Effect of electrolyte redox potentials on the

MOLECULAR CRYSTALS AND LIQUID CRYSTALS , VOL. , – https://doi.org/./..

Effect of electrolyte redox potentials on the photovoltaic performance in dye-sensitized solar cells based on porphyrin dye Sue Kyung Kim, Phuong Ho, Ji Won Lee, So Yeon Jeon, Suresh Thogiti, Rajesh Cheruku, Hyun-Jun Jo, and Jae Hong Kim

Downloaded by [University of New England] at 04:28 08 November 2017

School of chemical engineering, Yeungnam University, Gyeongsan, Gyeongbuk, Republic of Korea

ABSTRACT

KEYWORDS

We have investigated the effect of two different kinds of cobalt electrolytes on the photovoltaic performance in dye-sensitized solar cells (DSSCs) using the established porphyrin dye (YD-2). We have characterized the photovoltaic performance of DSSCs containing iodide/triiodide and two different Co3+/2+ electrolytes to confirm the relationship between redox potential of cobalt(III/II) tris(2,2 -bipridine), [Co(bpy)3 ]3+/2+ , cobalt(III/II) tris(1,10-phenanthroline), [Co(phen)3 ]3+/2+ and the photovoltaic characteristics in DSSCs. Compared to the iodine-based electrolyte, the photovoltage of the DSSCs based on [Co(phen)3 ]3+/2+ was increased by 15%. The current densities in [Co(bpy)3 ]3+/2+ and [Co(phen)3 ]3+/2+ -based devices decrease due to their larger size and decrease in difference bewteen HOMO level of dye and redox potential of electrolytes.

porphyrin YD- dye; cobalt electrolyte; open circuit voltage

Introduction The sunlight harvesting technology using photovoltaic devices has received great attention as an indispensable component of future global energy production. Among various photovoltaic technologies, the dye-sensitized solar cell (DSSC) introduced by O’Regan and Grätzel has been extensively studied as a progressive solar energy conversion technology because of its low cost, simple fabrication, and eco-friendliness [1]. In traditional DSSC, ruthenium-based complexes (for example, N719 and Z907) has been used as the sensitizing dye (SD) and ruthenium sensitizers have been distinguished by attaining more than 11% efficiencies [2, 3]. These ruthenium-based dyes have wide absorption spectra (λ ࣈ 350 nm), but low molar extinction coefficients (5,000–20,000 M−1 cm−1 ). When molar extinction coefficient of the SD is low, the mesoporous TiO2 layer must be thick to increase adsorption amount of the SD. Organic sensitizers have recently received great attention due to modest cost, ease of synthesis and modification, large molar absorption coefficients, and satisfactory stability. Porphyrins dyes have strong absorption in the visible region as well as available HOMO and LUMO energy levels. Dye materials with various structures based on porphyrin have been CONTACT Prof. Jae Hong Kim [email protected] School of Chemical Engineering, Yeungnam University, , Daehak-Ro, Gyeongsan, Gyeongbuk, , Korea (ROK). Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/gmcl. ©  Taylor & Francis Group, LLC

MOLECULAR CRYSTALS AND LIQUID CRYSTALS

85


Figure . Molecular structures of (a) porphyrin YD- dye and (b) iodine and cobalt complexes.

designed and investigated to enhance energy conversion efficiency [4–6]. The best conversion efficiencies in porphyrin based DSSC using iodine electrolyte system have been reported to 11% [7]. On the other hand, iodide/triiodide electrolyte has advantages such as easy penetration into TiO2 , high regeneration rate of dye and relatively slow recombination. However, the iodine electrolyte system has several issues such as volatile, leakage, and low mechanical properties. In addition, triiodide ion has wide absorption region in visible light region and competes with dye on light absorption, therefore, the short circuit current density (Jsc) is limited. To replace these iodine electrolyte system, many studies are in progress [8–10]. The cobalt electrolyte has attracted attention as the viable alternative electrolyte for iodide/triiodide electrolyte. A cobalt electrolyte has low and narrow absorption region compared to triiodide ion, and their redox potentials can be controlled easily by introduction of various ligands, electron donor or acceptor group [11]. Therefore, they have the advantage such as optimization of dye regeneration driving force and a high Voc. In this paper, cobalt-based redox systems were introduced to replace the iodide/triiodide redox system. As a photosensitizer, YD-2 porphyrin dye was employed [7]. We have characterized the photovoltaic performance of DSSCs containing iodide/triiodide and two different Co3+/2+ electrolytes. The redox species were cobalt(III/II) tris(2,2-bipridine), [Co(bpy)3 ]3+/2+ and cobalt(III/II) tris(1,10-phenanthroline), [Co(phen)3 ]3+/2+ . Figure 1 shows the molecular structures of porphyrin YD-2 dye and electrolytes. The DSSC sensitized with YD-2 using [Co(phen)3 ]3+/2+ redox couple exhibited VOC of 0.78 V under 1 sun, which is 15% higher than the device with iodine-based electrolyte.

