Chapter 4 Towards efficient tetrapod-like ZnO dye-sensitized solar cells
4.4 Efficient electron transporting in tetrapod-like ZnO photoanode
In this section, the EIS measurement is also employed to analyze the performance
of the tetrapod porous network photoanode, and the Nyquist Cole–Cole plot of the 42.2 μm tetrapod-like ZnO (T-ZnO) DSC is shown in Fig. 4-9. The effective electron
transporting coefficient (Deff) of 42.2 μm T-ZnO DSC is 1.533 x10-3 cm2 sec-1 which is
61
much larger than that of traditional TiO2-based DSCs (about 10-5 cm2 sec-1).[75] And the effective electron diffusion length for the 42.2 μm T-ZnO DSC is estimated by the relation: Ln = (Deff x τn), to be 45.57 μm. It means the optimal thickness of photoanode is around 46 μm that is consistent with the results of thickness dependent JSC measurement discussed in last section. The series resistance (RStot) estimated from EIS fitting is 30.2 Ω via the well-known relation (Eq. 4-3) [76] indicating consistent with 28.9 Ω calculated from the I-V curve of Fig. 4-6 (a).
(Eq. 4-2)
Fig. 4-7. Nyquist plot of the 42.2 μm tetrapod-like DSCs sensitized by Di49 dye. The empty circles in are the measurement data points, and the solid curve is the fitting result based on the equivalent circuit model as shown in Fig 2-7. The data was collect by applying a bias of the open circuit voltage (VOC) under solar simulator (AM 1.5G, 100 mW cm-2). The electrolyte (EL 1) of the device was a mixture of
0.5 M PMII, 0.03 M I2, and 0.5 M TBP in AN.
62
4-5 Influence of photoanode thickness on the photovoltaic performance of ionic liquid device
In order to obtain the high performance DSCs, the thickness of T-ZnO photoanode were varied deposited to 42 μm in pervious section. Under one sun irradiation, the results of photovoltaic characteristics with filling AN-based and IL-based electrolytes by varying different PE thickness are shown in Fig. 4-8. AN-based electrolyte (EL 1) is a mixture of 0.5 M PMII, 0.03 M I2, and 0.5 M TBP in AN. IL-based electrolyte (EL 2) is composed by 0.2 M I2, 0.5M TBP in PMII/C8MImPF6 mixture (v/v=35:65).
Figure 4-8(a) gives the open-circuit voltage (VOC) results for DSCs with two types of electrolytes, the VOC of both types decrease linearly with the increase of ZnO film thickness. The inhomogeneous light density through the porous ZnO films contributes an inhomogeneous quasi-Fermi level of photoanode. A 200mV VOC drops in IL-based DSCs relative to that of AN-based for the same thickness is due to the higher I2 concentration in IL-based DSCs, and can be explained by charge recombination model [55] between photoanode and the I /3 I couples in electrolyte with the following equation,
, (4-3)
where k is Boltzmann constant, T is the absolute temperature, q is the electronic elementary charge, Iinj is the flux of charge from sensitized dyes, ncb is electron
63
concentration in the ZnO, ket is the reaction rate constant of the dark current from ZnO to tri-iodide ions in the electrolyte, and [I3] is the concentration of tri-iodide ions in the electrolyte. The behaviors of the short-circuit photocurrent (JSC) are illustrated in Fig. 4-8(b). JSC of AN-based DSCs increases continuously with film thickness, and reaches a maximum value of 11.4 mA cm-2 at thickness of 26 μm. In contrast, JSC of IL-based DSCs has a plateau value close to 5 mA cm-2 in the device whose thickness is about 18 μm.
Fig. 4-8. Dependences of cell performances on film thickness including (a) open-circuit voltage (VOC), (b) short-circuit photocurrent density (JSC), (c) filling factor (FF), and (d) solar conversion efficiency.
The open-circle and open-square symbols correspond to the DSCs filling by AN-based and IL-based electrolytes, respectively. The lines are plotted to guide the eyes, and the error bars from the data of two devices are also shown in the figures. AN-based electrolyte (EL 1) is a mixture of 0.5 M PMII, 0.03 M
I2, and 0.5 M TBP in AN. IL-based electrolyte (EL 2) is composed by 0.2 M I2, 0.5M TBP in PMII/C8MImPF6 mixture (v/v=35:65). All data was obtained under one sun irradiation.
