Perfluoro anion based binary and ternary ionic liquids as electrolytes for dye-sensitized solar cells

Perfluoro anion based binary and ternary ionic liquids as electrolytes for dye-sensitized solar cells

Journal of Power Sources 311 (2016) 167e174 Contents lists available at ScienceDirect Journal of Power Sources journal homepage: www.elsevier.com/lo...

1MB Sizes 0 Downloads 1 Views

Journal of Power Sources 311 (2016) 167e174

Contents lists available at ScienceDirect

Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour

Perfluoro anion based binary and ternary ionic liquids as electrolytes for dye-sensitized solar cells HsieHsin Lin a, JiaeDe Peng a, V. Suryanarayanan b, D. Velayutham b, KuoeChuan Ho a, c, * a

Department of Chemical Engineering, National Taiwan University, Taipei 10617, Taiwan Electroorganic Division, CSIReCentral Electrochemical Research Institute, Karaikudi 630006, India c Institute of Polymer Science and Engineering, National Taiwan University, Taipei 10617, Taiwan b

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

 Eight triethylammonium and nmethylpiperidinium cations based ILs are synthesized.  The ILs have different carbon chain lengths of perfluoro carboxylate anions (PFC).  The cell efficiency decreases with the increase of the carbon chain length of PFC.  DSSC with ternary IL as electrolyte shows an efficiency of 6.01%.  It shows an unfailing long-term stability up to 1200 h stored in dark at 50  C.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 19 November 2015 Received in revised form 18 January 2016 Accepted 8 February 2016 Available online xxx

In this work, eight new ionic liquids (ILs) based on triethylammonium (TEA) or n-methylpiperidinium (NMP) cations and perfluoro carboxylate (PFC) anions having different carbon chain lengths are synthesized and their physico-chemical properties such as density, decomposition temperature, viscosity and conductivity are determined. Photovoltaic characteristics of dye-sensitized solar cells (DSSCs) with binary ionic liquids electrolytes, containing the mixture of the synthesized ILs and 1-methyl-3-propyl imidazolium iodide (PMII) (v/v ¼ 35/65), are evaluated. Among the different ILs, solar cells containing NMP based ILs show higher VOC than that of TEA, whereas, higher JSC is noted for the DSSCs incorporated with the latter when compared to the former. Further, the photo-current of the DSSCs decreases with the increase of the carbon chain length of perfluoro carboxylate anionic group of ILs. The cell performance of the DSSC containing ternary ionic liquids-based electrolytes compose of NMP-2C/TEA-2C/PMII (v/v/ v ¼ 28/7/65) exhibits a JSC of 12.99 mA cm2, a VOC of 639.0 mV, a FF of 0.72, and a cell efficiency of 6.01%. The extraordinary durability of the DSSC containing the above combination of electrolytes stored in dark at 50  C is proved to be unfailing up to 1200 h. © 2016 Elsevier B.V. All rights reserved.

Keywords: Ionic liquid Dyeesensitized solar cell Perfluoro carboxylate Long-term stability Gel electrolytes

1. Introduction

* Corresponding author. Department of Chemical Engineering, National Taiwan University, Taipei 10617, Taiwan. E-mail address: [email protected] (K.-C. Ho). http://dx.doi.org/10.1016/j.jpowsour.2016.02.029 0378-7753/© 2016 Elsevier B.V. All rights reserved.

In order to convert sunlight into electric current, dye-sensitized solar cells (DSSCs) have been developed owing to their easy fabrication, low cost, and high absorption factor [1e3]. To date, the cell