Experimental 1. Preparation of the cobalt complexes The cobalt complexes [Co(bpy)3 ](PF6 )2 , [Co(bpy)3 ](PF6)3 (bpy = 2,2 -bipyridine), [Co(phen)3 ](PF6 )2 and [Co(phen)3 ](PF6 )3 , (phen = 1,10-phenanthroline) were synthesized according to a previously reported procedure. Lithium perchlorate (LiClO4 ), 4-tert-butylpyridine (tBP), iodine (I2 ), 1,2-dimethyl-3-propylimidazolium iodide (PMII), 1-butyl-3-methylimidazolium iodide (BMII) and solvents were used as received from Sigma-Aldrich and TCI. The detailed synthesis processes are presented in ref. 12. 2. Solar cell fabrication and characterization of the DSSCs FTO glass was cleaned by sequential sonication in methanol, distilled water and acetone. A thin first layer of TiO2 was deposited by the “doctor-blade” technique. The film was dried at


86

S. K. KIM ET AL.

Figure . Energy level diagram of a DSSC sensitized with YD-, employing the different electrolytes.

70°C for 30 min, followed by heating to 450°C for 30 min. A second layer of 200 nm-sized light scattering particles was then coated on the top of the first layer, which was followed by heating to 450°C for 30 min. The TiO2 electrodes were immersed in a 0.2 mM solution of YD-2 porphyrin dye in ethanol at room temperature for 18 h before being removed and rinsed in ethanol immediately before cell fabrication. The cells were fabricated by sandwiching the sensitized TiO2 electrode and a thermally platinized FTO counter electrode together with a hot-melt polymer (Surlyn, 60 μm). The current density-voltage (J-V) characteristics of the prepared DSSCs were measured under a 1 sunlight intensity (100 mW/cm2 , AM 1.5) using a Keithley model 2400 Source Meter and a Newport 91192 solar simulator system (equipped with a 1 kW xenon arc lamp, Oriel). Light intensity was adjusted to 1 sun (100 mW/cm2 ) with a Radiant Power Energy Meter (model 70260, Oriel). The incident photon-to-current conversion efficiency (IPCE) results acquired from IPCE G1218a (PV Measurement). This system applies monochromatic light from a 75 W xenon arc lamp (Ushio UXL-75XE) filtered by a dual-grafting monochromator and individual filters onto the test devices. Electrochemical impedance spectroscopy (EIS) was performed using an electronic-chemical analyzer (Iviumstat Tec.).

Results and discussion A ruthenium-free porphyrin YD-2 dye has higher molar extinction coefficients (312,000 M−1 cm−1 ) and 11% solar-to-electric power conversion efficiency under 1 sunlight intensity (100 mW/cm2 , AM 1.5) was reported using iodine-based electrolyte by Bessho et al. [7]. We introduced cobalt electrolyte to improve conversion efficiency through increase in VOC . An energy level diagram of a DSSC sensitized with YD-2, employing the different cobalt redox couples is shown in Fig. 2. The J-V curves of the DSSCs using YD-2 photosensitizers with iodine based electrolyte and two different cobalt electrolytes are presented in Fig. 3. The measurements were carried out at AM 1.5G, and the photovoltaic parameters of the cells are listed in Table 1.



87

Figure . Current density–voltage characteristics for DSSCs with different electrolytes.

