Control of the oxidation level of graphene oxide for high efficiency polymer solar cells

Rui Wua, Ying Wanga, Lie Chenab, Liqiang Huanga and Yiwang Chen*ab
aCollege of Chemistry/Institute of Polymers, Nanchang University, 999 Xuefu Avenue, Nanchang 330031, China. E-mail: ywchen@ncu.edu.cn
bJiangxi Provincial Key Laboratory of New Energy Chemistry, Nanchang University, 999 Xuefu Avenue, Nanchang 330031, China

Received 3rd February 2015 , Accepted 7th May 2015

First published on 7th May 2015


Abstract

Graphene oxides (GOs) have been used as interfacial layers for fabricating more stable organic solar cells (OSCs). However, the influence of the degree of oxidation of GOs on their optoelectronic properties has been ignored. In this article, a series of GOs with different degrees of oxidation were successfully synthesized, by controlling the amount of oxidant KMnO4 during the oxidation process of graphite. With increasing oxidation level, more oxygenated functional groups were attached to the carbon basal plane and more defects were introduced into the GO sheets, resulting in an increased work function (WF) and a decreased conductivity. Meanwhile, the film-forming property of GOs was improved with increasing oxidation level, which is attributed to the adequate exfoliation of the GO sheets. After carefully controlling the oxidation level of GOs, the OSCs with GOs as the hole transport layer (HTL) show an efficiency value of 3.0%, comparable to that with poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) (3.2%), originating from the good film-forming property, appropriate work function and high conductivity.


Introduction

Organic solar cells (OSCs) represent an exciting class of renewable energy technology. They are under intensive investigation in both academia and industry due to their great potential to enable mass manufacture of flexible and low-cost devices through roll-to-roll techniques.1 Power conversion efficiencies (PCE) of OSCs have increased continuously over the last few years. The PCE of OSCs has exceeded 10% for a single cell2 and 11% for a tandem cell.3 In bulk heterojunction OSC devices, the interface between the active layer and the electrodes play an important role for their high efficiency and stability.4 In order to improve the efficiency and stability of OSCs, hole transport layer (HTL) and electron transport layer (ETL) materials are used at the anode interface and cathode interface, to minimize carrier recombination loss and current leakage at the interface between electrodes and semiconducting materials.5

For the anode interface, PEDOT:PSS is the most commonly used HTL material. Several problems of the PEDOT:PSS include high acidity and hygroscopic properties, resulting in poor long-term stability of OSCs. Li et al.6 first reported that graphene oxide (GO) can be used as a hole transport layer material for fabricating more stable OSCs, potentially replacing PEDOT:PSS. GO is a graphene sheet functionalized with oxygen groups including epoxy, hydroxyl, carboxyl and carbonyl.7 The availability of reactive carboxyl and epoxy/hydroxyl groups of GO sheets facilitates further functionalization of GO, permitting tunability of optoelectronic properties.8 Besides, GO can be processed in solution at a large scale with low cost, particularly attractive for massive applications. Due to the advantages of GO, it is a promising interface modification material for OSCs. Over the past few years, there has been significant progress in the application of GO in OSCs.9 Both pristine GO and its derivatives as interfacial layers of OSCs generally need a change in work function through functionalization or doping with molecules, to favor an energy alignment in the devices.10 Stratakis et al.11 demonstrated that GO’s work function can be tuned when treated with ultraviolet laser and chlorine gas. Similarly, sulfated12 and fluorine-functionalized GO13 with increased work function were also used as the HTL in OSCs. Moreover, the work function of GO can be effectively reduced, triggering its use as an ETL.14 Besides, the use of conductive fillers to improve the conductivity of GO has also been investigated.15

Actually, GOs synthesized by different methods and conditions have different degrees of oxidation. However, in recent years, most efforts have mainly focused on how to improve the performance of the GO by functionalization or doping with other molecules, but the effect of the modulation of the degree of oxidation of GO on its optoelectronic properties has been ignored. Clearly, it is a critical issue for further applications of GO in OSCs. In this study, a series of graphene oxides (GOs) with different degrees of oxidation were successfully synthesized, by carefully controlling the amount of oxidant KMnO4 during the oxidation process of graphite, and the effect of the oxidation level of GO on their optoelectronic properties has been systematically studied.