168

H.-H. Lin et al. / Journal of Power Sources 311 (2016) 167e174

efficiency of a DSSC achieved is around 14.3% [4]. Irradiated under the sunlight, the dye molecules in a DSSC inject electrons into the conduction band of the semiconductor, which then flow through the semiconductor followed by the external load and the counter electrode in order to trigger the reduction of triiodide ions to iodide ions, which in turn regenerate the oxidized dye and thereby complete the circuit. The related scheme of a DSSC is shown in Fig. S1. It is common to use organic solvents with high conductivities and low viscosities to prepare electrolytes, but high volatility of organic solvents may cause the leakage of electrolyte during sealing and this may pose great challenge for their practical use in the DSSCs [5e7]. In order to solve this problem, different strategies were proposed such as the employment of solvents with high boiling point [8], addition of inorganic metal [9,10] or polymers [11,12] as gelators to form gel-like electrolytes, incorporation of polymer matrices to inhibit the evaporation of electrolytes [13,14], or the application of room-temperature ionic liquids (RT-ILs) with high conductivity and good thermal stability [15e19]. In recent years, RT-ILs are found to be attractive in replacing the organic solvents due to several advantages such as their ease of preparation, facile penetration ability, high thermal stability (>100  C), wide electrochemical window [20], and high ionic conductivity [21,22]. Among the various types of RT-ILs based electrolytes widely reported for the DSSCs, 1-methyl 3propylimidazolium iodide (PMII) is a potential candidate for application in DSSC [23,24], as it shows a high conductivity and a good thermal stability; however, its high viscosity (1024 cP) [25] poses a big problem for its penetration through the TiO2 film (the semiconductor layer). Alternatively, addition of low viscosity RTILs into the PMII-based electrolyte, namely binary ionic liquid (bi-IL), increases their penetration in the TiO2 films leading to enhanced cell performance [26e29]. Recently, Lee et al. have synthesized three new ILs based on siloxane diimidazolium iodides and then mixed with guanidinium thiocyanate (GNCS), iodine (I2), and tert-butylpyridine (tBP) in 3methoxypropionitrile, The pertinent DSSC demonstrated higher cell efficiency (average cell efficiency ~5.8%) than that of DMII/tBP based DSSC (average cell efficiency ~5.0%) [30]. Vijayakumar et al. employed a polymer membrane for the absorption of the ionic liquid electrolyte, where the cell performance of DSSC demonstrated good long-term stability (~30 days without an obvious decline) and high efficiency (~6.42%); however, a complicated multi-step process is needed to fabricate a polymer matrix [31]. To the best of our knowledge, the state-of-the-art DSSCs with ILsbased or gel electrolytes show cell efficiencies of about 5e7% [11,13,30,31]. In this work, eight new ILs based on triethylammonium (TEA) or n-methylpiperidinium (NMP) cations and perfluoro carboxylate (PFC) anions having different carbon chain lengths were synthesized and their physical properties have been measured. The chemical structures of anions and cations of the ILs to be synthesized are shown in Table 1. The photo-voltaic characteristics of the DSSCs with binary ILs containing NMP-based PFC or TEA-based PFC mixed with PMII and ternary ionic liquids (ter-IL) containing the mixture of NMP-2C, TEA-2C and PMII were investigated. The experimental results show that DSSCs with the ter-IL based electrolytes can reach an outstanding performance (6.01%) by combining the advantages of PMII, NMP and TEA-based ILs and remain an unfailing long-term stability up to 1200 h. 2. Experimental section 2.1. Materials Diethyl ether was bought from SRL chemical company, India.

Iodine (I2, synthetical grade) and 1-methyl-3-propyl imidazolium iodide (PMII) were obtained from Merck; titanium (IV) tetraisopropoxide (TTIP, 97%), tert-butyl alcohol (tBA, 99.5%), 2methoxyethanol (99.95%), trifluoroacetic acid (99%), perfluorobutyric acid (99%), perfluorooctanoic acid (96%), guanidine thiocyanate (GuSCN), dimethyl sulfoxide (DMSO, 99.7%), acetonitrile (ACN, 99.99%), neutral cleaner, acetone (99.5%), and isopropyl alcohol (IPA, 99.5%), and triethylamine (99.5%) were obtained from SigmaeAldrich; Perfluorohexanoic acid (97%), and 1methylpiperidine (99%) were bought from Alfa Aesar, England; Nmethylbenzimidazole (NMBI, 99%) was bought from Acros. N719 dye was purchased from Luminescence Technology Corp, Taiwan. 2.2. Preparation of RTILs A series of RTILs containing N-methyl piperidinium and triethylammonium cations incorporated with various anions such as trifluoroacetate, perfluorobutyrate, perfluorohexanoate and perfluorooctanoate were prepared. Equimolar amount of perfluoro acid was added to N-methyl piperidine or triethylamine dropwise with constant stirring under nitrogen atmosphere at 0  C. The reaction mixture was stirred for 6 h at 30  C. The reaction vessel is shown in Fig. S2. Remnant amine in the raw product was removed under vacuum at 80  C for 24 h and the reactions mixture was cooled to room temperature. Further, the residual perfluoro acid and moisture in the product was removed by washing with diethyl ether for several times and then dried at 40  C in vacuum for 24 h. The chemical structure of synthesized ILs were confirmed by 1H and 13C NMR (Bruker Avance III HD-600 MHz NMR Spectrometer with CDCl3 as solvent and tetramethylsilane (TMS) as internal reference). The NMR characteristics are shown in Fig. S3. 2.3. Preparation of binary and ter-IL based electrolytes The binary and ternary IL based electrolytes were prepared by mixing 0.2 M I2, 0.4 M NMBI, 0.15 M GuSCN with PMII and the mixture was treated with either TEA-based PFC or NMP-based PFC (v/v ¼ 35/65) and NMP-2C/TEA-2C (v/v/v ¼ 0e35/35e0/65) respectively under constant stirring at 300 rpm for 24 h under 60  C. We maintained a volume ratio of 35/65 for the case of bi-ILs or ter-ILs (NMPþTEA)/PMII. This volume ratio was optimized in our previous studies [27,32,33], which gave the highest photovoltaic properties of the DSSC. 2.4. Assembly of the DSSCs The working electrode was prepared by thorough mixing of TTIP and 2-methoxyethanol (1:3 ¼ wt/wt) and was coated on a fluorineedoped SnO2 conducting glasses (FTO, TECe7, 7 U sq.1, NSG America, Inc., New Jersey, USA). Prior to the coating, it was cleaned by a neutral cleaner, and then washed with deionized water, acetone, and isopropanol, where each of treatments prolonged for 30 min. Precursor was coated on the washed conducting glasses by spin coating (3000 rpm for 30 s). Coated glass was gradually heated to 500  C and maintained at 500  C for 30 min to form contact layers. TiO2 films were made by the following two steps. First, a commercial TiO2 paste (TieNanoxide HT/SP, Solaronix SA, Switzerland) was coated on the contact layer by doctor-blade method as the transparent layer and heated at 500  C for 30 min. After cooling to room temperature, a self-made TiO2 paste was coated on the previous TiO2 as the scattering layer and annealed by the same process. The procedure for the preparation of the scattering layer paste had been described in our previous publication [32,33]. The thickness of TiO2 films was measured by a surface profilometer (Sloan Dektak

