- Email: [email protected]
Analysis of adsorption properties of N719 dye molecules on nanoporous TiO2 surface for dye-sensitized solar cell
Applied Surface Science 256 (2010) 5428–5433
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Analysis of adsorption properties of N719 dye molecules on nanoporous TiO2 surface for dye-sensitized solar cell Kyung-Jun Hwang a,1, Wang-Geun Shim b,1, Sung-Hoon Jung a, Seung-Joon Yoo c, Jae-Wook Lee a,* a
Department of Biochemical Engineering, Chosun University, Gwangju 501-759, Republic of Korea School of Applied Chemical Engineering, Chonnam National University, Gwangju 550-757, Republic of Korea c Department of Environmental and Chemical Engineering, Seonam University, Namwon 590-711, 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: Available online 28 December 2009
Ordered nanoporous TiO2 materials (MK-TiO2, MS-TiO2, and MU-TiO2) were synthesized for the dyesensitized solar cell (DSSC) by using different silica templates such as KIT-6, SBA-15, and MSU-H. To prepare a photoelectrode in DSSC, cis-bis(isothiocyanato)bis(2,20 -bipyridyl-4,40 -dicarboxylato)-ruthenium(II)bis-tetrabutylammonium dye (N719) was adsorbed onto the synthesized nanoporous TiO2 materials. The samples were characterized by XRD, TEM, FE-SEM, AFM, and N2 adsorption analyses. The photovoltaic performance of DSSC was evaluated from the overall conversion efficiency, fill factor, opencircuit voltage, and short-circuit current from the I–V curves measured. It was found that the photoelectric performance is highly dependent on the adsorption properties of N719 dye molecules on the nanoporous TiO2 replicas (MK-TiO2, MS-TiO2, and MU-TiO2) synthesized from different silica templates. ß 2010 Elsevier B.V. All rights reserved.
Keywords: Adsorption Nanoporous TiO2 N719 Dye-sensitized solar cells SBA-15 MSU-H KIT-6
1. Introduction Dye sensitized solar cells (DSSCs) consist of sensitizing dye, TiO2 porous film (anode electrode), electrolyte, and the opposite electrode (cathode electrode) [1,2]. For photovoltaic cells in DSSCs, ruthenium(II) complexes containing the polypyridyl ligands dye adsorbed on the TiO2 porous film are excited by absorbing the visible light. It has been reported that the efficiency of electron injection to the conduction band of TiO2 is highly dependent on the bonding structure of the dye adsorbed on the semiconductor TiO2 [3]. In addition, the electron transfer in DSSC is strongly influenced by the electrostatic and chemical interactions between the TiO2 surface and the adsorbed dye molecules [4,5]. Unfortunately, systematic studies on the influence of the adsorption properties between N719 molecules and TiO2 (1 0 1) surface on the energy conversion efficiency of DSSCs are very limited [6]. In this work, the influence of the adsorption properties between N719 molecules and the nanoporous TiO2 surface on the photovoltaic performance of DSSC was systematically investigated. For this investigation, we synthesized the ordered nanoporous TiO2 materials (MK-TiO2, MSTiO2, and MU-TiO2) using different silica templates of KIT-6 (bicontinuous cubic Ia3d), SBA-15 (2-D hexagonal P6mm), and MSU-H (2-D hexagonal P6mm). A dye molecule of cis-bis(isothio-
* Corresponding author. Tel.: +82 62 260 7151; fax: +82 62 230 7226. E-mail address: [email protected] (J.-W. Lee). 1 These authors contributed equally to this work. 0169-4332/$ – see front matter ß 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.apsusc.2009.12.128
cyanato)bis (2,20 -bipyridyl-4,40 -dicarboxylato)-ruthenium(II)bistetrabutylammonim (N719) used in this work has two bipyridyl ligands with two carboxyl groups at the 4 and 40 position of the bipyridyl group. TiO2 thin films adsorbed N719 dye on the synthesized TiO2 materials (MK-TiO2, MS-TiO2, and MU-TiO2) for DSSCs. The samples were characterized by XRD, FE-SEM, HR-TEM, AFM, and N2 adsorption analyses. The influence of the adsorption properties between N719 molecules and TiO2 surfaces (MK-TiO2, MS-TiO2, and MU-TiO2) on the energy conversion efficiency of DSSCs was also investigated on the basis of photovoltaic performance calculated from the I–V curves. 2. Experimental procedures Applying reported methods [7,8], periodic mesoporous silica templates (KIT-6, SBA-15, and MSU-H) were synthesized using a triblock copolymer (Pluronic P123, EO20PO70EO20, Mav: 5800, Aldrich) for the SBA-15 and KIT-6 and sodium silicate solution (10 wt % of SiO2, Na/Si: 2.5) for MSU-H as the structure directing agent materials. For the synthesis of TiO2 replica, the mesoporous silica materials were used after calcinations according to the standard procedure [8]. In a typical nano-replication, 0.75 g of Ti(OEt)4 and 30 ml of water are mixed, giving a white precipitate. The precipitate was collected by centrifuge and decantation of supernatant, and dissolved with 1.0 g of HCl (35 wt %) at room temperature. The clear TiO2 precursor sol was impregnated into 1.0 g of the mesoporous silica templates by the simple incipient wetness method, and the composites were dried at 433.15 K for
K.-J. Hwang et al. / Applied Surface Science 256 (2010) 5428–5433
