nanorod composite photoanode for dye-sensitized solar cells

Current Applied Physics 14 (2014) 294e299

Contents lists available at ScienceDirect

Current Applied Physics journal homepage: www.elsevier.com/locate/cap

Electrochemical properties of TiO2 nanoparticle/nanorod composite photoanode for dye-sensitized solar cells Chang Kook Hong a, b, Young Hee Jung b, Hyung Jin Kim c, Kyung Hee Park d, * a

The Research Institute for Catalysis, Chonnam National University, Gwangju 500-757, Republic of Korea School of Applied Chemical Engineering, Chonnam National University, Gwangju 500-757, Republic of Korea c Department of Advanced & Chemicals, Chonnam National University, Gwangju 700-757, Republic of Korea d The Research Institute of Advanced Engineering Technology, Chosun University, Gwangju 501-759, Republic of Korea b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 4 October 2013 Received in revised form 2 December 2013 Accepted 2 December 2013 Available online 14 December 2013

A unique composite of TiO2 nanoparticles (NPs) and nanorods (NRs) has been used to fabricate a photoelectrode for developing dye-sensitized solar cells (DSSCs) with higher sensitivity. The TiO2 nanorods were synthesized using a mechanical process, in which electrospun TiO2 nanofibers was grinded in a controlled way to obtain uniform size distribution. The characteristics of electron transport, recombination lifetime and charge collection were investigated by intensity-modulated photocurrent spectroscopy (IMPS) and intensity-modulated photovoltage spectroscopy (IMVS). Photoelectrodes prepared with the composites of NRs and NPs showed significant improvements in electron transportation compared to only NP photoelectrodes, which would enhance the photovoltaic performance of DSSCs. IMPS and IMVS measurements show that fast electron transport and slightly decreased recombination lifetime resulted in the improvement of efficiency. The highest energy conversion efficiency obtained from the photoelectrodes fabricated with the as-prepared rutile TiO2 nanofibers at 5 wt% NR content was up to 6.1% under AM1.5G solar illumination. The results demonstrate that the composite nanostructure can take advantage of both the fast electron transport of the nanorods and the high surface area of the nanoparticles. Crown Copyright Ó 2013 Published by Elsevier B.V. All rights reserved.

Keywords: Electrospinning Dye-sensitized solar cells TiO2 electrode Light scattering Photovoltaic performance

1. Introduction The uses of different types of nanostructured materials in dyesensitized solar cells (DSSC) have attracted worldwide attention as a low cost alternative to traditional photovoltaic devices [1,2]. Typically, nanoporous film based on TiO2 nanoparticles (w20 nm of diameter) has shown high energy conversion efficiency due to large amounts of adsorbed dye molecules [3]. But when the photoexcited electrons are excited and pass through the interparticle grain boundaries or suffer surface trapping/detrapping effects, the subsequent electron accumulation can enhance the recombination probability, causing restricted performance of DSSCs [4,5]. It was found that using one-dimensional (1-D) TiO2 (in the form of nanotubes, nanowires, nonofibers and nanorods) as the photoanode material for DSSCs significantly improved the charge collection, by providing a highway to transport photoelectrons, which consequently minimizes the recombination and prolongs the electron lifetime [6]. However, the photoelectro

* Corresponding author. Tel.: þ82 62 530 0322. E-mail addresses: [email protected], [email protected] (K.H. Park).

chemical properties of NP/NR-composite-based electrodes have many considerable factors that must be considered, since decrease of the NP contents leads to smaller surface area for dye adsorption, while the incorporation of NRs can enhance the electron transport rate and light scattering effect [7e11]. Saji et al. reported that the incorporation of 10% nanorod content into a nanoparticle matrix leads to reduced charge transport resistance [12]. This is because the optimum material should have higher surface area, better light scattering properties and higher yield for electron transportation. In this study, we use a double layer in TiO2 photoanodes for taking advantage of both the fast electron transport of NRs and the high surface area of NPs. Dense TiO2 electrodes were prepared with TiO2 NPs on FTO glass, and the NP/NR compositions of the photoanode were varied. This double layer consistently provides superior capability for both dye attachment and charge transport without surface area leakage of TiO2 NPs. DSSCs constructed using TiO2 and variations of the composite electrodes were evaluated for photovoltaic performance. The dye absorption amount and interfacial electron transport were systematically investigated by the UVeVis spectra, intensity-modulated photocurrent spectroscopy (IMPS) and intensity-modulated photovoltage spectroscopy (IMVS).