H.-H. Lin et al. / Journal of Power Sources 311 (2016) 167e174

169

Table 1 Chemical structures of ionic liquids. Name

Cation

Anion

n-methylpiperidiniumtrifluoroacetate (NMP-2C)

n-methylpiperidiniumperfluorobutyrate (NMP-4C)

n-methylpiperidiniumperfluorohexanoate (NMP-6C)

n-methylpiperidiniumperfluorooctanoate (NMP-8C)

Triethylammoniumtrifluoroacetate (TEA-2C)

Triethylammoniumperfluorobutyrate (TEA-4C)

Triethylammoniumperfluorohexanoate (TEA-6C)

Triethylammoniumperfluorooctanoate (TEA-8C)

3030), and the final thickness of mesoscopic TiO2 was maintained at ~15 mm. After sintering at 500  C and cooling to 80  C, the TiO2 electrode was immersed in a 5  104 M solution of N719 dye in acetonitrile (ACN) and tertebutyl alcohol (tBA) (1:1 ¼ v/v), at room temperature for 24 h. The photoanode of a DSSC was thus obtained. The counter electrode was prepared by sputtering Pt on the conducting glasses. The thickness of Pt film was controlled to be ~50 nm, which was obtained from the calibration curve of the sputtering. The DSSC was fabricated by sandwiching the above prepared TiO2 photoanode and the counter electrode. A 25 mmethick Surlyn® tape was used to separate and to seal them latter by heating. After the introduction of the electrolyte through the gap between the two electrodes by means of a capillary technique, the gap was closed using a hotemelt glue.

2.5. Instrumentation and measurements The active area of a DSSC was 0.33 cm2 covered by a mask with a light-illuminated area of 0.16 cm2 to reduce the scattered light and illuminated by a class-A quality solar simulator (XES-301S, AM1.5G, San-Ei Electric Co., Ltd.). Incident light intensity (100 mW cm2) was calibrated with the reference solar cell and meter (Oriel Instrument, model 91105). Photocurrent density-voltage curves of the DSSCs were obtained with a potentiostat/galvanostat (PGSTAT 30, Autolab, Eco-Chemie, the Netherlands). Electrochemical impedance spectra (EIS) were obtained by the above-mentioned potentiostat/galvanostat, equipped with an FRA2 module, under a

constant light illumination of 100 mW cm2. The frequency range explored was set at 10 mHz to 65 kHz. The applied bias voltage was set at the open-circuit voltage of the DSSC, between the counter electrode and the FTO-TiO2-dye working electrode, starting from the short-circuit condition; the corresponding ac amplitude was 10 mV. The impedance spectra were analyzed using an equivalent circuit model. The ionic conductivity (sS) of the synthesized ILs was determined from EIS analysis with the FTO/Pt/electrolyte/Pt/FTO symmetric cell using the following formula:

sS ¼ dS =ðAS  RS Þ

(1)

