Research Express@NCKU - Articles Digest
Research Express@NCKU Volume 12 Issue 7 - February 5, 2010 [ http://research.ncku.edu.tw/re/articles/e/20100205/2.html ]
A Hybride PVDF-HFP/Nanoparticles Gel Electrolyte
for Dye-Sensitized Solar Cell Application
Yuh-Lang Lee
*, Yu-Jen Shen, Yu-Min Yang
Department of Chemical Engineering, College of Engineering, National Cheng Kung University [email protected]
Nanotechnology 19 455201 (2008)
I
ntroductionDye-sensitized solar cells (DSSCs) have received significant attention in the past decade and energy conversion efficiencies up to 10-12 % have been reported for the DSSCs. However, such efficiency can only be obtained using a liquid electrolyte. The presence of a liquid electrolyte in a DSSC raises several practical problems including the solvent evaporation and leakage of the liquid electrolyte.Many efforts have been made to solve the problems by replacing the liquid electrolyte with p-type
semiconductors, organic hole-transport materials, and gel materials
incorporating I3-/I- as a redox couple. However, the energy conversion efficiencies of the solid state or
quasi-solid-state DSSCs were lower than the liquid versions.
Poly(vinylidene fluoride-co-hexafluoropropylene) (PVDF-HFP) and its derivatives were commonly used to solidify a liquid electrolyte for DSSCs applications. The small ionic radius and large electronegativity of the fluorine atoms in the PVDF-HFP are expected to increase the ionic conductivity in the solid electrolyte. To improve the interfacial and conductivity properties of polymer gel/solid electrolytes, nanoparticles such as TiO2 and SiO2 were introduced into the polymer matrix. These inorganic fillers were known to reduce the crystallinity of the polymer, enhance the ionic conductivity, and therefore, increase the overall conversion efficiency of a solid-state DSSCs.
In this work, graphite nano-particle (GNP) was used as an alternative filler of a PVDF-HFP based polymer gel electrolyte (PGE) for the DSSC applications. The GNP has been proved to be an efficient filler to enhance the energy conversion efficiency of a DSSC.
Experimental Section
TiO2 paste (Degussa P25) was spin-coated on the indium-tin oxide (ITO, about 13 Ω/sq) substrate and sintered at 450 ˚C for 30 minutes. The thickness of the TiO2 film was about 5.5 μm. The TiO2 substrate was immersed in an ethanol solution containing 0.3 mM Ruthenium-535-bis-TBA (Solaronix, N719 dye) for 20 h at room
temperature. The gel electrolyte was prepared by mixing PVDF-HFP (Atochem, KynarFlex 2801) and a 3-methoxypropionitrile (MPN) solution containing 0.1 M LiI, 0.05 M I2, 0.5 M 4-tert-butylpyridine (TBP), and 0.5 M 1-propyl-2,3-dimethylimidazolium iodide (DMPII). In this study, GNP (Alfa Aeser) and TiO2 (Degussa P25) are introduced to enhance the performance of the gel electrolyte.
The TiO2 photoelectrode and a Pt-coated counter electrode were sandwiched using a 60 μm thick sealing material (SX-1170-60, Solaronix SA) under pressure at about 100 ˚C. Two holes were left on the sealing spacer for
Research Express@NCKU - Articles Digest
electrolyte injection. The hot gel electrolyte (80 ˚C) was filled into the space of the cell using a vacuum pump, followed by sealing the holes using epoxy resin. The active area of the cell was 0.16 cm2. The performances of the
cells were measured under an illumination of a solar simulator (Newport, Oriel class A, 91160A) at 100% sun (AM1.5, 100 mW/cm2).
Results and Discussion
Figure 1a shows the I-V characteristics of PGE-DSSCs using GNP as a filler. The overall energy conversion efficiency of the cell using a liquid electrolyte is 6.39%. The efficiency of the cell decreases to 4.69% when 10 wt % of PVDF-HFP was added. When GNP was introduced, the efficiency increases first with increasing content of GNP and the maximum efficiency (η = 6.04 %) obtained at 0.25 wt% GNP. With further increase of GNP content, the efficiency decreases steadily. These results indicate that introduction a tiny amount of GNP (below 0.5 wt%) did have significant effect in improving the performance of the PGE-based DSSCs. Compared with the system without filler, both ISC and VOC increase due to the GNP introduction. The I-V characteristics of the PGE-DSSCs using TiO2 filler are shown in Figure 1b. These results show that introduction a small amount of TiO2
nanoparticles also has an effect to enhance the performance of the PGE-DSSCs. However, the promoting effect of TiO2 filler is not as significant as that of GNP. The maximum efficient of the PGE-DSSC using TiO2 filler is 5.29 % obtained at a TiO2 concentration of 0.5 wt%.
Figure 1. I-V characteristics of DSSCs prepared using PVDH-HFP-based PGE containing various concentrations of graphite (a) and TiO2 (b). The measurement was performed under the illumination of one sun (AM1.5, 100 mW/ cm2).
To study the role of the inorganic fillers on the efficiency enhancement of the PGE-DSSCs, the diffusivity of I3- ion in the PGE was measured first.