Experimental

Preparation of graphene oxide (GO)

GOs were prepared by oxidation of graphite powder according to a modified Hummer’s method.16 Briefly, graphite (1.0 g) was added to concentrated sulfuric acid (25 ml) under stirring at room temperature, then sodium nitrate (0.5 g) was added, and the mixture was cooled to 0 °C. Under violent agitation, potassium permanganate (0.5 g, 50 wt%) was added slowly to keep the temperature of the reaction system lower than 20 °C. Then, the suspension was transferred to a 35 °C water bath and vigorously stirred for about 1.0 h. Subsequently, 50 ml of water was added, and the solution was stirred for 15 min at 90 °C. An additional 180 ml of water was added and followed by a slow addition of 5 ml of H2O2 (3%), turning the color of the suspension from dark brown to yellow. The solution was filtered and washed with 1[thin space (1/6-em)]:[thin space (1/6-em)]10 HCl aqueous solution (100 ml) to remove metal ions followed by washing with 100 ml of water to remove the acid. In the end, it was purified by dialysis for several days to remove the remaining metal species. A similar process was followed for GOs with different degrees of oxidation by adding 0.5 g (50 wt%), 1 g (100 wt%), 4 g (400 wt%) and 6 g (600 wt%) of KMnO4.

Characterization

The resultant GOs with different degrees of oxidation were investigated by Raman spectroscopy (LabRam-1B), Fourier transform infrared spectroscopy (FT-IR Prestige-21), X-ray photoelectron spectroscopy (Thermo-VG Scientific ESCALAB 250) and wide-angle X-ray diffraction (Bruker D-8). Transmittance spectra were analyzed by UV-vis spectroscopy (PerkinElmer Lambda 750). The morphology of the GOs was characterized by using scanning electron microscopy (QuanTA-200F environmental scanning electron microscope), atomic force microscopy (Agilent 5500) and transmission electron microscopy (JEOL, JEM-2100F). Ultraviolet photoelectron spectroscopy (UPS) was carried out using an AXISULTRA DLD spectrometer (Kratos Analytical Ltd.). The current–voltage characteristics of devices under illumination were tested by a Keithley 2400 Source Meter. The light intensity was 100 mW cm−2.

Device fabrication

All the devices were fabricated with the structure of ITO/HTL/P3HT:PCBM/LiF/Al. The ITO substrates were cleaned by ultrasonication in soap water, deionized water, acetone and isopropanol. After drying the ITO substrates and treating the surface with UV ozone for 15 minutes, the interface layers of the devices were deposited on the ITO substrates using the following spin-coating conditions. The PEDOT:PSS layer was spin-coated from the solution at 4000 rpm for 60 s, followed by heating at 140 °C for 15 min. The GO layers were spin-coated from their solution in water (0.5 mg ml−1) at 2000 rpm for 60 s, followed by heating at 140 °C for 10 min. The active layer was sequentially spin-coated from the solution of P3HT/PCBM = 1/1 in o-dichlorobenzene (20 mg ml−1) at 800 rpm for 30 s, followed by thermal annealing at 150 °C for 10 min. After spin-coating of the organic layers, the devices were transferred into a vacuum chamber for thermal deposition of LiF (0.7 nm) and Al (100 nm) at a pressure of 10−7 Torr. The surface area of each device was 0.04 cm2, as determined by the overlap of the ITO and Al.