The device constant (dS/AS) was calculated from a standard cell calibration based on standard NaCl solution of 12.9 mS cm1 (Model 011006, Thermo Orion), and the ohmic serial resistance (RS) was obtained from Nyquist plot. The incident photo-to-current conversion efficiency (IPCE) curves were obtained at the shortcircuit condition. The light source was a class A quality solar simulator (PEC-L11, AM1.5G, Peccell Technologies, Inc.); light was focused through a monochromator (Oriel Instrument, model 74100) onto the photovoltaic cell. The monochromator was incremented through the visible spectrum to generate the IPCE (l) as defined below,

IPCE ðlÞ ¼ 1240 ðJSC =l4Þ

(2)

where l is the wavelength, JSC is the short-circuit photocurrent density (mA cm2) recorded with a potentiostat/galvanostat, and 4

170

H.-H. Lin et al. / Journal of Power Sources 311 (2016) 167e174

is the incident radiative flux (W m2) measured with an optical detector (Oriel Instrument, model 71580) and power meter (Oriel Instrument, model 70310). The density was determined by measuring the weight of prepared ILs (5.0 mL) at 25  C. Viscosity was measured with an AR2000ex stress-controlled rheometer (TA Instruments, USA) with parallel plate geometry at room temperature. Thermal measurement was performed with Thermo Plus 2 TG-8120 Dynamic TG (Rigaku Corporation, Japan).

3.2. Photovoltaic characteristics of the DSSCs with binary ionic liquid based electrolytes The photovoltaic performance of the bi-IL based DSSCs are presented in Fig. 1 and the corresponding parameters are shown in Table 3. Among all cases, NMP-2C and TEA-2C show the highest cell performance of 5.36% and 4.99%, respectively. Further, the NMPbased DSSCs show higher cell performance of DSSCs than those of the TEA-based cells, ascribing to the high JSC value of NMP-based cells, as a result of their high conductivity. Further, the cell

3. Results and discussion 3.1. Physical properties of RTILs The physico-chemical properties of synthesized ILs (Table 1) such as density, decomposition temperature (Tdecom), viscosity, and conductivity are displayed in Table 2. Among the different ILs, NMP8C and TEA-8C show high density as a result of increase in their chain length. Further, the NMP-based ILs show lower density than the TEA-based one. The changes in the thermal Tdecom measured by TG-DTA for the ILs are shown in Fig. S4 and the corresponding values are tabulated (Table 2). The table shows that the ILs possess Tdecom in the range of 168e177  C, indicating their superior thermal stability. Moreover, viscosity increases with the increase of alkyl chain length for the ILs based on PFC anions [32]. The lowest viscosities of 36.2 cP and 23.7 cP were noted for NMP-2C and TEA-2C respectively. On the other hand, NMP-based ILs show higher conductivities than the TEA-based ILs and the conductivity decreases with the increase in chain length of alkyl group in both classes of ILs. Highest conductivities of 21.41 mS cm1 and 15.96 mS cm1 were noted for NMP-based and TEA-based ILs, respectively. The difference in the viscosity and conductivity may be associated with various geometrical shapes and steric hindrance of the individual interacting sites of ILs, which cause a difference in their polarity leading to variation in the electron-pair accepting ability, as evidenced by Tokuda et al. [34,35]. Besides, the plots of conductivity vs. temperature for comparing the activation energies of pure ionic liquids (Ea) are shown in Fig. S5. The plots of conductivity vs. temperature for the ionic liquids studied are essentially linear in the temperature range of 20  Ce100  C. As shown in Table 2, the Arrhenius fits of the data in Fig. S5 give activation energies of 17.8 kJ mol1, 19.5 kJ mol1, 23.8 kJ mol1, 27.6 kJ mol1, 20.1 kJ mol1, 22.8 kJ mol1, 26.0 kJ mol1, and 31.4 kJ mol1 for the conductivities of NMP-2C, NMP-4C, NMP-6C, and NMP-8C, TEA-2C, TEA-4C, TEA6C, and TEA-8C, respectively. The trend in the activation energies is consistent with the apparent viscosities of the ionic liquids. Interestingly, NMP-2C has a lower activation energy and higher conductivity than TEA-2C, although it has higher apparent viscosity.

Table 2 Physico-chemical properties of ionic liquids.

Fig. 1. Photovoltaic properties of the DSSCs composed of the bi-IL based electrolytes (a) electrolytes which contain PMII and NMP based ionic liquids (b) electrolytes which contain PMII and TEA based ionic liquids.