10 min. This impregnation-drying process is repeated 9 times to maximize the amounts of TiO2 precursor within the mesopores of silica template using 1.0 M NaOH aqueous solution, nanoporous TiO2 materials (i.e., MK-TiO2, MS-TiO2, and MU-TiO2 replicas from KIT-6, SBA-15, and MSU-H) were obtained. Nanoporous TiO2 films were fabricated using TiO2 materials synthesized in this work (MK-TiO2, MS-TiO2, and MU-TiO2). For the preparation of nanoporous TiO2 thin film, TiO2 slurry was prepared by adding 2 g TiO2 particles, 0.68 ml 10% [v/v] acetyl acetone, 1 g hydroxypropyl cellulose (Mw 80,000, Aldrich), and 10.68 ml water for 12 h at 300 rpm using a paste mixer (PDM-300, Korea mixing technology Co.). Then, a nanoporous TiO2 film was fabricated by coating a precursor paste onto the fluorine-doped SnO2 conducting glass plates (FTO, 10 V/cm2, Asahi glass Co.) using the squeeze printing technique (adhesive tape was used as spacer of ca.65 m thickness). The nanoporous TiO2 film was treated by heating at 773.15 K for 2 h. The nanoporous TiO2 film formed thus on the FTO glass is 6–8 mm thickness and 0.5 cm 0.5 cm in size. The adsorbed Na+ metal ions resulting from the removal of mesoporous silica template in the synthesis process of nanoporous TiO2 particles act as the resistance of electrons in DSSCs. Therefore, TiO2 film was treated with the aqueous 0.1 M HCl solution for 4 h at 353.15 K for complete removal of Na+ metal ions, which was followed by the dye adsorption. To fabricate the DSSCs, the prepared thin film electrode was immersed in the N719 dye (Solaronix Co.) solution of 3 104 M at 333.15 K for 12 h, rinsed with anhydrous ethanol and dried. Pt coated glass-SnO2:F electrode was prepared as a counter electrode with an active area of 0.64 cm2. The Pt electrode was placed over the dye-adsorbed nanoporous TiO2 electrode, and the edges of the cell were sealed with 5 mm wide stripers of 60 m thick sealing sheet (SX 1170-60, Solaronix) by hot-pressing the two electrodes together at 353.15 K. The redox electrolyte was inserted into the cell through the small holes and sealed with a small square of sealing sheet. The redox electrolyte consists of 0.3 M 1,2-dimethyl-3-propylimidazolium iodide (Solaronix), 0.5 M LiI (Aldrich), 0.05 M I2 (Aldrich), and 0.5 M 4-tertbutylpyridine (4-TBP, Aldrich) and 3-metoxypropionitrile (3-MPN, Fluka) as a solvent. Adsorption equilibrium experiments were carried out by contacting a given amount of nanoporous TiO2 film with N719 dye solution of 0.01–0.3 mM in a shaking incubator at 333.15 K. The adsorption capacity of nonoporous TiO2 film was measured by completely desorbing the adsorbed dye molecules from TiO2 film using 0.1 M NaOH solution/ ethanol (50/50 vol%). The concentration of N719 dye solution was analyzed on a UV spectrometer (UV-160A, Shimadzu) at 522 nm.
5429
diffraction patterns of nanoporous TiO2 particles from the KIT-6, SBA-15, and MSU-H template showed that the nanoporous TiO2 replica materials exhibit ordered materials which are very similar to those of silica templates [8,9]. In the case of MU-TiO2 from MSUH, the two or more intense diffraction peaks are characteristic of a 2-D hexagonal (P6mm) structure. The well resolved three peaks of (2 1 1), (2 2 0), and (3 3 2) were observed, indicating the bicontinuous cubic Ia3d symmetry of KIT-6. The framework structures of mesoporous TiO2 materials are highly crystalline with mainly
3. Results and discussion 3.1. Characterization of nanoporous TiO2 particles The synthesized nanoporous TiO2 materials (MK-TiO2, MS-TiO2, and MU-TiO2) were characterized by XRD, FE-SEM, TEM, and BET analyses. The crystallinity of the synthesized nanoporous TiO2 was characterized by using a transmission electron microscope (TEM; F20, Tecnai) and an X-ray diffractometer (D/MAX-1200, Rigaku) applying CuKa X-ray and a Ni filter at 35 kV and 15 mA. The film thickness and surface morphology were measured using a fieldemission scanning electron microscope (FE-SEM; S-4700, Hitachi). The morphology of the nanoporous TiO2 electrode was also examined by means of an atomic force microscope (AFM; CP-2, Veeco) in the non-contact model. Nitrogen adsorption and desorption isotherms were measured at 77 K using a Micromeritics ASAP 2010 automatic analyzer. We confirmed that the TiO2 materials ordered with crystalline frameworks are successfully synthesized via the nano-replication route from various mesoporous silica templates. The X-ray
Fig. 1. HR-TEM images and electron diffraction pattern (inlet) of nanoporous TiO2 particles: (a) MK-TiO2, (b) MS-TiO2, and (c) MU-TiO2.
K.-J. Hwang et al. / Applied Surface Science 256 (2010) 5428–5433
5430
Table 1 Textural properties of nanoporous TiO2 replicas obtained from mesoporous silica.
MK-TiO2 MS-TiO2 MU-TIO2 a b c d
SBETa (m2/g)
Vtb (cm3/g)
388 256 160
0.488 0.301 0.444
VDFTc V<2 nm
V2–4 nm
V>4 nm
0.118 0.050 0.024
0.074 0.082 0.042
0.105 0.085 0.116
APDd nm
Particle size
0.4 2.9 4.8
Unknown 0.5–1 mm 200–300 nm
Specific BET surface area (P/P0 = 0.1–0.2). Total pore volume at P/P0 = 0.99. DFT pore volume. Average pore width.