Results and discussion

We synthesized four GOs with different degrees of oxidation, which were named GO (50 wt% KMnO4), GO (100 wt% KMnO4), GO (400 wt% KMnO4) and GO (600 wt% KMnO4), according to the different added amounts of oxidant KMnO4. Fig. 1a shows the X-ray photoelectron spectroscopy (XPS) survey spectra of the four GOs. Two intense characteristic peaks of C and O are observed. Fig. 1b shows the high-resolution XPS C1s spectra. With increasing amount of oxidant KMnO4, the oxidation level of GO increases, as evidenced by the decrease in C/O.17 The resultant Fourier transform infrared spectroscopy (FT-IR) of the prepared GOs with different degrees of oxidation is shown in Fig. 2a. The band at about 1579 cm−1 is due to the presence of C–C stretching in graphitic domains. With increasing oxidation level, the FT-IR spectrum reveals more oxygenated functional groups, including C–OH (∼1399 cm−1), C–O–C (∼1249 cm−1), C[double bond, length as m-dash]O (∼1727 cm−1), and C–O (∼1049 cm−1).18 It indicates that more oxygenated functional groups were attached to the carbon basal plane. Then, GOs were analyzed by Raman spectroscopy, as shown in Fig. 2b. Two prominent peaks were observed in the GO spectra, which correspond to D-peak and G-peak. The D peak exhibits the defect nature of GO, while the G peak is the characteristic peak of the sp2 carbon atom vibration. And the ratio (ID/IG) of the intensity of the D peak (ID) and G peak (IG) increased with the increase in oxidation level. The reason is that there were more defects in the GO sheet with the addition of oxygenated functional groups. Fig. 3 shows the X-ray diffraction (XRD) patterns of the GOs with different degrees of oxidation. The XRD pattern of the original graphite shows a diffraction peak at 2θ = 26.5° associated with an interlayer spacing of about 3.36 Å. However, with increasing degree of oxidation, the intensity of this peak starts to decrease, and finally disappears at higher oxidation levels. Meanwhile, it also can be observed that a new peak starts to appear at a lower angle, corresponding to the diffraction pattern of GOs. The interlayer spacing (d) of the GO increases from 6.81 Å to 7.90 Å as the oxidation level increases (Table S1), due to the addition of oxygenated functional groups weakening the π–π stacking between GO sheets.19
image file: c5ra02099a-f1.tif
Fig. 1 (a) XPS survey spectra of GOs with different degrees of oxidation (50 wt% KMnO4, 100 wt% KMnO4, 400 wt% KMnO4 and 600 wt% KMnO4 represent the input of KMnO4 corresponding to 1 g graphite). (b) High-resolution XPS C1s spectra of GOs with different degrees of oxidation (C/O represents the ratio of C and O in GOs with different degrees of oxidation).

image file: c5ra02099a-f2.tif
Fig. 2 (a) Fourier transform infrared spectra and (b) Raman spectra of GOs with different degrees of oxidation such as 50 wt% KMnO4, 100 wt% KMnO4, 400 wt% KMnO4 and 600 wt% KMnO4.

image file: c5ra02099a-f3.tif
Fig. 3 X-ray diffraction patterns of GOs with different degrees of oxidation such as 50 wt% KMnO4, 100 wt% KMnO4, 400 wt% KMnO4 and 600 wt% KMnO4.

Interfacial morphology and its contact with the active layer are crucial for OSC performance, thus the effect of the degree of oxidation on the morphology and surface wettability of the GO has been investigated. The surface contact angles of the GOs with different degrees of oxidation are shown in Fig. S1. Attaching the hydrophilic oxygenated functional groups on graphene results in a decrease in the surface contact angle of GOs, especially in the GO with a high degree of oxidation. However, compared to the surface contact angle of PEDOT:PSS (16°, Fig. S1e), these GOs still keep a moderate surface wettability with a surface contact angle of 60–70°, which can improve the interface compatibility of the active layer and HTL in OSCs. An atomic force microscope (AFM) was then used to analyze the surface morphology of the GOs with different degrees of oxidation. The results show that all samples of GOs have similar roughness values (Fig. S2). However, the difference in morphology caused by the degree of oxidation can be clearly detected by transmission electron microscopy (TEM), as shown in Fig. 4. It is obvious that all the samples of GOs have a lamellar morphology. GO (50 wt% KMnO4, Fig. 4a) shows serious aggregation, due to the partially oxidized graphene oxide leading to unsuccessful exfoliation. With increasing degree of oxidation, the dispersion of GOs is improved greatly, and the transparent GO sheet can be obviously observed, implying adequate exfoliation of the samples. This observation is also correlated with the scanning electron microscopy (SEM) images in Fig. S3. Compared with the GO with a low degree of oxidation containing closely aggregated graphene oxide sheet (50 wt% KMnO4, Fig. S3a), the other three GOs (100 wt% KMnO4, Fig. S3b, 400 wt% KMnO4, Fig. S3c and 600 wt% KMnO4, Fig. S3d) show much smoother surfaces, which is in favor of the film-forming process of following active layer and the formation of a stable interface.


image file: c5ra02099a-f4.tif
Fig. 4 Transmission electron microscopy images of GOs with different degrees of oxidation. (a) 50 wt% KMnO4, (b) 100 wt% KMnO4, (c) 400 wt% KMnO4 and (d) 600 wt% KMnO4.

