Journal of the Korean Physical Society, Vol. 59, No. 2, August 2011, pp. 362∼366
Temperature and Light Intensity Dependence of Polymer Solar Cells with MoO3 and PEDOT:PSS as a Buffer Layer Yongju Park, Seunguk Noh, Donggu Lee, Jun Young Kim and Changhee Lee∗ School of Electrical Engineering and Computer Science, Inter-University Semiconductor Research Center, Seoul National University, Seoul 151-744, Korea (Received 6 April 2011, in final form 13 June 2011) We analyzed the electrical characteristics of polymer solar cells with different buffer layers between the anode and the active layer. The current-voltage characteristics of the solar cells were measured in the temperature range from 200 K to 350 K at illumination intensities from 1 to 100 mW/cm2 . Devices structure with ITO/buffer layer/poly (3-hexylthiophene) (P3HT): [6,6]-phenyl-C61 -butyric acid methyl ester (PCBM)/LiF/Al - buffer layer being molybdenum oxide (MoO3 ) or poly(3,4ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) - showed different electronic transport behaviors with temperature change. A strong temperature dependence of short circuit current density of ITO/PEDOT:PSS/P3HT:PCBM/LiF/Al solar cells indicates that the device is limited by the transport properties of the hole injection layer, PEDOT:PSS. A thermally - activated transport of photo - generated charge carriers, influenced by recombination with shallow traps at the interface, describes the variation of the photo current. PACS numbers: 85.60.Bt, 72.40.+w, 72.80.Le, 72.20.Jv Keywords: Polymer solar cell, Hole Injection Layer, Temperature dependence, Light Intensity dependence, Buffer Layer DOI: 10.3938/jkps.59.362
similar at room temperature; however, the electric properties of the HILs are totally different when the temperature and the light intensity change. In a recent paper, MoO3 was reported to act as a hole generation layer [7] but PEDOT:PSS layer is known to be a hole injection layer. In this research, we analyzed different electrical characteristics of MoO3 as a hole generation layer and PEDOT:PSS as a hole injection layer with the structure of ITO/buffer Layer/P3HT:PCBM/LiF/Al in the temperature range from 200 K to 350 K for illumination intensities from 1 to 100 mW/cm2 .
I. INTRODUCTION Polymer solar cells have some advantages of fabrication of low cost as well as flexible and large panel manufacte [1,2]. However, the power conversion efficiencies (PCE) of polymer solar cells are about ∼8% under AM 1.5 illumination condition, and these efficiencies are much lower than through inorganic solar cells produced for practical applications [3-5]. There are many efforts to enhance the power conversion efficiency (PCE) of polymer solar cells with engineering device structure such as enhanced interpenetrating structure of bulk heterojunction with thermal/solvent-assisted annealing, additional addictive materials and buffer layers for improved charge transport characteristics. In order to increase the PCE of polymer solar cells, we have widely used MoO3 fabricated by using vacuum evaporation or PEDOT:PSS fabricated by using spin coating as the hole injection layer based on poly (3-hexylthiophene) (P3HT) and [6,6]-phenyl-C61 -butyric acid methyl ester (PCBM) solar cells [6]. The current density (J) - voltage (V) characteristics of solar cells with different Hole Injection Layer (HIL) (molybdenum oxide (MoO3 ) or poly(3,4-ethylenedioxythiophene):poly (styrenesulfonate) (PEDOT:PSS)) seem to be almost ∗ E-mail:
II. EXPERIMENTAL Poly (3-hexylthiophene) (P3HT) and [6,6]-phenyl-C61 butyric acid methyl ester (PCBM) were obtained from Rieke Metals Inc. and American Dye Source Inc., respectively. Chlorobenzene (C6 H5 Cl, anhydrous, 99.8%) was purchased from Sigma-Aldrich Chemical Co. Inc. The organic solar cells were prepared on commercial indium-tin-oxide (ITO) coated substrates. The ITO substrates were subsequently cleaned using isopropyl alcohol, de-ionized water, acetone and methanol in an ultrasonic bath and dried in a vacuum oven at 120 ◦ C for 30 minutes. A P3HT:PCBM blend solution was prepared with a weight ratio of 1:0.8 in chlorobenzene. The solu-
[email protected]; Fax: +82-2-877-6668
-362-
Temperature and Light Intensity Dependence of Polymer Solar Cells with MoO3 · · · – Yongju Park et al.