anatase phases (1 0 1, 0 0 4, 2 0 0, 1 0 5, 2 1 1) from wide-angle XRD patterns. Because of the space limitation, the results of XRD patterns are not shown. The electron diffraction of TiO2 displays the Debye-Scherrer rings of anatase (Fig. 1 (inlet)). The lattice fringes corresponding to the (1 0 1) plane of anatase phase can be seen in the HR-TEM image (Fig. 1). The nanoporous TiO2 synthesized via nano-replication using various kinds of mesoporous silicas as the structure directing templates consists mainly of anatase phases from as observed in the TEM and XRD patterns. The ordered meso-structure was observed in the TEM image of nanoporous TiO2 in Fig. 1 (outlet). The size of anatase phase crystal prepared from KIT-6, SBA-15, and MSU-H is about 10 nm. The morphology and the mesopore structure of the synthesized nanoporous TiO2 particles resemble the templates of KIT-6, KIT-15, and MSU-H. However, the average particle size measurement of MK-TiO2 could not be obtained because of the irregular cubic morphology of KIT-6 template. However, a relatively uniform morphology of MS-TiO2 and MUTiO2 was obtained using SBA-15 and MSU-H templates (Table 1). Fig. 2 shows the nitrogen adsorption–desorption isotherms (Fig. 2(a)), the pore size distributions (Fig. 2(b)) and the adsorption energy distribution curves (Fig. 2(c)) of MK-TiO2, MS-TiO2, and MUTiO2. The adsorption isotherm corresponds to the type IV isotherms according to the IUPAC. Well-defined steps appear in the adsorption–desorption curves between the relative pressures, P/P0, of 0.4– 0.7. The important physical properties of nanoporous TiO2 are listed in Table 1. The specific BET surface areas of MK-TiO2, MS-TiO2, and MU-TiO2 are about 388 m2/g, 256 m2/g, and 160 m2/g, respectively. In addition, the total pore volume and the average pore size of MKTiO2, MS-TiO2, and MU-TiO2 are in the range of 0.30–0.49 cm3/g and 0.4–4.8 nm when measured by using the DFT method. These results indicate that various particle sizes of nanoporous TiO2 with an ordered structure and high surface area can be successfully synthesized via direct replication route from the hard template. On the other hand, it has been known that the shape and the intensity of the adsorption energy distribution curve are highly related to the physical property (i.e., geometrical heterogeneity) and chemical characteristics (i.e., energetic heterogeneity) of nanocrystalline TiO2 for DSSC. Fig. 2(c) compares the adsorption energy distributions (AEDs) of nitrogen on nanoporous TiO2 thin films of MK-TiO2, MS-TiO2, and MU-TiO2. To calculate the AEDs from the low-pressure nitrogen adsorption isotherm data (P/P0 < 0.1), the Flower–Guggenheim isotherm was adopted as a kernel function and the generalized nonlinear regularization [11,12]. As shown in Fig. 2(c), all the AED curves showed two peaks indicating the existence of two energetically different adsorption sites. The first and second adsorption energy curves are distributed mainly in the range of 3–7 and 8–15 kJ/mol, respectively. 3.2. Nanoporous TiO2 thin film for DSSCs The TiO2 thin films using the nanoporous TiO2 were prepared for the photoelectrode in DSSCs. FE-SEM and AFM were used to
Fig. 2. N2 adsorption and desorption isotherms for MK-TiO2, MS-TiO2, and MU-TiO2 (a), pore size distributions (b), and adsorption energy distributions (c) of nanoporous TiO2 particles.
K.-J. Hwang et al. / Applied Surface Science 256 (2010) 5428–5433
5431
Fig. 3. FE-SEM (left side) and AFM (right side) analysis of TiO2 thin films: (a and d) MK-TiO2, (b and e) MS-TiO2, and (c and f) MU-TiO2.
examine the surface morphology of the TiO2 thin films. Fig. 3 shows the FE-SEM (left panel; a–c) and AFM (right panel; d–f) images. The primary particles of TiO2 of MS-TiO2 (c) and MU-TiO2 thin film (d) have a high surface area and uniform structure in the aggregated thin film because of the nature of the used mesoporous silica templates. However, the result of MK-TiO2 (b) thin film was different from the irregular mesoporous template. The aggregated particle size of MK-TiO2 is still unknown. The film thickness of the photoelectode prepared by using the nanoporous TiO2 replicas from the mesoporous silica templates was approximately in the range of 6–8 mm (Fig. 3 (left panel; a–c, inlet)). Fig. 3(d–f) illustrates the 3-D AFM images of the surface morphology of TiO2 thin film. The roughness of nanoporous films was higher except the MK-TiO2 thin film and the FE-SEM images are equally observed in the AFM images. 3.3. Adsorption properties of N719 on nanoporous TiO2 thin films Fig. 4(a) shows the comparison of N719 adsorption isotherms and the corresponding Langmuir–Freundlich (LF) isotherm model for three different nanoporous TiO2 thin films. The LF equation and the determined isotherm parameters are listed in Table 2. The experimental results clearly show that MU-TiO2 has a substantially higher adsorption capacity for N719 than that of MS-TiO2 and MK-
TiO2. Namely, the adsorption amount of N719 on three different nanoporous TiO2 is in the following sequence: MU-TiO2 > MKTiO2 > MS-TiO2. Contrary to our expectations however, there is no clear correlation between the adsorption amounts and the basic textural properties such as the BET surface area, the total pore volume and the average pore width. On the other hand, the molecular diameter of the N719 dye was reported about 1.76 nm, which was based on the crystallographic data of N3 dye. Recently Chen et al. have also reported that the adsorption amount of organic dyes (or Ru(II)-based complexes) on TiO2 is highly dependent on the molecular size of dyes employed [10]. Thus, in this work, to examine the relation between the pore volume and the N719 adsorption amount, the pore volumes were divided into three regions: (1) volume of pore width below 2 nm (V<2 nm), (2) volume of pore width between 2 and 4 nm (V2–4 nm), and (3) volume of pore width greater than 4 nm (V>4 nm). As shown in Fig. 4(b), a notable relationship was observed between the adsorption capacity of N719 and the volume of pore width greater than 4 nm (V>4 nm), indicating that the dye adsorption has close connection with the molecular size and the proportion of pores having optimum size. In this work, the order of adsorption capacity of N719 was in good agreement with that of V>4 nm. It was found that the desirable pore widths required for achieving the highest N719 adsorption are about 2 times greater than the diameter of
5432
K.-J. Hwang et al. / Applied Surface Science 256 (2010) 5428–5433 Table 2 Langmuir–Freundlich isotherm equation (a), adsorption energy distribution function (b) and Langmuir–Freundlich isotherm parameters (c). (a) Langmuir–Freundlich isotherm equation n
m ðKcÞ q ¼ q1þ ðKcÞn
where q is the adsorbed amount, qm is the monolayer adsorption capacity, K is the equilibrium constant, c is the solute equilibrium concentration and n is the system heterogeneity parameter. (b) Adsorption energy distribution function E12max Z Fx expðE12 =RT Þ F ðE12 ÞdE12 uðcÞ ¼ nno ¼ 1þFx expðE12 =RT Þ E
where u(c) is the total fraction coverage of solute, F ¼ Fðc; uÞ is a model dependent function, E12 = E1 E2 is the energy difference between the solute and water, F(E12) is the energy distribution function, T is the absolute temperature, R is the gas constant; x is c/csol, where csol is the solubility of the solute in water. (c) Langmuir–Freundlich isotherm equation parameters MK-TiO2 qm K n E (%)a a
MS-TiO2
38.4 7.39 0.840 0.772
29.9 9.62 0.935 0.491 n X jqexp;k qcal;k j 100 Average percent error: Eð%Þ ¼ n . q k¼1
MU-TiO2 41.5 8.61 0.772 0.414
exp;k
Fig. 5. I–V curves of nanoporous TiO2 films for MK-TiO2, MS-TiO2, and MU-TiO2.
Fig. 4. Adsorption equilibrium amount of N719 on nanoporous TiO2 films at 333.15 K (a), pore volumes in comparison with N719 adsorption as a function of nanoporous TiO2 films and (b) adsorption energy distributions of N719 on nanoporous TiO2 films (c).