1567-1739/$ e see front matter Crown Copyright Ó 2013 Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.cap.2013.12.003

C.K. Hong et al. / Current Applied Physics 14 (2014) 294e299

2. Experimental 2.1. Preparation of TiO2 nanorods Titanium nanofiber was fabricated using the electrospinning technique [13]. The electrospinning technique has been recognized as a versatile and effective method for the production of fibers with small diameters and with high surface-to-volume ratio [14]. It has been demonstrated that titanium tetraisopropoxide (TTIP, 98%, Aldrich) can be added directly to an alcohol solution containing poly(vinylpyrrolidone) (PVP, M.W. ¼ 1,300,000 g/mol, Aldrich). To suppress the hydrolysis reaction of the solegel precursor, acetic acid as well as PVP solution in ethanol must be added. 6 mL of TTIP was mixed with 12 mL of acetic acid and 12 mL of ethanol. After 60 min, this solution was added to 30 g of ethanol that contained 10 wt% PVP, followed by magnetic stirring for 24 h. The TiO2 sol was loaded in a syringe equipped with a 24-gauge stainless steel needle, and an electrical potential of 20 kV was applied by a high-voltage power supply. The flow rate was controlled to 50 mL/min using a syringe pump (Model 100, KD Scientific, USA), and the working distance was fixed at 15 cm. The prepared electrospun fiber was calcinated at 450e850  C. As-spun TiO2 nanofibers were formed with diameter in the range of 400e500 nm, which was decreased to 300e400 nm after annealing due to the removal of organic fraction. The TiO2 nanorods were obtained by grinding the electrospun TiO2 nanofibers mechanically. 2.2. DSSC fabrication A transparent film made of a TiO2 paste with about 18-nm-sized particles (18-NRT, Dyesol) was deposited by screen printing on top of conductive FTO glass substrates (8 U cm2, Pilkington). The electrolyte composition was 0.05 M I2 (Aldrich, 99.99%), 0.5 M LiI (Aldrich, 99.9%), 0.3 M 1,2-dimethyl-3-propylimidazolium iodide (Solaronix), and 0.5 M 4-tert-butylpyridine (Aldrich, 99%), using 3methoxy propionitrile (Fluka, 99%) as a solvent. Then, the hole in the DSSC was sealed with a thermoplastic polymer and a glass cover slide. To obtain TiO2 NP/NR composite electrodes, different wt.% ratios of TiO2 NRs were added to TiO2 paste and ball mill mixed for 2 h. Five composite pastes containing pastes of NPs and NRs with NR concentrations of 1, 3, 5, and 10 wt% were also prepared. TiO2 NP/ NR composite pastes were re-coated onto coated electrodes with TiO2 NPs on the FTO glass using squeeze printing technique, followed by sintering at 450  C for 30 min. About 13 mm thick TiO2 film was deposited on 0.25 cm2 FTO glass substrate. The TiO2 NP/NR composite electrodes were immersed overnight (ca. 24 h) in a 5  104 mol/L ethanol solution of Ru(dcbpy)2(NCS)2 (N719, Solaronix), rinsed with anhydrous ethanol, and dried. 2.3. Characterization Field emission scanning electron microscopy (FE-SEM, Hitachi, S-4700) was used to examine the microstructures of films. The incident photon-to-current conversion efficiency (IPCE) was measured as a function of wavelength from 350 nm to 900 nm using a specially designed IPCE system for dye-sensitized solar cells (PV Measurements, Inc.). The amount of dye adsorbed on the TiO2 film was determined from the N719 solution by direct adsorption measurements with a UVeVis spectrophotometer (Shimadzu UV1601A, Japan). The photovoltaic properties were investigated by measuring the J-V characteristics under irradiation with white light from a 150 W Xenon lamp (McSciencs, Korea). The incident light intensity and the active cell area were 100 mW cm2 and 0.25 cm2. Intensity-modulated photocurrent spectroscopy (IMPS) and