ILs

Density (g cm3)

Tdecoma ( C)

Viscosity (cP)

Conductivityb (mS cm1)

Activation energyc (kJ mol1)

NMP-2C NMP-4C NMP-6C NMP-8C TEA-2C TEA-4C TEA-6C TEA-8C

1.51 1.60 1.72 1.83 1.61 1.69 1.78 1.89

171.6 173.9 175.0 177.6 168.8 171.8 173.5 175.2

36.2 70.8 169.0 429.8 23.7 47.0 90.4 232.7

21.41 16.82 11.39 7.48 15.96 12.30 8.32 5.10

17.8 19.5 23.8 27.6 20.1 22.8 26.0 31.4

a b c

Tdecom was determined by the peak temperature of DTA measurement. Conductivity was measured by EIS using the cell configuration of FTO/Pt/electrolyte/Pt/FTO. The activation energy was determined by the Arrhenius fit of the data in Fig. S5.

H.-H. Lin et al. / Journal of Power Sources 311 (2016) 167e174

171

Table 3 Photovoltaic performances of the bi-IL based DSSCs. Bi-IL based electrolytes with

JSC

NMP-2C NMP-4C NMP-6C NMP-8C TEA-2C TEA-4C TEA-6C TEA-8C

12.05 9.23 7.45 5.68 10.46 8.35 7.06 4.61

VOC ± ± ± ± ± ± ± ±

0.07 0.13 0.12 0.18 0.15 0.17 0.16 0.21

630.9 630.2 628.9 626.3 657.0 644.3 636.4 622.0

ha

FF ± ± ± ± ± ± ± ±

4.6 4.3 4.5 3.7 6.2 5.6 5.3 5.2

0.71 0.71 0.70 0.70 0.72 0.71 0.69 0.68

± ± ± ± ± ± ± ±

0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01

5.36 4.13 3.28 2.49 4.99 3.82 3.10 1.95

± ± ± ± ± ± ± ±

0.10 0.12 0.12 0.19 0.13 0.14 0.17 0.16

a The power conversion efficiency (h) was calculated using the following relationship: h ¼ (JSC  VOC  FF) (Pin)1  100%, where JSC is the short-circuit photocurrent density, VOC is the open-circuit voltage, FF is the fill factor, and Pin is the light power per unit area.

performance decreases with the increase of carbon chain length of PFC irrespective of nature of cation present in the ILs (pertinent cell performance decreases from 5.36%, 4.13%, 3.28%, to 2.49% for NMP2C, -4C, -6C, to -8C based DSSCs). The result may be ascribed to the fact that the NMP-based PFC ILs having extended carbon chain length show low ionic conductivity leading to a large charge transfer resistance (Rct2) and Warburg resistance (ZW) in the electrolyte and as a result of this, the DSSC shows low JSC value. On the other hand, high viscosity ILs show low penetration ability into TiO2 films, which may retard the reaction of I/I 3 redox couple. Similar tendency can be seen in the cases of TEA-based DSSCs, where the cell performance decreases from 4.99%, 3.82%, 3.10%, to 1.95% for ILs such as TEA-2C, -4C, -6C, to -8C with the increase of alkyl chain length. It is noted that TEA-based DSSCs show slightly higher VOC values (622.0 mVe657.0 mV) than those of the NMP-based ILs (626.3 mVe630.9 mV). Besides, the VOC value decreases with the increase of carbon chain length of ILs. Fig. 2 shows the dark current measurement for the DSSCs based on bi-IL electrolytes. This measurement related to the characteristic of DSSC which prevents the electrons from being captured by electrolyte. The VOC value would be high if electrons find difficulty in their recombination. The present study reveals that extended carbon chain length leads to the lower VOC values, it could be as a result of high charge recombination. As evidenced from Fig. 2a and b, lower recombination takes place in DSSCs with TEA-based electrolytes than that of NMP, and the result also matches the variations in the VOC values. 3.3. Photovoltaic characteristics of the DSSCs with ternary ionic liquid based electrolytes In order to further enhance the cell performance, a ter-IL based electrolyte system containing different volume ratios of NMP-2C, TEA-2C and PMII was proposed by combining the advantages of NMP-based (high JSC value) and TEA-based ILs (high VOC value). A volume ratio of 35/65 for the case of ter-ILs (NMPþTEA)/PMII is chosen according to our previous studies [27,32,33]. The photovoltaic performances of the ter-IL based DSSCs are shown in Fig. 3 and the pertinent parameters are listed in Table 4. When the volume ratio of TEA-2C in the ter-IL based electrolytes becomes high, the relative VOC value becomes large due to the superior characteristic of TEA cations. On the other hand, when the volume ratio of NMP-2C in a ter-IL based electrolyte becomes high, the JSC value enhances due to the high conductivity of NMP-2C. According to the above results, an optimized ter-IL based electrolyte was obtained by optimizing the volume ratio of TEA-2C and NMP-2C. Finally, the terIL based electrolyte containing NMP-2C/TEA-2C/PMII (v/v/v ¼ 28/7/ 65) shows the best cell performance of 6.01%, averaged for three cells. Fig. 4 illustrates the IPCE spectra of DSSCs containing different volume ratios of the ter-IL based electrolytes. The integrated photocurrent density calculated from IPCE spectrum (JIPCE) can be estimated by using the following equation [36]:

Fig. 2. Dark current of the DSSCs composed of the bi-IL based electrolytes (a) electrolytes which contain PMII and NMP based ionic liquids (b) electrolytes which contain PMII and TEA based ionic liquids.