N719 molecules (1.76 nm). Although MU-TiO2 has the lowest BET surface area, the total pore volume and the largest average pore width, they reveal the highest N719 adsorption because of their largest amount of V>4 nm. Thus it is appropriate to explain that the N719 adsorption capacity is substantially dependent on the proportion of volume of pore width greater than 4 nm. The obtained result reveals that the pore volumes over a diameter of 4 nm may contribute more adsorption than the micropore volumes below a width of 4 nm. Fig. 4(c) shows the AEDs for N719 on three different nanoporous TiO2 films. To calculate the energy distributions, we used the data
below the monolayer coverage which were determined from the LF equation and the generalized nonlinear regularization method which sets no assumption on position and shape of the solution (or the energy distribution function) (Table 2) [11,12]. The detailed procedures for determining the AEDs were described in elsewhere [11–13]. As shown in Fig. 4(c), similar shapes of AEDs were observed. All samples have two distinct AED peaks, which may indicate that two different types of surface energy for N719 exist mainly on the surface of the adsorbent. In the cases of MK-TiO2 and MS-TiO2, the lower AED peaks are located between 0 and 9.2 kJ/ mol with the maximum at about 3.15 (MS-TiO2) and 3.36 (MKTiO2) kJ/mol, whereas the higher AED peaks appeared between 9.2 and 16.0 kJ/mol with the maximum at about 11.76 (MS-TiO2) and 11.97 (MK-TiO2) kJ/mol. In contrast, the AED for the MU-TiO2 differ slightly from the case of MK-TiO2 and MS-TiO2. The AED peak shape, height, and location of MU-TiO2 are greatly broader, lower, and shifted toward relatively higher adsorption energy. The main and shoulder (or small) AED curves are in the range of 0–11.6 and 12.4–22.1 kJ/mol, where the pronounced maxima were located at 4.4 and 17.0 kJ/mol, respectively. To evaluate the photovoltaic performance of nanoporous TiO2 thin film, the current–voltage curves were obtained. The photo-
K.-J. Hwang et al. / Applied Surface Science 256 (2010) 5428–5433 Table 3 Photovoltaic performance of DSSCs using nanoporous TiO2 films.
MK-TiO2 film MS-TiO2 film MU-TiO2 film
Voc (V)
Isc (mA/cm2)
Fill factor (FF)
heff (%)
0.59 0.59 0.60
8.8 5.2 9.5
0.64 0.57 0.64
3.3 1.7 3.6
current–voltage (I–V) curves were measured using a source measure unit under irradiation of white light from a 1000 W Xenon lamp (Thermo Oriel Instruments). The incident light intensity and the active cell area were 100 mW/cm2 and 0.25 cm2, respectively. The I–V curves were used to calculate the short-circuit current (Isc), open-circuit voltage (Voc), fill factor (FF), and overall conversion efficiency (heff) of DSSC. It was observed that the overall conversion efficiency (heff) is in the following order: MU-TiO2 film (3.6%) > MK-TiO2 film (3.3%) > MS-TiO2 film (1.7%). This trend is identical to that of the N719 adsorption on nanoporous TiO2. Therefore, it is reasonable to conclude that the overall photovoltaic performance was substantially dependent on the adsorption amount of N719 (Fig. 5 and Table 3). 4. Conclusions Ordered nanoporous TiO2 particles (MK-TiO2, MS-TiO2, MUTiO2) were synthesized via the nano-replication route using different silica templates of KIT-6 (bicontinuous cubic Ia3d), SBA-15(2-D hexagonal P6mm), and MSU-H (2-D hexagonal P6mm), respectively. TiO2 films of MK-TiO2, MS-TiO2, MU-TiO2 were prepared and characterized by XRD, TEM, FE-SEM, AFM, and BET. The photovoltaic performance was evaluated from the overall
5433
conversion efficiency (heff), fill factor (FF), open-circuit voltage (Voc) and short-circuit current (Isc) from the I–V curves measured. The overall conversion efficiency (heff) is in the following order: MU-TiO2 film (3.6%) >MK-TiO2 film (3.3%) >MS-TiO2 film (1.7%). The results of adsorption properties obtained in this work led us to conclude that the photoelectric performance is strongly dependent on the adsorption amount of dye molecules and the pore size distributions of MK-TiO2, MS-TiO2, and MU-TiO2. Acknowledgements This study was supported by the research funds from Chosun University, 2009. References [1] B. O’Regan, M. Gratzel, Nature 353 (1991) 737. [2] L.L. Kazmerski, J. Electron. Spectrosc. Relat. Phenomena 150 (2006) 105. [3] Z. Zhang, S.M. Zak eeruddin, B.C. O’Regan, R. Humphry-Bak er, M. Gratzel, J. Phys. Chem. B 109 (2005) 2818. [4] P. Persson, S. Lunnell, Sol. Energy Mater. Sol. Cells 63 (2000) 139. [5] M.K. Nazeeruddin, A. Kay, I. Rodicio, R. Humphry-Baker, E. Mueller, P. Liska, N. Vlachopoulos, M. Gratzel, J. Am. Chem. Soc. 115 (1993) 6382. [6] J.W. Lee, K.J. Hwang, W.G. Shim, K.H. Park, H.B. Gu, K.H. Kwun, Korean J. Chem. Eng. 24 (2007) 847. [7] S.S. Kim, H.I. Lee, J.M. Kim, Chem. Lett. 37 (2008) 140. [8] F. Kleitz, S.H. Choi, R. Ryoo, Chem. Commun. (2003) 2136. [9] K.J. Hwang, S.J. Yoo, S.S. Kim, J.M. Kim, W.G. Shim, S.I. Kim, J.W. Lee, J. Nanosci. Nanotechnol. 8 (2008) 4976. [10] K.F. Chen, Y.C. Hsu, Q. Wu, S.S. Sun, Org. Lett. 11 (2) (2009) 377. [11] W.G. Shim, H.C. Kang, C. Kim, S.C. Kim, J.W. Lee, C.J. Lee, H. Moon, J. Nanosci. Nanotechnol. 6 (2006) 3583. [12] K.J. Hwang, S.J. Yoo, S.H. Jung, D.W. Park, S.I. Kim, J.W. Lee, Bull. Korean Chem. Soc. 30 (1) (2009) 172. [13] K. La´szlo´, P. Podkos´cielny, A. Da˛browski, Appl. Surf. Sci. 252 (2009) 5752.