295

intensity-modulated photovoltage spectroscopy (IMVS) measurements were used to investigate transportation characteristics and the electron lifetime, respectively. The solar cells were illuminated under a red LED (l ¼ 625 nm) as a light source, in which the amplitude of the AC light modulation was 10% of the DC light intensity at a frequency ranging from 100 kHz to 0.1 Hz. The light intensity was adjusted with a calibrated Si photodiode. 3. Results and discussion The morphologies of the TiO2 NRs and NPeNR films after sintering at 450  C and 850  C were examined using a field-emission scanning electron microscope (FE-SEM). Fig. 1a and d shows the SEM images of TiO2 nanorods after sintering at 450  C (a) and 850  C (d) for 30 min, respectively. The NRs sintered at 450  C (a) have an average diameter of around 400 nm, revealing that the TiO2 NRs were composed of closely packed anatase single crystallites (or grains) with size less that 15 nm. However, TiO2 NRs sintered at 850  C (d) were composed of nanocrystalline aggregates with size of 30e50 nm. Fig. 1b shows an SEM image of the TiO2 NP/NR composite electrode using TiO2 NPs and NRs sintered at 450  C, which reveals the presence of randomly oriented interconnecting NRs. Fig. 1b and e shows low magnification SEM images of the asfabricated NP/NR composite film, showing many small particles randomly oriented in NR arrays. Closer observation reveals that there are still enough interspaces in the as-prepared composite film, which can make it easy to load dye into the film. A crosssectional SEM image of a TiO2 NP/NR composite electrode on a dense TiO2 NP electrode with 7 mm thickness on FTO glass is shown in Fig. 1c (with TiO2 NRs sintered at 450  C) and Fig. 1f (with TiO2 NRs sintered at 850  C). The thickness of the TiO2 NP/NR composite electrode was 5.5e6 mm and it was confirmed that pores were obtained due to the composited TiO2 nanorods. In the case of nanorods composited in photoanodes, many researchers have reported a decrease in the amount of dye adsorption compared to photoanodes with TiO2 NPs only. However, in this study small TiO2 grains have large surface-to-volume ratio, which favors dye adsorption from the dense TiO2 NP layer as well as the distributed nanopores of the NP/NR composite electrode. Adsorption amount of dye is a key factor influencing the photovoltaic conversion efficiency of DSSCs. Many studies on adsorption equilibrium and kinetics have been conducted based on the differences in the amount of dye adsorbed and then subsequently desorbed on TiO2 thin films. However, in this work, the adsorption kinetic data were directly obtained in the absence of the desorption step in a small adsorption chamber [15]. Fig. 2 shows the results of adsorption kinetic experiments at different samples (NR0, NR1, NR3, NR5 and NR10). The results show that the amount of adsorbed N719 dye was in the following order of NR5 > NR3 > NR10 > NR1 > NR0. The kinetics data were analyzed with a pseudo-second-order model [16].

dq=dt ¼ k2 ðqe  qÞ2 ;

(1)

where k2 (g/mg/min) is the second-order rate constant determined by the plot of t/qt vs. t. As listed in Table 1, the determined rate constants of qe and k2 were in the range of 477.5e749.4 mg/g and 3.004  105e5.716  105 g/mg/min, respectively. Note that the correlation coefficients (R2) of the pseudo-first-order model for the linear plots of TiO2 are very close to 1. This result implies that adsorption kinetics can be successfully described by the pseudosecond-order model. The incident photon to current conversion efficiency (IPCE) spectra of the NR0 and NP/NR with NR contents of 1, 3, 5, 10 wt% are shown in Fig. 3. The maximum efficiency at the 530 nm wavelength

