-evaporated and annealed cases, MoO3 is slightly n-doped after the CB:DIO spin-coating as a result of oxygen vacancies in the bulk of the material due to the adsorption of various species from the solvent onto MoO328. Despite the slight decrease in work function of both types of the film, the work function values still remain relatively high suitable for the hole transporting material. Therefore, we suggest that the degraded device performance (Fig. 1c, blue) primarily originates from the drastic morphology changes of MoO3 resulting upon the interaction with CB:DIO (Fig. 2a-h), which can be remedied via simple anneal treatment introduced in this work. In addition to the graphene anode-based devices discussed above, we also investigated inverted cathode-based PSC configuration that can provide improved device stability by avoiding easily oxidized low work function metal electrodes such as Al or Ca8, 29. In this structure, n-type semiconducting metal oxides such as ZnO or TiOx can be utilized as an effective electron transporting path from the photoactive polymer layer to the cathode. After achieving uniform coverage of the electron transporting ZnO layer directly on graphene surface by a simple spin-coating method (see Supplementary Fig. S2), graphene cathode-based inverted PSCs and their ITO-based counterparts were demonstrated with the following device structure: graphene (or ITO)/ZnO/PTB7:PC71BM/MoO3/Ag. The device structure, corresponding band diagram, and measured J–V characteristics are shown in Fig. 3. As illustrated in Fig. 3c, both graphene- and ITO-based inverted solar cells show great similarities in device performance with PCEs of 6.9% and 7.6%, respectively. We note that the inverted cathode-based PSCs fabricated on graphene electrodes without the ZnO layer exhibit negligible photoresponse (Supplementary Fig. S3), indicating the importance of electron transporting layer on the graphene cathode.
ACS Paragon Plus Environment
7
Nano Letters
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 29
With growing interests in flexible electronic devices30, as the final step we have explored the potential of our device structures to realize flexible graphene-based PSCs. We have realized both anode- and cathode-based device architectures on PEN substrates and tested their performance under mechanical tensile bending conditions. For that purpose, the devices were subjected to consecutive flexing cycles at 5 mm radius, corresponding to the strain of ~4.3%. The resulting graphene PSCs on PEN substrates show excellent device performance for both anode (PCE=6.1%, Fig. 4a and Supplementary Table S1) and cathode (PCE=7.1%, Fig. 4b and Supplementary Table S2) configurations. Our graphene-based flexible PSCs are robust under mechanical deformations, which is highly desirable for low-cost productions such as roll-to roll processing and applications that require flexibility. As shown in Fig. 4c, the device (graphene anode) did not display any significant performance changes up to 100 tensile flexing cycles. Key photovoltaic parameters at intermediate flexing cycles are shown in Fig.4e-f and Supplementary Table S3. We demonstrated highly efficient flexible graphene-based PSCs with record power conversion efficiency of 7.1%. The performances of our graphene-based devices are the highest reported in the literature: Figure 5 summarizes the device performance of graphene-based OSCs with various polymer acceptors reported in the literature along with our results for both normal (graphene anode) and inverted (graphene cathode) cells with low bandgap PTB7. The successful demonstration of both types of graphene electrode-based flexible devices in this work is an important milestone toward fully graphene electrode integrated flexible solar cells. This work is accomplished via thermal treatment of MoO3 hole transporting layer on graphene anode, which resolves the compatibility issue of pristine MoO3 with CB:DIO solvent for polymer deposition. Furthermore, we developed a simple spin-coating method for direct deposition of ZnO electron
ACS Paragon Plus Environment
8
Page 9 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Nano Letters
transporting layer on graphene for the cathode application. The demonstration of high-efficiency graphene-based flexible PSCs with remarkable mechanical robustness opens up a bright future for variety of graphene-based flexible optoelectronic devices.
Methods Graphene synthesis. Graphene films were synthesized via low pressure chemical vapor deposition on Cu foil (25 µm in thickness). The growth chamber was evacuated to a base pressure of ~50 mTorr and the Cu was annealed at 1000°C for 30 min under hydrogen gas (9 sccm, ~300 mTorr). Subsequently, methane gas (17 sccm, ~800 mTorr) was added and the graphene growth was performed for 30 min. Graphene sheets were transferred to quartz substrates using PMMA13 and the PMMA layer was consequently removed by acetone. Repeated transfers were performed to obtain three-layer graphene stacks. PEDOT:PSS modification for graphene film. PEDOT:PSS (Clevios™ P VP AI 4083) solution was modified by isopropyl alcohol via thorough mixing: after filtering the PEDOT::PSS (0.45 µm), it was mixed with IPA (3:1, v/v ratio) followed by 24 h of rigorous stirring at room temperature. This mixed solution was spin-coated in air at 4000 rpm for 60 s, and annealed at 170°C for 5 min in air. Solar cell fabrication. Pre-patterned ITO substrates (Thin Film Devices, 150 nm thick, 20 Ω/sq, 85%T) were cleaned by sonication in soap water, DI water, acetone and isopropanol, followed by oxygen plasma cleaning. Patterned graphene substrates were rinsed by acetone and IPA. ZnO layer was deposited by spin-coating 300 mM of zinc acetate dehydrate methanol solution followed by annealing at 175ºC for 10 min. Polymer blend solution was prepared by dissolving PTB7 (12 mg/ml, 1d-material) and PC71BM (40mg/ml, Sigma-Aldrich) in CB:DIO (97%:3% by
ACS Paragon Plus Environment
9
Nano Letters
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 29
volume) and mixing them at 2:1 volume ratio. The mixed solution was spin-coated at 1000 rpm for 2 min in nitrogen-filled glovebox. MoO3, Ag, Ca, and Al were thermally evaporated through the shadow mask at 1.0 Å/s. MoO3 was annealed at 150 ºC for 10 min. The device area (1.21 mm2) was defined by the overlap between the top and bottom electrodes. The substrate size was 12.5×12.5 mm for both ITO and graphene. Measurement and characterization. Current-voltage characteristics of the solar cells were recorded using a Keithley 6487 picoammeter source-meter in a nitrogen-filled glovebox. 100 mW/cm2 illumination was provided by a 150 W xenon arc-lamp (Newport 96000) equipped with an AM 1.5G filter. For the bending test of flexible graphene anode devices, J–V measurements were recorded after 20, 50, and 100 compressive flexing cycles at ~5 mm radius. The surface morphologies of MoO3 were characterized from Digital Instruments Veeco Dimension 3100 AFM operated in tapping mode. STEM images were obtained using a JEOL 2010F with an accelerating voltage of 200 kV. MoO3 (20 nm) films were deposited on rigid TEM grids with 30 nm SiNx windows, and were subjected to identical solvent and thermal treatment as the AFM analysis. UPS measurement. After thermally evaporating the MoO3 (20 nm) on Au coated glass substrate, the samples were transferred to UPS analysis chamber (