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Analysis of adsorption properties of N719 dye molecules on nanoporous TiO2 surface for dye-sensitized solar cell Kyung-Jun Hwang a,1, Wang-Geun Shim b,1, Sung-Hoon Jung a, Seung-Joon Yoo c, Jae-Wook Lee a,* a
Department of Biochemical Engineering, Chosun University, Gwangju 501-759, Republic of Korea School of Applied Chemical Engineering, Chonnam National University, Gwangju 550-757, Republic of Korea c Department of Environmental and Chemical Engineering, Seonam University, Namwon 590-711, 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: Available online 28 December 2009
Ordered nanoporous TiO2 materials (MK-TiO2, MS-TiO2, and MU-TiO2) were synthesized for the dyesensitized solar cell (DSSC) by using different silica templates such as KIT-6, SBA-15, and MSU-H. To prepare a photoelectrode in DSSC, cis-bis(isothiocyanato)bis(2,20 -bipyridyl-4,40 -dicarboxylato)-ruthenium(II)bis-tetrabutylammonium dye (N719) was adsorbed onto the synthesized nanoporous TiO2 materials. The samples were characterized by XRD, TEM, FE-SEM, AFM, and N2 adsorption analyses. The photovoltaic performance of DSSC was evaluated from the overall conversion efficiency, fill factor, opencircuit voltage, and short-circuit current from the I–V curves measured. It was found that the photoelectric performance is highly dependent on the adsorption properties of N719 dye molecules on the nanoporous TiO2 replicas (MK-TiO2, MS-TiO2, and MU-TiO2) synthesized from different silica templates. ß 2010 Elsevier B.V. All rights reserved.
Keywords: Adsorption Nanoporous TiO2 N719 Dye-sensitized solar cells SBA-15 MSU-H KIT-6
1. Introduction Dye sensitized solar cells (DSSCs) consist of sensitizing dye, TiO2 porous film (anode electrode), electrolyte, and the opposite electrode (cathode electrode) [1,2]. For photovoltaic cells in DSSCs, ruthenium(II) complexes containing the polypyridyl ligands dye adsorbed on the TiO2 porous film are excited by absorbing the visible light. It has been reported that the efficiency of electron injection to the conduction band of TiO2 is highly dependent on the bonding structure of the dye adsorbed on the semiconductor TiO2 [3]. In addition, the electron transfer in DSSC is strongly influenced by the electrostatic and chemical interactions between the TiO2 surface and the adsorbed dye molecules [4,5]. Unfortunately, systematic studies on the influence of the adsorption properties between N719 molecules and TiO2 (1 0 1) surface on the energy conversion efficiency of DSSCs are very limited [6]. In this work, the influence of the adsorption properties between N719 molecules and the nanoporous TiO2 surface on the photovoltaic performance of DSSC was systematically investigated. For this investigation, we synthesized the ordered nanoporous TiO2 materials (MK-TiO2, MSTiO2, and MU-TiO2) using different silica templates of KIT-6 (bicontinuous cubic Ia3d), SBA-15 (2-D hexagonal P6mm), and MSU-H (2-D hexagonal P6mm). A dye molecule of cis-bis(isothio-
* Corresponding author. Tel.: +82 62 260 7151; fax: +82 62 230 7226. E-mail address: [email protected] (J.-W. Lee). 1 These authors contributed equally to this work. 0169-4332/$ – see front matter ß 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.apsusc.2009.12.128
cyanato)bis (2,20 -bipyridyl-4,40 -dicarboxylato)-ruthenium(II)bistetrabutylammonim (N719) used in this work has two bipyridyl ligands with two carboxyl groups at the 4 and 40 position of the bipyridyl group. TiO2 thin films adsorbed N719 dye on the synthesized TiO2 materials (MK-TiO2, MS-TiO2, and MU-TiO2) for DSSCs. The samples were characterized by XRD, FE-SEM, HR-TEM, AFM, and N2 adsorption analyses. The influence of the adsorption properties between N719 molecules and TiO2 surfaces (MK-TiO2, MS-TiO2, and MU-TiO2) on the energy conversion efficiency of DSSCs was also investigated on the basis of photovoltaic performance calculated from the I–V curves. 2. Experimental procedures Applying reported methods [7,8], periodic mesoporous silica templates (KIT-6, SBA-15, and MSU-H) were synthesized using a triblock copolymer (Pluronic P123, EO20PO70EO20, Mav: 5800, Aldrich) for the SBA-15 and KIT-6 and sodium silicate solution (10 wt % of SiO2, Na/Si: 2.5) for MSU-H as the structure directing agent materials. For the synthesis of TiO2 replica, the mesoporous silica materials were used after calcinations according to the standard procedure [8]. In a typical nano-replication, 0.75 g of Ti(OEt)4 and 30 ml of water are mixed, giving a white precipitate. The precipitate was collected by centrifuge and decantation of supernatant, and dissolved with 1.0 g of HCl (35 wt %) at room temperature. The clear TiO2 precursor sol was impregnated into 1.0 g of the mesoporous silica templates by the simple incipient wetness method, and the composites were dried at 433.15 K for
K.-J. Hwang et al. / Applied Surface Science 256 (2010) 5428–5433
10 min. This impregnation-drying process is repeated 9 times to maximize the amounts of TiO2 precursor within the mesopores of silica template using 1.0 M NaOH aqueous solution, nanoporous TiO2 materials (i.e., MK-TiO2, MS-TiO2, and MU-TiO2 replicas from KIT-6, SBA-15, and MSU-H) were obtained. Nanoporous TiO2 films were fabricated using TiO2 materials synthesized in this work (MK-TiO2, MS-TiO2, and MU-TiO2). For the preparation of nanoporous TiO2 thin film, TiO2 slurry was prepared by adding 2 g TiO2 particles, 0.68 ml 10% [v/v] acetyl acetone, 1 g hydroxypropyl cellulose (Mw 80,000, Aldrich), and 10.68 ml water for 12 h at 300 rpm using a paste mixer (PDM-300, Korea mixing technology Co.). Then, a nanoporous TiO2 film was fabricated by coating a precursor paste onto the fluorine-doped SnO2 conducting glass plates (FTO, 10 V/cm2, Asahi glass Co.) using the squeeze printing technique (adhesive tape was used as spacer of ca.65 m thickness). The nanoporous TiO2 film was treated by heating at 773.15 K for 2 h. The nanoporous TiO2 film formed thus on the FTO glass is 6–8 mm thickness and 0.5 cm 0.5 cm in size. The adsorbed Na+ metal ions resulting from the removal of mesoporous silica template in the synthesis process of nanoporous TiO2 particles act as the resistance of electrons in DSSCs. Therefore, TiO2 film was treated with the aqueous 0.1 M HCl solution for 4 h at 353.15 K for complete removal of Na+ metal ions, which was followed by the dye adsorption. To fabricate the DSSCs, the prepared thin film electrode was immersed in the N719 dye (Solaronix Co.) solution of 3 104 M at 333.15 K for 12 h, rinsed with anhydrous ethanol and dried. Pt coated glass-SnO2:F electrode was prepared as a counter electrode with an active area of 0.64 