296

C.K. Hong et al. / Current Applied Physics 14 (2014) 294e299

Fig. 1. SEM images of electrospun nanorods: TiO2eNR after calcinations at 450  C (a) and 850  C (d), plan view of TiO2 NP and 5 wt% NR (NR5) composited electrode calcinated 450  C (b) and 850  C (e), cross-sectional SEM images of TiO2 NP and NR5 composited electrode calcinated at 450  C (c) and 850  C (f) on TiO2 NP electrode.

coincides with the maximum absorption wavelength of the N719 dye. In the entire wavelength region, NP/NR reveals higher external quantum efficiency (EQE) than NR0, which is compatible with the Jsc value obtained by photocurrentevoltage measurement. The IPCE peak height at 530 nm for the NP/NR5 electrode with 5 wt% NR is 70.6%, which is much higher than the value of 68.3% obtained for NR0 without NR composite, and 67% for the NP/NR10 electrode with 10 wt% NR. In order to estimate the light scattering effect, the NP/NR composite electrode showed higher EQE in the long wavelength region around 600e700 nm, suggesting that the NP/NR composite electrode provides an appreciable scattering effect. The IPCE peak at 630 nm for the NP/NR5 electrode with 5 wt% NR is 51.3%, which is much higher than the value of 32.7% obtained for NR0 without NR. The stronger scattering increased light absorption by up to 57%.

1.0

C/C0

0.9

0.8

NR0 NR1 NR3

Table 1 The pseudo-second-order kinetic parameters of N719 dye adsorption on TiO2 films.

NR5 NR10

0.7

0

60

Samples

120

180

240

300

360

420

Time, min Fig. 2. Amount of dye adsorption in photoanode prepared for TiO2 nanoparticles of one layer (1NP), two layers (2NP), nanoparticles and nanorods calcinated at 450  C composited on TiO2 nanoparticles (2NP/NR450) with various photoanodes.

Pseudo-second-order-kinetics qe (mg/g)

NR0 NR1 NR3 NR5 NR10

477.5 552.3 687.0 749.4 644.3

k2 (g/mg/min) 4.273 5.716 4.638 3.004 5.398

    

5

10 105 105 105 105

R2 0.99 0.99 0.99 0.99 0.99

C.K. Hong et al. / Current Applied Physics 14 (2014) 294e299

Table 2 Photovoltaic performance of the DSSCs with photoanodes containing various percentages of nanorods. x is nanorod composite percentage ratio in TiO2 nanoparticles. Thickness of NP electrode and NP/NR composite are 7 mm and 6 mm, respectively. Total thickness is 13 mm.

80 NR0 NR1 NR3 NR5 NR10

70 60

EQE (%)

50 40 Light scattering

30

297

Photoanodes (2NP/x)

Voc (V)

Jsc (mA cm2)

FF

h (%)

NR0 NR1 NR3 NR5 NR10

0.66 0.66 0.67 0.68 0.69

13.2 13.7 14.3 16.1 13.2

0.51 0.52 0.53 0.52 0.54

4.7 4.9 5.3 6.1 5.4

20 10

(a)

0

24

20

-10 500

600

700

800

Wavelength (nm) Fig. 3. Incident proton to current conversion efficiency (IPCE) curves for the DSSCs fabricated from NR/NP samples with NR percentages of 0, 1, 3, 5, and 10 wt%. Thickness of TiO2 NP electrode is 7 mm, and NP/NR composite thickness is 6 mm. Total thickness is 13 mm.