64
The FF remains a constant for the AN-based DSCs, while the IL-based DSCs show a prominent decline in FF with increasing film thickness. The different behavior observed in JSC and FF results from the higher viscosity of the ionic liquid that limits the mass-transport of the I /3 I couples in electrolyte and increases the resistance of the electrolyte. The maximal conversion efficiency of the AN-based and IL-based DSCs, which can be calculated by under one sun irradiation, are 4.73% and 1.39% for 26 μm and 12 μm thick T-ZnO, respectively.
Therefore, in the IL-based DSCs, the thickness of the photoanode isn‟t the strongest influence factor to optimize the device; on the contrary, the characteristics of the ionic liquid and the morphology of the photoanode should be paid more attention to.
To realize the differences between the T-ZnO framework as the PE and the conventional PE made of spherical C-ZnO nanopowders, the J-V curves of these two DSC devices are shown in Fig. 4-9 (a). The photovoltaic parameters obtained with AN-based DSCs using 26 μm T-ZnO photoanode (open square symbol) are VOC=0.622V, JSC=11.55 mA cm-2, FF=0.67, and η=4.81%. In IL-based DSCs, however, the lower performances are obtained, in which the photovoltaic parameters of
26 μm T-ZnO DSCs (open triangle symbol) are VOC=0.448V, JSC=5.37 mA cm-2, FF=0.48, and η=1.15%. And those of 26 μm C-ZnO DSCs (solid triangle symbol) are
VOC=0.444V, JSC=3.95 mA cm-2, FF=0.49, and η=0.87%. The high photocurrent of
65
11.55 mA cm-2 are attributed to the high absorption coefficient of D149 in AN-based system, but the photocurrent in IL-based system (triangle symbol) reduces to less than a half of that in AN-based system (square symbol). It is believed that the higher viscosity of the IL-based electrolyte leads to the ionic-transport limitation of the photocurrent. On the other hand, the photocurrent of IL-based DSCs using T-ZnO photoanode is almost 1.3 times larger than that using C-ZnO photoanode, due to the lager pore size in the tetrapod-like framework provides an efficient ionic diffusion pathway to improve the ionic-transport limitation. Moreover, the IPCE spectra provides a conversion efficiency measurement of wavelength-dependent photons that incident on the DSCs to photocurrent flowing between the photoanode and counter electrodes. The IPCE spectra of these devices are shown in Fig. 4-9(b). High IPCE are obtained with T-ZnO DSCs filled with AN-based electrolyte; the peak values are observed at about 550 nm and 390 nm, which are contributed from the absorption peak of D149 dye and the absorption of ZnO, respectively. Although D149 has a strong green light absorption (68700 Mol-1 cm-1 at 526 nm), the plateau IPCE value about 30%
is observed from 500 nm to 600 nm in T-ZnO DSCs filled with IL-based electrolyte. It results in the slow ionic diffusion in the high velocity IL electrolyte that slows down the oxidation-reduction reaction of the dye molecule. Furthermore, the IPCE reduction of all spectral range is also observed. Additionally, almost 4 times IPCE decreasing from
66
400 to 450 nm between two electrolyte systems is also due to the light absorption of higher iodine concentration in the IL-based system [77]. JSC agrees with the value integrated by IPCE spectrum with standard AM 1.5G sunlight, the reasons from IPCE can explain the JSC difference of 26 μm-thick DSCs beteen AN-based and IL-based electrolytes.
Fig. 4-9. Cell performances of AN-based (square symbol) and IL-based (triangle symbol) DSCs based on 26μm-thick ZnO nanostructure photoanodes that were constructed by commercial (solid symbol, C-ZnO)
and tetrapod-like (open symbol, T-ZnO) ZnO powders. (a) The J-V curve of these DSC devices (100 mWcm-2, AM 1.5G) and (b) The IPCE spectra of these DSC devices. AN-based electrolyte (EL 1) is a
mixture of 0.5 M PMII, 0.03 M I2, and 0.5 M TBP in AN. IL-based electrolyte (EL 2) is composed by 0.2 M I2, 0.5M TBP in PMII/C8MImPF6 mixture (v/v=35:65).
67