The diffusion coefficients measured for the gel electrolytes were shown in Figure 2. The presence of nanoparticles in a solid state polymer electrolyte was known to reduce the crystallization of a polymer. For a PGE in which the polymer was not solidified
completely, the added nanoparticle may also play a role to reduce the interaction between polymer chains
Research Express@NCKU - Articles Digest
Figure 2. Diffusivity of I3- in the PVDF-HFP-based PGE as a function of filler concentration
and inhibit the local crystallization of the polymer in a PGE. This effect is inferred to be responsible for the increase of ion diffusivity at the presence of TiO2. However, the slight increase of the diffusivity for the
GNP is not consistent to the large increase of the short-circuit current found for the GNP-contained devices, indicating the importance of other effects.
Table 1. The conductivities of the PGE at 30 ˚C for various filler concentrations.
Table 1 shows the conductivity of the PGE measured at room temperature.The effect of GNP addition on conductivity enhancement is much significant than its effect on the I3- diffusivity. Similar results were also
found in the literature. It was reported that a gelation process decreases the diffusion coefficient of iodide ions to ca. one quarter of that observed in acetonitrile, but did not significantly influence the charge transport or ionic conductivityof the electrolyte. Eguchi and co-worker also found that increase the concentration of I3
-in the electrolyte decreases the diffusion constant of I3-, but the impedance of the electrolyte decreases,
indicating an increase of the ion conductivity. These results suggest that diffusivity is not the only factor determining the charge transport in the electrolyte and other mechanisms should be involved. In the literature, a widely accepted mechanism contributes to the charge transport in an I-/I3- redox system is the charge exchange
reaction (I- + I3- → I3- + I-), i.e. the Grotthus-type charge-transfer mechanism. The electron exchange between
redox couples was reported to have important contribution to the electronic conduction of a PGE system. In the present system the electron exchange could occur between I-, I3-, and polyiodides, coupling chemical bond
exchange and electron hopping processes. The Grotthus-type charge-transfer mechanism has been used to explain the performance enhancement of a gel-electrolyte containing TiO2 nanofiller. In was proposed that the adsorbed cations on the TiO2 surface can align the anionic redox couples by electrostatic force, forming an electron transport path to facilitate the charge-exchange mechanism.
Furthermore, because ion conductivity is determined by all ionic species presented in the electrolyte including anionic and cationic ions, the higher conductivity of GNP-contained electrolyte may be attributed to species other than I3-. It is well known that cations such as Li+ and DMPII ions prefer to adsorb on TiO2 surface, which is
supposed to decrease the concentration of cations in the bulk phase. Therefore, the free cation concentration in a PGE containing TiO2 filler is lower than that in a GNP-contained PGE, which may be a possible reason leading to a higher conductivity of GNP system. In addition, the inherent conductivity of GNP may also contribute to the conductivity of the PGE.
The EIS measurements were carried out in the dark conditions by applying a sinusoidal perturbation of 10 mV at -0.65V. The Nyquist plots of the EIS spectra were shown in Figure 3 for the PGE-DSSCs without and with 0.25 wt% fillers. All the spectra exhibit three semicircles, which were assigned, from left to right, to the electrochemical reaction at the Pt counter electrode (Z1), the reaction at the TiO2/dye/electrolyte interface
Research Express@NCKU - Articles Digest
Figure 3. Nyquist plots of the EIS spectra measured under dark conditions for DSSCs prepared using PVDH-HFP-based PGEs with and without fillers. The solid lines are the fitting results for obtaining the real part of the impedances
(Z2), and the diffusion of I3- at the Pt/electrolyte
interface (Z3).
The effect of GNP on decreasing of the electron recombination at the TiO2-photoelectrode/electrolyte interface can be attributed to the high charge transport properties of the GNP-contained PGE. It is inferred that the TiO2 filler in the PGE prefers to adsorb onto the TiO2 photoelectrode, leading to a better contact between PGE and TiO2 photoelectrode. However, the
direct contact of theTiO2 filler to a dye-sensitized electrode is supposed to be detrimental to the performance of a DSSC device. Due to the electron collection property of TiO2 filler (P25), the adsorbed filler particles may act as a bridge for an excited electron to recombine with oxidized species in the PGE. This effect is taken as a
responsibility of the lower VOC and efficiency of the device using TiO2 filler.
The results of EIS analysis indicate that introduction of GNP filler can enhance the charge conductivity of the PGE and inhibit the electron recombination at the TiO2-photoelectrode/electrolyte interface. The two effects are
responsible for the higher performance of the GNP-contained gel-state DSSC. References :
1. Mohammad K N, Filippo D A, Simona F, Annabella S, Guido V, Paul L, Seigo I, Bessho T and Grätzel M 2005 J. Am. Chem. Soc. 127 16835
2. Durrant J R and Haque S A 2003 Nature Mater 2 362
3. Han H W, Liu W, Zhang J and Zhao X Z 2005 Adv. Funct. Mater. 15, 1940 4. Fan F R F, Liu H Y and Bard A J 1985 J. Phys. Chem 89 4418
5. Dabestani R, Wang X L, Bard A J, Campion A, Fox M A, Webber S E and White J M 1986 J. Phys. Chem. 90 2729
6. Kavan L and Grätzel M. 1989 Electrochim. Acta 34 1327
7. Wang P, Zakeeruddin S M, Exnar I and Grätzel M 2002 Chem. Commun. 2972