Z JIPCE ¼

IPCE ðlÞ efAM1:5G ðlÞ dl

(3)

where the parameter e is the elementary charge and fAM1.5G is the photon flux at AM 1.5G (100 mW cm2). Table 4 indicates the values of JIPCE obtained by this equation. It can be seen that the JIPCE values follow the same trend as that of the JSC values obtained by actual measurements with discrepancy of less than 10%.

172

H.-H. Lin et al. / Journal of Power Sources 311 (2016) 167e174

spectroscopy (EIS) analysis and the result is shown in Fig. 5a and Table 4. There are two semicircles in Fig. 5a, the small one corresponds to the resistance at the counter electrode/electrolyte interface (Rct1); the big one associates with the resistance at the TiO2/dye/electrolyte interface (Rct2). It can be seen from the figure that resistance of Rct1 is found to be relatively small because of its low resistance. As shown in Table 4, the DSSC with NMP-2C/TEA2C/PMII (v/v/v ¼ 28/7/65) based electrolyte shows the lowest Rct2 (12.52 U) and highest JSC, when compared to the DSSCs with different volume ratios of ILs. This is correlated with the appropriate combination of outstanding conductivity of NMP-2C and low charge recombination of TEA-2C. Besides this, the value of Rct2 is related to JSC value, which can be inferred from the conductivity and viscosity values. The conductivity of NMP-2C is higher than that of TEA-2C, but the viscosity of NMP-2C is larger than TEA-2C, and this would cause difficulty in their penetration into TiO2 film. Hence, a high volume ratio of NMP-2C in electrolyte still leads to a large Rct2 value. Fig. 5b shows the Bode plot diagrams of the ter-IL based DSSCs, where frequency peaks relate Rct2 and ZW parameters at the TiO2

Fig. 3. Photovoltaic properties of the DSSCs composed of the ter-IL based electrolytes which contain NMP-2C, TEA-2C and PMII in different volume ratios.

Table 4 Photovoltaic performance of the ter-IL based DSSCs. Ter-IL based electrolytes NMP-2C/TEA-2C/PMII

JSC (mA cm2)

0/35/65 7/28/65 14/21/65 21/14/65 28/7/65 35/0/65

10.46 11.16 12.01 12.40 12.99 12.05

a b c d

± ± ± ± ± ±

0.15 0.08 0.08 0.14 0.16 0.07

JIPCEa (mA cm2)

VOC (mV)

9.64 10.21 10.89 11.57 11.95 11.01

657.0 652.3 649.7 643.3 639.0 630.9

h

FF

(%) ± ± ± ± ± ±

6.2 9.9 7.1 3.5 4.2 4.6

0.72 0.71 0.70 0.71 0.72 0.71

± ± ± ± ± ±

0.01 0.01 0.01 0.01 0.01 0.01

4.99 5.17 5.49 5.64 6.01 5.36

± ± ± ± ± ±

0.13 0.18 0.13 0.13 0.14 0.10

Rct2b ( U)

ZWc (U)

ted

18.10 15.60 13.57 13.36 12.52 12.99

37.32 32.43 30.38 26.32 25.43 25.86

9.11 7.92 6.88 5.99 5.21 4.54

(ms)

JIPCE is calculated by Equation (3). Rct2 is the resistance of electrolyte which is measured by DSSC. ZW is the resistance of electrolyte which is measured by a symmetric cell (FTO/Pt/electrolyte/Pt/FTO). te is calculated by the equation te ¼ (2pfmax)1, fmax is calculated from Fig. 5b.

Fig. 4. IPCE action spectra of DSSC containing different volume ratios of ionic liquid in the ter-IL based DSSCs.