Prior to employing GOs with different degrees of oxidation as the hole transport layers in OSCs, the work functions of the GOs were measured with ultraviolet photoelectron spectroscopy (UPS) (Fig. 5a). The work functions of GOs coated on ITO glass with different degrees of oxidation can be calculated from the following equation.20

WF = EF + Ecutoff
where is the He(I) excitation energy equal to 21.2 eV, EF is the Fermi level, and Ecutoff is the high-binding energy cutoff. The work functions of GOs oxidized from 50 wt%, 100 wt%, 400 wt% and 600 wt% KMnO4 were 4.8 eV, 4.9 eV, 5.1 eV and 5.1 eV, respectively. These values were higher than the typical value (4.6 eV) obtained for pristine graphene. The higher values for the GOs are due to the surface Cδ+–Oδ dipoles. The induced polar character of Cδ+–Oδ bonds is responsible for the downward shift of the Fermi level of GOs, and the subsequent increase in work function from 4.6 eV to 5.1 eV. With increasing number of oxygenated functional groups in the base plane of carbon, the intensity of the surface Cδ+–Oδ dipoles is enhanced, resulting in the further rise of the GO work functions. Obviously, the WF of oxidized GOs from 400 wt% and 600 wt% KMnO4 matches the HOMO level of P3HT better (5.1 eV vs. 5.2 eV),21 as shown in Fig. 5b, thus the improved energy alignment is preferable to lower the Schottky barrier and form an ohmic contact at the ITO/active layer interface for charge extraction and collection.22 The existence of the oxidized groups is expected to impact greatly on the conductivity of the GO, so the conductivity of the GO layers with different degrees of oxidation was also evaluated with the device structure of ITO/GO/Al (Fig. 6a). As expected, due to the growth of the interlayer spacing (Table S1) and increase in the number of GO sheet defects (Fig. 2b), the conductivity of the GO layers declines with increasing oxidation level.


image file: c5ra02099a-f5.tif
Fig. 5 (a) Photoemission cutoff obtained via UPS for GOs with different degrees of oxidation. (b) The energy levels of all the materials used in the OSC cells.

image file: c5ra02099a-f6.tif
Fig. 6 (a) IV characteristics of GOs with different degrees of oxidation (the inset represents the structure of the device). With increasing oxidation level, the conductivity of the GOs with different degrees of oxidation decreases. (b) The structure of the OSC device. (c) JV characteristics of P3HT:PCBM photovoltaic devices with different HTLs.

To determine the performance of GOs with different degrees of oxidation as hole transport layers in OSCs, OSC devices with the configuration of ITO/HTL/P3HT:PCBM (200 nm)/LiF (0.7 nm)/Al (100 nm) were fabricated (Fig. 6b). Prior to it, the transmittance of the GOs was investigated. Fig. S4 shows that all the GO samples have favorable transmittance (>80%), indicating that the GOs can be used as HTL materials. Fig. 6c shows the illuminated current density–voltage (JV) curves of the P3HT:PCBM-based OSC devices with GOs and PEDOT:PSS as the HTL. The detailed photovoltaic characteristics are summarized in Table 1. The device with PEDOT:PSS HTL exhibits an average open-circuit voltage (VOC) of 0.60 V, short-circuit current density (JSC) of 8.32 mA cm−2, fill factor (FF) of 0.604 and PCE of 3.1%, which is consistent with reported values.23 Meanwhile, the device with GO (400 wt% KMnO4) exhibits an average VOC of 0.56 V, JSC of 7.71 mA cm−2, and FF of 0.636, leading to a PCE of 2.7%, and the best efficiency value (3.0%) of the device with GO (400 wt% KMnO4) as the HTL is comparable to that with PEDOT:PSS (3.2%). Comparing the performance of the four GOs as HTLs, a clear trend of the power conversion efficiency with increasing oxidation level can be observed. For the first three GOs (50 wt% KMnO4, 100 wt% KMnO4 and 400 wt% KMnO4), with increasing oxidation level, the average VOC increases from 0.34 V to 0.56 V and the FF increases from 0.334 to 0.636, yielding a PCE from 0.9% to 2.7%. The enhanced VOC with increasing oxidation level can be attributed to gradually increased work functions of the GOs. Obviously, the 5.1 eV of GO (400 wt% KMnO4) is almost equal to the HOMO level of P3HT (5.2 eV), leading to the largest VOC (0.56 V). At the same time, the much more homogeneous dispersion and smoother surface of GO sheets (400 wt% KMnO4) induces improved FF, as depicted by SEM and TEM observation. However, when the amount of KMnO4 increases to 600 wt%, the PCE together with all of the parameters of the device with GO (600 wt% KMnO4) drop obviously, probably owing to its significant decrease in conductivity (Fig. 6a). From the results we can see that careful control of the degree of oxidation of GO can fine-tune the optoelectronic and film-forming properties, such as work function, conductivity and morphology, consequently leading to satisfactory device performances.