DSSCs with various percentages of TiO2 NRs were fabricated and tested. The DSSC made using the composite with 5 wt% NR achieved a short-circuit current density (Jsc) of 16.1 mA cm2 and an energy conversion efficiency (h) of 6.1% (Fig. 4 and Table 2). In comparison, the DSSC made using the TiO2 NPs had Jsc and h values of 13.2 mA cm2 and 4.7%, respectively, for the same photoanode thickness (w13 mm). These results indicate that the 30% improvement in h mainly resulted from the 22% increase in Jsc. The increase in Jsc is probably due to both the light scattering caused by the NRs and good dye uptake. Recently, numerous techniques used for studying electron dynamics have been reported. Among them, intensity-modulated photocurrent spectroscopy (IMPS) and intensity-modulated

τt(ms)

400

16

12

8 0

(b)

0.3

0.4

0.5

0.6

0.5

0.6

NR0 NR1 NR3 NR5 NR10

8.0E-05 6.0E-05 4.0E-05 2.0E-05 0

15

0.2

JSC (mA cm-2)

20 NR0 NR1 NR3 NR5 NR10

0.1

1.0E-04

Diffusion coefficint (cm 2 s-1)

300

0.1

0.2

0.3

0.4

JSC (mA cm-2)

(c) 110

NR0 NR1 NR3 NR5 NR10

90 τr(ms)

Current density (mA cm-2)

NR0 NR1 NR3 NR5 NR10

10

5

70

50

30

0 0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Voltage (V) Fig. 4. Photocurrentevoltage (JeV) curves for the DSSCs fabricated from NR/NP samples with NR percentages of 0, 1, 3, 5, and 10 wt%. Thickness of TiO2 NP electrode is 7 mm and NP/NR composite thickness is 6 mm. Total thickness is 13 mm.

0

0.1

0.2

0.3

JSC (mA

0.4

0.5

0.6

cm-2)

Fig. 5. Electron transport time (st) (a), electron diffusion coefficients (Dn) (b) and recombination lifetime (sr) (c) for the DSSCs fabricated from NR/NP samples with nanorod percentages of 0, 1, 3, 5, and 10 wt%. Thickness of TiO2 NP electrode is 7 mm, and NP/NR composite thickness is 6 mm. Total thickness is 13 mm.

298

C.K. Hong et al. / Current Applied Physics 14 (2014) 294e299

Table 3 Electron dynamic parameters estimated by IMPS/IMVS. x is nanorod composite percentage ratio in TiO2 nanoparticles. Thickness of NP electrode and NP/NR composite are 7 mm and 6 mm, respectively. Total thickness is 13 mm. Photoanodes (2NP/x)

Dn (105 cm2 s1)

st (ms)

sr (ms)

Ln (mm)

hcoll (%)