-363-
Fig. 1. (Color online) Semi logarithmic representation of the dark and illuminated (100 mW/cm2 ) current density voltage characteristics for ITO/MoO3 /P3HT:PCBM/LiF/Al and ITO/ PEDOT:PSS /P3HT:PCBM /LiF/Al.
tion was heated to 50 ◦ C and continuously stirred for 3 h. After that, we deposited 10 nm of molybdenumoxide (MoO3 ) onto the ITO substrate under a high vacuum condition (2 × 10−6 Torr), and we deposited on ∼50 nm layer of poly(ethylene-dioxythiophene) doped with poly(styrene sulfonate) (PEDOT : PSS), Batron P (Bayer AG, Germany), was spin coated onto the ITO substrate. The P3HT:PCBM solution was spin-coated on the MoO3 layer to a thickness of 120 nm. Then, the lithium fluoride (LiF) and aluminium (Al) electrodes (100 nm) were evaporated under a high vacuum condition (2 × 10−6 Torr). The samples were heated at 150 ◦ C for 30 minute inside a glove box filled with Ar gas. The overlap of the active area between the ITO and the Al electrode is 4 mm × 5 mm. The photocurrent-voltage characteristics were measured by using a Keithley 237 source measurement unit in the dark and under illumination from a solar simulator (Newport, 91160A). The Light intensity was changed between 1 and 100 mW/cm2 by using neutral density filters.
III. RESULTS AND DISCUSSION The polymer solar cell structures with ITO/buffer layer/P3HT:PCBM/LiF/Al were studied in the temperature range from 200 K to 350 K for illumination intensities from 1 to 100 mW/cm2 . MoO3 and PEDOT : PSS layers were adopted as buffer layers in order to lower the energy barrier located in the ITO/active layer interface. LiF was used as an electron injection layer [8]. Figure 1 shows the current - voltage characteristics (J-V) in a semi-logarithmic representation for bulkheterojunction solar cells based on different buffer layers. As the electrode materials and the active layer materials used in both devices are the same, these differences are determined by the different electrochemical properties
Fig. 2. (Color online) Short circuit current density - voltage characteristics of an organic solar cell using (a) MoO3 and (b) PEDOT:PSS as a buffer layer as a function of voltage at the temperature indicated in the legend.
of MoO3 and PEDOT : PSS. The photocurrents (JSC ) measured under light illumination of 100 mW/cm2 are almost the same. However, the dark J-V characteristics are quite different for the two buffer layers, especially under reverse bias and under forward bias up to the turnon voltage region where characteristics are dominated by the leakage current. The device with PEDOT : PSS shows a large leakage current compared to the deivce with MoO3 ; this characteristic is caused by large lateral current conductivity property [9,10]. Figure 2 show the J-V characteristics of organic solar cells using (a) MoO3 and (b) PEDOT:PSS as a buffer layer as a function of voltage at the temperatures indicated in the legend. The current density of the device with PEDOT:PSS changes more drastically with temperature than that of the device with MoO3 . Several origins can be suggested. First of all, PEDOT:PSS acts as a hole transporting layer; therefore, holes pass through High Occupied Molecular Orbital (HOMO) of the PEDOT:PSS layer, unlike a thin MoO3 layer, which acts as hole generation layer [7]; moreover, the PEDOT : PSS layer is thicker than the MoO3 layer. Second of all, the conductivity of the PEDOT : PSS film decreases as the temperature is decreased [11]. As a result, the JV characteristics of solar cells with PEDOT : PSS show poor performance compare to those of the solar cell with MoO3 at low temperatures. Figure 3(a) shows the short-circuit current density
-364-
Journal of the Korean Physical Society, Vol. 59, No. 2, August 2011
Fig. 3. (Color online) Variation of (a) the short circuit density, (b) the open-circuit voltage, (c) the fill factor and (d) the power conversion efficiency (PCE) with temperature for ITO/MoO3 /P3HT:PCBM/LiF/Al and ITO/PEDOT:PSS/P3HT:PCBM/LiF/Al under illumination with PLight = 100 mW/cm2 .