3.4. Electrochemical impedance characteristics of the DSSCs with ter-IL based electrolyte The charge transfer resistance behavior in DSSCs with the ter-IL based electrolytes is investigated using electrochemical impedance

film/electrolyte interface and the bulk electrolyte, respectively. The maximum frequencies of charge transfer (fmax) is also marked in the figure. The electron lifetime (te) for a recombination can be calculated by te ¼ (2pfmax)1, and te values are shown in Table 4. Typically, a large te value corresponds to a high VOC value because of its lower electron recombination leading to an increase of charge transportation in the TiO2 films. The result shows that, with the TEA-2C content increases, the te becomes long, resulting in large VOC value. It is understandable that the bi-IL based electrolyte containing higher volume ratio of TEA-2C corresponds to larger VOC value due to the longer te. The charge transfer resistance of the DSSCs based on ter-IL electrolytes is investigated using EIS of symmetric cells (FTO/Pt/ electrolyte/Pt/FTO) and the result is shown in Fig. 5c. The EIS spectra of symmetric cells show two circles. The first circle with high frequency corresponds to the charge transfer resistance (Rct) between counter electrode and electrolyte, and the second circle with lower frequency corresponds to ZW of I/I 3 in the electrolyte. In Fig. 5c, the ter-IL based DSSC with electrolytes NMP-2C/TEA-2C/ PMII in the volume ratio of 28/7/65 shows the lowest ZW among the electrolytes containing different compositions, which reveals that the resistance is the lowest at this optimum volume ratio. Fig. 5d shows the durability data of the ter-IL based DSSC with the electrolyte NMP-2C/TEA-2C/PMII in the volume ratio of 28/7/ 65. In this experiment, the cells were sealed by Surlyn® (25 mm) and UV glue to enhance the long-term stability. The cells were stored in dark at 50  C, and efficiency was measured once at different periods

H.-H. Lin et al. / Journal of Power Sources 311 (2016) 167e174

173

Fig. 5. (a) Nyquist plots of EIS of the DSSCs contain different volume ratios of ionic liquid in the ter-IL based DSSCs. (b) Bode plots of DSSCs with different volume ratios of ionic liquid in the ter-IL based electrolytes, measured under 100 mW cm2. (c) EIS spectra of the symmetric cell (FTO/Pt/electrolytes/Pt/FTO) with different volume ratios of ionic liquid in the ter-IL based electrolytes. (d) Durability data of the ter-IL based cell (NMP-2C/TEA-2C/PMII ¼ 28/7/65) which is stored in 50  C for 1200 h.

of time, and the result shows that the DSSC exhibits an extraordinary durability and unfailing cell efficiency up to 1200 h.

Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jpowsour.2016.02.029.

4. Conclusions In summary, eight novel ILs containing PFC as anions with different carbon chain lengths and TEA or NMP as cations were synthesized and their physico-chemical properties were determined. Among the different ILs, NMP-2C and TEA-2C possesses the highest ionic conductivity and the lowest viscosity respectively. Solar cells containing NMP based ILs show higher VOC than that of TEA, whereas, higher JSC was noted for the DSSCs incorporated with the latter when compared to the former. The photovoltaic characteristics of the DSSCs with NMP-2C/PMII and TEA-2C/PMII binary IL electrolytes exhibit cell efficiencies of 5.36% and 4.99% respectively, whereas, the DSSC with NMP/TEA/PMII ter-IL electrolytes shows the highest performance of 6.01%. Impedance analysis shows that the ter-IL based DSSCs can effectively decrease Rct2 and ZW when compared to that of the bi-IL based DSSCs. Finally, the long-term stability of the ter-IL based DSSC stored in dark at 50  C is proved to be unfailing up to 1200 h.

Acknowledgements

References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16]

This work was supported in part by the Ministry of Science and Technology (MOST) of Taiwan under grant numbers 100-2923-E002-004-MY3 and 102-2221-E-002-186-MY3. Dr. V. Suryanarayanan and Dr. D. Velayutham thank GITA (CII), New Delhi for the international travel grant under DST Indo-Taiwan Exchange Programme.

[17] [18] [19] [20]