Table 1 Summary of the photovoltaic parameters of the fabricated OSCsa
HTL JSC [mA cm−2] VOC [V] FF [%] PCE [%]
a Key: the structure of the OSCs is ITO/HTL/P3HT:PCBM/LiF/Al. All values represent an average from fourteen devices on a single chip.b The best device PCE.
PEDOT:PSS 8.32 ± 0.22 0.60 ± 0.01 60.4 ± 2.8 3.1 ± 0.1(3.2)b
GO (50 wt% KMnO4) 7.93 ± 0.36 0.34 ± 0.02 33.4 ± 3.0 0.9 ± 0.1(1.1)b
GO (100 wt% KMnO4) 7.69 ± 0.35 0.41 ± 0.02 52.2 ± 3.7 1.7 ± 0.1(2.0)b
GO (400 wt% KMnO4) 7.71 ± 0.22 0.56 ± 0.01 63.6 ± 2.7 2.7 ± 0.2(3.0)b
GO (600 wt% KMnO4) 7.67 ± 0.34 0.54 ± 0.01 61.8 ± 1.8 2.5 ± 0.1(2.7)b


Conclusions

In conclusion, a series of GOs with different degrees of oxidation have been successfully prepared, and the effect of GO oxidation level on their optoelectronic properties has also been systematically studied. With increasing oxidation level, more oxygenated functional groups were attached to the carbon basal plane and more defects were introduced into the GO sheets, resulting in an increasing work function (from 4.8 eV to 5.1 eV) and decreasing conductivity. Meanwhile, the film-forming property was improved with increasing oxidation level, which is attributed to a more adequate exfoliation of the GO sheets. Due to its good film-forming property, appropriate work function and high enough conductivity, the OSC device with the GO (400 wt% KMnO4) as the HTL showed the best PCE (3.0%), which is comparable with the PEDOT:PSS device (3.2%). In view of the significant effect of the oxidation level on GO optoelectronic properties including film-forming property, work function and conductivity, this work indicates that controlling oxidation levels plays a key role in achieving high performance OSCs with GO HTLs, and it would promote further studies and applications of GO in OSCs.

Acknowledgements

This work was financially supported by the National Science Fund for Distinguished Young Scholars (51425304), National Natural Science Foundation of China (51273088, 51263016 and 51473075), and National Basic Research Program of China (973 Program 2014CB260409).