Fig. 4. (Color online) Temperature dependence of the (a) series resistance, RS , and (b) parallel resistance, RSh , for ITO/ MoO3 /P3HT:PCBM/LiF/Al and ITO/PEDOT:PSS/P3HT:PCBM/LiF/Al under illumination with PLight = 100 mW/cm2 .
(JSC ) as a function of temperature. The JSC of the studied samples increases with increasing temperature, especially the device with PEDOT : PSS. The temperature dependence of the current density in polymer solar cells can be due to either the temperature-dependent mobility or to the steady state charge carrier concentration [12], where might means that holes may not pass through the interlayer between the active layer and the anode because of small conductivity of carriers due to the low temperature. This phenomenon happens more in PEDOT:PSS, because the impedance and the resistance of PEDOT:PSS are higher than those of MoO3 . The open-circuit voltage manifests an almost linear de-
crease with increasing temperature. Consistent with inorganic solar cell theory, an expression for VOC is obtained, which contained a temperature-dependent factor that affects the open-circuit voltage. First, the diffusion of charge will increase more rapidly with increasing temperature than the charge mobility. Thus, if the drift and the diffusion currents are to be balanced at open circuit, a higher voltage must be applied across the device as the temperature increases [13]; Eg kT I o max Voc = − ln , (1) e e Isc where Eg is the energy gap of the material, Isc the short-
Temperature and Light Intensity Dependence of Polymer Solar Cells with MoO3 · · · – Yongju Park et al.
-365-
Fig. 5. (Color online) (a) Activation energy fitting data from the Arehenius equation and (b) activation energy values for ITO/ MoO3 /P3HT:PCBM/LiF/Al and ITO/PEDOT : PSS/P3HT:PCBM/LiF/Al under illumination with PLight from 1 mW/cm2 to 100 mW/cm2 .
Fig. 6. (Color online) Double-logarithmic representation of the short circuit current density as a function of light intensity for different temperatures of the device using (a) MoO3 and (b) PEDOT:PSS as a buffer layer. The recombination coefficient α is indicated in the inset for T = 200 K and T = 350 K.
circuit current, Iomax the maximum saturation current, k the Boltzmann constant, T the absolute temperature and e the electronic charge. There are no significant differences in V OC with MoO3 and PEDOT:PSS - layerbased solar cells, as shown in Fig. 3(b). We observe the fill factor (FF) and the PCE of the device in Figs. 3(c) and (d). The FF of the device using PEDOT:PSS increases almost linear with increasing temperature, but no significant change is observed in the device using MoO3 . The FF and the PCE are relatively smaller at low temperatures in the device using PEDOT:PSS, but the FF and the PCE seem to be relatively constant with increasing temperature for the device using MoO3 . From the J-V curves of the illuminated sample, the series (RS ) and the parallel (RSh ) resistance were deduced and are shown in Fig. 4. RS is primarily governed by the Ohmic resistance. RS in the investigated PEDOT:PSSlayer-based solar cell decreases in the temperature range 200 K - 350 K, despite there being no significant change in the MoO3 -layer-based solar cell, whereas RSh remains almost constant in both devices. These two parameters determine the overall J-V shape and show themselves in the fill factor (FF). The FF of the PEDOT:PSS-layerbased solar cell change significantly with temperature change, indicating different transport properties between
the PEDOT:PSS layer as a hole-injection layer and the MoO3 layer as a hole-generation layer. We derived the activation energy from the Arehenius equation, which means a minimum- energy chemical response can happen, in Fig. 5(a). The activation energy is related to the trap depth of the device in organic solar cells, and carrier is difficult in a device with a large activation energy [14]. The activation energy of the PEDOT:PSS device is much bigger than that of the MoO3 device in whole illumination range, as displayed in Fig. 5(b). This activation energy affects carrier transport at the interlayer between the active layer and the electrode. We can explain the FF of the PEDOT:PSS case by the temperature still being much lower than 240 K even though the series resistance of the PEDOT:PSS device is higher than that of the MoO3 device by using the activation energy results; ∆ . (2) Jsc (T, PLight ) = J0 (PLight ) exp − kT Figure 6 displays the short circuit current density as a function of the light intensity PLight as a doublelogarithmic scale. The JSC follows the power law dependence JSC = Pα Light the recombination coefficient α is obtained for T = 200 K and T = 350 K representa-
-366-
Journal of the Korean Physical Society, Vol. 59, No. 2, August 2011
tion in the two devices. The nearly linear dependence of JSC (α = 1) indicates that charge-carrier losses in the absorber layer are dominated by monomolecular recombination. If pure bimolecular recombination is dominant, the JSC follows a square root dependence of PLight with α = 0.5 [15]. In the MoO3 device, α did not change much with temperature, as shown in Fig. 6(a). However, we could obtain an increasing α value with increasing temperature increased in the PEDOT:PSS device, as shown in Fig. 6(b). This phenomenon can be interpreted as being due to the carrier recombination characteristics, reflect the different J-V characteristics of solar cells with different buffer layers (PEDOT:PSS and MoO3 ), depending on the temperature.