DeVries et al. found that the VOC increased as the formal potential of the redox mediator became more positive for a series of cobalt complexes. In our study, the VOC values also increased with the Nernst potential of the redox mediators and the highest VOC was obtained using [Co(phen)3 ]3+/2+ . However, unlike what we expected the efficiency was reduced. The lower performance attained by the cobalt electrolytes compared to iodine electrolyte is mainly due to the loss of the photocurrent and fill factor. In general, low photocurrents and fill factors were observed for all cobalt-based cells compared to iodine based cells. This pattern of low photocurrents and fill factors with cobalt electrolyte has been reported because the size of cobalt complexes are larger than that of iodine which could bring out slow mass transport through the nanoparticle network and electrolyte [13]. In addition, fast recombination of electrons in the TiO2 to the Co3+ complex decreases photocurrent and fill factor [14, 15]. These recombination rate of electron between the TiO2 and the Co+3 complex increases as the potential difference between Fermi level of the TiO2 and the energy level of the Co+3 complex increases, therefore the JSC for [Co(phen)3 ]3+/2+ is lower than that for [Co(bpy)3 ]3+/2+ . Incident photon-to-current conversion efficiency (IPCE) measurements were performed for three DSSCs containing [Co(bpy)3 ]3+/2+ , [Co(phen)3 ]3+/2+ and I− /I3 − redox mediators, and the IPCE results are presented in Fig. 4. The shapes of the IPCE spectra for the DSSCs using all cobalt redox mediators are similar to that using iodine, but the intensities of cobalt electrolytes are smaller than that of iodine electrolyte. These poor IPCE results in [Co(bpy)3 ]3+/2+ and [Co(phen)3 ]3+/2+ electrolytes are good agreement in the J-V results and could be attributed to its lower conductivity as shown in Table II and decrease in difference bewteen the HOMO level of dye and redox potential of electrolytes. The spectra of cobalt electrolytes in the short wavelength region (from 300 to 350 nm) more slowly reduced comparing to the long wavelength region. If all absorption spectra are due to the absorption of the dye, the IPCE spectra should be reduced at a constant ratio in all regions. However, Table . Device performance of the dye-sensitized solar cell with the various electrolytes. Electrolyte

JSC (mA/cm )

VOC (V)

FF (%)

Efficiency (%)

Iodine Co(bpy) Co(phen)

. . .

. . .

. . .

. . .


88

S. K. KIM ET AL.

Figure . Incident photon-to-current conversion efficiency (IPCE) for DSSCs with different electrolytes.

the signals in the short wavelength region are dominantly attributed to the absorption in the mesoporous TiO2 layer and the electrolyte supplies electrons to the valence band (VB) of the mesoporous TiO2 . When the electrolyte does not supply electron to the VB, the TiO2 cannot absorb the light and the IPCE spectra related to the TiO2 should disappear. If the long alkly chains completely prevented the TiO2 from electrolyte, the absorption spectra at short wavelength should disappear. These IPCE spectra at short wavelength mean that YD2 dyes do not prevent the TiO2 from the electrolyte. On the other hand, the VOC in cobalt electrolyte based DSSCs was increased as the redox potential of electrolytes decreased as shown in Fig. 2. Electrochemical impedance spectroscopy (EIS) was then applied to investigate the interfacial charge transfer processes of the devices, especially for understanding the charge transfer dynamics at the dye coated TiO2 photoelectrode/electrolyte [16]. Fig. 5 shows the Nyquist plot of iodine and cobalt electrolytes under dark conditions and the corresponding values are listed in Table 2. The first smaller semicircle at higher frequency represents the charge transfer resistances at the Pt/electrolyte interface (Rce ), second larger one at medium frequency

Figure . Electrochemical impedance spectra (EIS) for different electrolytes.


89


Table . EIS analysis for the various electrolytes. Electrolyte

RS ()

Rce ()

Rct ()

Ws ()

Iodine Co(bpy) Co(phen)

. . .

. . .

. . .

. . .

indicates charge recombination resistance at the TiO2 /dye/electrolyte interface (Rct ) and third one at lower frequency indicates diffusion of the redox shuttle through the electrolyte (Ws ). Rs is the series resistance and the semicircles related to electrochemical charge transfer on the photoanode and Pt counter electrode are mostly convoluted in the first semicircle at a higher frequency. Due to the poor electron-transfer kinetics between Co-electrolyte and Pt, the charge transfer resistance at the Pt counter electrode, Rce , is another significant source of resistance. From the Figure 5 and Table 2, it is clear that the recombination resistance (Rct ) obtained at the working electrode interface (TiO2 /dye/electrolyte) is increased in the order of I− /I3 − < [Co(bpy)3 ]3+/2+ < [Co(phen)3 ]3+/2+ , which indicate that the recombination reaction rate of the transferred electrons with the oxidized triiodide species in the electrolyte is decreased in the order of I− /I3 − > [Co(bpy)3 ]3+/2+ > [Co(phen)3 ]3+/2+ . The results are consistent with the obtained VOC values.