NR0 NR1 NR3 NR5 NR10

5.4 5.9 7.1 6.8 5.6

16 17 16 18 19

82 96 85 91 80

18.1 19.1 19.8 18.7 16.7

80 82 81 81 76

photovoltage spectroscopy (IMVS) have been proven as very useful methods, providing valuable information. Devices are illuminated under a sinusoidal modulated light of various frequency ranges, and represented as polar coordinates in complex plane plots of the response of amplitude and phase variation [17]. The charge collection and electron transport in TiO2 photoanodes are highly dependent on morphology, surface area and crystalline. In order to understand the substantial electron transportation and recombination characteristics in the TiO2 NP/NRs-based devices prepared in this study, IMPS and IMVS measurements have been employed for all the devices, and the plots obtained are shown in Fig. 5. From the IMPS plots in Fig 5(a), the transport time (st) of the photoelectrons traveling through the TiO2 film was calculated using the following equation: st ¼ 1/(2pfmin), where fmin is the frequency at the minimum imaginary component of the plot [18,19]. Fig 5(b) shows that the corresponding transit time is inversely proportional to the electron diffusion coefficient (Dn), which can be estimated with the st calculated from the formula of Dn ¼ d2/(2.35st), where d is the thickness of TiO2 film [19]. The data obtained for the individual devices are listed in Table 3. The lowest Dn value (w5.4  105 cm2 s1) was obtained from the NR0-based device, and was increased as the NR composite content was increased, reaching the highest value for NR3 (w7.1  105 cm2 s1). The value decreased to 6.8  105 cm2 s1 and 5.6  105 cm2 s1 for NR5 and NR10, respectively. The smaller Dn value might be due to the more tortuous pathway for electrons, affecting the trapping/detrapping effect [20e22]. From Fig. 5(c), which shows the IMVS plots of the NRx-based devices, the corresponding electron lifetimes (sr) were calculated with equation sr ¼ 1/(2pfmin), where fmin is the frequency at the minimum imaginary component in the IMVS plot [17]. The electron lifetime in the TiO2 photoanode is profoundly dependent on the back reaction with I 3 . A significant increase in sr was found as x was increased from 82 ms to 91 ms. It is clear that the increasing trend of sr is the same as that of Voc listed in Table 2, revealing a decrease of the electron recombination with I-3 in the electrolyte. The electron lifetime in TiO2 photoanodes were longer for NR5 (91 ms) than for NR0 (82 ms). This also suggests that the irregularity of the nanorods caused longer electron transport paths, enhancing the back reaction rate. Another parameter for the estimation of competition between electron diffusion and recombination is the electron average diffusion length (Ln), which can be calculated by the relation Ln ¼ (Dn$sr)1/2. Ln in TiO2 photoanodes became longer for NR3 (19.8 mm) than for NR0 (18.1 mm). 4. Conclusions Dye-sensitized solar cells have been fabricated based on composites of TiO2 electrospun NRs and NPs. By rationally tuning the weight ratio of TiO2 NRs and TiO2 NPs, the energy conversion efficiency and electrochemical properties were investigated. The NR5 (containing 5 wt% TiO2 NRs) had the greatest efficiency value,