References

  1. X. Hu, L. Chen, Y. Zhang, Q. Hu, J. Yang and Y. Chen, Chem. Mater., 2014, 26(21), 6293–6302 CrossRef CAS.
  2. J. D. Chen, C. Cui, Y. Q. Li, L. Zhou, Q. D. Ou, C. Li, Y. Li and J. X. Tang, Adv. Mater., 2015, 27, 1035–1041 CrossRef CAS PubMed.
  3. C. C. Chen, W. H. Chang, K. Yoshimura, K. Ohya, J. You, J. Gao, Z. Hong and Y. Yang, Adv. Mater., 2014, 26(32), 5670–5677 CrossRef CAS PubMed.
  4. H.-L. Yip and A. K. Y. Jen, Energy Environ. Sci., 2012, 5(3), 5994 CAS.
  5. R. Po, C. Carbonera, A. Bernardi and N. Camaioni, Energy Environ. Sci., 2011, 4(2), 285 CAS.
  6. S. S. Li, K. H. Tu, C. C. Lin, C. W. Chen and M. Chhowalla, ACS Nano, 2010, 4, 3169–3174 CrossRef CAS PubMed.
  7. Y. Gao, H. L. Yip, K. S. Chen, K. M. O’Malley, O. Acton, Y. Sun, G. Ting, H. Chen and A. K. Jen, Adv. Mater., 2011, 23(16), 1903–1908 CrossRef CAS PubMed.
  8. J. Liu, M. Durstock and L. Dai, Energy Environ. Sci., 2014, 7(4), 1297 CAS.
  9. (a) D. H. Wang, J. K. Kim, J. H. Seo, I. Park, B. H. Hong, J. H. Park and A. J. Heeger, Angew. Chem., Int. Ed., 2013, 52(10), 2874–2880 CrossRef CAS PubMed; (b) Y. Gao, H.-L. Yip, S. K. Hau, K. M. O’Malley, N. C. Cho, H. Chen and A. K. Y. Jen, Appl. Phys. Lett., 2010, 97(20), 203306 CrossRef PubMed.
  10. (a) D. Yang, L. Zhou, L. Chen, B. Zhao, J. Zhang and C. Li, Chem. Commun., 2012, 48(65), 8078–8080 RSC; (b) D. Yang, L. Zhou, W. Yu, J. Zhang and C. Li, Adv. Energy Mater., 2014, 4(15) DOI:10.1002/aenm.201400591.
  11. E. Stratakis, K. Savva, D. Konios, C. Petridis and E. Kymakis, Nanoscale, 2014, 6(12), 6925–6931 RSC.
  12. J. Liu, Y. Xue and L. Dai, J. Phys. Chem. Lett., 2012, 3(14), 1928–1933 CrossRef CAS.
  13. S. H. Kim, C. H. Lee, J. M. Yun, Y. J. Noh, S. S. Kim, S. Lee, S. M. Jo, H. I. Joh and S. I. Na, Nanoscale, 2014, 6(13), 7183–7187 RSC.
  14. J. Liu, Y. Xue, Y. Gao, D. Yu, M. Durstock and L. Dai, Adv. Mater., 2012, 24(17), 2228–2233 CrossRef CAS PubMed.
  15. (a) J. Kim, V. C. Tung and J. Huang, Adv. Energy Mater., 2011, 1(6), 1052–1057 CrossRef CAS PubMed; (b) B. Yin, Q. Liu, L. Yang, X. Wu, Z. Liu, Y. Hua, S. Yin and Y. Chen, J. Nanosci. Nanotechnol., 2010, 10(3), 1934–1938 CrossRef CAS PubMed.
  16. Y. Xu, L. Zhao, H. Bai, W. Hong, C. Li and G. Shi, J. Am. Chem. Soc., 2009, 131, 13490 CrossRef CAS PubMed.
  17. C. K. Chua and M. Pumera, Chem. Soc. Rev., 2014, 43, 291–312 RSC.
  18. K. Krishnamoorthy, M. Veerapandian, K. Yun and S. J. Kim, Carbon, 2013, 53, 38–49 CrossRef CAS PubMed.
  19. S. Thangavel and G. Venugopal, Powder Technol., 2014, 257, 141–148 CrossRef CAS PubMed.
  20. H. Ishii, K. Sugiyama, E. Ito and K. Seki, Adv. Mater., 1999, 11(8), 605–625 CrossRef CAS.
  21. T. Ji, L. Tan and Y. Chen, Phys. Chem. Chem. Phys., 2015, 17, 4137–4145 RSC.
  22. C. Duan, K. Zhang, C. Zhong, F. Huang and Y. Cao, Chem. Soc. Rev., 2013, 42(23), 9071–9104 RSC.
  23. X. Liu, H. Kim and L. J. Guo, Org. Electron., 2013, 14(2), 591–598 CrossRef CAS PubMed.

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra02099a

This journal is © The Royal Society of Chemistry 2015
Click here to see how this site uses Cookies. View our privacy policy here.