cm2. The Pt electrode was placed over the dye-adsorbed nanoporous TiO2 electrode, and the edges of the cell were sealed with 5 mm wide stripers of 60 m thick sealing sheet (SX 1170-60, Solaronix) by hot-pressing the two electrodes together at 353.15 K. The redox electrolyte was inserted into the cell through the small holes and sealed with a small square of sealing sheet. The redox electrolyte consists of 0.3 M 1,2-dimethyl-3-propylimidazolium iodide (Solaronix), 0.5 M LiI (Aldrich), 0.05 M I2 (Aldrich), and 0.5 M 4-tertbutylpyridine (4-TBP, Aldrich) and 3-metoxypropionitrile (3-MPN, Fluka) as a solvent. Adsorption equilibrium experiments were carried out by contacting a given amount of nanoporous TiO2 film with N719 dye solution of 0.01–0.3 mM in a shaking incubator at 333.15 K. The adsorption capacity of nonoporous TiO2 film was measured by completely desorbing the adsorbed dye molecules from TiO2 film using 0.1 M NaOH solution/ ethanol (50/50 vol%). The concentration of N719 dye solution was analyzed on a UV spectrometer (UV-160A, Shimadzu) at 522 nm.
5429
diffraction patterns of nanoporous TiO2 particles from the KIT-6, SBA-15, and MSU-H template showed that the nanoporous TiO2 replica materials exhibit ordered materials which are very similar to those of silica templates [8,9]. In the case of MU-TiO2 from MSUH, the two or more intense diffraction peaks are characteristic of a 2-D hexagonal (P6mm) structure. The well resolved three peaks of (2 1 1), (2 2 0), and (3 3 2) were observed, indicating the bicontinuous cubic Ia3d symmetry of KIT-6. The framework structures of mesoporous TiO2 materials are highly crystalline with mainly
3. Results and discussion 3.1. Characterization of nanoporous TiO2 particles The synthesized nanoporous TiO2 materials (MK-TiO2, MS-TiO2, and MU-TiO2) were characterized by XRD, FE-SEM, TEM, and BET analyses. The crystallinity of the synthesized nanoporous TiO2 was characterized by using a transmission electron microscope (TEM; F20, Tecnai) and an X-ray diffractometer (D/MAX-1200, Rigaku) applying CuKa X-ray and a Ni filter at 35 kV and 15 mA. The film thickness and surface morphology were measured using a fieldemission scanning electron microscope (FE-SEM; S-4700, Hitachi). The morphology of the nanoporous TiO2 electrode was also examined by means of an atomic force microscope (AFM; CP-2, Veeco) in the non-contact model. Nitrogen adsorption and desorption isotherms were measured at 77 K using a Micromeritics ASAP 2010 automatic analyzer. We confirmed that the TiO2 materials ordered with crystalline frameworks are successfully synthesized via the nano-replication route from various mesoporous silica templates. The X-ray
Fig. 1. HR-TEM images and electron diffraction pattern (inlet) of nanoporous TiO2 particles: (a) MK-TiO2, (b) MS-TiO2, and (c) MU-TiO2.
K.-J. Hwang et al. / Applied Surface Science 256 (2010) 5428–5433
5430
Table 1 Textural properties of nanoporous TiO2 replicas obtained from mesoporous silica.
MK-TiO2 MS-TiO2 MU-TIO2 a b c d
SBETa (m2/g)
Vtb (cm3/g)
388 256 160
0.488 0.301 0.444
VDFTc V<2 nm
V2–4 nm
V>4 nm
0.118 0.050 0.024
0.074 0.082 0.042
0.105 0.085 0.116
APDd nm
Particle size
0.4 2.9 4.8
Unknown 0.5–1 mm 200–300 nm
Specific BET surface area (P/P0 = 0.1–0.2). Total pore volume at P/P0 = 0.99. DFT pore volume. Average pore width.
anatase phases (1 0 1, 0 0 4, 2 0 0, 1 0 5, 2 1 1) from wide-angle XRD patterns. Because of the space limitation, the results of XRD patterns are not shown. The electron diffraction of TiO2 displays the Debye-Scherrer rings of anatase (Fig. 1 (inlet)). The lattice fringes corresponding to the (1 0 1) plane of anatase phase can be seen in the HR-TEM image (Fig. 1). The nanoporous TiO2 synthesized via nano-replication using various kinds of mesoporous silicas as the structure directing templates consists mainly of anatase phases from as observed in the TEM and XRD patterns. The ordered meso-structure was observed in the TEM image of nanoporous TiO2 in Fig. 1 (outlet). The size of anatase phase crystal prepared from KIT-6, SBA-15, and MSU-H is about 10 nm. The morphology and the mesopore structure of the synthesized nanoporous TiO2 particles resemble the templates of KIT-6, KIT-15, and MSU-H. However, the average particle size measurement of MK-TiO2 could not be obtained because of the irregular cubic morphology of KIT-6 template. However, a relatively uniform morphology of MS-TiO2 and MUTiO2 was obtained using SBA-15 and MSU-H templates (Table 1). Fig. 2 shows the nitrogen adsorption–desorption isotherms (Fig. 2(a)), the pore size distributions (Fig. 2(b)) and the adsorption energy distribution curves (Fig. 2(c)) of MK-TiO2, MS-TiO2, and MUTiO2. The adsorption isotherm corresponds to the type IV isotherms according to the IUPAC. Well-defined steps appear in the adsorption–desorption curves between the relative pressures, P/P0, of 0.4– 0.7. The important physical properties of nanoporous TiO2 are listed in Table 1. The specific BET surface areas of MK-TiO2, MS-TiO2, and MU-TiO2 are about 388 m2/g, 256 m2/g, and 160 m2/g, respectively. In addition, the total pore volume and the average pore size of MKTiO2, MS-TiO2, and MU-TiO2 are in the range of 0.30–0.49 cm3/g and 0.4–4.8 nm when measured by using the DFT method. These results indicate that various particle sizes of nanoporous TiO2 with an ordered structure and high surface area can be successfully synthesized via direct replication route from the hard template. On the other hand, it has been known that the shape and the intensity of the adsorption energy distribution curve are highly related to the physical property (i.e., geometrical heterogeneity) and chemical characteristics (i.e., energetic heterogeneity) of nanocrystalline TiO2 for DSSC. Fig. 2(c) compares the adsorption energy distributions (AEDs) of nitrogen on nanoporous TiO2 thin films of MK-TiO2, MS-TiO2, and MU-TiO2. To calculate the AEDs from the low-pressure nitrogen adsorption isotherm data (P/P0 < 0.1), the Flower–Guggenheim isotherm was adopted as a kernel function and the generalized nonlinear regularization [11,12]. As shown in Fig. 2(c), all the AED curves showed two peaks indicating the existence of two energetically different adsorption sites. The first and second adsorption energy curves are distributed mainly in the range of 3–7 and 8–15 kJ/mol, respectively. 3.2. Nanoporous TiO2 thin film for DSSCs The TiO2 thin films using the nanoporous TiO2 were prepared for the photoelectrode in DSSCs. FE-SEM and AFM were used to
Fig. 2. N2 adsorption and desorption isotherms for MK-TiO2, MS-TiO2, and MU-TiO2 (a), pore size distributions (b), and adsorption energy distributions (c) of nanoporous TiO2 particles.