which increased by about 22% compared to those composed of NR0 (pure double layer TiO2 NP solar cells). In addition, intensitymodulated photocurrent spectroscopy (IMPS) and intensitymodulated photovoltage spectroscopy (IMVS) proved that NR3 had the greatest electron diffusion coefficient. Faster interfacial charge transfer and higher dye adsorption amount led to higher power conversion efficiency. In summary, this study reveals that the composite made of electrospun TiO2 NRs and conventional TiO2 NPs can be used to construct an innovative type of photoanode that noticeably improves the harvesting of light without substantially sacrificing the dye uptake in DSSCs. Acknowledgments This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF), funded by the Ministry of Education, Science and Technology (2012010655). This work was also supported by the Priority Research Centers Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2009-0094055). References [1] M. Grätzel, Perspectives for dye-sensitized nanocrystalline solar cells, Prog. Photovolt. Res. Appl. 8 (2000) 171e185. [2] B. O’Regan, M. Grätzel, A low-cost, high-efficiency solar cell based on dyesensitized colloidal TiO2 films, Nature 353 (1991) 737e740. [3] P. Joshi, L. Zhang, D. Davoux, Z. Zhu, D. Galipeau, H. Fong, Q. Qiao, Composite of TiO2 nanofibers and nanoparticles for dye-sensitized solar cells with significantly improved efficiency, Energy Environ. Sci. 3 (2010) 1507e1510. [4] L. Yang, W.W.F. Leung, Application of a bilayer TiO2 nanofiber photoanode for optimization of dye-sensitized solar cells, Adv. Mater. 23 (2011) 4559e 4562. [5] B. Liu, E.S. Aydil, Growth of oriented single-crystalline rutile TiO2 nanorods on transparent conducting substrates for dye-sensitized solar cells, J. Am. Chem. Soc. 131 (2009) 3985e3990. [6] L. Francis, A.S. Nair, R. Jose, S. Ramakrishna, V. Thavasi, E. Marsano, Fabrication and characterization of dye-sensitized solar cells from rutile nanofibers and nanorods, Energy 36 (2010) 627e632. [7] G. Ai, W.T. Sun, Y.L. Zhang, L.M. Peng, Nanoparticle and nanorod TiO2 composite photoelectrodes with improved performance, Chem. Commun. 47 (2011) 6608e6610. [8] P. Poudel, Q. Qiao, One dimensional nanostructure/nanoparticle composites as photoanodes for dye-sensitized solar cells, Nanoscale 4 (2012) 2826e 2838. [9] K.H. Park, E.M. Jin, H.B. Gu, S.D. Yoon, E.M. Han, J.J. Yun, 204% enhanced efficiency of ZrO2 nanofibers doped dye-sensitized solar cells, Appl. Phys. Lett. 97 (2010) 023302. [10] Z. Liu, X. Zhang, S. Nishimoto, M. Jin, D.A. Tryk, T. Murakami, A. Fujishima, Anatase TiO2 nanoparticles on rutile TiO2 nanorods: a heterogeneous nanostructure via layer-by-layer assembly, Langmuir 23 (2007) 10916e10919. [11] Y.H. Jung, K.H. Park, J.S. Oh, D.H. Kim, C.K. Hong, Effect of TiO2 rutile nanorods on the photoelectrodes of dye-sensitized solar cells, Nanoscale Res. Lett. 8 (2013) 37. [12] V.S. Saji, M. Pyo, Dye sensitized solar cell of TiO2 nanoparticle/nanorod composites prepared via low-temperature synthesis in oleic acid, Thin Solid Films 518 (2010) 6542e6546. [13] J.Y. Park, I.H. Lee, G.N. Bea, Optimization of the electrospinning conditions for preparation of nanofibers from polyvinylacetate (PVAc) in ethanol solvent, J. Ind. Eng. Chem. 14 (2008) 707e713. [14] S.M. Jo, M.Y. Song, Y.R. Ahn, C.R. Park, D.Y. Kim, Nanofibril formation of electrospun TiO2 fibers and its application to dye-sensitized solar cells, J. Macromol. Sci. A 42 (2005) 1529e1540. [15] T.Y. Kim, J.W. Lee, E.M. Jin, J.Y. Park, J.H. Kim, K.H. Park, In situ measurement of dye adsorption on TiO2 thin films for dye-sensitized solar cells, Measurements 46 (2013) 1692e1697. [16] K.H. Park, T.Y. Kim, J.Y. Park, E.M. Jin, S.H. Yim, D.Y. Choi, J.W. Lee, Adsorption characteristics of gardenia yellow as natural photosensitizer for dyesensitized solar cells, Dyes Pigm. 96 (2013) 595e601. [17] D. Zhang, T. Yoshida, T. Oekermann, K. Furuta, H. Minoura, Room-temperature synthesis of porous nanoparticulate TiO2 films for flexible dye-sensitized solar cells, Adv. Funct. Mater. 16 (2006) 1228e1234. [18] B.H. Lee, M.Y. Song, S.Y. Jang, S.M. Jo, S.Y. Kwak, D.Y. Kim, Charge transport characteristics of high efficiency dye-sensitized solar cells based on electrospun TiO2 nanorod photoelectrodes, J. Phys. Chem. C 113 (2009) 21453e 21457.

C.K. Hong et al. / Current Applied Physics 14 (2014) 294e299 [19] P.S. Archana, R. Jose, C. Vijila, S. Ramakrisshna, Improved electron diffusion coefficient in electrospun TiO2 nanowires, J. Phys. Chem. C 113 (2009) 21538e 21542. [20] J.V.D. Lagemaat, A.J. Frank, Nonthermalized electron transport in dyesensitized nanocrystalline TiO2 films: transient photocurrent and randomwalk modeling studies, J. Phys. Chem. B 105 (2001) 11194e11205.

299

[21] M.J. Cass, A.B. Walker, D. Martinez, L.M. Peter, Grain morphology and trapping effects on electron transport in dye-sensitized nanocrystalline solar cells, J. Phys. Chem. B 109 (2005) 5100e5107. [22] L. Liang, S. Dai, L. Hu, F. Kong, W. Xu, K. Wang, Porosity effects on electron transport in TiO2 films and its application to dye-sensitized solar cells, J. Phys. Chem. B 110 (2006) 12404e12409.