Conclusion In this study, we employed [Co(bpy)3 ]3+/2+ and [Co(phen)3 ]3+/2+ electrolytes to improve photovoltaic performance in DSSCs based on porphyrin YD-2 dye through the increase in VOC . The DSSC sensitized with YD-2 using [Co(phen)3 ]3+/2+ redox couple showed VOC of 0.78 V under 1 sun. This value was 15% higher than that of the iodine-based electrolyte because of high potential difference. The current densities and incident photon-to-current conversion efficiency (IPCE) of DSSCs with [Co(bpy)3 ]3+/2+ and [Co(phen)3 ]3+/2+ decreased due to their lower conductivity and decrease in difference bewteen HOMO level of dye and redox potential of electrolytes. The recombination resistance (Rct ) obtained at the working electrode interface (TiO2 /dye/electrolyte) is increased in the order of I− /I3 − < [Co(bpy)3 ]3+/2+ < [Co(phen)3 ]3+/2+ and the recombination reaction rate of the transferred electrons with the oxidized triiodide species in the electrolyte is decreased in the order of I− /I3 − > [Co(bpy)3 ]3+/2+ > [Co(phen)3 ]3+/2+ .

Acknowledgments This work was supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) and the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (No. 20163010012310).

References [1] O’Regan, B. & Grätzel, M. (1991). Nature, 353, 737. [2] Nazeeruddin, M. K., Angelis, F. D., Fantacci, S., Selloni, A., Viscardi, G., Liska, P., Ito, S., Takeru, B., & Grätzel, M. (2005). J. Am. Chem. Soc., 127, 16835. [3] Wang, P., Zakeeruddin, S. M., Moser, J. E., Nazeeruddin, M. K., Sekiguchi, T., & Grätzel, M. (2003). Nat. Mater., 2, 402.


90

S. K. KIM ET AL.

[4] Collin, J.-P., Harriman, A., Heitz, V., Odobel, F., & Sauvage, J.-P. (1994). J. Am. Chem. Soc., 116, 5679. [5] Gust, D., Moore, T. A., & Moore, A. L. (2001). Acc. Chem. Res., 34, 40. [6] Duncan, T. V., Wu, S. P., & Therien, M. J. (2006). J. Am. Chem. Soc., 128, 10423. [7] Bessho, T., Zakeeruddin, S. M., Yeh, C.-Y., Diau, E. W.-G., & Grätzel, M. (2010). Angew. Chem. Int. Ed., 49, 6646. [8] Hattori, S., Wada, Y., Yanagida, S., & Fukuzumi, S. (2005). J. Am. Chem. Soc., 127, 9648. [9] Li, D., Li, H., Luo, Y., Li, K., Meng, Q., Armand, M., & Chen, L. (2010). Adv. Funct. Mater., 20, 3358. [10] Phuong, H., Suresh, T., Yong, H. L., & Jae, H. K. (2017). Sci. Rep., 7, 2272. [11] Nelson, J., Amick, T. J., & Elliott, C. M. (2008). J. Phys. Chem. C, 112, 18255. [12] Lee, D. K., Ahn, K.-S., Thogiti, S., & Kim, J. H. (2015). Dyes and Pigments, 117, 83. [13] Feldt, S. M., Gibson, E. A., Gabrielsson, E., Sun, L., Boschloo, G. & Hagfeldt, A. (2010). J. Am. Chem. Soc., 32, 16714. [14] Nelson, J. J., Amick, T. J., & Elliott, C. M. (2008). J. Phys. Chem. C, 112, 18255. [15] Klahr, B. M. & Hamann, T. W. (2009). J. Phys. Chem. C, 113, 14040. [16] Nguyen, T. H., Suresh, T., & Kim, J. H. (2016). Org. Electron., 30, 40.