€tzel, J. Photochem, Photobiol. A Chem. 164 (2004) 3e14. M. Gra B. O'Regan, M. Gr€ atzel, Nature 353 (1991) 737e739. €tzel, Nature 414 (2001) 338e344. M. Gra K. Kakiage, Y. Aoyama, T. Yano, K. Oya, J. Fujisawa, M. Hanaya, Chem. Commun. 51 (2015) 15894e15897. E. Stathatos, P. Lianos, Adv. Mater. 14 (2002) 354e357. J.H. Wu, S.C. Hao, Z. Lan, J.M. Lin, M.L. Huang, Y.F. Huang, L. Fang, S. Yin, T. Sato, Adv. Funct. Mater. 17 (2007) 2645e2652. W. Kubo, K. Murakoshi, T. Kitamura, S. Yoshida, M. Haruki, K. Hanabusa, H. Shirai, Y. Wada, S. Yanagida, J. Phys. Chem. B 105 (2001) 12809e12815. A. Hess, G. Barber, C. Chen, T.E. Mallouk, H.R. Allcock, ACS Appl. Mater. Interfaces 5 (2013) 13029e13034. S.J. Lue, Y.L. Wu, Y.L. Tung, C.M. Shih, Y.C. Wang, J.R. Li, J. Power Sources 274 (2015) 1283e1291. N. Yamanaka, R. Kawano, W. Kubo, N. Masaki, T. Kitamura, Y. Wada, M. Watanabe, S. Yanagida, J. Phys. Chem. B 111 (2007) 4763e4769. P.Z. Yang, Q.W. Tang, Electrochim. Acta 188 (2016) 560e565. J. Yoon, D.K. Kang, J. Won, J.Y. Park, Y.S. Kang, J. Power Sources 201 (2012) 395e401. S.S. Yang, Q.W. Tang, B.L. He, L.M. Yu, J. Power Sources 273 (2015) 1148e1155. Q.H. Li, Q.W. Tang, B.L. He, P.Z. Yang, J. Power Sources 264 (2014) 83e91. ^te, H. Pettersson, N. Papageorgiou, Y. Athanassov, M. Armand, P. Bonho €tzel, J. Electrochem. Soc. 143 (1996) 3099e3108. A. Azam, M. Gra H. Matsumoto, T. Matsuda, T. Tsuda, R. Hagiwara, Y. Ito, Y. Miyazaki, Chem. Lett. 30 (2001) 26e27. M. Gorlov, L. Kloo, Dalton Trans. (2008) 2655e2666. M. Ghavre, O. Byrne, L. Altes, P.K. Surolia, M. Spulak, B. Quilty, K.R. Thampi, N. Gathergood, Green Chem. 16 (2014) 2252e2265. R. Kawano, H. Matsui, C. Matsuyama, A. Sato, A.B.H. Susana, N. Tanabe, M. Watanabe, J. Photochem. Photobiol. A Chem. 164 (2004) 87e92. J.M. Pringle, J. Golding, C.M. Forsyth, G.B. Deacon, M. Forsyth, D.R. MacFarlane, J. Mater. Chem. 12 (2002) 3475e3480.

174

H.-H. Lin et al. / Journal of Power Sources 311 (2016) 167e174

[21] M.C. Buzzeo, R.G. Evans, R.G. Compton, ChemPhysChem 5 (2004) 1106e1120. [22] H. Tokuda, K. Hayamizu, K. Ishii, A.B.H. Susan, M. Watanabe, J. Phys. Chem. B 108 (2004) 16593e16600. [23] P.K. Singh, K.W. Kim, H.W. Rhee, Polym. Eng. Sci. (2009) 862e865. [24] M. Berginc, U.O. Krasovec, M. Ho cevar, M. Topi c, Thin Solid Films 516 (2008) 7155e7159. [25] K. Kalyanasundaram, Dye Sensitized Solar Cells, Fundamental Sciences Chemistry, EPFL Press, 2010. €tzel, J. Am. Chem. Soc. 128 [26] D. Kuang, P. Wang, S. Ito, S.M. Zakeeruddin, M. Gra (2006) 7732e7733. [27] C.P. Lee, J.D. Peng, D. Velayutham, J. Chang, P.W. Chen, V. Suryanarayanan, K.C. Ho, Electrochim. Acta 114 (2013) 303e308. [28] F. Hao, H. Lin, Y.Z. Liu, G. Yang, G. Wang, J. Li, Electrochim. Acta 56 (2011) 5605e5610.

[29] M. Bidikoudi, L.F. Zubeir, P. Falaras, J. Mater. Chem. A 2 (2014) 15326e15336. [30] S.H. Lee, Y.T. Jeon, Y.D. Lim, A. Hossain, S.Y. Lee, Y.G. Cho, H.C. Ju, W.G. Kim, Electrochim. Acta 107 (2013) 675e680. [31] E. Vijayakumar, A. Subramania, Z.F. Fei, P.J. Dyson, J. Appl. Polym. Sci. 132 (2015), 42032/1e7. [32] C.P. Lee, P.Y. Chen, R. Vittal, K.C. Ho, J. Mater. Chem. 20 (2010) 2356e2361. [33] J.D. Peng, C.P. Lee, D. Velayutham, V. Suryanarayanan, K.C. Ho, J. Mater. Chem. A 3 (2015) 6383e6391. [34] H. Tokuda, K. Hayamizu, K. Ishii, Md. A. B. H. Susan, M. Watanabe, J. Phys. Chem. B 109 (2005) 6103e6110. [35] H. Tokuda, K. Ishii, A.B.H. Susan, S. Tsuzuki, K. Hayamizu, M. Watanabe, J. Phys. Chem. B 110 (2006) 2833e2839. [36] A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo, H. Pettersson, Chem. Rev. 110 (2010) 6595e6663.