IV. CONCLUSION In conclusion, performance was demonstrated for bulk heterojunction solar cell based on MoO3 and PEDOT:PSS materials, a notable difference in the value of the short-circuit current density - voltage change with increasing temperature. This difference is attributed to the electronic transport properties of the buffer-layer material studied. We obtained relatively different electrical properties changes between the PEDOT:PSS-layer- and the MoO3 layer-based solar cells due to different hole transporting patterns. A strong temperature dependence of the short-circuit current density of the PEDOT:PSS-layerbased solar cells indicates that the device is limited by the transport properties of the hole injection layer, PEDOT:PSS. Therefore, this might be indicated by the bimolecular probability of the PEDOT:PSS device decreasing because the hole mobility is increased by a temperature change, so charge carriers at the PEDOT:PSS interface flow easily. This bimolecular recombination probability decrease is also related to the larger current density change with increasing temperature, especially in the PEDOT:PSS device.
ACKNOWLEDGMENTS
This work was financially supported by the National Research Foundation (NRF) of Korea through the Acceleration Research Program (R0A-2008-000-20108-0) and also supported in part by the Ministry of Education, Science, and Technology (MEST) through the BK21 Program.
REFERENCES
[1] G. Yu, J. Gao, J. C. Hummelen, F. Wudl and A. J. Heeger, Science 270, 1789 (1995). [2] C. J. Brabec, N. S. Sariciftci and J. C. Hummelen, Adv. Funct. Mater. 11, 15 (2001). [3] S. Sista, M. H. Park, Z. Hong, Y. Wu, J. Hou, W. L. Kwan, G. Li and Y. Yang, Adv. Mater. 22, 380 (2010). [4] H. Hoppe, N. S. Sariciftci and D. Meissner, Mol. Cryst. Liq. Cryst. 385, 113 (2002). [5] J. Y. Kim, K. Lee, N. E Coates, D. Moses, T. Nguyen, M. Dante and A. J. Heeger, Science 317, 222 (2007). [6] J. H. Park, T.-W. Lee, B.-D. Chin, D. H. Wang and O. O. Park, Macromol. Rapid Commun. 31, 2095 (2010). [7] M. Kr¨ oger, S. Hamwi, J. Meyer, T. Riedl, W. Kowalsky and A. Kahn, Appl. Phys. Lett. 95, 123301 (2009). [8] A. Godoy et al., Sol. Energy Mater. Sol. Cells 94, 648 (2010). [9] S. C. Veenstra, U. Stalmach, V. V. Krasnikov, G. Hadziioannou, H. T. Jonkman, A. Heeres and G. A. Sawatzky, Appl. Phys. Lett. 76, 2253 (2000). [10] S. Steudel, S. De Vusser, K. Myny, M. Lenes, J. Genoe and P. Heremans, J. Appl. Phys. 99, 114519 (2006). [11] J. Y. Kim, J. H. Jung, D. E. Lee and J. Joo, Synth. Met. 126, 311 (2002). [12] I. Riedel, V. Dyakonov, J. Parisi, L. Lutsen, D. Vanderzande and J. C. Hummelen, Adv. Funct. Mater. 14, 38 (2004). [13] D. Chirvase, Z. Chiguvare, M. Knipper, J. Parisi, V. Dyakonov and J. C. Hummelen, Synth. Met. 138, 299 (2003). [14] G. Boschloo and A. Hagfeldt, J. Phys. Chem. B. 109, 12093 (2005). [15] V. D. Mihailetchi, P. W. M. Blom, J. C. Hummelen and M. T. Rispens, J. Appl. Phys. 94, 15 (2003).