K.-J. Hwang et al. / Applied Surface Science 256 (2010) 5428–5433
5431
Fig. 3. FE-SEM (left side) and AFM (right side) analysis of TiO2 thin films: (a and d) MK-TiO2, (b and e) MS-TiO2, and (c and f) MU-TiO2.
examine the surface morphology of the TiO2 thin films. Fig. 3 shows the FE-SEM (left panel; a–c) and AFM (right panel; d–f) images. The primary particles of TiO2 of MS-TiO2 (c) and MU-TiO2 thin film (d) have a high surface area and uniform structure in the aggregated thin film because of the nature of the used mesoporous silica templates. However, the result of MK-TiO2 (b) thin film was different from the irregular mesoporous template. The aggregated particle size of MK-TiO2 is still unknown. The film thickness of the photoelectode prepared by using the nanoporous TiO2 replicas from the mesoporous silica templates was approximately in the range of 6–8 mm (Fig. 3 (left panel; a–c, inlet)). Fig. 3(d–f) illustrates the 3-D AFM images of the surface morphology of TiO2 thin film. The roughness of nanoporous films was higher except the MK-TiO2 thin film and the FE-SEM images are equally observed in the AFM images. 3.3. Adsorption properties of N719 on nanoporous TiO2 thin films Fig. 4(a) shows the comparison of N719 adsorption isotherms and the corresponding Langmuir–Freundlich (LF) isotherm model for three different nanoporous TiO2 thin films. The LF equation and the determined isotherm parameters are listed in Table 2. The experimental results clearly show that MU-TiO2 has a substantially higher adsorption capacity for N719 than that of MS-TiO2 and MK-
TiO2. Namely, the adsorption amount of N719 on three different nanoporous TiO2 is in the following sequence: MU-TiO2 > MKTiO2 > MS-TiO2. Contrary to our expectations however, there is no clear correlation between the adsorption amounts and the basic textural properties such as the BET surface area, the total pore volume and the average pore width. On the other hand, the molecular diameter of the N719 dye was reported about 1.76 nm, which was based on the crystallographic data of N3 dye. Recently Chen et al. have also reported that the adsorption amount of organic dyes (or Ru(II)-based complexes) on TiO2 is highly dependent on the molecular size of dyes employed [10]. Thus, in this work, to examine the relation between the pore volume and the N719 adsorption amount, the pore volumes were divided into three regions: (1) volume of pore width below 2 nm (V<2 nm), (2) volume of pore width between 2 and 4 nm (V2–4 nm), and (3) volume of pore width greater than 4 nm (V>4 nm). As shown in Fig. 4(b), a notable relationship was observed between the adsorption capacity of N719 and the volume of pore width greater than 4 nm (V>4 nm), indicating that the dye adsorption has close connection with the molecular size and the proportion of pores having optimum size. In this work, the order of adsorption capacity of N719 was in good agreement with that of V>4 nm. It was found that the desirable pore widths required for achieving the highest N719 adsorption are about 2 times greater than the diameter of
5432
K.-J. Hwang et al. / Applied Surface Science 256 (2010) 5428–5433 Table 2 Langmuir–Freundlich isotherm equation (a), adsorption energy distribution function (b) and Langmuir–Freundlich isotherm parameters (c). (a) Langmuir–Freundlich isotherm equation n
m ðKcÞ q ¼ q1þ ðKcÞn
where q is the adsorbed amount, qm is the monolayer adsorption capacity, K is the equilibrium constant, c is the solute equilibrium concentration and n is the system heterogeneity parameter. (b) Adsorption energy distribution function E12max Z Fx expðE12 =RT Þ F ðE12 ÞdE12 uðcÞ ¼ nno ¼ 1þFx expðE12 =RT Þ E
where u(c) is the total fraction coverage of solute, F ¼ Fðc; uÞ is a model dependent function, E12 = E1 E2 is the energy difference between the solute and water, F(E12) is the energy distribution function, T is the absolute temperature, R is the gas constant; x is c/csol, where csol is the solubility of the solute in water. (c) Langmuir–Freundlich isotherm equation parameters MK-TiO2 qm K n E (%)a a
MS-TiO2
38.4 7.39 0.840 0.772
29.9 9.62 0.935 0.491 n X jqexp;k qcal;k j 100 Average percent error: Eð%Þ ¼ n . q k¼1
MU-TiO2 41.5 8.61 0.772 0.414
exp;k
Fig. 5. I–V curves of nanoporous TiO2 films for MK-TiO2, MS-TiO2, and MU-TiO2.
Fig. 4. Adsorption equilibrium amount of N719 on nanoporous TiO2 films at 333.15 K (a), pore volumes in comparison with N719 adsorption as a function of nanoporous TiO2 films and (b) adsorption energy distributions of N719 on nanoporous TiO2 films (c).
N719 molecules (1.76 nm). Although MU-TiO2 has the lowest BET surface area, the total pore volume and the largest average pore width, they reveal the highest N719 adsorption because of their largest amount of V>4 nm. Thus it is appropriate to explain that the N719 adsorption capacity is substantially dependent on the proportion of volume of pore width greater than 4 nm. The obtained result reveals that the pore volumes over a diameter of 4 nm may contribute more adsorption than the micropore volumes below a width of 4 nm. Fig. 4(c) shows the AEDs for N719 on three different nanoporous TiO2 films. To calculate the energy distributions, we used the data
below the monolayer coverage which were determined from the LF equation and the generalized nonlinear regularization method which sets no assumption on position and shape of the solution (or the energy distribution function) (Table 2) [11,12]. The detailed procedures for determining the AEDs were described in elsewhere [11–13]. As shown in Fig. 4(c), similar shapes of AEDs were observed. All samples have two distinct AED peaks, which may indicate that two different types of surface energy for N719 exist mainly on the surface of the adsorbent. In the cases of MK-TiO2 and MS-TiO2, the lower AED peaks are located between 0 and 9.2 kJ/ mol with the maximum at about 3.15 (MS-TiO2) and 3.36 (MKTiO2) kJ/mol, whereas the higher AED peaks appeared between 9.2 and 16.0 kJ/mol with the maximum at about 11.76 (MS-TiO2) and 11.97 (MK-TiO2) kJ/mol. In contrast, the AED for the MU-TiO2 differ slightly from the case of MK-TiO2 and MS-TiO2. The AED peak shape, height, and location of MU-TiO2 are greatly broader, lower, and shifted toward relatively higher adsorption energy. The main and shoulder (or small) AED curves are in the range of 0–11.6 and 12.4–22.1 kJ/mol, where the pronounced maxima were located at 4.4 and 17.0 kJ/mol, respectively. To evaluate the photovoltaic performance of nanoporous TiO2 thin film, the current–voltage curves were obtained. The photo-
K.-J. Hwang et al. / Applied Surface Science 256 (2010) 5428–5433 Table 3 Photovoltaic performance of DSSCs using nanoporous TiO2 films.
MK-TiO2 film MS-TiO2 film MU-TiO2 film
Voc (V)
Isc (mA/cm2)
Fill factor (FF)
heff (%)
0.59 0.59 0.60
8.8 5.2 9.5
0.64 0.57 0.64
3.3 1.7 3.6
current–voltage (I–V) curves were measured using a source measure unit under irradiation of white light from a 1000 W Xenon lamp (Thermo Oriel Instruments). The incident light intensity and the active cell area were 100 mW/cm2 and 0.25 cm2, respectively. The I–V curves were used to calculate the short-circuit current (Isc), open-circuit voltage (Voc), fill factor (FF), and overall conversion efficiency (heff) of DSSC. It was observed that the overall conversion efficiency (heff) is in the following order: MU-TiO2 film (3.6%) > MK-TiO2 film (3.3%) > MS-TiO2 film (1.7%). This trend is identical to that of the N719 adsorption on nanoporous TiO2. Therefore, it is reasonable to conclude that the overall photovoltaic performance was substantially dependent on the adsorption amount of N719 (Fig. 5 and Table 3). 4. Conclusions Ordered nanoporous TiO2 particles (MK-TiO2, MS-TiO2, MUTiO2) were synthesized via the nano-replication route using different silica templates of KIT-6 (bicontinuous cubic Ia3d), SBA-15(2-D hexagonal P6mm), and MSU-H (2-D hexagonal P6mm), respectively. TiO2 films of MK-TiO2, MS-TiO2, MU-TiO2 were prepared and characterized by XRD, TEM, FE-SEM, AFM, and BET. The photovoltaic performance was evaluated from the overall
5433
conversion efficiency (heff), fill factor (FF), open-circuit voltage (Voc) and short-circuit current (Isc) from the I–V curves measured. The overall conversion efficiency (heff) is in the following order: MU-TiO2 film (3.6%) >MK-TiO2 film (3.3%) >MS-TiO2 film (1.7%). The results of adsorption properties obtained in this work led us to conclude that the photoelectric performance is strongly dependent on the adsorption amount of dye molecules and the pore size distributions of MK-TiO2, MS-TiO2, and MU-TiO2. Acknowledgements This study was supported by the research funds from Chosun University, 2009. References [1] B. O’Regan, M. Gratzel, Nature 353 (1991) 737. [2] L.L. Kazmerski, J. Electron. Spectrosc. Relat. Phenomena 150 (2006) 105. [3] Z. Zhang, S.M. Zak eeruddin, B.C. O’Regan, R. Humphry-Bak er, M. Gratzel, J. Phys. Chem. B 109 (2005) 2818. [4] P. Persson, S. Lunnell, Sol. Energy Mater. Sol. Cells 63 (2000) 139. [5] M.K. Nazeeruddin, A. Kay, I. Rodicio, R. Humphry-Baker, E. Mueller, P. Liska, N. Vlachopoulos, M. Gratzel, J. Am. Chem. Soc. 115 (1993) 6382. [6] J.W. Lee, K.J. Hwang, W.G. Shim, K.H. Park, H.B. Gu, K.H. Kwun, Korean J. Chem. Eng. 24 (2007) 847. [7] S.S. Kim, H.I. Lee, J.M. Kim, Chem. Lett. 37 (2008) 140. [8] F. Kleitz, S.H. Choi, R. Ryoo, Chem. Commun. (2003) 2136. [9] K.J. Hwang, S.J. Yoo, S.S. Kim, J.M. Kim, W.G. Shim, S.I. Kim, J.W. Lee, J. Nanosci. Nanotechnol. 8 (2008) 4976. [10] K.F. Chen, Y.C. Hsu, Q. Wu, S.S. Sun, Org. Lett. 11 (2) (2009) 377. [11] W.G. Shim, H.C. Kang, C. Kim, S.C. Kim, J.W. Lee, C.J. Lee, H. Moon, J. Nanosci. Nanotechnol. 6 (2006) 3583. [12] K.J. Hwang, S.J. Yoo, S.H. Jung, D.W. Park, S.I. Kim, J.W. Lee, Bull. Korean Chem. Soc. 30 (1) (2009) 172. [13] K. La´szlo´, P. Podkos´cielny, A. Da˛browski, Appl. Surf. Sci. 252 (2009) 5752.